Databk

R
, XILINX, XACT, XC2064, XC3090, XC4005, 4,750,155; 4,758,985; 4,820,937; 4,821,233; 4,835,418;
XC-DS501, FPGA Architect, FPGA Foundry, NeoCAD, 4,853,626; 4,855,619; 4,855,669; 4,902,910; 4,940,909;
NeoCAD EPIC, NeoCAD PRISM, NeoROUTE, Plus Logic, 4,967,107; 5,012,135; 5,023,606; 5,028,821; 5,047,710;
Plustran, P+, Timing Wizard, and TRACE are registered 5,068,603; 5,140,193; 5,148,390; 5,155,432; 5,166,858;
trademarks of Xilinx, Inc. 5,224,056; 5,243,238; 5,245,277; 5,267,187; 5,291,079;
5,295,090; 5,302,866; 5,319,252; 5,319,254; 5,321,704;
, all XC-prefix product designations, XACTstep, XACT-
5,329,174; 5,329,181; 5,331,220; 5,331,226; 5,332,929;
step Advanced, XACTstep Foundry, XACT-Floorplanner,
5,337,255; 5,343,406; 5,349,248; 5,349,249; 5,349,250;
XACT-Performance, XAPP, XAM, X-BLOX, X-BLOX plus,
5,349,691; 5,357,153; 5,360,747; 5,361,229; 5,362,999;
XChecker, XDM, XDS, XEPLD, XPP, XSI, Foundation
5,365,125; 5,367,207; 5,386,154; 5,394,104; 5,399,924;
Series, BITA, Configurable Logic Cell, CLC, Dual Block,
5,399,925; 5,410,189; 5,410,194; 5,414,377; 5,422,833;
FastCLK, FastCONNECT, FastFLASH, FastMap, Hard-
5,426,378; 5,426,379; 5,430,687; 5,432,719; 5,448,181;
Wire, LCA, Logic Cell, LogiCore, LogicProfessor, MicroVia,
5,448,493; 5,450,021; 5,450,022; 5,453,706; 5,466,117;
PLUSASM, PowerGuide, PowerMaze, Select-RAM,
5,469,003; 5,475,253; 5,477,414; 5,481,206; 5,483,478;
SMARTswitch, TrueMap, UIM, VectorMaze, VersaBlock,
5,486,707; 5,486,776; 5,488,316; 5,489,858; 5,489,866;
VersaRing, Spartan, Spartan-XL and ZERO+ are trade-
5,491,353; 5,495,196; 5,498,979; 5,498,989; 5,499,192;
marks of Xilinx, Inc.
5,500,608; 5,500,609; 5,502,000; 5,502,440; RE 34,363,
The Programmable Logic Company and The Programma- RE 34,444, and RE 34,808. Other U.S. and foreign patents
ble Gate Array Company are service marks of Xilinx, Inc. pending. Xilinx, Inc. does not represent that devices shown
All other trademarks are the property of their respective or products described herein are free from patent infringe-
owners. ment or from any other third party right. Xilinx assumes no
obligation to correct any errors contained herein or to
Xilinx does not assume any liability arising out of the appli- advise any user of this text of any correction if such be
cation or use of any product described or shown herein; nor made. Xilinx will not assume any liability for the accuracy or
does it convey any license under its patents, copyrights, or correctness of any engineering or software support or
maskwork rights or any rights of others. Xilinx reserves the
assistance provided to a user.
right to make changes, at any time, in order to improve reli-
ability, function or design and to supply the best product Xilinx products are not intended for use in life support appli-
possible. Xilinx will not assume responsibility for the use of ances, devices, or systems. Use of a Xilinx product in such
any circuitry described herein other than circuitry entirely applications without the written consent of the appropriate
embodied in its products. Xilinx devices and products are Xilinx officer is prohibited.
protected under one or more of the following U.S. Patents: Copyright 1999 Xilinx, Inc. All Rights Reserved.
4,642,487; 4,695,740; 4,706,216; 4,713,557; 4,746,822;
R
The Programmable Logic

Data Book
Xilinx Home Page (WWW): http://www.xilinx.com/
(“Answers” Database, File Download)
XDOCS E-mail Document Server: xdocs@xilinx.com
send E-mail with help in the header
Application Service Centers
North America Hotline: 1-408-879-5199 (USA, Xilinx headquarters)
1-800-255-7778 (North America)
Fax: 1-408-879-4442
Email: hotline@xilinx.com
United Kingdom Hotline: (44) 1932-820821

Fax: (44) 1932-828522
Email: ukhelp@xilinx.com
France Hotline: (33) 1-3463-0100

Fax: (33) 1-3463-0959
Email: frhelp@xilinx.com
Germany Hotline: (49) 89-93088-130

Fax: (49) 89-93088-188
Email: dlhelp@xilinx.com
Japan Hotline: (81) 3-3297-9163

Fax: (81) 3-3297-0067
Email: jhotline@xilinx.com
Korea Hotline: (82) 2-761-4277

Fax: (82) 2-761-4278
Email: korea@xilinx.com
Hong Kong Hotline: (85) 2-2424-5200

Fax: (85) 2-2424-7159
Email: hongkong@xilinx.com
Software Authorization and Licensing: 1-800-624-4782
On-Line Authorization: www.xilinx.com
2100 Logic Drive

San Jose, California 95124
United States of America
Telephone: (408) 559-7778
Fax: (408) 559-7114
On behalf of the employees of Xilinx, our sales representatives, our distributors, and our
manufacturing partners, welcome to our 1999 Data Book, and thank you for your interest in
Xilinx products and services.
As the inventor of Field Programmable Gate Array technology and the world’s leading
supplier of programmable logic, we would like to pledge our continuing commitment to
providing you, our users, with the best possible integrated circuit components, development
systems, and technical and sales support.
Over the past year, we have substantially expanded our product line with the full introduction
of 3.3V and 2.5V device offerings. These include the XC4000XLA, XC4000XLT, 9500XL,
SpartanXL, and Virtex series of programmable logic. We have continued to enhance our
leading-edge products as well, with new speed grades and improved pricing. The Alliance
and Foundation series products have set a new standard for functionality and ease-of-use in
programmable logic development systems. You can expect this pace of innovation to
continue, and even increase, as we maintain our leadership role in bringing leading-edge
programmable logic solutions to the market.
We look forward to satisfying all of your programmable logic needs.
Sincerely,
Wim Roelandts
Chief Executive Officer
R
Section Titles
1 Introduction
2 Development System Products and CORE Solutions Products
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
7 XC3000 & XC5200 Products
8 SPROM Products and Programming Support
9 HardWire FpgASIC Products
10 High-Reliability and QML Military Products
11 Packages and Thermal Characteristics
12 Testing, Quality, and Reliability
13 Technical Support and Services
14 Product Technical Information
15 Index
16 Sales Offices, Sales Representatives, and Distributors

Table of Contents
Section Titles
Introduction
An Introduction to Xilinx Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
Development System Products and CORE Solutions Products
Development Systems: Products Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Development Systems: Product Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
CORE Solutions Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13
Virtex Products
Virtex Products
Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
Virtex™ 2.5 V
Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
Spartan Products
Spartan Products
Spartan and SpartanXL Families

Spartan and SpartanXL Families Field Programmable Gate Arrays . . . . . . . . . . . . . . . . 4-38
Spartan and SpartanXL Families Field Programmable Gate Arrays . . . . . . . . . . . . . . . . 4-56
Spartan and SpartanXL Families Field Programmable Gate Arrays . . . . . . . . . . . . . . . . 68
XC9500 Products
XC9500/XL Series Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
FastFLASH™ XC9500XL High-Performance CPLD Family . . . . . . . . . . . . . . . . . . . . . . . 5-5
XC9536XL High Performance CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-21
XC95144XL High Performance CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-37

XC9500 In-System Programmable CPLD Family. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-59
XC9536 In-System Programmable CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-75
XC95108 In-System Programmable CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-91
XC4000 Products
XC4000E and XC4000X Series

XC4000E and XC4000X Series Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . 6-7
XC4000XLA/XV Field
Programmable Gate Arrays
XC4000XLA/XV
XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-203

XC3000 & XC5200 Products
XC3000 Series Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
XC3000 Series
XC5200 Series Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-79
XC5200 Series
SPROM Products and

Programming Support
SPROM Products and Programming Support Table of Contents . . . . . . . . . . . . . . . . . . 8-1
XC1700L Series High Density Serial Configuration PROMs . . . . . . . . . . . . . . . . . . . . . . 8-3
XC1700E Family of
Serial Configuration PROMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-15
Spartan and SpartanXL Families of Serial Configuration PROMs. . . . . . . . . . . . . . . . . . 8-27
HW-130 Programmer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-37
HardWire FpgASIC Products
Xilinx HardWire™ Product Family Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
High-Reliability and
QML Military Products
QPRO™ QML Certified and Radiation Hardened Products . . . . . . . . . . . . . . . . . . . . . . . 10-1
QPRO™ XQ4000E/EX QML High-Reliability Field Programmable Gate Arrays Table of Contents
10-7
QPROTM XQ4000E/EX QML High Reliability Field Programmable Gate Arrays. . . . . . . . 10-11
QPROTM XQ4000E/EX QML High Reliability Field Programmable Gate Arrays . . . . . . 10-45
QPRO™ XQ4000XL Series QML High-Reliability FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . 10-53
QPRO™ XQR4000XL Radiation Hardened Field Programmable Gate Arrays . . . . . . . . . 10-73
Packages and Thermal Characteristics
Packages and Thermal Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-1
Package Drawings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-23

Testing, Quality, and Reliability
Quality Assurance and Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-3
Technical Support and Services
Technical Support And Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-1
Product Technical Information
Product Technical Information Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-1
Xilinx FPGAs: A Technical Overview for the First-Time User . . . . . . . . . . . . . . . . . . . . . 14-5
Choosing a Xilinx Product Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-7
I/O Characteristics of the ‘XL FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-13
XC4000 Series
Technical Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-15
XC3000 Series
Technical Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-19
FPGA Configuration Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-30
Configuring Mixed FPGA

Daisy Chains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-38
Configuration Issues:
Power-up, Volatility, Security, Battery Back-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-40
Dynamic Reconfiguration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-44
Metastable Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-46
Set-up and Hold Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-49
Overshoot and Undershoot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-50
Boundary Scan in XC4000 and XC5200 Series Devices . . . . . . . . . . . . . . . . . . . . . . . . . . 14-51

Virtex I/V Curves for Various
Output Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-62
I/V Curves for Various

Device Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-64
Index
Book Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-1
Sales Offices, Sales Representatives, and Distributors
Sales Offices, Sales Representatives, and Distributors. . . . . . . . . . . . . . . . . . . . . . . . . . 16-1

Introduction
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

Introduction Table of Contents
An Introduction to Xilinx Products

About this Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
Data Sheet Categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
Data Book Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
About the Company . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Product Line Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Programmable Logic vs. Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Faster Design and Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Design Changes without Penalty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Shortest Time-to-Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Component Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Field Programmable Gate Arrays (FPGAs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
Complex Programmable Logic Devices (CPLDs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
HardWire devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
Serial PROMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
High-Reliability Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
Development System Products. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
Technical Support and Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
XC3000 Series Product Selection Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6
Spartan Series Product Selection Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
HardWire XH3 Product Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
HardWire XH3L Product Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
0
R
February 2, 1999 0 1*
About this Book Chapter 2 contains a discussion of the overall design

methodology when using Xilinx programmable logic and
This Data Book provides a “snapshot in time” in its listing of descriptions of Xilinx development system products. This
IC devices and development system software available chapter also includes a description of the Xilinx CORE 1
from Xilinx as of late 1998. New devices, speed grades, solutions products.
package types and development system products are con-
tinually being added to the Xilinx product portfolio. Users Chapter 3 includes the product descriptions for the Xilinx
are encouraged to contact their local Xilinx sales represen- Virtex Series of high-density, high-speed Field Programma-
tative and consult the WebLINX World Wide Web site ble Gate Arrays (FPGAs).
(http://www.xilinx.com) and the quarterly XCELL newsletter Chapter 4 includes the product descriptions for the Xilinx
for the latest information regarding new product availability. Spartan Series of low-cost Field Programmable Gate
This book covers the current XC4000E/EX/XL, XC4000XV, Arrays (FPGAs).
XC4000XLT, Spartan/XL, Virtex, XC5200, XC3000A/L, Chapter 5 contains the product descriptions for the Xilinx
XC3100A/L, XC9500/XL, XC1700D/L and XC1701L. Complex Programmable Logic Device (CPLD) products,
The product specifications for several older Xilinx product including the XC9000 series.
families are not included in this Data Book. This does not Chapter 6 includes the product descriptions for the Xilinx
imply that these products are no longer available. However, XC4000 Series of high-speed Field Programmable Gate
for new designs, users are encouraged to use the newer Arrays (FPGAs).
products described in this book, which offer better perfor-
Chapter 7 includes the product descriptions for the Xilinx
mance at lower cost than the older technologies. Product
XC3000 and XC5200 Series of low-cost Field Programma-
specifications for the older products are available at
ble Gate Arrays (FPGAs).
WebLINX, the Xilinx site on the World Wide Web, or
through your local Xilinx sales representative. Chapter 8 holds the product descriptions for the XC1700
families of Serial PROM devices. These Serial PROMs pro-
Data Sheet Categories vide a convenient, low-cost means of storing configuration
programs for the SRAM-based FPGAs. Chapter 8 also
In order to provide the most up-to-date information, some
describes the HW130 device programmer for the XC170X
component products included in this book may not have
series of Serial PROMs and the XC9500 series of CPLDs.
been fully characterized at the time of publication. In these
cases, the AC and DC characteristics included in the data Chapter 9 contains a brief overview of the HardWire prod-
sheets will be marked as Advance or Preliminary informa- uct line. Detailed product specifications are available in
tion. (Not withstanding the definitions of such terms, all separate Xilinx data sheets.
specifications are subject to change without notice.) These
Chapter 10 is an overview of Xilinx High-Reliability/Military
designations have the following meaning:
products. Detailed product specifications are available in
• Advance — Initial estimates based on simulation separate Xilinx data sheets.
and/or extrapolation from other speed grades, devices,
Chapter 11 contains a description of all the physical pack-
or device families. Use as estimates, but not for final
ages for the various IC products, including information
production. about the thermal characteristics of those packages.
• Preliminary — Based on preliminary characterization.
Changes are possible, but not expected. Chapter 12 discusses the testing, quality, and reliability of
• Final (unmarked) — Specifications not identified as Xilinx component products.
either Advance or Preliminary are to be considered Chapter 13 includes a listing of all the technical support
final. facilities provided by Xilinx.
Data Book Contents Chapter 14 contains additional information about Xilinx

components that is not provided in the product specifica-
Chapter 1 is a general overview of the Xilinx product line, tions of the earlier chapters. This includes some additional
and is recommended reading for designers who are new to electrical parameters that are not in the product specifica-
the field of high-density programmable logic. tions because they are not part of the manufacturing test
February 2, 1999 1-1

R
program for the particular device, but may be of interest to ment to FPGAs, offering simpler design software and more
the user. Also included in this chapter is a discussion of the predictable timing.
JTAG boundary test scan logic found in several Xilinx com-
As the market leader in one of the fastest growing seg-
ponent families.
ments of the semiconductor industry, Xilinx strategy is to
The final two sections contain an index to the topics focus its resources on creating new ICs and development
included in this Data Book and a listing of Xilinx sales system software, providing world-class technical support,
offices, sales representatives, and distributors. developing markets, and building a diverse customer base
across a broad range of geographic and end-use applica-
About the Company tion segments. The company has avoided the large capital
commitment and overhead burden associated with sole
Xilinx, Inc., offers the industry’s broadest selection of pro-
ownership and operation of a wafer fabrication facility.
grammable logic devices. With 1997 revenues of over $560
Instead, Xilinx has established alliances with several
million, Xilinx is the world’s largest supplier of programma-
high-volume, state-of-the-art CMOS IC manufacturers.
ble logic, and the market leader in Field Programmable
Using standard, high-volume processes assures low man-
Gate Arrays (FPGAs).
ufacturing costs, produces programmable logic devices
Xilinx was founded in 1984, based on the revolutionary idea with well-established reliability, and provides for early
of combining the logic density and versatility of gate arrays access to advances in CMOS processing technology.
with the time-to-market advantages and convenience of
Xilinx headquarters are located in San Jose, California.
user-programmable standard parts. One year later, Xilinx
The company markets its products worldwide through a
introduced the world’s first Field Programmable Gate Array.
network of direct sales offices, manufacturers’ representa-
Since then, through a combination of architectural and
tives, and distributors (as listed in the back of this book).
manufacturing process improvements, the company has
The company has representatives and distributors in over
continually increased device performance, in terms of
38 countries.
capacity, speed, and ease-of-use, while lowering costs.
In 1992, Xilinx expanded its product line to include Product Line Overview
advanced Complex Programmable Logic Devices
Field Programmable Gate Arrays (FPGAs) and Complex
(CPLDs). For the user, CPLDs are an attractive comple-
Programmable Logic Devices (CPLDs) can be used in vir-
tually any digital logic system. Over 50 million Xilinx compo-
1-2 February 2, 1999

R
nents have been used in a wide variety of end-equipment market research firm McKinsey & Co. concluded that a
applications, ranging from supercomputers to hand-held six-month delay in getting to market can cost a product
instruments, from central office switches to centrifuges, and one-third of its lifetime potential profit. With mask-pro-
from missile guidance systems to guitar synthesizers. grammed gate arrays, design iterations can easily add that
much time, and more, to a product schedule.
Xilinx achieved its leading position through a continuing
commitment to provide a complete product solution. This Once the decision has been made to use Xilinx program-
encompasses a focus on all three critical areas of the mable logic, a choice must be made from a number of prod-
high-density programmable solution “triangle”: components uct families, device options, and product types. The
(silicon), software, and service (Figure 1). information in the product selection matrices that follow can
help guide that selection; detailed product specifications
Programmable Logic vs. Gate Arrays are available in subsequent chapters of this book. Since
1
Xilinx programmable logic devices provide the benefits of many component products are available in common pack-
high integration levels without the risks or expenses of ages with common footprints, designs often can be
semi-custom and custom IC development. Some of the migrated to higher or lower density devices, or even across
benefits of programmable logic versus mask-programmed some product families, without any printed circuit board
gate arrays are briefly discussed below. changes. Design ideas, represented in text or schematic
format, are converted into a configuration data file for an
Faster Design and Verification FPGA or CPLD device using the Xilinx development soft-
ware running on a PC or workstation.
Xilinx FPGAs and CPLDs can be designed and verified
quickly while the same process requires several weeks with
Component Products
gate arrays. There are no non-recurring engineering (NRE)
costs, no test vectors to generate, and no delay while wait- Xilinx offers the broadest line of programmable logic
ing for prototypes to be manufactured. devices available today, with hundreds of products featur-
ing various combinations of architectures, logic densities,
Design Changes without Penalty package types, and speed grades in commercial, industrial,
Because the devices are software-configured and and military grades. This breadth of product offerings
user-programmed, modifications are much less risky and allows the selection of the programmable logic device that
can be made anytime - in a manner of minutes or hours, as is best suited for the target application.
opposed to the weeks it would take with a gate array. This Xilinx programmable logic offerings include several families
results in significant cost savings in design and production. of reprogrammable FPGAs and FLASH-memory-based
CPLDs (Figure 2). HardWire devices are mask-pro-
Shortest Time-to-Market
grammed versions of the reprogrammable FPGAs, and
When designing with Xilinx programmable logic, provide a transparent, no-risk migration path to lower-cost
time-to-market is measured in days or a few weeks, not the devices for high-volume, stable designs. Additionally, a
months often required when using gate arrays. A study by family of Serial PROM devices is available to store configu-
• Optimized circuits/architectures • Powerful but easy

SO
• Highest performance/densities • Integrated across families

N
FT
CO
• Deep submicron processes • Seamless integration into

W
LI
customer CAE system

AR
SI
• Unmatched quality
E
and reliability
S E RV I C E
• Global world class sales/distribution support
• Global world class technical support: FAEs/support center/on-line/internet
• Global world class manufacturing: quality/capacity/delivery X5955
Figure 1: The Xilinx Programmable Solution Triangle

R
ration programs for the reprogrammable FPGA devices. Complex Programmable Logic Devices (CPLDs)
Many devices are available in military temperature range
Designers more comfortable with the speed, design sim-
and/or MIL-STD-883B versions, for high-reliability and mili-
plicity, and predictability of PALs may prefer CPLD devices.
tary applications.
Conceptually, CPLDs consist of multiple PAL-like function
Field Programmable Gate Arrays (FPGAs) blocks that can be interconnected through a switch matrix
(Figure 4). The XC9000 CPLD series features in-system
FPGA devices feature a gate-array-like architecture, with a
programmable FLASH technology, and, like most of the
matrix of logic cells surrounded by a periphery of I/O cells,
FPGA families, includes built-in JTAG boundary scan test
as diagrammed in Figure 3. Segments of metal intercon-
logic.
nect can be linked in an arbitrary manner by programmable
switches to form the desired signal nets between the cells. HardWire devices
FPGAs combine an abundance of logic gates, registers, Xilinx HardWire products are a family of .5µ to .35µ
and I/Os with fast system speed. Xilinx offers several state-of-the-art sea-of-gates multi-mask ASIC devices. The
families of reprogrammable, static-memory-based (SRAM- HardWire flow is the lowest risk method for cost reduction
based) FPGAs, including the Virtex, Spartan, XC3000, for FPGA based systems. HardWire uses a turnkey devel-
XC4000, and XC5000 series. opment flow. Customer engineering resources are used to
ASIC Alternatives discuss features implemented in the FPGA design and
review conversion reports. No customer engineering
Gate Arrays resources are used to convert the design. Each HardWire
Custom
Highest Density device is developed using the FPGA’s design files. This
Xilinx
ASIC Tools guarantees the HardWire device will be functionally com-
Product Line patible with the FPGA. No customer generated test vectors
are required with HardWire. Each HardWire device is
CPLD FPGA HardWire™ tested using a combination of industry standard and Xilinx
ISP Programmable Custom
PAL Architecture Gate Array Transparent Conversion
patented test methods in a full scan methodology. Hard-
Medium Density Architecture 100% Tested Wire process technologies, conversion methods and test-
Simple Tools High Density
ing procedures provide the most cost-effective alternative
ASIC Tools
to traditional gate arrays.
PAL Devices Serial PROMs
Programmable AND/OR
Architecture The XC1700 family features one-time programmable serial
Low Density
Simple Tools PROMs ranging in density from about 18,000 bits to over
X5957
260,000 bits. These serial PROMs are an easy-to-use,
cost-effective method for storing configuration data for the
Figure 2: Application-Specific IC Products SRAM-based FPGAs.
PROGRAMMABLE I/O BLOCKS

INTERCONNECT
X1153
LOGIC BLOCKS
Figure 3: FPGA Architecture

R

,
FB FB
FB FB
I/O Interconnect I/O

Matrix
1
FB FB
FB FB
X5956
Figure 4: CPLD Architecture
High-Reliability Devices forms include the ubiquitous PC and several popular

workstations.
Xilinx was the first company to offer high-reliability FPGAs
by introducing MIL-STD-883B qualified XC2000 and Technical Support and Service
XC3000 series devices in 1989. MIL-STD-883B members
of the XC4000 FPGA series are currently available, and Providing global, world-class manufacturing, technical sup-
qualified versions of additional Xilinx families are in devel- port, and sales/distribution support is an essential founda-
opment. The product line also includes Standard Microcir- tion of the Xilinx product strategy. Xilinx manufacturing
cuit Drawing (SMD) versions of several families. Some facilities have earned ISO9002 certification, and Xilinx
Xilinx devices are available in tested die form through quality and reliability achievements are among the world’s
arrangements with manufacturing partners. best - not just for programmable logic suppliers, but among
all semiconductor companies. Comprehensive technical
Development System Products support facilities include training courses, extensive prod-
uct documentation and application notes, a quarterly tech-
Xilinx offers a complete software environment for the imple-
nical newsletter, the WebLINX World Wide Web site,
mentation of logic designs in Xilinx programmable logic
technical support hotlines, and a cadre of Field Application
devices. This environment combines powerful technology
Engineers. Sales support is provided by a worldwide net-
with a flexible, easy-to-use graphical interface to help users
work of representatives and distributors.
achieve the best possible designs, regardless of experi-
ence level. The user has a wide range of choices between
a fully-automatic implementation and detailed involvement
in the layout process. The development system provides all
the implementation tools required to design with Xilinx logic
devices, including the following:
• libraries and interfaces for popular schematic editors,
logic synthesis tools, and simulators
• design manager/flow engine
• module generator
• map, place, and route compilation software
• static timing analyzer
• hardware debugger
Xilinx is committed to an “open system” approach to
front-end design creation, synthesis, and verification. Xilinx
devices are supported by the broadest number of EDA ven-
dors and synthesis vendors in the industry. Supported plat-

R
XC3000 Series Product Selection Matrix
XC3020A/L
XC3030A/L
XC3042A/L
XC3064A/L
XC3090A/L
XC3120A
XC3130A
XC3142A
XC3164A
XC3190A
XC3195A
XC3142L
XC3190L
DEVICES
XC3000 Series
FEATURES
Low Voltage
KEY
Low Cost/ (3.3 V)

High Performance
Low Power High
Performance
Max Logic Gates (K) 1.5 2 3 4.5 6 1.5 2 3 4.5 6 7.5 3 6
DENSITY
Max RAM Bits N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Typical Gate Range (K) 1-1.5 1.5-2 2-3 3.5-4.5 5-6 1-1.5 1.5-2 2-3 3.5-4.5 5-6 6.5-7.5 2-3 5-6
CLBs 64 100 144 224 320 64 100 144 224 320 484 144 320
Flip-Flops 256 360 480 688 928 256 360 480 688 928 1320 480 928
Output Drive (mA) 4 4 4 4 4 8 8 8 8 8 8 4 4
FEATURES
JTAG (IEEE 1149.1) N N N N N N N N N N N N N

Dedicated Arithmetic N N N N N N N N N N N N N
Quiescent Current (mA) 0.5/0.02 0.5/0.02 0.5/0.02 0.5/0.02 0.5/0.02 8 8 8 8 8 8 1.5 1.5
Fastest Speed Grade -6/-8 -6/-8 -6/-8 -6/-8 -6/-8 -09 -09 -09 -09 -09 -09 -2 -2
XC4028EX/XL
XC4036EX/XL
XC4005E/XL
XC4010E/XL
XC4013E/XL
XC4020E/XL
XC4044XL
XC4052XL
XC4062XL
XC4085XL
XC4003E
XC4006E
XC4008E
XC4025E
DEVICES
XC4000 Series
FEATURES
High Density
KEY
High Performance
Select-RAM™ Memory
System Gate Range* (Logic and RAM) (K) 2-5 3-9 4-12 6-15 7-20 10-30 13-40 15-45 18-50 22-65 27-80 33-100 40-130 55-180
Logic Cells 238 466 608 770 950 1368 1862 2432 2432 3078 3800 4598 5472 7448
DENSITY
Max Logic Gates, (no RAM) (K) 3 5 6 8 10 13 20 25 28 36 44 52 62 85

Max RAM Bits (no Logic) 3200 6272 8192 10368 12800 18432 25088 32768 32768 41472 51200 61952 73728 100352
CLBs 100 196 256 324 400 576 784 1024 1024 1296 1600 1936 2304 3136
Flip-Flops 360 616 768 936 1120 1536 2016 2560 2560 3168 3840 4576 5376 7168
Output Drive (mA) 12 12 12 12 12 12 12 12 12 12 12 12 12 12
FEATURES
JTAG (IEEE 1149.1) Y Y Y Y Y Y Y Y Y Y Y Y Y Y

Dedicated Arithmetic Y Y Y Y Y Y Y Y Y Y Y Y Y Y
Quiescent Current (mA) 10 10 10 10 10 10 10 10 10 10 10 10 10 10
Fastest Speed Grade -1 -09 -1 -1 -09 -09 -09 -1 -09 -09 -09 -09 -09 -09
*Maximum System gates assume 20% of CLBs used as RAM

R
Spartan Series Product Selection Matrix

DEVICES
XCS05 XCS10 XCS20 XCS30 XCS40

Spartan Series
XCS05XL XCS10XL XCS20XL XCS30XL XCS40XL
FEATURES
High Performance
KEY
Select-RAM™ Memory
Low Cost
System Gate Range* (Logic and RAM) (K) 2-5 3-10 7-20 10-30 13-40 1
Logic Cells 238 466 950 1368 1862
DENSITY
Max Logic Gates, (no RAM) (K) 3 5 10 13 20

Max RAM Bits (no Logic) 3200 6272 12800 18432 25088
CLBs 100 196 400 576 784
Flip-Flops 360 616 1120 1536 2016
Output Drive (mA) 12/24 12/24 12/24 12/24 12/24
FEATURES
JTAG (IEEE 1149.1) Y Y Y Y Y

Dedicated Arithmetic Y Y Y Y Y
Quiescent Current (mA) 3/0.1 3/0.1 3/0.1 3/0.1 3/0.1
Fastest Speed Grade -4/-5 -4/-5 -4/-5 -4/-5 -4/-5
*Maximum System gates assume 20% of CLBs used as RAM

DEVICES
XC5200 Series XC5202 XC5204 XC5206 XC5210 XC5215

FEATURES
KEY
Low Cost
Max Logic Gates (K) 3 6 10 16 23

Max RAM Bits N/A N/A N/A N/A N/A
DENSITY
Typical Gate Range (K) 2-3 4-6 6-10 10-16 15-23

CLBs/Logic Cells 64 120 196 324 484
Flip-Flops 256 480 784 1296 1936
Output Drive (mA) 8 8 8 8 8
FEATURES
JTAG (IEEE 1149.1) Y Y Y Y Y

Dedicated Arithmetic Y Y Y Y Y
Quiescent Current (mA) 15 15 15 15 15
Fastest Speed Grade -3 -3 -3 -3 -3

R

XC9536 XC9572 XC95108 XC95144 XC95216 XC95288
DEVICES
XC9500 Series CPLD

XC9536XL XC9572XL XC99144XL XC95288XL
High Performance
FEATURES
Low Cost
KEY
JTAG ISP
High Reliability
Gates (K) 0.8 1.6 2.4 3.2 4.8 6.4
DENSITY
Macrocells 36 72 108 144 216 288

Flip-Flops 36 72 108 144 216 288
Output Drive (mA) 8/24 8/24 24 8/24 24 8/24
FEATURES
JTAG (IEEE 1149.1) Y Y Y Y Y Y

Dedicated Arithmetic Y Y Y Y Y Y
Quiescent Current (mA) 10/30 20/65 100 45/160 – 85/300
Fastest Speed Grade (ns) 4/5 5/7 7 5/7 10 6/15
Note 1: XC9500 series is 5V
HardWire XH3 Product Family

FEATURES DENSITY FEATURES DEVICES
HardWire XH3 Series XH302 XH304 XH306 XH308 XH310 XH312
.5µ process
KEY
5.0-volt
Pin compatible FPGA replacement
# Pads 136 172 204 240 292 352
Largest Package for Base Array PQ100 PQ160 PQ208 PQ240 PQ240 BG432
DLM Usable Gates 14,500 28,000 42000 62,000 97,000 146,000
JTAG Y Y Y Y Y Y
CE Y Y Y Y Y Y
RAM Y Y Y Y Y Y
HardWire XH3L Product Family

FEATURES DENSITY FEATURES DEVICES
HardWire XH3L Series XH310L XH312L XH316L XH320L XH326L
.35µ process
KEY
3.3-volt
Pin compatible FPGA replacement
# Pads 220 284 352 432 560
Largest Package for Base Array PQ208 BG256 BG352 BG432 BG560
TLM Usable Gates 95,000 170,000 270,000 425,000 750,000
JTAG Y Y Y Y Y
CE Y Y Y Y Y
RAM Y Y Y Y Y

fq
Development System Products and

CORE Solutions Products
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

Development System Products and
CORE Solutions Products
Table of Contents
Development System Products and CORE Solutions Products
Development Systems: Products Overview

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Product Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Development Systems: Product Descriptions

Development Systems Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
Foundation Series: Foundation Base System (PC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
Foundation Series: Foundation Base-Express System with VHDL/Verilog Synthesis (PC) . . . . . . . . . . 2-5
Foundation Series: Foundation Standard System (PC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6
Foundation Series: Foundation Express System (PC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
Alliance Series: Alliance Base (PC or Workstation) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
Alliance Series: Alliance Standard (PC or Workstation) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
Alliance Series Options (PC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12
CORE Solutions Overview

Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13
CORE Solutions Products. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13
CORE Solutions Data Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13
LogiCORE Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13
AllianceCORE Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-15
0
Development Systems: Products

Overview
December 10, 1997 (Version 2.0) 0 2* Product Overview
Introduction all Xilinx programmable logic devices, featuring the

industry’s largest FPGA devices.
Leading-edge silicon products, state-of-the art software
solutions and world-class technical support make up the Foundation Series
total solution delivered by Xilinx. The software component
The Xilinx Foundation Series provides everything required
of this solution is critical to the success of every design
project. Xilinx Software Solutions provide powerful tools
to design a programmable logic device in an easy-to-use 2
environment. This fully integrated tool set allows users to
which make designing with programmable logic simple.
access design entry, synthesis, implementation and simu-
Push button design flows, integrated on-line help, multime-
lation tools in a ready-to-use package. Every step in the
dia tutorials, plus high performance automatic and
design process is accomplished using graphical tool bars,
auto-interactive tools, help designers achieve optimum
icons and pop-up menus supported by interactive tutorials
results. And the industry’s broadest array of programmable
and comprehensive on-line help.
logic technology and EDA integration options deliver unpar-
alleled design flexibility. The Xilinx Foundation Series features support for stan-
dards based HDL design. All configurations support the
Product Overview popular ABEL language, with integrated compilers opti-
mized for each target architecture. HDL configurations
Xilinx Software Solutions are available in two different prod-
include integrated VHDL/Verilog synthesis from Synopsys
uct series making it easy for designers to choose the right
system for their needs. These two series support the indus- with tutorials and graphical HDL design entry tools to turn
try’s broadest array of programmable logic IC families. This new users into experts quickly and easily.
allows users to standardize their design tools for all pro- HDL Configurations
grammable logic applications and use these tools to realize
the benefits of the industry’s highest performance and den- HDL configurations of the Foundation Series contain inte-
sity FPGAs and CPLDs. It also makes it easy to migrate grated VHDL/Verilog synthesis and graphical interactive
designs to new technologies and re-use existing designs in HDL entry tools with the following features:
new applications. • On-line tutorial teaches the art of VHDL design.
The Xilinx Foundation Series provides designers with a • Xilinx HDL Editor provides color coding, syntax
complete, ready-to-use solution for programmable logic checking and single click error navigation making it
design. easy to create and debug VHDL, Verilog and ABEL
designs.
The Xilinx Alliance Series provides designers powerful • Graphical State Machine editor makes the design of
integration of Xilinx design tools with their existing EDA simple or complex state machines simple and intuitive.
environment. • HDL Language Assistant provides libraries of common
functions with optimized VHDL, Verilog and ABEL code.
Flexible Configurations • FPGA and CPLD specific synthesis and optimization
Xilinx Software Solutions are available in two device config- from Synopsys tools produce high-utilization,
urations giving designers a cost-effective way to match high-performance results
their tools to the design methodologies they require. These
configurations are available for both the Foundation and Alliance Series
Alliance Series. The Alliance Series provides powerful and integrated
• Base configurations provide push button design flows design tools for users who require a quality solution for their
and support a broad array of FPGA and CPLD devices chosen EDA design solution. With the Alliance Series,
targeted for low density and high volume applications. users can choose from a wide range of design techniques
• Standard configurations combine push button flows with including schematic capture, module-based design and
powerful auto-interactive tools. These tools give HDL design solutions. With standard based design inter-
designers more influence and control over faces including EDIF, VITAL, VHDL, Verilog and SDF, this
implementation while maintaining the benefits of design series provides maximum flexibility, portability, mixed ven-
automation. Standard configurations include support for dor support, and design reuse.
December 10, 1997 (Version 2.0) 2-1

R
Development Systems: Products Overview
Quality integration with leading EDA vendors such as which delivers push-button design flows and incremental
ALDEC, Exemplar, Cadence, Mentor Graphics, Model design capabilities. These Xilinx-exclusive capabilities
Technology, OrCAD, Synopsys, Synplicity, Veribest and leverage results from previous design iterations to reduce
VIEWlogic provide tightly-coupled environments that make runtimes and shorter design iterations to less than ten min-
it easy to move through the design process and through a utes. As engineers design complex circuits incrementally,
mixed EDA vendor flow. The EDA vendors are supported this technology allows them to work in their preferred meth-
through the Xilinx Alliance Program, insuring high quality odology.
tools and accuracy of results. Information on Xilinx Alliance
M1 Technology also delivers advanced timing driven
Program vendors can be found on the Xilinx WEB page
place-and route capabilities to deliver maximum design
www.xilinx.com.
performance through push-button flows.
The Alliance Series includes an enhanced set of
easy-to-use features including, design manager, flow M1 Technical Benefits
engine, installation, on-line documentation, and answer
database. In addition, the Alliance Series includes a power-
Maximum Design Performance
ful and complete implementation toolset, LogiBLOX (next M1 technology enables the user to achieve maximum
generation module generation), fully integrated EDA ven- design performance by providing a unique combination of
dor support, and a powerful gate-level optimizer. Also advanced algorithms and interactive tools. Designer pro-
included are new advanced place and route software that ductivity is greatly enhanced through use of simple,
has incremental design capabilities and SMARTspecs (a push-button flows and optional auto-interactive tools. Cus-
robust timing constraint language). Users can achieve up to tomer testing has shown that M1 technology used with
25% performance improvements with no additional elapse XC4000XL/XV devices results in 70 percent shorter run
time through the use of the Alliance Series Turns Engine. times, up to a 25 percent performance improvement, and
The Turns Engine uses networked workstations to run mul- the ability to place and route devices with up to 100 percent
tiple place and route passes for a single design. This fea- utilization with a push-button flow.
ture is included with the Alliance Series BASE and
Standard workstation development systems. The libraries Modular Software System
and interface provide Xilinx Unified Library schematic sym- The modular architecture of the Xilinx M1 technology
bols, HDL synthesis libraries, VITAL(VHDL) and Verilog allows rapid delivery of incremental technologies, new fea-
simulation models with timing information and translators tures, device support, and versions of its leading software
through a standard netlist format. All of these tools provide product families. New feature sets can now be released
a complete spectrum of high density design methodologies independently resulting in users’ ability to quickly complete
from fully-automatic to hand-crafted and close integration designs without having to re-learn new tools as enhance-
with Xilinx LogiCores and AllianceCore partners. ments are made. The investment Xilinx has made in the M1
technology ensures that the continuous delivery of innova-
Alliance Series Options tive device architectures and improved software solutions
VIEWlogic Workview Office Development System options can be done more rapidly, and predictably than previous
as part of the Alliance Series are intended for users who software versions.
want the integration of a complete solution with the power
Methodology Flexibility
to access board and system level design tools. These prod-
ucts include VIEWlogic Workview Office schematic capture High-level design methodologies are becoming the meth-
and simulation tools. odology choice for the design of complex programmable
logic. M1 technology delivers programmable logic specific
Xilinx M1 Software Technology high-level flows. The flows provide high-quality, high perfor-
mance optimized results, and afford fast, flexible design
M1 technology represents Xilinx’s next generation software
technology. This advanced technology developed as a changes and iterations to match the way engineers design.
result of the Xilinx merger with NeoCAD Inc., enables digi- Designers employ a mixture of graphical and lan-
tal system designers to increase design performance, guage-based design entry methods while providing an
leverage standards-based, high-level design methodolo- easy-to-learn environment for Hardware Description Lan-
guage (HDL) based design. Xilinx recognizes that design
gies and quickly receive new software features and device
environments are variant and, therefore, has created a flex-
support through Xilinx Foundation Series and Alliance
Series software solutions. ible system enabling the customer to choose the best meth-
odology for their environment or design challenge.
Increased Design Performance
The M1 technology provides dramatically improved design
performance through advanced place-and-route software
2-2 December 10, 1997 (Version 2.0)

0
Development Systems: Product

Descriptions
November 25, 1997 (Version 2.0) 0 2*
Development Systems Descriptions

It’s simple to order a Xilinx Development System. Just
choose a Foundation or Alliance Series and a few options.
Give your local Xilinx Sales Office a call for information
about our evaluation kits.
2
Foundation Series
• Foundation Base System (PC)
• Foundation Base-Express System (PC)
• Foundation Standard System (PC)
• Foundation Express System (PC)
Alliance Series
• Alliance Base (PC or Workstation)
• Alliance Standard (PC or Workstation)
Alliance Series Options

• VIEWlogic Workview Office Standard Development
System Options (PC)
November 25, 1997 (Version 2.0) 2-3

R
Foundation Series: Foundation Base System (PC)

Overview Package Features - Foundation Base
The Foundation Series provides a complete, ready-to-use System
design system for the design of Xilinx programmable logic
FND FND FND FND
devices. The Foundation Base System provides design Feature
BAS STD BSX EXP
entry (schematic and Abel HDL), simulation, and device
CPLD Devices √ √ √ √
implementation tools for a broad array of FPGA and CPLD
1
devices targeted for low density and high volume applica- FPGA Devices √ √ √1 √
tions. Libraries and Interface √ √ √ √
Schematic Editor √ √ √ √
System Features HDL Editor √ √ √ √
• Project manager Graphical State Editor √ √ √ √
• Schematic editor ABEL 6 Entry / Synthesis √ √ √ √
• Integrated HDL editor with support for the Abel 6 HDL
VHDL Entry / Synthesis √ √
• Functional and timing simulator
Verilog Entry / Synthesis √ √
• EDIF, VHDL (VITAL compliant), and Verilog / SDF
design interfaces Schematic-centric Synthesis √ √ √ √
• Device implementation software for Xilinx CPLDs and HDL-centric Synthesis √
FPGAs Simulator √ √ √ √
• Comprehensive on-line help, on-line documentation, Device Implementation √ √ √ √
and software tutorials Maintenance2 √ √ √ √
• Software maintenance, including hotline support and 11/12/97
software updates Notes: 1. Spartan, XC3x00A/X, XC4000E/X up to
XC4010E/X, and XC5200 up to XC5210.
Device Support 2. A period of maintenance is included with new
design system licenses, after which annual
• CPLDs: maintenance contracts may be purchased.
- XC9500 Contact your Xilinx sales representative for more
• FPGAs: information.
- XC4000E/X Up to XC4010E/X
- Spartan
- XC3x00A/L
- XC5200 Up to XC5210 FPGAs
Required Hardware Environment

• Windows 95 and Windows NT 4.0 compatible PCs
• Minimum memory requirements: 32 MB RAM, 32-64
MB Virtual Memory
• CD-ROM drive
2-4 November 25, 1997 (Version 2.0)

R
Foundation Series: Foundation Base-Express System with VHDL/Verilog

Synthesis (PC)
Overview Package Features - Foundation
The Foundation Series provides a complete, ready-to-use Base-Express System
FND FND FND FND
devices. The Foundation Express System incorporates Feature
BAS STD BSX EXP
advanced synthesis technology from Synopsys, and pro-
vides design entry (schematic and HDL), VHDL and Verilog
FPGA Devices √ 1 √ √1 √
synthesis, simulation, and device implementation tools for a
broad array of FPGA and CPLD devices targeted for low Libraries and Interface √ √ √ √
density and high volume applications. Schematic Editor √ √ √ √
2
HDL Editor √ √ √ √
System Features
Graphical State Editor √ √ √ √
• Project manager ABEL 6 Entry / Synthesis √ √ √ √
• Schematic editor
• Integrated HDL editor with support for VHDL, Verilog,
and Abel 6 HDL
• VHDL and Verilog synthesis, including compilation and Schematic-centric Synthesis √ √ √ √
optimization HDL-centric Synthesis √
• Functional and timing simulator Simulator √ √ √ √
• EDIF, VHDL (VITAL compliant), and Verilog / SDF Device Implementation √ √ √ √
design interfaces Maintenance2 √ √ √ √
• Device implementation software for Xilinx CPLDs and 11/12/97
FPGAs Notes: 1. Spartan, XC3x00A/L, XC4000E/X up to
• Comprehensive on-line help, on-line documentation, XC4010E/X, and XC5200 up to XC5210.
and software tutorials 2. A period of maintenance is included with new
• Software maintenance, including hotline support and design system licenses, after which annual
maintenance contracts may be purchased.
software updates Contact your Xilinx sales representative for more
information.
Device Support
• CPLDs:
- XC9500
• FPGAs:
- XC4000E/X Up to XC4010E/X
- Spartan
- XC3x00A/L
- XC5200 Up to XC5210 FPGAs

• Minimum memory requirements: 32 MB RAM, 32-64
MB Virtual Memory
• CD-ROM drive

R
Foundation Series: Foundation Standard System (PC)

FND FND FND FND
devices. The Foundation Standard System provides design Feature
BAS STD BSX EXP
entry (schematic and Abel HDL), simulation, and device
implementation tools for all Xilinx CPLDs and Xilinx
1
FPGAs. FPGA Devices √ √ √1 √
Libraries and Interface √ √ √ √
System Features Schematic Editor √ √ √ √
• Project manager HDL Editor √ √ √ √
• Schematic editor Graphical State Editor √ √ √ √
• Integrated HDL editor with support for the Abel 6 HDL ABEL 6 Entry / Synthesis √ √ √ √
• Functional and timing simulator
• EDIF, VHDL (VITAL compliant), and Verilog / SDF
design interfaces
• Device implementation software for Xilinx CPLDs and Schematic-centric Synthesis √ √ √ √
FPGAs HDL-centric Synthesis √
• Comprehensive on-line help, on-line documentation, Simulator √ √ √ √
and software tutorials Device Implementation √ √ √ √
• Software maintenance, including hotline support and Maintenance2 √ √ √ √
software updates 11/12/97
Notes: 1. Spartan, XC3x00A/L, XC4000E/X up to
Device Support XC4010E/X, and XC5200 up to XC5210.
• CPLDs: 2. A period of maintenance is included with new
design system licenses, after which annual
- XC9500 maintenance contracts may be purchased.
• FPGAs: Contact your Xilinx sales representative for more
- XC4000E/X information.
- Spartan
- XC3x00A/L
- XC5200

• Minimum memory requirements
- Small Devices (< 10K gates): 32 MB RAM, 32-64
MB Virtual Memory
- Medium Devices (10K to 30K gates): 64 MB RAM,
64-128 MB Virtual Memory
- Large Devices (> 30K gates): 128 MB RAM, 128-256
MB Virtual Memory
• CD-ROM drive

R
Foundation Series: Foundation Express System (PC)

FND FND FND FND
devices. The Foundation Express System incorporates Feature
BAS STD BSX EXP
advanced synthesis technology from Synopsys, and pro-
vides design entry (schematic and HDL), VHDL and Verilog
1
synthesis, simulation, and device implementation tools for FPGA Devices √ √ √1 √
all Xilinx CPLDs and Xilinx FPGAs. Libraries and Interface √ √ √ √
Schematic Editor √ √ √ √
System Features HDL Editor √ √ √ √
• Project manager Graphical State Editor √ √ √ √ 2
• Schematic editor ABEL 6 Entry / Synthesis √ √ √ √
• Integrated HDL editor with support for VHDL, Verilog,
and Abel 6 HDL
• VHDL and Verilog synthesis, including compilation and
optimization Schematic-centric Synthesis √ √ √ √
• Functional and timing simulator HDL-centric Synthesis √
• EDIF, VHDL (VITAL compliant), and Verilog / SDF Simulator √ √ √ √
design interfaces Device Implementation √ √ √ √
• Device implementation software for Xilinx CPLDs and Maintenance2 √ √ √ √
FPGAs 11/12/97
• Comprehensive on-line help, on-line documentation, Notes: 1. Spartan, XC3x00A/L, XC4000E/X up to
and software tutorials XC4010E/X, and XC5200 up to XC5210.
• Software maintenance, including hotline support and 2. A period of maintenance is included with new
software updates design system licenses, after which annual
maintenance contracts may be purchased.
Contact your Xilinx sales representative for more
Device Support information.
• CPLDs:
- XC9500
• FPGAs:
- XC4000E/X
- Spartan
- XC3x00A/L
- XC5200

• Minimum memory requirements
- Small Devices (< 10K gates): 32 MB RAM, 32-64
MB Virtual Memory
- Medium Devices (10K to 30K gates): 64 MB RAM,
64-128 MB Virtual Memory
- Large Devices (> 30K gates): 128 MB RAM, 128-256
MB Virtual Memory
• CD-ROM drive

R
Alliance Series: Alliance Base (PC or Workstation)

Overview Libraries and Interfaces
Next generation FPGA/CPLD design solutions leveraging Cadence
“Open Systems” integration with premier EDA partners for
• Concept schematic libraries and Verilog-XL simulation
devices up to 10,000 gates.
models
Base System Features:
Mentor
• EDA Libraries & Interfaces • Falcon Framework schematic capture library and
• Design Manager and Flow Engine ModelSim simulation models
• LogiBLOX Module Generator • Leonardo synthesis libraries and interfaces are
• Gate Optimizer available from Mentor or Exemplar Logic
• Complete HDL design methodology support
• Incremental design capabilities Synopsys
• Place and route utilizing SMARTspecs • HDL Design Solutions (VHDL and Verilog)
• Re-entrant router • Design Compiler, FPGA Compiler II, FPGA Express,
• Multi-pass PAR VSS
• Timing Analyzer • Vital Simulation models
• Standard netlist and backannotation (EDIF, SDF, VITAL • DesignWare arithmetic modules
VHDL and Verilog) * No libraries required to support FPGA Express
• Xchecker Hardware Debugger (workstation only) VIEWlogic
• Workview Office schematic capture library and
Package Includes:
functional and timing simulation interface
• Alliance Quick Start Guide
• Alliance Release Document Exemplar
• Answer Database • Leonardo and Galileo synthesis libraries and interfaces
• Core Technology CD are available from Exemplar Logic
• CAE Libraries CD
• On-line Documentation CD with DynaText browser
Synplicity
• Synplify synthesis libraries and interfaces are available
• Hardware Cable
from Synplicity
• Demoboard
Model Technology
Device Support: • ModelSim, V-System HDL simulation libraries and
• CPLDs: interface
- XC9500
• FPGAs:
- XC4000E/X Up to XC4010E/X Contact your local EDA sales office to purchase these EDA
- Spartan tools.
- XC3x00A/L
- XC5200 Up to XC5210 FPGAs Support and Updates Include:
• Answers Database - http://www.xilinx.com or Answers
electronic book included.
• Hotline Telephone Support
• Apps FAX and E-Mail
• Online Documentation
• World Wide Web Access
• Technical Newletter
• Extensive Application Notes
• Software Updates (for in-maintenance customers)
- A period of maintenance is included with new design
system licenses, after which annual maintenance
contracts may be purchased. Contact your Xilinx
sales representative for more information

R
Required Hardware Environment (PC) Required Hardware Environment

• Fully IBM compatible PC486/Pentium (Workstation)
- NEC98 supported • Ultra Sparc (or equivalent)
• Windows 95, Windows NT 4.0 - Sun OS 4.1.3 and 4.1.4
- Chinese, Korean and Japanese versions - Solaris 2.5
• Minimum 300 Mbytes hard-disk space • HP715 (or equivalent)
• CD-ROM drive - HP-UX 10.2
• VGA display • RS6000
• Serial port mouse - AIX 4.1.5 (no GUIs)
• One parallel and two serial ports • 64 MB RAM (Use additional to increase performance)
• 32 MB RAM (Use additional RAM to increase • 64MB min Swap Space
performance) • Color Monitor
• 32 MB - 64 MB Virtual Memory 2

R
Alliance Series: Alliance Standard (PC or Workstation)

Overview Libraries and Interfaces
Next generation FPGA/CPLD design solutions leveraging Cadence
“Open Systems” integration with premier EDA partners for • Concept schematic libraries and Verilog-XL simulation
unlimited gate capacity. models
Base System Features: Mentor
• EDA Libraries & Interfaces • Falcon Framework schematic capture library and
• Design Manager and Flow Engine ModelSim simulation models
• LogiBLOX Module Generator • Leonardo synthesis libraries and interfaces are
• Gate Optimizer available from Mentor or Exemplar Logic
• Full HDL design methodology support
Synopsys
• Incremental design capabilities
• HDL Design Solutions (VHDL and Verilog)
• Place and route utilizing SMARTspecs
• Design Compiler, FPGA Compiler II, FPGA Express,
• Re-entrant router
VSS
• Multi-pass PAR
• Vital Simulation models
• Timing Analyzer
• DesignWare arithmetic modules
• Standard netlist and backannotation (EDIF, SDF, VITAL
* No libraries required to support FPGA Express
VHDL and Verilog)
• Xchecker Hardware Debugger (workstation only) VIEWlogic
• Workview Office schematic capture library and
Package Includes: functional and timing simulation interface
• Alliance Quick Start Guide
Exemplar
• Alliance Release Document
• Leonardo and Galileo synthesis libraries and interfaces
• Answer Database
are available from Exemplar Logic
• Core Technology CD
• CAE Libraries CD Synplicity
• On-line Documentation CD with DynaText browser • Synplify synthesis libraries and interfaces are available
• Hardware Cable from Synplicity
• Demoboard
Model Technology
Device Support: • ModelSim, V-System HDL simulation libraries and
• CPLDs: interface
- XC9500
• FPGAs:
Contact your local EDA sales office to purchase these EDA
- XC4000E/X
tools.
- Spartan
- XC3x00A/L Support and Updates Include:
- XC5200
• Answers Database - http://www.xilinx.com or Answers
electronic book included.
• Hotline Telephone Support
• Online Documentation
• World Wide Web Access
• Technical Newletter
• Extensive Application Notes
• Software Updates (for in-maintenance customers)
- A period of maintenance is included with new design
system licenses, after which annual maintenance
contracts may be purchased. Contact your Xilinx
sales representative for more information

R
Required Hardware Environment (PC) Required Hardware Environment

• Fully IBM compatible PC486/Pentium (Workstation)
- NEC98 supported • Ultra Sparc (or equivalent)
• Windows 95, Windows NT 4.0 - Sun OS 4.1.3 and 4.1.4
- Chinese, Korean and Japanese versions - Solaris 2.5
• Minimum 300 Mbytes hard-disk space • HP715 (or equivalent)
• CD-ROM drive - HP-UX 10.2
• VGA display • RS6000
• Serial port mouse - AIX 4.1.5 (no GUIs)
• One parallel and two serial ports • Small Devices: (28K or <) XC4000E;
• Small Devices: (8K or <) XC9536 - XC95108; XC4028EX - XC4036EX; XC4005XL - XC4028XL
XC4003E - XC4008E; XC4005XL - XC4008XL - 64 MB RAM (Use additional RAM to increase
- 32 MB RAM (Use additional RAM to increase performance) 2
performance) - 64MB min Swap Space
- 32 -64 MB Virtual Memory • Large Devices (36K or >) XC4036XL - XC4062XL
• Medium Devices: (10K- 28K) XC95144 - XC95216; - 128 MB RAM (Use additional RAM to increase
XC4010E - XC4025E; XC4028EX - XC4036EX; performance)
XC4010XL - XC4028XL - 128 MB min Swap Space
- 64 MB RAM (Use additional RAM to increase • Color Monitor
performance)
- 64-128 MB Virtual Memory
• Large Devices: (36K or >) XC4036XL - XC4062XL
- 128K RAM (Use additional RAM to increase
performance)
- 128 - 256 MB Virtual Memory

R
Alliance Series Options (PC)

Overview Support and Updates Include:
VIEWlogic Workview Office schematic capture and gate • Answers Database - http://www.xilinx.com or Answers
simulator development system with libraries and interfaces electronic book included.
for Xilinx FPGAs and CPLDs. • Hotline Telephone Support
Workview Office Standard Features: • Software Updates (for in-maintenance customers)
• Workview Office schematic editor • Online Documentation
• Workview Office gate simulator • World Wide Web Access
• Libraries and interfaces • Technical Newletter
• Hotline support • Extensive Application Notes
• Software maintenance for 90-days
Required Hardware Environment:
Libraries Support: • Fully IBM compatible PC486/Pentium
• CPLDs: • Windows 95, Windows NT 4.0
- XC9500 - Chinese, Korean and Japanese versions
• FPGAs: • Minimum 500 Mbytes hard-disk space
- XC4000E/X • CD-ROM drive
- Spartan • VGA display
- XC3x00A/L • Serial port mouse
- XC5200 • One parallel and two serial ports
• 64 Mbytes RAM recommended (increase to improve
performance)

0
R
November 24, 1998 (Version 2.0) 0 2* Product Overview
Background view) which lists all of the functions available today. This
table will be your best guide to locating a specific product. If
The ASIC core industry has been developing for over a you don't see what you need, check the AllianceCORE
decade. Today there exists a wealth of intellectual property Partner Profiles, Areas of Expertise section, for each of our
(IP) that is readily available from numerous sources. During AllianceCORE partners. Our partners will be more than
this time, however, programmable logic did not have the willing to discuss the possibility of producing a core specif-
density or the performance needed to accommodate large 2
ically for your needs.
IP cores.
Today, things have changed considerably. Xilinx is shipping
Data Book Contents
FPGAs like the XL family that have usable densities up to The contents of the data book are as follows:
1,00,000 gate Virtex. Now, not only is the use of • Introduction
pre-defined logic functions in programmable logic a possi- - Program Overview
bility, it is becoming a requirement to meet ever-shrinking
- Product Listing by Application Segment
product development cycles.
• LogiCORE Products, sold and supported by Xilinx
As a result, many ASIC core vendors and system designers - Product Overview
are beginning to look at using cores for their programmable - PCI
logic designs. It is for this reason that Xilinx created the - DSP
CORE Solutions portfolio of products. - CORE Generator products
• AllianceCORE Products, sold and supported by Xilinx'
CORE Solutions Products Partners
- Program Overview
CORE Solutions products support four application areas.
The application areas are as follows: - Products
- AllianceCORE Partner Profiles
• Standard Bus Interfaces - such as PCI, PCMCIA, • LogiBLOX, GUI-based small function generator
USB and Plug-and-Play ISA. • Reference Designs
• DSP Functions - These range from small building • Sales Offices, Representatives and Distributors
blocks such adders, registers and multipliers, to larger
system-level functions such as FIR filters and Ordering Information
Reed-Solomon coders.
To order a copy, request the CORE Solutions Data Book
• Telecom and Networking - building blocks for popular
from the Xilinx Literature Department. In the US call
communications standards.
1-800-231-3386. For international locations call
• Base-Level Functions - a broad category of functions
1-408-879-5017 or you can send an E-mail request to:
used across many application segments. These include
the every small parameterizable LogiBLOX macros up literature@xilinx.com.
through larger functions such as UARTs and DMA An electronic version of the CORE Solutions Data Book
controllers. (1.2M Adobe Acrobat.pdf format) can also be downloaded
from:
CORE Solutions Data Book
www.xilinx.com/products/logicore/core_sol.pdf
The goal of the CORE Solutions portfolio of products is to
provide cores with the shortest time-to-market and best LogiCORE Products
possible device utilization the programmable logic industry
has to offer. Xilinx has published a brand new data book LogiCORE products are sold, licensed and supported by
focused entirely on programmable logic cores and related Xilinx. They are developed internally by Xilinx or jointly with
products. Now there is one definitive sourcebook with a partner.
detailed descriptions of all Xilinx CORE Solutions. Typically, LogiCORE designs use Smart-IP technology to
When you receive your copy of the CORE Solutions Data achieve a highly predictable functionality and performance
Book, become familiar with the Product Listing by Applica- of the core. For example, the performance of the core will
tion Segment Table, (reproduced at the end of this over-

R
not be affected by the adjacent custom logic, the user’s the time-to-market benefits are maintained for high-com-
choice of EDA development tools, or coding style. As a plexity FPGAs.
result, you can save several months in design and verifica-
Xilinx PCI Solutions
tion time.
Xilinx’ PCI solution includes devices, tools and cores
The cores that Xilinx provides as LogiCORE products typi-
needed to build a cost-effective single-chip PCI system in
cally fall into one of two categories.The first are high-perfor-
record time.
mance cores that require a thorough understanding and
control of the FPGA technology and implementation soft- • LogiCORE PCI Designs - With over 400 licensed cores,
ware in order to achieve the desired performance and com- Xilinx LogiCORE PCI have become the industry's most
plexity. An example of a core in this category is the used PCI solution. The product line includes both 32-
LogiCORE PCI interface. and 64-bit, 33 to 66MHz fully compliant PCI interface
designs with 0 wait-state performance and up to
The second category are cores that benefit from a very
528MB/s sustained bandwidth. The critical PCI min and
specialized implementation in the FPGA. An example is the
max timing is guaranteed by use of Smart-IP, which will
LogiCORE DSP modules that are implemented using
save the user significant development time and enable
unique algorithms. Those algorithms fit the
100% PCI compliance.
lookup-table-based architecture of the FPGA. The result is
• Virtex - The industry's fastest FPGAs allow you to
outstanding performance and device utilization, often more
integrate all standard PCI interface variations including
than 10 times better than generic HDL descriptions.
the high-performance, 64-bit 66MHz core.
Smart-IP • Spartan XL - offers the lowest cost 32-bit, 33MHz PCI
solution. Integrate a 0 wait-state PCI interface with your
Cores made with Smart-IP technology provide flexibility own design, at a price below standard PCI chips.
while maintaining their performance and predictability • XC4000XLA - allows you to integrate a high
regardless of device size and the number of cores used in performance 32-bit, 33MHz PCI interface with up to
the device. Smart-IP technology combines the power of the 60K gates of user designed logic.
Xilinx implementation software with the unique features of • Web-based configuration - provides you with an easy
the Xilinx FPGA architectures to give your cores with con- way to configure the PCI core according to your needs,
sistent performance, no matter where you place them. and gives you instant access to the latest design files
and documentation.
Xilinx CORE Generator System • XPERTS Design Services - gives you a worldwide
In addition to actual cores, Xilinx is committed to develop network of specially trained consultants providing
enabling design tools and methodologies to facilitate design services for Xilinx PCI.
“Design Reuse” with FPGAs. The first products available in PCI is an extremely high-performance and complex specifi-
this category are the CORE Generator System available on cation that is challenging to meet in any technology. To
CD) and the web-based PCI Configuration tool. This inno- meet the stringent PCI specification the core is carefully
vative methodology for acquiring and using cores combines hand-tuned for the targeted architecture. Placement and
the benefits of: routing for the critical parts of the core is locked down to
• A firm core with parameterizable and predictable ensure that timing can be met every time the core is used.
performance To achieve our goals, the LogiCORE development team is
• Flexibility of system level design, facilitated by working closely with both the IC and Software teams. As an
behavioral languages such as VHDL and Verilog. example of this teamwork, new methodologies for charac-
Additionally Xilinx, using the web as a distribution mecha- terizing and modeling our FPGAs have been developed.
nism, gives the user access to the latest versions and The result is access to state of the art technology and
enhancements of cores at: expertise, that allows you to complete your PCI application
in record time.
www.xilinx.com/products/logicore/coregen/corelinx.htm
Since the introduction of the first core in early 1996, Xilinx
LogiCORE products are customized to fit your specific
has built up solid knowledge about PCI. We are committed,
application using an intuitive graphical user interface.
and will continuously develop our PCI products to remain
Based on your inputs, the CORE Generator System gener-
state of the art.
ates a proven core with highly predictable timing which can
be integrated into any VHDL-, Verilog- or schematic-based Xilinx DSP Solutions
design flow. As a result, you can integrate several individu-
ally proven cores with given performance into one system Using an FPGA to implement high performance DSP func-
on a single FPGA. Because each core is already verified, tions often allows a radical performance advantage over
fixed processors while maintaining maximum flexibility and

R
the shortest time-to-market. Until now, tools to automate AllianceCORE Criteria

the design process have been lacking and most designs
A core must meet a minimum set of criteria before it can
have been completed manually by experienced FPGA
receive the AllianceCORE label.
designers.
With the introduction of Xilinx’ CORE Generator System for Core Selection
DSP, complex parameterized DSP building blocks can be The AllianceCORE program looks at cores from a practical
implemented automatically with the performance and den- point of view. A programmable logic version of a core must
sity of a hand-tuned implementation. LogiCORE DSP mod- have value over an ASIC or standard product version of the
ules can be used with VHDL-, Verilog- or schematic based same function. It must be cost effective and make sense for
design methodologies. use in a programmable device in a production system. If a
Higher level DSP cores and DSP prototyping boards are candidate core does not pass these simple test, then it
available from our AllianceCORE partners. does not make sense to invest the effort to convert it to an
AllianceCORE module. 2
DSP system level modeling tools are available to mathe-
matically model LogiCORE based FPGA designs and aid in Core Qualification
determining optimal core parameters. systemView by Ela-
Generic, synthesizable cores offer maximum flexibility for
nix supports the Xilinx DSP LogiCORE libraries with
users with unique requirements. This is typically the format
bit-true modeling and is integrated with Xilinx CORE Gen-
for cores provided to the ASIC market. With programmable
erator System software.
logic, however, this flexibility can come at the expense of
The Xilinx DSP Solution consists of: efficiency and performance. It can take a considerable
• Standard Xilinx FPGA components - XC4000, Spartan, amount of effort to get a specific core to synthesize in a way
Virtex that meets density and timing requirements. Time spent
• DSP parameterizable LogiCORE FPGA cores trying to accomplish this can quickly reduce the
• AllianceCORE DSP cores and prototyping boards time-to-market advantage of using programmable logic and
• DSP system level tools integration cores in the first place.
• DSP starter kit Xilinx is not interested in promoting generic, synthesizable
• Regional dedicated DSP Field Applications Engineers functions as AllianceCOREs. Instead, AllianceCOREs are
• DSP Ph.D. level support generally provided as parameterizable black-boxes that
allow customization in critical areas. This guarantees that
Acquiring LogiCORE Products the implementation is optimized for density while still meet-
LogiCORE products are available from your local Xilinx ing performance, preserving the time-to-market value of
sales representative similar to other Xilinx software prod- programmable logic. Flexibility is provided by allowing you
ucts. Xilinx and your local sales representative will also be to quickly implement your unique logic on the same device.
your primary source for support of the core, the devices Source code versions of the cores are also available from
and the design tools. the partners at additional cost for those who need ultimate
flexibility.
You can also send email questions to:
Announced AllianceCOREs have been implemented and
logicore@xilinx.com.
verified in a Xilinx device. They are available immediately
for purchase in a Xilinx-specific format. Timing-critical
AllianceCORE Overview cores designed to adhere to an industry standard also
The AllianceCORE program is a cooperative effort between come with appropriate constraints files in order to guaran-
Xilinx and independent third-party core developers. It is tee functionality and compliance. AllianceCOREs origi-
designed to produce a broad selection of industry-standard nated from either schematic or HDL entry tools.
solutions dedicated for use in Xilinx programmable logic.
Core Integration
Xilinx takes an active role with its partners in the process of
productizing AllianceCOREs. This is unique to the Alli- AllianceCOREs are not just cores, they are complete solu-
anceCORE program. Because the process is so involved, tions for system designs. While cores by themselves have
we work closely with our partners to select the right cores value, in many cases it is often not enough to just supply a
first. This naturally limits the number of partners we can generic core. You may need additional tools such as sys-
work with at any one time and subsequently the number of tem software and prototyping equipment to help you rapidly
available cores. At the same time it raises the quality and integrate the core into your design, perform system debug
usability of the cores that are offered. in a real-world environment, and then quickly convert the
prototype to a production unit. This is particularly true of
complex functions.

R
Many AllianceCORE functions are supported by Xil- If you want additional information about the AllianceCORE
inx-based demonstration or prototyping boards. Some also program or are interested in becoming a partner, contact
have system simulation models or debug software. All of Xilinx directly.
this allows you to evaluate and work with the function
Xilinx, Inc.
before you have to layout your board. These tools are pro-
2100 Logic Drive
vided by the AllianceCORE partner, usually at additional
San Jose, CA 95124
cost. Descriptions of the support tools available for each
Attn: Mark Bowlby, AllianceCORE Product Manager
core are included in the CORE Solutions Data Book.
Phone: +1 408-879-5381
Complete solutions like these help preserve the value of Fax: +1 408-879-4780
using programmable logic while minimizing the support E-mail: alliancecore@xilinx.com
burden for the core provider. URL: www.xilinx.com/products/logicore
/alliance/tblpart.htm
Acquiring AllianceCORE Products
AllianceCORE products are sold and serviced directly by
the AllianceCORE partners since they are the experts for
their particular products. They are responsible for pricing,
licensing terms, delivery and technical support. Contact
information for each partner is included in the Alli-
anceCORE Partner Profiles section of the CORE Solutions
Data Book.
Table 1: AllianceCORE Products
Check www.xilinx.com/products/logicore/tblcores.htm for the latest listing of available Cores

Function Name Provider
Standard Bus Interface Products
CAN Bus Interface SICAN Microelectronics
FireWire Demo Board I2P
FireWire Link Layer Core I2P
IIC Interface Memec Design Services
PCMCIA Card Interfaces Mobile Media Research
PCMCIA Development Tools Mobile Media Research
USB 3-Port Hub Controller Mentor
USB Development Tools Mentor
USB Function Controller Cores Mentor
Digital Signal Processing
YCrCb2RGB Color Space Converter Perigee, LLC
Communications & Networking Products
Networking
Cell Assembler CoreEl MicroSystems
Cell Delineation CoreEl MicroSystems
CRC10/32 Cores CoreEl MicroSystems
10/100 Ethernet MACCores CoreEl MicroSystems
UTOPIA Master & Slave Cores CoreEl MicroSystems
Forward Error Correction
Reed-Solomon Codecs Integrated Silicon Systems
Reed-Solomon Encoder Memec Design Services
Viterbi Decoder CAST, Inc.
Telecommunications
Data Encryption Standard Engine Memec Design Services
HDLC Protocol Core Integrated Silicon Systems
PPP8 HDLC CoreEl MicroSystems

R
Table 1: AllianceCORE Products
Check www.xilinx.com/products/logicore/tblcores.htm for the latest listing of available Cores

Function Name Provider
MT1F T1 Framer Virtual IP Group
Base-Level Products
Processor Products
C2901 Microprocessor Slice CAST, Inc.
C2910a Microprogram Controller CAST, Inc.
RISC CPU Cores T7L Technology Inc.
RISC Development Tools T7L Technology Inc.
V8 uRISC 8-bit RISC Microprocessor VAutomation
2
IntellicoreTM Prototyping System VAutomation
Process Peripherals
M8237 DMA Controller Virtual IP Group
M8254 Programmable Timer Virtual IP Group
M8255A Peripheral Interface CAST, Inc.
M8255 Peripheral Interface Virtual IP Group
XF8255 Peripheral Interface Memec Design Services
XF8256 Multifunction Controller Memec Design Services
M8259 Interrupt Controller CAST, Inc.
M8259 Interrupt Controller Virtual IP Group
XF8279 Keyboard Display Interface Memec Design Services
XF9128 Video Terminal Logic Controller Memec Design Services
SDRAM Controller NMI Electronics
DRAM Controller NMI Electronics
UARTs
Compact UART CAST, Inc.
C6850 ACIA CAST, Inc.
XF8250 UART Memec Design Services
C8251 UART CAST, Inc.
C16450 UART CAST, Inc.
M16450 UART Virtual IP Group
C16550 UART with RAM CAST, Inc.
M16550A UART with RAM Virtual IP Group
Generic Core Development Tools
DSP Prototyping Boards GV & Associates
FPGA Development Module Memec Design Services
Core Evaluation Card NMI Electronics

R
Table 2: LogiCORE Products Table 3: Reference Designs
Function Name Function Name

Standard Bus Interface Products Standard Bus Interface Products
PCI Synthesizable PCI Bridge Design Example
PCI32 4000 (V2.0) ISA Plug and Play Interface
PCI32 Spartan Digital Signal Processing
PCI64/66 Virtex 16-Tap, 8-Bit FIR Filter
PCI32 Virtex Base-Level Products
PCI32 Bridge Design 4Mb Virtual SPROM
PCI64 Bridge Design Adders, Subtractors, Accumulators
PCI32 Power Management Configuring FPGAs over a Processor Bus
HotPCI Spartan Prototyping Board Counters (3 types)
Vireo Device Driver Kits Dynamic Microcontroller
Digital Signal Processing FIFOs (2 types)
Correlators Frequency/Phase Comparator for PLL
1D RAM/ROM Based Correlators Harmonic Frequency Synthesizer and FSK Modulator
Filters Multiplexers, Barrel Shifters
Comb Filter Pulse-Width Modulation
SDA FIR Filter Serial Code Conversion between BCD and Binary
Dual Channel SDA FIR Filter
PDA FIR Filter
Table 4: LogiBLOX Products
Transforms
Fast Fourier Transforms Function Name
DSP Building Blocks Base-Level Products
SDA FIR Control Logic Basic Elements
Sine/Cosine Clock Divider
Time Skew Buffers (3 types) Comparator
Base-Level Products Constant
Basic Elements Counter
Constant Data Register
2/3/4-Input Multiplexers Decoder
Register Simple Gates
Math Functions Input/Output
1’s and 2’s Complement Multiplexer
Scaled by 1/2 Accumulator Shift Register
Adders (4 types) Tristate
Integrator Math Functions
Multipliers (4 types) Accumulator
Square Root Adder/Subtractors
Subtracters (2 types) Memories
Memories ROM, RAM, Synch-RAM, Dual Port RAM
Delay Element
Synchronous FIFO
Look-up Tables (2)
RAM and ROM
Generic Core Development Tools
Xilinx CORE Generator System

R
Virtex Products
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

0
R
Virtex Products
Table of Contents
0 3*
Virtex™ 2.5 V Field Programmable Gate Arrays

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
Virtex Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
Higher Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
Architectural Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
Virtex Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
Input/Output Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
Input Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6 3
Output Path. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
I/O Banking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
Configurable Logic Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Look-Up Tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Storage Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Additional Logic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
Arithmetic Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
BUFTs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
Block RAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
Programmable Routing Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
Local Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
General Purpose Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
I/O Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
Dedicated Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-11
Global Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-11
Clock Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
Delay-Locked Loop (DLL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-13
Development System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-13
Design Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14
Design Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14
Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14
Configuration Modes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-15
Slave Serial Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-15
Master Serial Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-17
SelectMAP Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-17
Write . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-19
Boundary-Scan Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-21
Configuration Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-21
Delaying Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
Start-Up Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
Data Stream Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
Readback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
Pin Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-23
Virtex DC Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24
Definition of Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24
All specifications are subject to change without notice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24
Virtex Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24
Virtex Recommended Operating Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24
Virtex DC Characteristics Over Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . 3-25
3-1
R
Virtex Products Table of Contents
Virtex DC Input and Output levels. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-25

Virtex Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-26
Virtex IOB Input Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-26
Virtex IOB Output Switching Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-27
Virtex Pad-to-Pad Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-28
Virtex Clock Distribution Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-28
Virtex Clock Distribution Switching Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-28
Virtex CLB Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-29
Virtex CLB Arithmetic Switching Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-30
Virtex CLB SelectRAM Switching Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-30
Virtex BLOCKRAM Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-31
Virtex TBUF Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-31
Virtex Test Access Port Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-31
Virtex Pin Outs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-32
Pin-Out Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-32
Pin-Out Diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-42
TQ144 Pin-out Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-43
PQ240/HQ240 Pin-out Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-44
BG256 Pin-out Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-45
Device/Package Combinations and Maximum I/O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-49
Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-49
Revision Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-50
3-2
0
R
Virtex™ 2.5 V
Field Programmable Gate Arrays
January 27, 1999 (Version 1.2) 0 3* Advance Product Specification
Features • Supported by FPGA Foundation™ and Alliance

Development Systems
• Fast, high-density Field-Programmable Gate Arrays - Complete support for Unified Libraries, Relationally
- Densities from 50k to 1M system gates Placed Macros, and Design Manager
- System performance up to 200 MHz - Wide selection of PC and workstation platforms
- 66-MHz PCI Compliant • SRAM-based in-system configuration
- Hot-swappable for Compact PCI - Unlimited reprogrammability
• Multi-standard SelectIO™ interfaces - Four programming modes
- 16 high-performance interface standards • 0.22-µm five-layer metal process 3
- Connects directly to ZBTRAM devices • 100% factory tested
• Built-in clock-management circuitry
- Four dedicated delay-locked loops (DLLs) for Description
advanced clock control
- Four primary low-skew global clock distribution nets, The Virtex FPGA family delivers high-performance,
plus 24 secondary global nets high-capacity programmable logic solutions. Dramatic
• Hierarchical memory system increases in silicon efficiency result from optimizing the
- LUTs configurable as 16-bit RAM, 32-bit RAM, 16-bit new architecture for place-and-route efficiency and exploit-
dual-ported RAM, or 16-bit Shift Register ing an aggressive 5-layer-metal 0.22-µm CMOS process.
- Configurable synchronous dual-ported 4k-bit RAMs These advances make Virtex FPGAs powerful and flexible
- Fast interfaces to external high-performance RAMs alternatives to mask-programmed gate arrays. The Virtex
• Flexible architecture that balances speed and density family comprises the nine members shown in Table 1.
- Dedicated carry logic for high-speed arithmetic Building on experience gained from previous generations
- Dedicated multiplier support of FPGAs, the Virtex family represents a revolutionary step
- Cascade chain for wide-input functions forward in programmable logic design. Combining a wide
- Abundant registers/latches with clock enable, and variety of programmable system features, a rich hierarchy
dual synchronous/asynchronous set and reset of fast, flexible interconnect resources, and advanced pro-
- Internal 3-state bussing cess technology, the Virtex family delivers a high-speed
- IEEE 1149.1 boundary-scan logic and high-capacity programmable logic solution that
- Die-temperature sensing device enhances design flexibility while reducing time-to-market.
Table 1: Virtex Field-Programmable Gate Array Family Members.
System Maximum Max Select

Device CLB Array Logic Cells BlockRAM Bits
Gates Available I/O RAM Bits
XCV50 57,906 16x24 1,728 180 32,768 24,576
XCV100 108,904 20x30 2,700 180 40,960 38,400
XCV150 164,674 24x36 3,888 260 49,152 55,296
XCV200 236,666 28x42 5,292 284 57,344 75,264
XCV300 322,970 32x48 6,912 316 65,536 98,304
XCV400 468,252 40x60 10,800 404 81,920 153,600
XCV600 661,111 48x72 15,552 500 98,304 221,184
XCV800 888,439 56x84 21,168 514 114,688 301,056
XCV1000 1,124,022 64x96 27,648 514 131,072 393,216
January 27, 1999 (Version 1.2) 3-3

R
Virtex Architecture Architectural Description

Virtex devices feature a flexible, regular architecture that
Virtex Array
comprises an array of configurable logic blocks (CLBs) sur-
rounded by programmable input/output blocks (IOBs), all The Virtex user-programmable gate array, shown in
interconnected by a rich hierarchy of fast, versatile routing Figure 1, comprises two major configurable elements: con-
resources. The abundance of routing resources permits the figurable logic blocks (CLBs) and input/output blocks
Virtex family to accommodate even the largest and most (IOBs).
complex designs. • CLBs provide the functional elements for constructing
Virtex FPGAs are SRAM-based, and are customized by logic
loading configuration data into internal memory cells. In • IOBs provide the interface between the package pins
some modes, the FPGA reads its own configuration data and the CLBs
from an external PROM (master serial mode). Otherwise, CLBs interconnect through a general routing matrix (GRM).
the configuration data is written into the FPGA (Select- The GRM comprises an array of routing switches located at
MAP™ and slave serial modes). the intersections of horizontal and vertical routing chan-
The standard Xilinx Foundation™ and Alliance Series™ nels. Each CLB nests into a VersaBlock™ that also pro-
Development systems deliver complete design support for vides local routing resources to connect the CLB to the
Virtex, covering every aspect from behavioral and sche- GRM.
matic entry, through simulation, automatic design transla- The VersaRing™ I/O interface provides additional routing
tion and implementation, to the creation, downloading, and resources around the periphery of the device. This routing
readback of a configuration bit stream. improves I/O routability and facilitates pin locking.
Higher Performance The Virtex architecture also includes the following circuits
that connect to the GRM.
Virtex devices provide better performance than previous
generations of FPGA. Designs can achieve synchronous • Dedicated block memories of 4096 bits each
system clock rates up to 200 MHz including I/OVirtex inputs • Clock DLLs for clock-distribution delay compensation
and outputs comply fully with PCI specifications, and inter- and clock domain control
faces can be implemented that operate at 33 MHz or 66 • 3-State buffers (BUFTs) associated with each CLB that
MHz. Additionally, Virtex supports the hot-swapping drive dedicated segmentable horizontal routing
requirements of Compact PCI. resources
Xilinx thoroughly benchmarked the Virtex family. While per- Values stored in static memory cells control the config-
formance is design-dependent, many designs operated urable logic elements and interconnect resources. These
internally at speeds in excess of 100 MHz and can achieve values load into the memory cells on power-up, and can
200 MHz. Table 2 shows performance data for representa- reload if necessary to change the function of the device.
tive circuits, using worst-case timing parameters..
DLL IOBs DLL
Table 2: Performance for Common Circuit Functions
VersaRing
Function Bits Virtex -6
Register-to-Register
16 5.0 ns
Adder
64 7.2 ns
VersaRing
VersaRing
BRAMs
BRAMs
8x8 5.1 ns
IOBs
IOBs
Pipelined Multiplier CLBs

16 x 16 6.0 ns
16 4.4 ns
Address Decoder
64 6.4 ns
16:1 Multiplexer 5.4 ns
9 4.1 ns
VersaRing
Parity Tree 18 5.0 ns
36 6.9 ns IOBs
DLL DLL
Chip-to-Chip
HSTL Class IV 200 MHz vao_b.eps
LVTTL,16mA, fast slew 180 MHz

Figure 1: Virtex Architecture Overview
3-4 January 27, 1999 (Version 1.2)

R
Supplementary Description documents on the WebLINX™ The three IOB storage elements function either as
website (http://www.xilinx.com/partinfo/databook.htm) aug- edge-triggered D-type flip-flops or as level sensitive
ment the following description of the various Virtex-archi- latches. Each IOB has a clock signal (CLK) shared by the
tecture components. The Supplementary Descriptions three flip-flops and independent clock enable signals for
provide more detailed information and cover the following each flip-flop.
topics. In addition to the CLK and CE control signals, the three
• Input/Output Block flip-flops share a Set/Reset (SR). For each flip-flop, this sig-
• Configurable Logic Block nal can be independently configured as a synchronous Set,
• Memory Resources a synchronous Reset, an asynchronous Preset, or an asyn-
• Clock Distribution chronous Clear.
• Routing Resources
The input and output buffers and all of the IOB control sig-
• Configuration and Readback nals have independent polarity controls.
• Boundary Scan
• Power Consumption All pads are protected against damage from electrostatic
discharge (ESD) and from over-voltage transients. Two
Input/Output Block forms of over-voltage protection are provided, one that per- 3
mits 5-V compliance, and one that does not. For 5-V com-
The Virtex IOB, Figure 2, features SelectIO™ inputs and
pliance, a zener-like structure connected to ground turns
outputs that support a wide variety of I/O signalling stan-
on when the output rises to approximately 6.5 V. When 5-V
dards, see Table 3. These high-speed inputs and outputs
compliance is not required, a conventional clamp diode
are capable of supporting PCI interfaces up to 66 MHz.
may be connected to the output supply voltage, VCCO. The
Table 3: Supported Select I/O Standards type of over-voltage protection can be selected indepen-
dently for each pad.
Input Output Board
Reference Source Termination Optional pull-up and pull-down resistors and an optional
I/O Standard weak-keeper circuit are attached to each pad. Prior to con-
Voltage Voltage Voltage
(VREF) (VCCO) (VTT) figuration all outputs not involved in configuration are forced
LVTTL N/A 3.3 N/A into their high-impedance state. The pull-down resistors
2 – 24 mA and the weak-keeper circuits are inactive, but inputs may
optionally be pulled up.
LVCMOS2 N/A 2.5 N/A
PCI N/A 3.3 N/A The activation of pull-up resistors prior to configuration is
GTL 0.8 N/A 1.2 controlled on a global basis by the configuration mode pins.
If the pull-up resistors are not activated, all the pins will
GTL+ 1.0 N/A 1.5
float. Consequently, external pull-up or pull-down resistors
HSTL Class I 0.75 1.5 1.5 must be provided on pins required to be at a well-defined
HSTL Class 0.75 1.5 1.5 logic level prior to configuration.
III
All Virtex IOBs support IEEE 1149.1-compatible boundary
HSTL Class 0.75 1.5 1.5
scan testing.
IV
SSTL3 1.5 3.3 1.5
Class I and II
SSTL2 1.125 2.5 1.125
Class I and II
CTT 1.5 3.3 1.5
AGP 1.32 3.3 N/A

R
PS
D Q
T
TCE EC
L Weak
Keeper
SR
PS
PAD
O D Q
OCE EC OBUF
L
SR
PS
IQ Q D Programmable
Delay
EC
L IBUF
Vref
SR
SR
CLK
ICE
iob_c.eps
Figure 2: Virtex Input/Output Block (IOB)
Input Path Each output driver can be individually programmed for a

wide range of low-voltage signalling standards. Each out-
A buffer In the Virtex IOB input path routes the input signal
put buffer can source up to 24 mA and sink up to 48mA.
either directly to internal logic or through an optional input
Drive strength and slew rate controls minimize bus tran-
flip-flop.
sients.
An optional delay element at the D-input of this flip-flop
In most signalling standards, the output High voltage
eliminates pad-to-pad hold time. The delay is matched to
depends on an externally supplied VCCO voltage. The need
the internal clock-distribution delay of the FPGA, and when
to supply VCCO imposes constraints on which standards
used, assures that the pad-to-pad hold time is zero.
can be used in close proximity to each other. See “I/O
Each input buffer can be configured to conform to any of Banking” on page 6.
the low-voltage signalling standards supported. In some of
An optional weak-keeper circuit is connected to each out-
these standards the input buffer utilizes a user-supplied
put. When selected, the circuit monitors the voltage on the
threshold voltage, VREF. The need to supply VREF imposes
pad and weakly drives the pin High or Low to match the
constraints on which standards can used in close proximity
input signal. If the pin is connected to a multiple-source sig-
to each other. See “I/O Banking” on page 6.
nal, the weak keeper holds the signal in its last state if all
There are optional pull-up and pull-down resistors at each drivers are disabled. Maintaining a valid logic level in this
input for use after configuration. Their value is in the range way eliminates bus chatter.
50 – 150 kohms.
Because the weak-keeper circuit uses the IOB input buffer
Output Path to monitor the input level, an appropriate VREF voltage must
be provided if the signalling standard requires one. The
The output path includes a 3-state output buffer that drives
provision of this voltage must comply with the I/O banking
the output signal onto the pad. The output signal can be
rules.
routed to the buffer directly from the internal logic or
through an optional IOB output flip-flop. I/O Banking
The 3-state control of the output can also be routed directly Some of the I/O standards described above require VCCO
from the internal logic or through a flip-flip that provides and/or VREF voltages. These voltages externally and con-
synchronous enable and disable. nected to device pins that serve groups of IOBs, called

R
banks. Consequently, restrictions exist about which I/O more I/O pins convert to VREF pins. Since these are always
standards can be combined within a given bank. a superset of the VREF pins used for smaller devices, it is
possible to design a PCB that permits migration to a larger
Eight I/O banks result from separating each edge of the
device if necessary. All the VREF pins for the largest device
FPGA into two banks, as shown in Figure 3. Each bank has
anticipated must be connected to the VREF voltage, and not
multiple VCCO pins, all of which must be connected to the
same voltage. This voltage is determined by the output used for I/O.
standards in use. In smaller devices, some VCCO pins used in larger devices
do not connect within the package. These unconnected
Within a bank, output standards may be mixed only if they
pins may be left unconnected externally, or may be con-
use the same VCCO. Compatible standards are shown in
nected to the VCCO voltage to permit migration to a larger
Table 4. GTL and GTL+ appear under all voltages because
device if necessary.
their open-drain outputs do not depend on VCCO.
In TQ144 and PQ/HQ240 packages, all VCCO pins are
Table 4: Compatible Output Standards
bonded together internally, and consequently the same
VCCO Compatible Standards VCCO voltage must be connected to all of them. The VREF
3.3 V PCI, LVTTL, SSTL3 I, SSTL3 II, CTT, AGP, pins remain internally connected as eight banks, and may
3
GTL, GTL+ be used as described previously.
2.5 V SSTL2 I, SSTL2 II, LVCMOS2, GTL, GTL+
Configurable Logic Block
1.5 V HSTL I, HSTL III, HSTL IV, GTL, GTL+
The basic building block of the Virtex CLB is the logic cell
Some input standards require a user-supplied threshold (LC). An LC includes a 4-input function generator, carry
voltage, VREF. In this case, certain user-I/O pins are auto- logic, and a storage element. The output from the function
matically configured as inputs for the VREF voltage. Approx- generator in each LC drives both the CLB output and the D
imately one in six of the I/O pins in the bank assume this input of the flip-flop. Each Virtex CLB contains four LCs,
role. organized in two similar slices, as shown in Figure 4.
The VREF pins within a bank are interconnected internally Figure 5 shows a more detailed view of a single slice.
and consequently only one VREF voltage can be used In addition to the four basic LCs, the Virtex CLB contains
within each bank. All VREF pins in the bank, however, must logic that combines function generators to provide func-
be connected to the external voltage source for correct tions of five or six inputs. Consequently, when estimating
operation. the number of system gates provided by a given device,
Within a bank, inputs that require VREF can be mixed with each CLB counts as 4.5 LCs.
those that do not. However, only one VREF voltage may be Look-Up Tables
used within a bank.Input buffers that use VREF are not
5V-tolerant. Virtex function generators are implemented as 4-input
look-up tables (LUTs). In addition to operating as a function
The VCCO and VREF pins for each bank appear in the generator, each LUT can provide a 16 x 1-bit synchronous
device pin-out tables and diagrams. The diagrams also
RAM. Furthermore, the two LUTs within a slice can be
show the bank affiliation of each I/O.
combined to create a 16 x 2-bit or 32 x 1-bit synchronous
Within a given package, the number of VREF and VCCO pins RAM, or a 16x1-bit dual-port synchronous RAM.
can vary depending on the size of device. In larger devices,
The Virtex LUT can also provide a 16-bit shift register that
is ideal for capturing high-speed or burst-mode data. This
mode can also be used to store data in applications such
Bank 0 Bank 1
as Digital Signal Processing.
GCLK3 GCLK2
Bank 7
Bank 2
Storage Elements
Virtex The storage elements in the Virtex slice can be configured
Device either as edge-triggered D-type flip-flops or as level-sensi-
tive latches. The D inputs can be driven either by the func-
Bank 6
Bank 3
tion generators within the slice or directly from slice inputs,

GCLK1 GCLK0 bypassing the function generators.
Bank 5 Bank 4
In addition to Clock and Clock Enable signals, each Slice
has synchronous set and reset signals (SR and BY). SR
X8778_b forces a storage element into the initialization state speci-
Figure 3: Virtex I/O Banks fied for it in the configuration. BY forces it into the opposite

R
COUT COUT
YB YB
Y Y
G4 G4
G3 SP G3 SP
LUT Carry & D Q YQ LUT Carry & D Q YQ
G2 Control G2 Control
EC EC
G1 G1
RC RC
BY BY
XB XB
X X
F4 F4
F3 SP F3 SP
LUT Carry & LUT Carry & D Q XQ
D Q XQ F2
F2 Control Control
EC EC
F1 F1
RC RC
BX BX
Slice 1 Slice 0
slice_b.eps
CIN CIN
Figure 4: 2-slice Virtex CLB
state. Alternatively, these signals may be configured to The dedicated carry path can also be used to cascade
operate asynchronously. function generators for implementing wide logic functions.
All of the control signals are independently invertible, and BUFTs
are shared by the two flip-flops within the slice.
Each Virtex CLB contains two 3-state drivers (BUFTs) that
Additional Logic can drive on-chip busses. See “Dedicated Routing” on
The F5 multiplexer in each slice combines the function gen- page 11. Each Virtex BUFT has an independent 3-state
erator outputs. This combination provides either a function control pin and an independent input pin.
generator that can implement any 5-input function, a 4:1 Block RAM
multiplexer, or selected functions of up to nine inputs.
Virtex FPGAs incorporate several large BlockSelectRAM+
Similarly, the F6 multiplexer combines the outputs of all four memories. These complement the distributed SelectRAM+
function generators in the CLB by selecting one of the LUTRAMs that provide shallow RAM structures imple-
F5-multiplexer outputs. This permits the implementation of mented in CLBs.
any 6-input function, an 8:1 multiplexer, or selected func-
tions of up to 19 inputs. BlockSelectRAM+ memory blocks are organized in col-
umns. All Virtex devices contain two such columns, one
Each CLB has four direct feedthrough paths, one per LC. along each vertical edge. These columns extend the full
These paths provide extra data input lines or additional height of the chip. Each memory block is four CLBs high,
local routing that does not consume logic resources. and consequently, a Virtex device 64 CLBs high will
Arithmetic Logic contain16 memory blocks per column, and a total of 32
blocks.
Dedicated carry logic provides fast arithmetic carry capabil-
ity for high-speed arithmetic functions. The Virtex CLB sup-
ports two separate carry chains, one per Slice. The height
of the carry chains is two bits per CLB.
The arithmetic logic includes an XOR gate that allows a
1-bit full adder to be implemented within an LC. In addition,
a dedicated AND gate improves the efficiency of multiplier
implementation.

R
COUT
YB
CY
G4 I3 Y
G3 I2 O
G2 I1
LUT INIT
G1 I0 D Q YQ
WE DI
0 EC
1 REV
BY
XB
F5IN
F6
CY
F5 F5
CK WSO BY DG
WE X
A4 WSH BX DI
INIT
D Q XQ
3
BX EC
I3 WE DI
F4
F3 I2 O REV
F2 I1 LUT
F1 I0
1
SR
CLK
CE
CIN
viewslc4.eps
Figure 5: Detailed View of Virtex Slice
Table 5 shows the amount of Block SelectRAM+ memory two ports can be configured independently, providing
that is available in each Virtex device. built-in bus-width conversion.
Table 5: Virtex Block SelectRAM+ Amounts .
Total Block RAMB4_S#_S#

Virtex Device # of Blocks
SelectRAM+ Bits
WEA
XCV50 8 32,768 ENA
RSTA DOA[#:0]
XCV100 10 40,960 CLKA
XCV150 12 49,152 ADD[<#:0]
DIA[#:0]
XCV200 14 57,344
XCV300 16 65,536
WEB
XCV400 20 81,920 ENB
RSTB DOB[#:0]
XCV600 24 98,304 CLKB
XCV800 28 114,688 ADDRB[#:0]
DIB[#:0]
XCV1000 32 131,072
ramb_a.eps
Each Block SelectRAM+ cell, as illustrated in Figure 6, is a
fully synchronous dual-ported 4096-bit RAM with indepen-
dent control signals for each port. The data widths of the Figure 6: Dual-Port Block SelectRam+

R
Table 6 shows the depth and width aspect ratios for the General Purpose Routing
Block SelectRAM+
Most Virtex signals are routed on the general purpose rout-
Table 6: Block SelectRAM+ Port Aspect Ratios ing, and consequently, the majority of interconnect
resources are associated with this level of the routing hier-
Width Depth ADDR Bus Data Bus
archy. The general routing resources are located in hori-
1 4096 ADDR<11:0> DATA<0> zontal and vertical routing channels associated with the
2 2048 ADDR<10:0> DATA<1:0> rows and columns CLBs. The general-purpose routing
4 1024 ADDR<9:0> DATA<3:0> resources are listed below.
8 512 ADDR<8:0> DATA<7:0> • Adjacent to each CLB is a General Routing Matrix
16 256 ADDR<7:0> DATA<15:0> (GRM). The GRM is the switch matrix through which
horizontal and vertical routing resources connect, and
The Virtex block RAM also includes dedicated routing to
is also the means by which the CLB gains access to the
provide an efficient interface with both CLBs and other
general purpose routing.
block RAMs.
• 24 single-length lines route GRM signals to adjacent
GRMs in each of the four directions.
Programmable Routing Matrix
• 96 buffered Hex lines route GRM signals to another
It is the longest delay path that limits the speed of any GRMs six-blocks away in each one of the four
worst-case design. Consequently, the Virtex routing archi- directions. Organized in a staggered pattern, Hex lines
tecture and its place-and-route software were defined in a may be driven only at their endpoints. Hex-line signals
single optimization process. This joint optimization mini- can be accessed either at the endpoints or at the
mizes long-path delays, and consequently, yields the best midpoint (three blocks from the source). One third of the
system performance. Hex lines are bidirectional, while the remaining ones
The joint optimization also reduces design compilation are uni-directional.
times because the architecture is software-friendly. Design • 12 Longlines are buffered, bidirectional wires that
cycles are correspondingly reduced due to shorter design distribute signals across the device quickly and
iteration times. efficiently. Vertical Longlines span the full height of the
device, and horizontal ones span the full width of the
Local Routing device.
The VersaBlock provides local routing resources, as shown I/O Routing
in Figure 7, providing the following three types of connec-
tions. Virtex devices have additional routing resources around
their periphery that form an interface between the CLB
• Interconnections among the LUTs, flip-flops, and GRM array and the IOBs. This additional routing, called the Ver-
• Internal CLB feedback paths that provide high-speed saRing, facilitates pin-swapping and pin-locking, such that
connections to LUTs within the same CLB, chaining logic redesigns can adapt to existing PCB layouts.
them together with minimal routing delay Time-to-market is reduced, since PCBs and other system
• Direct paths that provide high-speed connections components can be manufactured while the logic design is
between horizontally adjacent CLBs, eliminating the still in progress.
delay of the GRM.

R
To Adjacent
GRM
To Adjacent To Adjacent
GRM GRM GRM
3
To Adjacent
GRM
Direct Connection Direct Connection

To Adjacent CLB To Adjacent
CLB CLB
X8794b
Figure 7: Virtex Local Routing
Dedicated Routing Global Routing

Some classes of signal require dedicated routing resources Global Routing resources distribute clocks and other sig-
to maximize performance. In the Virtex architecture, dedi- nals with very high fanout throughout the device. Virtex
cated routing resources are provided for two classes of sig- devices include two tiers of global routing resources
nal. referred to as primary and secondary global routing
resources.
• Horizontal routing resources are provided for on-chip
3-state busses. Four partitionable bus lines are • The primary global routing resources are four dedicated
provided per CLB row, permitting multiple busses within global nets with dedicated input pins that are designed
a row, as shown in Figure 8. to distribute high-fanout clock signals with minimal
• Two dedicated nets per CLB propagate carry signals skew. Each global clock net can drive all CLB, IOB, and
vertically to the adjacent CLB.
Tri-State
Lines
CLB CLB CLB CLB
buft_c.eps
Figure 8: BUFT Connections to Dedicated Horizontal Bus Lines

R
block RAM clock pins. The primary global nets may only between the clock input pad and internal clock-input pins
be driven by global buffers. There are four global throughout the device. Each DLL can drive two global clock
buffers, one for each global net. networks.The DLL monitors the input clock and the distrib-
• The secondary global routing resources consist of 24 uted clock, and automatically adjusts a clock delay ele-
backbone lines, 12 across the top of the chip and 12 ment. Additional delay is introduced such that clock edges
across bottom. From these lines, up to 12 unique reach internal flip-flops exactly one clock period after they
signals per column can be distributed via the 12 arrive at the input. This closed-loop system effectively elim-
longlines in the column. These secondary resources inates clock-distribution delay by ensuring that clock edges
are more flexible than the primary resources since they arrive at internal flip-flops in synchronism with clock edges
are not restricted to routing only to clock pins. arriving at the input.
In addition to eliminating clock-distribution delay, the DLL
Clock Distribution
provides advanced control of multiple clock domains. The
Virtex provides high-speed, low-skew clock distribution DLL provides four quadrature phases of the source clock,
through the primary global routing resources described can double the clock, or divide the clock by 1.5, 2, 2.5, 3, 4,
above. A typical clock distribution net is shown in Figure 9. 5, 8, or 16. It has six outputs.
Four global buffers are provided, two at the top center of the The DLL also operates as a clock mirror. By driving the out-
device and two at the bottom center. These drive the four put from a DLL off-chip and then back on again, the DLL
primary global nets that in turn drive any clock pin. can be used to deskew a board level clock among multiple
Four dedicated clock pads are provided, one adjacent to Virtex devices.
each of the global buffers. The input to the global buffer is In order to guarantee that the system clock is operating cor-
selected either from these pads or from signals in the gen- rectly prior to the FPGA starting up after configuration, the
eral purpose routing. DLL can delay the completion of the configuration process
until after it has achieved lock.
Delay-Locked Loop (DLL)
Associated with each global clock input buffer is a fully dig-
ital Delay-Locked Loop (DLL) that can eliminate skew
GCLKPAD3 GCLKPAD2
Global Clock Rows GCLKBUF3 GCLKBUF2 Global Clock Column
Global Clock Spine
GCLKBUF1 GCLKBUF0
GCLKPAD1 GCLKPAD0
gclkbu_2.eps
Figure 9: Global Clock Distribution Network

R
Boundary Scan
Table 7: Boundary-Scan Instructions
Virtex devices support all the mandatory boundary-scan
instructions specified in the IEEE standard 1149.1. A Test Boundary-Scan Binary
Description
Access Port (TAP) and registers are provided that imple- Command Code(4:0)
ment the EXTEST, SAMPLE/PRELOAD, and BYPASS EXTEST 00000 Enables boundary-scan
instructions. The TAP also supports two USERCODE EXTEST operation
instructions and internal scan chains. SAMPLE 00001 Enables boundary-scan
The TAP uses dedicated package pins that always operate SAMPLE operation
using LVTTL. For TDO to operate using LVTTL, the VCCO USR1 00010 Access user-defined regis-
for Bank 2 must be 3.3 V. Otherwise, TDO switches ter 1
rail-to-rail between ground and VCCO. USR2 00011 Access user-defined reg-
Boundary-scan operation is independent of individual IOB ister 2
configurations, and unaffected by package type. All IOBs, CFG_OUT 00100 Access the configuration
including unbonded ones, are treated as independent bus for Readback
3-state bidirectional pins in a single scan chain. Retention CFG_IN 00101 Access the configuration 3
of the bidirectional test capability after configuration facili- bus for Configuration
tates the testing of external interconnections. INTEST 00111 Enables boundary-scan
Table 7 lists the boundary-scan instructions supported in INTEST operation
Virtex FPGAs. Internal signals can be captured during USRCODE 01000 Enables shifting out USER
EXTEST by connecting them to unbonded or unused IOBs. code
They may also be connected to the unused outputs of IOBs IDCODE 01001 Enables shifting out of ID
defined as unidirectional input pins. This technique partially Code
compensates for the absence of INTEST support. HIZ 01010 Tri-states output pins while
The public boundary-scan instructions are available prior to enabling the Bypass Reg-
configuration. After configuration, the public instructions ister
remain available together with any USERCODE instruc- BUS_RST 01011 Reset the Configuration
tions installed during the configuration. While the SAMPLE Bus
and BYPASS instructions are available during configura- JSTART 01100 Clock the start-up se-
tion, it is recommended that boundary-scan operations not quence when StartupClk is
be performed during this transitional period. TCK
In addition to the test instructions outlined above, the BYPASS 11111 Enables BYPASS
boundary-scan circuitry can be used to configure the RESERVED All other Xilinx reserved instructions
FPGA, and also to read back the configuration data. codes
To facilitate internal scan chains, the User Register pro- Application programs ranging from schematic capture to
vides three outputs (Reset, Update, and Shift) that repre- Placement and Routing (PAR) can be accessed through
sent the corresponding states in the boundary-scan the XDM software. The program command sequence is
internal state machine. generated prior to execution, and stored for documentation.
Several advanced software features facilitate Virtex design.
Development System RPMs, for example, are schematic-based macros with rela-
Virtex FPGAs are supported by the Xilinx Foundation and tive location constraints to guide their placement. They help
Alliance CAE tools. The basic methodology for Virtex ensure optimal implementation of common functions.
design consists of three interrelated steps: design entry, For HDL design entry, the Xilinx FPGA Foundation devel-
implementation, and verification. Industry-standard tools opment system provides interfaces to the following synthe-
are used for design entry and simulation (for example, Syn- sis design environments.
opsys FPGA Express), while Xilinx provides proprietary
architecture-specific tools for implementation. • Synopsys (FPGA Compiler, FPGA Express)
• Exemplar (Spectrum)
The Xilinx development system is integrated under the Xil- • Synplicity (Synplify)
inx Design Manager (XDM™) software, providing design-
ers with a common user interface regardless of their choice For schematic design entry, the Xilinx FPGA Foundation
of entry and verification tools. The XDM software simplifies and alliance development system provides interfaces to the
the selection of implementation options with pull-down following schematic-capture design environments.
menus and on-line help. • Mentor Graphics V8 (Design Architect, QuickSim II)

R
• Viewlogic Systems (Viewdraw) these user-specified requirements and accommodate

them.
Third-party vendors support many other environments.
A standard interface-file specification, Electronic Design Timing requirements are entered on a schematic in a form
directly relating to the system requirements, such as the
Interchange Format (EDIF), simplifies file transfers into and
targeted clock frequency, or the maximum allowable delay
out of the development system.
between two registers. In this way, the overall performance
Virtex FPGAs supported by a unified library of standard of the system along entire signal paths is automatically tai-
functions. This library contains over 400 primitives and lored to user-generated specifications. Specific timing infor-
macros, ranging from 2-input AND gates to 16-bit accumu- mation for individual nets is unnecessary.
lators, and includes arithmetic functions, comparators,
counters, data registers, decoders, encoders, I/O functions, Design Verification
latches, Boolean functions, multiplexers, shift registers, and
In addition to conventional software simulation, FPGA
barrel shifters.
users can use in-circuit debugging techniques. Because
The “soft macro” portion of the library contains detailed Xilinx devices are infinitely reprogrammable, designs can
descriptions of common logic functions, but does not con- be verified in real time without the need for extensive sets of
tain any partitioning or placement information. The perfor- software simulation vectors.
mance of these macros depends, therefore, on the
The development system supports both software simula-
partitioning and placement obtained during implementa-
tion and in-circuit debugging techniques. For simulation,
tion.
the system extracts the post-layout timing information from
RPMs, on the other hand, do contain predetermined parti- the design database, and back-annotates this information
tioning and placement information that permits optimal into the netlist for use by the simulator. Alternatively, the
implementation of these functions. Users can create their user can verify timing-critical portions of the design using
own library of soft macros or RPMs based on the macros the TRACE® static timing analyzer.
and primitives in the standard library.
For in-circuit debugging, the development system includes
The design environment supports hierarchical design entry, a download and readback cable. This cable connects the
with high-level schematics that comprise major functional FPGA in the target system to a PC or workstation. After
blocks, while lower-level schematics define the logic in downloading the design into the FPGA, the designer can
these blocks. These hierarchical design elements are auto- single-step the logic, readback the contents of the flip-flops,
matically combined by the implementation tools. Different and so observe the internal logic state. Simple modifica-
design entry tools can be combined within a hierarchical tions can be downloaded into the system in a matter of min-
design, thus allowing the most convenient entry method to utes.
be used for each portion of the design.
Configuration
Design Implementation
Virtex devices are configured by loading configuration data
The place-and-route tools (PAR) automatically provide the into the internal configuration memory. Some of the pins
implementation flow described in this section. The parti- used for this are dedicated configuration pins, while others
tioner takes the EDIF netlist for the design and maps the may be re-used as general purpose inputs and outputs
logic into the architectural resources of the FPGA (CLBs once configuration is complete.
and IOBs, for example). The placer then determines the
best locations for these blocks based on their interconnec- The dedicated pins are the mode pins (M2, M1, M0), the
tions and the desired performance. Finally, the router inter- configuration clock pin (CCLK), the PROGRAM pin, the
connects the blocks. DONE pin and the boundary-scan pins (TDI, TDO, TMS,
TCK). Depending on the configuration mode chosen,
The PAR algorithms support fully automatic implementa- CCLK may be an output generated by the FPGA, or may be
tion of most designs. For demanding applications, however, generated externally, and provided to the FPGA as an
the user can exercise various degrees of control over the input.
process. User partitioning, placement, and routing informa-
tion is optionally specified during the design-entry process. Note that some configuration pins that may act as outputs.
The implementation of highly structured designs can bene- For correct operation, these pins may require a VCCO of
3.3 V to permit LVTTL operation. All the pins affected fall in
fit greatly from basic floorplanning.
banks 2 or 3.
The implementation software incorporates Timing Wizard®
timing-driven placement and routing. Designers specify For a more detailed description than that given below, see
timing requirements along entire paths during design entry. the Supplementary Description on Configuration and
The timing path analysis routines in PAR then recognize Readback.

R
Configuration Modes having the IOB pins either pulled up or left floating prior to
configuration. The selection codes are listed in Table 8.
Virtex supports the following four configuration modes.
Configuration through the boundary-scan port is always
• Slave-serial mode
available, independent of the mode selection. Selecting the
• Master-serial mode boundary-scan mode simply turns off the other modes. The
• SelectMAP mode three mode pins have internal pull-up resistors, and default
• Boundary-scan mode
to a logic High if left unconnected.
The Configuration mode pins (M2, M1, M0) select among
these configuration modes with the option in each case of
Table 8: Configuration Codes
CCLK Data Serial Pre-configuration

Configuration Mode M2 M1 M0
Direction Width Dout Pull-ups
Master-serial mode 0 0 0 Out 1 Yes No
Boundary-scan mode 1 0 1 N/A 1 No No
3
SelectMAP mode 1 1 0 In 8 No No
Slave-serial mode 1 1 1 In 1 Yes No
Master-serial mode 1 0 0 Out 1 Yes Yes
Boundary-scan mode 0 0 1 N/A 1 No Yes
SelectMAP mode 0 1 0 In 8 No Yes
Slave-serial mode 0 1 1 In 1 Yes Yes
Slave Serial Mode configuration chains. This change was made to improve
serial-configuration rates for Virtex only chains.
In slave serial mode, the FPGA receives configuration data
in bit-serial form from a serial PROM or other source of Figure 10 shows a full master/slave system. A Virtex device
serial configuration data. The serial bitstream must be in slave serial mode should be connected as shown in the
setup at the DIN input pin a short time before each rising third device from the left
edge of an externally generated CCLK. Slave-serial mode is selected by applying <111> to the
Multiple FPGAs can be daisy-chained for configuration mode pins (M2, M1, M0). A weak pull-up on the mode pins
from a single source. After a particular FPGA has been makes slave serial the default mode if the pins are left
configured, the data for the next device is routed to the unconnected. Figure 11 shows slave-serial configuration
DOUT pin. The data on the DOUT pin changes on the ris- timing.
ing edge of CCLK. Table 9 provides more detail about the characteristics
The capture of DIN on the rising edge of CCLK differs from shown in Figure 11. Configuration must be delayed until the
previous families, but will not cause a problem for mixed INIT pins of all daisy-chained FPGAs are High.
Table 9: Serial Mode Programming Switching
Description Symbol Units

DIN setup/hold, slave mode 1/2 TDCC/TCCD 5.0/0.0 ns, min
DIN setup/hold, master mode 1/2 TDSCK/TSCKD 5.0/0.0 ns, min
DOUT 3 TCCO 12.0 ns, max
High time 4 TCCH 5.0 ns, min
CCLK
Low time 5 TCCL 5.0 ns, min
Maximum Frequency FCC 66 MHz, max
Frequency Tolerance, master mode with +45%
respect to nominal -30%

R
N/C
3.3V
M0 M1 4.7 K M0 M1
M2 N/C M2
DOUT DIN DOUT
CCLK
VIRTEX
MASTER XC1701L VIRTEX,
SERIAL XC4000XL,
CCLK CLK SLAVE
DIN DATA
PROGRAM CE CEO PROGRAM

DONE INIT RESET/OE DONE INIT
(Low Reset Option Used)
PROGRAM X9025_d
Figure 10: Master/Slave Serial Mode Circuit Diagram
DIN
1 TDCC 2 TCCD 5 TCCL
CCLK
4 TCCH
3 TCCO
DOUT
(Output)
X5379_a
Figure 11: Slave Serial Mode Programming Switching Characteristics

R
Master Serial Mode remaining devices operate in slave-serial mode. The

SPROM RESET pin is driven by INIT, and CE input is
In master serial mode, the CCLK output of the FPGA drives
driven by DONE. There is, therefore, the potential for con-
a Xilinx Serial PROM that feeds bit-serial data to the DIN
tention on the DONE pin, depending on the start-up
input. The FPGA accepts this data on each rising CCLK
sequence options chosen.
edge. After the FPGA has been loaded, the data for the
next device in a daisy-chain is presented on the DOUT pin The sequence of operations necessary to configure a Vir-
after the rising CCLK edge. The preamble is also forwarded tex FPGA serially appears in Figure 12.
to other devices in the daisy-chain. Figure 13 shows the timing of master-serial configuration.
The interface is identical to slave serial except that an inter- Master serial mode is selected by a <000> on the mode
nal oscillator is used to generate the configuration clock pins (M2, M1, M0). Table 9 shows the timing information for
(CCLK). A wide range of frequencies can be selected for Figure 13
CCLK which always starts at a slow default frequency. Con- At power-up, Vcc must rise from 2.0 V to Vcc min in less
figuration bits then switch CCLK to a higher frequency for than 25 ms, otherwise delay configuration by pulling PRO-
the remainder of the configuration. Switching to a lower fre- GRAM Low until Vcc is valid.
quency is prohibited. 3
The CCLK frequency is set using the ConfigRate option in
SelectMAP Mode
the bitstream generation software. The maximum CCLK The SelectMAP mode is the fastest configuration option.
frequency that can be selected is 60 MHz. When selecting Byte-serial data is written into the FPGA with a BUSY flag
a CCLK frequency, ensure that the serial PROM and any controlling the flow of data.
daisy-chained FPGAs are fast enough to support the clock An external data source provides a byte stream, CCLK, a
rate. Chip Select (CS) signal and a Write signal (WRITE). If
On power-up, the CCLK frequency is 2.5 MHz. This fre- BUSY is asserted (High) by the FPGA, the data must be
quency is used until the ConfigRate bits have been loaded held until BUSY goes Low.
when the frequency changes to the selected ConfigRate. Data can also be read using the SelectMAP mode. If
Unless a different frequency is specified in the design, the WRITE is not asserted, configuration data is read out of the
default ConfigRate is 2.5 MHz.
FPGA as part of a readback operation.
Figure 10 shows a full master/slave system. In this system,
the leftmost device operates in master-serial mode. The

R
Apply Power
FPGA starts to clear
configuration memory.
If used to delay
Set PROGRAM = High configuration
FPGA makes a final
clearing pass and releases
If used to delay
INIT when finished. Release INIT
configuration
Low
INIT?
High
Load a Configuration Bit
Once a Frame,
FPGA checks data using CRC
and pulls INIT Low on error. No
End of
Bitstream?
If no CRC errors found, Yes

FPGA enters start-up phase
causing DONE to go High. Configuration Completed
X8793_a
Figure 12: Serial Configuration Flowchart
.
CCLK
(Output)
2 TCKDS
1 TDSCK
Serial Data In
Serial DOUT
(Output)
X3223_a
Figure 13: Master Serial Mode Programming Switching Characteristics

R
After configuration, the pins of the SelectMAP port can be Multiple Virtex FPGAs can be configured using the Select-
used as additional user I/O. Alternatively, the port may be MAP mode, and be made to start-up simultaneously. To
retained to permit high-speed 8-bit readback. configure multiple devices in this way, wire the individual
CCLK, Data, WRITE, and BUSY pins of all the devices in
Retention of the SelectMAP port is selectable on a
parallel. The individual devices are loaded separately by
design-by-design basis when the bitstream is generated. If
asserting the CS pin of each device in turn and writing the
retention is selected, PROHIBIT constraints are required to
appropriate data.
prevent the SelectMAP-port pins from being used as user
I/O. Note: See Table 10 for SelectMAP Write Timing
Characteristics.
Table 10: SelectMAP Write Timing Characteristics
Description Symbol Units

D0-7 Setup/Hold 1/2 TSMDCC/TSMCCD 5.0/0.0 ns, min
CS Setup/Hold 3/4 TSMCSCC/TSMCCCS 7.0/0.0 ns, min
WRITE Setup/Hold 5/6 TSMCCW/TSMWCC 7.0/0.0 ns, min
CCLK
3
BUSY Propagation Delay 7 TSMCKBY 12.0 ns, max
Maximum Frequency FCC 66 MHz, max
Maximum Frequency with FCCNH 50 MHz, max
no handshake
Write can be split into many such sequences. The packet does
not have to complete within one assertion of CS, illustrated
Write operations send packets of configuration data into the
in Figure 14.
FPGA. The sequence of operations for a multi-cycle write
operation is shown below. Note that a configuration packet
CCLK
CS 3 4
WRITE 5 6
1 2
DATA[7:0]
7
BUSY
No Write Write No Write Write

X8796_b
Figure 14: Write Operations
1. Assert WRITE and CS Low. Note that when CS is 3. At the rising edge of CCLK: If BUSY is Low, the data is
asserted on successive CCLKs, WRITE must remain accepted on this clock. If BUSY is High from a previous
either asserted or deasserted. Otherwise an abort will write, the data is not be accepted. Acceptance will
be initiated, as described below. instead occur on the first clock after BUSY goes Low,
and the data must be held until this has happened.
2. Drive data onto D[7:0]. Note that to avoid contention, the
data source should not be enabled while CS is Low and 4. Repeat steps 2 and 3 until all the data has been sent.
WRITE is High. Similarly, while WRITE is High, no more
5. Deassert CS and WRITE.
that one CS should be asserted.

R
A flowchart for the write operation appears in Figure 15. essary, and data can simply be entered into the FPGA
Note that if CCLK is slower than fCCNH, the FPGA will never every CCLK cycle.
assert BUSY, In this case, the above handshake is unnec-
Apply Power
FPGA starts to clear
configuration memory.
If used to delay
Set PROGRAM = High configuration
FPGA makes a final
clearing pass and releases
If used to delay
INIT when finished. Release INIT configuration
Low
INIT?
High
Set WRITE = Low
Enter Data Source

Sequence A
Set CS = Low On first FPGA
Apply Configuration Byte
Once a Frame,
FPGA checks data using CRC
and pulls INIT Low on error. High
Busy?
Low
No
End of Data?
If no errors, Yes
first FPGAs enter start-up phase
releasing DONE.
Set CS = High On first FPGA
If no errors, Repeat Sequence A For any other FPGAs

later FPGAs enter start-up phase
releasing DONE.
Disable Data Source
Set WRITE = High

When all DONE pins
are released, DONE goes High
and start-up sequences complete. Configuration Completed
X8795_a
Figure 15: SelectMAP Flowchart for Write Operation
Abort aries, and the FPGA requires a new synchronization word

prior to accepting any new packets.
During a given assertion of CS, the user cannot switch from
a write to a read, or vice-versa. This action causes the cur- To initiate an abort during a write operation, deassert
rent packet command to be aborted. The device will remain WRITE. At the rising edge of CCLK, an abort is initiated, as
BUSY until the aborted operation has completed. Following shown in Figure 16.
an abort, data is assumed to be unaligned to word bound-

R
CCLK
CS
WRITE
DATA[7:0]
BUSY 3
Abort
X8797_b
Figure 16: SelectMAP Write Abort Waveforms
Boundary-Scan Mode figuration data is loaded into the memory, and finally, the
logic is activated by a start-up process.
In the boundary-scan mode, no non-dedicated pins are
required, configuration being done entirely through the Configuration is automatically initiated on power-up unless
IEEE 1149.1 Test Access Port. it is delayed by the user, as described below. The configu-
ration process may also be initiated by asserting PRO-
Configuration through the TAP uses the special CFG_IN
GRAM. The end of the memory-clearing phase is signalled
instruction. This instruction allows data input on TDI to be
by INIT going High, and the completion of the entire pro-
converted into data packets for the internal configuration
cess is signalled by asserting DONE.
bus.
The power-up timing of configuration signals is shown in
The following steps are required to configure the FPGA
Figure 17. The corresponding timing characteristics are
through the boundary-scan port.
listed in Table 11.
1. Load the CFG_IN instruction into the boundary-scan Vcc TPOR
instruction register (IR)
2. Enter the Shift-DR (SDR) state PROGRAM
TPI
3. Shift a standard configuration bitstream into TDI INIT
4. Return to Run-Test-Idle (RTI) TICCK
CCLK OUTPUT or INPUT

5. Load the JSTART instruction into IR
6. Enter the SDR state M0, M1,
VALID
(Required)
7. Clock TCK for the length of the sequence (the length is 98122302
programmable) Figure 17: Power-up Timing Configuration Signals

8. Return to RTI
As noted above, configuration and readback is always
Table 11: Power-up Timing Characteristics
available. The boundary-scan mode simply locks out the
other modes. The boundary-scan mode is selected by a Description Symbol Units
<101> on the mode pins (M2, M1, M0). Power-on Reset TPOR 2.0 ms, max
Program Latency TPL 100.0 µs, max
Configuration Sequence
CCLK (output) TICCK 0.5 µs, min
The configuration of Virtex devices is a three-phase pro- delay 4.0 µs, max
cess. First, the configuration memory is cleared. Next, con-

R
Delaying Configuration Data Stream Format

Configuration of the FPGA can be delayed by holding the Virtex devices are configured by sequentially loading
PROGRAM pin Low until the system is ready for the device frames of data that have been concatenated into a bit-
to configure. During the memory clearance phase, the con- stream. Table 12 lists the total number of bits required to
figuration sequences continuously cycles through the con- configure each device.
figuration memory clearing all addresses. This activity
continues until the completion of one full address cycle Table 12: Virtex Bit-stream Lengths
after the PROGRAM pin goes High. Thus, configuration is
delayed by extending the memory clearance phase. Device # of Configuration Bits
Alternatively, INIT can be held Low using an open-drain
XCV50 559,232
driver. An open-drain is required since INIT is a bidirec-
tional open-drain pin that is held Low by the FPGA while the XCV100 781,248
configuration memory is being cleared. Extending the time
that the pin is Low causes the configuration sequencer to XCV150 1,041,128
act as if the configuration memory is still being cleared.
Thus, configuration is delayed by preventing entry into the XCV200 1,335,872
phase where data is loaded.
XCV300 1,751,840
Start-Up Sequence
XCV400 2,546,080
The default Start-up sequence is that one CCLK cycle after
DONE goes High, the global tri-state signal (GTS) is XCV600 3,608,000
released. This permits device outputs to turn on as neces-
sary. XCV800 4,715,684
One CCLK cycle later, the Global Set/Reset (GSR) and XCV1000 6,127,776
Global Write Enable (GWE) signals are released. This per-
mits the internal storage elements to begin changing state
in response to the logic and the user clock. Readback
The relative timing of these events may be changed. In The configuration data stored in the Virtex configuration
addition, the GTS, GSR, and GWE events may be made memory can be readback for verification. Along with the
dependent on the DONE pins of multiple devices all going configuration data it is possible to readback the contents all
High, forcing the devices to start in synchronism. The flip-flops/latches, LUTRAMs, and block RAMs. This capa-
sequence may also be paused at any stage until lock has bility is used for real-time debugging.
been achieved on any or all DLLs. For more detailed information contact the factory for a copy
of the “Supplementary Description on Configuration and
Readback”.

R
Pin Definitions
Table 13: Special Purpose Pins
Dedicated
Pin Name Direction Description
Pin
GCK0, GCK1, Yes Input Clock input pins that connect to Global Clock Buffers. These
GCK2, GCK3 pins become user inputs when not needed for clocks.
M0, M1, M2 Yes Input Mode pins are used to specify the configuration mode.
CCLK Yes Input or The configuration Clock I/O pin: it is an input for SelectRAM and
Output slave-serial modes, and output in master-serial mode
PROGRAM Yes Input Initiates a configuration sequence when asserted Low.

3
DONE Yes Bidirectional Indicates that configuration loading is complete, and that the
start-up sequence is in progress. The output may be open drain.
INIT No Bidirectional When Low, indicates that the configuration memory is being
(Open-drain) cleared. The pin becomes a user I/O after configuration.
BUSY/ No Output In SelectMAP mode, BUSY controls the rate at which configura-
DOUT tion data is loaded. The pin becomes a user I/O after configura-
tion unless the SelectMAP port is retained.
In bit-serial modes, DOUT provides preamble and configuration
data to downstream devices in a daisy-chain. The pin becomes
a user I/O after configuration.
D0/DIN, No Input In SelectMAP mode, D0-7 are configuration data input pins.
D1, D2, These pins become user I/Os after configuration unless the Se-
D3, D4, lectMAP port is retained.
D5, D6, In bit-serial modes, DIN is the single data input. This pin be-
D7 comes a user I/O after configuration.
WRITE No Input In SelectMAP mode, the active-low Write Enable signal. The pin
becomes a user I/O after configuration unless the SelectMAP
port is retained.
CS No Input In SelectMAP mode, the active-low Chip Select signal. The pin
becomes a user I/O after configuration unless the SelectMAP
port is retained.
TDI, TDO, Yes Mixed Boundary-scan Test-Access-Port pins, as defined in IEEE

TMS, TCK 1149.1.
DXN, DXP Yes N/A Temperature-sensing diode pins. (Anode: DXP, Cathode: DXN)
VCCINT Yes Input Power-supply pins for the internal core logic.
VCCO Yes Input Power-supply pins for the output drivers (subject to banking
rules)
VREF No Input Input threshold voltage pins. Become user I/Os when an exter-
nal threshold voltage is not needed (subject to banking rules).
GND Yes Input Ground

R
Virtex DC Characteristics
Definition of Terms
Data sheets may be designated as Advance or Preliminary. The status of specifications in these data sheets is as follows:
Advance: Initial estimates based on simulation and/or extrapolation from other speed grades, devices, or families.
Values are subject to change. Use as estimates, not for production.
Preliminary: Based on preliminary characterization. Further changes are not expected.
Unmarked: Data sheets not identified as either Advance or Preliminary are to be considered final.
All specifications are representative of worst-case supply voltage and junction temperature conditions. The parameters
included are common to popular designs and typical applications. Contact the factory for design considerations requiring
more detailed information.
All specifications are subject to change without notice.
Virtex Absolute Maximum Ratings

Symbol Description Units
VCCINT Supply voltage relative to GND -0.5 to 3.0 V
VCCO Supply voltage relative to GND -0.5 to 4.0 V
VREF Input Reference Voltage -0.5 to 3.6 V
Input voltage relative to GND Using VREF -0.5 to 3.6 V
VIN
Internal threshold -0.5 to 5.5 V
VTS Voltage applied to 3-state output -0.5 to 5.5 V
VCC Longest Supply Voltage Rise Time from 1 V to 3V 50 ms
TSTG Storage temperature (ambient) -65 to +150 °C
TSOL Maximum soldering temperature (10 s @ 1/16 in. = 1.5 mm) +260 °C
Junction temperature Ceramic packages +125 °C
TJ
Plastic packages +125 °C
Notes: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress
ratings only, and functional operation of the device at these or any other conditions beyond those listed under Operating
Conditions is not implied. Exposure to Absolute Maximum Ratings conditions for extended periods of time may affect device
reliability.
Power supplies may turn on in any order.
For protracted periods (e.g., longer than a day), VIN should not exceed VCCO by more that 3.6 V.
Virtex Recommended Operating Conditions

Symbol Description Min Max Units
Supply voltage relative to GND, TJ = 0 °C to +85°C Commercial 2.5 - 5% 2.5 + 5% V
VCCINT
Supply voltage relative to GND, TJ = -40°C to +100°C Industrial 2.5 - 5% 2.5 + 5% V
Supply voltage relative to GND, TJ = 0 °C to +85°C Commercial 3.6 V
VCCO
Supply voltage relative to GND, TJ = -40°C to +100°C Industrial 3.6 V
TIN Input signal transition time 250 ns
Notes: Correct operation is guaranteed with a minimum VCCINT of 2.25 V (Nominal VCCINT -10%). Below the minimum value stated
above, all delay parameters increase by 3% for each 50-mV reduction in VCCINT below the specified range.
At junction temperatures above those listed as Operating Conditions, all delay parameters increase by 0.35% per °C.
Input and output measurement threshold is ~50% of VCC.

R
Virtex DC Characteristics Over Recommended Operating Conditions

VDRINT Data Retention VCCINT Voltage (below which configuration data may be lost) 2.0 V
VDRIO Data Retention VCCO Voltage (below which configuration data may be lost) 1.2 V
ICCINTQ Quiescent VCCINT supply current (Note 1) mA
ICCOQ Quiescent VCCO supply current (Note 1) mA
IREF VREF current per VREF pin 20 µA
IL Input or output leakage current -10 +10 µA
CIN Input capacitance (sample tested) BGA, PQ, HQ, packages 8 pF
IRPU Pad pull-up (when selected) @ Vin = 0 V, VCCO = 3.3 V (sample tested) Note 2 0.25 mA
IRPD Pad pull-down (when selected) @ Vin = 3.6 V (sample tested) Note 2 0.15 mA
Note 1: With no output current loads, no active input pull-up resistors, all I/O pins Tri-stated and floating. 3
Note 2: Internal pull-up and pull-down resistors guarantee valid logic levels at unconnected input pins. These pull-up and pull-down
resistors do not guarantee valid logic levels when input pins are connected to other circuits.
Virtex DC Input and Output levels

Values for VIL and VIH are recommended input voltages. Values for VOL and VOH are guaranteed output voltages over the
recommended operating conditions. Only selected standards are tested. These are chosen to ensure that all standards
meet their specifications. The selected standards are tested at minimum VCCO with the respective IOL and IOH currents
shown. Other standards are sample tested.
Input/Output VIL VIH VOL VOH IOL IOH

Standard V, min V, max V, min V, max V, Max V, Min mA mA
LVTTL (Note 1) - 0.5 0.8 2.0 5.5 0.4 2.4 24 - 24
LVCMOS2 - 0.5 0.7 1.7 5.5 10% VCCO 90% VCCO 1.5 - 0.5
PCI, 3.3 V - 0.5 44% VCCINT 60% VCCINT VCCO + 0.5 10% VCCO 90% VCCO Note 2 Note 2
PCI, 5.0 V - 0.5 0.8 2.0 5.5 0.55 2.4 Note 2 Note 2
GTL - 0.5 VREF - 0.05 VREF + 0.05 3.6 0.4 n/a 40 n/a
GTL+ - 0.5 VREF - 0.1 VREF + 0.1 3.6 0.6 n/a 36 n/a
HSTL I - 0.5 VREF - 0.1 VREF + 0.1 3.6 0.4 VCCO - 0.4 8 -8
HSTL III - 0.5 VREF - 0.1 VREF + 0.1 3.6 0.4 VCCO - 0.4 24 -8
HSTL IV - 0.5 VREF - 0.1 VREF + 0.1 3.6 0.4 VCCO - 0.4 48 -8
SSTL3 I - 0.5 VREF - 0.2 VREF + 0.2 3.6 VREF - 0.6 VREF + 0.6 8 -8
SSTL3 II - 0.5 VREF - 0.2 VREF + 0.2 3.6 VREF - 0.8 VREF + 0.8 16 -16
SSTL2 I - 0.5 VREF - 0.2 VREF + 0.2 3.6 VREF - 0.50 VREF + 0.50 7.6 -7.6
SSTL2 II - 0.5 VREF - 0.2 VREF + 0.2 3.6 VREF - 0.50 VREF + 0.50 15.2 -15.2
CTT - 0.5 VREF - 0.2 VREF + 0.2 3.6 VREF - 0.4 VREF + 0.4 8 -8
AGP - 0.5 VREF - 0.2 VREF + 0.2 3.6 10% VCCO 90% VCCO Note 2 Note 2
Note 1: VOL and VOH for lower drive currents are sample tested.
Note 2: Tested according to the relevant specifications.

R
Virtex Switching Characteristics

Testing of switching parameters is modeled after testing methods specified by MIL-M-38510/605. All devices are 100%
functionally tested. Internal timing parameters are derived from measuring internal test patterns. Listed below are
representative values. For more specific, more precise, and worst-case guaranteed data, use the values reported by the
static timing analyzer (TRCE in the Xilinx Development System) and back-annotated to the simulation netlist. All timing
parameters assume worst-case operating conditions (supply voltage and junction temperature). Values apply to all Virtex
devices unless otherwise noted.
Virtex IOB Input Switching Characteristics

Input delays associated with the pad are specified for LVTTL levels. For other standards, these delays typically vary by less
than 0.3 ns. Precise values are provided by the timing analyzer.
Speed Grade
Description Symbol -6 -5 -4 Units
Propagation Delays
Pad to I output, no delay All TIOPI 0.8 0.9 1.0 ns, max
Pad to I output, with delay XCV50 TIOPID 1.3 1.5 1.7 ns, max
XCV100 1.3 1.5 1.7 ns, max
XCV150 1.3 1.5 1.7 ns, max
XCV200 1.3 1.5 1.8 ns, max
XCV300 1.3 1.5 1.8 ns, max
XCV400 1.6 1.8 2.1 ns, max
XCV600 1.6 1.8 2.1 ns, max
XCV800 2.0 2.3 2.6 ns, max
XCV1000 2.0 2.3 2.7 ns, max
Pad to output IQ via transparent latch, no delay All TIOPLI 1.7 1.9 2.2 ns, max
Pad to output IQ via transparent latch, with delay XCV50 TIOPLID 3.0 3.5 4.0 ns, max
XCV100 3.0 3.5 4.0 ns, max
XCV150 3.1 3.6 4.2 ns, max
XCV200 3.2 3.7 4.3 ns, max
XCV300 3.2 3.7 4.3 ns, max
XCV400 3.4 3.9 4.5 ns, max
XCV600 3.5 4.0 4.6 ns, max
XCV800 3.7 4.3 4.9 ns, max
XCV1000 3.9 4.5 5.1 ns, max
Sequential Delays
Clock CLK to output IQ TIOCKIQ 1.1 1.2 1.4 ns, max
Setup and Hold Times with respect to Clock CLK Setup Time/Hold Time
Pad, no delay TIOPICK/TIOICKP 1.9/0.0 2.2/0.0 2.5/0.0 ns, min
Pad, with delay (Note 1) TIOPICKD/TIOICKPD 3.5/<0 4.1/<0 4.7/<0 ns, min
ICE input TIOICECK/TIOCKICE 0.8/0.0 0.9/0.0 1.0/0.0 ns, min
SR input (IFF, synchronous) TIOSRCKI/TIOCKISR 0.8/0.0 1.0/0.0 1.1/0.0 ns, min
Set/Reset Delays
SR input to IQ (asynchronous) TIOSRIQ 1.2 1.4 1.6 ns, max
GSR to output IQ TGSRQ 10.0 11.0 13.0 ns, max
Note 1: With delay, the IOB hold time is negative. This reduces or eliminates pad-to-pad hold time.

R
Virtex IOB Output Switching Characteristics

Output delays terminating at a pad are specified for LVTTL with 12 mA drive and slow slew rate, the default output standard.
For other standards, adjust the delays by adding the values shown. Tri-state turn-off delays should not be adjusted.
Speed Grade
Propagation Delays
O input to Pad TIOOP 4.1 5.3 6.1 ns, max
O input to Pad via transparent latch TIOOLP 4.3 5.5 6.4 ns, max
3-State Delays
T input to Pad high-impedance TIOTHZ 1.1 1.3 1.5 ns, max
T input to valid data on Pad TIOTON 5.0 5.8 6.6 ns, max
T input to Pad high-impedance via transparent latch TIOTLPHZ 1.7 1.9 2.2 ns, max
T input to valid data on Pad via transparent latch TIOTLPON 5.0 5.8 6.6 ns, max
GTS to Pad high impedance TGTS 5.0 5.8 6.7 ns, max
Sequential Delays
Clock CLK to Pad TIOCKP 5.8 6.7 7.7 ns, max 3
Clock CLK to Pad high-impedance (synchronous) TIOCKHZ 2.1 2.4 2.8 ns, max
Clock CLK to valid data on Pad (synchronous) TIOCKON 5.8 6.7 7.7 ns, max
Setup Times before Clock CLK
O input TIOOCK 0.5 0.5 0.6 ns, min
OCE input TIOOCECK 0.8 0.9 1.0 ns, min
SR input (OFF) TIOSRCKO 0.8 1.0 1.1 ns, min
3-State Setup Times
TCE input TIOTCECK 0.8 0.9 1.0 ns, min
T input TIOTCK 0.2 0.2 0.3 ns, min
SR input (TFF) TIOSRCKT 0.8 1.0 1.1 ns, min
Hold Times after Clock CLK
All Hold Times 0.0 0.0 0.0 ns, min
Set/Reset Delays
SR input to Pad (asynchronous) TIOSRP 5.0 5.8 6.6 ns, max
SR input to Pad high-impedance (asynchronous) TIOSRHZ 2.3 2.6 3.0 ns, max
SR input to valid data on Pad (asynchronous) TIOSRON 6.0 6.8 7.9 ns, max
GSR to Pad TGSRQ 10.0 11.0 13.0 ns, max
Output Delay Adjustments
Standard-specific increments for delays terminating at pads LVTTL, Slow, 2 mA 13.3 15.2 17.5 ns
4 mA 5.8 6.6 7.6 ns
6 mA 3.0 3.5 4.0 ns
8 mA 1.2 1.4 1.6 ns
12 mA 0.0 0.0 0.0 ns
16 mA - 0.2 - 0.2 - 0.3 ns
24 mA - 0.6 - 0.7 - 0.8 ns
LVTTL, Fast, 2 mA 11.5 13.3 15.3 ns
4 mA 3.5 4.1 4.7 ns
6 mA 1.3 1.5 1.7 ns
8 mA - 0.9 - 1.0 - 1.2 ns
12 mA - 1.9 - 2.2 - 2.5 ns
16 mA - 2.0 - 2.3 - 2.7 ns
24 mA - 2.3 - 2.6 - 3.0 ns
LVCMOS2 - 2.0 - 2.3 - 2.7 ns
PCI, 33 MHz, 3.3 V 0.3 0.4 0.4 ns
PCI, 33 MHz, 5.0 V - 0.9 - 1.1 - 1.2 ns
PCI, 66 MHz, 3.3 V - 2.7 - 3.1 - 3.6 ns
GTL - 2.6 - 3.0 - 3.5 ns
GTL+ - 1.9 - 2.2 - 2.5 ns
HSTL I - 2.8 - 3.2 - 3.7 ns
HSTL III - 2.8 - 3.3 - 3.7 ns
HSTL IV - 3.0 - 3.4 - 3.9 ns
SSTL3 I - 2.5 - 2.8 - 3.3 ns
SSTL3 II - 3.0 - 3.4 - 3.9 ns
SSTL2 I - 2.4 - 2.7 - 3.2 ns
SSTL2 II - 2.8 - 3.2 - 3.7 ns
CTT - 2.5 - 2.9 - 3.4 ns
AGP - 2.8 - 3.3 - 3.8 ns

R
Virtex Pad-to-Pad Switching Characteristics

Output delays terminating at the pad are specified for LVTTL levels with 12 mA drive and slow slew rate (the default output
standard). For other standards, these delays must be adjusted by adding the values shown in the Virtex IOB Output
Switching Characteristics table on page 27.
Speed Grade
Using a DLL
Pad-to-pad input data setup time before the clock ns, min
Pad-to-pad input data hold time after the clock ns, min
Pad-to-pad delay from clock input to data output ns, max
Without a DLL
Pad-to-pad input data setup time before the clock ns, min
Pad-to-pad input data hold time after the clock ns, min
Pad-to-pad delay from clock input to data output Note 1 Note 1 Note 1 ns, max
Note 1: These dfelays may be calculated by adding the Clock-to-Pad delay from the Virtex IOB Input Switching Characteristics table on
page 26 to the GCLK-pad-to-flip-flop delay shown in the Virtex Clock Distribution Guidelines table on page 28.
Virtex Clock Distribution Guidelines

Speed Grade
GCLK Distribution
From GCLK pad to any flip-flop XCV50 ns, max
XCV100 ns, max
Note: These clock-distribution delays are pro- XCV150 ns, max
vided for guidance only. They reflect the de- XCV200 ns, max
lays encountered in a typical design under XCV300 ns, max
worst-case conditions. Precise values for a XCV400 ns, max
particular design are provided by the timing XCV600 ns, max
analyzer. XCV800 ns, max
XCV1000 ns, max
Virtex Clock Distribution Switching Characteristics

Speed Grade
GCLK IOB and Buffer
Global Clock PAD to output. TGPID 1.0 1.1 1.3 ns, max
IN input to OUT output TGIO 0.9 1.0 1.2 ns, max

R
Virtex CLB Switching Characteristics

Delays originating at F/G inputs vary slightly according to the input used. The values listed below are worst-case. Precise
values are provided by the timing analyzer.
Speed Grade
Combinatorial Delays
4-input function: F/G inputs to X/Y outputs TILO 0.6 0.7 0.8 ns, max
5-input function: F/G inputs to F5 output TIF5 1.0 1.1 1.2 ns, max
5-input function: F/G inputs to X output TIF5X 1.0 1.2 1.3 ns, max
6-input function: F/G inputs to Y output via F6 MUX TIF6Y 1.2 1.4 1.6 ns, max
6-input function: F5IN input to Y output TF5INY 0.4 0.5 0.6 ns, max
Incremental delay routing through transparent latch to XQ/YQ out- TIFNCTL 0.4 0.5 0.6 ns, max
puts
BY input to YB output TBYYB 0.5 0.6 0.7 ns, max
Sequential Delays
FF Clock CLK to XQ/YQ outputs TCKO 1.1 1.3 1.4 ns, max 3
Latch Clock CLK to XQ/YQ outputs TCKLO 0.7 0.7 0.9 ns, max
4-input function: F/G Inputs TICK 1.0 1.1 1.2 ns, min
5-input function: F/G inputs TIF5CK 1.4 1.6 1.8 ns, min
6-input function: F5IN input TF5INCK 0.8 0.9 1.0 ns, min
6-input function: F/G inputs via F6 MUX TIF6CK 1.6 1.8 2.0 ns, min
BX/BY inputs TDICK 1.6 1.8 2.0 ns, min
CE input TCECK 0.8 0.9 1.0 ns, min
SR/BY inputs (synchronous) TRCK 1.3 1.5 1.7 ns, min
Clock CLK
Minimum Pulse Width, High TCH 2.0 2.3 2.6 ns, min
Minimum Pulse Width, Low TCL 2.0 2.3 2.6 ns, min
Set/Reset
Minimum Pulse Width, SR/BY inputs TRPW 2.9 3.4 3.9 ns, min
Delay from SR/BY inputs to XQ/YQ outputs (asynchronous) TRQ 1.6 1.9 2.2 ns, max
Delay from GSR to XQ/YQ outputs TGSRQ 10.0 11.0 13.0 ns, max

R
Virtex CLB Arithmetic Switching Characteristics

Setup times not listed explicitly can be approximated by decreasing the combinatorial delays by the setup time adjustment
listed. Precise values are provided by the timing analyzer.
Speed Grade
F operand inputs to X via XOR TOPX 0.8 0.9 1.0 ns, max
F operand input to XB output TOPXB 1.2 1.4 1.6 ns, max
F operand input to Y via XOR TOPY 1.6 1.9 2.2 ns, max
F operand input to YB output TOPYB 1.3 1.5 1.7 ns, max
F operand input to COUT output TOPCYF 1.3 1.5 1.7 ns, max
G operand inputs to Y via XOR TOPGY 1.0 1.1 1.3 ns, max
G operand input to YB output TOPGYB 1.4 1.6 1.9 ns, max
G operand input to COUT output TOPCYG 1.4 1.6 1.8 ns, max
BX initialization input to COUT TBXCY 0.8 0.9 1.0 ns, max
CIN input to X output via XOR TCINX 0.5 0.5 0.6 ns, max
CIN input to XB TCINXB 0.1 0.1 0.1 ns, max
CIN input to Y via XOR TCINY 0.5 0.6 0.7 ns, max
CIN input to YB TCINYB 0.2 0.2 0.2 ns, max
CIN input to COUT output
Multiplier Operation TBYP 0.1 0.2 0.2 ns, max
F1/2 operand inputs to XB output via AND TFANDXB 0.4 0.5 0.6 ns, max
F1/2 operand inputs to YB output via AND TFANDYB 0.5 0.6 0.6 ns, max
F1/2 operand inputs to COUT output via AND TFANDCY 0.5 0.5 0.6 ns, max
G1/2 operand inputs to YB output via AND TGANDYB 0.4 0.4 0.5 ns, max
G1/2 operand inputs to COUT output via AND TGANDCY 0.4 0.4 0.5 ns, max
CIN input to FFX TCCKX 0.8 0.9 1.1 ns, min
CIN input to FFY TCCKY 0.9 1.0 1.1 ns, min
Setup Time Adjustment ns
Virtex CLB SelectRAM Switching Characteristics

Speed Grade
Sequential Delays
Clock CLK to X/Y outputs (WE active) TSHCKO ns, max
Shift-Register Mode
Clock CLK to X/Y outputs ns, max
F/G address inputs TAS/TAH 0.6 0.7 0.8 ns, min
BX/BY data inputs (DIN) TDS/TDH 1.0 1.2 1.3 ns, min
CE input (WE) TWS/TWH 0.6 0.6 0.7 ns, min
Shift-Register Mode
BX/BY data inputs (DIN) TSHDICK ns, min
CE input (WS) TSHCECK ns, min
Clock CLK
Minimum Pulse Width, High TWPH 2.9 3.4 3.9 ns, min
Minimum Pulse Width, Low TWPL 2.9 3.4 3.9 ns, min
Minimum clock period to meet address write cycle time TWC 5.8 6.7 7.7 ns, min

R
Virtex BLOCKRAM Switching Characteristics

Speed Grade
Sequential Delays
Clock CLK to DOUT output TBCKO 3.3 3.8 4.4 ns, max
ADDR inputs TBACK 1.2 1.4 1.6 ns, min
DIN inputs TBDCK 1.2 1.4 1.6 ns, min
EN input TBECK 2.7 3.1 3.6 ns, min
RST input TBRCK 2.5 2.9 3.3 ns, min
WEN input TBWCK 2.4 2.8 3.2 ns, min
Clock CLK
Minimum Pulse Width, High TBPWH 2.0 2.3 2.6 ns, min
Minimum Pulse Width, Low TBPWL 2.0 2.3 2.6 ns, min 3
CLKA -> CLKB setup time for different ports TBCCS 3.0 3.5 4.0 ns, min
Virtex TBUF Switching Characteristics

Speed Grade
IN input to OUT output TIO 0.2 0.2 0.2 ns, max
TRI input to OUT output high-impedance TOFF 0.2 0.2 0.2 ns, max
Tri input to valid data on OUT output TON 0.2 0.2 0.2 ns, max
Virtex Test Access Port Switching Characteristics

Speed Grade
TMS and TDI Setup times before TCK TTAPTCK 4.0 4.0 4.0 ns, min
TMS and TDI Hold times after TCK TTCKTAP 2.0 2.0 2.0 ns, min
Output delay from clock TCK to output TDO TTCKTDO 11.0 11.0 11.0 ns, max
Maximum TCK clock frequency FTCK 33 33 33 MHz, max

R
Virtex Pin Outs

Pin-Out Tables
See the Xilinx WebLINX website (http://www.xilinx.com/partinfo/databook.htm) for updates or additional pin-out information.
For convenience, Table 14 and Table 15 list the locations of special-purpose and power-supply pins. Pins not listed are user
I/Os.
Table 14: Virtex Pin-out Tables (QFP Packages)
Pin Name Device TQ144 PQ/HQ240
GCK0 All 90 92
GCK1 All 93 89
GCK2 All 19 210
GCK3 All 16 213
M0 All 110 60
M1 All 112 58
M2 All 108 62
CCLK All 38 179
PROGRAM All 72 122
DONE All 74 120
INIT All 71 123
BUSY/DOUT All 39 178
D0/DIN All 40 177
D1 All 45 167
D2 All 47 163
D3 All 51 156
D4 All 59 145
D5 All 63 138
D6 All 65 134
D7 All 70 124
WRITE All 32 185
CS All 33 184
TDI All 34 183
TDO All 36 181
TMS All 143 2

R
Table 14: Virtex Pin-out Tables (QFP Packages) (Continued)
TCK All 2 239
VCCINT All 10, 15, 16, 32,

25, 57, 43, 77,
84, 94, 88, 104,
99, 126 137, 148,
164, 198,
214, 225
VCCO, Unbanked All 1, 17, 15, 30,

37, 55, 44, 61,
73, 92, 76, 90,
109, 128 105, 121, 3
136, 150,
165, 180,
197, 212,
226, 240
VREF, Bank 0 XCV50 5, 13 218, 232
(VREF pins are listed in- XCV100/150 ... + 7 ... + 229

crementally. Connect all
pins listed for both the XCV200/300 ... + 236
required device and all
smaller devices listed in XCV400 ... + 215
the same package.)
XCV600 ... + 230
XCV800 ... + 222
VREF, Bank 1 XCV50 22, 30 191, 205

the same package.)
XCV600 ... + 193
XCV800 ... + 201
VREF, Bank 2 XCV50 42, 50 157, 171

the same package.)
XCV600 ... + 169
XCV800 ... + 161

R
VREF, Bank 3 XCV50 60, 68 130, 144

the same package.)
XCV600 ... + 132
XCV800 ... + 140
VREF, Bank 4 XCV50 79, 87 97, 111

the same package.)
XCV600 ... + 109
XCV800 ... + 101
VREF, Bank 5 XCV50 96, 104 70, 84

the same package.)
XCV600 ... + 72
XCV800 ... +80
VREF, Bank 6 XCV50 116, 123 36, 50

the same package.)
XCV600 ... + 48
XCV800 ... + 40

R
VREF, Bank 7 XCV50 133, 140 9, 23

the same package.)
XCV600 ... + 11
XCV800 ... + 19
GND All 9, 18, 1, 8,

26, 35, 14, 22, 3
46, 54, 29, 37,
64, 75, 45, 51,
83, 91, 59, 69,
100, 111, 75, 83,
120, 129, 91, 98,
136, 144, 106, 112,
119, 129,
135, 143,
151, 158,
166, 172,
182, 190,
196, 204,
211, 219,
227, 233

R
Table 15: Virtex Pin-out Tables (BGA)
Pin Name Device BG256 BG352 BG432 BG560
GCK0 All Y11 AE13 AL16 AL17
GCK1 All Y10 AF14 AK16 AJ17
GCK2 All A10 B14 A16 D17
GCK3 All B10 D14 D17 A17
M0 All Y1 AD24 AH28 AJ29
M1 All U3 AB23 AH29 AK30
M2 All W2 AC23 AJ28 AN32
CCLK All B19 C3 D4 C4
PROGRAM All Y20 AC4 AH3 AM1
DONE All W19 AD3 AH4 AJ5
INIT All U18 AD2 AJ2 AH5
BUSY/DOUT All D18 E4 D3 D4
D0/DIN All C19 D3 C2 E4
D1 All E20 G1 K4 K3
D2 All G19 J3 K2 L4
D3 All J19 M3 P4 P3
D4 All M19 R3 V4 W4
D5 All P19 U4 AB1 AB5
D6 All T20 V3 AB3 AC4
D7 All V19 AC3 AG4 AJ4
WRITE All A19 D5 B4 D6
CS All B18 C4 D5 A2
TDI All C17 B3 B3 D5
TDO All A20 D4 C4 E6
TMS All D3 D23 D29 B33
TCK All A1 C24 D28 E29
DXN All W3 AD23 AH27 AK29
DXP All V4 AE24 AK29 AJ28

R
Table 15: Virtex Pin-out Tables (BGA) (Continued)
VCCINT XCV50/100/ C10, D6, A20, C14,

150/200 D15, F4, D10, J24,
(VCCINT pins are listed in- F17, L3, K4, P2,
pins listed for both the re- L18, R4, P25, V24,
quired device and all R17, U6, W2, AC10,
smaller devices listed in U15, V10 AE14, AE19
the same package.)
XCV300 ... + B16, D12, A10, A17,
L1, L25, B23, C14,
R23, T1, C19, K3,
AF11, AF16 K29, N2,
N29, T1,
T29, W2, 3
W31, AB2,
AB30, AJ10,
AJ16, AK13,
AK19, AK22
XCV400/600 ... + B26, C7, A21, B14,

F1, F30, B18, B28,
AE29, AF1, C24, E9,
AH8, AH24 E12, F2,
H30, J1,
K32, N1,
N33, U5,
U30, Y2,
Y31, AD2,
AD32, AG3,
AG31, AK8,
AK11, AK17,
AK20, AL14,
AL27, AN25
XCV800/1000 ... + B12, C22,

M3, N29,
AB2, AB32,
AJ13, AL22,
VCCO, Bank 0 All D7, D8 A17, B25, A21, C29, A22, A26,
D19 D21 A30, B19,
B32
VCCO, Bank 1 All D13, D14 A10, D7, A1, A11, A10, A16,
D13 D11 B13, C3,
E5
VCCO, Bank 2 All G17, H17 B2, H4, C3, L1, B2, D1,
K1 L4 H1, M1,
R2
VCCO, Bank 3 All N17, P17 P4, U1, AA1, AA4, V1, AA2,
Y4 AJ3 AD1, AK1,
AL2

R
VCCO, Bank 4 All U13, U14 AC8, AE2, AH11, AL1, AM2, AM15,
AF10 AL11 AN4, AN8,
AN12
VCCO, Bank 5 All U7, U8 AC14, AC20, AH21, AJ29, AL31, AM21,
AF17 AL21 AN18, AN24,
AN30
VCCO, Bank 6 All N4, P4 U26, W23, AA28, AA31, W32, AB33,
AE25 AL31 AF33, AK33,
AM32
VCCO, Bank 7 All G4, H4 G23, K26, A31, L28, C32, D33,
N23 L31 K33, N32,
T33
VREF, Bank 0 XCV50 A8, B4

(VREF pins are listed in- XCV100/150 ... + A4 A16,C19,
pins listed for both the re- C21
quired device and all
smaller devices listed in XCV200/300 ... + D21 B19, D22,
the same package.) D24, D26
XCV400 ... + B15 ... + C18 A19, D20,

D26, E23,
E27
XCV600 ... + C24 ... + E24
XCV800 ... + B21 ... + E21
XCV1000 ... + D29
VREF, Bank 1 XCV50 A17, B12
(VREF pins are listed in- XCV100/150 ... + B15 B6, C9,
crementally. Connect all C12
pins listed for both the re-
quired device and all XCV200/300 ... + D6 A13, B7,
smaller devices listed in
the same package.) C6, C10
XCV400 ... + C13 ... + B15 A6, D7,

D11, D16,
E15
XCV600 ... + D10 ... + D10
XCV800 ... + B12 ... + D13
XCV1000 ... + E7

R
VREF, Bank 2 XCV50 C20, J18
(VREF pins are listed in- XCV100/150 ... + F19 E2, H2,
crementally. Connect all M4
quired device and all XCV200/300 ... + D2 E2, G3,
the same package.) J2, N1
XCV400 ... + M1 ... + R3 G5, H4,

L5, P4,
R1
XCV600 ... + H1 ... + K5 3

XCV800 ... + M3 ... + N5
XCV1000 ... + B3
VREF, Bank 3 XCV50 M18, V20
(VREF pins are listed in- XCV100/150 ... + R19 R4, V4,
crementally. Connect all Y3
quired device and all XCV200/300 ... + AC2 V2, AB4,
the same package.) AD4, AF3
XCV400 . + R1 ... + U2 V4, W5,

AD3, AE5,
AK2
XCV600 ... + AC3 ... + AF1
XCV800 ... + Y3 ... + AA4
XCV1000 ... + AH4
VREF, Bank 4 XCV50 V12, Y18
(VREF pins are listed in- XCV100/150 ... + W15 AC12, AE5,
crementally. Connect all AE8,
quired device and all XCV200/300 ... + AE4 AJ7, AL4,
the same package.) AL8, AL13
XCV400 ... + AF12 ... + AK15 AL7, AL10,

AL16, AM4,
AM14
XCV600 ... + AK8 ... + AL9
XCV800 ... + AJ12 ... + AK13
XCV1000 ... + AN3

R
VREF, Bank 5 XCV50 V9, Y3
(VREF pins are listed in- XCV100/150 ... + W6 AC15, AC18,

crementally. Connect all AD20
quired device and all XCV200/300 ... + AE23 AJ18, AJ25,
the same package.) AK23, AK27
XCV400 ... + AF15 ... + AJ17 AJ18, AJ25,

AL20, AL24,
AL29
XCV600 ... + AL24 ... + AM26
XCV800 ... + AH19 ... + AN23
XCV1000 ... + AK28
VREF, Bank 6 XCV50 M2, R3
(VREF pins are listed in- XCV100/150 ...+ T1 R24, Y26,

crementally. Connect all AA25,
quired device and all XCV200/300 ... + AD26 V28, AB28,
the same package.) AE30, AF28
XCV400 ... + P24 ... + U28 V29, Y32,

AD31, AE29,
AK32
XCV600 ... + AC28 ... + AE31
XCV800 ... + Y30 ... + AA30
XCV1000 ... + AH30
VREF, Bank 7 XCV50 G3, H1
(VREF pins are listed in- XCV100/150 ... + D1 D26, G26,

crementally. Connect all L26
quired device and all XCV200/300 ... + E24 F28, F31,
the same package.) J30, N30
XCV400 ... + M25 ... + R31 E31, G31,

K31, P31,
T31
XCV600 ... + J28 ... + H32
XCV800 ... + M28 ... + L33
XCV1000 ... + D31

R
GND All C3, C18, A1, A2, A2, A3, A1, A7,
D4, D5, A5, A8, A7, A9, A12, A14,
D9, D10, A14, A19, A14, A18, A18, A20,
D11, D12, A22, A25, A23, A25, A24, A29,
D16, D17. A26, B1, A29, A30, A32, A33,
E4, E17, B26, E1, B1, B2, B1, B6,
J4, J9, E26, H1, B30, B31, B9, B15,
J10, J11, H26, N1, C1, C31, B23, B27,
J12, J17, P26, W1, D16, G1, B31, C2,
K4, K9, W26, AB1, G31, J1, E1, F32,
K10, K11, AB26, AE1, J31, P1, G2, G33,
K12, K17, AE26, AF1, P31, T4, J32, K1,
3
L4, L9, AF2, AF5, T28, V1, L2, M33,
L10, L11, AF8, AF13, V31, AC1, P1, P33,
L12, L17, AF19, AF22, AC31, AE1, R32, T1,
M4, M9, AF25, AF26 AE31, AH16, V33, W2,
M10, M11, AJ1, AJ31, Y1, Y33,
M12, M17, AK1, AK2, AB1, AC32,
T4, T17, AK30, AK31, AD33, AE2,
U4, U5, AL2, AL3, AG1, AG32,
U9, U10, AL7, AL9 AH2, AJ33,
U11, U12, AL14, AL18 AL32, AM3,
U16, U17, AL23, AL25, AM7, AM11,
V3, V18 AL29, AL30 AM19, AM25,
AM28, AM33,
AN1, AN2,
AN5, AN10,
AN14, AN16,
AN20, AN22,
AN27, AN33
No Connect C31, AC2,

AK4, AL3

R
Pin-Out Diagrams
The following diagrams, pages 44 through 48, illustrate the Table 16 lists the symbols used in these diagrams. The dia-
locations of special-purpose pins on Virtex FPGAs. grams also show I/O-bank boundaries.
Table 16: Pin-out Diagram Symbols
Symbol Pin Function

✳ General I/O
V VCCINT
v Device-dependent VCCINT, n/c on smaller devices
O VCCO
R VREF
r Device-dependent VREF, remains I/O on smaller devices
G Ground
Ø, 1, 2, 3 Global Clocks
❿, ❶, ❷ M0, M1, M2
➉, ➀, ➁, ➂, D0/DIN, D1, D2, D3, D4, D5, D6, D7
➃, ➄, ➅, ➆
B DOUT/BUSY
D DONE
P PROGRAM
I INIT
K CCLK
W WRITE
S CS
T Boundary-scan Test Access Port
+ Temperature diode, anode
– Temperature diode, cathode
n No connect

R
TQ144 Pin-out Diagram
117
116
115
114
113
112
111
110
109
144
143
142
141
140
139
138
137
136
135
134
133
132
131
130
129
128
127
126
125
124
123
122
121
120
119
118
G T ✳✳R ✳ r ✳G✳✳R ✳✳✳GO✳ V ✳✳R ✳✳G✳ r ✳R ✳✳✳ ❶G ❿O
1 O ❷ 108
2 T ✳ 107
3 ✳ Bank 7 Bank 6 ✳ 106
4 ✳ ✳ 105
5 R R 104
6 ✳ ✳ 103
7 r r 102
8 ✳ ✳ 101
9 G G 100
10 V Bank 0 Bank 5 V 99
11 ✳ ✳ 98 3
12 ✳ ✳ 97
13 R R 96
14 ✳ ✳ 95
15 V V 94
16 3 TQ144 1 93
17 O O 92
18 G (Top view) G 91
19 2 Ø 90
20 ✳ ✳ 89
21 ✳ ✳ 88
22 R R 87
23 ✳ ✳ 86
24 ✳ ✳ 85
25 V V 84
26 G Bank 1 Bank 4 G 83
27 ✳ ✳ 82
28 r r 81
29 ✳ ✳ 80
30 R R 79
31 ✳ ✳ 78
32 W ✳ 77
33 S ✳ 76
34 T Bank 2 Bank 3 G 75
35 G D 74
36 T O 73
O K B ➉ ✳ R ✳ r ➀ G ➁ ✳ ✳ R ➂ ✳ ✳ G O ✳ V ✳➃ R ✳ ✳ ➄ G ➅ r ✳ R ✳ ➆ I P
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72

R
PQ240/HQ240 Pin-out Diagram
233
223
213
203
193
183
239
229
219
209
199
189
235
225
215
205
195
185
231
221
211
201
191
181
237
227
217
207
197
187
T ✳ ✳ G ✳ r G V ✳ ✳ G ✳ r 3 G ✳ ✳ R ✳ r ✳ O ✳ r R ✳ r W T T
O ✳ r ✳ R r ✳ O ✳ r ✳ R ✳ V O 2 r ✳ G ✳ ✳ V G r ✳ G ✳ ✳ S G
1 G Bank 0 Bank 1 O
T K 179
3 ✳ B
✳ ➉ 177
5 r ✳
✳ r 175
7 ✳ ✳
G ✳ 173
9 R G
✳ R 171
11 r ✳
r r 169
13 ✳ r
G ➀ 167
15 O Bank 7 Bank 2 G
V O 165
17 ✳ V
✳ ➁ 163
19 r ✳
✳ r 161
21 ✳ ✳
G ✳ 159
23 R G
✳ PQ240/HQ240 R 157
25 ✳ ➂
r (Top view) ✳ 155
27 ✳ r
✳ ✳ 153
29 G ✳
O
Pins are shown staggered G 151
31 ✳ O
V for readability ✳ 149
33 r V
✳ r 147
35 ✳ ✳
R ➃ 145
37 G R
✳ G 143
39 ✳ ✳
r ✳ 141
41 ✳ r
✳ Bank 6 Bank 3 ✳ 139
43 V ➄
O V 137
45 G O
✳ G 135
47 r ➅
r r 133
49 ✳ r
R ✳ 131
51 G R
✳ G 129
53 ✳ ✳
r ✳ 127
55 ✳ r
✳ ✳ 125
57 ✳ ➆
❶ Bank 5 Bank 4 I 123
59 G P
❿ O 121
❷ ✳ r ✳ R r ✳ O ✳ r ✳ R ✳ V O Ø r ✳ G ✳ ✳ V G r ✳ G ✳ ✳ ✳ D
O ✳ ✳ ✳ G ✳ r G V ✳ ✳ G ✳ r 1 G ✳ ✳ R ✳ r ✳ O ✳ r R ✳ r ✳ G
63
73
83
93
103
113
67
77
87
97
107
117
61
71
81
91
101
111
65
75
85
95
105
115
69
79
89
99
109
119

R
BG256 Pin-out Diagram
10
11
12
13
14
15
16
17
18
19
20
1
2
3
4
5
6
7
8
9
A T ✳ ✳ r ✳ ✳ ✳ R✳ 2 ✳ ✳ ✳ ✳ ✳ ✳ R ✳ W T A
B ✳ ✳ ✳ R ✳ ✳ ✳ ✳ ✳ 3 ✳ R✳ ✳ r ✳ ✳ S K✳ B
C ✳ ✳ G ✳ ✳ ✳ ✳ ✳ ✳ V ✳ ✳ ✳ ✳ ✳ ✳ T G ➉ R C
D r ✳ T G G V O O G G G G O O V G G B ✳ ✳ D
E ✳ ✳ ✳ G Bank 0 Bank 1 G ✳ ✳ ➀ E
F ✳ ✳ ✳ V V ✳ r ✳ F
G ✳ ✳ R O BG256 O ✳ ➁✳ G
H R ✳ ✳ O Bank 7 Bank 2 O ✳ ✳ ✳ H 3
J ✳ ✳ ✳ G G G G G G R ➂✳ J
K ✳ ✳ ✳ G G G G G G ✳ ✳ ✳ K
L ✳ ✳ V G G G G G G V ✳ ✳ L
M ✳ R ✳ G G G G G G R ➃✳ M
N ✳ ✳ ✳ O Bank 6 Bank 3 O ✳ ✳ ✳ N
P ✳ ✳ ✳ O (Top View) O ✳ ➄✳ P
R ✳ ✳ R V V ✳ r ✳ R
T r ✳ ✳ G Bank 5 Bank 4 G ✳ ✳ ➅ T
U ✳ ✳ ❶ G G V O O G G G G O O V G G I ✳ ✳ U
V ✳ ✳ G + ✳ ✳ ✳ ✳ R V ✳ R✳ ✳ ✳ ✳ ✳ G ➆ R V
W ✳ ❷ – ✳ ✳ r ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ r ✳ ✳ ✳ D✳ W
Y ❿ ✳ R ✳ ✳ ✳ ✳ ✳ ✳ 1 Ø ✳ ✳ ✳ ✳ ✳ ✳ R ✳ P Y
10
11
12
13
14
15
16
17
18
19
20
1
2
3
4
5
6
7
8
9

R
24
25
26
10
11
12
13
14
15
16
17
18
19
20
21
22
23
1
2
3
4
5
6
7
8
9
A G G ✳ ✳ G ✳ ✳ G ✳ O ✳ ✳ ✳ G ✳ R O ✳ G V ✳ G ✳ ✳ G G A
B G O T ✳ ✳ R ✳ ✳ ✳ ✳ ✳ ✳ ✳ 2 ✳ v ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ O G B
C ✳ ✳ K S✳ ✳ ✳ ✳ R ✳ ✳ R ✳ V ✳ ✳ ✳ ✳ R ✳ R ✳ ✳ T ✳ ✳ C
D ✳ r ➉ T W r O ✳ ✳ V ✳ v O 3 ✳ ✳ ✳ ✳ O ✳ r ✳ T ✳ ✳ R D
E G R ✳ B ✳ r ✳ G E
F ✳ ✳ ✳ ✳ Bank 1 Bank 0 ✳ ✳ ✳ ✳ F
G ➀✳ ✳ ✳ O ✳ ✳ R G
H G R ✳ O ✳ ✳ ✳ G H
J ✳ ✳ ➁✳ ✳ V ✳ ✳ J
K O ✳ ✳V Bank 2 Bank 7 ✳ ✳ ✳ O K
L v ✳ ✳ ✳ ✳ ✳ v R L
M ✳ ✳ ➂R ✳ ✳ ✳ ✳ M
N G ✳ ✳ ✳ BG352 O ✳ ✳ ✳ N
P ✳V ✳ O (Top View) ✳ ✳ V G P
R ✳ ✳ ➃R v R ✳ ✳ R
T v ✳ ✳ ✳ ✳ ✳ ✳ ✳ T
U O ✳ ✳ ➄ Bank 3 Bank 6 ✳ ✳ ✳ O U
V ✳ ✳ ➅R ✳ V ✳ ✳ V
W GV ✳ ✳ O ✳ ✳ G W
Y ✳ ✳ R O ✳ ✳ ✳ R Y
AA ✳ ✳ ✳ ✳ Bank 4 Bank 5 ✳ ✳ R ✳ AA
AB G ✳ ✳ ✳ ❶ ✳ ✳ G AB
AC ✳ r ➆P ✳ ✳ ✳ O ✳ V ✳ R ✳ O R ✳ ✳ R ✳ O ✳ ✳ ❷ ✳ ✳ ✳ AC
AD ✳ I D ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ R ✳ ✳ – ❿ ✳ r AD
AE G O ✳ r R ✳ ✳ R ✳ ✳ ✳ ✳ Ø V ✳ ✳ ✳ ✳ V ✳ ✳ ✳ r + O G AE
AF G G ✳ ✳ G ✳ ✳ G ✳ O v ✳ G 1 ✳ v O ✳ G ✳ ✳ G ✳ ✳ G G AF
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
1
2
3
4
5
6
7
8
9

R
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
1
2
3
4
5
6
7
8
9
A O G G ✳ ✳ ✳ G ✳ G V O ✳ R G ✳ 2 V G ✳ ✳ O ✳ G ✳ G ✳ ✳ ✳ G G O A
B G G T W ✳ ✳ R ✳ ✳ ✳ ✳ r ✳ ✳ r ✳ ✳ ✳ R ✳ r ✳ V ✳ ✳ v ✳ ✳ ✳ G G B
C G ➉ O T ✳ R v ✳ ✳ R ✳ ✳ ✳ V ✳ ✳ ✳ r V ✳ ✳ ✳ ✳ r ✳ ✳ ✳ ✳ O ✳ G C
D ✳ ✳ B K S ✳ ✳ ✳ ✳ r O ✳ ✳ ✳ ✳ G 3 ✳ ✳ ✳ O R ✳ R ✳ R ✳ T T ✳ ✳ D
E ✳ R ✳ ✳ ✳ ✳ ✳ ✳ E
F v ✳ ✳ ✳ Bank 1 Bank 0 R ✳ v R F
G G ✳ R ✳ ✳ ✳ ✳ G G
H r ✳ ✳ ✳ ✳ ✳ ✳ ✳ H 3
J G R ✳ ✳ r ✳ R G J
K ✳ ➁ V ➀ Bank 2 Bank 7 ✳ V ✳ ✳ K
L O ✳ ✳ O O ✳ ✳ O L
M ✳ ✳ r ✳ r ✳ ✳ ✳ M
N R V ✳ ✳ ✳ V R ✳ N
P G ✳ ✳ ➂ ✳ ✳ ✳ G P
R ✳ ✳ r ✳ BG432 ✳ ✳ ✳ r R
T V ✳ ✳ G (Top View) G V ✳ ✳ T
U ✳ r ✳ ✳ r ✳ ✳ ✳ U
V G R ✳ ➃ R ✳ ✳ G V
W ✳ V ✳ ✳ ✳ ✳ ✳ V W
Y ✳ ✳ r ✳ ✳ ✳ r ✳ Y
AA O ✳ ✳ O O ✳ ✳ O AA
AB ➄ V ➅ R Bank 3 Bank 6 R ✳ V ✳ AB
AC G ✳ r ✳ r ✳ ✳ G AC
AD ✳ ✳ ✳ R ✳ ✳ ✳ ✳ AD
AE G ✳ ✳ ✳ ✳ v R G AE
AF v ✳ R ✳ Bank 4 Bank 5 R ✳ ✳ ✳ AF
AG ✳ ✳ ✳ ➆ ✳ ✳ ✳ ✳ AG
AH ✳ ✳ P D ✳ ✳ ✳ v ✳ ✳ O ✳ ✳ ✳ ✳ G ✳ ✳ r ✳ O ✳ ✳ v ✳ ✳ – ❿ ❶ ✳ ✳ AH
AJ G I O ✳ ✳ ✳ R ✳ ✳ V ✳ r ✳ ✳ ✳ V r R ✳ ✳ ✳ ✳ ✳ ✳ R ✳ ✳ ❷ O ✳ G AJ
AK G G ✳ ✳ ✳ ✳ ✳ r ✳ ✳ ✳ ✳ V ✳ r 1 ✳ ✳ V ✳ ✳ V R ✳ ✳ ✳ R ✳ + G G AK
AL O G G R ✳ ✳ G R G ✳ O ✳ R G ✳ Ø ✳ G ✳ ✳ O ✳ G r G ✳ ✳ ✳ G G O AL
29
30
31
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1
2
3
4
5
6
7
8
9

R
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
1
2
3
4
5
6
7
8
9
A G S ✳ ✳ ✳ R G ✳ ✳ O ✳ G ✳ G ✳ O 3 G R G V O ✳ G ✳ O ✳ ✳ G O ✳ G G A
B G O r ✳ ✳ G ✳ ✳ G ✳ ✳ v O V G ✳ ✳ V O ✳ ✳ ✳ G ✳ ✳ ✳ G V ✳ ✳ G O T B
C ✳ G OK ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ v ✳V ✳ ✳ ✳ ✳ ✳ ✳ n O ✳ C
D O ✳ ✳ B T W R ✳ ✳ r R ✳ r ✳ ✳ R 2 ✳ ✳ R ✳ ✳ ✳ ✳ ✳ R ✳ ✳ r ✳ r ✳ O D
E G ✳ ✳ ➉O T r ✳V ✳ ✳ V ✳ ✳ R ✳ ✳ ✳ ✳ ✳ r ✳ R r ✳ ✳ R ✳ T ✳ R ✳ ✳ E
F ✳ V ✳ ✳ ✳ Bank 1 Bank 0 ✳ ✳ ✳ G ✳ F
G ✳ G ✳ ✳ R ✳ ✳ R ✳ G G
H O ✳ ✳ R ✳ ✳ V ✳ r ✳ H
J V ✳ ✳ ✳ ✳ ✳ ✳ ✳ G ✳ J
K G ✳ ➀✳ r Bank 2 Bank 7 ✳ ✳ R V O K
L ✳ G ✳ ➁R ✳ ✳ ✳ ✳ r L
M O ✳ v ✳ ✳ ✳ ✳ ✳ ✳ G M
N V ✳ ✳ ✳ r v ✳ ✳ O V N
P G ✳ ➂R ✳ ✳ ✳ R ✳ G P
R R O ✳ ✳ ✳ BG560 ✳ ✳ ✳ G ✳ R
T G ✳ ✳ ✳ ✳ (Top View) ✳ ✳ R ✳ O T
U ✳ ✳ ✳ ✳V ✳ V ✳ ✳ ✳ U
V O ✳ ✳ R ✳ R ✳ ✳ ✳ G V
W ✳ G ✳ ➃R ✳ ✳ ✳ O ✳ W
Y G V ✳ ✳ ✳ ✳ ✳ V R G Y
AA ✳ O ✳ r ✳ ✳ r ✳ ✳ ✳ AA
AB G v ✳ ✳ ➄ ✳ ✳ ✳ v O AB
AC ✳ n ✳ ➅ ✳ Bank 3 Bank 6 ✳ ✳ ✳ G ✳ AC
AD O V R ✳ ✳ ✳ ✳ R V G AD
AE ✳ G ✳ ✳ R R ✳ r ✳ ✳ AE
AF r ✳ ✳ ✳ ✳ ✳ ✳ ✳ ✳ O AF
AG G ✳ V ✳ ✳ Bank 4 Bank 5 ✳ ✳ V G ✳ AG
AH ✳ G ✳ r I ✳ r ✳ ✳ ✳ AH
AJ ✳ ✳ ✳ ➆D ✳ ✳ ✳ ✳ ✳ ✳ ✳ v ✳ ✳ ✳ 1 R ✳ ✳ ✳ ✳ ✳ ✳ R ✳ ✳ + ❿ ✳ ✳ ✳ G AJ
AK O R ✳ n ✳ ✳ ✳V ✳ ✳V ✳ r ✳ ✳ ✳ V ✳ ✳ V ✳ ✳ ✳ ✳ ✳ ✳ ✳ r – ❶ ✳ R O AK
AL ✳ O n ✳ ✳ ✳ R ✳ r R ✳ ✳ ✳ V ✳ R Ø ✳ ✳ R ✳ v ✳ R ✳ ✳ V ✳ R ✳ O G ✳ AL
AM P O G R ✳ ✳ G ✳ ✳ ✳ G ✳ ✳ R O ✳ ✳ ✳ G ✳ O ✳ ✳ ✳ G r ✳ G ✳ ✳ ✳ O G AM
AN G G r O G ✳ ✳ O ✳ G ✳ O ✳ G ✳ G ✳ O ✳ G ✳ G r O V ✳ G ✳ ✳ O ✳ ❷ G AN
33
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
1
2
3
4
5
6
7
8
9

R
Device/Package Combinations and Maximum I/O

Maximum User I/O
Package
XCV50 XCV100 XCV150 XCV200 XCV300 XCV400 XCV600 XCV800 XCV1000
CS144 94 94
TQ144 94 94
PQ240 164 164 164 164 164
HQ240 164 164 164
BG256 180 180
BG352 260 260 260
BG432 316 316 316 316
BG560 404 404 404 404
FG256 176 176 176 176
FG456 260 284 312
3
FG600 404 404 404
FG680 500 514 514
Ordering Information
Example: XCV300 -6 PQ 240 C

Device Type Temperature Range
C = Commercial (TJ = 0°C to +85°C)
I = Industrial (TJ = -40°C to +100°C)
Speed Grade
-4
-5 Number of Pins
-6
Package Types
BG = Ball Grid Array
FG = Fine-pitch Ball Grid Array
PQ = Plastic Quad Flat Pack
HQ = High Heat Dissipation QFP
TQ = Thin Quad Flat Pack
CS = Chip-scale Package

R
Revision Table
Date Revision
11/98 Initial document release.
1/99 Update of package drawings, updated specifications

R
Spartan Products
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

0
R
Spartan Products
Table of Contents
0 4*
Spartan and SpartanXL Families Field Programmable Gate Arrays

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
Spartan Series Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
Additional SpartanXL Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
General Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4
Logic Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
Configurable Logic Blocks (CLBs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
Function Generators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
Flip-Flops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
Latches (SpartanXL only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
CLB Signal Flow Control. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7
Input/Output Blocks (IOBs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7 4
IOB Input Signal Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7
IOB Output Signal Path. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9
Routing Channel Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
CLB Routing Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-13
I/O Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-14
Global Nets and Buffers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-14
Advanced Features Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15
Distributed RAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15
Memory Configuration Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15
Fast Carry Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-18
3-State Long Line Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20
Three-State Buffer Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20
On-Chip Oscillator. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-21
Global Signals: GSR and GTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-21
Global Set/Reset. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-21
Global 3-State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-21
Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-21
Data Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-22
Instruction Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-22
Bit Sequence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-23
Including Boundary Scan in a Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-23
Boundary Scan Enhancements (SpartanXL only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-24
Power Down (SpartanXL Only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-24
Configuration and Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-25
Configuration Mode Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-25
Master Serial Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-26
Slave Serial Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27
Serial Daisy Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27
Express Mode (SpartanXL only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-28
Pseudo Daisy Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-28
Setting CCLK Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-30
Cyclic Redundancy Check (CRC) for Configuration and Readback . . . . . . . . . . . . . . . . . . 4-31
Configuration Memory Clear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32
Initialization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-33
Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-33
4-1
R
Spartan Products Table of Contents
Delaying Configuration After Power-Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-34

Configuration Through the Boundary Scan Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-34
Readback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-35
Readback Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-35
Readback with the XChecker Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-36
Spartan Program Readback Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . 4-36
Configuration Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-37
Master Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-37
Slave Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-37
Spartan Detailed Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-38
Spartan Absolute Maximum Ratings1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-38
Spartan Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-38
Spartan DC Characteristics Over Operating Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-39
Spartan Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-39
Spartan CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-40
Spartan CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines . . . . . . . . . . . . 4-41
Spartan Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-43
Spartan Output Flip-Flop, Clock-to-Out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-43
Spartan Pin-to-Pin Input Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-44
Spartan Primary and Secondary Setup and Hold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-44
Spartan IOB Input Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-45
Spartan IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-46
SpartanXL Detailed Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-47
SpartanXL Absolute Maximum Ratings1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-47
SpartanXL Recommended Operating Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-47
SpartanXL DC Characteristics Over Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-48
SpartanXL Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 4-48
SpartanXL CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-49
SpartanXL CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines . . . . . . . . . . 4-50
SpartanXL Pin-to-Pin Output Parameter Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-52
SpartanXL Output Flip-Flop, Clock-to-Out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-52
SpartanXL Pin-to-Pin Input Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-52
SpartanXL Setup and Hold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-52
SpartanXL IOB Input Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-54
SpartanXL IOB Output Switching Characteristic Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-55
Pin Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-56
Device-Specific Pinout Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-58
Pin Locations for XCS05 & XCS05XL Devices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-58
Additional XCS10/XL Package Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-60
Product Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-68
4-2
marc
0
R
Spartan and SpartanXL Families
January 6, 1999 (Version 1.4) 0 4 Preliminary Product Specification
Introduction • System level features

- Available in both 5.0 Volt and 3.3 Volt versions
The SpartanTM Series is the first high-volume production - On-chip SelectRAMTM memory
FPGA solution to deliver all the key requirements for ASIC - Fully PCI compliant
replacement up to 40,000 gates. These requirements - Low power segmented routing architecture
include high performance, on-chip RAM, Core Solutions - Full readback capability for program verification and
and prices that, in high volume, approach and in many internal node observability
cases are equivalent to mask programmed ASIC devices. - Dedicated high-speed carry logic
The Spartan Series is the result of more than thirteen years - Internal 3-state bus capability
of FPGA design experience and feedback from thousands - 8 global low-skew clock or signal networks
of customers. By streamlining the Spartan feature set, - IEEE 1149.1-compatible boundary scan logic
leveraging advanced hybrid process technologies and • Versatile I/O and packaging 4
focusing on total cost management, the Spartan Series - Low cost plastic packages available in all densities
delivers the key features required by ASIC and other high - Footprint compatibility in common packages
volume logic users while avoiding the initial cost, long - Individually programmable output slew-rate control
development cycles and inherent risk of conventional maximizes performance and reduces noise
ASICs. The Spartan Series currently has 10 members, as - Zero input register hold time simplifies system timing
shown in Table 1. • Fully supported by powerful Xilinx development system
- Foundation series: Integrated, shrink-wrap software
Spartan Series Features - Alliance series: Over 100 PC and workstation 3RD
party development systems supported
Note: The Spartan Series devices described in this data
- Fully automatic mapping, placement and routing
sheet include the 5 V Spartan family of devices and the
- Interactive design editor for design optimization
3.3 V SpartanXL™ family of devices.
• Next generation ASIC replacement technology Additional SpartanXL Features
- First ASIC replacement FPGA for high-volume • 3.3V supply for low power with 5V tolerant I/Os
production with on-chip RAM • Power down input
- Advanced process technology • Higher performance
- Density up to 1862 logic cells or 40,000 system • Faster carry logic
gates • More flexible high-speed clock network
- Streamlined feature set based on XC4000 • Latch capability in Configurable Logic Blocks
architecture • Input fast capture latch
- System performance beyond 80 MHz • Optional mux or 2-input function generator on outputs
- Broad set of AllianceCORE™ and LogiCORE™ • 12 mA or 24 mA output drive
pre-defined solutions available • 5V/3.3V PCI compatible
- Unlimited reprogrammability • Enhanced Boundary Scan
- Low cost • Express Mode configuration
Table 1: Spartan and SpartanXL Series Field Programmable Gate Arrays
Max Typical Number Max.

Logic System Gate Range CLB Total of Available
Device Cells Gates (Logic and RAM)* Matrix CLBs Flip-Flops User I/O
XCS05 & XCS05XL 238 5,000 2,000 - 5,000 10 x 10 100 360 77
XCS10 & XCS10XL 466 10,000 3,000 - 10,000 14 x 14 196 616 112
XCS20 & XCS20XL 950 20,000 7,000 - 20,000 20 x 20 400 1,120 160
XCS30 & XCS30XL 1368 30,000 10,000 - 30,000 24 x 24 576 1,536 192
XCS40 & XCS40XL 1862 40,000 13,000 - 40,000 28 x 28 784 2,016 205
* Max values of Typical Gate Range include 20-30% of CLBs used as RAM.

R
General Overview shortening design and development cycles, and also offer a
cost-effective solution for production rates well beyond
Spartan Series FPGAs are implemented with a regular, 50,000 systems per month.
flexible, programmable architecture of Configurable Logic
Blocks (CLBs), interconnected by a powerful hierarchy of Spartan Series devices achieve high-performance,
versatile routing resources (routing channels), and sur- low-cost operation through the use of an advanced archi-
rounded by a perimeter of programmable Input/Output tecture and semiconductor technology. Spartan and Spar-
Blocks (IOBs), as seen in Figure 1. They have generous tanXL devices provide system clock rates exceeding
routing resources to accommodate the most complex inter- 80 MHz and internal performance in excess of 150 MHz.
connect patterns. In contrast to other FPGA devices, Spartan FPGAs offer
the most cost-effective solution while maintaining lead-
The devices are customized by loading configuration data ing-edge performance. In addition to the conventional ben-
into internal static memory cells. Re-programming is possi- efit of high volume programmable logic solutions Spartan
ble an unlimited number of times. The values stored in FPGAs also offer on-chip edge-triggered single-port and
these memory cells determine the logic functions and inter- dual-port RAM, clock enables on all flip-flops, fast carry
connections implemented in the FPGA. The FPGA can logic, and many other features.
either actively read its configuration data from an external
serial PROM (Master Serial mode), or the configuration The Spartan Series leverages the highly successful
data can be written into the FPGA from an external device XC4000 architecture with many of that family’s features and
(Slave Serial mode). benefits. Technology advancements have been derived
from the XC4000XLA and XC4000XV process develop-
Spartan FPGAs can be used where hardware must be ments.
adapted to different user applications. FPGAs are ideal for
B-
OSC
IOB
IOB
IOB
IOB
IOB
IOB
IOB
IOB
SCAN
IOB IOB
CLB CLB CLB CLB
IOB IOB
IOB IOB
CLB CLB CLB CLB
IOB IOB
Routing Channels
IOB IOB
CLB CLB CLB CLB
IOB IOB
IOB IOB
CLB CLB CLB CLB
IOB IOB
IOB
IOB
IOB
IOB
IOB
IOB
IOB
IOB
START
RDBK
-UP
VersaRing Routing Channel
Rev 2.0
Figure 1: Basic FPGA Block Diagram

R
Logic Functional Description fied block diagram in Figure 2. There are three look-up
tables (LUT) which are used as logic function generators,
The Spartan Series uses a standard FPGA structure as two flip-flops and two groups of signal steering multiplexers.
shown in Figure 1 on page 4. The FPGA consists of an There are also some more advanced features provided by
array of configurable logic blocks (CLBs) placed in a matrix the CLB which will be covered in the “Advanced Features
of routing channels. The input and output of signals is Description” on page 15.
achieved through a set of input/output blocks (IOBs) form-
ing a ring around the CLBs and routing channels. Function Generators
• CLBs provide the functional elements for implementing Two 16x1 memory look-up tables (F-LUT and G-LUT) are
the user’s logic. used to implement 4-input function generators, each offer-
• IOBs provide the interface between the package pins ing unrestricted logic implementation of any Boolean func-
and internal signal lines. tion of up to four independent input signals (F1 to F4 or G1
• Routing channels provide paths to interconnect the to G4). Using memory look-up tables the propagation delay
inputs and outputs of the CLBs and IOBs. is independent of the function implemented.
The functionality of each circuit block is customized during A third 3-input function generator (H-LUT) can implement
configuration by programming internal static memory cells. any Boolean function of its three inputs. Two of these inputs
The values stored in these memory cells determine the are controlled by programmable multiplexers (see box “A”
logic functions and interconnections implemented in the of Figure 2). These inputs can come from the F-LUT or
4
FPGA. G-LUT outputs or from CLB inputs. The third input always
comes from a CLB input. The CLB can, therefore, imple-
Configurable Logic Blocks (CLBs) ment certain functions of up to nine inputs, like parity
The CLBs are used to implement most of the logic in an checking. The three LUTs in the CLB can also be combined
FPGA. The principal CLB elements are shown in the simpli- to do any arbitrarily defined Boolean function of five inputs.
B
G-LUT
G4 G4 SR
D Q YQ
Logic
G3 G3 Function
of G CK
G2 G2 G1-G4
EC
G1 G1 H-LUT
SR G Y
Logic
Function
H1 H1 of H
F,G,H1
DIN F
F4 F4 SR
Logic A D Q XQ
F3 F3 Function
of F CK
F2 F2 F1-F4
EC
F1 F1
F-LUT X
Multiplexer Controlled
K by Configuration Program
EC
Rev 1.0
Figure 2: Spartan Simplified CLB Logic Diagram (some features not shown)

R
A CLB can implement any of the following functions:

SR
• Any function of up to four variables, plus any second
function of up to four unrelated variables, plus any third
function of up to three unrelated variables1
• Any single function of five variables
• Any function of four variables together with some GND
functions of six variables
• Some functions of up to nine variables. GSR
Implementing wide functions in a single block reduces both
the number of blocks required and the delay in the signal SD
path, achieving both increased capacity and speed. D D Q Q
The versatility of the CLB function generators significantly

improves system speed. In addition, the design-software CK
tools can deal with each function generator independently.
This flexibility improves cell usage. RD
EC
Flip-Flops
Vcc
Each CLB contains two flip-flops that can be used to regis- Rev 1.1
ter (store) the function generator outputs. The flip-flops and Multiplexer Controlled
function generators can also be used independently (see by Configuration Program
Figure 2 on page 5). The CLB input DIN can be used as a
direct input to either of the two flip-flops. H1 can also drive Figure 3: CLB Flip-Flop Functional Block Diagram
either flip-flop via the H-LUT with a slight additional delay.
The two flip-flops have common clock (CK), clock enable Clock Input
(EC) and set/reset (SR) inputs. Internally both flip-flops are
Each flip-flop can be triggered on either the rising or falling
also controlled by a global initialization signal (GSR) which
clock edge. The CLB clock line is shared by both flip-flops.
is described in detail in “Global Signals: GSR and GTS” on
However, the clock is individually invertible for each flip-flop
page 21.
(see CK path in Figure 3). Any inverter placed on the clock
Latches (SpartanXL only) line in the design is automatically absorbed into the CLB.
The SpartanXL CLB storage elements can also be config- Clock Enable
ured as latches. The two latches have common clock (K)
The clock enable line (EC) is active High. The EC line is
and clock enable (EC) inputs. Functionality of the storage
shared by both flip-flops in a CLB. If either one is left dis-
element is described in Table 2.
connected, the clock enable for that flip-flop defaults to the
Table 2: CLB Storage Element Functionality active state. EC is not invertible within the CLB. The clock
enable is synchronous to the clock and must satisfy the
Mode CK EC SR D Q setup and hold timing specified for the device.
Power-Up or
X X X X SR
GSR Set/Reset
X X 1 X SR The set/reset line (SR) is an asynchronous active High con-
Flip-Flop
__/ 1* 0* D D trol of the flip-flop. SR can be configured as either set or
Operation
0 X 0* X Q reset at each flip-flop. This configuration option determines
Latch Operation 1 1* 0* X Q the state in which each flip-flop becomes operational after
(SpartanXL) configuration. It also determines the effect of a GSR pulse
0 1* 0* D D
during normal operation, and the effect of a pulse on the
Both X 0 0* X Q
SR line of the CLB. The SR line is shared by both flip-flops.
Legend:
If SR is not specified for a flip-flop the set/reset for that
X Don’t care
__/ Rising edge (clock not inverted) flip-flop defaults to the inactive state. SR is not invertible
SR Set or Reset value. Reset is default. within the CLB.
0* Input is Low or unconnected (default value)
1* Input is High or unconnected (default value)
1. When three separate functions are generated, one of the function outputs must be captured in a flip-flop internal to the CLB. Only two
unregistered function generator outputs are available from the CLB.

R
CLB Signal Flow Control Input/Output Blocks (IOBs)

In addition to the H-LUT input control multiplexers (shown User-configurable input/output blocks (IOBs) provide the
in box “A” of Figure 2 on page 5) there are signal flow con- interface between external package pins and the internal
trol multiplexers (shown in box “B” of Figure 2) which select logic. Each IOB controls one package pin and can be con-
the signals which drive the flip-flop inputs and the combina- figured for input, output, or bidirectional signals. Figure 5 on
torial CLB outputs (X and Y). page 8 shows a simplified functional block diagram of the
Each flip-flop input is driven from a 4:1 multiplexer which Spartan IOB.
selects among the three LUT outputs and DIN as the data IOB Input Signal Path
source.
The input signal to the IOB can be configured to either go
Each combinatorial output is driven from a 2:1 multiplexer directly to the routing channels (via I1 and I2 in Figure 5) or
which selects between two of the LUT outputs. The X out- to the input register. The input register can be programmed
put can be driven from the F-LUT or H-LUT, the Y output
as either an edge-triggered flip-flop or a level-sensitive
from G-LUT or H-LUT.
latch. The functionality of this register is shown in Table 3,
Control Signals and a simplified block diagram of the register can be seen
in Figure 6.
There are four signal control multiplexers on the input of the
CLB. These multiplexers allow the internal CLB control sig- Table 3: Input Register Functionality
nals (H1, DIN, SR, and EC in Figure 2 and Figure 4) to be
4
Mode CK EC D Q
driven from any of the four general control inputs (C1 - C4
Power-Up or X X X SR
in Figure 4) into the CLB. Any of these inputs can drive any
GSR
of the four internal control signals.
Flip-Flop __/ 1* D D
The four internal control signals are: 0 X X Q
• EC - Enable Clock Latch 1 1* X Q
• SR - Asynchronous Set/Reset or H function generator 0 1* D D
Input 0
Both X 0 X Q
• DIN - Direct In or H function generator Input 2
Legend:
• H1 - H function generator Input 1. X Don’t care
__/ Rising edge (clock not inverted)
SR Set or Reset value. Reset is default.
DIN
GSR
H1
SD
C1 D D Q Q
C2
SR CK
C3
RD
C4
EC
Vcc
EC
Rev 1.1
Rev 1.1
by Configuration Program
Figure 6: IOB Flip-Flop/Latch Functional Block
Figure 4: CLB Control Signal Interface Diagram

R
The register choice is made by placing the appropriate exit the IOB can each carry either the direct or registered
library symbol. For example, IFD is the basic input flip-flop input signal.
(rising edge triggered), and ILD is the basic input latch
The 5V Spartan input buffers can be globally configured for
(transparent-High). Variations with inverted clocks are also either TTL (1.2 V) or CMOS (Vcc/2) thresholds, using an
available. The clock signal inverter is also shown in option in the bitstream generation software. The Spartan
Figure 6 on the CK line. output levels are also configurable; the two global adjust-
The Spartan IOB data input path has a one-tap delay elements of input threshold and output level are independent.
ment: either the delay is inserted (default), or it is not. The The inputs of Spartan devices can be driven by the outputs
SpartanXL IOB data input path has a two-tap delay ele- of any 3.3 V device, if the Spartan inputs are in TTL mode.
ment, with choices of a full delay, a partial delay, or no There is a slight input hysteresis of about 300 mV. Spar-
delay. The added delay guarantees a zero hold time with tanXL inputs are TTL compatible and 3.3 V CMOS compat-
respect to clocks routed through the Spartan global clock ible.
buffers. (See “Global Nets and Buffers” on page 14 for a
Supported sources for Spartan Series device inputs are
description of the global clock buffers in the Spartan shown in Table 4.
Series.) For a shorter input register setup time, with positive
hold-time, attach a NODELAY attribute or property to the SpartanXL I/Os are fully 5V tolerant even though the VCC
flip-flop. is 3.3 Volts. This allows 5V signals to directly connect to the
SpartanXL inputs without damage, as shown in Table 4. In
The output of the input register goes to the routing chan- addition, the 3.3 Volt VCC can be applied before or after 5V
nels (via I1 and I2 in Figure 5). The I1 and I2 signals that
signals are applied to the I/Os. This makes the SpartanXL
devices immune to power supply sequencing problems.
GTS
O D Q
OUTPUT DRIVER
CK Programmable Slew Rate
OK Programmable TTL/CMOS Drive
EC
Package
Pad
I1
INPUT BUFFER
I2 Delay
D Q
Programmable
Pull-Up/
IK CK
Pull-Down
Network
EC EC Multiplexer Controlled
Rev 1.1
Figure 5: Simplified Spartan IOB Block Diagram

R
SpartanXL VCC Clamping

Table 4: Supported Sources for Spartan Series Inputs
SpartanXL FPGAs have an optional clamping diode con-
Spartan SpartanXL nected from each I/O to VCC. When enabled they clamp
Inputs Inputs ringing transients back to the 3.3 V supply rail. This clamp-
Source 5.0 V, 5.0 V, 3.3 V ing action is required in 3.3 V PCI applications. VCC clamp-
TTL CMOS CMOS ing is a global option affecting all I/O pins.
Any device, Vcc = 3.3 V, SpartanXL devices are fully 5V TTL I/O compatible if VCC
√ √
CMOS outputs clamping is not enabled. With VCC clamping enabled, the
Unreli-
Spartan Series, Vcc = 5 V, SpartanXL devices will begin to clamp input voltages to one
√ able √
TTL outputs diode voltage drop above VCC. If enabled, TTL I/O compat-
Data
Any device, Vcc = 5 V, ibility is maintained but full 5V I/O tolerance is sacrificed.
√ √ The user may select either 5V tolerance (default) or 3.3 V
TTL outputs (Voh ≤ 3.7 V)
PCI compatibility. In both cases negative voltage is
Any device, Vcc = 5 V, √ (default
√ √ clamped to one diode voltage drop below ground.
CMOS outputs mode)
SpartanXL devices are compatible with TTL, LVTTL, PCI
3V, PCI 5V and LVCMOS signalling. The various standards
are illustrated in Table 5.
4
Table 5: I/O Standards Supported by SpartanXL FPGAs
Signaling VCC
Standard Clamping Output Drive VIH MAX VIH MIN VIL MAX VOH MIN VOL MAX
TTL Not allowed 12/24 mA 5.5 2.0 0.8 2.4 0.4
LVTTL OK 12/24 mA 3.6 2.0 0.8 2.4 0.4
PCI5V Not allowed 24 mA 5.5 2.0 0.8 2.4 0.4
PCI3V Required 12 mA 3.6 50% of VCC 30% of VCC 90% of VCC 10% of VCC
LVCMOS 3V OK 12/24 mA 3.6 50% of VCC 30% of VCC 90% of VCC 10% of VCC
Additional Fast Capture Input Latch (SpartanXL only) Table 6: Output Flip-Flop Functionality
The SpartanXL IOB has an additional optional latch on the Clock

input. This latch is clocked by the clock used for the output Mode Clock Enable T D Q
flip-flop rather than the input clock. Therefore, two different Power-Up X X 0* X SR
clocks can be used to clock the two input storage elements. or GSR
This additional latch allows the fast capture of input data, X 0 0* X Q
which is then synchronized to the internal clock by the IOB
Flip-Flop __/ 1* 0* D D
flip-flop or latch.
X X 1 X Z
To place the Fast Capture latch in a design, use one of the 0 X 0* X Q
special library symbols, ILFFX or ILFLX. ILFFX is a trans- Legend:
parent-Low Fast Capture latch followed by an active-High X Don’t care
input flip-flop. ILFLX is a transparent-Low Fast Capture __/ Rising edge (clock not inverted)
latch followed by a Transparent-High input latch. Any of the SR Set or Reset value. Reset is default.
clock inputs can be inverted before driving the library ele- 1* Input is High or unconnected (default value)
ment, and the inverter is absorbed into the IOB. Z 3-state
IOB Output Signal Path Output Multiplexer/2-Input Function Generator

Output signals can be optionally inverted within the IOB, (SpartanXL only)
and can pass directly to the output buffer or be stored in an The output path in the SpartanXL IOB contains an addi-
edge-triggered flip-flop and then to the output buffer. The tional multiplexer not available in the Spartan IOB. The mul-
functionality of this flip-flop is shown in Table 6. tiplexer can also be configured as a 2-input function
generator, implementing a pass gate, AND gate, OR gate,
or XOR gate, with 0, 1, or 2 inverted inputs.

R
When configured as a multiplexer, this feature allows two Output Slew Rate
output signals to time-share the same output pad; effec-
The slew rate of each output buffer is, by default, reduced,
tively doubling the number of device outputs without requir-
to minimize power bus transients when switching non-criti-
ing a larger, more expensive package. The select input is
cal signals. For critical signals, attach a FAST attribute or
the pin used for the output flip-flop clock, OK.
property to the output buffer or flip-flop.
When the multiplexer is configured as a 2-input function
Table 7: Supported Destinations
generator, logic can be implemented within the IOB itself.
for Spartan Series Outputs
Combined with a Global buffer, this arrangement allows
very high-speed gating of a single signal. For example, a SpartanXL Spartan
wide decoder can be implemented in CLBs, and its output Outputs Outputs
gated with a Read or Write Strobe Driven by a Global Destination 3.3 V, 5.0 V, 5.0 V,
buffer. CMOS TTL CMOS
The user can specify that the IOB function generator be Any device, Vcc = 3.3 V, √ √ Some1
used by placing special library symbols beginning with the CMOS-threshold inputs
letter “O.” For example, a 2-input AND gate in the IOB func- Any device, Vcc = 5.0 V, √ √ √
tion generator is called OAND2. Use the symbol input pin TTL-threshold inputs
labelled “F” for the signal on the critical path. This signal is Any device, Vcc = 5.0 V, Unreliable √
placed on the OK pin — the IOB input with the shortest CMOS-threshold inputs Data
delay to the function generator. Two examples are shown in
Figure 7. 1. Only if destination device has 5-V tolerant inputs
Spartan Series devices have a feature called “Soft

OMUX2
F D0 Start-up,” designed to reduce ground bounce when all out-
O
D1
puts are turned on simultaneously at the end of configura-
tion. When the configuration process is finished and the
OAND2 S0 device starts up, the first activation of the outputs is auto-
X6598
X6599
matically slew-rate limited. Immediately following the initial
Figure 7: AND & MUX Symbols in SpartanXL IOB activation of the I/O, the slew rate of the individual outputs
is determined by the individual configuration option for each
Output Buffer IOB.
An active-High 3-state signal can be used to place the out-
put buffer in a high-impedance state, implementing 3-state Pull-up and Pull-down Network
outputs or bidirectional I/O. Under configuration control, the Programmable pull-up and pull-down resistors are used for
output (O) and output 3-state (T) signals can be inverted. tying unused pins to Vcc or Ground to minimize power con-
The polarity of these signals is independently configured sumption and reduce noise sensitivity. The configurable
for each IOB (see Figure 5 on page 8). pull-up resistor is a p-channel transistor that pulls to Vcc.
By default, a 5V Spartan device output buffer pull-up struc- The configurable pull-down resistor is an n-channel transis-
ture is configured as a TTL-like totem-pole. The High driver tor that pulls to Ground. The value of these resistors is typ-
is an n-channel pull-up transistor, pulling to a voltage one ically 20 kΩ − 100 kΩ (See “Spartan DC Characteristics
transistor threshold below Vcc. Alternatively, the outputs Over Operating Conditions” on page 39.). This high value
can be globally configured as CMOS drivers, with addi- makes them unsuitable as wired-AND pull-up resistors.
tional p-channel pull-up transistors pulling to Vcc. This After configuration, voltage levels of unused pads, bonded
option, applied using the bitstream generation software, or unbonded, must be valid logic levels, to reduce noise
applies to all outputs on the device. It is not individually pro- sensitivity and avoid excess current. Therefore, by default,
grammable. unused pads are configured with the internal pull-up resis-
All SpartanXL device outputs are configured as CMOS tor active. Alternatively, they can be individually configured
drivers, therefore driving rail-to-rail. The SpartanXL outputs with the pull-down resistor, or as a driven output, or to be
are individually programmable for 12 mA or 24 mA output driven by an external source. To activate the internal
drive. pull-up, attach the PULLUP library component to the net
attached to the pad. To activate the internal pull-down,
Any 5V Spartan device with its outputs configured in TTL attach the PULLDOWN library component to the net
mode can drive the inputs of any typical 3.3 V device. (For attached to the pad.
a detailed discussion of how to interface between 5.0 V and
3.3 V devices, see the 3V Products section of The Pro-
grammable Logic Data Book.) Supported destinations for
Spartan Series device outputs are shown in Table 7.

R
Set/Reset Routing Channel Description

As with the CLB registers, the GSR signal can be used to All internal routing channels are composed of metal seg-
set or clear the input and output registers, depending on ments with programmable switching points and switching
the value of the INIT attribute or property. The two flip-flops matrices to implement the desired routing. A structured,
can be individually configured to set or clear on reset and hierarchical matrix of routing channels is provided to
after configuration. Other than the global GSR net, no achieve efficient automated routing.
user-controlled set/reset signal is available to the I/O
This section describes the routing channels available in
flip-flops (Figure 6). The choice of set or reset applies to
Spartan Series devices. Figure 8 shows a general block
both the initial state of the flip-flop and the response to the
diagram of the CLB routing channels. The implementation
GSR pulse.
software automatically assigns the appropriate resources
Independent Clocks based on the density and timing requirements of the
design. The following description of the routing channels is
Separate clock signals are provided for the input (IK) and
for information only and is simplified with some minor
output (OK) flip-flops. The clock can be independently
details omitted. For an exact interconnect description the
inverted for each flip-flop within the IOB, generating either
designer should open a design in the EPIC design editor
falling-edge or rising-edge triggered flip-flops. The clock
and review the actual connections in this tool.
inputs for each IOB are independent.
The routing channels will be discussed as follows;
Common Clock Enables 4
• CLB routing channels which run along each row and
The input and output flip-flops in each IOB have a common column of the CLB array.
clock enable input (see EC signal in Figure 6), which • IOB routing channels which form a ring (called a
through configuration, can be activated individually for the VersaRing) around the outside of the CLB array. It
input or output flip-flop, or both. This clock enable operates connects the I/O with the CLB routing channels.
exactly like the EC signal on the Spartan Series CLB. It • Global routing consists of dedicated networks primarily
cannot be inverted within the IOB. designed to distribute clocks throughout the device with
minimum delay and skew. Global routing can also be
Three-State Register (SpartanXL Only)
used for other high-fanout signals.
SpartanXL devices incorporate an optional register control-
ling the three-state enable in the IOBs. The use of the
three-state control register can significantly improve output
enable and disable time.

R
PSM PSM PSM 8 Singles
3 Longs
CLB CLB
2 Doubles
PSM PSM PSM
Rev 1.1
2 Doubles 3 Longs 8 Singles 3 Longs 2 Doubles
Figure 8: Spartan Series CLB Routing Channels and Interface Block Diagram

R
CLB Routing Channels
YQ
G4
C4
F4
The routing channels around the CLB are derived from
three types of interconnects; single-length, double-length,
and longlines. At the intersection of each vertical and hori-
zontal routing channel is a signal steering matrix called a CIN Y
Programmable Switch Matrix (PSM). Figure 8 shows the
COUT G3
basic routing channel configuration showing single-length
lines, double-length lines and longlines as well as the CLBs G1
and PSMs. The CLB to routing channel interface is shown C3
as well as how the PSMs interface at the channel intersec-
tions.
C1
CLB
K
CLB Interface
F3
A block diagram of the CLB interface signals is shown in
F1
Figure 9. The input signals to the CLB are distributed
evenly on all four sides providing maximum routing flexibil- X
Rev 1.1
ity. In general, the entire architecture is symmetrical and
regular. It is well suited to established placement and rout- 4
F2
XQ
ing algorithms. Inputs, outputs, and function generators can
G2
C2
freely swap positions within a CLB to avoid routing conges-
tion during the placement and routing operation. The
Figure 9: CLB Interconnect Signals
exceptions are the clock (K) input and CIN/COUT signals.
The K input is routed to dedicated global vertical lines as For example, a single-length signal entering on the right
well as 4 single-length lines and is on the left side of the side of the switch matrix can be routed to a single-length
CLB. The CIN/COUT signals are routed through dedicated line on the top, left, or bottom sides, or any combination
interconnects which do not interfere with the general rout- thereof, if multiple branches are required. Similarly, a dou-
ing structure. The output signals from the CLB are available ble-length signal can be routed to a double-length line on
to drive both vertical and horizontal channels. any or all of the other three edges of the programmable
switch matrix.
Programmable Switch Matrices
The horizontal and vertical single- and double-length lines Single-Length Lines
intersect at a box called a programmable switch matrix Single-length lines provide the greatest interconnect flexi-
(PSM). Each PSM consists of programmable pass transis- bility and offer fast routing between adjacent blocks. There
tors used to establish connections between the lines (see are eight vertical and eight horizontal single-length lines
Figure 10). associated with each CLB. These lines connect the switch-
ing matrices that are located in every row and column of
CLBs.
Six Pass Transistors Per

Switch Matrix Interconnect Point
Figure 10: Programmable Switch Matrix

R
Single-length lines are connected by way of the program- Routing connectivity of the longlines is shown in Figure 8.
mable switch matrices, as shown in Figure 10. Routing The longlines also interface to some 3-state buffers which
connectivity is shown in Figure 8. is described later in “3-State Long Line Drivers” on
page 20.
Single-length lines incur a delay whenever they go through
a PSM. Therefore, they are not suitable for routing signals I/O Routing
for long distances. They are normally used to conduct sig-
nals within a localized area and to provide the branching for Spartan Series devices have additional routing around the
nets with fanout greater than one. IOB ring. This routing is called a VersaRing. The VersaRing
facilitates pin-swapping and redesign without affecting
Double-Length Lines board layout. Included are eight double-length lines, and
The double-length lines consist of a grid of metal segments, four longlines.
each twice as long as the single-length lines: they run past
Global Nets and Buffers
two CLBs before entering a PSM. Double-length lines are
grouped in pairs with the PSMs staggered, so that each line The Spartan Series devices have dedicated global net-
goes through a PSM at every other row or column of CLBs works. These networks are designed to distribute clocks
(see Figure 8). and other high fanout control signals throughout the
devices with minimal skew.
There are four vertical and four horizontal double-length
lines associated with each CLB. These lines provide faster Four vertical longlines in each CLB column are driven
signal routing over intermediate distances, while retaining exclusively by special global buffers. These longlines are in
routing flexibility. addition to the vertical longlines used for standard intercon-
nect. In the 5V Spartan devices, the four global lines can be
Longlines driven by either of two types of global buffers; Primary Glo-
Longlines form a grid of metal interconnect segments that bal buffers (BUFGP) or Secondary Global buffers
run the entire length or width of the array. Longlines are (BUFGS). Each of these lines can be accessed by one par-
intended for high fan-out, time-critical signal nets, or nets ticular Primary Global buffer, or by any of the Secondary
that are distributed over long distances. Global buffers, as shown in Figure 11. In the 3V SpartanXL
devices, the four global lines can be driven by any of the
Each Spartan Series longline has a programmable splitter
eight Global Low-Skew Buffers (BUFGLS). The clock pins
switch at its center. This switch can separate the line into
of every CLB and IOB can also be sourced from local inter-
two independent routing channels, each running half the
connect.
width or height of the array.
IOB IOB IOB IOB
locals
locals
locals
locals
BUFGS BUFGP
PGCK1 SGCK4
SGCK1 PGCK4
4
4
BUFGP BUFGS
4
4 locals locals
CLB CLB
IOB IOB
locals locals
X4 Any BUFGS X4 X4 Any BUFGS X4
locals locals
One BUFGP One BUFGP
IOB per Global Line per Global Line IOB
locals CLB CLB locals
BUFGS BUFGP
PGCK2 SGCK3
SGCK2 PGCK3
locals
locals
locals
locals
BUFGP BUFGS
IOB IOB IOB IOB X6604
Figure 11: 5V Spartan Series Global Net Distribution

R
The four Primary Global buffers offer the shortest delay and preceding description). There is one data input, one
negligible skew. Four Secondary Global buffers have data output and one address decoder for each array.
slightly longer delay and slightly more skew due to poten- These arrays can be addressed independently.
tially heavier loading, but offer greater flexibility when used • The 32 x 1 Single-Port configuration contains a RAM
to drive non-clock CLB inputs. The eight Global Low-Skew array with 32 locations, each one-bit wide. There is one
buffers in the SpartanXL devices combine short delay, neg- data input, one data output, and one 5-bit address
ligible skew, and flexibility. decoder.
• The Dual-Port mode 16 x 1 configuration contains a
The Primary Global buffers must be driven by the
RAM array with 16 locations, each one-bit wide. There
semi-dedicated pads (PGCK1-4). The Secondary Global
are two 4-bit address decoders, one for each port. One
buffers can be sourced by either semi-dedicated pads
port consists of an input for writing and an output for
(SGCK1-4) or internal nets. Each corner of the device has
reading, all at a selected address. The other port
one Primary buffer and one Secondary buffer. The Spar-
consists of one output for reading from an
tanXL family has eight global low-skew buffers, two in each
independently selected address.
corner. All can be sourced by either semi-dedicated pads
(GCK1-8) or internal nets. Table 8: CLB Memory Configurations
Using the library symbol called BUFG results in the soft-
ware choosing the appropriate clock buffer, based on the Mode 16 x 1 (16 x 1) x 2 32 x 1
timing requirements of the design. A global buffer should be Single-Port √ √ √ 4
specified for all timing-sensitive global signal distribution. Dual-Port √
To use a global buffer, place a BUFGP (primary buffer),
BUFGS (secondary buffer), BUFGLS (SpartanXL global The appropriate choice of RAM configuration mode for a
low-skew buffer), or BUFG (any buffer type) element in a given design should be based on timing and resource
schematic or in HDL code. requirements, desired functionality, and the simplicity of the
design process. Selection criteria include the following:
Advanced Features Description Whereas the 32 x 1 Single-Port, the (16 x 1) x 2 Single-Port
and the 16 x 1 Dual-Port configurations each use one entire
Distributed RAM CLB, the 16 x 1 Single-Port configuration uses only one half
Optional modes for each CLB allow the function generators of a CLB. Due to its simultaneous read/write capability, the
(F-LUT and G-LUT) to be used as Random Access Mem- Dual-Port RAM can transfer twice as much data as the Sin-
ory (RAM). gle-Port RAM, which permits only one data operation at
any given time.
Read and write operations are significantly faster for this
on-chip RAM than for off-chip implementations. This speed CLB memory configuration options are selected by using
advantage is due to the relatively short signal propagation the appropriate library symbol in the design entry.
delays within the FPGA.
Single-Port Mode
Memory Configuration Overview
There are three CLB memory configurations for the Sin-
There are two available memory configuration modes: single-Port RAM: 16 x 1, (16 x 1) x 2, and 32 x 1, the functional
gle-port RAM and dual-port RAM. For both these modes, organization of which is shown in Figure 12.
write operations are synchronous (edge-triggered), while
read operations are asynchronous. In the Single-Port The Single-Port RAM signals and the CLB signals
Mode, a single CLB can be configured as either a 16 x 1, (Figure 2 on page 5) from which they are originally derived
(16 x 1) x 2 or 32 x 1 RAM array. In the Dual-Port mode, a are shown in Table 9.
single CLB can be configured only as one 16 x 1 RAM Table 9: Single-Port RAM Signals
array. The different CLB memory configurations are sum-
marized in Table 8. Any of these possibilities can be individ- RAM Signal Function CLB Signal
ually programmed into a Spartan Series CLB. D Data In DIN or H1
A[3:0] Address F1-F4 or G1-G4
• The 16 x 1 Single-Port configuration contains a RAM
array with 16 locations, each one-bit wide. One 4-bit A4 (32 x 1 only) Address H1
address decoder determines the RAM location for write WE Write Enable SR
and read operations. There is one input for writing data WCLK Clock K
and one output for reading data, all at the selected SPO Single Port Out FOUT or GOUT
address. (Data Out)
• The (16 x 1) x 2 Single-Port configuration combines two
16 x 1 Single Port configurations (each according to the

R
WRITE ROW
READ ROW
●
●
SELECT
SELECT
16 x 1
A[n-1:0] 32 x 1
RAM ARRAY
INPUT REGISTER
n ●
●
WE WRITE READ SPO

CONTROL OUT
D0 or D1
WCLK ●
Figure 12: Logic Diagram for the Single-Port RAM

NOTE: 1. The (16 x 1) x 2 configuration combines two 16 x 1 Single Port RAMs, each with its own independent address bus and
data input. The same WE and WCLK signals are connected to both RAMs.
2. n = 4 for the 16 x 1 and (16 x 1) x 2 configurations. n = 5 for the 32 x 1 configuration.
Writing data to the Single-Port RAM is essentially the same WCLK can be configured as active on either the rising edge
as writing to a data register. It is an edge-triggered (syn- (default) or the falling edge. While the WCLK input to the
chronous) operation performed by applying an address to RAM accepts the same signal as the clock input to the
the A inputs and data to the D input during the active edge associated CLB’s flip-flops, the sense of this WCLK input
of WCLK while WE is High. can be inverted with respect to the sense of the flip-flop
clock inputs. Consequently, within the same CLB, data at
The timing relationships are shown in Figure 13. The High
the RAM’s SPO line can be stored in a flip-flop with either
logic level on WE enables the input data register for writing.
the same or the inverse clock polarity used to write data to
The active edge of WCLK latches the address, input data,
the RAM.
and WE signals. Then, an internal write pulse is generated
that loads the data into the memory cell. The WE input is active-High and cannot be inverted within
the CLB.
TWPS Allowing for settling time, the data on the SPO output
reflects the contents of the RAM location currently
WCLK (K)
addressed. When the address changes, following the asyn-
TWSS TWHS chronous delay TILO, the data stored at the new address
location will appear on SPO. If the data at a particular RAM
WE
address is overwritten, after the delay TWOS, the new data
TDSS TDHS will appear on SPO.
DATA IN
Dual-Port Mode
In dual-port mode, the function generators (F-LUT and
TASS TAHS G-LUT) are used to create a 16 x 1 Dual-Port memory. Of
the two data ports available, one permits read and write
ADDRESS
operations at the address specified by A[3:0] while the sec-
ond provides only for read operations at the address spec-
TILO TILO
TWOS
ified independently by DPRA[3:0]. As a result,
simultaneous read/write operations at different addresses
DATA OUT OLD NEW
(or even at the same address) are supported.
The functional organization of the 16 x 1 Dual-Port RAM is
X6461
shown in Figure 14.
Figure 13: Data Write and Access Timing for RAM

R
WRITE ROW
●
READ ROW
SELECT
●
SELECT
16 x 1
A[3:0]
RAM
INPUT REGISTER
4 4 ●
●
WE ● WRITE READ SPO

● CONTROL OUT
D
●
WCLK ●
4
WRITE ROW
READ ROW
●
SELECT
●
SELECT
16 x 1
DPRA[3:0]
RAM
● 4
●
WRITE READ DPO

CONTROL OUT
Figure 14: Logic Diagram for the Dual-Port RAM
The Dual-Port RAM signals and the CLB signals from for the lower memory. Therefore, SPO reflects the data at
which they are originally derived are shown in Table 10. address A[3:0].
Table 10: Dual-Port RAM Signals The other address port, labeled DPRA[3:0] for Dual Port
Read Address, supplies the read address for the upper
CLB
RAM Signal Function memory. The write address for this memory, however,
Signal
comes from the address A[3:0]. Dual Port Out (DPO)
D Data In DIN
serves as the data output for the upper memory. Therefore,
A[3:0] Read Address for Single-Port. F1-F4
DPO reflects the data at address DPRA[3:0].
Write Address for Single-Port
and Dual-Port. By using A[3:0] for the write address and DPRA[3:0] for the
DPRA[3:0] Read Address for Dual-Port G1-G4 read address, and reading only the DPO output, a FIFO
WE Write Enable SR that can read and write simultaneously is easily generated.
WCLK Clock K The simultaneous read/write capability possible with the
SPO Single Port Out FOUT Dual-Port RAM can provide twice the effective data
(addressed by A[3:0]) throughput of a Single-Port RAM alternating read and write
DPO Dual Port Out GOUT operations.
(addressed by DPRA[3:0])
The timing relationships for the Dual-Port RAM mode are
The RAM16X1D primitive used to instantiate the Dual-Port shown in Figure 13.
RAM consists of an upper and a lower 16 x 1 memory array.
Note that write operations to RAM are synchronous
The address port labeled A[3:0] supplies both the read and
(edge-triggered); however, data access is asynchronous.
write addresses for the lower memory array, which behaves
the same as the 16 x 1 Single-Port RAM array described
previously. Single Port Out (SPO) serves as the data output

R
Initializing RAM at FPGA Configuration

Both RAM and ROM implementations of the Spartan CLB CLB CLB CLB
Series are initialized during device configuration. The initial
contents are defined via an INIT attribute or property
attached to the RAM or ROM symbol, as described in the
schematic library guide. If not defined, all RAM contents
are initialized to zeros, by default. CLB CLB CLB CLB
RAM initialization occurs only during device configuration.

The RAM content is not affected by GSR.
More Information on using RAM inside CLBs
CLB CLB CLB CLB
Three application notes are available from Xilinx that dis-
cuss synchronous (edge-triggered) RAM: “Xilinx Edge-Trig-
gered and Dual-Port RAM Capability,” “Implementing
FIFOs in Xilinx RAM,” and “Synchronous and Asynchro-
nous FIFO Designs.” All three application notes apply to CLB CLB CLB CLB
both the Spartan and the SpartanXL Series.
Fast Carry Logic X6610
Each CLB F-LUT and G-LUT contains dedicated arithmetic

Figure 15: Available Spartan Carry Propagation
logic for the fast generation of carry and borrow signals.
Paths
This extra output is passed on to the function generator in
the adjacent CLB. The carry chain is independent of nor-
The carry chain in Spartan devices can run either up or
mal routing resources. (See Figure 15.)
down. At the top and bottom of the columns where there
Dedicated fast carry logic greatly increases the efficiency are no CLBs above and below, the carry is propagated to
and performance of adders, subtractors, accumulators, the right. The default is always to propagate up the column,
comparators and counters. It also opens the door to many as shown in the figures. The carry chain in SpartanXL
new applications involving arithmetic operation, where the devices can only run up the column, providing even higher
previous generations of FPGAs were not fast enough or too speed.
inefficient. High-speed address offset calculations in micro-
Figure 16 on page 19 shows a Spartan Series CLB with
processor or graphics systems, and high-speed addition in
dedicated fast carry logic. The carry logic shares operand
digital signal processing are two typical applications.
and control inputs with the function generators. The carry
The two 4-input function generators can be configured as a outputs connect to the function generators, where they are
2-bit adder with built-in hidden carry that can be expanded combined with the operands to form the sums.
to any length. This dedicated carry circuitry is so fast and
Figure 17 on page 20 shows the details of the Spartan
efficient that conventional speed-up methods like carry
carry logic. This diagram shows the contents of the box
generate/propagate are meaningless even at the 16-bit
labeled “CARRY LOGIC” in Figure 16.
level, and of marginal benefit at the 32-bit level. This fast
carry logic is one of the more significant features of the The fast carry logic can be accessed by placing special
Spartan Series, speeding up arithmetic and counting func- library symbols, or by using Xilinx Relationally Placed Mac-
tions. ros (RPMs) that already include these symbols.

R
C OUT D IN
CARRY
LOGIC
Y
G H
CARRY
G4
G3
G
G2 DIN
H S/R
G D Q YQ
F 4
G1
EC
COUT0
H1 H
DIN
F H S/R
CARRY G D Q XQ
F
EC
F4
F3
F
F2
H
F1
X
F
K S/R EC
CIN
S6699_01
Figure 16: Fast Carry Logic in Spartan CLB

R
C OUT
M
G1
M
1
0 1 G2
I 0
G4
G3
C OUT0
TO
M FUNCTION
GENERATORS
F2
M
1
0 1
F1
M 0
F4
M 0 1
M 3
F3 1 M
M 0
M C IN
S2000_01
Figure 17: Detail of Spartan Dedicated Carry Logic
3-State Long Line Drivers Three-State Buffer Example

A pair of 3-state buffers is associated with each CLB in the Figure 18 shows how to use the 3-state buffers to imple-
array. These 3-state buffers (BUFT) can be used to drive ment a multiplexer. The selection is accomplished by the
signals onto the nearest horizontal longlines above and buffer 3-state signal.
below the CLB. They can therefore be used to implement Pay particular attention to the polarity of the T pin when
multiplexed or bidirectional buses on the horizontal lon- using these buffers in a design. Active-High 3-state (T) is
glines, saving logic resources. identical to an active-Low output enable, as shown in
There is a weak keeper at each end of these two horizontal Table 11.
longlines. This circuit prevents undefined floating levels. Table 11: Three-State Buffer Functionality
However, it is overridden by any driver.
IN T OUT
The buffer enable is an active-High 3-state (i.e. an
active-Low enable), as shown in Table 11. X 1 Z
IN 0 IN
Z = DA • A + D B • B + D C • C + D N • N
~100 kΩ
DA DB DC DN
BUFT BUFT BUFT BUFT
A B C N
X6466
"Weak Keeper"
Figure 18: 3-State Buffers Implement a Multiplexer

R
On-Chip Oscillator Global 3-State

Spartan Series devices include an internal oscillator. This A separate Global 3-State line (GTS) as shown in Figure 5
oscillator is used to clock the power-on time-out, for config- on page 8 forces all FPGA outputs to the high-impedance
uration memory clearing, and as the source of CCLK in state, unless boundary scan is enabled and is executing an
Master configuration mode. The oscillator runs at a nominal EXTEST instruction. GTS does not compete with other
8 MHz frequency that varies with process, Vcc, and tem- routing resources; it uses a dedicated distribution network.
perature. The output frequency falls between 4 MHz and GTS can be driven from any user-programmable pin as a
10 MHz. global 3-state input. To use this global net, place an input
The oscillator output is optionally available after configura- pad and input buffer in the schematic or HDL code, driving
tion. Any two of four resynchronized taps of a built-in divider the GTS pin of the STARTUP symbol. This is similar to what
are also available. These taps are at the fourth, ninth, four- is shown in Figure 19 for GSR except the IBUF would be
teenth and nineteenth bits of the divider. Therefore, if the connected to GTS. A specific pin location can be assigned
primary oscillator output is running at the nominal 8 MHz, to this input using a LOC attribute or property, just as with
the user has access to an 8 MHz clock, plus any two of any other user-programmable pad. An inverter can option-
500 kHz, 16 kHz, 490 Hz and 15 Hz. These frequencies ally be inserted after the input buffer to invert the sense of
can vary by as much as -50% or +25%. the Global 3-State signal. Alternatively, GTS can be driven
from any internal node.
These signals can be accessed by placing the OSC library
element in a schematic or in HDL code. The oscillator is
4
Boundary Scan
automatically disabled after configuration if the OSC sym-
bol is not used in the design. The ‘bed of nails’ has been the traditional method of testing
electronic assemblies. This approach has become less
Global Signals: GSR and GTS appropriate, due to closer pin spacing and more sophisti-
cated assembly methods like surface-mount technology
Global Set/Reset and multi-layer boards. The IEEE Boundary Scan Standard
A separate Global Set/Reset line, as shown in Figure 3 on 1149.1 was developed to facilitate board-level testing of
page 6 for the CLB and Figure 6 on page 7 for the IOB, sets electronic assemblies. Design and test engineers can
or clears each flip-flop during power-up, reconfiguration, or imbed a standard test logic structure in their device to
when a dedicated Reset net is driven active. This global net achieve high fault coverage for I/O and internal logic. This
(GSR) does not compete with other routing resources; it structure is easily implemented with a four-pin interface on
uses a dedicated distribution network. any boundary scan-compatible IC. IEEE 1149.1-compati-
ble devices may be serial daisy-chained together, con-
Each flip-flop is configured as either globally set or reset in
nected in parallel, or a combination of the two.
the same way that the local set/reset (SR) is specified.
Therefore, if a flip-flop is set by SR, it is also set by GSR. The Spartan Series implements IEEE 1149.1-compatible
Similarly, if in reset mode, it is reset by both SR and GSR. BYPASS, PRELOAD/SAMPLE and EXTEST boundary
scan instructions. When the boundary scan configuration
GSR can be driven from any user-programmable pin as a option is selected, three normal user I/O pins become ded-
global reset input. To use this global net, place an input pad icated inputs for these functions. Another user output pin
and input buffer in the schematic or HDL code, driving the becomes the dedicated boundary scan output. The details
GSR pin of the STARTUP symbol. (See Figure 19.) A spe- of how to enable this circuitry are covered later in this sec-
cific pin location can be assigned to this input using a LOC tion.
attribute or property, just as with any other user-program-
By exercising these input signals, the user can serially load
mable pad. An inverter can optionally be inserted after the
commands and data into these devices to control the driv-
input buffer to invert the sense of the GSR signal. Alterna-
ing of their outputs and to examine their inputs. This
tively, GSR can be driven from any internal node.
method is an improvement over bed-of-nails testing. It
avoids the need to over-drive device outputs, and it reduces
the user interface to four pins. An optional fifth pin, a reset
STARTUP
for the control logic, is described in the standard but is not
PAD GSR Q2 implemented in Xilinx devices.
GTS Q3 The dedicated on-chip logic implementing the IEEE 1149.1
IBUF
Q1Q4 functions includes a 16-state machine, an instruction regis-
CLK DONEIN ter and a number of data registers. The functional details
can be found in the IEEE 1149.1 specification and are also
X5260
discussed in the Xilinx application note: “Boundary Scan in
Figure 19: Schematic Symbols for Global Set/Reset FPGA Devices.”

R
Figure 20 is a diagram of the Spartan Series boundary The other standard data register is the single flip-flop
scan logic. It includes three bits of Data Register per IOB, BYPASS register. It synchronizes data being passed
the IEEE 1149.1 Test Access Port controller, and the through the FPGA to the next downstream boundary scan
Instruction Register with decodes. device.
Spartan Series devices can also be configured through the The FPGA provides two additional data registers that can
boundary scan logic. See “Configuration Through the be specified using the BSCAN macro. The FPGA provides
Boundary Scan Pins” on page 34. two user pins (BSCAN.SEL1 and BSCAN.SEL2) which are
the decodes of two user instructions. For these instructions,
Data Registers two corresponding pins (BSCAN.TDO1 and
The primary data register is the boundary scan register. For BSCAN.TDO2) allow user scan data to be shifted out on
each IOB pin in the FPGA, bonded or not, it includes three TDO. The data register clock (BSCAN.DRCK) is available
bits for In, Out and 3-State Control. Non-IOB pins have for control of test logic which the user may wish to imple-
appropriate partial bit population for In or Out only. PRO- ment with CLBs. The NAND of TCK and RUN-TEST-IDLE
GRAM, CCLK and DONE are not included in the boundary is also provided (BSCAN.IDLE).
scan register. Each EXTEST CAPTURE-DR state captures
all In, Out, and 3-state pins.
Instruction Set
The Spartan Series boundary scan instruction set also
The data register also includes the following non-pin bits:
includes instructions to configure the device and read back
TDO.T, and TDO.O, which are always bits 0 and 1 of the
the configuration data. The instruction set is coded as
data register, respectively, and BSCANT.UPD, which is
shown in Table 12.
always the last bit of the data register. These three bound-
ary scan bits are special-purpose Xilinx test signals.
DATA IN
IOB.T 0
1 sd
D Q D Q 1
0
IOB IOB IOB IOB IOB LE
IOB IOB sd
1
D Q D Q
0
IOB IOB
LE
IOB IOB
1
IOB.I
0
IOB IOB
1 sd
IOB IOB D Q D Q
0
LE
IOB IOB
1
0
IOB IOB IOB.Q
BYPASS
REGISTER
IOB.T 0
M TDO
U 1 sd
INSTRUCTION REGISTER D Q D Q 1
TDI
X 0
LE
1 sd
D Q D Q
0
LE
1
IOB.I
0
DATAOUT UPDATE EXTEST

SHIFT/ CLOCK DATA
CAPTURE REGISTER
X9016
Figure 20: Spartan Series Boundary Scan Logic

R
Table 12: Boundary Scan Instructions BSDL (Boundary Scan Description Language) files for
Spartan Series devices are available on the Xilinx web site
Instruction Test I/O Data in the File Download area. Note that the 5V Spartan
TDO Source
I2 I1 I0 Selected Source devices and 3V SpartanXL devices have different BSDL
0 0 0 EXTEST DR DR files.
0 0 1 SAMPLE/PR DR Pin/Logic
ELOAD
Including Boundary Scan in a Design
0 1 0 USER 1 BSCAN. User Logic If boundary scan is only to be used during configuration, no
TDO1 special schematic elements need be included in the sche-
0 1 1 USER 2 BSCAN. User Logic matic or HDL code. In this case, the special boundary scan
TDO2 pins TDI, TMS, TCK and TDO can be used for user func-
tions after configuration.
1 0 0 READBACK Readback Data Pin/Logic
1 0 1 CONFIGURE DOUT Disabled To indicate that boundary scan remain enabled after config-
1 1 0 Reserved — — uration, place the BSCAN library symbol and connect the
TDI, TMS, TCK and TDO pad symbols to the appropriate
1 1 1 BYPASS Bypass Register —
pins, as shown in Figure 22.
Bit Sequence Even if the boundary scan symbol is used in a schematic,
The bit sequence within each IOB is: In, Out, 3-State. The the input pins TMS, TCK, and TDI can still be used as 4
input-only pins contribute only the In bit to the boundary inputs to be routed to internal logic. Care must be taken not
scan I/O data register, while the output-only pins contrib- to force the chip into an undesired boundary scan state by
utes all three bits. inadvertently applying boundary scan input patterns to
these pins. The simplest way to prevent this is to keep TMS
The first two bits in the I/O data register are TDO.T and High, and then apply whatever signal is desired to TDI and
TDO.O, which can be used for the capture of internal sig- TCK.
nals. The final bit is BSCANT.UPD, which can be used to
drive an internal net. These locations are primarily used by Avoiding Inadvertent Boundary Scan
Xilinx for internal testing.
If TMS or TCK is used as user I/O, care must be taken to
From a cavity-up view of the chip (as shown in EPIC), start- ensure that at least one of these pins is held constant dur-
ing in the upper right chip corner, the boundary scan ing configuration. In some applications, a situation may
data-register bits are ordered as shown in Figure 21. The occur where TMS or TCK is driven during configuration.
device-specific pinout tables for the Spartan Series include This may cause the device to go into boundary scan mode
the boundary scan locations for each IOB pin. and disrupt the configuration process.
To prevent activation of boundary scan during configura-
Bit 0 ( TDO end) TDO.T tion, do either of the following:
Bit 1 TDO.O
Bit 2 • TMS: Tie High to put the Test Access Port controller
Top-edge IOBs (Right to Left)
in a benign RESET state
• TCK: Tie High or Low—do not toggle this clock input.
Left-edge IOBs (Top to Bottom)
For more information regarding boundary scan, refer to the
Xilinx Application Note, “Boundary Scan in FPGA Devices.“
MODE.I
Optional To User
Logic
IBUF
Bottom-edge IOBs (Left to Right)
BSCAN
Right-edge IOBs (Bottom to Top) TDI TDI TDO TDO
TMS TMS DRCK

(TDI end) BSCANT.UPD
TCK TCK IDLE
To User
S6075_02 From TDO1 SEL1 Logic
User Logic TDO2 SEL2
Figure 21: Boundary Scan Bit Sequence X2675
Figure 22: Boundary Scan Schematic Example

R
Boundary Scan Enhancements (SpartanXL only) 3.3V SpartanXL family adds a dedicated active-Low Power
Down pin (PWRDWN) to reduce supply current to 100 µA
SpartanXL devices have improved boundary scan function-
typical. The PWRDWN pin takes advantage of one of the
ality and performance in the following areas:
unused Don’t Connect locations on the 5V Spartan. The
IDCODE: The IDCODE register is now supported. By using user must de-select the “5V Tolerant I/Os” option in the
the IDCODE, the device connected to the JTAG port can be Configuration Options to achieve the specified Power Down
determined. The use of the IDCODE enables selective concurrent. The PWRDWN pin has a default internal pull-up
figuration dependent on the FPGA found. resistor, allowing it to be left unconnected if unused.
The IDCODE register has the following binary format: VCC must continue to be supplied during Power Down, and
vvvv:ffff:fffa:aaaa:aaaa:cccc:cccc:ccc1 configuration data is maintained. When the PWRDWN pin
is pulled Low, the input and output buffers are disabled. The
where inputs are internally forced to a logic Low level, including
c = the company code (49h for Xilinx) the MODE pins, DONE, CCLK, and TDO, and all internal
pull-up resistors are turned off. The PROGRAM pin is not
a = the array dimension in CLBs (ranges from 0Ah for
affected by Power Down. The GSR net is asserted during
XCS05XL to 1Ch for XCS40XL)
Power Down, initializing all the flip-flops to their start-up
f = the family code (02h for SpartanXL family) state.
v = the die version number (currently 0h) PWRDWN has a minimum pulse width of 50 ns. On enter-
Table 13: IDCODEs Assigned to SpartanXL FPGAs ing the Power Down state, the inputs will be disabled and
the flip-flops set/reset, and then the outputs are disabled
FPGA IDCODE about 10 ns later. The user may prefer to assert the GTS or
XCS05XL 0040A093h GSR signals before PWRDWN to affect the order of events.
XCS10XL 0040E093h When the PWRDWN signal is returned High, the inputs will
XCS20XL 00414093h be enabled first, followed immediately by the release of the
GSR signal initializing the flip-flops. About 10 ns later, the
XCS30XL 00418093h
outputs will be enabled. Allow 50 ns after the release of
XCS40XL 0041C093h PWRDWN before using the device.
Configuration State: The configuration state is available to Power Down retains the configuration, but loses all data
JTAG controllers. stored in the device. All inputs are interpreted as Low, but
the internal combinatorial logic is fully functional. Make
Configuration Disable: The JTAG port can be prevented
sure that the combination of all inputs Low and all flip-flops
from configuring the FPGA.
set or reset in your design will not generate internal oscilla-
TCK Startup: TCK can now be used to clock the start-up tions, or create permanent bus contention by activating
block in addition to other user clocks. internal bus drivers with conflicting data onto the same long
CCLK Holdoff: Changed the requirement for Boundary line.
Scan Configure or EXTEST to be issued prior to the During configuration, the PWRDWN pin must be High. If
release of INIT pin and CCLK cycling. the Power Down state is entered before or during configu-
Reissue Configure: The Boundary Scan Configure can be ration, the device will re-start configuration once the
reissued to recover from an unfinished attempt to configure PWRDWN signal is removed. Note that the configuration
the device. pins are affected by Power Down and may not reflect their
normal function. If there is an external pull-up resistor on
Bypass FF: Bypass FF and IOB is modified to provide the DONE pin, it will be High during Power Down even if the
DRCLOCK only during BYPASS for the bypass flip-flop, device is not yet configured. Similarly, if PWRDWN is
and during EXTEST or SAMPLE/PRELOAD for the IOB asserted before configuration is completed, the INIT pin will
register. not indicate status information.
Power Down (SpartanXL Only) Note that the PWRDWN pin is not part of the Boundary
Scan chain. Therefore, the SpartanXL family has a sepa-
All Spartan Series devices use a combination of efficient rate set of BSDL files than the 5V Spartan family. Boundary
segmented routing and advanced process technology to scan logic is not usable during Power Down.
provide low power consumption under all conditions. The

R
Configuration and Test Table 14: Pin Functions During Configuration (Spartan
Configuration is the process of loading design-specific pro- only)
gramming data into one or more FPGAs to define the func-
tional operation of the internal blocks and their CONFIGURATION MODE
<MODE Pin>
interconnections. This is somewhat like loading the com-
SLAVE MASTER
mand registers of a programmable peripheral chip. Spartan USER
SERIAL SERIAL
OPERATION
Series devices use several hundred bits of configuration <High> <Low>
data per CLB and its associated interconnects. Each con- MODE (I) MODE (I) MODE
figuration bit defines the state of a static memory cell that HDC (HIGH) HDC (HIGH) I/O
controls either a function look-up table bit, a multiplexer LDC (LOW) LDC (LOW) I/O
input, or an interconnect pass transistor. The Xilinx devel- INIT INIT I/O
opment system translates the design into a netlist file. It DONE DONE DONE
automatically partitions, places and routes the logic and PROGRAM (I) PROGRAM (I) PROGRAM
generates the configuration data in PROM format. CCLK (I) CCLK (O) CCLK (I)
DIN (I) DIN (I) I/O
Configuration Mode Control DOUT DOUT SGCK4-I/O
TDI TDI TDI-I/O
5V Spartan Series devices have two configuration modes. TCK TCK TCK-I/O
• MODE = 1 sets Slave Serial mode TMS TMS TMS-I/O 4
• MODE = 0 sets Master Serial mode TDO TDO TDO-(O)
ALL OTHERS
3V SpartanXL Series devices have three configuration Notes 1. A shaded table cell represents the internal pull-up
modes. used before and during configuration.
2. (I) represents an input; (O) represents an output.
• M1/M0 = 11 sets Slave Serial mode 3. INIT is an open-drain output during configuration.
• M1/M0 = 10 sets Master Serial mode
• M1/M0 = 0X sets Express mode Table 15: Pin Functions During Configuration
In addition to these modes, the device can be configured (SpartanXL only)
through the Boundary Scan logic (See “Configuration CONFIGURATION MODE <M1:M0>
Through the Boundary Scan Pins” on page 34.). SLAVE MASTER USER
The Mode pins are sampled prior to starting configuration SERIAL SERIAL EXPRESS OPERATION
<1:1> <1:0> <0:X>
to determine the configuration mode. After configuration,
M1(HIGH) (I) M1(HIGH) (I) M1(LOW) (I) M1
these pin are unused. The Mode pins have a weak pull-up M0(HIGH) (I) M0(LOW) (I) M0 (I) M0
resistor of 20 kΩ to 100 kΩ turned on during configuration. HDC (HIGH) HDC (HIGH) HDC (HIGH) I/O
With the Mode pins High, Slave Serial mode is selected, LDC (LOW) LDC (LOW) LDC (LOW) I/O
which is the most popular configuration mode. Therefore, INIT INIT INIT I/O
for the most common configuration mode, the Mode pins DONE DONE DONE DONE
can be left unconnected. If the Master Serial mode is PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM
desired, the MODE/M0 pin should be connected directly to CCLK (I) CCLK (O) CCLK (I) CCLK (I)
GND, or through a pull-down resistor of 1 KΩ or less. DATA 7 (I) I/O
DATA 6 (I) I/O
During configuration, some of the I/O pins are used tempo-
DATA 5 (I) I/O
rarily for the configuration process. All pins used during
DATA 4 (I) I/O
configuration are shown in Table 14 and Table 15.
DATA 3 (I) I/O
DATA 2 (I) I/O
DATA 1 (I) I/O
DIN (I) DIN (I) DATA 0 (I) I/O
DOUT DOUT DOUT SGCK4-I/O
TDI TDI TDI TDI-I/O
TCK TCK TCK TCK-I/O
TMS TMS TMS TMS-I/O
TDO TDO TDO TDO-(O)
CS1 I/O
ALL OTHERS
Notes 1. A shaded table cell represents the internal pull-up
used before and during configuration.
3. INIT is an open-drain output during configuration.

R
Master Serial Mode falling CCLK edge, and the next FPGA in the daisy chain
accepts data on the subsequent rising CCLK edge. See the
The Master serial mode uses an internal oscillator to gen- timing diagram in Figure 23.
erate a Configuration Clock (CCLK) for driving potential
slave devices and the Xilinx serial-configuration PROM In the bitstream generation software, the user can specify
(SPROM). The CCLK speed is selectable as either 1 MHz Fast Configuration Rate, which, starting several bits into the
(default) or 8 MHz. Configuration always starts at the first frame, increases the CCLK frequency by a factor of
default slow frequency, then can switch to the higher fre- eight. For actual timing values please refer to the specifica-
quency during the first frame. Frequency tolerance is -50% tion section. Be sure that the serial PROM and slaves are
to +25%. fast enough to support this data rate. Devices such as
XC3000A and XC3100A do not support the Fast Configura-
In Master Serial mode, the CCLK output of the device tion Rate option.
drives a Xilinx SPROM that feeds the FPGA DIN input.
Each rising edge of the CCLK output increments the Serial The SPROM CE input can be driven from either LDC or
PROM internal address counter. The next data bit is put on DONE. Using LDC avoids potential contention on the DIN
the SPROM data output, connected to the FPGA DIN pin. pin, if this pin is configured as user-I/O, but LDC is then
The FPGA accepts this data on the subsequent rising restricted to be a permanently High user output after con-
CCLK edge. figuration. Using DONE can also avoid contention on DIN,
provided the early DONE option is invoked.
When used in a daisy-chain configuration the Master Serial
FPGA is placed as the first device in the chain and is Figure 24 shows a full master/slave system. The leftmost
referred to as the lead FPGA. The lead FPGA presents the device is in Master Serial mode, all other devices in the
preamble data, and all data that overflows the lead device, chain are in Slave Serial mode.
on its DOUT pin. There is an internal pipeline delay of 1.5
CCLK periods, which means that DOUT changes on the
CCLK
(Output)
2 TCKDS
1 TDSCK
Serial Data In n n+1 n+2
Serial DOUT n–3 n–2 n–1 n

(Output)
X3223
Description Symbol Min Max Units

DIN setup 1 TDSCK 20 ns
CCLK
DIN hold 2 TCKDS 0 ns
Notes: 1. At power-up, Vcc must rise from 2.0 V to Vcc min in less than 25 ms, otherwise delay configuration by pulling PROGRAM
Low until Vcc is valid.
2. Master Serial mode timing is based on testing in slave mode.

R
Slave Serial Mode Serial Daisy Chain

In Slave Serial mode, the FPGA receives serial configura- Multiple devices with different configurations can be con-
tion data on the rising edge of CCLK and, after loading its nected together in a “daisy chain,” and a single combined
configuration, passes additional data out, resynchronized bitstream used to configure the chain of slave devices.
on the next falling edge of CCLK.
To configure a daisy chain of devices, wire the CCLK pins
In this mode, an external signal drives the CCLK input of of all devices in parallel, as shown in Figure 24. Connect
the FPGA (most often from a Master Serial device). The the DOUT of each device to the DIN of the next. The lead or
serial configuration bitstream must be available at the DIN master FPGA and following slaves each passes resynchro-
input of the lead FPGA a short setup time before each ris- nized configuration data coming from a single source. The
ing CCLK edge. header data, including the length count, is passed through
and is captured by each FPGA when it recognizes the 0010
The lead FPGA then presents the preamble data—and all
preamble. Following the length-count data, each FPGA
data that overflows the lead device—on its DOUT pin.
outputs a High on DOUT until it has received its required
There is an internal delay of 0.5 CCLK periods, which
number of data frames.
means that DOUT changes on the falling CCLK edge, and
the next FPGA in the daisy chain accepts data on the sub- After an FPGA has received its configuration data, it
sequent rising CCLK edge. passes on any additional frame start bits and configuration
data on DOUT. When the total number of configuration
Figure 24 shows a full master/slave system. A Spartan 4
Series device in Slave Serial mode should be connected as clocks applied after memory initialization equals the value
of the 24-bit length count, the FPGAs begin the start-up
shown in the third device from the left.
sequence and become operational together. FPGA I/O are
Slave Serial is the default mode if the Mode pins are left normally released two CCLK cycles after the last configura-
unconnected, as they have weak pull-up resistors during tion bit is received.
configuration.
The daisy-chained bitstream is not simply a concatenation
Multiple slave devices with identical configurations can be of the individual bitstreams. The PROM file formatter must
wired with parallel DIN inputs. In this way, multiple devices be used to combine the bitstreams for a daisy-chained con-
can be configured simultaneously. figuration.
NOTE:
M2, M1, M0 can be shorted
to VCC if not used as I/O
VCC
4.7 K 4.7 K 4.7 K
M0 M1 PWRDN
MODE N/C MODE M2
DOUT DIN DOUT DIN DOUT
CCLK CCLK
Spartan VCC
MASTER XC17S00 +5 V Spartan FPGA
4.7 K
SERIAL SLAVE SLAVE
CCLK CLK VPP
DIN DATA
PROGRAM LDC CE CEO PROGRAM RESET

DONE INIT RESET/OE DONE INIT D/P INIT
PROGRAM S9025_02

R
DIN Bit n Bit n + 1
CCLK
4 TCCH 3 TCCO
DOUT
Bit n - 1 Bit n
(Output)
X5379

DIN setup 1 TDCC 20 ns
DIN hold 2 TCCD 0 ns
DIN to DOUT 3 TCCO 30 ns
CCLK
High time 4 TCCH 45 ns
Low time 5 TCCL 45 ns
Frequency FCC 10 MHz
Note: Configuration must be delayed until the INIT pins of all daisy-chained FPGAs are High.
Express Mode (SpartanXL only) device’s configuration memory is not already full. The lead
device in the chain has its CS1 input tied High (or floating,
Express mode is similar to Slave Serial mode, except that since there is an internal pullup). The status pin DOUT is
data is processed one byte per CCLK cycle instead of one pulled Low after the header is received by all devices, and
bit per CCLK cycle. An external source is used to drive
remains Low until the device’s configuration memory is full.
CCLK, while byte-wide data is loaded directly into the con-
DOUT is then pulled High to signal the next device in the
figuration data shift registers (Figure 26). A CCLK fre- chain to accept the configuration data on the D0-D7 bus.
quency of 1 MHz is equivalent to a 8 MHz serial rate,
because eight bits of configuration data are loaded per The DONE pins of all devices in the chain should be tied
CCLK cycle. Express mode does not support CRC error together, with one or more active internal pull-ups. If a large
checking, but does support constant-field error checking. A number of devices are included in the chain, deactivate
length count is not used in Express mode. some of the internal pull-ups, since the Low-driving DONE
pin of the last device in the chain must sink the current from
Express mode must be specified as an option to the BitGen
all pull-ups in the chain. The DONE pull-up is activated by
program, which generates the bitstream. The Express
default. It can be deactivated using a BitGen option.
mode bitstream is not compatible with the other configura-
tion modes. (See Table 16 on page 31.) Express mode is The requirement that all DONE pins in a daisy chain be
selected by a <0X> on the mode pins (M1, M0). wired together applies only to Express mode, and only if all
devices in the chain are to become active simultaneously.
The first byte of parallel configuration data must be avail- All SpartanXL devices in Express mode are synchronized
able at the D inputs of the FPGA a short setup time before to the DONE pin. User I/Os for each device become active
the second rising CCLK edge. Subsequent data bytes are after the DONE pin for that device goes High. (The exact
clocked in on each consecutive rising CCLK edge
timing is determined by BitGen options.) Since the DONE
(Figure 27).
pin is open-drain and does not drive a High value, tying the
Pseudo Daisy Chain DONE pins of all devices together prevents all devices in
the chain from going High until the last device in the chain
Multiple devices with different configurations can be config- has completed its configuration cycle. If the DONE pin of a
ured in a pseudo daisy chain provided that all of the devices device is left unconnected, the device becomes active as
are in Express mode. A single combined bitstream is used soon as that device has been configured.
to configure the chain of Express mode devices. CCLK pins
are tied together and D0-D7 pins are tied together for all Because only SpartanXL, XC4000XLA/XV, and XC5200
devices along the chain. A status signal is passed from devices support Express mode, only these devices can be
DOUT to CS1 of successive devices along the chain. used to form an Express mode daisy chain.
Frame data is accepted only when CS1 is High and the

R
VCC
To Additional
M0 M1 M0 M1 Optional
Daisy-Chained
Devices
CS1 DOUT CS1 DOUT
8 8
DATA BUS D0-D7 D0-D7
Optional
VCC Daisy-Chained
SpartanXL SpartanXL
4.7KΩ
PROGRAM PROGRAM PROGRAM
INIT INIT DONE INIT DONE
CCLK CCLK
4
To Additional
Optional
Daisy-Chained
CCLK
Devices
X6611_b
Figure 26: Express Mode Circuit Diagram

R

INIT (High) setup time 1 TIC 5 µs
D0 - D7 setup time 2 TDC 20 ns
D0 - D7 hold time 3 TCD 0 ns
CCLK
CCLK High time TCCH 45 ns
CCLK Low time TCCL 45 ns
CCLK Frequency FCC 10 MHz
CCLK
1 TIC
INIT
TCD 3
2 T
DC
D0-D7 BYTE BYTE BYTE

0 1 6
DOUT
Header Received FPGA Filled
X6710_m
Note: If not driven by the preceding DOUT, CS1 must remain High until the device is fully configured.
Figure 27: Express Mode Programming Switching Characteristics
Setting CCLK Frequency Table 16. Bit-serial data is read from left to right. Express
mode data is shown with D0 at the left and D7 at the right.
In Master mode, CCLK can be generated in either of two
frequencies. In the default slow mode, the frequency The configuration data stream begins with a string of eight
ranges from 0.5 MHz to 1.25 MHz for Spartan Series ones, a preamble code, followed by a 24-bit length count
devices. In fast CCLK mode, the frequency ranges from and a separator field of ones (or 24 fill bits, in SpartanXL
4 MHz to 10 MHz for Spartan Series devices. The fre- Express mode). This header is followed by the actual con-
quency is changed to fast by an option when running the figuration data in frames. The length and number of frames
bitstream generation software. depends on the device type (see Table 17). Each frame
begins with a start field and ends with an error check. In
Data Stream Format serial modes, a postamble code is required to signal the
end of data for a single device. In all cases, additional
The data stream (“bitstream”) format is identical for both
start-up bytes of data are required to provide four clocks for
serial configuration modes, but different for the SpartanXL
the startup sequence at the end of configuration. Long
Express mode. In Express mode, the device becomes
daisy chains require additional startup bytes to shift the last
active when DONE goes High, therefore no length count is
data through the chain. All startup bytes are don’t-cares.
required. Additionally, CRC error checking is not supported
in Express mode. The data stream format is shown in

R
A selection of CRC or non-CRC error checking is allowed

by the bitstream generation software. The SpartanXL
Table 16: Spartan Series Data Stream Formats
Express mode only supports non-CRC error checking. The
Express Mode non-CRC error checking tests for a designated
Serial Modes
Data Type (D0-D7) end-of-frame field for each frame. For CRC error checking,
(D0...)
(SpartanXL only) the software calculates a running CRC and inserts a
Fill Byte 11111111b FFFFh unique four-bit partial check at the end of each frame. The
Preamble Code 0010b 11110010b 11-bit CRC check of the last frame of an FPGA includes the
last seven data bits.
Length Count COUNT(23:0) COUNT(23:0)1
Fill Bits 1111b — Detection of an error results in the suspension of data load-
ing before DONE goes High, and the pulling down of the
Field Check Code — 11010010b
INIT pin. In Master serial mode, CCLK continues to operate
Start Field 0b 11111110b
externally. The user must detect INIT and initialize a new
Data Frame DATA(n-1:0) DATA(n-1:0) configuration by pulsing the PROGRAM pin Low or cycling
CRC or Constant xxxx (CRC) 11010010b Vcc.
Field Check or 0110b
Extend Write Cycle — FFFFFFFFFFh Cyclic Redundancy Check (CRC) for Configura-
Postamble 01111111b — tion and Readback
4
Start-Up Bytes FFh FFFFFFFFFFFFFFh2 The Cyclic Redundancy Check is a method of error detec-
tion in data transmission applications. Generally, the trans-
LEGEND: mitting system performs a calculation on the serial
Unshaded Once per bitstream bitstream. The result of this calculation is tagged onto the
data stream as additional check bits. The receiving system
Light Once per data frame
performs an identical calculation on the bitstream and com-
Dark Once per device pares the result with the received checksum.
Note 1: Not used by configuration logic.
Note 2: Development system may add more start-up Each data frame of the configuration bitstream has four
bytes. error bits at the end, as shown in Table 16. If a frame data
error is detected during the loading of the FPGA, the con-
figuration process with a potentially corrupted bitstream is
terminated. The FPGA pulls the INIT pin Low and goes into
a Wait state.

R
Table 17: Spartan Program Data
Device XCS05 XCS10 XCS20 XCS30 XCS40

Max System 5,000 10,000 20,000 30,000 40,000
Gates
CLBs 100 196 400 576 784
(Row x Col.) (10 x 10) (14 x 14) (20 x 20) (24 x 24) (28 x 28)
IOBs 80 112 160 192 224
Part Number XCS05 XCS05XL XCS10 XCS10XL XCS20 XCS20XL XCS30 XCS30XL XCS40 XCS40XL
Supply Voltage 5V 3.3 V 5V 3.3 V 5V 3.3 V 5V 3.3 V 5V 3.3 V
Bits per Frame 126 127 166 167 226 227 266 267 306 307
Frames 428 429 572 573 788 789 932 933 1,076 1,077
Program Data 53,936 54,491 94,960 95,699 178,096 179,111 247,920 249,119 329,264 330,647
PROM Size 53,984 54,536 95,008 95,744 178,144 179,160 247,968 249,168 329,312 330,696
(bits)
Serial PROM 17S05 17S05XL 17S10 17S10XL 17S20 17S20XL 17S30 17S30XL 17S40 17S40XL
Express Mode 79,040 128,456 221,024 298,664 387,824
PROM Size
(bits)
Notes: 1.Bits per Frame = (10 x number of rows) + 7 for the top + 13 for the bottom + 1 + 1 start bit + 4 error check bits (+ 1 for
SpartanXL device)
Number of Frames = (36 x number of columns) + 26 for the left edge + 41 for the right edge + 1 (+ 1 for SpartanXL device)
Program Data = (Bits per Frame x Number of Frames) + 8 postamble bits
PROM Size = Program Data + 40 (header) + 8, rounded up to the nearest byte
2.The user can add more “one” bits as leading dummy bits in the header, or, if CRC = off, as trailing dummy bits at the end of
any frame, following the four error check bits. However, the Length Count value must be adjusted for all such extra “one”
bits, even for extra leading ones at the beginning of the header.
3. Express mode adds 57 (XCS05XL, XCS10XL), or 53 (XCS20XL, XCS30XL, XCS40XL) bits per frame.
During Readback, 11 bits of the 16-bit checksum are added When Vcc reaches an operational level, and the circuit
to the end of the Readback data stream. The checksum is passes the write and read test of a sample pair of configu-
computed using the CRC-16 CCITT polynomial, as shown ration bits, a time delay is started. This time delay is nomi-
in Figure 28. The checksum consists of the 11 most signif- nally 16 ms. The delay is four times as long when in Master
icant bits of the 16-bit code. A change in the checksum indi- Serial Mode to allow ample time for all slaves to reach a
cates a change in the Readback bitstream. A comparison stable Vcc. When all INIT pins are tied together, as recom-
to a previous checksum is meaningful only if the readback mended, the longest delay takes precedence. Therefore,
data is independent of the current device state. CLB out- devices with different time delays can easily be mixed and
puts should not be included (Readback Capture option not matched in a daisy chain.
used), and if RAM is present, the RAM content must be
This delay is applied only on power-up. It is not applied
unchanged.
when reconfiguring an FPGA by pulsing the PROGRAM pin
Statistically, one error out of 2048 might go undetected.
X2 X15
X16
There are four major steps in the Spartan Series power-up 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
configuration sequence.
SERIAL DATA IN
• Configuration Memory Clear Polynomial: X16 + X15 + X2 + 1
• Initialization
• Configuration 1 1 1 1 1 0 15 14 13 12 11 10 9 8 7 6 5
• Start-Up
START BIT
LAST DATA FRAME CRC – CHECKSUM
The full process is illustrated in Figure 29.

X1789
Readback Data Stream
Configuration Memory Clear
When power is first applied or is reapplied to an FPGA, an
Figure 28: Circuit for Generating CRC-16
internal circuit forces initialization of the configuration logic.

R
Low. During this time delay, or as long as the PROGRAM

input is asserted, the configuration logic is held in a Config-
uration Memory Clear state. The configuration-memory
VCC No
Boundary Scan Valid frames are consecutively initialized, using the internal oscil-
Instructions
Available: lator.
Yes
At the end of each complete pass through the frame
Test MODE, Generate addressing, the power-on time-out delay circuitry and the
One Time-Out Pulse PROGRAM
of 16 or 64 ms = Low level of the PROGRAM pin are tested. If neither is asserted,
Yes
the logic initiates one additional clearing of the configura-
tion frames and then tests the INIT input.
Keep Clearing
Configuration Memory
Initialization
EXTEST* During initialization and configuration, user pins HDC, LDC,
SAMPLE/PRELOAD
BYPASS
Completely Clear INIT and DONE provide status outputs for the system inter-
Configuration Memory ~1.3 µs per Frame
CONFIGURE*
(* if PROGRAM = High)
Once More face. The outputs LDC, INIT and DONE are held Low and
HDC is held High starting at the initial application of power.
INIT
The open drain INIT pin is released after the final initializa-
No
High? if
Master
tion pass through the frame addresses. There is a deliber- 4
Master Delays Before
ate delay before a Master-mode device recognizes an
Yes
Sampling Mode Line inactive INIT. Two internal clocks after the INIT pin is recog-
Sample
nized as High, the device samples the MODE pin to deter-
Mode Line mine the configuration mode. The appropriate interface
Master CCLK lines become active and the configuration preamble and
Goes Active
data can be loaded.
LDC Output = L, HDC Output = H
Load One
Configuration
Data Frame
Configuration
The 0010 preamble code indicates that the following 24 bits
represent the length count for serial modes. The length
Frame Yes
Error
Pull INIT Low
and Stop
count is the total number of configuration clocks needed to
load the complete configuration data. (Four additional con-
No
figuration clocks are required to complete the configuration
SAMPLE/PRELOAD Config- process, as discussed below.) After the preamble and the
BYPASS uration No
memory length count have been passed through to any device in the
Full
daisy chain, its DOUT is held High to prevent frame start
Yes
bits from reaching any daisy-chained devices. In Spar-
Pass tanXL Express mode, the length count bits are ignored, and
Configuration
Data to DOUT DOUT is held Low, to disable the next device in the pseudo
daisy chain.
CCLK A specific configuration bit, early in the first frame of a mas-
Count Equals No
Length ter device, controls the configuration-clock rate and can
Count increase it by a factor of eight. Therefore, if a fast configu-
Yes
ration clock is selected by the bitstream, the slower clock
rate is used until this configuration bit is detected.
Start-Up
Sequence
Each frame has a start field followed by the frame-configu-
F
ration data bits and a frame error field. If a frame data error
I/O Active
Operational
EXTEST is detected, the FPGA halts loading, and signals the error
SAMPLE PRELOAD
BYPASS by pulling the open-drain INIT pin Low. After all configura-
USER 1 If Boundary Scan
USER 2 is Selected tion frames have been loaded into an FPGA using a serial
CONFIGURE
READBACK
mode, DOUT again follows the input data so that the
s6076_01
remaining data is passed on to the next device. In Spar-
tanXL Express mode, when the first device is fully pro-
grammed, DOUT goes High to enable the next device in the
Figure 29: Power-up Configuration Sequence chain.

R
Delaying Configuration After Power-Up Configuration Through the Boundary Scan

There are two methods of delaying configuration after Pins
power-up: put a logic Low on the PROGRAM input, or pull Spartan Series devices can be configured through the
the bidirectional INIT pin Low, using an open-collector boundary scan pins. The basic procedure is as follows:
(open-drain) driver. (See Figure 29 on page 33.)
• Power up the FPGA with INIT held Low (or drive the
A Low on the PROGRAM input is the more radical PROGRAM pin Low for more than 300 ns followed by a
approach, and is recommended when the power-supply High while holding INIT Low). Holding INIT Low allows
rise time is excessive or poorly defined. As long as PRO- enough time to issue the CONFIG command to the
GRAM is Low, the FPGA keeps clearing its configuration FPGA. The pin can be used as I/O after configuration if
memory. When PROGRAM goes High, the configuration a resistor is used to hold INIT Low.
memory is cleared one more time, followed by the begin- • Issue the CONFIG command to the TMS input
ning of configuration, provided the INIT input is not exter- • Wait for INIT to go High
nally held Low. Note that a Low on the PROGRAM input • Sequence the boundary scan Test Access Port to the
automatically forces a Low on the INIT output. The Spartan SHIFT-DR state
Series PROGRAM pin has a permanent weak pull-up. • Toggle TCK to clock data into TDI pin.
Using an open-collector or open-drain driver to hold INIT The user must account for all TCK clock cycles after INIT
Low before the beginning of configuration causes the goes High, as all of these cycles affect the Length Count
FPGA to wait after completing the configuration memory compare.
clear operation. When INIT is no longer held Low exter-
nally, the device determines its configuration mode by cap- For more detailed information, refer to the Xilinx application
turing the state of the MODE pin, and is ready to start the note, “Boundary Scan in FPGA Devices.” This application
configuration process. A master device waits up to an addi- note applies to Spartan and SpartanXL devices.
tional 300 µs to make sure that any slaves in the optional
daisy chain have seen that INIT is High.

R
IF UNCONNECTED,
DEFAULT IS CCLK
CLK DATA READ_DATA
READBACK OBUF
READ_TRIGGER TRIG RIP
IBUF s1786_01
Figure 30: Readback Schematic Example
Readback flip-flops and the input signals I1 and I2. Note that while
the bits describing configuration (interconnect, function
The user can read back the content of configuration mem- generators, and RAM content) are not inverted, the CLB
ory and the level of certain internal nodes without interfer- and IOB output signals are inverted. RDBK.TRIG is located
ing with the normal operation of the device. in the lower-left corner of the device.
Readback not only reports the downloaded configuration When the Readback Capture option is not selected, the val-
bits, but can also include the present state of the device, ues of the capture bits reflect the configuration data origi-
represented by the content of all flip-flops and latches in nally written to those memory locations. If the RAM
CLBs and IOBs, as well as the content of function genera- 4
capability of the CLBs is used, RAM data are available in
tors used as RAMs.
readback, since they directly overwrite the F and G func-
Readback of SpartanXL Express mode bitstreams results tion-table configuration of the CLB.
in data that does not resemble the original bitstream,
because the bitstream format differs from other modes. Readback Abort
Spartan Series Readback does not use any dedicated pins, When the Readback Abort option is selected, a
but uses four internal nets (RDBK.TRIG, RDBK.DATA, High-to-Low transition on RDBK.TRIG terminates the read-
RDBK.RIP and RDBK.CLK) that can be routed to any IOB. back operation and prepares the logic to accept another
To access the internal Readback signals, place the READ- trigger.
BACK library symbol and attach the appropriate pad sym- After an aborted readback, additional clocks (up to one
bols, as shown in Figure 30. readback clock per configuration frame) may be required to
After Readback has been initiated by a Low-to-High transi- re-initialize the control logic. The status of readback is indi-
tion on RDBK.TRIG, the RDBK.RIP (Read In Progress) cated by the output control net RDBK.RIP. RDBK.RIP is
output goes High on the next rising edge of RDBK.CLK. High whenever a readback is in progress.
Subsequent rising edges of this clock shift out Readback
Clock Select
data on the RDBK.DATA net.
CCLK is the default clock. However, the user can insert
Readback data does not include the preamble, but starts
another clock on RDBK.CLK. Readback control and data
with five dummy bits (all High) followed by the Start bit
are clocked on rising edges of RDBK.CLK. If readback
(Low) of the first frame. The first two data bits of the first
must be inhibited for security reasons, the readback control
frame are always High.
nets are simply not connected. RDBK.CLK is located in the
Each frame ends with four error check bits. They are read lower right chip corner.
back as High. The last seven bits of the last frame are also
read back as High. An additional Start bit (Low) and an Violating the Maximum High and Low Time
11-bit Cyclic Redundancy Check (CRC) signature follow, Specification for the Readback Clock
before RDBK.RIP returns Low. The readback clock has a maximum High and Low time
specification. In some cases, this specification cannot be
Readback Options met. For example, if a processor is controlling readback, an
Readback options are: Readback Capture, Readback interrupt may force it to stop in the middle of a readback.
Abort, and Clock Select. They are set with the bitstream This necessitates stopping the clock, and thus violating the
generation software. specification.
Readback Capture The specification is mandatory only on clocking data at the

end of a frame prior to the next start bit. The transfer mech-
When the Readback Capture option is selected, the read- anism will load the data to a shift register during the last six
back data stream includes sampled values of CLB and IOB clock cycles of the frame, prior to the start bit of the follow-
signals. The rising edge of RDBK.TRIG latches the ing frame. This loading process is dynamic, and is the
inverted values of the four CLB outputs, the IOB output source of the maximum High and Low time requirements.

R
Therefore, the specification only applies to the six clock Readback with the XChecker Cable
cycles prior to and including any start bit, including the
The XChecker Universal Download/Readback Cable and
clocks before the first start bit in the readback data stream.
Logic Probe uses the readback feature for bitstream verifi-
At other times, the frame data is already in the register and
cation. It can also display selected internal signals on the
the register is not dynamic. Thus, it can be shifted out just
computer screen, acting as a low-cost in-circuit emulator.
like a regular shift register.
The user must precisely calculate the location of the read-
back data relative to the frame. The system must keep track
of the position within a data frame, and disable interrupts
before frame boundaries. Frame lengths and data formats
are listed in Table 16 and Table 17.
Spartan Program Readback Switching Characteristic Guidelines

Testing of the switching parameters is modeled after testing methods specified by MIL-M-38510/605. All devices are 100%
functionally tested. Internal timing parameters are not measured directly. They are derived from benchmark timing patterns
that are taken at device introduction, prior to any process improvements.
The following guidelines reflect worst-case values over the recommended operating conditions.
Finished
Internal Net
rdbk.TRIG
TRCRT TRCRT 2
1 TRTRC 2 1 TRTRC
rdclk.I
4 TRCL TRCH 5
rdbk.RIP
6
TRCRR
rdbk.DATA DUMMY DUMMY VALID VALID
TRCRD
7 X1790
Spartan and SpartanXL Readback

rdbk.TRIG rdbk.TRIG setup to initiate and abort Readback 1 TRTRC 200 - ns
rdbk.TRIG hold to initiate and abort Readback 2 TRCRT 50 - ns
rdclk.1 rdbk.DATA delay 7 TRCRD - 250 ns
rdbk.RIP delay 6 TRCRR - 250 ns
High time 5 TRCH 250 500 ns
Low time 4 TRCL 250 500 ns
Note 1: Timing parameters apply to all speed grades.
Note 2: If rdbk.TRIG is High prior to Finished, Finished will trigger the first Readback.

R
Configuration Switching Characteristics
Vcc TPOR
RE-PROGRAM
>300 ns
PROGRAM
TPI
INIT
TICCK TCCLK
<300 ns
Mode Pins
VALID DONE RESPONSE
(Required)
x1532_01
<300 ns
I/O
4
Master Mode
Power-On Reset TPOR 40 130 ms
Program Latency TPI 30 200 µs per
CLB column
CCLK (output) Delay TICCK 40 250 µs
CCLK (output) Period, slow TCCLK 640 2000 ns
CCLK (output) Period, fast TCCLK 80 250 ns
Slave Mode
CLB column
CCLK (input) Delay (required) TICCK 4 µs
CCLK (input) Period (required) TCCLK 100 ns

Spartan and SpartanXL Families Field Programmable Gate Arrays January 6, 1999 (Version 1.4) 4
Spartan Detailed Specifications

Definition of Terms
In the following tables, some specifications may be designated as Advance or Preliminary. These terms are defined as
follows:
Unmarked: Specifications not identified as either Advance or Preliminary are to be considered Final.
Notwithstanding the definition of the above terms, all specifications are subject to change without notice.
Except for pin-to-pin input and output parameters, the a.c. parameter delay specifications included in this document are
derived from measuring internal test patterns. All specifications are representative of worst-case supply voltage and junction
temperature conditions. The parameters included are common to popular designs and typical applications.
Spartan Absolute Maximum Ratings1

Symbol Description Value Units
VCC Supply voltage relative to GND -0.5 to +7.0 V
VIN Input voltage relative to GND (Note 2, 3) -0.5 to VCC +0.5 V
VTS Voltage applied to 3-state output (Note 2, 3) -0.5 to VCC +0.5 V
TJ Junction temperature Plastic packages +125 °C
Note 1: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are
stress ratings only, and functional operation of the device at these or any other conditions beyond those listed under
Operating Conditions is not implied. Exposure to Absolute Maximum Ratings conditions for extended periods of time may
affect device reliability.
2: Maximum DC overshoot (above VCC) or undershoot (below GND) must be limited to either 0.5 V or 10 mA, whichever is
easier to achieve.
3: Maximum AC (during transitions) conditions are as follows; the device pins may undershoot to -2.0 V or overshoot to + 7.0 V,
provided this overshoot or undershoot lasts no more than 11 ns with a forcing current no greater than 100 mA.
Spartan Recommended Operating Conditions

VCC Supply voltage relative to GND, TJ = 0°C to +85°C Commercial 4.75 5.25 V
Supply voltage relative to GND, TJ = -40°C to +100°C Industrial 4.5 5.5 V
VIH High-level input voltage TTL inputs 2.0 VCC V
CMOS inputs 70% 100% VCC
VIL Low-level input voltage TTL inputs 0 0.8 V
CMOS inputs 0 20% VCC
Note 1: At junction temperatures above those listed as Recommended Operating Conditions, all delay parameters increase by 0.35%
per °C.
Note 2: Input and output Measurement thresholds are: 1.5 V for TTL and 2.5 V for CMOS.

R
Spartan DC Characteristics Over Operating Conditions

VOH High-level output voltage @ IOH = -4.0mA, VCC min TTL outputs 2.4 V
High-level output voltage @ IOH = -1.0mA, VCC min CMOS outputs VCC-0.5 V
VOL Low-level output voltage @ IOL = 12.0mA, VCC min TTL outputs 0.4 V
(Note 1) CMOS outputs 0.4 V
ICCO Quiescent FPGA supply current (Note 2) Commercial 3.0 mA
Industrial 6.0 mA
CIN Input capacitance (sample tested) 10 pF
IRPU Pad pull-up (when selected) @ VIN = 0V (sample tested) 0.02 0.25 mA
IRPD Pad pull-down (when selected) @ VIN = 5V (sample tested) 0.02 mA
Note 1: With 50% of the outputs simultaneously sinking 12mA, up to a maximum of 64 pins.
Note 2: With no output current loads, no active input pull-up resistors, all package pins at Vcc or GND, and the FPGA configured with a
Tie option.
4
Spartan Global Buffer Switching Characteristic Guidelines
representative values where one global clock input drives one vertical clock line in each accessible column, and where all
accessible IOB and CLB flip-flops are clocked by the global clock net.
When fewer vertical clock lines are connected, the clock distribution is faster; when multiple clock lines per column are driven
from the same global clock, the delay is longer. For more specific, more precise, and worst-case guaranteed data, reflecting
the actual routing structure, use the values provided by the static timing analyzer (TRCE in the Xilinx Development System)
and back-annotated to the simulation netlist. These path delays, provided as a guideline, have been extracted from the static
timing analyzer report. All timing parameters assume worst-case operating conditions (supply voltage and junction
temperature).
Speed Grade -4 -3
Units
Description Symbol Device Max Max
From pad through Primary buffer, to any clock K TPG XCS05 2.0 4.0 ns
XCS10 2.4 4.3 ns
XCS20 2.8 5.4 ns
XCS30 3.2 5.8 ns
XCS40 3.5 6.4 ns
From pad through Secondary buffer, to any clock K TSG XCS05 2.5 4.4 ns
XCS10 2.9 4.7 ns
XCS20 3.3 5.8 ns
XCS30 3.6 6.2 ns
XCS40 3.9 6.7 ns

R
Spartan CLB Switching Characteristic Guidelines

parameters assume worst-case operating conditions (supply voltage and junction temperature). Values apply to all Spartan
devices and expressed in nanoseconds unless otherwise noted.
Speed Grade -4 -3
Units
Description Symbol Min Max Min Max
Clocks
Clock High time TCH 3.0 4.0 ns
Clock Low time TCL 3.0 4.0 ns
F/G inputs to X/Y outputs TILO 1.2 1.6 ns
F/G inputs via H to X/Y outputs TIHO 2.0 2.7 ns
C inputs via H1 via H to X/Y outputs THH1O 1.7 2.2 ns
CLB Fast Carry Logic
Operand inputs (F1, F2, G1, G4) to COUT TOPCY 1.7 2.1 ns
Add/Subtract input (F3) to COUT TASCY 2.8 3.7 ns
Initialization inputs (F1, F3) to COUT TINCY 1.2 1.4 ns
CIN through function generators to X/Y outputs TSUM 2.0 2.6 ns
CIN to COUT, bypass function generators TBYP 0.5 0.6 ns
Sequential Delays
Clock K to Flip-Flop outputs Q TCKO 2.1 2.8 ns
Setup Time before Clock K
F/G inputs TICK 1.8 2.4 ns
F/G inputs via H TIHCK 2.9 3.9 ns
C inputs via H1 through H THH1CK 2.3 3.3 ns
C inputs via DIN TDICK 1.3 2.0 ns
C inputs via EC TECCK 2.0 2.6 ns
C inputs via S/R, going Low (inactive) TRCK 2.5 4.0 ns
Hold Time after Clock K
All Hold times, all devices 0.0 0.0 ns
Set/Reset Direct
Width (High) TRPW 3.0 4.0 ns
Delay from C inputs via S/R, going High to Q TRIO 3.0 4.0 ns
Global Set/Reset
Minimum GSR Pulse Width TMRW 11.5 13.5 ns
Delay from GSR input to any Q TMRQ See page 45 for TRRI values per device.
Toggle Frequency (MHz) FTOG 166 125 MHz
(for export control purposes)

R
Spartan CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines

devices and are expressed in nanoseconds unless otherwise noted.
Speed Grade -4 -3
Single Port RAM Units
Size Symbol Min Max Min Max
Write Operation
Address write cycle time (clock K period) 16x2 TWCS 8.0 11.6 ns
32x1 TWCTS 8.0 11.6 ns
Clock K pulse width (active edge) 16x2 TWPS 4.0 5.8 ns
32x1 TWPTS 4.0 5.8 ns
Address setup time before clock K 16x2 TASS 1.5 2.0 ns 4
32x1 TASTS 1.5 2.0 ns
Address hold time after clock K 16x2 TAHS 0.0 0.0 ns
32x1 TAHTS 0.0 0.0 ns
DIN setup time before clock K 16x2 TDSS 1.5 2.7 ns
32x1 TDSTS 1.5 1.7 ns
DIN hold time after clock K 16x2 TDHS 0.0 0.0 ns
32x1 TDHTS 0.0 0.0 ns
WE setup time before clock K 16x2 TWSS 1.5 1.6 ns
32x1 TWSTS 1.5 1.6 ns
WE hold time after clock K 16x2 TWHS 0.0 0.0 ns
32x1 TWHTS 0.0 0.0 ns
Data valid after clock K 16x2 TWOS 6.5 7.9 ns
32x1 TWOTS 7.0 9.3 ns
Read Operation
Address read cycle time 16x2 TRC 2.6 2.6 ns
32x1 TRCT 3.8 3.8 ns
Data Valid after address change (no Write 16x2 TILO 1.2 1.6 ns
Enable) 32x1 TIHO 2.0 2.7 ns
Address setup time before clock K 16x2 TICK 1.8 2.4 ns
32x1 TIHCK 2.9 3.9 ns
Note: Timing for 16 x 1 RAM option is identical to 16 x 2 RAM timing.

R
Spartan CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines (continued)

devices and are expressed in nanoseconds unless otherwise noted.
Speed Grade -4 -3
Dual Port RAM Units
Write Operation
Address write cycle time (clock K period) 16x1 TWCDS 8.0 11.6 ns
Clock K pulse width (active edge) 16x1 TWPDS 4.0 5.8 ns
Address setup time before clock K 16x1 TASDS 1.5 2.1 ns
Address hold time after clock K 16x1 TAHDS 0.0 0.0 ns
DIN setup time before clock K 16x1 TDSDS 1.5 1.6 ns
DIN hold time after clock K 16x1 TDHDS 0.0 0.0 ns
WE setup time before clock K 16x1 TWSDS 1.5 1.6 ns
WE hold time after clock K 16x1 TWHDS 0.0 0.0 ns
Data valid after clock K 16x1 TWODS 6.5 7.0 ns
Note 1: Read Operation Timing for 16x1 dual-port RAM option is identical to 16x2 single-port RAM timing.
Spartan CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS TWPDS
WCLK (K) WCLK (K)
TWSS TWHS TWSDS TWHDS
WE WE
TDSS TDHS TDSDS TDHDS
DATA IN DATA IN
TASS TAHS TASDS TAHDS
ADDRESS ADDRESS
TILO TILO TILO

TILO
TWOS
TWODS
DATA OUT OLD NEW DATA OUT OLD NEW
X6461 X6474
Single Port Dual Port

R
Spartan Pin-to-Pin Output Parameter Guidelines

functionally tested. Pin-to-pin timing parameters are derived from measuring external and internal test patterns and are
guaranteed over worst-case operating conditions (supply voltage and junction temperature). Listed below are representative
values for typical pin locations and normal clock loading. For more specific, more precise, and worst-case guaranteed data,
reflecting the actual routing structure, use the values provided by the static timing analyzer (TRCE in the Xilinx Development
System) and back-annotated to the simulation netlist. These path delays, provided as a guideline, have been extracted from
the static timing analyzer report.
Spartan Output Flip-Flop, Clock-to-Out

Speed Grade -4 -3
Units
Global Primary Clock to TTL Output using OFF
Fast TICKOF XCS05 5.3 8.7 ns
XCS10 5.7 9.1 ns
XCS20 6.1 9.3 ns
XCS30 6.5 9.4 ns 4
XCS40 6.8 10.2 ns
Slew-rate limited TICKO XCS05 9.0 11.5 ns
XCS10 9.4 12.0 ns
XCS20 9.8 12.2 ns
XCS30 10.2 12.8 ns
XCS40 10.5 12.8 ns
Global Secondary Clock to TTL Output using OFF
Fast TICKSOF XCS05 5.8 9.2 ns
XCS10 6.2 9.6 ns
XCS20 6.6 9.8 ns
XCS30 7.0 9.9 ns
XCS40 7.3 10.7 ns
Slew-rate limited TICKSO XCS05 9.5 12.0 ns
XCS10 9.9 12.5 ns
XCS20 10.3 12.7 ns
XCS30 10.7 13.2 ns
XCS40 11.0 14.3 ns
Delay Adder for CMOS Outputs Option
Fast TCMOSOF All devices 0.8 1.0 ns
Slew-rate Limited TCMOSO All devices 1.5 2.0 ns
OFF = Output Flip-Flop
Note 1: Listed above are representative values where one global clock input drives one vertical clock line in each accessible column,
and where all accessible IOB and CLB flip-flops are clocked by the global clock net.
Note 2: Output timing is measured at ~50% VCC threshold with 50 pF external capacitive load. For different loads, see Figure 31.
Capacitive Load Factor Figure 31: Delay Factor at Various Capacitive Loads
3
Figure 31 shows the relationship between I/O output delay
and load capacitance. It allows a user to adjust the speci- 2
Delta Delay (ns)
fied output delay if the load capacitance is different than 1

50 pF. For example, if the actual load capacitance is
0
120 pF, add 2.5 ns to the specified delay. If the load capac-
itance is 20 pF, subtract 0.8 ns from the specified output -1
delay. Figure 31 is usable over the specified operating con- -2

ditions of voltage and temperature and is independent of 0 20 40 60 80 100 120 140
Capacitance (pF)
the output slew rate control. X8257

R
Spartan Pin-to-Pin Input Parameter Guidelines

values for typical pin locations and normal clock loading.
Spartan Primary and Secondary Setup and Hold

Speed Grade -4 -3
Units
Description Symbol Device Min Min
Input Setup/Hold Times Using Primary Clock and IFF
No Delay TPSUF/TPHF XCS05 1.2 / 1.7 1.8 / 2.5 ns
XCS10 1.0 / 2.3 1.5 / 3.4 ns
XCS20 0.8 / 2.7 1.2 / 4.0 ns
XCS30 0.6 / 3.0 0.9 / 4.5 ns
XCS40 0.4 / 3.5 0.6 / 5.2 ns
With Delay TPSU/TPH XCS05 4.3 / 0.0 6.0 / 0.0 ns
XCS10 4.3 / 0.0 6.0 / 0.0 ns
XCS20 4.3 / 0.0 6.0 / 0.0 ns
XCS30 4.3 / 0.0 6.0 / 0.0 ns
XCS40 5.3 / 0.0 6.8 / 0.0 ns
Input Setup/Hold Times Using Secondary Clock and IFF
No Delay TSSUF/TSHF XCS05 0.9 / 2.2 1.5 / 3.0 ns
XCS10 0.7 / 2.8 1.2 / 3.9 ns
XCS20 0.5 / 3.2 0.9 / 4.5 ns
XCS30 0.3 / 3.5 0.6 / 5.0 ns
XCS40 0.1 / 4.0 0.3 / 5.7 ns
With Delay TSSU/TSH XCS05 4.0 / 0.0 5.7 / 0.0 ns
XCS10 4.0 / 0.0 5.7 / 0.0 ns
XCS20 4.0 / 0.5 5.7 / 0.5 ns
XCS30 4.0 / 0.5 5.7 / 0.5 ns
XCS40 5.0 / 0.0 6.5 / 0.0 ns
IFF = Input Flip-Flop or Latch
Note 1: Setup time is measured with the fastest route and the lightest load. Hold time is measured using the furthest distance and a
reference load of one clock pin per IOB/CLB.

R
Spartan IOB Input Switching Characteristic Guidelines

static timing analyzer (TRCE in the Xilinx Development System) and back-annotated to the simulation netlist. These path
delays, provided as a guideline, have been extracted from the static timing analyzer report. All timing parameters assume
worst-case operating conditions (supply voltage and junction temperature).
Speed Grade -4 -3
Units
Description Symbol Device Min Max Min Max
Setup Times - TTL Inputs (Note 1)
Clock Enable (EC) to Clock (IK), no delay TECIK All devices 1.6 2.1 ns
Pad to Clock (IK), no delay TPICK All devices 1.5 2.0 ns
Hold Times
Clock Enable (EC) to Clock (IK), no delay TIKEC All devices 0.0 0.9 ns
All Other Hold Times All devices 0.0 0.0 ns
Propagation Delays - TTL Inputs (Note 1)
Pad to I1, I2 TPID All devices 1.5 2.0 ns 4
Pad to I1, I2 via transparent input latch, no delay TPLI All devices 2.8 3.6 ns
Clock (IK) to I1, I2 (flip-flop) TIKRI All devices 2.7 2.8 ns
Clock (IK) to I1, I2 (latch enable, active Low) TIKLI All devices 3.2 3.9 ns
Delay Adder for Input with Delay Option
TECIKD = TECIK + TDelay TDelay XCS05 3.6 4.0 ns
TPICKD = TPICK + TDelay XCS10 3.7 4.1 ns
TPDLI = TPLI + TDelay XCS20 3.8 4.2 ns
XCS30 4.5 5.0 ns
XCS40 5.5 5.5 ns
Global Set/Reset
Minimum GSR Pulse Width TMRW All devices 11.5 13.5 ns
Delay from GSR input to any Q TRRI XCS05 9.0 11.3 ns
XCS10 9.5 11.9 ns
XCS20 10.0 12.5 ns
XCS30 10.5 13.1 ns
XCS40 11.0 13.8 ns
Note 1: Delay adder for CMOS Inputs option: for -3 speed grade, add 0.4 ns; for -4 speed grade, add 0.2 ns.
Note 2: Input pad setup and hold times are specified with respect to the internal clock (IK). For setup and hold times with respect to the
clock input, see the pin-to-pin parameters in the Pin-to-Pin Input Parameters table.
Note 3: Voltage levels of unused pads, bonded or unbonded, must be valid logic levels. Each can be configured with the internal pull-up
(default) or pull-down resistor, or configured as a driven output, or can be driven from an external source.

R
Spartan IOB Output Switching Characteristic Guidelines

worst-case operating conditions (supply voltage and junction temperature). Values are expressed in nanoseconds unless
otherwise noted.
Speed Grade -4 -3
Units
Clocks
Clock High TCH All devices 3.0 4.0 ns
Clock Low TCL All devices 3.0 4.0 ns
Propagation Delays - TTL Outputs (Notes 1, 2)
Clock (OK) to Pad, fast TOKPOF All devices 3.3 4.5 ns
Clock (OK to Pad, slew-rate limited TOKPOS All devices 6.9 7.0 ns
Output (O) to Pad, fast TOPF All devices 3.6 4.8 ns
Output (O) to Pad, slew-rate limited TOPS All devices 7.2 7.3 ns
3-state to Pad hi-Z (slew-rate independent) TTSHZ All devices 3.0 3.8 ns
3-state to Pad active and valid, fast TTSONF All devices 6.0 7.3 ns
3-state to Pad active and valid, slew-rate limited TTSONS All devices 9.6 9.8 ns
Setup and Hold Times
Output (O) to clock (OK) setup time TOOK All devices 2.5 3.8 ns
Output (O) to clock (OK) hold time TOKO All devices 0.0 0.0 ns
Clock Enable (EC) to clock (OK) setup time TECOK All devices 2.0 2.7 ns
Clock Enable (EC) to clock (OK) hold time TOKEC All devices 0.0 0.5 ns
Global Set/Reset
Minimum GSR pulse width TMRW All devices 11.5 13.5 ns
Delay from GSR input to any Pad TRPO XCS05 12.0 15.0 ns
XCS10 12.5 15.7 ns
XCS20 13.0 16.2 ns
XCS30 13.5 16.9 ns
XCS40 14.0 17.5 ns
Note 1: Delay adder for CMOS Outputs option (with fast slew rate option): for -3 speed grade, add 1.0 ns; for -4 speed grade, add 0.8 ns.
Note 2: Delay adder for CMOS Outputs option (with slow slew rate option): for -3 speed grade, add 2.0 ns; for -4 speed grade, add 1.5 ns.
Note 3: Output timing is measured at ~50% VCC threshold, with 50 pF external capacitive loads including test fixture. Slew-rate limited
output rise/fall times are approximately two times longer than fast output rise/fall times.

R
SpartanXL Detailed Specifications

Definition of Terms
follows:
Advance: Initial estimates based on simulation and/or extrapolation from other speed grades, devices, or device
families. Values are subject to change. Use as estimates, not for production.
Notwithstanding the definition of the above terms, all specifications are subject to change without notice.
SpartanXL Absolute Maximum Ratings1

VCC Supply voltage relative to GND -0.5 to 4.0 V
5V Tolerant I/O Checked2, 3 -0.5 to 5.5 V
VIN Input voltage relative to GND (Note 2, 3, 4, 5)
Not 5V Tolerant I/Os4, 5 -0.5 to VCC + 0.5 V 4
5V Tolerant I/O Checked2, 3 -0.5 to 5.5 V
VTS Voltage applied to 3-state output (Note 2, 3, 4, 5)
Not 5V Tolerant I/Os4, 5 -0.5 to VCC + 0.5 V
VCCt Longest Supply Voltage Rise Time from 1V to 3V 50 ms
TJ Junction temperature Plastic packages +125 °C
Note 1. Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are
2: With 5V Tolerant I/Os Selected, the Maximum DC overshoot (above VCC) must be limited to either 5.5 V or 10 mA and
undershoot (below GND) must be limited to either 0.5 V or 10 mA, whichever is easier to achieve.
3: With 5V Tolerant I/Os Selected, the Maximum AC (during transitions) conditions are as follows; the device pins may
undershoot to -2.0 V or overshoot to + 7.0 V, provided this overshoot or undershoot lasts no more than 11 ns with a forcing
current no greater than 100 mA.
4: Without 5V Tolerant I/Os Selected, the Maximum DC overshoot or undershoot must be limited to either 0.5 V or 10 mA,
whichever is easier to achieve.
5: Without 5V Tolerant I/Os Selected, the Maximum AC conditions are as follows; the device pins may undershoot to -2.0 V or
overshoot to VCC + 2.0 V, provided this overshoot or undershoot lasts no more than 11 ns with a forcing current no greater
than 100 mA.
SpartanXL Recommended Operating Conditions

Supply voltage relative to GND, TJ = 0°C to +85°C Commercial 3.0 3.6 V
VCC Supply voltage relative to GND, TJ = -40°C to Industrial 3.0 3.6 V
+100°C
VIH High-level input voltage 50% of VCC 5.5 V
VIL Low-level input voltage 0 30% of VCC V
Notes: At junction temperatures above those listed as Operating Conditions, all delay parameters increase by 0.35% per °C.

R
SpartanXL DC Characteristics Over Operating Conditions

High-level output voltage @ IOH = -4.0 mA, VCC min (LVTTL) 2.4 V
VOH
High-level output voltage @ IOH = -500 µA, (LVCMOS) 90% VCC V
Low-level output voltage @ IOL = 12.0 mA, VCC min (LVTTL) (Note 1) 0.4 V
VOL
Low-level output voltage @ IOL = 1500 µA, (LVCMOS) 10% VCC V
VDR Data Retention Supply Voltage (below which configuration data may be lost) 2.5 V
ICCO Quiescent FPGA supply current (Note 3) 5 mA
ICCPD Power Down FPGA supply current (Note 4) 100 µA
IRPU Pad pull-up (when selected) @ Vin = 0V (sample tested) 0.02 0.25 mA
IRPD Pad pull-down (when selected) @ Vin = 3.3V (sample tested) 0.02 mA
Note 1: With up to 64 pins simultaneously sinking 12 mA (default mode).
Note 2: With up to 64 pins simultaneously sinking 24mA (with 24 mA option selected).
Note 3: With no output current loads, no active input pull-up resistors, all package pins at Vcc or GND, and the FPGA configured with
the Tie option.
Note 4: With PWRDWN active and 5V tolerance not selected. Typical value.
SpartanXL Global Buffer Switching Characteristic Guidelines

temperature).
Speed Grade -5 -4
Units
From pad through buffer, to any clock K TGLS XCS05XL 1.1 1.5 ns
XCS10XL 1.5 1.8 ns
XCS20XL 1.8 2.1 ns
XCS30XL 2.1 2.5 ns
XCS40XL 2.4 2.8 ns
Advance Preliminary

R
SpartanXL CLB Switching Characteristic Guidelines

parameters assume worst-case operating conditions (supply voltage and junction temperature). Values apply to all
SpartanXL devices and expressed in nanoseconds unless otherwise noted.
Speed Grade -5 -4
Units
Clocks
F/G inputs via transparent latch to Q outputs TITO 1.5 1.8 ns
C inputs via H1 via H to X/Y outputs THH1O 1.5 1.8 ns 4
Sequential Delays
Clock K to Flip-Flop or latch outputs Q TCKO 1.2 1.4 ns
All Hold times, all devices 0.0 0.0 ns
Set/Reset Direct
Global Set/Reset
Minimum GSR Pulse Width TMRW 10.5 11.5 ns
Toggle Frequency (MHz) FTOG 200 179 MHz
(for export control purposes)
Advance Preliminary

R
SpartanXL CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines

SpartanXL devices and are expressed in nanoseconds unless otherwise noted.
Speed Grade -5 -4
Size1 Symbol Min Max Min Max
Write Operation
32x1 TWCTS 7.7 8.4 ns
Clock K pulse width (active edge) 16x2 TWPS 3.1 3.6 ns
32x1 TWPTS 3.1 3.6 ns
Address setup time before clock K 16x2 TASS 1.3 1.5 ns
32x1 TASTS 1.5 1.7 ns
32x1 TDSTS 1.8 2.1 ns
32x1 TWSTS 1.3 1.5 ns
All hold times after clock K 0.0 0.0 ns
32x1 TWOTS 5.4 6.3 ns
Read Operation
32x1 TRCT 3.8 5.5 ns
Data Valid after address change (no Write 16x2 TILO 1.0 1.1 ns
Enable) 32x1 TIHO 1.7 2.0 ns
32x1 TIHCK 1.3 1.6 ns
Advance Preliminary
Note 1: Timing for 16 x 1 RAM option is identical to 16 x 2 RAM timing.

R
SpartanXL CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines (cont.)

SpartanXL devices and are expressed in nanoseconds unless otherwise noted.
Speed Grade -5 -4
Dual Port RAM Units
Size1 Symbol Min Max Min Max
Write Operation
Clock K pulse width (active edge) 16x1 TWPDS 3.9 3.6 ns
All hold time after clock K 16x1 0.0 0.0 ns 4
Advance Preliminary
Note 1: Read Operation Timing for 16x1 dual-port RAM option is identical to 16x2 single-port RAM timing.
Spartan CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS TWPDS
WCLK (K) WCLK (K)
WE WE
DATA IN DATA IN
ADDRESS ADDRESS
TILO TILO TILO

TILO
TWOS
TWODS
X6461 X6474
Single Port Dual Port

R
SpartanXL Pin-to-Pin Output Parameter Guidelines

SpartanXL Output Flip-Flop, Clock-to-Out

Speed Grade -5 -4
Units
Global Clock to Output using OFF
Fast TICKOF XCS05XL 4.4 5.2 ns
XCS10XL 4.7 5.5 ns
XCS20XL 5.0 5.8 ns
XCS30XL 5.3 6.2 ns
XCS40XL 5.6 6.5 ns
Slew Rate Adjustment
For output slow option add TSLOW All Devices 1.5 1.7 ns
OFF = Output Flip Flop Advance Prelim
Note 1: Output delays are representative values where one global clock input drives one vertical clock line in each accessible column,
Note 2: Output timing is measured at ~50% VCC threshold with 50 pF external capacitive load.
SpartanXL Pin-to-Pin Input Parameter Guidelines

SpartanXL Setup and Hold

Speed Grade -5 -4
Units
Description Symbol Device Min Min
Input Setup/Hold Times Using Global Clock and IFF
No Delay TSUF/THF XCS05XL 1.1/2.0 1.6/2.6 ns
XCS10XL 1.0/2.2 1.5/2.8 ns
XCS20XL 0.9/2.4 1.4/3.0 ns
XCS30XL 0.8/2.6 1.3/3.2 ns
XCS40XL 0.7/2.8 1.2/3.4 ns
Full Delay TSU/TH XCS05XL 3.9/0.0 5.1/0.0 ns
XCS10XL 4.1/0.0 5.3/0.0 ns
XCS20XL 4.3/0.0 5.5/0.0 ns
XCS30XL 4.5/0.0 5.7/0.0 ns
XCS40XL 4.7/0.0 5.9/0.0 ns
IFF = Input Flip-Flop or Latch Advance Prelim
Note 3: Setup time is measured with the fastest route and the lightest load. Hold time is measured using the furthest distance and a
reference load of one clock pin per IOB/CLB.

R
Capacitive Load Factor 3
Figure 32 shows the relationship between I/O output delay 2
Delta Delay (ns)

and load capacitance. It allows a user to adjust the speci- 1
fied output delay if the load capacitance is different than
0
120 pF, add 2.5 ns to the specified delay. If the load capac- -1
itance is 20 pF, subtract 0.8 ns from the specified output
-2
delay. Figure 32 is usable over the specified operating con- 0 20 40 60 80 100 120 140
ditions of voltage and temperature and is independent of Capacitance (pF)
X8257
the output slew rate control.

Figure 32: Delay Factor at Various Capacitive Loads

R
SpartanXL IOB Input Switching Characteristic Guidelines

Speed Grade -5 -4
Units
Setup Times
Clock Enable (EC) to Clock (IK) TECIK All devices 0.0 0.0 ns
Pad to Fast Capture Latch Enable (OK), no delay TPOCK All devices 0.7 0.8 ns
Hold Times
All Hold Times All devices 0.0 0.0 ns
Propagation Delays
Pad to I1, I2 TPID All devices 0.9 1.1 ns
Pad to I1, I2 via transparent input latch, no delay TPLI All devices 2.1 2.5 ns
Delay Adder for Input with Delay Option
TPICKD = TPICK + TDelay TDelay XCS05XL 4.0 4.7 ns
TPDLI = TPLI + TDelay XCS10XL 4.8 5.6 ns
XCS20XL 5.0 5.9 ns
XCS30XL 5.5 6.5 ns
XCS40XL 6.5 7.6 ns
Global Set/Reset
Minimum GSR Pulse Width TMRW All devices 10.5 11.5 ns
Delay from GSR input to any Q TRRI XCS05XL 9.0 10.5 ns
XCS10XL 9.5 11.0 ns
XCS20XL 10.0 11.5 ns
XCS30XL 11.0 12.5 ns
XCS40XL 12.0 13.5 ns
Advance Preliminary
Note 1: Input pad setup and hold times are specified with respect to the internal clock (IK). For setup and hold times with respect to the
clock input, see the pin-to-pin parameters in the Pin-to-Pin Input Parameters table.

R
SpartanXL IOB Output Switching Characteristic Guidelines

worst-case operating conditions (supply voltage and junction temperature). Values are expressed in nanoseconds unless
otherwise noted.
Speed Grade -5 -4
Units
Clocks
Clock High TCH All devices 2.0 2.3 ns
Clock Low TCL All devices 2.0 2.3 ns
Propagation Delays
Clock (OK) to Pad, fast TOKPOF All devices 3.2 3.7 ns
Output (O) to Pad, fast TOPF All devices 2.5 2.9 ns 4
3-state to Pad hi-Z (slew-rate independent) TTSHZ All devices 2.8 3.3 ns
3-state to Pad active and valid, fast TTSONF All devices 2.6 3.0 ns
Output (O) to Pad via Output Mux, fast TOFPF All devices 3.7 4.4 ns
Select (OK) to Pad via Output Mux, fast TOKFPF All devices 3.3 3.9 ns
For Output SLOW option add TSLOW All devices 1.5 1.7 ns
Output (O) to clock (OK) setup time TOOK All devices 0.5 0.5 ns
Output (O) to clock (OK) hold time TOKO All devices 0.0 0.0 ns
Clock Enable (EC) to clock (OK) setup time TECOK All devices 0.0 0.0 ns
Clock Enable (EC) to clock (OK) hold time TOKEC All devices 0.1 0.1 ns
Global Set/Reset
Minimum GSR pulse width TMRW All devices 10.5 11.5 ns
Delay from GSR input to any Pad TRPO XCS05XL 11.9 14.0 ns
XCS10XL 12.4 14.5 ns
XCS20XL 12.9 15.0 ns
XCS30XL 13.9 16.0 ns
XCS40XL 14.9 17.0 ns
Advance Preliminary
Note 1: Output timing is measured at ~50% VCC threshold, with 50 pF external capacitive loads including test fixture. Slew-rate limited
output rise/fall times are approximately two times longer than fast output rise/fall times.
Note 2: Voltage levels of unused pads, bonded or unbonded, must be valid logic levels. Each can be configured with the internal
pull-up (default) or pull-down resistor, or configured as a driven output, or can be driven from an external source.

Pin Descriptions unused it is configured as an input with the I/O pull-up

resistor network remaining activated.
There are three types of pins in the Spartan Series devices:
Any user I/O can be configured to drive the Global
• Permanently dedicated pins Set/Reset net GSR or the global three-state net, GTS. See
• User I/O pins that can have special functions “Global Signals: GSR and GTS” on page 21 for more infor-
• Unrestricted user-programmable I/O pins. mation.
Before and during configuration, all outputs not used for the Device pins for Spartan Series devices are described in
configuration process are 3-stated with the I/O pull-up Table 18.
resistor network activated. After configuration, if an IOB is
Table 18: Pin Descriptions
I/O I/O
During After
Pin Name Config. Config. Pin Description
Permanently Dedicated Pins
Eight or more (depending on package) connections to the nominal +5 V supply voltage
VCC X X (+3.3 V for SpartanXL devices). All must be connected, and each must be decoupled
with a 0.01 - 0.1 µF capacitor to Ground.
Eight or more (depending on package type) connections to Ground. All must be con-
GND X X
nected.
During configuration, Configuration Clock (CCLK) is an output in Master mode and is an
input in Slave mode. After configuration, CCLK has a weak pull-up resistor and can be
selected as the Readback Clock. There is no CCLK High or Low time restriction on
CCLK I or O I
Spartan Series devices, except during Readback. See “Violating the Maximum High
and Low Time Specification for the Readback Clock” on page 35 for an explanation of
this exception.
DONE is a bidirectional signal with an optional internal pull-up resistor. As an output, it
indicates the completion of the configuration process. As an input, a Low level on DONE
DONE I/O O can be configured to delay the global logic initialization and the enabling of outputs.
The optional pull-up resistor is selected as an option in the program that creates the
configuration bitstream. The resistor is included by default.
PROGRAM is an active Low input that forces the FPGA to clear its configuration mem-
ory. It is used to initiate a configuration cycle. When PROGRAM goes High, the FPGA
finishes the current clear cycle and executes another complete clear cycle, before it
PROGRAM I I
goes into a WAIT state and releases INIT.
The PROGRAM pin has a permanent weak pull-up, so it need not be externally pulled
up to Vcc.
The Mode input(s) are sampled after INIT goes High to determine the configuration
MODE
mode to be used.
(Spartan)
I X During configuration, these pins have a weak pull-up resistor. For the most popular con-
M0, M1
figuration mode, Slave Serial, the mode pins can be left unconnected. For Master Serial
(SpartanXL)
mode, connect the Mode/M0 pin directly to system ground.
PWRDWN is an active Low input that forces the FPGA into the Power Down state and
reduces power consumption. When PWRDWN is Low, the FPGA disables all I/O and
initializes all flip-flops. All inputs are interpreted as Low independent of their actual level.
VCC must be maintained, and the configuration data is maintained. PWRDWN halts
PWRDWN I I
configuration if asserted before or during configuration, and re-starts configuration when
removed. When PWRDWN returns High, the FPGA becomes operational by first en-
abling the inputs and flip-flops and then enabling the outputs. PWRDWN has a default
internal pull-up resistor.
Don’t Connect Pins reserved for factory testing and possible future enhancements. Pins must be left
X X
(Spartan) floating.

R
Table 18: Pin Descriptions (Continued)
I/O I/O
During After
User I/O Pins That Can Have Special Functions
If boundary scan is used, this pin is the Test Data Output. If boundary scan is not used,
this pin is a 3-state output without a register, after configuration is completed.
TDO O O
To use this pin, place the library component TDO instead of the usual pad symbol. An
output buffer must still be used.
If boundary scan is used, these pins are Test Data In, Test Clock, and Test Mode Select
inputs respectively. They come directly from the pads, bypassing the IOBs. These pins
can also be used as inputs to the CLB logic after configuration is completed.
I/O
TDI, TCK, If the BSCAN symbol is not placed in the design, all boundary scan functions are inhib-
I or I
TMS ited once configuration is completed, and these pins become user-programmable I/O.
(JTAG)
In this case, they must be called out by special library elements. To use these pins,
place the library components TDI, TCK, and TMS instead of the usual pad symbols. In-
put or output buffers must still be used.
High During Configuration (HDC) is driven High until the I/O go active. It is available as 4
HDC O I/O a control output indicating that configuration is not yet completed. After configuration,
HDC is a user-programmable I/O pin.
Low During Configuration (LDC) is driven Low until the I/O go active. It is available as a
LDC O I/O control output indicating that configuration is not yet completed. After configuration,
LDC is a user-programmable I/O pin.
Before and during configuration, INIT is a bidirectional signal. A 1 kΩ - 10 kΩ external
pull-up resistor is recommended.
As an active-Low open-drain output, INIT is held Low during the power stabilization and
internal clearing of the configuration memory. As an active-Low input, it can be used
INIT I/O I/O
to hold the FPGA in the internal WAIT state before the start of configuration. Master
mode devices stay in a WAIT state an additional 30 to 300 µs after INIT has gone High.
During configuration, a Low on this output indicates that a configuration data error has
occurred. After the I/O go active, INIT is a user-programmable I/O pin.
Four Primary Global inputs each drive a dedicated internal global net with short delay
and minimal skew. If not used to drive a global buffer, any of these pins is a user-pro-
PGCK1 -
Weak grammable I/O.
PGCK4 I or I/O
Pull-up The PGCK1-PGCK4 pins drive the four Primary Global Buffers. Any input pad symbol
(Spartan)
connected directly to the input of a BUFGP symbol is automatically placed on one of
these pins.
Four Secondary Global inputs each drive a dedicated internal global net with short delay
and minimal skew. These internal global nets can also be driven from internal logic. If
SGCK1 -
Weak not used to drive a global net, any of these pins is a user-programmable I/O pin.
SGCK4 I or I/O
Pull-up The SGCK1-SGCK4 pins provide the shortest path to the four Secondary Global Buff-
(Spartan)
ers. Any input pad symbol connected directly to the input of a BUFGS symbol is auto-
matically placed on one of these pins.
Eight Global inputs each drive a dedicated internal global net with short delay and min-
imal skew. These internal global nets can also be driven from internal logic. If not used
GCK1 - GCK8 Weak to drive a global net, any of these pins is a user-programmable I/O pin.
I or I/O
(SpartanXL) Pull-up The GCK1-GCK8 pins provide the shortest path to the eight Global Low-Skew Buffers.
Any input pad symbol connected directly to the input of a BUFGLS symbol is automati-
cally placed on one of these pins.
CS1 During Express configuration, CS1 is used as a serial-enable signal for daisy-chaining.
I I/O
(SpartanXL)
D0-D7 During Express configuration, these eight input pins receive configuration data. After
I I/O
(SpartanXL) configuration, they are user-programmable I/O pins.

R
I/O I/O
During After
During Slave Serial or Master Serial configuration, DIN is the serial configuration data
DIN I I/O input receiving data on the rising edge of CCLK. After configuration, DIN is a user-pro-
grammable I/O pin.
During Slave Serial or Master Serial configuration, DOUT is the serial configuration data
output that can drive the DIN of daisy-chained slave FPGAs. DOUT data changes on
the falling edge of CCLK, one-and-a-half CCLK periods after it was received at the DIN
DOUT O I/O input.
In SpartanXL Express mode, DOUT is the status output that can drive the CS1 of dai-
sy-chained FPGAs, to enable and disable downstream devices.
After configuration, DOUT is a user-programmable I/O pin.
Unrestricted User-Programmable I/O Pins
These pins can be configured to be input and/or output after configuration is completed.
Weak
I/O I/O Before configuration is completed, these pins have an internal high-value pull-up resis-
Pull-up
tor network that defines the logic level as High.
Device-Specific Pinout Tables

Device-specific tables include all packages for each Spartan and SpartanXL device. They follow the pad locations around
the die, and include boundary scan register locations.
Pin Locations for XCS05 & XCS05XL Devices

XCS05/XL XCS05/XL
PC84 VQ100 Bndry Scan PC84 VQ100 Bndry Scan
Pad Name Pad Name
VCC P2 P89 - VCC P33 P25 -
I/O P3 P90 32 Don’t Connect †, P34 P26 126 †
I/O P4 P91 35 PWRDWN ††
I/O - P92 38 I/O, PGCK2 †, GCK3 †† P35 P27 127 ‡
I/O - P93 41 I/O (HDC) P36 P28 130 ‡
I/O P5 P94 44 I/O - P29 133 ‡
I/O P6 P95 47 I/O (LDC) P37 P30 136 ‡
I/O P7 P96 50 I/O P38 P31 139 ‡
I/O P8 P97 53 I/O P39 P32 142 ‡
I/O P9 P98 56 I/O - P33 145 ‡
I/O, SGCK1 †, GCK8 †† P10 P99 59 I/O - P34 148 ‡
VCC P11 P100 - I/O P40 P35 151 ‡
GND P12 P1 - I/O (INIT) P41 P36 154 ‡
I/O, PGCK1 †, GCK1 †† P13 P2 62 VCC P42 P37 -
I/O P14 P3 65 GND P43 P38 -
I/O, TDI P15 P4 68 I/O P44 P39 157 ‡
I/O, TCK P16 P5 71 I/O P45 P40 160 ‡
I/O, TMS P17 P6 74 I/O - P41 163 ‡
I/O P18 P7 77 I/O - P42 166 ‡
I/O - P8 83 I/O P46 P43 169 ‡
I/O P19 P9 86 I/O P47 P44 172 ‡
I/O P20 P10 89 I/O P48 P45 175 ‡
GND P21 P11 - I/O P49 P46 178 ‡
VCC P22 P12 - I/O P50 P47 181 ‡
I/O P23 P13 92 I/O, SGCK3 †, GCK4 †† P51 P48 184 ‡
I/O P24 P14 95 GND P52 P49 -
I/O - P15 98 DONE P53 P50 -
I/O P25 P16 104 VCC P54 P51 -
I/O P26 P17 107 PROGRAM P55 P52 -
I/O P27 P18 110 I/O (D7 ††) P56 P53 187 ‡
I/O - P19 113 I/O, PGCK3 †, GCK5 †† P57 P54 190 ‡
I/O P28 P20 116 I/O (D6 ††) P58 P55 193 ‡
I/O, SGCK2 †, GCK2 †† P29 P21 119 I/O - P56 196 ‡
Don’t Connect †, M1 †† P30 P22 122 I/O (D5 ††) P59 P57 199 ‡
GND P31 P23 - I/O P60 P58 202 ‡
MODE †, M0 †† P32 P24 125 I/O - P59 205 ‡
I/O - P60 208 ‡

R
XCS05/XL XCS05/XL
PC84 VQ100 Bndry Scan PC84 VQ100 Bndry Scan
Pad Name Pad Name
I/O (D4 ††) P61 P61 211 ‡ I/O, PGCK4 †, GCK7 †† P78 P79 5
I/O P62 P62 214 ‡ I/O (CS1 ††) P79 P80 8
VCC P63 P63 - I/O P80 P81 11
GND P64 P64 - I/O P81 P82 14
I/O (D3 ††) P65 P65 217 ‡ I/O P82 P83 17
I/O P66 P66 220 ‡ I/O - P84 20
I/O - P67 223 ‡ I/O - P85 23
I/O (D2 ††) P67 P68 229 ‡ I/O P83 P86 26
I/O P68 P69 232 ‡ I/O P84 P87 29
I/O (D1 ††) P69 P70 235 ‡ GND P1 P88 -
I/O P70 P71 238 ‡ 11/20/98
I/O (D0 ††, DIN) P71 P72 241 ‡ † = 5V Spartan only
I/O, SGCK4 †, GCK6 †† P72 P73 244 ‡
(DOUT) †† = 3V SpartanXL only
CCLK P73 P74 -
VCC P74 P75 -
‡ The “PWRDWN” on the XCS05XL is not part of the Boundary
Scan chain. For the XCS05XL, subtract 1 from all Boundary Scan
O, TDO P75 P76 0
GND P76 P77 -
numbers from GCK3 on (127 and higher).
I/O P77 P78 2
4
XCS10/XL Bndry XCS10/XL Bndry
PC84 VQ100 TQ144 PC84 VQ100 TQ144
Pad Name Scan Pad Name Scan
VCC P2 P89 P128 - GND - - P27 -
I/O P3 P90 P129 44 I/O P27 P18 P28 152
I/O P4 P91 P130 47 I/O - P19 P29 155
I/O - P92 P131 50 I/O - - P30 158
I/O - P93 P132 53 I/O - - P31 161
I/O P5 P94 P133 56 I/O P28 P20 P32 164
I/O P6 P95 P134 59 I/O, SGCK2 †, P29 P21 P33 167
I/O - - P135 62 GCK2 ††
I/O - - P136 65 Don’t Connect †, M1 †† P30 P22 P34 170
GND - - P137 - GND P31 P23 P35 -
I/O P7 P96 P138 68 MODE †, M0 †† P32 P24 P36 173
I/O P8 P97 P139 71 VCC P33 P25 P37 -
I/O - - P140 74 Don’t Connect †, P34 P26 P38 174 †
PWRDWN ††
I/O - - P141 77
I/O, PGCK2 †, GCK3 †† P35 P27 P39 175 ‡
I/O P9 P98 P142 80
I/O (HDC) P36 P28 P40 178 ‡
I/O, SGCK1 †, GCK8 †† P10 P99 P143 83
I/O - - P41 181 ‡
VCC P11 P100 P144 -
I/O - - P42 184 ‡
GND P12 P1 P1 -
I/O - P29 P43 187 ‡
I/O, PGCK1 †, GCK1 †† P13 P2 P2 86
I/O (LDC) P37 P30 P44 190 ‡
I/O P14 P3 P3 89
GND - - P45 -
I/O - - P4 92
I/O - - P46 193 ‡
I/O - - P5 95
I/O - - P47 196 ‡
I/O, TDI P15 P4 P6 98
I/O P38 P31 P48 199 ‡
I/O, TCK P16 P5 P7 101
I/O P39 P32 P49 202 ‡
GND - - P8 -
I/O - P33 P50 205 ‡
I/O - - P9 104
I/O - P34 P51 208 ‡
I/O - - P10 107
I/O P40 P35 P52 211 ‡
I/O, TMS P17 P6 P11 110
I/O (INIT) P41 P36 P53 214 ‡
I/O P18 P7 P12 113
VCC P42 P37 P54 -
I/O - - P13 116
GND P43 P38 P55 -
I/O - P8 P14 119
I/O P44 P39 P56 217 ‡
I/O P19 P9 P15 122
I/O P45 P40 P57 220 ‡
I/O P20 P10 P16 125
I/O - P41 P58 223 ‡
GND P21 P11 P17 -
I/O - P42 P59 226 ‡
VCC P22 P12 P18 -
I/O P46 P43 P60 229 ‡
I/O P23 P13 P19 128
I/O P47 P44 P61 232 ‡
I/O P24 P14 P20 131
I/O - - P62 235 ‡
I/O - P15 P21 134
I/O - - P63 238 ‡
I/O - - P22 137
GND - - P64 -
I/O P25 P16 P23 140
I/O P48 P45 P65 241 ‡
I/O P26 P17 P24 143
I/O P49 P46 P66 244 ‡
I/O - - P25 146
I/O - - P67 247 ‡
I/O - - P26 149
I/O - - P68 250 ‡

R

PC84 VQ100 TQ144 PC84 VQ100 TQ144
I/O P50 P47 P69 253 ‡ I/O, SGCK4 †, GCK6 †† P72 P73 P106 340 ‡
I/O, SGCK3 †, GCK4 †† P51 P48 P70 256 ‡ (DOUT)
GND P52 P49 P71 - CCLK P73 P74 P107 -
DONE P53 P50 P72 - VCC P74 P75 P108 -
VCC P54 P51 P73 - O, TDO P75 P76 P109 0
PROGRAM P55 P52 P74 - GND P76 P77 P110 -
I/O (D7 ††) P56 P53 P75 259 ‡ I/O P77 P78 P111 2
I/O, PGCK3 †, GCK5 †† P57 P54 P76 262 ‡ I/O, PGCK4 †, GCK7 †† P78 P79 P112 5
I/O - - P77 265 ‡ I/O - - P113 8
I/O - - P78 268 ‡ I/O - - P114 11
I/O (D6 ††) P58 P55 P79 271 ‡ I/O (CS1 ††) P79 P80 P115 14
I/O - P56 P80 274 ‡ I/O P80 P81 P116 17
GND - - P81 - GND - - P118 -
I/O - - P82 277 ‡ I/O - - P119 20
I/O - - P83 280 ‡ I/O - - P120 23
I/O (D5 ††) P59 P57 P84 283 ‡ I/O P81 P82 P121 26
I/O P60 P58 P85 286 ‡ I/O P82 P83 P122 29
I/O - P59 P86 289 ‡ I/O - P84 P123 32
I/O - P60 P87 292 ‡ I/O - P85 P124 35
I/O (D4 ††) P61 P61 P88 295 ‡ I/O P83 P86 P125 38
I/O P62 P62 P89 298 ‡ I/O P84 P87 P126 41
VCC P63 P63 P90 - GND P1 P88 P127 -
GND P64 P64 P91 - 11/20/98
I/O (D3 ††) P65 P65 P92 301 ‡ † = 5V Spartan only
I/O P66 P66 P93 304 ‡
I/O - P67 P94 307 ‡ †† = 3V SpartanXL only
I/O - - P95 310 ‡
‡ The “PWRDWN” on the XCS10XL is not part of the Boundary
I/O (D2 ††) P67 P68 P96 313 ‡
I/O P68 P69 P97 316 ‡ Scan chain. For the XCS10XL, subtract 1 from all Boundary Scan
I/O - - P98 319 ‡ numbers from GCK3 on (175 and higher).
I/O - - P99 322 ‡
GND - - P100 - Additional XCS10/XL Package Pins
I/O (D1 ††) P69 P70 P101 325 ‡
I/O P70 P71 P102 328 ‡ TQ144
I/O - - P103 331 ‡ Not Connected Pins
I/O - - P104 334 ‡ P117 - - - - -
5/5/97
I/O (D0 ††, DIN) P71 P72 P105 337 ‡

R

VQ100 TQ144 PQ208 VQ100 TQ144 PQ208
VCC P89 P128 P183 - I/O P19 P29 P45 227
I/O P90 P129 P184 62 I/O - P30 P46 230
I/O P91 P130 P185 65 I/O - P31 P47 233
I/O P92 P131 P186 68 I/O P20 P32 P48 236
I/O P93 P132 P187 71 I/O, SGCK2 †, GCK2 †† P21 P33 P49 239
I/O - - P188 74 Don’t Connect †, M1 †† P22 P34 P50 242
I/O - - P189 77 GND P23 P35 P51 -
I/O P94 P133 P190 80 MODE †, M0 †† P24 P36 P52 245
I/O P95 P134 P191 83 VCC P25 P37 P53 -
VCC †† - - P192 - Don’t Connect †, P26 P38 P54 246 †
I/O - P135 P193 86 PWRDWN ††
I/O - P136 P194 89 I/O, PGCK2 †, GCK3 †† P27 P39 P55 247 ‡
GND - P137 P195 - I/O (HDC) P28 P40 P56 250 ‡
I/O - - P196 92 I/O - P41 P57 253 ‡
I/O - - P197 95 I/O - P42 P58 256 ‡
I/O - - P198 98 I/O P29 P43 P59 259 ‡
I/O - - P199 101 I/O (LDC) P30 P44 P60 262 ‡
I/O P96 P138 P200 104 I/O - - P61 265 ‡
I/O P97 P139 P201 107 I/O - - P62 268 ‡ 4
I/O - P140 P204 110 I/O - - P63 271 ‡
I/O - P141 P205 113 I/O - - P64 274 ‡
I/O P98 P142 P206 116 GND - P45 P66 -
I/O, SGCK1 †, GCK8 †† P99 P143 P207 119 I/O - P46 P67 277 ‡
VCC P100 P144 P208 - I/O - P47 P68 280 ‡
GND P1 P1 P1 - I/O P31 P48 P69 283 ‡
I/O, PGCK1 †, GCK1 †† P2 P2 P2 122 I/O P32 P49 P70 286 ‡
I/O P3 P3 P3 125 VCC †† - - P71 -
I/O - P4 P4 128 I/O - - P72 289 ‡
I/O - P5 P5 131 I/O - - P73 292 ‡
I/O, TDI P4 P6 P6 134 I/O P33 P50 P74 295 ‡
I/O, TCK P5 P7 P7 137 I/O P34 P51 P75 298 ‡
I/O - - P8 140 I/O P35 P52 P76 301 ‡
I/O - - P9 143 I/O (INIT) P36 P53 P77 304 ‡
I/O - - P10 146 VCC P37 P54 P78 -
I/O - - P11 149 GND P38 P55 P79 -
GND - P8 P13 - I/O P39 P56 P80 307 ‡
I/O - P9 P14 152 I/O P40 P57 P81 310 ‡
I/O - P10 P15 155 I/O P41 P58 P82 313 ‡
I/O, TMS P6 P11 P16 158 I/O P42 P59 P83 316 ‡
I/O P7 P12 P17 161 I/O - - P84 319 ‡
VCC †† - - P18 - I/O - - P85 322 ‡
I/O - - P19 164 VCC †† - - P86 -
I/O - - P20 167 I/O P43 P60 P87 325 ‡
I/O - P13 P21 170 I/O P44 P61 P88 328 ‡
I/O P8 P14 P22 173 I/O - P62 P89 331 ‡
I/O P9 P15 P23 176 I/O - P63 P90 334 ‡
I/O P10 P16 P24 179 GND - P64 P91 -
GND P11 P17 P25 - I/O - - P93 337 ‡
VCC P12 P18 P26 - I/O - - P94 340 ‡
I/O P13 P19 P27 182 I/O - - P95 343 ‡
I/O P14 P20 P28 185 I/O - - P96 346 ‡
I/O P15 P21 P29 188 I/O P45 P65 P97 349 ‡
I/O - P22 P30 191 I/O P46 P66 P98 352 ‡
I/O - - P31 194 I/O - P67 P99 355 ‡
I/O - - P32 197 I/O - P68 P100 358 ‡
VCC †† - - P33 - I/O P47 P69 P101 361 ‡
I/O P16 P23 P34 200 I/O, SGCK3 †, GCK4 †† P48 P70 P102 364 ‡
I/O P17 P24 P35 203 GND P49 P71 P103 -
I/O - P25 P36 206 DONE P50 P72 P104 -
I/O - P26 P37 209 VCC P51 P73 P105 -
GND - P27 P38 - PROGRAM P52 P74 P106 -
I/O - - P40 212 I/O (D7 ††) P53 P75 P107 367 ‡
I/O - - P41 215 I/O, PGCK3 †, GCK5 †† P54 P76 P108 370 ‡
I/O - - P42 218 I/O - P77 P109 373 ‡
I/O - - P43 221 I/O - P78 P110 376 ‡
I/O P18 P28 P44 224 I/O (D6 ††) P55 P79 P112 379 ‡
I/O P56 P80 P113 382 ‡

R

VQ100 TQ144 PQ208 VQ100 TQ144 PQ208
I/O - - P114 385 ‡ VCC P75 P108 P156 -
I/O - - P115 388 ‡ O, TDO P76 P109 P157 0
I/O - - P116 391 ‡ GND P77 P110 P158 -
I/O - - P117 394 ‡ I/O P78 P111 P159 2
GND - P81 P118 - I/O, PGCK4 †, GCK7 †† P79 P112 P160 5
I/O - P82 P119 397 ‡ I/O - P113 P161 8
I/O - P83 P120 400 ‡ I/O - P114 P162 11
VCC †† - - P121 - I/O (CS1 ††) P80 P115 P163 14
I/O (D5 ††) P57 P84 P122 403 ‡ I/O P81 P116 P164 17
I/O P58 P85 P123 406 ‡ I/O - P117 P166 20
I/O - - P124 409 ‡ I/O - - P167 23
I/O - - P125 412 ‡ I/O - - P168 26
I/O P59 P86 P126 415 ‡ I/O - - P169 29
I/O P60 P87 P127 418 ‡ GND - P118 P170 -
I/O (D4 ††) P61 P88 P128 421 ‡ I/O - P119 P171 32
I/O P62 P89 P129 424 ‡ I/O - P120 P172 35
VCC P63 P90 P130 - VCC †† - - P173 -
GND P64 P91 P131 - I/O P82 P121 P174 38
I/O (D3 ††) P65 P92 P132 427 ‡ I/O P83 P122 P175 41
I/O P66 P93 P133 430 ‡ I/O - - P176 44
I/O P67 P94 P134 433 ‡ I/O - - P177 47
I/O - P95 P135 436 ‡ I/O P84 P123 P178 50
I/O - - P136 439 ‡ I/O P85 P124 P179 53
I/O - - P137 442 ‡ I/O P86 P125 P180 56
I/O (D2 ††) P68 P96 P138 445 ‡ I/O P87 P126 P181 59
I/O P69 P97 P139 448 ‡ GND P88 P127 P182 -
VCC †† - - P140 - 11/20/98
I/O - P98 P141 451 ‡

I/O - P99 P142 454 ‡
Additional XCS20/XL Package Pins
GND - P100 P143 - PQ208
I/O - - P145 457 ‡ Not Connected Pins
I/O - - P146 460 ‡ P12 P18 † P33 † P39 P65 P71 †
I/O - - P147 463 ‡ P86 † P92 P111 P121 † P140 † P144
I/O - - P148 466 ‡ P165 P173 † P192 † P202 P203 -
I/O (D1 ††) P70 P101 P149 469 ‡ 9/16/98
I/O P71 P102 P150 472 ‡ † = 5V Spartan only

I/O - P103 P151 475 ‡
†† = 3V SpartanXL only
I/O - P104 P152 478 ‡
I/O (D0 ††, DIN) P72 P105 P153 481 ‡ ‡ The “PWRDWN” on the XCS20XL is not part of the Boundary
I/O, SGCK4 †, GCK6 †† P73 P106 P154 484 ‡ Scan chain. For the XCS20XL, subtract 1 from all Boundary Scan
(DOUT) numbers from GCK3 on (247 and higher).
CCLK P74 P107 P155 -

VQ100 TQ144 PQ208 PQ240 BG256 VQ100 TQ144 PQ208 PQ240 BG256
VCC P89 P128 P183 P212 VCC* - I/O - - P202 P234 D5 128
I/O P90 P129 P184 P213 C10 74 I/O - - P203 P235 C4 131
I/O P91 P130 P185 P214 D10 77 I/O - P140 P204 P236 B3 134
I/O P92 P131 P186 P215 A9 80 I/O - P141 P205 P237 B2 137
I/O P93 P132 P187 P216 B9 83 I/O P98 P142 P206 P238 A2 140
I/O - - P188 P217 C9 86 I/O, SGCK1 †, P99 P143 P207 P239 C3 143
I/O - - P189 P218 D9 89 GCK8 ††
I/O P94 P133 P190 P220 A8 92 VCC P100 P144 P208 P240 VCC* -
I/O P95 P134 P191 P221 B8 95 GND P1 P1 P1 P1 GND* -
VCC - - P192 P222 VCC* - I/O, PGCK1 †, P2 P2 P2 P2 B1 146
GCK1 ††
I/O - - - P223 A6 98
I/O P3 P3 P3 P3 C2 149
I/O - - - P224 C7 101
I/O - P4 P4 P4 D2 152
I/O - P135 P193 P225 B6 104
I/O - P5 P5 P5 D3 155
I/O - P136 P194 P226 A5 107
I/O, TDI P4 P6 P6 P6 E4 158
GND - P137 P195 P227 GND* -
I/O, TCK P5 P7 P7 P7 C1 161
I/O - - P196 P228 C6 110
I/O - - P8 P8 D1 164
I/O - - P197 P229 B5 113
I/O - - P9 P9 E3 167
I/O - - P198 P230 A4 116
I/O - - P10 P10 E2 170
I/O - - P199 P231 C5 119
I/O - - P11 P11 E1 173
I/O P96 P138 P200 P232 B4 122
I/O - - P12 P12 F3 176
I/O P97 P139 P201 P233 A3 125

R

I/O - - - P13 F2 179 I/O - - P72 P81 Y8 343 ‡
GND - P8 P13 P14 GND* - I/O - - P73 P82 U9 346 ‡
I/O - P9 P14 P15 G3 182 I/O - - - P84 Y9 349 ‡
I/O - P10 P15 P16 G2 185 I/O - - - P85 W10 352 ‡
I/O, TMS P6 P11 P16 P17 G1 188 I/O P33 P50 P74 P86 V10 355 ‡
I/O P7 P12 P17 P18 H3 191 I/O P34 P51 P75 P87 Y10 358 ‡
VCC - - P18 P19 VCC* - I/O P35 P52 P76 P88 Y11 361 ‡
I/O - - - P20 H2 194 I/O (INIT) P36 P53 P77 P89 W11 364 ‡
I/O - - - P21 H1 197 VCC P37 P54 P78 P90 VCC* -
I/O - - P19 P23 J2 200 GND P38 P55 P79 P91 GND* -
I/O - - P20 P24 J1 203 I/O P39 P56 P80 P92 V11 367 ‡
I/O - P13 P21 P25 K2 206 I/O P40 P57 P81 P93 U11 370 ‡
I/O P8 P14 P22 P26 K3 209 I/O P41 P58 P82 P94 Y12 373 ‡
I/O P9 P15 P23 P27 K1 212 I/O P42 P59 P83 P95 W12 376 ‡
I/O P10 P16 P24 P28 L1 215 I/O - - P84 P96 V12 379 ‡
GND P11 P17 P25 P29 GND* - I/O - - P85 P97 U12 382 ‡
VCC P12 P18 P26 P30 VCC* - I/O - - - P99 V13 385 ‡
I/O P13 P19 P27 P31 L2 218 I/O - - - P100 Y14 388 ‡
I/O P14 P20 P28 P32 L3 221 VCC - - P86 P101 VCC* -
I/O P15 P21 P29 P33 L4 224 I/O P43 P60 P87 P102 Y15 391 ‡
I/O - P22 P30 P34 M1 227 I/O P44 P61 P88 P103 V14 394 ‡
I/O - - P31 P35 M2 230 I/O - P62 P89 P104 W15 397 ‡ 4
I/O - - P32 P36 M3 233 I/O - P63 P90 P105 Y16 400 ‡
I/O - - - P38 N1 236 GND - P64 P91 P106 GND* -
I/O - - - P39 N2 239 I/O - - - P107 V15 403 ‡
VCC - - P33 P40 VCC* - I/O - - P92 P108 W16 406 ‡
I/O P16 P23 P34 P41 P1 242 I/O - - P93 P109 Y17 409 ‡
I/O P17 P24 P35 P42 P2 245 I/O - - P94 P110 V16 412 ‡
I/O - P25 P36 P43 R1 248 I/O - - P95 P111 W17 415 ‡
I/O - P26 P37 P44 P3 251 I/O - - P96 P112 Y18 418 ‡
GND - P27 P38 P45 GND* - I/O P45 P65 P97 P113 U16 421 ‡
I/O - - - P46 T1 254 I/O P46 P66 P98 P114 V17 424 ‡
I/O - - P39 P47 R3 257 I/O - P67 P99 P115 W18 427 ‡
I/O - - P40 P48 T2 260 I/O - P68 P100 P116 Y19 430 ‡
I/O - - P41 P49 U1 263 I/O P47 P69 P101 P117 V18 433 ‡
I/O - - P42 P50 T3 266 I/O, SGCK3 †, P48 P70 P102 P118 W19 436 ‡
I/O - - P43 P51 U2 269 GCK4 ††
I/O P18 P28 P44 P52 V1 272 GND P49 P71 P103 P119 GND* -
I/O P19 P29 P45 P53 T4 275 DONE P50 P72 P104 P120 Y20 -
I/O - P30 P46 P54 U3 278 VCC P51 P73 P105 P121 VCC* -
I/O - P31 P47 P55 V2 281 PROGRAM P52 P74 P106 P122 V19 -
I/O P20 P32 P48 P56 W1 284 I/O (D7 ††) P53 P75 P107 P123 U19 439 ‡
I/O, SGCK2 †, P21 P33 P49 P57 V3 287 I/O, PGCK3 †, P54 P76 P108 P124 U18 442 ‡
GCK2 †† GCK5 ††
Don’t Connect †, P22 P34 P50 P58 W2 290 I/O - P77 P109 P125 T17 445 ‡
M1 †† I/O - P78 P110 P126 V20 448 ‡
GND P23 P35 P51 P59 GND* - I/O - - - P127 U20 451 ‡
MODE †, M0 †† P24 P36 P52 P60 Y1 293 I/O - - P111 P128 T18 454 ‡
VCC P25 P37 P53 P61 VCC* - I/O (D6 ††) P55 P79 P112 P129 T19 457 ‡
Don’t Connect †, P26 P38 P54 P62 W3 294 † I/O P56 P80 P113 P130 T20 460 ‡
PWRDWN †† I/O - - P114 P131 R18 463 ‡
I/O, PGCK2 †, P27 P39 P55 P63 Y2 295 ‡ I/O - - P115 P132 R19 466 ‡
GCK3 †† I/O - - P116 P133 R20 469 ‡
I/O (HDC) P28 P40 P56 P64 W4 298 ‡ I/O - - P117 P134 P18 472 ‡
I/O - P41 P57 P65 V4 301 ‡ GND - P81 P118 P135 GND* -
I/O - P42 P58 P66 U5 304 ‡ I/O - - - P136 P20 475 ‡
I/O P29 P43 P59 P67 Y3 307 ‡ I/O - - - P137 N18 478 ‡
I/O (LDC) P30 P44 P60 P68 Y4 310 ‡ I/O - P82 P119 P138 N19 481 ‡
I/O - - P61 P69 V5 313 ‡ I/O - P83 P120 P139 N20 484 ‡
I/O - - P62 P70 W5 316 ‡ VCC - - P121 P140 VCC* -
I/O - - P63 P71 Y5 319 ‡ I/O (D5 ††) P57 P84 P122 P141 M17 487 ‡
I/O - - P64 P72 V6 322 ‡ I/O P58 P85 P123 P142 M18 490 ‡
I/O - - P65 P73 W6 325 ‡ I/O - - P124 P144 M20 493 ‡
I/O - - - P74 Y6 328 ‡ I/O - - P125 P145 L19 496 ‡
GND - P45 P66 P75 GND* - I/O P59 P86 P126 P146 L18 499 ‡
I/O - P46 P67 P76 W7 331 ‡ I/O P60 P87 P127 P147 L20 502 ‡
I/O - P47 P68 P77 Y7 334 ‡ I/O (D4 ††) P61 P88 P128 P148 K20 505 ‡
I/O P31 P48 P69 P78 V8 337 ‡ I/O P62 P89 P129 P149 K19 508 ‡
I/O P32 P49 P70 P79 W8 340 ‡ VCC P63 P90 P130 P150 VCC* -
VCC - - P71 P80 VCC* - GND P64 P91 P131 P151 GND* -

R

I/O (D3 ††) P65 P92 P132 P152 K18 511 ‡ I/O P85 P124 P179 P208 A11 65
I/O P66 P93 P133 P153 K17 514 ‡ I/O P86 P125 P180 P209 A10 68
I/O P67 P94 P134 P154 J20 517 ‡ I/O P87 P126 P181 P210 B10 71
I/O - P95 P135 P155 J19 520 ‡ GND P88 P127 P182 P211 GND* -
I/O - - P136 P156 J18 523 ‡ 11/20/98
I/O - - P137 P157 J17 526 ‡ * Pads labelled GND* or VCC* are internally bonded to Ground or
I/O (D2 ††) P68 P96 P138 P159 H19 529 ‡ VCC planes within the package.
I/O P69 P97 P139 P160 H18 532 ‡
† = 5V Spartan only
VCC - - P140 P161 VCC* -
I/O - P98 P141 P162 G19 535 ‡ †† = 3V SpartanXL only
I/O - P99 P142 P163 F20 538 ‡ ‡ The “PWRDWN” on the XCS30XL is not part of the Boundary
I/O - - - P164 G18 541 ‡ Scan chain. For the XCS30XL, subtract 1 from all Boundary Scan
I/O - - - P165 F19 544 ‡
numbers from GCK3 on (295 and higher).
GND - P100 P143 P166 GND* -
I/O - - - P167 F18 547 ‡
I/O - - P144 P168 E19 550 ‡
I/O - - P145 P169 D20 553 ‡ PQ240
I/O - - P146 P170 E18 556 ‡ GND Pins
I/O - - P147 P171 D19 559 ‡ P22 P37 P83 P98 P143 P158
I/O - - P148 P172 C20 562 ‡ P204 P219 - - - -
I/O (D1 ††) P70 P101 P149 P173 E17 565 ‡ Not Connected Pins
I/O P71 P102 P150 P174 D18 568 ‡ P195 - - - - -
I/O - P103 P151 P175 C19 571 ‡ 2/12/98
I/O - P104 P152 P176 B20 574 ‡
I/O (D0 ††, DIN) P72 P105 P153 P177 C18 577 ‡
I/O, SGCK4 †, P73 P106 P154 P178 B19 580 ‡ BG256
GCK6 †† (DOUT) VCC Pins
CCLK P74 P107 P155 P179 A20 - C14 D6 D7 D11 D14 D15
VCC P75 P108 P156 P180 VCC* - E20 F1 F4 F17 G4 G17
O, TDO P76 P109 P157 P181 A19 0 K4 L17 P4 P17 P19 R2
GND P77 P110 P158 P182 GND* - R4 R17 U6 U7 U10 U14
I/O P78 P111 P159 P183 B18 2 U15 V7 W20 - - -
I/O, PGCK4 †, P79 P112 P160 P184 B17 5 GND Pins
GCK7 ††
A1 B7 D4 D8 D13 D17
I/O - P113 P161 P185 C17 8
G20 H4 H17 N3 N4 N17
I/O - P114 P162 P186 D16 11
U4 U8 U13 U17 W14 -
I/O (CS1) †† P80 P115 P163 P187 A18 14
Not Connected Pins
I/O P81 P116 P164 P188 A17 17
A7 A13 C8 D12 H20 J3
I/O - - P165 P189 C16 20
J4 M4 M19 V9 W9 W13
I/O - - - P190 B16 23
Y13 - - - - -
I/O - P117 P166 P191 A16 26 6/4/97
I/O - - P167 P192 C15 29
I/O - - P168 P193 B15 32
I/O - - P169 P194 A15 35
GND - P118 P170 P196 GND* -
I/O - P119 P171 P197 B14 38
I/O - P120 P172 P198 A14 41
I/O - - - P199 C13 44
I/O - - - P200 B13 47
VCC - - P173 P201 VCC* -
I/O P82 P121 P174 P202 C12 50
I/O P83 P122 P175 P203 B12 53
I/O - - P176 P205 A12 56
I/O - - P177 P206 B11 59
I/O P84 P123 P178 P207 C11 62

R

PQ208 PQ240 BG256 PQ208 PQ240 BG256
VCC P183 P212 VCC* - I/O P30 P34 M1 263
I/O P184 P213 C10 86 I/O P31 P35 M2 266
I/O P185 P214 D10 89 I/O P32 P36 M3 269
I/O P186 P215 A9 92 I/O - - M4 272
I/O P187 P216 B9 95 I/O - P38 N1 278
I/O P188 P217 C9 98 I/O - P39 N2 281
I/O P189 P218 D9 101 VCC P33 P40 VCC* -
I/O P190 P220 A8 104 I/O P34 P41 P1 284
I/O P191 P221 B8 107 I/O P35 P42 P2 287
I/O - - C8 110 I/O P36 P43 R1 290
I/O - - A7 113 I/O P37 P44 P3 293
VCC P192 P222 VCC* - GND P38 P45 GND* -
I/O - P223 A6 116 I/O - P46 T1 296
I/O - P224 C7 119 I/O P39 P47 R3 299
I/O P193 P225 B6 122 I/O P40 P48 T2 302
I/O P194 P226 A5 125 I/O P41 P49 U1 305
GND P195 P227 GND* - I/O P42 P50 T3 308
I/O P196 P228 C6 128 I/O P43 P51 U2 311
I/O P197 P229 B5 131 I/O P44 P52 V1 320 4
I/O P198 P230 A4 134 I/O P45 P53 T4 323
I/O P199 P231 C5 137 I/O P46 P54 U3 326
I/O P200 P232 B4 140 I/O P47 P55 V2 329
I/O P201 P233 A3 143 I/O P48 P56 W1 332
I/O P202 P234 D5 152 I/O, SGCK2 †, P49 P57 V3 335
I/O P203 P235 C4 155 GCK2 ††
I/O P204 P236 B3 158 Don’t Connect †, P50 P58 W2 338
M1 ††
I/O P205 P237 B2 161
GND P51 P59 GND* -
I/O P206 P238 A2 164
MODE †, M0 †† P52 P60 Y1 341
I/O, SGCK1 †, P207 P239 C3 167
GCK8 †† VCC P53 P61 VCC* -
VCC P208 P240 VCC* - Don‘t Connect †, P54 P62 W3 342†
PWRDWN ††
GND P1 P1 GND* -
I/O, PGCK2 †, P55 P63 Y2 343 ‡
I/O, PGCK1 †, P2 P2 B1 170
GCK3 ††
GCK1 ††
I/O (HDC) P56 P64 W4 346 ‡
I/O P3 P3 C2 173
I/O P57 P65 V4 349 ‡
I/O P4 P4 D2 176
I/O P58 P66 U5 352 ‡
I/O P5 P5 D3 179
I/O P59 P67 Y3 355 ‡
I/O, TDI P6 P6 E4 182
I/O (LDC) P60 P68 Y4 358 ‡
I/O, TCK P7 P7 C1 185
I/O P61 P69 V5 367 ‡
I/O P8 P8 D1 194
I/O P62 P70 W5 370 ‡
I/O P9 P9 E3 197
I/O P63 P71 Y5 373 ‡
I/O P10 P10 E2 200
I/O P64 P72 V6 376 ‡
I/O P11 P11 E1 203
I/O P65 P73 W6 379 ‡
I/O P12 P12 F3 206
I/O - P74 Y6 382 ‡
I/O - P13 F2 209
GND P66 P75 GND* -
GND P13 P14 GND* -
I/O P67 P76 W7 385 ‡
I/O P14 P15 G3 212
I/O P68 P77 Y7 388 ‡
I/O P15 P16 G2 215
I/O P69 P78 V8 391 ‡
I/O, TMS P16 P17 G1 218
I/O P70 P79 W8 394 ‡
I/O P17 P18 H3 221
VCC P71 P80 VCC* -
VCC P18 P19 VCC* -
I/O P72 P81 Y8 397 ‡
I/O - P20 H2 224
I/O P73 P82 U9 400 ‡
I/O - P21 H1 227
I/O - - V9 403 ‡
I/O - - J4 230
I/O - - W9 406 ‡
I/O - - J3 233
I/O - P84 Y9 409 ‡
I/O P19 P23 J2 236
I/O - P85 W10 412 ‡
I/O P20 P24 J1 239
I/O P74 P86 V10 415 ‡
I/O P21 P25 K2 242
I/O P75 P87 Y10 418 ‡
I/O P22 P26 K3 245
I/O P76 P88 Y11 421 ‡
I/O P23 P27 K1 248
I/O (INIT) P77 P89 W11 424 ‡
I/O P24 P28 L1 251
VCC P78 P90 VCC* -
GND P25 P29 GND* -
GND P79 P91 GND* -
VCC P26 P30 VCC* -
I/O P80 P92 V11 427 ‡
I/O P27 P31 L2 254
I/O P81 P93 U11 430 ‡
I/O P28 P32 L3 257
I/O P82 P94 Y12 433 ‡
I/O P29 P33 L4 260
I/O P83 P95 W12 436 ‡

R

PQ208 PQ240 BG256 PQ208 PQ240 BG256
I/O P84 P96 V12 439 ‡ I/O (D2 ††) P138 P159 H19 619 ‡
I/O P85 P97 U12 442 ‡ I/O P139 P160 H18 622 ‡
I/O - - Y13 445 ‡ VCC P140 P161 VCC* -
I/O - - W13 448 ‡ I/O P141 P162 G19 625 ‡
I/O - P99 V13 451 ‡ I/O P142 P163 F20 628 ‡
I/O - P100 Y14 454 ‡ I/O - P164 G18 631 ‡
VCC P86 P101 VCC* - I/O - P165 F19 634 ‡
I/O P87 P102 Y15 457 ‡ GND P143 P166 GND* -
I/O P88 P103 V14 460 ‡ I/O - P167 F18 637 ‡
I/O P89 P104 W15 463 ‡ I/O P144 P168 E19 640 ‡
I/O P90 P105 Y16 466 ‡ I/O P145 P169 D20 643 ‡
GND P91 P106 GND* - I/O P146 P170 E18 646 ‡
I/O - P107 V15 469 ‡ I/O P147 P171 D19 649 ‡
I/O P92 P108 W16 472 ‡ I/O P148 P172 C20 652 ‡
I/O P93 P109 Y17 475 ‡ I/O (D1 ††) P149 P173 E17 655 ‡
I/O P94 P110 V16 478 ‡ I/O P150 P174 D18 658 ‡
I/O P95 P111 W17 481 ‡ I/O P151 P175 C19 667 ‡
I/O P96 P112 Y18 484 ‡ I/O P152 P176 B20 670 ‡
I/O P97 P113 U16 493 ‡ I/O (D0 ††, DIN) P153 P177 C18 673 ‡
I/O P98 P114 V17 496 ‡ I/O, SGCK4 †, P154 P178 B19 676 ‡
I/O P99 P115 W18 499 ‡ GCK6 †† (DOUT)
I/O P100 P116 Y19 502 ‡ CCLK P155 P179 A20 -
I/O P101 P117 V18 505 ‡ VCC P156 P180 VCC* -
I/O, SGCK3 †, P102 P118 W19 508 ‡ O, TDO P157 P181 A19 0
GCK4 †† GND P158 P182 GND* -
GND P103 P119 GND* - I/O P159 P183 B18 2
DONE P104 P120 Y20 - I/O, PGCK4 †, P160 P184 B17 5
VCC P105 P121 VCC* - GCK7 ††
PROGRAM P106 P122 V19 - I/O P161 P185 C17 8
I/O (D7 ††) P107 P123 U19 511 ‡ I/O P162 P186 D16 11
I/O, PGCK3 †, P108 P124 U18 514 ‡ I/O (CS1 ††) P163 P187 A18 14
GCK5 †† I/O P164 P188 A17 17
I/O P109 P125 T17 517 ‡ I/O P165 P189 C16 26
I/O P110 P126 V20 520 ‡ I/O - P190 B16 29
I/O - P127 U20 523 ‡ I/O P166 P191 A16 32
I/O P111 P128 T18 526 ‡ I/O P167 P192 C15 35
I/O (D6 ††) P112 P129 T19 535 ‡ I/O P168 P193 B15 38
I/O P113 P130 T20 538 ‡ I/O P169 P194 A15 41
I/O P114 P131 R18 541 ‡ GND P170 P196 GND* -
I/O P115 P132 R19 544 ‡ I/O P171 P197 B14 44
I/O P116 P133 R20 547 ‡ I/O P172 P198 A14 47
I/O P117 P134 P18 550 ‡ I/O - P199 C13 50
GND P118 P135 GND* - I/O - P200 B13 53
I/O - P136 P20 553 ‡ VCC P173 P201 VCC* -
I/O - P137 N18 556 ‡ I/O - - A13 56
I/O P119 P138 N19 559 ‡ I/O - - D12 59
I/O P120 P139 N20 562 ‡ I/O P174 P202 C12 62
VCC P121 P140 VCC* - I/O P175 P203 B12 65
I/O (D5 ††) P122 P141 M17 565 ‡ I/O P176 P205 A12 68
I/O P123 P142 M18 568 ‡ I/O P177 P206 B11 71
I/O - - M19 574 ‡ I/O P178 P207 C11 74
I/O P124 P144 M20 577 ‡ I/O P179 P208 A11 77
I/O P125 P145 L19 580 ‡ I/O P180 P209 A10 80
I/O P126 P146 L18 583 ‡ I/O P181 P210 B10 83
I/O P127 P147 L20 586 ‡ GND P182 P211 GND* -
I/O (D4 ††) P128 P148 K20 589 ‡ 11/20/98
I/O P129 P149 K19 592 ‡ * Pads labelled GND* or VCC* are internally bonded to Ground or
VCC P130 P150 VCC* - VCC planes within the package.
GND P131 P151 GND* -
I/O (D3 ††) P132 P152 K18 595 ‡ † = 5V Spartan only
I/O P133 P153 K17 598 ‡
I/O P134 P154 J20 601 ‡
†† = 3V SpartanXL only
I/O P135 P155 J19 604 ‡ ‡ The “PWRDWN” on the XCS40XL is not part of the Boundary
I/O P136 P156 J18 607 ‡ Scan chain. For the XCS40XL, subtract 1 from all Boundary Scan
I/O P137 P157 J17 610 ‡ numbers from GCK3 on (343 and higher).
I/O - - H20 613 ‡

R

PQ240
GND Pins
P22 P37 P83 P98 P143 P158
P204 P219 - - - -
Not Connected Pins
P195 - - - - -
2/12/98
BG256
VCC Pins
C14 D6 D7 D11 D14 D15
E20 F1 F4 F17 G4 G17
K4 L17 P4 P17 P19 R2
R4 R17 U6 U7 U10 U14
U15 V7 W20 - - -
GND Pins
A1 B7 D4 D8 D13 D17
G20 H4 H17 N3 N4 N17
U4 U8 U13 U17 W14 -
6/17/97

Product Availability
Table 19 shows the packages and speed grades for Spartan Series devices. Table 20 shows the number of user I/Os
available for each device/package combination.
Table 19: Component Availability Chart for Spartan Series FPGAs
PINS 84 100 144 208 240 256

TYPE Plast. PLCC Plast. VQFP Plast. TQFP Plast. PQFP Plast. PQFP Plast. BGA
Device CODE PC84 VQ100 TQ144 PQ208 PQ240 BG256
-3 C C, I
XCS05
-4 C C
-3 C C, I C
XCS10
-4 C C C
-3 C C, I C, I
XCS20
-4 C C C
-3 C C, I C, I C C
XCS30
-4 C C C C C
-3 C, I C C
XCS40
-4 C C C
-4 C C, I
XCS05XL
-5 (C) (C)
-4 C C, I C
XCS10XL
-5 (C) (C) (C)
-4 C C, I C, I
XCS20XL
-5 (C) (C) (C)
-4 C C, I C, I C C
XCS30XL
-5 (C) (C) (C) (C) (C)
-4 C, I C C
XCS40XL
-5 (C) (C) (C)
11/20/98
C = Commercial TJ = 0° to +85°C ( ) Parentheses indicate future product plans

I = Industrial TJ = -40°C to +100°C
Table 20: User I/O Chart for Spartan Series FPGAs
Max Package Type

Device I/O PC84 VQ100 TQ144 PQ208 PQ240 BG256
XCS05 80 61 77
XCS10 112 61 77 112
XCS20 160 77 113 160
XCS30 192 77 113 169 192 192
XCS40 224 169 192 205
XCS05XL 80 61 77
XCS10XL 112 61 77 112
XCS20XL 160 77 113 160
XCS30XL 192 77 113 169 192 192
XCS40XL 224 169 192 205
11/20/98

R
Example: XCS20XL-4 PQ208C

C = Commercial (TJ = 0 to +85oC)
Speed Grade
I = Industrial (TJ = -40oC to +100oC)
-3
-4
-5 Number of Pins
Package Type
BG = Ball Grid Array VQ = Very Thin Quad Flat Pack
PC = Plastic Lead Chip Carrier TQ = Thin Quad Flat Pack
PQ = Plastic Quad Flat Pack
Table 21: Revisions
Version Description
4/2/98 Rev. 1.0 (Preliminary) Added timing specifications for 5V devices, updated PQ208 pinout tables, im-
proved description.
5/31/98 Rev. 1.1 (Preliminary) Updated all timing specifications.
9/28/98 Rev. 1.2 (Preliminary) Added SpartanXL architecture description, changed speed grades to -4/-5
11/20/98 Rev. 1.3 (Preliminary) Added SpartanXL specs and Power Down
1/6/99 Rev. 1.4 (Preliminary) All SpartanXL -4 specs designated Preliminary with no changes

XC9500 Products
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

XC9500 Products
Table of Contents
XC9500 Products
XC9500/XL Series Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
FastFLASH™ XC9500XL High-Performance CPLD Family . . . . . . . . . . . . . . . . . . . . . . . 5-5
XC9500 In-System Programmable CPLD Family. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-59

0
R
XC9500/XL Series Table of Contents
0 5*
XC9500 Products
FastFLASH™ XC9500XL High-Performance CPLD Family

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
Family Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
Architecture Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5
Function Block. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-7
Macrocell. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
Product Term Allocator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-10
FastCONNECT II Switch Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-13
I/O Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-14
5V Tolerant I/Os . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-15
Pin-Locking Capability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-15
In-System Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-16 5
External Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-16
Reliability and Endurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-16
IEEE 1149.1 Boundary-Scan (JTAG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-17
Design Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-17
Low Power Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-17
Timing Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-18
Power-Up Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-19
Development System Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-19
FastFLASH Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-19
XC9536XL High Performance CPLD

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-21
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-21
Power Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-21
Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-23
Recommended Operation Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-23
Quality and Reliability Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-23
DC Characteristic Over Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-23
AC Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-24
Internal Timing Parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-25
XC9536XL I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-26
XC9536XL Global, JTAG and Power Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-26
Component Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-27

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-29
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-29
5-1
R

XC9572XL I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-34
XC9572XL Global, JTAG and Power Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-35

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-37
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-37
DC Characteristics Over Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-39
XC95144XL I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-42
XC95144XL Global, JTAG and Power Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-44
Revision History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-45

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-47
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-47
XC95288XL I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-52
XC95288XL Global, JTAG and Power Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-56
XC9500 In-System Programmable CPLD Family

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-59
Family Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-59
Architecture Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-59
Function Block. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-61
Macrocell. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-62
Product Term Allocator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-64
FastCONNECT Switch Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-67
I/O Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-68
Pin-Locking Capability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-69
In-System Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-70
External Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-70
Endurance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-70
IEEE 1149.1 Boundary-Scan (JTAG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-70
Design Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-70
Low Power Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-71
Timing Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-71
Power-Up Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-72
Development System Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-73
FastFLASH Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-73
5-2
R
XC9536 In-System Programmable CPLD

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-75
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-75
Power Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-75
Recommended Operating Conditions1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-77
Endurance Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-77
XC9536 I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-80
XC9536 Global, JTAG and Power Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-80
Revision Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-81

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-83
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-83
Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-85 5
Recommended Operation Conditions1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-85
XC9572 I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-88
XC9572 Global, JTAG and Power Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-89

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-91
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-91
XC95108 I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-96
XC95108 Global, JTAG and Power Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-97

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-99
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-99
5-3
R

XC95144 I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-104

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-109
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-109
XC95288 I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-114
5-4
0
FastFLASH™ XC9500XL
R
High-Performance CPLD
Family
October 2, 1998 (Version 1.1) 0 5* Preliminary Product Specification
Features family is designed to work closely with the Xilinx Spar-

tan-XL and XC4000XL FPGA families, allowing system
• Optimized for high-performance 3.3V systems designers to partition logic optimally between fast interface
- 4 ns pin-to-pin logic delays, with internal system circuitry and high-density general purpose logic. As shown
frequency up to 200 MHz in Table 1, logic density of the XC9500XL devices ranges
- Small footprint packages including VQFPs, TQFPs from 800 to 6400 usable gates with 36 to 288 registers,
and CSPs (Chip Scale Package) respectively. Multiple package options and associated I/O
- Lower power operation capacity are shown in Table 2. The XC9500XL family mem-
- 5V tolerant I/O pins accept 5V, 3.3V, and 2.5V bers are fully pin-compatible, allowing easy design migra-
signals tion across multiple density options in a given package
- 3.3V or 2.5V output capability footprint.
- Advanced 0.35 micron feature size CMOS
FastFLASHTM technology The XC9500XL architectural features address the require-
• Advanced system features ments of in-system programmability. Enhanced pin-locking
- In-system programmable capability avoids costly board rework. In-system program- 5
- Superior pin-locking and routability with ming throughout the full commercial operating range and a
FastCONNECT IITM switch matrix high programming endurance rating provide worry-free
- Extra wide 54-input Function Blocks reconfigurations of system field upgrades. Extended data
- Up to 90 product-terms per macrocell with individual retention supports longer and more reliable system operat-
product-term allocation ing life.
- Local clock inversion with 3 global and one Advanced system features include output slew rate control
product-term clocks and user-programmable ground pins to help reduce system
- Individual output enable per output pin with local noise. Each user pin is compatible with 5V, 3.3V, and 2.5V
inversion inputs, and the outputs may be configured for 3.3V or 2.5V
- Input hysteresis on all user and boundary-scan pin operation. The XC9500XL device exhibits symmetric full
inputs 3.3V output voltage swing to allow balanced rise and fall
- Bus-hold circuitry on all user pin inputs times.
- Full IEEE Standard 1149.1 boundary-scan (JTAG)
support on all devices Architecture Description
• Four pin-compatible device densities
Each XC9500XL device is a subsystem consisting of multi-
- 36 to 288 macrocells, with 800 to 6400 usable gates
ple Function Blocks (FBs) and I/O Blocks (IOBs) fully inter-
• Fast concurrent programming
connected by the FastCONNECT II switch matrix. The IOB
• Slew rate control on individual outputs
provides buffering for device inputs and outputs. Each FB
• Enhanced data security features
provides programmable logic capability with extra wide 54
• Excellent quality and reliability
inputs and 18 outputs. The FastCONNECT II switch matrix
- 10,000 program/erase cycles endurance rating
connects all FB outputs and input signals to the FB inputs.
- 20 year data retention
For each FB, up to 18 outputs (depending on package
• Pin-compatible with 5V core XC9500 family in common
pin-count) and associated output enable signals drive
package footprints
directly to the IOBs. See Figure 1.
Family Overview
The FastFLASH XC9500XL family is a 3.3V CPLD family
targeted for high-performance, low-voltage applications in
leading-edge communications and computing systems,
where high device reliability and low power dissipation is
important. Each XC9500XL device supports in-system pro-
gramming (ISP) and the full IEEE 1149.1 (JTAG) bound-
ary-scan, allowing superior debug and design iteration
capability for small form-factor packages. The XC9500XL
October 2, 1998 (Version 1.1) 5-5

R
3
JTAG
JTAG Port In-System Programming Controller
Controller
54
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O
FastCONNECT II Switch Matrix

I/O 54
Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
54
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 54
1 Function
I/O/GSR 18 Block N
2 or 4 Macrocells
I/O/GTS 1 to 18
X5877_01
Figure 1: XC9500XL Architecture

Note: Function block outputs (indicated by the bold lines) drive the I/O blocks directly.
Table 1: XC9500XL Device Family

Macrocells 36 72 144 288
Usable Gates 800 1,600 3,200 6,400
Registers 36 72 144 288
tPD (ns) 4 5 5 6
tSU (ns) 3 3.7 3.7 4.1
tCO (ns) 3 3.5 3.5 4.3
fSYSTEM (MHz) 200 178 178 151
5-6 October 2, 1998 (Version 1.1)

R
Table 2: XC9500XL Packages and User I/O Pins (not including 4 dedicated JTAG pins)

44-Pin PLCC 34 34
64-Pin VQFP 36 52
100-Pin TQFP 72 81
144-Pin TQFP 117 117
208-Pin PQFP 168
48-Pin CSP 36 38
144-Pin CSP 117
352-Pin BGA 192
Function Block
Each Function Block, as shown in Figure 2 is comprised of Logic within the FB is implemented using a sum-of-prod-
18 independent macrocells, each capable of implementing ucts representation. Fifty-four inputs provide 108 true and
a combinatorial or registered function. The FB also complement signals into the programmable AND-array to
receives global clock, output enable, and set/reset signals. form 90 product terms. Any number of these product terms,
The FB generates 18 outputs that drive the FastCONNECT up to the 90 available, can be allocated to each macrocell
switch matrix. These 18 outputs and their corresponding by the product term allocator.
output enable signals also drive the IOB.
5
Macrocell 1
Programmable Product
AND-Array Term
Allocators 18 To FastCONNECT II
From Switch Matrix
54
FastCONNECT II
Switch Matrix 18 OUT
To I/O Blocks
18 PTOE
Macrocell 18
1 3
Global Global
Set/Reset Clocks X5878_01
Figure 2: XC9500XL Function Block

R
Macrocell
Each XC9500XL macrocell may be individually configured The product term allocator associated with each macrocell
for a combinatorial or registered function. The macrocell selects how the five direct terms are used.
and associated FB logic is shown in Figure 3. The macrocell register can be configured as a D-type or
Five direct product terms from the AND-array are available T-type flip-flop, or it may be bypassed for combinatorial
for use as primary data inputs (to the OR and XOR gates) operation. Each register supports both asynchronous set
to implement combinatorial functions, or as control inputs and reset operations. During power-up, all user registers
including clock, clock enable, set/reset, and output enable. are initialized to the user-defined preload state (default to 0
if unspecified).
54 Global Global
Set/Reset Clocks
Additional
Product
Terms
(from other
macrocells)
Product Term Set
0
To
S FastCONNECTII
Switch Matrix
D/T Q
EC
Product
Term
Product Term Clock Enable R
Allocator
Product Term Clock
Product Term Reset
OUT
To
Product Term OE PTOE I/O Blocks
Additional
Product
Terms
(from other
macrocells)
X5879_new
Figure 3: XC9500XL Macrocell Within Function Block

All global control signals are available to each individual term clock. Both true and complement polarities of the
macrocell, including clock, set/reset, and output enable sig- selected clock source can be used within each macrocell. A
nals. As shown in Figure 4, the macrocell register clock GSR input is also provided to allow user registers to be set
originates from either of three global clocks or a product to a user-defined state.

R
Macrocell
Product Term Set
S
D/T
Product Term Clock EC
R
Product Term Reset
I/O/GSR
Global Set/Reset
I/O/GCK1
Global Clock 1
5
I/O/GCK2
Global Clock 2
I/O/GCK3
Global Clock 3 99011202
Figure 4: Macrocell Clock and Set/Reset Capability

R
Product Term Allocator

The product term allocator controls how the five direct Note that the incremental delay affects only the product
product terms are assigned to each macrocell. For exam- terms in other macrocells. The timing of the direct product
ple, all five direct terms can drive the OR function as shown terms is not changed.
in Figure 5. Product Term
Allocator
Macrocell
Product Term
Logic
Product Term
Allocator
X5894
Figure 5: Macrocell Logic Using Direct Product Term
The product term allocator can re-assign other product

terms within the FB to increase the logic capacity of a mac-
Macrocell Logic
rocell beyond five direct terms. Any macrocell requiring With 15 P-Terms
additional product terms can access uncommitted product
terms in other macrocells within the FB. Up to 15 product
terms can be available to a single macrocell with only a
small incremental delay of tPTA, as shown in Figure 6.
Product Term
Allocator
X5895
Figure 6: Product Term Allocation With 15 Product

Terms

R
The product term allocator can re-assign product terms In this example, the incremental delay is only 2*tPTA. All 90
from any macrocell within the FB by combining partial sums product terms are available to any macrocell, with a maxi-
of products over several macrocells, as shown in Figure 7. mum incremental delay of 8*tPTA.
Product Term
Allocator
Macrocell Logic
With 2
Product Terms
Product Term
Allocator
Product Term
Allocator
Macrocell Logic
With 18
Product Terms
Product Term
Allocator
X5896
Figure 7: Product Term Allocation Over Several Macrocells

R
The internal logic of the product term allocator is shown in Figure 8.
From Upper To Upper

Macrocell Macrocell
Product Term
Allocator
Product Term Set
Global Set/Reset
S
D/T Q
EC
Global Clocks
R
Product Term Clock
Product Term Reset
Global Set/Reset
Product Term OE
From Lower To Lower

X5881_01
Macrocell Macrocell
Figure 8: Product Term Allocator Logic

R

The FastCONNECT II Switch Matrix connects signals to FastCONNECT II matrix. Any of these (up to a fan-in limit of
the FB inputs, as shown in Figure 9. All IOB outputs (corre- 54) may be selected to drive each FB with a uniform delay.
sponding to user pin inputs) and all FB outputs drive the
FastCONNECT II
Switch Matrix Function Block
I/O Block
(36) 18
D/T Q
I/O
Function Block
I/O Block
(36)
18
D/T Q
I/O
Wired-AND Capability
X5882_01
Figure 9: FastCONNECT II Switch Matrix

R
I/O Block
The I/O Block (IOB) interfaces between the internal logic and user programmable ground control. See Figure 10 for
and the device user I/O pins. Each IOB includes an input details.
buffer, output driver, output enable selection multiplexer,
To other
Macrocells
I/O Block
To FastCONNECT
Switch Matrix
Macrocell Bus-Hold
OUT I/O
(Inversion in
AND-array) 1 User-
Product Term OE PTOE Programmable
Ground
0
Slew Rate
Control
I/O/GTS1 Global OE 1
Available in XC95144XL
I/O/GTS3
Global OE 3 and XC95288XL
X5899_01
Figure 10: I/O Block and Output Enable Capability

The input buffer is compatible with 5V CMOS, 5V TTL, 3.3V (50 mV typical) to help reduce system noise for input sig-
CMOS, and 2.5V CMOS signals. The input buffer uses the nals with slow rise or fall edges.
internal 3.3V voltage supply (VCCINT) to ensure that the Each output driver is designed to provide fast switching with
input thresholds are constant and do not vary with the minimal power noise. All output drivers in the device may
VCCIO voltage. Each input buffer provides input hysteresis
be configured for driving either 3.3V CMOS levels (which
are compatible with 5V TTL levels as well) or 2.5V CMOS

R
levels by connecting the device output voltage supply regardless of the internal macrocell signal, so the internal
(VCCIO) to a 3.3V or 2.5V voltage supply. Figure 11 shows macrocell logic is unaffected by the programmable ground
how the XC9500XL device can be used in 3.3V only sys- pin capability.
tems and mixed voltage systems with any combination of Each IOB also provides for bus-hold circuitry that is active
5V, 3.3V and 2.5V power supplies. during valid user operation. The bus-hold feature elimi-
Each output driver can also be configured for slew-rate lim- nates the need to tie unused pins either high or low by hold-
ited operation. Output edge rates may be slowed down to ing the last known state of the input until the next input
reduce system noise (with an additional time delay of signal is present. The bus-hold circuit drives back the same
tSLEW) under user control. See Figure 12. state via a nominal resistance (RBH) of 50k ohms. See
Figure 13. Note the bus-hold output will drive no higher
The output enable may be generated from one of four
than VCCIO to prevent overdriving signals when interfacing
options: a product term signal from the macrocell, any of
to 2.5V components.
the global output enable signals (GTS), always “1,” or
always “0.” There are two global output enables for devices When the device is not in valid user operation, the bus-hold
with 72 or fewer macrocells, and four global output enables circuit defaults to an equivalent 50k ohm pull-up resistor in
for devices with 144 or more macrocells. Any selected out- order to provide a known repeatable device state. This
put enable signal may be inverted locally at each pin output occurs when the device is in the erased state, in program-
to provide maximal design flexibility. ming mode, in JTAG INTEST mode, or during initial
power-up. A pull-down resistor (1k ohm) may be externally
Each IOB provides user programmable ground pin capabil-
added to any pin to override the default RBH resistance to
ity. This allows device I/O pins to be configured as addi-
force a low state during power-up or any of these other
tional ground pins in order to force otherwise unused pins
modes. 5
to a low voltage state, as well as provide for additional
device grounding capability. This grounding of the pin is
achieved by internal logic that forces a logic low output
5 V CMOS 3.3 V 5 V CMOS 3.3 V 2.5 V

5V 5V
VCCINT VCCIO VCCINT VCCIO
0V 0V
5 V TTL or 3.3 V CMOS, 5 V TTL 5 V TTL or 2.5 V CMOS
3.6 V 3.3 V 3.6 V 2.5 V

XC9500XL XC9500XL
IN CPLD OUT IN CPLD OUT
0V 0V 0V 0V
3.3 V CMOS or 3.3 V CMOS or

3.3 V 3.3 V
GND GND
0V 0V
2.5V CMOS 2.5V CMOS

2.5 V (a) 2.5 V (b) X5901_01
0V 0V
Figure 11: XC9500XL Devices in (a) 3.3V only and (b) Mixed 5V/3.3V/2.5V Systems
5V Tolerant I/Os Pin-Locking Capability

The I/Os on each XC9500XL device are fully 5V tolerant The capability to lock the user defined pin assignments dur-
even though the core power supply is 3.3 volts. This allows ing design iteration depends on the ability of the architec-
5V CMOS signals to connect directly to the XC9500XL ture to adapt to unexpected changes. The XC9500XL
inputs without damage. In addition, the 3.3V VCCINT power devices incorporate architectural features that enhance the
supply can be applied before or after 5V signals are applied ability to accept design changes while maintaining the
to the I/Os. In mixed 5V/3.3V/2.5V systems, the user pins, same pinout.
the core power supply (VCCINT), and the output power sup-
The XC9500XL architecture provides for superior pin-lock-
ply (VCCIO) may have power applied in any order. This
ing characteristics with a combination of large number of
makes the XC9500XL devices immune to power supply
routing switches in the FastCONNECT II switch matrix, a
sequencing problems.
54-wide input Function Block, and flexible, bi-directional
product term allocation within each macrocell. These fea-

R
tures address design changes that require adding or For extensive design changes requiring higher logic capac-
changing internal routing, including additional signals into ity than is available in the initially chosen device, the new
existing equations, or increasing equation complexity, design may be able to fit into a larger pin-compatible device
respectively. using the same pin assignments. The same board may be
used with a higher density device without the expense of
board rework.
Output Output
Voltage Voltage
VCCIO
Standard
Slew-Rate Limited
Slew-Rate Limited
tSLEW tSLEW
1.5 V 1.5 V
Standard
Time Time
0 0
(a) (b) X5900_01
Figure 12: Output Slew-Rate Control For (a) Rising and (b) Falling Outputs
must remain low during this time, then a pulldown resistor

may be added to the pin.
Set to PIN
during valid user
operation External Programming
Drive to
VCCIO level XC9500XL devices can also be programmed by the Xilinx
HW-130 device programmer as well as third-party program-
0
mers. This provides the added flexibility of using pre-pro-
PIN
grammed devices during manufacturing, with an in-system
RBH
programmable option for future enhancements and design
changes.
Reliability and Endurance

I/O All XC9500XL CPLDs provide a minimum endurance level
of 10,000 in-system program/erase cycles and a minimum
data retention of 20 years. Each device meets all func-
99011200
tional, performance, and data retention specifications
Figure 13: Bus-Hold Logic within this endurance limit.
IEEE 1149.1 Boundary-Scan (JTAG)

In-System Programming XC9500XL devices fully support IEEE 1149.1 bound-
ary-scan (JTAG). EXTEST, SAMPLE/PRELOAD, BYPASS,
One or more XC9500XL devices can be daisy chained
USERCODE, INTEST, IDCODE, HIGHZ and CLAMP
together and programmed in-system via a standard 4-pin
instructions are supported in each device. Additional
JTAG protocol, as shown in Figure 14. In-system program-
instructions are included for in-system programming opera-
ming offers quick and efficient design iterations and elimi-
tions.
nates package handling. The Xilinx development system
provides the programming data sequence using a Xilinx
download cable, a third-party JTAG development system,
Design Security
JTAG-compatible board tester, or a simple microprocessor XC9500XL devices incorporate advanced data security
interface that emulates the JTAG instruction sequence. features which fully protect the programming data against
unauthorized reading or inadvertent device erasure/repro-
All I/Os are 3-stated and pulled high by the bus-hold cir-
gramming. Table 3 shows the four different security settings
cuitry during in-system programming. If a particular signal
available.

R
The read security bits can be set by the user to prevent the Table 3: Data Security Options
internal programming pattern from being read or copied.
When set, they also inhibit further program operations but
allow device erase. Erasing the entire device is the only Read Security
way to reset the read security bit.
Default Set
The write security bits provide added protection against
Read Allowed Read Inhibited
accidental device erasure or reprogramming when the
Write Security
JTAG pins are subject to noise, such as during system Default Program/Erase Program Inhibit
power-up. Once set, the write-protection may be deacti- Allowed Erase Allowed
vated when the device needs to be reprogrammed with a
valid pattern with a specific sequence of JTAG instructions.
Set Program/Erase
Program/Erase
Allowed Inhibited
V CC
GN D
(a) (b) X5902
Figure 14: In-System Programming Operation (a) Solder Device to PCB and (b) Program Using Download Cable
Low Power Mode Timing Model

All XC9500XL devices offer a low-power mode for individ- The uniformity of the XC9500XL architecture allows a sim-
ual macrocells or across all macrocells. This feature allows plified timing model for the entire device. The basic timing
the device power to be significantly reduced. model, shown in Figure 15, is valid for macrocell functions
that use the direct product terms only, with standard power
Each individual macrocell may be programmed in
setting, and standard slew rate setting. Table 4 shows how
low-power mode by the user. Performance-critical parts of
each of the key timing parameters is affected by the product
the application can remain in standard power mode, while
term allocator (if needed), low-power setting, and slew-lim-
other parts of the application may be programmed for
ited setting.
low-power operation to reduce the overall power dissipa-
tion. Macrocells programmed for low-power mode incur The product term allocation time depends on the logic span
additional delay (tLP) in pin-to-pin combinatorial delay as of the macrocell function, which is defined as one less than
well as register setup time. Product term clock to output the maximum number of allocators in the product term
and product term output enable delays are unaffected by path. If only direct product terms are used, then the logic
the macrocell power-setting. span is 0. The example in Figure 6 shows that up to 15
product terms are available with a span of 1. In the case of
Figure 7, the 18 product term function has a span of 2.

R
Detailed timing information may be derived from the full tim- for each parameter are given in the individual device data
ing model shown in Figure 16. The values and explanations sheets.
tSU
Combinatorial Combinatorial
Logic Logic D/T Q
tCO
Propagation Delay = tPD Setup Time = tSU Clock to Out Time = tCO
(a) (b)
tPSU
Combinatorial
Logic D/T Q
Combinatorial
P-Term Clock Logic D/T Q
Path
tPCO
Setup Time = tPSU Clock to Out Time = tPCO Internal System Cycle Time = tSYSTEM
(c) (d)
99011201
Figure 15: Basic Timing Model
tF
tLOGILP S*tPTA tSLEW

tPDI
tIN tLOGI
D/T Q tOUT
tSUI tCOI
tHI
tPTCK
EC t
AOI
tEN
tGCK > tRAI
tPTSR SR
tGSR
tPTTS
Macrocell
tGTS
Figure 16: Detailed Timing Model

R
Power-Up Characteristics ABEL), and simulation capabilities. It supports the

XC9500XL family as well as other CPLD and FPGA fami-
The XC9500XL devices are well behaved under all operat- lies.
ing conditions. During power-up each XC9500XL device
employs internal circuitry which keeps the device in the qui- The Alliance Series includes CPLD and FPGA implementa-
escent state until the VCCINT supply voltage is at a safe tion technology as well as all necessary libraries and inter-
level (approximately 2.5 V). During this time, all device pins faces for Alliance partner EDA solutions.
and JTAG pins are disabled and all device outputs are dis-
abled with the pins weakly pulled high, as shown in Table 5. FastFLASH Technology
When the supply voltage reaches a safe level, all user reg- An advanced 0.35 micron feature size CMOS Flash process is
isters become initialized (typically within 200 µs), and the used to fabricate all XC9500XL devices. The FastFLASH pro-
device is immediately available for operation, as shown in cess provides high performance logic capability, fast pro-
Figure 17. gramming times, and superior reliability and endurance
If the device is in the erased state (before any user pattern ratings.
is programmed), the device outputs remain disabled with VCCINT
weak pull-up. The JTAG pins are enabled to allow the
device to be programmed at any time. All devices are
shipped in the erased state from the factory. 2.5V
3.8 V
(Typ)
(Typ)
If the device is programmed, the device inputs and outputs
take on their configured states for normal operation. The
JTAG pins are enabled to allow device erasure or bound- 5
ary-scan tests at any time.
0V
Development System Support No
Power
Quiescent
State User Operation
Quiescent
State
No
Power
The XC9500XL family and associated in-system program- Initialization of User Registers X5904
ming capabilities are fully supported in either software solu-
tions available from Xilinx. Figure 17: Device Behavior During Power-up
The Foundation Series is an all-in-one development sys-
tem containing schematic entry, HDL (VHDL, Verilog, and
Table 4: Timing Model Parameters
Product Term Macrocell Output Slew-Limited

Description Parameter
Allocator1 Low-Power Setting Setting
Propagation Delay tPD + tPTA * S + tLP + tSLEW
Global Clock Setup Time tSU + tPTA * S + tLP –
Global Clock-to-output tCO – – + tSLEW
Product Term Clock Setup tPSU + tPTA * S + tLP –
Time
Product Term Clock-to-output tPCO – – + tSLEW
Internal System Cycle Period tSYSTEM + tPTA * S + tLP –
Note: 1. S = the logic span of the function, as defined in the text.
Table 5: XC9500XL Device Characteristics
Device Quiescent Erased Device Valid User

Circuitry State Operation Operation
IOB Bus-Hold Pull-up Pull-up Bus-Hold
Device Outputs Disabled Disabled As Configured
Device Inputs and Clocks Disabled Disabled As Configured
Function Block Disabled Disabled As Configured
JTAG Controller Disabled Enabled Enabled

R

0
R
XC9536XL High Performance
CPLD
September 28, 1998 (Version 1.0) 0 5* Preliminary Product Specification
Features Power Estimation

• 4 ns pin-to-pin logic delays Power dissipation in CPLDs can vary substantially depend-
• System frequency up to 200 MHz ing on the system frequency, design application and output
• 36 macrocells with 800 usable gates loading. To help reduce power dissipation, each macrocell
• Available in small footprint packages in a XC9500XL device may be configured for low-power
- 44-pin PLCC (34 user I/O pins) mode (from the default high-performance mode). In addi-
- 48-pin CSP (36 user I/O pins) tion, unused product-terms and macrocells are automati-
- 64-pin VQFP (36 user I/O pins) cally deactivated by the software to further conserve power.
• Optimized for high-performance 3.3 V systems
For a general estimate of ICC, the following equation may
- Low power operation
be used:
- 5 V tolerant I/O pins accept 5 V, 3.3 V, and 2.5 V
signals ICC (mA) = MCHP(0.5) + MCLP(0.3) + MC(0.0045 mA/MHz) f
- 3.3 V or 2.5 V output capability Where:
- Advanced 0.35 micron feature size CMOS 5
MCHP = Macrocells in high-performance (default) mode
FastFLASH™ technology
• Advanced system features MCLP = Macrocells in low-power mode
- In-system programmable MC = Total number of macrocells used
- Superior pin-locking and routability with
f = Clock frequency (MHz)
FastCONNECT II™ switch matrix
- Extra wide 54-input Function Blocks This calculation is based on typical operating conditions
- Up to 90 product-terms per macrocell with individual using a pattern of 16-bit up/down counters in each Function
product-term allocation Block with no output loading. The actual ICC value varies
- Local clock inversion with 3 global and one with the design application and should be verified during
product-term clocks normal system operation.
- Individual output enable per output pin Figure 1 shows the above estimation in a graphical form.
- Input hysteresis on all user and boundary-scan pin
inputs
- Bus-hold circuitry on all user pin inputs
- Full IEEE Standard 1149.1 boundary-scan (JTAG) 60
• Fast concurrent programming 200 MHz

50
• Enhanced data security features
Typical ICC (mA)
40
• Excellent quality and reliability nce
ma 125 MHz
- Endurance exceeding 10,000 program/erase cycles e rfor
30 hP
- 20 year data retention H ig
er
- ESD protection exceeding 2,000 V Pow
20 Low
• Pin-compatible with 5 V-core XC9536 device in the
44-pin PLCC package and the 48-pin CSP package 10
Description
0 50 100 150 200
The XC9536XL is a 3.3 V CPLD targeted for high-perfor- Clock Frequency (MHz)
X5836xl
mance, low-voltage applications in leading-edge communi-
cations and computing systems. It is comprised of two
Figure 1: Typical Icc vs. Frequency for XC9536XL
54V18 Function Blocks, providing 800 usable gates with
propagation delays of 4 ns. See Figure 2 for architecture
overview.
September 28, 1998 (Version 1.0) 5-21

R
3
JTAG
JTAG Port 1 In-System Programming Controller
Controller
54
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O

I/O 54
Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
I/O
I/O
I/O
3
I/O/GCK
1
I/O/GSR
2
I/O/GTS
X5922C_1
Function Block outputs (indicated by the bold line) drive the I/O Blocks directly.
5-22 September 28, 1998 (Version 1.0)

R
Absolute Maximum Ratings

VIN Input voltage relative to GND (Note 1) -0.5 to 5.5 V
VTS Voltage applied to 3-state output (Note 1) -0.5 to 5.5 V
o
TSTG Storage temperature (ambient) -65 to +150 C
TSOL Maximum soldering temperature (10s @ 1/16 in. = 1.5 mm) +260 oC
o
TJ Junction temperature +150 C
Note 1: Maximum DC undershoot below GND must be limited to either 0.5 V or 10 mA, whichever is easier to achieve. During
transitions, the device pins may undershoot to -2.0 V or overshoot to +7.0 V, provided this over- or undershoot lasts less
than 10 ns and with the forcing current being limited to 200 mA.
Note 2: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress
reliability.
Recommended Operation Conditions

Symbol Parameter Min Max Units
VCCINT Supply voltage for internal logic Commercial TA = 0oC to 70oC 3.0 3.6 V 5
and input buffers Industrial TA = -40oC to +85oC 3.0 3.6 V
VCCIO Supply voltage for output drivers for 3.3 V operation 3.0 3.6 V
Supply voltage for output drivers for 2.5 V operation 2.3 2.7 V
VIL Low-level input voltage 0 0.80 V
VIH High-level input voltage 2.0 5.5 V
VO Output voltage 0 VCCIO V
Quality and Reliability Characteristics

tDR Data Retention 20 - Years
NPE Program/Erase Cycles (Endurance) 10,000 - Cycles
VESD Electrostatic Discharge (ESD) 2,000 - Volts
DC Characteristic Over Recommended Operating Conditions

Symbol Parameter Test Conditions Min Max Units
VOH Output high voltage for 3.3 V outputs IOH = -4.0 mA 2.4 V
Output high voltage for 2.5 V outputs IOH = -500 µA 90% VCCIO V
VOL Output low voltage for 3.3 V outputs IOL = 8.0 mA 0.4 V
Output low voltage for 2.5 V outputs IOL = 500 µA 0.4 V
IIL Input leakage current VCC = Max ± 10.0 µA
VIN = GND or VCC
IIH I/O high-Z leakage current VCC = Max ± 10.0 µA
VIN = GND or VCC
CIN I/O capacitance VIN = GND 10.0 pF
f = 1.0 MHz
ICC Operating Supply Current VI = GND, No load 10(Typ) mA
(low power mode, active) f = 1.0 MHz

R
AC Characteristics
XC9536XL-4 XC9536XL-5 XC9536XL-7 XC9536XL-10
Symbol Parameter Units
Min1 Max1 Min Max Min Max Min Max
tPD I/O to output valid 4.0 5.0 7.5 10.0 ns
tSU I/O setup time before GCK 3.0 3.7 4.8 6.5 ns
tH I/O hold time after GCK 0.0 0.0 0.0 0.0 ns
tCO GCK to output valid 3.0 3.5 4.5 5.8 ns
fSYSTEM Multiple FB internal operating 200.0 178.6 125.0 100.0 MHz
frequency
tPSU I/O setup time before p-term 1.2 1.7 1.6 2.1 ns
clock input
tPH I/O hold time after p-term clock 1.8 2.0 3.2 4.4 ns
input
tPCO P-term clock output valid 4.8 5.5 7.7 10.2 ns
tOE GTS to output valid 3.5 4.0 5.0 7.0 ns
tOD GTS to output disable 4.0 5.0 7.0 ns
tPOE Product term OE to output en- 6.5 7.0 9.5 11.0 ns
abled
tPOD Product term OE to output dis- 7.0 9.5 11.0 ns
abled
tAO GSR to output valid 7.6 10.0 12.0 14.5 ns
tPAO P-term S/R to output valid 8.4 10.5 12.6 15.3 ns
tWLH GCK pulse width (High or Low) 2.5 2.8 4.0 4.5 ns
tPLH P-term clock pulse width (High 5.0 5.0 6.5 7.0 ns
or Low)
Advance Preliminary
Note 1:Please contact Xilinx for up-to-date information on advance specifications.
VTEST
R1
Output Type VCCIO VTEST R1 R2 CL
Device Output 3.3 V 3.3 V 320 Ω 360 Ω 35 pF
2.5 V 2.5 V 250 Ω 660 Ω 35 pF
R2 CL
X5906A
Figure 3: AC Load Circuit

R
Internal Timing Parameters

XC9536XL-4 XC9536XL-5 XC9536XL-7 XC9536XL-10
Min1 Max1 Min Max Min Max Min Max
Buffer Delays
tIN Input buffer delay 1.2 1.5 2.3 3.5 ns
tGCK GCK buffer delay 1.0 1.1 1.5 1.8 ns
tGSR GSR buffer delay 1.2 2.0 3.1 4.5 ns
tGTS GTS buffer delay 3.5 4.0 5.0 7.0 ns
tOUT Output buffer delay 1.6 2.0 2.5 3.0 ns
tEN Output buffer enable/disable 0.0 0.0 0.0 0.0 ns
delay
Product Term Control Delays
tPTCK Product term clock delay 1.6 1.6 2.4 2.7 ns
tPTSR Product term set/reset delay 0.8 1.0 1.4 1.8 ns
tPTTS Product term 3-state delay 5.3 5.5 7.2 7.5 ns
Internal Register and Combinatorial Delays
tPDI Combinatorial logic propagation 0.4 0.5 1.3 1.7 ns
delay
5
tSUI Register setup time 2.0 2.3 2.6 3.0 ns
tHI Register hold time 1.0 1.4 2.2 3.5 ns
tECSU Register clock enable setup 2.0 2.3 2.6 3.0 ns
time
tECHO Register clock enable hold time 1.0 1.4 2.2 3.5 ns
tCOI Register clock to output valid 0.4 0.4 0.5 1.0 ns
time
tAOI Register async. S/R to output 4.8 6.0 6.4 7.0 ns
delay
tRAI Register async. S/R recover be- 5.0 7.5 10.0 ns
fore clock
tLOGI Internal logic delay 0.8 1.0 1.4 1.8 ns
tLOGILP Internal low power logic delay 3.8 5.0 6.4 7.3 ns
Feedback Delays
tF FastCONNECT II™ feedback 1.8 1.9 3.5 4.2 ns
delay
Time Adders
tPTA Incremental product term allo- 0.6 0.7 0.8 1.0 ns
cator delay
tSLEW Slew-rate limited delay 2.5 3.0 4.0 4.5 ns
Advance Preliminary
Note 1: Please contact Xilinx for up-to-date information on advance specifications.

R
XC9536XL I/O Pins

Function BScan Function BScan
Macrocell PC44 CS48 VQ64 Notes Macrocell PC44 CS48 VQ64 Notes
Block Order Block Order
1 1 2 D6 9 105 2 1 1 D7 8 51
1 2 3 C7 10 102 2 2 44 E5 7 48
1 3 5 B7 15 99 [1] 2 3 42 E6 5 45 [1]
1 4 4 C6 11 96 2 4 43 E7 6 42
1 5 6 B6 16 93 [1] 2 5 40 F6 2 39 [1]
1 6 8 A6 19 90 2 6 39 G7 64 36 [1]
1 7 7 A7 17 87 [1] 2 7 38 G6 63 33
1 8 9 C5 20 84 2 8 37 F5 62 30
1 9 11 B5 22 81 2 9 36 G5 61 27
1 10 12 A4 24 78 2 10 35 F4 60 24
1 11 13 B4 25 75 2 11 34 G4 57 21
1 12 14 A3 27 72 2 12 33 E3 56 18
1 13 18 B2 33 69 2 13 29 F2 50 15
1 14 19 B1 35 66 2 14 28 G1 48 12
1 15 20 C2 36 63 2 15 27 F1 45 9
1 16 22 C3 38 60 2 16 26 E2 44 6
1 17 24 D2 42 57 2 17 25 E1 43 3
1 18 - D3 39 54 2 18 - E4 49 0
Note 1: Global control pin.
XC9536XL Global, JTAG and Power Pins

Pin Type PC44 CS48 VQ64
I/O/GCK1 5 B7 15
I/O/GCK2 6 B6 16
I/O/GCK3 7 A7 17
I/O/GTS1 42 E6 5
I/O/GTS2 40 F6 2
I/O/GSR 39 G7 64
TCK 17 A1 30
TDI 15 B3 28
TDO 30 G2 53
TMS 16 A2 29
VCCINT 3.3 V 21, 41 C1, F7 3, 37
VCCIO 2.5 V/3.3 V 32 G3 55
GND 10, 23, 31 A5, D1, F3 21, 41, 54
No Connects – – 1, 4, 12, 13, 14, 18,
23, 26, 31, 32, 34, 40,
46, 47, 51, 52, 58, 59

R
Example: XC9536XL -5 PC 44 C
Number of Pins
Speed Grade Package Type
Speed Options Packaging Options

-10 10 ns pin-to-pin delay PC44 44-Pin Plastic Lead Chip Carrier (PLCC)
-7 7.5 ns pin-to-pin delay
CS48 48-Pin Chip Scale Package
-5 5 ns pin-to-pin delay
-4 4 ns pin-to-pin delay VQ64 64-Pin Quad Flat Pack (VQFP)
Temperature Options
C = Commercial TA = 0oC to +70oC
I = Industrial TA = -40oC to +85oC 5
Component Availability
Pins 44 48 64
Type Plastic Plastic Plastic
PLCC CSP VQFP
Code PC44 CS48 VQ64
-10 C, I - C, I
-7 C C C
XC9536XL
-5 C C C
-4 (C) - (C)
C = Commercial (TA = 0oC to +70oC) I = Industrial (TA = -40oC to +85oC)
( ) Parenthesis indicate future planed products. Please contact Xilinx for up-to-date
information.

R

0
R
CPLD
September 28, 1998 (Version 1.0) 0 5* Preliminary Product Specification

• 72 macrocells with 1,600 usable gates loading. To help reduce power dissipation, each macrocell
- 44-pin PLCC (34 user I/O pins) mode (from the default high-performance mode). In addi-
- 48-pin CSP (38 user I/O pins) tion, unused product-terms and macrocells are automati-
- 64-pin VQFP (52 user I/O pins) cally deactivated by the software to further conserve power.
- 100-Pin TQFP (72 user I/O pins)
be used:
- 5 V tolerant I/O pins accept 5 V, 3.3 V, and 2.5 V ICC (mA) = MCHP(0.5) + MCLP(0.3) + MC(0.0045 mA/MHz) f
signals Where:
- 3.3 V or 2.5 V output capability 5
- Advanced 0.35 micron feature size CMOS
FastFLASH™ technology MCLP = Macrocells in low-power mode
• Advanced system features MC = Total number of macrocells used
- In-system programmable
FastCONNECT II™ switch matrix This calculation is based on typical operating conditions
- Extra wide 54-input Function Blocks using a pattern of 16-bit up/down counters in each Function
- Up to 90 product-terms per macrocell with individual Block with no output loading. The actual ICC value varies
product-term allocation with the design application and should be verified during
- Local clock inversion with 3 global and one normal system operation.
product-term clocks Figure 1 shows the above estimation in a graphical form.
- Individual output enable per output pin
inputs
- Bus-hold circuitry on all user pin inputs 100
178 MHz
• Fast concurrent programming 80
• Enhanced data security features ce
60 an
rm
• Excellent quality and reliability rf o 104 MHz
Typical ICC (mA)
e
hP
- Endurance exceeding 10,000 program/erase cycles Hig
40 r
- 20 year data retention Po
we
w
- ESD protection exceeding 2,000 V Lo
• Pin-compatible with 5 V-core XC9572 device in the 20
44-pin PLCC package and the 100-pin TQFP package
Description 0 50 100
Clock Frequency (MHz)
150 200
X5872xl
The XC9572XL is a 3.3 V CPLD targeted for high-perfor-

Figure 1: Typical Icc vs. Frequency for XC9572XL
cations and computing systems. It is comprised of four
54V18 Function Blocks, providing 1,600 usable gates with
overview.

R
3
JTAG
Controller
54
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O

I/O 54
Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
54
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 54
1 Function
I/O/GSR 18 Block 4
4 Macrocells
I/O/GTS 1 to 18
54
Function
18 Block 4
Macrocells
1 to 18
Figure 2: XC9572XL Architecture X5922D

R

o
o
Conditions is not implied. Exposure to Absolute Maximum Ratings conditions for extended periods of time may affect
device reliability.



VIN = GND or VCC
VIN = GND or VCC
f = 1.0 MHz
ICC Operating Supply Current VI = GND, No load 20 (Typ) mA

R
AC Characteristics
XC9572XL-5 XC9572XL-7 XC9572XL-10
Min1 Max1 Min Max Min Max
tPD I/O to output valid 5.0 7.5 10.0 ns
tSU I/O setup time before GCK 3.7 4.8 6.5 ns
tH I/O hold time after GCK 0.0 0.0 0.0 ns
tCO GCK to output valid 3.5 4.5 5.8 ns
fSYSTEM Multiple FB internal operating fre- 178.6 125.0 100.0 MHz
quency
tPSU I/O setup time before p-term clock 1.7 1.6 2.1 ns
input
tPH I/O hold time after p-term clock input 2.0 3.2 4.4 ns
tPCO P-term clock output valid 5.5 7.7 10.2 ns
tOE GTS to output valid 4.0 5.0 7.0 ns
tPOE Product term OE to output enabled 7.0 9.5 11.0 ns
tPOD Product term OE to output disabled 7.0 9.5 11.0 ns
tAO GSR to output valid 10.0 12.0 14.5 ns
tPAO P-term S/R to output valid 10.5 12.6 15.3 ns
tWLH GCK pulse width (High or Low) 2.8 4.0 4.5 ns
tPLH P-term clock pulse width (High or 5.0 6.5 7.0 ns
Low)
Advance Preliminary
VTEST
R1
2.5 V 2.5 V 250 Ω 660 Ω 35 pF
R2 CL
X5906A

R

XC9572XL-5 XC9572XL-7 XC9572XL-10
Buffer Delays
tIN Input buffer delay 1.5 2.3 3.5 ns
tGCK GCK buffer delay 1.1 1.5 1.8 ns
tGSR GSR buffer delay 2.0 3.1 4.5 ns
tGTS GTS buffer delay 4.0 5.0 7.0 ns
tOUT Output buffer delay 2.0 2.5 3.0 ns
tEN Output buffer enable/disable delay 0.0 0.0 0.0 ns
tPTCK Product term clock delay 1.6 2.4 2.7 ns
tPTSR Product term set/reset delay 1.0 1.4 1.8 ns
tPTTS Product term 3-state delay 5.5 7.2 7.5 ns
tPDI Combinatorial logic propagation de- 0.5 1.3 1.7 ns
lay
tSUI Register setup time 2.3 2.6 3.0 ns 5
tHI Register hold time 1.4 2.2 3.5 ns
tECSU Register clock enable setup time 2.3 2.6 3.0 ns
tECHO Register clock enable hold time 1.4 2.2 3.5 ns
tCOI Register clock to output valid time 0.4 0.5 1.0 ns
tAOI Register async. S/R to output delay 6.0 6.4 7.0 ns
tRAI Register async. S/R recover before 5.0 7.5 10.0 ns
clock
tLOGI Internal logic delay 1.0 1.4 1.8 ns
tLOGILP Internal low power logic delay 5.0 6.4 7.3 ns
Feedback Delays
tF FastCONNECT II™ feedback delay 1.9 3.5 4.2 ns
Time Adders
tPTA Incremental product term allocator 0.7 0.8 1.0 ns
delay
tSLEW Slew-rate limited delay 3.0 4.0 4.5 ns
Advance Preliminary

R
XC9572XL I/O Pins

Func- Func-
Macro- BScan Macro- BScan
tion PC44 CS48 VQ64 TQ100 Notes tion PC44 CS48 VQ64 TQ100 Notes
cell Order cell Order
Block Block
1 1 - - - 16 213 3 1 - - - 41 105
1 2 1 D7 8 13 210 3 2 11 B5 22 32 102
1 3 - D4 12 18 207 3 3 - C4 31 49 99
1 4 - - 13 20 204 3 4 - - 32 50 96
1 5 2 D6 9 14 201 3 5 12 A4 24 35 93
1 6 3 C7 10 15 198 3 6 - - 34 53 90
1 7 - - - 25 195 3 7 - - - 54 87
1 8 4 C6 11 17 192 3 8 13 B4 25 37 84
1 9 5 B7 15 22 189 [1] 3 9 14 A3 27 42 81
1 10 - - 18 28 186 3 10 - D3 39 60 78
1 11 6 B6 16 23 183 [1] 3 11 18 B2 33 52 75
1 12 - - 23 33 180 3 12 - - 40 61 72
1 13 - - - 36 177 3 13 - - - 63 69
1 14 7 A7 17 27 174 [1] 3 14 19 B1 35 55 66
1 15 8 A6 19 29 171 3 15 20 C2 36 56 63
1 16 - - - 39 168 3 16 24 D2 42 64 60
1 17 9 C5 20 30 165 3 17 22 C3 38 58 57
1 18 - - - 40 162 3 18 - - - 59 54
2 1 - - - 87 159 4 1 - - - 65 51
2 2 35 F4 60 94 156 4 2 25 E1 43 67 48
2 3 - - 58 91 153 4 3 - - 46 71 45
2 4 - - 59 93 150 4 4 - - 47 72 42
2 5 36 G5 61 95 147 4 5 26 E2 44 68 39
2 6 37 F5 62 96 144 4 6 - E4 49 76 36
2 7 - - - 3 141 [2] 4 7 - - - 77 33
2 8 38 G6 63 97 138 4 8 27 F1 45 70 30
2 9 39 G7 64 99 135 [1] 4 9 - - - 66 27
2 10 - - 1 1 132 4 10 - - 51 81 24
2 11 40 F6 2 4 129 [1] 4 11 28 G1 48 74 21
2 12 - - 4 6 126 4 12 - - 52 82 18
2 13 - - - 8 123 4 13 - - - 85 15
2 14 42 E6 5 9 120 [3] 4 14 29 F2 50 78 12
2 15 43 E7 6 11 117 4 15 33 E3 56 89 9
2 16 - - - 10 114 4 16 - - - 86 6
2 17 44 E5 7 12 111 4 17 34 G4 57 90 3
2 18 - - - 92 108 4 18 - - - 79 0

Note 2: GTS1 for TQ100
Note 3: GTS1 for PC44, CS48, VQ64

R

Pin Type PC44 CS48 VQ64 TQ100
I/O/GCK1 5 B7 15 22
I/O/GCK2 6 B6 16 23
I/O/GCK3 7 A7 17 27
I/O/GTS1 42 E6 5 3
I/O/GTS2 40 F6 2 4
I/O/GSR 39 G7 64 99
TCK 17 A1 30 48
TDI 15 B3 28 45
TDO 30 G2 53 83
TMS 16 A2 29 47
VCCINT 3.3 V 21, 41 C1, F7 3, 37 5, 57, 98
VCCIO 2.5 V/3.3 V 32 G3 26, 55 26, 38, 51, 88
10, 23, 31 A5, D1, F3 14, 21, 41, 54 21, 31, 44, 62,
GND
69, 75, 84, 100
No Connects - – – 2, 7, 19, 24, 34,
43, 46, 73, 80
5
Example: XC9572XL -7 TQ 100 C

Number of Pins

-10 10 ns pin-to-pin delay PC44 44-Pin Plastic Lead Chip Carrier (PLCC)
-7 7.5 ns pin-to-pin delay
CS48 48-Pin Chip Scale Package (CSP)
VQ64 64-Pin Very Thin Quad Flat Pack (VQFP)
TQ100 100-Pin Thin Quad Flat Pack (TQFP)
Temperature Options
C = Commercial TA = 0oC to +70oC
I = Industrial TA = -40oC to +85oC

R
Pins 44 48 64 100
Type Plastic Plastic Plastic Plastic
PLCC CSP VQFP TQFP
Code PC44 CS48 VQ64 TQ100
-10 C, I - C, I C, I
XC9572XL -7 C C C C
-5 (C) - (C) (C)
( ) Parenthesis indicate future planned products. Please contact XIlinx for up-to-date availability information.

0
R
CPLD
November 13, 1998 (Version 1.2) 0 5* Preliminary Product Specification

• 5 ns pin-to-pin logic delays Power dissipation in CPLDs can very substantially depend-
• System frequency up to 178 MHz ing on the system frequency, design application, and output
- 100-pin TQFP (81 user I/O pins) mode (from the default high-performance mode). In addi-
- 144-pin TQFP (117 user I/O pins) tion, unused product-terms and macrocells are automati-
- 144-pin CSP (117 user I/O pins) cally deactivated by the software to further conserve power.
- Low power operation For a general estimate of ICC, the following equation may
- 5 V tolerant I/O pins accept 5 V, 3.3 V, and 2.5 V be used:
FastFLASH™ technology MCHP = Macrocells in high-performance (default) mode
- In-system programmable
MC = Total number of macrocells used
FastCONNECT II™ switch matrix f = Clock frequency (MHz)
- Individual output enable per output pin with local
inversion Figure 1 shows the above estimation in graphical form.
inputs
200
- Bus-hold ciruitry on all user pin inputs 178 MHz
• Fast concurrent programming nce
150 rma
• Slew rate control on individual outputs erfo
Typical ICC (mA)
h P
• Enhanced data security features Hig
104 MHz
• Excellent quality and reliability
100
- Endurance exceeding 10,000 program/erase cycles er
- 20 year data retention P ow
Low
- ESD protection exceeding 2,000 V 50
• Pin-compatible with 5 V-core XC95144 device in the
100-pin TQFP package
Description 0 50 100 150 200

The XC95144XL is a 3.3 V CPLD targeted for high-perfor- X5898C

cations and computing systems. It is comprised of eight Figure 1: Typical Icc vs. Frequency for XC95144XL
overview.

R
3
JTAG
Controller
54
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O

I/O 54
Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
54
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 54
1 Function
I/O/GSR 18 Block 4
4 Macrocells
I/O/GTS 1 to 18
54
Function
18 Block 8
Macrocells
1 to 18
X5922B

R

o
o
reliability.


DC Characteristics Over Recommended Operating Conditions

VIN = GND or VCC
VIN = GND or VCC
f = 1.0 MHz
ICC Operating Supply Current VI = GND, No load 45 ma

R
AC Characteristics
XC95144XL-5 XC95144XL-7 XC95144XL-10
fSYSTEM Multiple FB internal operating frequency 178.6 125.0 100.0 MHz
tPSU I/O setup time before p-term clock input 1.7 1.6 2.1 ns
tPCO P-term clock output valid 5.5 7.7 10.2 ns
tAO GSR to output valid 10.0 12.0 14.5 ns
tPAO P-term S/R to output valid 10.5 12.6 15.3 ns
tPLH P-term clock pulse width (High or Low) 5.0 6.5 7.0 ns
Advance Preliminary
Note 1:Please contact Xilinx for up-to-date information on advance specifications.
VTEST
R1
2.5 V 2.5 V 250 Ω 660 Ω 35 pF
R2 CL
X5906A

R

XC95144XL-5 XC95144XL-7 XC95144XL-10
Buffer Delays
tPDI Combinatorial logic propagation delay 0.5 1.3 1.7 ns
tSUI Register setup time 2.3 2.6 3.0 ns
tHI Register hold time 1.4 2.2 3.5 ns 5
tECSU Register clock enable setup time 2.3 2.6 3.0 ns
tECHO Register clock enable hold time 1.4 2.2 3.5 ns
tRAI Register async. S/R recover before clock 5.0 7.5 10.0 ns
Feedback Delays
tF FastCONNECT II™ feedback delay 1.9 3.5 4.2 ns
Time Adders
tPTA Incremental product term allocator delay 0.7 0.8 1.0 ns
Advance Preliminary

R
XC95144XL I/O Pins

Macrocell TQ100 TQ144 CS144 Notes Macrocell TQ100 TQ144 CS144 Notes
1 1 – 23 H3 429 3 1 – 39 M3 321
1 2 11 16 F1 426 3 2 23 32 L1 318 [1]
1 3 12 17 G2 423 3 3 – 41 K4 315
1 4 – 25 J1 420 3 4 – 44 N4 312
1 5 13 19 G3 417 3 5 24 33 L2 309
1 6 14 20 G4 414 3 6 25 34 L3 306
1 7 - - - 411 3 7 – 46 L5 303
1 8 15 21 H1 408 3 8 27 38 N2 300 [1]
1 9 16 22 H2 405 3 9 28 40 N3 297
1 10 – 31 K3 402 3 10 – 48 N5 294
1 11 17 24 H4 399 3 11 29 43 M4 291
1 12 18 26 J2 396 3 12 30 45 K5 288
1 13 – - - 393 3 13 – - - 285
1 14 19 27 J3 390 3 14 32 49 K6 282
1 15 20 28 J4 387 3 15 33 50 L6 279
1 16 – 35 M1 384 3 16 – - - 276
1 17 22 30 K2 381 [1] 3 17 34 51 M6 273
1 18 – – - 378 3 18 – – - 270
2 1 – 142 C3 375 4 1 – 118 C9 267
2 2 99 143 A2 372 [1] 4 2 87 126 A7 264
2 3 – - - 369 4 3 – 133 A5 261
2 4 – 4 C1 366 4 4 – - - 258
2 5 1 2 B1 363 [1] 4 5 89 128 D7 255
2 6 2 3 C2 360 [1] 4 6 90 129 A6 252
2 7 – - - 357 4 7 – - - 249
2 8 3 5 D4 354 [1] 4 8 91 130 B6 246
2 9 4 6 D3 351 [1] 4 9 92 131 C6 243
2 10 – 7 D2 348 4 10 – 135 C5 240
2 11 6 9 E4 345 4 11 93 132 D6 237
2 12 7 10 E3 342 4 12 94 134 B5 234
2 13 – 12 E1 339 4 13 – 137 A4 231
2 14 8 11 E2 336 4 14 95 136 D5 228
2 15 9 13 F4 333 4 15 96 138 B4 225
2 16 – 14 F3 330 4 16 – 139 C4 222
2 17 10 15 F2 327 4 17 97 140 A3 219
2 18 – – - 324 4 18 - - - 216

R
XC95144XL (Continued)

Macrocell TQ100 TQ144 CS144 Notes Macrocell TQ100 TQ144 CS144 Notes
5 1 - - - 213 7 1 – – - 105
5 2 35 52 N6 210 7 2 50 71 N12 102
5 3 – 59 L8 207 7 3 – 75 L12 99
5 4 – - - 204 7 4 – - - 96
5 5 36 53 M7 201 7 5 52 74 M13 93
5 6 37 54 N7 198 7 6 53 76 L13 90
5 7 – 66 M10 195 7 7 – 77 K10 87
5 8 39 56 K7 192 7 8 54 78 K11 84
5 9 40 57 N8 189 7 9 55 80 K13 81
5 10 – 68 N11 186 7 10 – 79 K12 78
5 11 41 58 M8 183 7 11 56 82 J11 75
5 12 42 60 K8 180 7 12 58 85 H10 72
5 13 – 70 L11 177 7 13 – 81 J10 69
5 14 43 61 N9 174 7 14 59 86 H11 66 5
5 15 46 64 K9 171 7 15 60 87 H12 63
5 16 – - - 168 7 16 – 83 J12 60
5 17 49 69 M11 165 7 17 61 88 H13 57
5 18 – – - 162 7 18 – – - 54
6 1 – – - 159 8 1 – – - 51
6 2 74 106 C11 156 8 2 63 91 G11 48
6 3 – - - 153 8 3 – 95 F11 45
6 4 – 111 B11 150 8 4 – 97 E13 42
6 5 76 110 A12 147 8 5 64 92 G10 39
6 6 77 112 A11 144 8 6 65 93 F13 36
6 7 – - - 141 8 7 – - - 33
6 8 78 113 D10 138 8 8 66 94 F12 30
6 9 79 116 A10 135 8 9 67 96 F10 27
6 10 – 115 B10 132 8 10 – 101 D13 24
6 11 80 119 B9 129 8 11 68 98 E12 21
6 12 81 120 A9 126 8 12 70 100 E10 18
6 13 – - - 123 8 13 – 103 D11 15
6 14 82 121 D8 120 8 14 71 102 D12 12
6 15 85 124 A8 117 8 15 72 104 C13 9
6 16 – 117 D9 114 8 16 – 107 B13 6
6 17 86 125 B7 111 8 17 73 105 C12 3
6 18 – – - 108 8 18 – – - 0

R

Pin Type TQ100 TQ144 CS144
I/O/GCK1 22 30 K2
I/O/GCK2 23 32 L1
I/O/GCK3 27 38 N2
I/O/GTS1 3 5 D4
I/O/GTS2 4 6 D3
I/O/GTS3 1 2 B1
I/O/GTS4 2 3 C2
I/O/GSR 99 143 A2
TCK 48 67 L10
TDI 45 63 L9
TDO 83 122 C8
TMS 47 65 N10
VCCINT 3.3 V 5, 57, 98 8, 42, 84, 141 B3, D1, J13, L4
VCCIO 2.5 V/3.3 V 26, 38, 51, 88 1, 37, 55, 73, 109, 127 A1, A13, C7, L7, N1, N13
21, 31, 44, 62, 69, 75, 84, 18, 29, 36, 47, 62, 72, 89, B2, B8, B12, C10, E11,
GND 100 90, 99, 108, 114, 123, 144 G1, G12, G13, K1, M2,
M5, M9, M12
No Connects - – –

Number of Pins

-10 10 ns pin-to-pin delay TQ100 100-Pin Thin Quad Flat Pack (TQFP)
-7 7.5 ns pin-to-pin delay TQ144 144-Pin Thin Quad Flat Pack (TQFP)
-5 5 ns pin-to-pin delay CS144 144-Pin Chip Scale Package (CSP)
Temperature Options
C= Commercial TA = 0oC to +70oC

R
Pins 100 144 144
Type Plastic Plastic Chip Scale Package
TQFP TQFP CSP
Code TQ100 TQ144 CS144
-10 C, I C, I -
XC95144XL -7 C C (C)
-5 (C) (C) -
( ) Parenthesis indicate future planned products. Please contact Xilinx for up-to-date availability
information.
Revision Control
Date Revision
10/30/98 Minor corrections in CS144 pinout table.
11/13/98 V1.2 Minor correction in CS144 pinout table.

R

0
R
CPLD
September 28, 1998 (Version 1.0) 0 5* Advance Product Specification

- 144-pin TQFP (117 user I/O pins) mode (from the default high-performance mode). In addi-
- 208-pin PQFP (168 user I/O pins) tion, unused product-terms and macrocells are automati-
- 352-pin BGA (192 user I/O pins) cally deactivated by the software to further conserve power.
be used:
- 5 V tolerant I/O pins accept 5 V, 3.3 V, and 2.5 V
FastFLASH™ technology
- In-system programmable MC = Total number of macrocells used
FastCONNECT II™ switch matrix
- Individual output enable per output pin Figure 1 shows the above estimation in a graphical form.
inputs
- Bus-hold ciruitry on all user pin inputs
- Full IEEE Standard 1149.1 boundary-scan (JTAG) 400
• Fast concurrent programming 350 151 MHz

• Enhanced data security features 300
ce
• Excellent quality and reliability an
rm
erfo
P
- Endurance exceeding 10,000 program/erase cycles 250
gh
Typical ICC (mA)
Hi
- 20 year data retention 94 MHz
200
- ESD protection exceeding 2,000 V
• Pin-compatible with 5 V-core XC95288 device in the 150 we
r
Po
w
208-pin HQFP package Lo
100
Description
50
The XC95288XL is a 3.3 V CPLD targeted for high-perfor-
0
mance, low-voltage applications in leading-edge communi- 50 100 150 200
cations and computing systems. It is comprised of sixteen Clock Frequency (MHz)
54V18 Function Blocks, providing 6,400 usable gates with x58288xl

overview. Figure 1: Typical Icc vs. Frequency for XC95288XL

R
3
JTAG
Controller
54
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O

I/O 54
Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
54
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 54
1 Function
I/O/GSR 18 Block 4
4 Macrocells
I/O/GTS 1 to 18
54
Function
18 Block 16
Macrocells
1 to 18
X5922E

R

o
o
reliability.



VIN = GND or VCC
VIN = GND or VCC
f = 1.0 MHz
ICC Operating Supply Current VI = GND, No load 85(Typ) mA

R
AC Characteristics
XC95288XL-6 XC95288XL-10
Min1 Max1 Min1 Max1
tPD I/O to output valid 6.0 10.0 ns
tSU I/O setup time before GCK 4.1 6.5 ns
tH I/O hold time after GCK 0.0 0.0 ns
tCO GCK to output valid 4.3 5.8 ns
fSYSTEM Multiple FB internal operating frequency 151.5 100.0 MHz
tPSU I/O setup time before p-term clock input 2.1 2.1 ns
tPH I/O hold time after p-term clock input 2.0 4.4 ns
tPCO P-term clock output valid 6.3 10.2 ns
tOE GTS to output valid 4.5 7.0 ns
tOD GTS to output disable 7.0 ns
tPOE Product term OE to output enabled 8.0 11.0 ns
tPOD Product term OE to output disabled 11.0 ns
tAO GSR to output valid 10.8 14.5 ns
tPAO P-term S/R to output valid 11.6 15.3 ns
tWLH GCK pulse width (High or Low) 3.3 4.5 ns
tPLH P-term clock pulse width (High or Low) 6.0 7.0 ns
Advance
VTEST
R1
2.5 V 2.5 V 250 Ω 660 Ω 35 pF
R2 CL
X5906A

R

XC95288XL-6 XC95288XL-10
Min1 Max1 Min1 Max1
Buffer Delays
tIN Input buffer delay 1.8 3.5 ns
tGCK GCK buffer delay 1.4 1.8 ns
tGSR GSR buffer delay 2.2 4.5 ns
tGTS GTS buffer delay 4.5 7.0 ns
tOUT Output buffer delay 2.4 3.0 ns
tEN Output buffer enable/disable delay 0.0 0.0 ns
tPTCK Product term clock delay 1.6 2.7 ns
tPTSR Product term set/reset delay 1.2 1.8 ns
tPTTS Product term 3-state delay 6.2 7.5 ns
tPDI Combinatorial logic propagation delay 0.6 1.7 ns
tSUI Register setup time 2.5 3.0 ns
tHI Register hold time 1.6 3.5 ns 5
tECSU Register clock enable setup time 2.5 3.0 ns
tECHO Register clock enable hold time 1.6 3.5 ns
tCOI Register clock to output valid time 0.5 1.0 ns
tAOI Register async. S/R to output delay 6.2 7.0 ns
tRAI Register async. S/R recover before clock 10.0 ns
tLOGI Internal logic delay 1.2 1.8 ns
tLOGILP Internal low power logic delay 5.2 7.3 ns
Feedback Delays
tF FastCONNECT II™ feedback delay 2.4 4.2 ns
Time Adders
tPTA Incremental product term allocator delay 0.8 1.0 ns
tSLEW Slew-rate limited delay 3.5 4.5 ns
Advance

R
XC95288XL I/O Pins

Macrocell TQ144 PQ208 BG352 Notes Macrocell TQ144 PQ208 BG352 Notes
1 1 – – – 861 3 1 – – – 753
1 2 – 28 N26 858 3 2 28 38 U24 750
1 3 – 29 P25 855 3 3 – 39 U23 747
1 4 – – – 852 3 4 – – – 744
1 5 20 30 P23 849 3 5 – 40 Y26 741
1 6 21 31 P24 846 3 6 – 41 W25 738
1 7 – – – 843 3 7 – – – 735
1 8 22 32 R26 840 3 8 – 43 AA26 732
1 9 – – R25 837 3 9 – – Y25 729
1 10 23 33 R24 834 3 10 30 44 Y24 726 [1]
1 11 – – R23 831 3 11 – – AA25 723
1 12 24 34 T26 828 3 12 31 45 AB25 720
1 13 – – – 825 3 13 – – – 717
1 14 25 35 T25 822 3 14 32 46 AA24 714 [1]
1 15 26 36 T23 819 3 15 33 47 Y23 711
1 16 – – – 816 3 16 – – – 708
1 17 27 37 V26 813 3 17 – 48 AC26 705
1 18 – – – 810 3 18 – – – 702
2 1 – – – 807 4 1 – – – 699
2 2 9 15 K23 804 4 2 2 3 E23 696 [1]
2 3 10 16 K24 801 4 3 – 4 C26 693
2 4 – – – 798 4 4 – – – 690
2 5 11 17 J25 795 4 5 3 5 E24 687 [1]
2 6 12 18 L24 792 4 6 4 6 F24 684
2 7 – – – 789 4 7 – – – 681
2 8 13 19 K25 786 4 8 5 7 E25 678 [1]
2 9 – – L25 783 4 9 – – D26 675
2 10 14 20 L26 780 4 10 – 8 G24 672
2 11 – – M23 777 4 11 – – F25 669
2 12 15 21 M24 774 4 12 6 9 F26 666 [1]
2 13 – – – 771 4 13 – – – 663
2 14 16 22 M25 768 4 14 7 10 H23 660
2 15 17 23 M26 765 4 15 – 12 G26 657
2 16 – – – 762 4 16 – – – 654
2 17 19 25 N25 759 4 17 – 14 H25 651
2 18 – – – 756 4 18 – – – 648
Notes: [1] Global control pin

R
XC95288XL I/O Pins (continued)

5 1 – – – 645 7 1 – – – 537
5 2 34 49 AA23 642 7 2 – 62 AC19 534
5 3 – 50 AB24 639 7 3 45 63 AD19 531
5 4 – – – 636 7 4 – – – 528
5 5 35 51 AD25 633 7 5 46 64 AE20 525
5 6 – 54 AE24 630 7 6 – 66 AC18 522
5 7 – – – 627 7 7 – – – 519
5 8 38 55 AD23 624 [1] 7 8 – 67 AD18 516
5 9 – – AC22 621 7 9 – – AE19 513
5 10 39 56 AF24 618 7 10 – 69 AD17 510
5 11 – – AD22 615 7 11 – – AE18 507
5 12 40 57 AE23 612 7 12 48 70 AF18 504
5 13 – – – 609 7 13 – – – 501
5 14 41 58 AE22 606 7 14 – 71 AE17 498
5 15 43 60 AE21 603 7 15 49 72 AE16 495
5 16 – – – 600 7 16 – – – 492
5 17 44 61 AF21 597 7 17 – 73 AF16 489
5 18 – – – 594 7 18 – – – 486 5
6 1 – – – 591 8 1 - – – 483
6 2 135 197 C19 588 8 2 130 186 A15 480
6 3 136 198 D18 585 8 3 131 187 B15 477
6 4 – – – 582 8 4 – – – 474
6 5 137 199 A21 579 8 5 132 188 C15 471
6 6 138 200 B20 576 8 6 – 189 D15 468
6 7 – – – 573 8 7 – – – 465
6 8 139 201 C20 570 8 8 133 191 A16 462
6 9 – – B21 567 8 9 – – B16 459
6 10 140 202 B22 564 8 10 134 192 C16 456
6 11 – – C21 561 8 11 – – B17 453
6 12 – 203 D20 558 8 12 – 193 C17 450
6 13 – – – 555 8 13 – – – 447
6 14 142 205 B24 552 8 14 – 194 B18 444
6 15 143 206 C23 549 [1] 8 15 – 195 A20 441
6 16 – – – 546 8 16 – – – 438
6 17 – 208 D22 543 8 17 – 196 B19 435
6 18 – – – 540 8 18 – – – 432
Note: [1] Global control pin

R

9 1 – – – 429 11 1 – – – 321
9 2 50 74 AE14 426 11 2 – 87 AD9 318
9 3 51 75 AF14 423 11 3 60 88 AC10 315
9 4 – – – 420 11 4 – – – 312
9 5 52 76 AE13 417 11 5 61 89 AF7 309
9 6 53 77 AC13 414 11 6 – 90 AE8 306
9 7 – – – 411 11 7 – – – 303
9 8 54 78 AD13 408 11 8 – 91 AD8 300
9 9 – – AF12 405 11 9 – – AE7 297
9 10 – 80 AE12 402 11 10 64 95 AD7 294
9 11 56 82 AD12 399 11 11 66 97 AE5 291
9 12 57 83 AC12 396 11 12 68 99 AC7 288
9 13 – – – 393 11 13 – – – 285
9 14 58 84 AF11 390 11 14 69 100 AE3 282
9 15 – 85 AE11 387 11 15 – 101 AD4 279
9 16 – – – 384 11 16 – – – 276
9 17 59 86 AE9 381 11 17 70 102 AC5 273
9 18 – – – 378 11 18 – – – 270
10 1 – – – 375 12 1 – – – 267
10 2 117 170 C10 372 12 2 110 158 B3 264
10 3 118 171 B9 369 12 3 111 159 A3 261
10 4 – – – 366 12 4 – – – 258
10 5 119 173 A9 363 12 5 112 160 D6 255
10 6 120 174 D11 360 12 6 – 161 C6 252
10 7 – – – 357 12 7 – – – 249
10 8 121 175 B11 354 12 8 113 162 B5 246
10 9 – – A11 351 12 9 – – A4 243
10 10 124 178 C12 348 12 10 115 164 B6 240
10 11 125 179 B12 345 12 11 – 165 A6 237
10 12 126 180 A12 342 12 12 116 166 D8 234
10 13 – – – 339 12 13 – – – 231
10 14 128 182 A13 336 12 14 – 167 B7 228
10 15 – 183 B14 333 12 15 – 168 A7 225
10 16 – – – 330 12 16 – – – 222
10 17 129 185 C14 327 12 17 – 169 D9 219
10 18 – – – 324 12 18 – – – 216

R

13 1 – – – 213 15 1 – – – 105
13 2 71 103 AD3 210 15 2 79 117 V3 102
13 3 – 106 AD2 207 15 3 80 118 W2 99
13 4 – – – 204 15 4 – – – 96
13 5 – 107 AC3 201 15 5 – 119 U4 93
13 6 – 109 AD1 198 15 6 – 120 U3 90
13 7 – – – 195 15 7 – – – 87
13 8 74 110 AA4 192 15 8 81 121 V2 84
13 9 – – AA3 189 15 9 – – V1 81
13 10 – 111 AB2 186 15 10 82 122 U2 78
13 11 75 112 AC1 183 15 11 83 123 T2 75
13 12 – 113 AA2 180 15 12 85 125 R4 72
13 13 – – – 177 15 13 – – – 69
13 14 76 114 AA1 174 15 14 86 126 R3 66
13 15 77 115 Y1 171 15 15 87 127 R2 63
13 16 – – – 168 15 16 – – – 60
13 17 78 116 V4 165 15 17 88 128 R1 57
13 18 – – – 162 15 18 – – – 54 5
14 1 – – – 159 16 1 – – – 51
14 2 – 144 K3 156 16 2 91 131 P1 48
14 3 100 145 G1 153 16 3 92 133 N2 45
14 4 – – – 150 16 4 – – – 42
14 5 101 146 H2 147 16 5 93 134 N4 39
14 6 102 147 H3 144 16 6 94 135 N3 36
14 7 – – – 141 16 7 – – – 33
14 8 103 148 J4 138 16 8 95 136 M1 30
14 9 – – F1 135 16 9 – – M2 27
14 10 104 149 G2 132 16 10 96 137 M3 24
14 11 105 150 G3 129 16 11 97 138 M4 21
14 12 – 151 F2 126 16 12 98 139 L1 18
14 13 – – – 123 16 13 – – – 15
14 14 106 152 E2 120 16 14 – 140 L2 12
14 15 107 154 D2 117 16 15 – 142 L3 9
14 16 – – – 114 16 16 – – – 6
14 17 – 155 F4 111 16 17 – 143 J1 3
14 18 – – – 108 16 18 – – – 0

R

Pin Type TQ144 PQ208 BG352
I/O/GCK1 30 44 Y24
I/O/GCK2 32 46 AA24
I/O/GCK3 38 55 AD23
I/O/GTS1 5 7 E25
I/O/GTS2 6 9 F26
I/O/GTS3 2 3 E23
I/O/GTS4 3 5 E24
I/O/GSR 143 206 C23
TCK 67 98 AD6
TDI 63 94 AF6
TDO 122 176 D12
TMS 65 96 AE6
VCCINT 3.3V 8, 42, 84, 141 11, 59, 124, 153, 204 J23, V24, AF23, AC15, AF15,
AD11, AD5, Y3, T1, J3, G4,
D5, D10, B13, D17, C22, H24
VCCIO 2.5V/3.3 V 1, 37, 55, 73, 109, 127 1, 26, 53, 65, 79, 92, 105, A10, A17, B2, B25, D7, D13,
132, 157, 172, 181, 184 D19, G23, H4, K1, K26, N23,
P4, U1, U26, W23, Y4, AC8,
AC14, AC20, AE25, AF10,
AF17
GND 18, 29, 36, 47, 62, 72, 89, 90, 2, 13, 24, 27, 42, 52, 68, 81, A1, A2, A5, A8, A14, A19,
99, 108, 114, 123, 144 93, 104,1 08, 129, 130, 141, A22, A25, A26, B1, B26, C7,
156, 163, 177, 190, 207 C9, C13, C18, D24, E1, E26,
H1, H26, K4, N1, N24, P3,
P26, V23, W1, W4, W26,
AB1, AB4, AB26, AC9, AD10,
AD14, AD15, AD20, AE1,
AE26, AF1, AF2, AF5, AF8,
AF13, AF19, AF22, AF25,
AF26
No Connects A18, A23, A24, B4, B8, B10,
B23, C1, C2, C3, C4, C5, C8,
C11, C24, C25, D1, D3, D4,
D14, D16, D21, D23, D25, E3,
E4, F3, F23, G25, J2, J24,
J26, K2, L4, L23, P2, T3, T4,
– – T24, U25, V25, W3, W24, Y2,
AB3, AB23, AC2, AC4, AC6,
AC11, AC16, AC17, AC21,
AC23, AC24, AC25, AD16,
AD21, AD24, AD26, AE2,
AE4, AE10, AE15, AF3, AF4,
AF9, AF20

R

Number of Pins

-10 10 ns pin-to-pin delay TQ144 144-Pin Thin Quad Flat Pack (TQFP)
-6 6 ns pin-to-pin delay PQ208 208-Pin Plastic Quad Flat Pack (PQFP)
BG352 352-Pin Plastic Ball Grid Array (BGA)
Temperature Options
C= Commercial TA = 0oC to +70oC
5
Pins 144 208 352
TQFP PQFP BGA
Code TQ144 PQ208 BG352
-10 C, I C, I (C)
XC95288XL
-6 (C) (C) (C)
( ) Parenthesis indicate future product plans. Please contact Xilinx for up-to-date availability information.

R

0
R XC9500 In-System Programmable

CPLD Family
December 14, 1998 (Version 3.0) 0 1*
Features Family Overview

• High-performance The XC9500 CPLD family provides advanced in-system
- 5 ns pin-to-pin logic delays on all pins programming and test capabilities for high performance,
- fCNT to 125 MHz general purpose logic integration. All devices are in-system
• Large density range programmable for a minimum of 10,000 program/erase
- 36 to 288 macrocells with 800 to 6,400 usable gates cycles. Extensive IEEE 1149.1 (JTAG) boundary-scan sup-
• 5 V in-system programmable port is also included on all family members.
- Endurance of 10,000 program/erase cycles
As shown in Table 1, logic density of the XC9500 devices
- Program/erase over full commercial voltage and
ranges from 800 to over 6,400 usable gates with 36 to 288
temperature range
registers, respectively. Multiple package options and asso-
• Enhanced pin-locking architecture
ciated I/O capacity are shown in Table 2. The XC9500 fam-
• Flexible 36V18 Function Block
ily is fully pin-compatible allowing easy design migration
- 90 product terms drive any or all of 18 macrocells
across multiple density options in a given package foot-
within Function Block 5
print.
- Global and product term clocks, output enables, set
and reset signals The XC9500 architectural features address the require-
• Extensive IEEE Std 1149.1 boundary-scan (JTAG) ments of in-system programmability. Enhanced pin-locking
support capability avoids costly board rework. An expanded JTAG
• Programmable power reduction mode in each instruction set allows version control of programming pat-
macrocell terns and in-system debugging. In-system programming
• Slew rate control on individual outputs throughout the full device operating range and a minimum
• User programmable ground pin capability of 10,000 program/erase cycles provide worry-free recon-
• Extended pattern security features for design protection figurations and system field upgrades.
• High-drive 24 mA outputs Advanced system features include output slew rate control
• 3.3 V or 5 V I/O capability and user-programmable ground pins to help reduce system
• Advanced CMOS 5V FastFLASH technology noise. I/Os may be configured for 3.3 V or 5 V operation. All
• Supports parallel programming of multiple XC9500 outputs provide 24 mA drive.
devices
Architecture Description
Each XC9500 device is a subsystem consisting of multiple
Function Blocks (FBs) and I/O Blocks (IOBs) fully intercon-
nected by the FastCONNECT switch matrix. The IOB pro-
vides buffering for device inputs and outputs. Each FB
provides programmable logic capability with 36 inputs and
18 outputs. The FastCONNECT switch matrix connects all
FB outputs and input signals to the FB inputs. For each FB,
12 to 18 outputs (depending on package pin-count) and
associated output enable signals drive directly to the IOBs.
See Figure 1.

R
3
JTAG
JTAG Port In-System Programming Controller
Controller
36
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O
36
FastCONNECT Switch Matrix

I/O Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
36
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 36
1 Function
I/O/GSR 18 Block N
2 or 4 Macrocells
I/O/GTS 1 to 18
X5877
Figure 1: XC9500 Architecture

Note: Function Block outputs (indicated by the bold line) drive the I/O Blocks directly.
Table 1: XC9500 Device Family
XC9536 XC9572 XC95108 XC95144 XC95216 XC95288

Macrocells 36 72 108 144 216 288
Usable Gates 800 1,600 2,400 3,200 4,800 6,400
Registers 36 72 108 144 216 288
tPD (ns) 5 7.5 7.5 7.5 10 15
tSU (ns) 3.5 4.5 4.5 4.5 6.0 8.0
tCO (ns) 4.0 4.5 4.5 4.5 6.0 8.0
fCNT (MHz) 100 125 125 125 111.1 92.2
fSYSTEM (MHz) 100 83.3 83.3 83.3 66.7 56.6
Note: fCNT = Operating frequency for 16-bit counters
fSYSTEM = Internal operating frequency for general purpose system designs spanning multiple FBs.

R
Table 2: Available Packages and Device I/O Pins (not including dedicated JTAG pins)
XC9536 XC9572 XC95108 XC95144 XC95216 XC95288

44-Pin VQFP 34
44-Pin PLCC 34 34
48-Pin CSP 34
84-Pin PLCC 69 69
100-Pin TQFP 72 81 81
100-Pin PQFP 72 81 81
160-Pin PQFP 108 133 133
208-Pin HQFP 166 168
352-Pin BGA 166 192
Function Block
Each Function Block, as shown in Figure 2, is comprised of form 90 product terms. Any number of these product terms,
18 independent macrocells, each capable of implementing up to the 90 available, can be allocated to each macrocell
a combinatorial or registered function. The FB also by the product term allocator.
receives global clock, output enable, and set/reset signals. Each FB (except for the XC9536) supports local feedback
The FB generates 18 outputs that drive the FastCONNECT
paths that allow any number of FB outputs to drive into its 5
switch matrix. These 18 outputs and their corresponding own programmable AND-array without going outside the
output enable signals also drive the IOB.
FB. These paths are used for creating very fast counters
Logic within the FB is implemented using a sum-of-prod- and state machines where all state registers are within the
ucts representation. Thirty-six inputs provide 72 true and same FB.
complement signals into the programmable AND-array to
Macrocell 1
Programmable Product
AND-Array Term
Allocators 18 To FastCONNECT
From Switch Matrix
36
FastCONNECT
Switch Matrix 18 OUT
To I/O Blocks
18 PTOE
Macrocell 18
1 3
Global Global
Set/Reset Clocks X5878
Figure 2: XC9500 Function Block

R
Macrocell
Each XC9500 macrocell may be individually configured for term allocator associated with each macrocell selects how
a combinatorial or registered function. The macrocell and the five direct terms are used.
associated FB logic is shown in Figure 3.
The macrocell register can be configured as a D-type or
Five direct product terms from the AND-array are available T-type flip-flop, or it may be bypassed for combinatorial
for use as primary data inputs (to the OR and XOR gates) operation. Each register supports both asynchronous set
to implement combinatorial functions, or as control inputs and reset operations. During power-up, all user registers
including clock, set/reset, and output enable. The product are initialized to the user-defined preload state (default to 0
if unspecified).
54 Global Global
Set/Reset Clocks
Additional
Product
Terms
(from other
macrocells)
Product Term Set
0
To
S FastCONNECTII
Switch Matrix
D/T Q
CE
Product
Term
Product Term Clock Enable R
Allocator
Product Term Clock
Product Term Reset
OUT
To
Product Term OE PTOE I/O Blocks
Additional
Product
Terms
(from other
macrocells)
X5879_new
Figure 3: XC9500 Macrocell Within Function Block

R
All global control signals are available to each individual term clock. Both true and complement polarities of a GCK
macrocell, including clock, set/reset, and output enable sig- pin can be used within the device. A GSR input is also pro-
nals. As shown in Figure 4, the macrocell register clock vided to allow user registers to be set to a user-defined
originates from either of three global clocks or a product state.
Macrocell
Product Term Set
S
D/T
Product Term Clock
R
Product Term Reset
I/O/GSR
Global Set/Reset
I/O/GCK1
5
Global Clock 1
I/O/GCK2
Global Clock 2
Global Clock 3
I/O/GCK3
X5880
Figure 4: Macrocell Clock and Set/Reset Capability

R

The product term allocator controls how the five direct Note that the incremental delay affects only the product
product terms are assigned to each macrocell. For exam- terms in other macrocells. The timing of the direct product
ple, all five direct terms can drive the OR function as shown terms is not changed.
in Figure 5. Product Term
Allocator
Product Term
Allocator
Macrocell
Product Term
Logic
Product Term
Allocator
X5894
Figure 5: Macrocell Logic Using Direct Product Term
The product term allocator can re-assign other product

terms within the FB to increase the logic capacity of a mac-
Macrocell Logic
rocell beyond five direct terms. Any macrocell requiring With 15 P-Terms
additional product terms can access uncommitted product
terms in other macrocells within the FB. Up to 15 product
terms can be available to a single macrocell with only a
small incremental delay of tPTA, as shown in Figure 6.
Product Term
Allocator
X5895
Figure 6: Product Term Allocation With 15 Product

Terms

R
The product term allocator can re-assign product terms In this example, the incremental delay is only 2*tPTA. All 90
from any macrocell within the FB by combining partial sums product terms are available to any macrocell, with a maxi-
of products over several macrocells, as shown in Figure 7. mum incremental delay of 8*tPTA.
Product Term
Allocator
Macrocell Logic
With 2
Product Terms
Product Term
Allocator
Product Term
Allocator
Macrocell Logic
With 18
Product Terms
Product Term
Allocator
X5896
Figure 7: Product Term Allocation Over Several Macrocells

R
The internal logic of the product term allocator is shown in Figure 8.

From Upper To Upper
Macrocell Macrocell
Product Term
Allocator
Product Term Set
Global Set/Reset
S
D/T Q
Global Clocks
Product Term Clock R
Product Term Reset
Global Set/Reset
Product Term OE
From Lower To Lower

X5881
Macrocell Macrocell
Figure 8: Product Term Allocator Logic

R

The FastCONNECT switch matrix connects signals to the The FastCONNECT switch matrix is capable of combining
FB inputs, as shown in Figure 9. All IOB outputs (corre- multiple internal connections into a single wired-AND out-
sponding to user pin inputs) and all FB outputs drive the put before driving the destination FB. This provides addi-
FastCONNECT matrix. Any of these (up to a FB fan-in limit tional logic capability and increases the effective logic
of 36) may be selected, through user programming, to drive fan-in of the destination FB without any additional timing
each FB with a uniform delay. delay. This capability is available for internal connections
originating from FB outputs only. It is automatically invoked
by the development software where applicable.
FastCONNECT II
Switch Matrix
Function Block
I/O Block
(54) 18
5
D/T Q
I/O
Function Block
I/O Block
(54)
18
D/T Q
I/O
X5882_01
Figure 9: FastCONNECT Switch Matrix

R
I/O Block
The I/O Block (IOB) interfaces between the internal logic The output enable may be generated from one of four
and the device user I/O pins. Each IOB includes an input options: a product term signal from the macrocell, any of
buffer, output driver, output enable selection multiplexer, the global OE signals, always “1”, or always “0”. There are
and user programmable ground control. See Figure 10 for two global output enables for devices with up to 144 mac-
details. rocells, and four global output enables for devices with 180
or more macrocells. Both polarities of any of the global
The input buffer is compatible with standard 5 V CMOS, 5 V
3-state control (GTS) pins may be used within the device.
TTL and 3.3 V signal levels. The input buffer uses the internal
5 V voltage supply (VCCINT) to ensure that the input thresh-
olds are constant and do not vary with the VCCIO voltage.
To other
Macrocells
I/O Block
To FastCONNECT
Switch Matrix
Macrocell Bus-Hold
OUT I/O
(Inversion in
AND-array) 1 User-
Product Term OE PTOE Programmable
Ground
0
Slew Rate
Control
Available in XC95144XL
I/O/GTS3 Global OE 3 and XC95288XL
X5899_01
Figure 10: I/O Block and Output Enable Capability

R
Each output has independent slew rate control. Output voltage supply. Figure 12 shows how the XC9500 device
edge rates may be slowed down to reduce system noise can be used in 5 V only and mixed 3.3 V/5 V systems.
(with an additional time delay of tSLEW) through program-
ming. See Figure 11. Pin-Locking Capability
Each IOB provides user programmable ground pin capabil- The capability to lock the user defined pin assignments dur-
ity. This allows device I/O pins to be configured as addi- ing design changes depends on the ability of the architec-
tional ground pins. By tying strategically located ture to adapt to unexpected changes. The XC9500 devices
programmable ground pins to the external ground connec- have architectural features that enhance the ability to
tion, system noise generated from large numbers of simul- accept design changes while maintaining the same pinout.
taneous switching outputs may be reduced. The XC9500 architecture provides maximum routing within
A control pull-up resistor (typically 10K ohms) is attached to the FastCONNECT switch matrix, and incorporates a flexi-
each device I/O pin to prevent them from floating when the ble Function Block that allows block-wide allocation of
device is not in normal user operation. This resistor is available product terms. This provides a high level of confi-
active during device programming mode and system dence of maintaining both input and output pin assign-
power-up. It is also activated for an erased device. The ments for unexpected design changes.
resistor is deactivated during normal operation. For extensive design changes requiring higher logic capac-
The output driver is capable of supplying 24 mA output ity than is available in the initially chosen device, the new
drive. All output drivers in the device may be configured for design may be able to fit into a larger pin-compatible device
either 5 V TTL levels or 3.3 V levels by connecting the using the same pin assignments. The same board may be
device output voltage supply (VCCIO) to a 5 V or 3.3 V used with a higher density device without the expense of
5
board rework.
Output Output
Voltage Voltage
Standard
Slew-Rated Limited
Slew-Rated Limited
tSLEW tSLEW
1.5 V 1.5 V
Standard
Time Time
0 0
(a) (b) X5900
Figure 11: Output Slew-Rate Control For (a) Rising and (b) Falling Outputs
5 V CMOS 5V 5 V CMOS 5V 3.3 V

5V 5V
VCCINT VCCIO VCCINT VCCIO
0V 0V
5 V TTL or 5 V TTL 5 V TTL or 3.3 V
3.6 V ~4V 3.6 V 3.3 V

XC9500 XC9500
IN CPLD OUT IN CPLD OUT
0V 0V 0V 0V
3.3 V or 3.3 V or
3.3 V 3.3 V
GND GND
0V 0V
(a) (b) X5901
Figure 12: XC9500 Devices in (a) 5 V Systems and (b) Mixed 3.3 V/5 V Systems

R
In-System Programming The TMS and TCK pins have dedicated pull-up resistors as
specified by the IEEE 1149.1 standard.
XC9500 devices are programmed in-system via a standard
4-pin JTAG protocol, as shown in Figure 13. In-system pro- Boundary Scan Description Language (BSDL) files for the
gramming offers quick and efficient design iterations and XC9500 are included in the development system and are
eliminates package handling. The Xilinx development sys- available on the Xilinx FTP site.
tem provides the programming data sequence using a Xil-
inx download cable, a third-party JTAG development Design Security
system, JTAG-compatible board tester, or a simple micro- XC9500 devices incorporate advanced data security fea-
processor interface that emulates the JTAG instruction tures which fully protect the programming data against
sequence. unauthorized reading or inadvertent device erasure/repro-
All I/Os are 3-stated and pulled high by the IOB resistors gramming. Table 3 shows the four different security set-
during in-system programming. If a particular signal must tings available.
remain low during this time, then a pulldown resistor may The read security bits can be set by the user to prevent the
be added to the pin. internal programming pattern from being read or copied.
When set, they also inhibit further program operations but
External Programming allow device erase. Erasing the entire device is the only
XC9500 devices can also be programmed by the Xilinx way to reset the read security bit.
HW130 device programmer as well as third-party program- The write security bits provide added protection against
mers. This provides the added flexibility of using pre-pro- accidental device erasure or reprogramming when the
grammed devices during manufacturing, with an in-system JTAG pins are subject to noise, such as during system
programmable option for future enhancements. power-up. Once set, the write-protection may be deacti-
vated when the device needs to be reprogrammed with a
Endurance valid pattern.
All XC9500 CPLDs provide a minimum endurance level of Table 3: Data Security Options
10,000 in-system program/erase cycles. Each device
meets all functional, performance, and data retention spec- Read Security
ifications within this endurance limit. Default Set

IEEE 1149.1 Boundary-Scan (JTAG)
Write Security
Default
Program/Erase Allowed Program Inhibited/Erase Allowed
XC9500 devices fully support IEEE 1149.1 boundary-scan
(JTAG). EXTEST, SAMPLE/PRELOAD, BYPASS, USER- Read Allowed Read Inhibited
CODE, INTEST, IDCODE, and HIGHZ instructions are sup- Set
Program/Erase Inhibited Program/Erase Inhibited
ported in each device. For ISP operations, five additional
instructions are added; the ISPEN, FERASE, FPGM, X5905
FVFY, and ISPEX instructions are fully compliant exten-

sions of the 1149.1 instruction set.

R
V CC
GND
(a) (b) X5902
Figure 13: In-System Programming Operation (a) Solder Device to PCB and (b) Program Using Download Cable
5
Low Power Mode Timing Model
All XC9500 devices offer a low-power mode for individual The uniformity of the XC9500 architecture allows a simpli-
macrocells or across all macrocells. This feature allows the fied timing model for the entire device. The basic timing
device power to be significantly reduced. model, shown in Figure 14, is valid for macrocell functions
that use the direct product terms only, with standard power
Each individual macrocell may be programmed in
setting, and standard slew rate setting. Table 4 shows how
low-power mode by the user. Performance-critical parts of
each of the key timing parameters is affected by the prod-
the application can remain in standard power mode, while
uct term allocator (if needed), low-power setting, and
other parts of the application may be programmed for
slew-limited setting.
low-power operation to reduce the overall power dissipa-
tion. Macrocells programmed for low-power mode incur The product term allocation time depends on the logic span
additional delay (tLP) in pin-to-pin combinatorial delay as of the macrocell function, which is defined as one less than
well as register setup time. Product term clock to output the maximum number of allocators in the product term
and product term output enable delays are unaffected by path. If only direct product terms are used, then the logic
the macrocell power-setting. span is 0. The example in Figure 6 shows that up to 15
product terms are available with a span of 1. In the case of
Figure 7, the 18 product term function has a span of 2.
Detailed timing information may be derived from the full
timing model shown in Figure 15. The values and explana-
tions for each parameter are given in the individual device
data sheets.

R
tSU
Combinatorial Combinatorial
Logic Logic D/T Q
tCO
Propagation Delay = tPD Setup Time = tSU Clock to Out Time = tCO
(a) (b)
tPSU
Combinatorial
Logic D/T Q
Combinatorial
P-Term Clock Logic D/T Q
Path
tPCO
Setup Time = tPSU Clock to Out Time = tPCO Internal System Cycle Time = tSYSTEM
(c) (d)
All resources within FB using local Feedback

Combinatorial
Logic
Combinatorial
Logic D/T Q
Combinatorial
Logic
Internal Cycle Time = tCNT

(e) Propagation Delay = tPD + tFBK
Setup With Feedback
Figure 14: Basic Timing Model Time
Combinatorial
(f)
Logic
Pin Feedback
tF
tLF
tLOGILP S*tPTA tSLEW
tPDI
tIN tLOGI
D/T Q tOUT
tSUI tCOI
tHI
tPTCK
tAOI
tEN
tGCK > tRAI
tPTSR SR
tGSR
tPTTS
tGTS
Figure 15: Detailed Timing Model
Power-Up Characteristics with the IOB pull-up resistors (~ 10K ohms) enabled, as
shown in Table 5. When the supply voltage reaches a safe
The XC9500 devices are well behaved under all operating level, all user registers become initialized (typically within
conditions. During power-up each XC9500 device employs 100 µs for 9536 - 95144, 200 µs for 95216 and 300 µs for
internal circuitry which keeps the device in the quiescent 95288), and the device is immediately available for opera-
state until the VCCINT supply voltage is at a safe level tion, as shown in Figure 16.
(approximately 3.8 V). During this time, all device pins and
JTAG pins are disabled and all device outputs are disabled

R
If the device is in the erased state (before any user pattern FastFLASH Technology
is programmed), the device outputs remain disabled with
the IOB pull-up resistors enabled. The JTAG pins are An advanced CMOS Flash process is used to fabricate all
enabled to allow the device to be programmed at any time. XC9500 devices. Specifically developed for Xilinx in-system pro-
grammable CPLDs, the FastFLASH process provides high
If the device is programmed, the device inputs and outputs performance logic capability, fast programming times, and
take on their configured states for normal operation. The endurance of 10,000 program/erase cycles.
JTAG pins are enabled to allow device erasure or bound-
VCCINT
ary-scan tests at any time.
Development System Support

3.8 V
The XC9500 CPLD family is fully supported by the develop- (Typ)
ment systems available from Xilinx and the Xilinx Alliance
Program vendors.
The designer can create the design using ABEL, schemat-
ics, equations, VHDL, or Verilog in a variety of software 0V
front-end tools. The development system can be used to No Quiescent Quiescent No
Power User Operation
State State Power
implement the design and generate a JEDEC bitmap which
can be used to program the XC9500 device. Each develop- Initialization of User Registers X5904
ment system includes JTAG download software that can be

used to program the devices via the standard JTAG inter- Figure 16: Device Behavior During Power-up
5
face and a download cable.
Table 4: Timing Model Parameters
Product Term Macrocell Output Slew-Limited

Description Parameter
Allocator1 Low-Power Setting Setting
Propagation Delay tPD + tPTA * S + tLP + tSLEW
Global Clock Setup Time tSU + tPTA * S + tLP –
Global Clock-to-output tCO – – + tSLEW
Product Term Clock Setup tPSU + tPTA * S + tLP –
Time
Product Term Clock-to-output tPCO – – + tSLEW
Internal System Cycle Period tSYSTEM + tPTA * S + tLP –
Note: 1. S = the logic span of the function, as defined in the text.
Table 5: XC9500 Device Characteristics
Device Quiescent Erased Device Valid User

Circuitry State Operation Operation
IOB Pull-up Resistors Enabled Enabled Disabled
Device Outputs Disabled Disabled As Configured
Device Inputs and Clocks Disabled Disabled As Configured
Function Block Disabled Disabled As Configured
JTAG Controller Disabled Enabled Enabled

R
Revision History
Version Date Revision
3.0 12/14/98 Revised datasheet to reflect new AC characteristics and Internal Timing Parmeters.

9
 XC9536 In-System Programmable

CPLD
November 30, 1998 (Version 5.0) 0 5* Product Specification
Features Power Management

• 5 ns pin-to-pin logic delays on all pins Power dissipation can be reduced in the XC9536 by config-
• fCNT to 100 MHz uring macrocells to standard or low-power modes of opera-
• 36 macrocells with 800 usable gates tion. Unused macrocells are turned off to minimize power
• Up to 34 user I/O pins dissipation.
• 5 V in-system programmable (ISP) Operating current for each design can be approximated for
- Endurance of 10,000 program/erase cycles specific operating conditions using the following equation:
temperature range ICC (mA) =
• Enhanced pin-locking architecture MCHP (1.7) + MCLP (0.9) + MC (0.006 mA/MHz) f
Where:
within Function Block MCHP = Macrocells in high-performance mode
5
- Global and product term clocks, output enables, set MCLP = Macrocells in low-power mode
and reset signals
• Extensive IEEE Std 1149.1 boundary-scan (JTAG)
support f = Clock frequency (MHz)
• Programmable power reduction mode in each
Figure 1 shows a typical calculation for the XC9536 device.
macrocell
• User programmable ground pin capability
• Extended pattern security features for design protection
• High-drive 24 mA outputs (83)
• 3.3 V or 5 V I/O capability ance
erform
High P
• Advanced CMOS 5V FastFLASH technology
Typical ICC (mA)
• Supports parallel programming of more than one (50)

(50)
XC9500 concurrently ower
Low P
• Available in 44-pin PLCC, 44-pin VQFP, and 48-pin
CSP packages (30)
Description
The XC9536 is a high-performance CPLD providing
0 50 100
advanced in-system programming and test capabilities for Clock Frequency (MHz)
X5920
general purpose logic integration. It is comprised of two
36V18 Function Blocks, providing 800 usable gates with Figure 1: Typical ICC vs. Frequency For XC9536
propagation delays of 5 ns. See Figure 2 for the architec-
ture overview.

R
3
JTAG
Controller
36
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O
36

I/O Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
I/O
I/O
I/O
3
I/O/GCK
1
I/O/GSR
2
I/O/GTS
X5919

Note: Function Block outputs (indicated by the bold line) drive the I/O Blocks directly

R

Symbol Parameter Value Units
VIN DC input voltage relative to GND -0.5 to VCC + 0.5 V
VTS Voltage applied to 3-state output with respect to GND -0.5 to VCC + 0.5 V
TSTG Storage temperature -65 to +150 °C
TSOL Max soldering temperature (10 s @ 1/16 in = 1.5 mm) +260 °C
Warning: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are
Recommended Operating Conditions is not implied. Exposure to Absolute Maximum Rating conditions for extended periods
of time may affect device reliability.
Recommended Operating Conditions 1

VCCINT Supply voltage for internal logic and input buffer 4.75 5.25 V
(4.5) (5.5)
VCCIO Supply voltage for output drivers for 5 V operation 4.75 (4.5) 5.25 (5.5) V
Supply voltage for output drivers for 3.3 V operation 3.0 3.6 V 5
VIH High-level input voltage 2.0 VCCINT +0.5 V
Note 1. Numbers in parenthesis are for industrial-temperature range versions.
Endurance Characteristics
NPE Program/Erase Cycles 10,000 - Cycles

R

VOH Output high voltage for 5 V operation IOH = -4.0 mA 2.4 V
VCC = Min
Output high voltage for 3.3 V operation IOH = -3.2 mA 2.4 V
VCC = Min
VOL Output low voltage for 5 V operation IOL = 24 mA 0.5 V
VCC = Min
Output low voltage for 3.3 V operation IOL = 10 mA 0.4 V
VCC = Min
IIL Input leakage current VCC = Max ±10.0 µA
VIN = GND or VCC
IIH I/O high-Z leakage current VCC = Max ±10.0 µA
VIN = GND or VCC
f = 1.0 MHz
ICC Operating Supply Current VI = GND, No load mA
30 (Typ)
AC Characteristics
XC9536-5 XC9536-6 XC9536-7 XC9536-10 XC9536-15
Min Max Min Max Min Max Min Max Min Max
tPD I/O to output valid 5.0 6.0 7.5 10.0 ns 15.0
tSU I/O setup time before GCK 3.5 3.5 4.5 6.0 8.0ns
tH I/O hold time after GCK 0.0 0.0 0.0 0.0 0.0ns
tCO GCK to output valid 4.0 4.0 4.5 6.0 8.0 ns
fCNT1 16-bit counter frequency 100.0 100.0 83.3 66.7 55.6 MHz
fSYSTEM 2 Multiple FB internal operating frequency 100.0 100.0 83.3 66.7 55.6 MHz
tPSU I/O setup time before p-term clock input 0.5 0.5 0.5 2.0 4.0 ns
tPH I/O hold time after p-term clock input 3.0 3.0 4.0 4.0 4.0 ns
tPCO P-term clock to output valid 7.0 7.0 8.5 10.0 12.0 ns
tOE GTS to output valid 5.0 5.0 5.5 6.0 11.0 ns
tOD GTS to output disable 5.0 5.0 5.5 6.0 11.0 ns
tPOE Product term OE to output enabled 9.0 9.0 9.5 10.0 14.0 ns
tPOD Product term OE to output disabled 9.0 9.0 9.5 10.0 14.0 ns
tWLH GCK pulse width (High or Low) 4.0 4.0 4.0 4.5 5.5 ns
Note: 1. fCNT is the fastest 16-bit counter frequency available.
fCNT is also the Export Control Maximum flip-flop toggle rate, fTOG.
2. fSYSTEM is the internal operating frequency for general purpose system designs spanning multiple FBs.

R
VTEST
R1
3.3 V 3.3 V 260 Ω 360 Ω 35 pF
R2 CL
X5906

XC9536-5 XC9536-6 XC9536-7 XC9536-10 XC9536-15
Min Max Min Max Min Max Min Max Min Max
Buffer Delays
tIN Input buffer delay 1.5 1.5 2.5 3.5 4.5 ns 5
tGCK GCK buffer delay 1.5 1.5 1.5 2.5 3.0 ns
tGSR GSR buffer delay 4.0 4.0 4.5 6.0 7.5 ns
tGTS GTS buffer delay 5.0 5.0 5.5 6.0 11.0 ns
tOUT Output buffer delay 2.0 2.0 2.5 3.0 4.5 ns
tEN Output buffer enable/disable delay 0.0 0.0 0.0 0.0 0.0 ns
tPTCK Product term clock delay 3.0 3.0 3.0 3.0 2.5 ns
tPTSR Product term set/reset delay 1.0 1.0 2.0 2.5 3.0 ns
tPTTS Product term 3-state delay 5.5 5.5 4.5 3.5 5.0 ns
Internal Register and Combinatorial delays
tPDI Combinatorial logic propagation delay 0.5 1.5 0.5 1.0 3.0 ns
tSUI Register setup time 2.5 2.5 1.5 2.5 3.5 ns
tHI Register hold time 1.0 1.0 3.0 3.5 4.5 ns
tCOI Register clock to output valid time 0.5 0.5 0.5 0.5 0.5 ns
tAOI Register async. S/R to output delay 6.0 6.0 6.5 7.0 8.0 ns
tRAI Register async. S/R recovery before clock 5.0 5.0 7.5 10.0 10.0 ns
tLOGI Internal logic delay 1.0 1.0 2.0 2.5 3.0 ns
tLOGILP Internal low power logic delay 9.0 9.0 10.0 11.0 11.5 ns
Feedback Delays
tF FastCONNECT matrix feeback delay 6.0 6.0 8.0 9.5 11.0 ns
Time Adders
tPTA3 Incremental Product Term Allocator delay 0.8 0.8 1.0 1.0 1.0 ns
tSLEW Slew-rate limited delay 3.5 3.5 4.0 4.5 5.0 ns
Note: 3. tPTA is multiplied by the span of the function as defined in the family data sheet.

R
XC9536 I/O Pins

Macrocell PC44 VQ44 CS48 Notes Macrocell PC44 VQ44 CS48 Notes
1 1 2 40 D6 105 2 1 1 39 D7 51
1 2 3 41 C7 102 2 2 44 38 E5 48
1 3 5 43 B7 99 [1] 2 3 42 36 E6 45 [1]
1 4 4 42 C6 96 2 4 43 37 E7 42
1 5 6 44 B6 93 [1] 2 5 40 34 F6 39 [1]
1 6 8 2 A6 90 2 6 39 33 G7 36 [1]
1 7 7 1 A7 87 [1] 2 7 38 32 G6 33
1 8 9 3 C5 84 2 8 37 31 F5 30
1 9 11 5 B5 81 2 9 36 30 G5 27
1 10 12 6 A4 78 2 10 35 29 F4 24
1 11 13 7 B4 75 2 11 34 28 G4 21
1 12 14 8 A3 72 2 12 33 27 E3 18
1 13 18 12 B2 69 2 13 29 23 F2 15
1 14 19 13 B1 66 2 14 28 22 G1 12
1 15 20 14 C2 63 2 15 27 21 F1 9
1 16 22 16 C3 60 2 16 26 20 E2 6
1 17 24 18 D2 57 2 17 25 19 E1 3
1 18 – – - 54 2 18 - - - 0
Note: [1] Global control pin Note: [1] Global control pin
XC9536 Global, JTAG and Power Pins

Pin Type PC44 VQ44 CS48
I/O/GCK1 5 43 B7
I/O/GCK2 6 44 B6
I/O/GCK3 7 1 A7
I/O/GTS1 42 36 E6
I/O/GTS2 40 34 F6
I/O/GSR 39 33 G7
TCK 17 11 A1
TDI 15 9 B3
TDO 30 24 G2
TMS 16 10 A2
VCCINT 5 V 21,41 15,35 C1,F7
VCCIO 3.3 V/5 V 32 26 G3
GND 23,10,31 17,4,25 A5, D1, F3
No Connects — — C4, D3, D4, E4

R
XC9536 -5 PC 44 C

Speed Number of Pins
Package Type

-15 15 ns pin-to-pin delay PC44 44-Pin Plastic Leaded Chip Carrier (PLCC)
-10 10 ns pin-to-pin delay VQ44 44-Pin Thin Quad Pack (VQFP)
-7 7.5 ns pin-to-pin delay CS48 48-Pin Chip Scale Package (CSP)
Temperature Options
C = Commercial (0°C to +70°C)
I = Industrial (–40°C to +85°C)
Pins 44 48
Plastic Plastic Plastic
Type
PLCC VQFP CSP
Code PC44 VQ44 CS48
–15 C,I C,I -
–10 C,I C,I C
XC9536 –7 C,I C,I C
–6 C C -
–5 C C C
C = Commercial (0°C to +70°C), I = Industrial (–40°C to +85°C)
Revision Control
Date Reason
6/3/98 Revise datasheet to reflect new CSP package pinouts & ordering code.
11/2/98 Revise datasheet to reflect new AC characteristics and Internal Timing Parameters.
11/30/98 Revise datasheet to remove PCI compliancy statement and remove tLF.

R

0

CPLD

• 7.5 ns pin-to-pin logic delays on all pins Power dissipation can be reduced in the XC9572 by config-
• fCNT to 125 MHz uring macrocells to standard or low-power modes of opera-
• 72 macrocells with 1,600 usable gates tion. Unused macrocells are turned off to minimize power
• Up to 72 user I/O pins dissipation.
Where:
5
and reset signals
Figure 1 shows a typical calculation for the XC9572 device.
macrocell
• Extended pattern security features for design protection 200
• High-drive 24 mA outputs
• 3.3 V or 5 V I/O capability
• Advanced CMOS 5V FastFLASH technology (160)
ance
• Supports parallel programming of more than one erform
High P
XC9500 concurrently
(125)
Typical Icc (ma)
• Available in 44-pin PLCC, 84-pin PLCC, 100-pin PQFP

and 100-pin TQFP packages 100 (100)
o wer
Low P
Description
The XC9572 is a high-performance CPLD providing (65)
advanced in-system programming and test capabilities for
general purpose logic integration. It is comprised of four
propagation delays of 7.5 ns. See Figure 2 for the architec-
ture overview. 0 50 100
Figure 1: Typical ICC vs. Frequency for XC9572

R
3
JTAG
Controller
36
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O
36

I/O Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
36
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 36
1 Function
I/O/GSR 18 Block 4
2 Macrocells
I/O/GTS 1 to 18
X5921

R

Recommended Operation Conditions 1

(4.5) (5.5)
VIL Low-level input voltage 0 0.80 V 5
Note: 1. Numbers in parenthesis are for industrial temperature range versions.

R

VCC = Min
VCC = Min
VCC = Min
VCC = Min
VIN = GND or VCC
VIN = GND or VCC
f = 1.0 MHz
ICC Operating Supply Current VI = GND, No load 65 (Typ) ma
AC Characteristics
XC9572-7 XC9572-10 XC9572-15
Min Max Min Max Min Max
fCNT1 16-bit counter frequency 125.0 111.1 95.2 MHz
fSYSTEM 2 Multiple FB internal operating frequency 83.3 66.7 55.6 MHz
tPCO P-term clock to output valid 8.5 10.0 12.0 ns
Note: 1. fCNT is the fastest 16-bit counter frequency available, using the local feedback when applicable.
VTEST
R1
3.3 V 3.3 V 260 Ω 360 Ω 35 pF
R2 CL
X5906

R

XC9572-7 XC9572-10 XC9572-15
Buffer Delays
5
tRAI Register async. S/R recovery before clock 7.5 10.0 10.0 ns
Feedback Delays
tF FastCONNECT matrix feedback delay 8.0 9.5 11.0 ns
tLF Function Block local feeback delay 4.0 3.5 3.5 ns
Time Adders
tPTA3 Incremental Product Term Allocator delay 1.0 1.0 1.0 ns

R
XC9572 I/O Pins

Function PC PC PQ TQ BScan Function PC PC PQ TQ BScan
Macrocell Notes Macrocell Notes
Block 44 84 100 100 Order Block 44 84 100 100 Order
1 1 – 4 18 16 213 3 1 – 25 43 41 105
1 2 1 1 15 13 210 3 2 11 17 34 32 102
1 3 – 6 20 18 207 3 3 – 31 51 49 99
1 4 – 7 22 20 204 3 4 – 32 52 50 96
1 5 2 2 16 14 201 3 5 12 19 37 35 93
1 6 3 3 17 15 198 3 6 – 34 55 53 90
1 7 – 11 27 25 195 3 7 – 35 56 54 87
1 8 4 5 19 17 192 3 8 13 21 39 37 84
1 9 5 9 24 22 189 [1] 3 9 14 26 44 42 81
1 10 – 13 30 28 186 3 10 – 40 62 60 78
1 11 6 10 25 23 183 [1] 3 11 18 33 54 52 75
1 12 – 18 35 33 180 3 12 – 41 63 61 72
1 13 – 20 38 36 177 3 13 – 43 65 63 69
1 14 7 12 29 27 174 [1] 3 14 19 36 57 55 66
1 15 8 14 31 29 171 3 15 20 37 58 56 63
1 16 – 23 41 39 168 3 16 – 45 67 65 60
1 17 9 15 32 30 165 3 17 22 39 60 58 57
1 18 – 24 42 40 162 3 18 – – 61 59 54
2 1 – 63 89 87 159 4 1 – 46 68 66 51
2 2 35 69 96 94 156 4 2 24 44 66 64 48
2 3 – 67 93 91 153 4 3 – 51 73 71 45
2 4 – 68 95 93 150 4 4 – 52 74 72 42
2 5 36 70 97 95 147 4 5 25 47 69 67 39
2 6 37 71 98 96 144 4 6 – 54 78 76 36
2 7 – 76 5 3 141 [2] 4 7 – 55 79 77 33
2 8 38 72 99 97 138 4 8 26 48 70 68 30
2 9 39 74 1 99 135 [1] 4 9 27 50 72 70 27
2 10 – 75 3 1 132 4 10 – 57 83 81 24
2 11 40 77 6 4 129 [1] 4 11 28 53 76 74 21
2 12 – 79 8 6 126 4 12 – 58 84 82 18
2 13 – 80 10 8 123 4 13 – 61 87 85 15
2 14 42 81 11 9 120 [3] 4 14 29 56 80 78 12
2 15 43 83 13 11 117 4 15 33 65 91 89 9
2 16 – 82 12 10 114 4 16 – 62 88 86 6
2 17 44 84 14 12 111 4 17 34 66 92 90 3
2 18 – – 94 92 108 4 18 – – 81 79 0

[2] Global control pin GTS1 for PC84, PQ100, and TQ100
[3] Global control pin GTS1 for PC44

R

Pin Type PC44 PC84 PQ100 TQ100
I/O/GCK1 5 9 24 22
I/O/GCK2 6 10 25 23
I/O/GCK3 7 12 29 27
I/O/GTS1 42 76 5 3
I/O/GTS2 40 77 6 4
I/O/GSR 39 74 1 99
TCK 17 30 50 48
TDI 15 28 47 45
TDO 30 59 85 83
TMS 16 29 49 47
VCCINT 5 V 21,41 38,73,78 7,59,100 5,57,98
VCCIO 3.3 V/5 V 32 22,64 28,40,53,90 26,38,51,88
8,16,27,42, 2,23,33,46,64,71, 100,21,31,44,62,69,
GND 10,23,31 49,60 77,86 75, 84
4,9,21,26,36,45,48, 2,7,19,24,34,43,46,
5
No Connects – — 75, 82 73, 80

R
XC9572 -7 PQ 100 C

Package Type

-7 7.5 ns pin-to-pin delay PQ100 100-Pin Plastic Quad Flat Pack (PQFP)
TQ100 100-Pin Very Thin Quad Flat Pack (TQFP)
Temperature Options
C Commercial0°C to +70°C
I Industrial–40°C to +85°C
Pins 44 84 100
Plastic Plastic Plastic Plastic
Type
PLCC PLCC PQFP TQFP
Code PC44 PC84 PQ100 TQ100
–15 C(I) C(I) C(I) C(I)
XC9572 –10 C(I) C(I) C(I) C(I)
–7 C C C C
C = Commercial = 0° to +70°C I = Industrial = –40° to +85°C
Revision Control
Date Revision
11/30/98 V 2.2 - Update AC Charateristics and Internal Parameters

0

CPLD

• 7.5 ns pin-to-pin logic delays on all pins Power dissipation can be reduced in the XC95108 by con-
• fCNT to 125 MHz figuring macrocells to standard or low-power modes of
• 108 macrocells with 2400 usable gates operation. Unused macrocells are turned off to minimize
• Up to 108 user I/O pins power dissipation.
Where:
5
and reset signals
Figure 1 shows a typical calculation for the XC95108
macrocell
device.
• Extended pattern security features for design protection 300
• 3.3 V or 5 V I/O capability (250)
rformance
• Advanced CMOS 5V FastFLASH technology H igh Pe
• Supports parallel programming of more than one 200
Typical ICC (mA)
XC9500 concurrently (180)

(170)
• Available in 84-pin PLCC, 100-pin PQFP, 100-pin TQFP
ower
and 160-pin PQFP packages Low P
Description 100

general purpose logic integration. It is comprised of six
36V18 Function Blocks, providing 2,400 usable gates with 0 50 100
propagation delays of 7.5 ns. See Figure 2 for the architec- Clock Frequency (MHz)
X5898
ture overview.
Figure 1: Typical ICC vs. Frequency for XC95108

R
3
JTAG
Controller
36
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O
36

I/O Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
36
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 36
1 Function
I/O/GSR 18 Block 4
2 Macrocells
I/O/GTS 1 to 18
36
Function
18 Block 5
Macrocells
1 to 18
36
Function
18 Block 6
Macrocells
1 to 18
X5897

R


(4.5) (5.5)
Note: 1. Numbers in parenthesis are for industrial-temperature range versions.

R

VCC = Min
Output high voltage for 3.3 V operation IOH = -3.2 mA V
VCC = Min 2.4
VCC = Min
VCC = Min
VIN = GND or VCC
VIN = GND or VCC
f = 1.0 MHz
AC Characteristics
XC95108-7 XC95108-10 XC95108-15 XC95108-20
Min Max Min Max Min Max Min Max
tPD I/O to output valid 7.5 10.0 15.0 20.0 ns
tSU I/O setup time before GCK 4.5 6.0 8.0 10.0 ns
tH I/O hold time after GCK 0.0 0.0 0.0 0.0 ns
tCO GCK to output valid 4.5 6.0 8.0 10.0 ns
fCNT1 16-bit counter frequency 125.0 111.1 95.2 83.3 MHz
fSYSTEM 2 Multiple FB internal operating frequency 83.3 66.7 55.6 50.0 MHz
tPSU I/O setup time before p-term clock input 0.5 2.0 4.0 4.0 ns
tPH I/O hold time after p-term clock input 4.0 4.0 4.0 6.0 ns
tPCO P-term clock to output valid 8.5 10.0 12.0 16.0 ns
tOE GTS to output valid 5.5 6.0 11.0 16.0 ns
tOD GTS to output disable 5.5 6.0 11.0 16.0 ns
tPOE Product term OE to output enabled 9.5 10.0 14.0 18.0 ns
tPOD Product term OE to output disabled 9.5 10.0 14.0 18.0 ns
tWLH GCK pulse width (High or Low) 4.0 4.5 5.5 5.5 ns

R
VTEST
R1
3.3 V 3.3 V 260 Ω 360 Ω 35 pF
R2 CL
X5906

XC95108-7 XC95108-10 XC95108-15 XC95108-20
5
Buffer Delays
tIN Input buffer delay 2.5 3.5 4.5 6.5 ns
tGCK GCK buffer delay 1.5 2.5 3.0 3.0 ns
tGSR GSR buffer delay 4.5 6.0 7.5 9.5 ns
tGTS GTS buffer delay 5.5 6.0 11.0 16.0 ns
tOUT Output buffer delay 2.5 3.0 4.5 6.5 ns
tEN Output buffer enable/disable delay 0.0 0.0 0.0 0.0 ns
tPTCK Product term clock delay 3.0 3.0 2.5 2.5 ns
tPTSR Product term set/reset delay 2.0 2.5 3.0 3.0 ns
tPTTS Product term 3-state delay 4.5 3.5 5.0 5.0 ns
tPDI Combinatorial logic propagation delay 0.5 1.0 3.0 4.0 ns
tSUI Register setup time 1.5 2.5 3.5 3.5 ns
tHI Register hold time 3.0 3.5 4.5 6.5 ns
tCOI Register clock to output valid time 0.5 0.5 0.5 0.5 ns
tAOI Register async. S/R to output delay 6.5 7.0 8.0 8.0 ns
tRAI Register async. S/R recovery before clock 7.5 10.0 10.0 10.0 ns
tLOGI Internal logic delay 2.0 2.5 3.0 3.0 ns
tLOGILP Internal low power logic delay 10.0 11.0 11.5 11.5 ns
Feedback Delays
tF FastCONNECT matrix feedback delay 8.0 9.5 11.0 13.0 ns
tLF Function Block local feeback delay 4.0 3.5 3.5 5.0 ns
Time Adders
tPTA3 Incremental Product Term Allocator delay 1.0 1.0 1.0 1.5 ns
tSLEW Slew-rate limited delay 4.0 4.5 5.0 5.5 ns

R
XC95108 I/O Pins

Macrocell PC84 PQ100 TQ100 PQ160 Notes Macrocell PC84 PQ100 TQ100 PQ160 Notes
1 1 – – – 25 321 3 1 – – – 45 213
1 2 1 15 13 21 318 3 2 14 31 29 47 210
1 3 2 16 14 22 315 3 3 15 32 30 49 207
1 4 – 21 19 29 312 3 4 – 36 34 57 204
1 5 3 17 15 23 309 3 5 17 34 32 54 201
1 6 4 18 16 24 306 3 6 18 35 33 56 198
1 7 – – – 27 303 3 7 – – – 50 195
1 8 5 19 17 26 300 3 8 19 37 35 58 192
1 9 6 20 18 28 297 3 9 20 38 36 59 189
1 10 – 26 24 36 294 3 10 – 45 43 69 186
1 11 7 22 20 30 291 3 11 21 39 37 60 183
1 12 9 24 22 33 288 [1] 3 12 23 41 39 62 180
1 13 – – – 34 285 3 13 – – – 52 177
1 14 10 25 23 35 282 [1] 3 14 24 42 40 63 174
1 15 11 27 25 37 279 3 15 25 43 41 64 171
1 16 12 29 27 42 276 [1] 3 16 26 44 42 68 168
1 17 13 30 28 44 273 3 17 31 51 49 77 165
1 18 – – – 43 270 3 18 – – – 74 162
2 1 – – – 158 267 4 1 – – – 123 159
2 2 71 98 96 154 264 4 2 57 83 81 134 156
2 3 72 99 97 156 261 4 3 58 84 82 135 153
2 4 – 4 2 4 258 4 4 – 82 80 133 150
2 5 74 1 99 159 255 [1] 4 5 61 87 85 138 147
2 6 75 3 1 2 252 4 6 62 88 86 139 144
2 7 – – – 9 249 4 7 – – – 128 141
2 8 76 5 3 6 246 [1] 4 8 63 89 87 140 138
2 9 77 6 4 8 243 [1] 4 9 65 91 89 142 135
2 10 – 9 7 12 240 4 10 – – – 147 132
2 11 79 8 6 11 237 4 11 66 92 90 143 129
2 12 80 10 8 13 234 4 12 67 93 91 144 126
2 13 – – – 14 231 4 13 – – – 153 123
2 14 81 11 9 15 228 4 14 68 95 93 146 120
2 15 82 12 10 17 225 4 15 69 96 94 148 117
2 16 83 13 11 18 222 4 16 – 94 92 145 114
2 17 84 14 12 19 219 4 17 70 97 95 152 111
2 18 – – – 16 216 4 18 – – – 155 108

R
XC95108 I/O Pins (continued)

Macrocell PC84 PQ100 TQ100 PQ160 Notes Macrocell PC84 PQ100 TQ100 PQ160 Notes
5 1 – – – 76 105 6 1 – – – 91 51
5 2 32 52 50 79 102 6 2 45 67 65 103 48
5 3 33 54 52 82 99 6 3 46 68 66 104 45
5 4 – 48 46 72 96 6 4 – 75 73 116 42
5 5 34 55 53 86 93 6 5 47 69 67 106 39
5 6 35 56 54 88 90 6 6 48 70 68 108 36
5 7 – – – 78 87 6 7 – – – 105 33
5 8 36 57 55 90 84 6 8 50 72 70 111 30
5 9 37 58 56 92 81 6 9 51 73 71 113 27
5 10 – – – 84 78 6 10 – – – 107 24
5 11 39 60 58 95 75 6 11 52 74 72 115 21
5 12 40 62 60 97 72 6 12 53 76 74 117 18
5 13 – – – 87 69 6 13 – – – 112 15
5 14 41 63 61 98 66 6 14 54 78 76 122 12
5 15 43 65 63 101 63 6 15 55 79 77 124 9
5 16 – 61 59 96 60 6 16 – 81 79 129 6 5
5 17 44 66 64 102 57 6 17 56 80 78 126 3
5 18 – – – 89 54 6 18 – – – 114 0

Pin Type PC84 PQ100 TQ100 PQ160
I/O/GCK1 9 24 22 33
I/O/GCK2 10 25 23 35
I/O/GCK3 12 29 27 42
I/O/GTS1 76 5 3 6
I/O/GTS2 77 6 4 8
I/O/GSR 74 1 99 159
TCK 30 50 48 75
TDI 28 47 45 71
TDO 59 85 83 136
TMS 29 49 47 73
VCCINT 5 V 38,73,78 7,59,100 5,57,98 10,46,94,157
VCCIO 3.3 V/5 V 22,64 28,40,53,90 26,38,51,88 1,41,61,81,121,141
GND 8,16,27,42,49,60 2,23,33,46,64,71,77,86 100,21,31,44,62,69,75,84 20,31,40,51,70,80,99
GND – – – 100,110,120,127,137
GND – – – 160
No connects – – – 3,5,7,32,38,39,48,53,55,6
5,66,67,83,85,93,109,
118,119,125,130,131,
132,149,150,151

R
XC95108 -7 PQ 160 C

Number of Pins
Speed Package Type

- 20 20 ns pin-to-pin delay PC84 84-Pin Plastic Leaded Chip Carrier (PLCC)
-10 10 ns pin-to-pin delay TQ100 100-Pin Very Thin Quad Flat Pack (TQFP)
Temperature Options
C Commercial 0°C to +70°C
I Industrial –40°C to +85°C
Pins 84 100 160
Plastic Plastic Plastic Plastic
Type
PLCC PQFP TQFP PQFP
Code PC84 PQ100 TQ100 PQ160
–20 C(I) C(I) C(I) C(I)
–15 C(I) C(I) C(I) C(I)
XC95108
–10 C(I) C(I) C(I) C(I)
–7 C(I) C(I) C(I) C(I)
Revision Control
Date Revision
11/30/98 V 2.1 - Update AC Characteristics and Internal Parameters

0

CPLD
Features Operating current for each design can be approximated for

specific operating conditions using the following equation:
• 7.5 ns pin-to-pin logic delays on all pins
• fCNT to 111 MHz ICC (mA) =
• 144 macrocells with 3,200 usable gates MCHP (1.7) + MCLP (0.9) + MC (0.006 mA/MHz) f
• Up to 133 user I/O pins Where:
• 5 V in-system programmable
- Endurance of 10,000 program/erase cycles MCHP = Macrocells in high-performance mode
- Program/erase over full commercial voltage and MCLP = Macrocells in low-power mode
temperature range
• Enhanced pin-locking architecture
• Flexible 36V18 Function Block f = Clock frequency (MHz)
- 90 product terms drive any or all of 18 macrocells Figure 1 shows a typical calculation for the XC95144
within Function Block device. 5
- Global and product term clocks, output enables, set
and reset signals
600
support
(480)
macrocell
nce
• Slew rate control on individual outputs orma
Perf
400 High
Typical ICC (mA)

• Extended pattern security features for design protection (320)
• High-drive 24 mA outputs (300)

Pow
er
Low
200
(160)
• Supports parallel programming of more than one
XC9500 concurrently
• Available in 100-pin PQFP, 100-pin TQFP, and 160-pin
PQFP packages
0 50 100
X5898B
Description
The XC95144 is a high-performance CPLD providing Figure 1: Typical Icc vs. Frequency for XC95144
general purpose logic integration. It is comprised of eight
propagation delays of 7.5 ns. See Figure 2 for the architec-
ture overview.
Power Management
Power dissipation can be reduced in the XC95144 by con-
figuring macrocells to standard or low-power modes of
operation. Unused macrocells are turned off to minimize
power dissipation.

R
3
JTAG
Controller
36
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O
36

I/O Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
36
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 36
1 Function
I/O/GSR 18 Block 4
2 Macrocells
I/O/GTS 1 to 18
36
Function
18 Block 8
Macrocells
1 to 18
X5922

R


(4.5) (5.5)

R

VCC = Min
VCC = Min
VCC = Min
VCC = Min
VIN = GND or VCC
VIN = GND or VCC
f = 1.0 MHz
AC Characteristics
XC95144-7 XC95144-10 XC95144-15

R
VTEST
R1
3.3 V 3.3 V 260 Ω 360 Ω 35 pF
R2 CL
X5906

XC95144-7 XC95144-10 XC95144-15
5
Buffer Delays
Feedback Delays
tLF Function Block local feedback delay 4.0 3.5 3.5 ns
Time Adders

R
XC95144 I/O Pins

Function TQ PQ PQ BScan Function TQ PQ PQ BScan
Block 100 100 160 Order Block 100 100 160 Order
1 1 – – 25 429 3 1 – – 43 321
1 2 11 13 18 426 3 2 23 25 35 318 [1]
1 3 12 14 19 423 3 3 – – 45 315
1 4 – – 27 420 3 4 – – 48 312
1 5 13 15 21 417 3 5 24 26 36 309
1 6 14 16 22 414 3 6 25 27 37 306
1 7 – – 32 411 3 7 – – 50 303
1 8 15 17 23 408 3 8 27 29 42 300 [1]
1 9 16 18 24 405 3 9 28 30 44 297
1 10 – – 34 402 3 10 – – 52 294
1 11 17 19 26 399 3 11 29 31 47 291
1 12 18 20 28 396 3 12 30 32 49 288
1 13 – – 38 393 3 13 – – 53 285
1 14 19 21 29 390 3 14 32 34 54 282
1 15 20 22 30 387 3 15 33 35 56 279
1 16 – – 39 384 3 16 – – 55 276
1 17 22 24 33 381 [1] 3 17 34 36 57 273
1 18 – – – 378 3 18 – – – 270
2 1 – – 158 375 4 1 – – 132 267
2 2 99 1 159 372 [1] 4 2 87 89 140 264
2 3 – – 3 369 4 3 – – 147 261
2 4 – – 5 366 4 4 – – 149 258
2 5 1 3 2 363 [1] 4 5 89 91 142 255
2 6 2 4 4 360 [1] 4 6 90 92 143 252
2 7 – – 7 357 4 7 – – 150 249
2 8 3 5 6 354 [1] 4 8 91 93 144 246
2 9 4 6 8 351 [1] 4 9 92 94 145 243
2 10 – – 9 348 4 10 – – 151 240
2 11 6 8 11 345 4 11 93 95 146 237
2 12 7 9 12 342 4 12 94 96 148 234
2 13 – – 14 339 4 13 – – 153 231
2 14 8 10 13 336 4 14 95 97 152 228
2 15 9 11 15 333 4 15 96 98 154 225
2 16 – – 16 330 4 16 – – 155 222
2 17 10 12 17 327 4 17 97 99 156 219
2 18 – – – 324 4 18 – – – 216
Notes: [1] Global control pin.

Macrocell outputs to package pins subject to change, contact factory for latest information. Power, GND, JTAG and Global
Signals are fixed.

R

Function TQ PQ PQ BScan Function TQ PQ PQ BScan
Block 100 100 160 Order Block 100 100 160 Order
5 1 – – 65 213 7 1 – – – 105
5 2 35 37 58 210 7 2 50 52 79 102
5 3 – – 66 207 7 3 – – 84 99
5 4 – – 67 204 7 4 – – 85 96
5 5 36 38 59 201 7 5 52 54 82 93
5 6 37 39 60 198 7 6 53 55 86 90
5 7 – – 74 195 7 7 – – 87 87
5 8 39 41 62 192 7 8 54 56 88 84
5 9 40 42 63 189 7 9 55 57 90 81
5 10 – – 76 186 7 10 – – 89 78
5 11 41 43 64 183 7 11 56 58 92 75
5 12 42 44 68 180 7 12 58 60 95 72
5 13 – – 78 177 7 13 – – 91 69
5 14 43 45 69 174 7 14 59 61 96 66
5 15 46 48 72 171 7 15 60 62 97 63
5 16 – – 83 168 7 16 – – 93 60 5
5 17 49 51 77 165 7 17 61 63 98 57
5 18 – – – 162 7 18 – – – 54
6 1 – – – 159 8 1 – – – 51
6 2 74 76 117 156 8 2 63 65 101 48
6 3 – – 119 153 8 3 – – 105 45
6 4 – – 123 150 8 4 – – 107 42
6 5 76 78 122 147 8 5 64 66 102 39
6 6 77 79 124 144 8 6 65 67 103 36
6 7 – – 125 141 8 7 – – 109 33
6 8 78 80 126 138 8 8 66 68 104 30
6 9 79 81 129 135 8 9 67 69 106 27
6 10 – – 128 132 8 10 – – 112 24
6 11 80 82 133 129 8 11 68 70 108 21
6 12 81 83 134 126 8 12 70 72 111 18
6 13 – – 130 123 8 13 – – 114 15
6 14 82 84 135 120 8 14 71 73 113 12
6 15 85 87 138 117 8 15 72 74 115 9
6 16 – – 131 114 8 16 – – 118 6
6 17 86 88 139 111 8 17 73 75 116 3
6 18 – – – 108 8 18 – – – 0

R

Pin Type TQ100 PQ100 PQ160
I/O/GCK1 22 24 33
I/O/GCK2 23 25 35
I/O/GCK3 27 29 42
I/O/GTS1 3 5 6
I/O/GTS2 4 6 8
I/O/GTS3 1 3 2
I/O/GTS4 2 4 4
I/O/GSR 99 1 159
TCK 48 50 75
TDI 45 47 71
TDO 83 85 136
TMS 47 49 73
VCCINT 5 V 5, 57, 98 7, 59, 100 10, 46, 94, 157
VCCIO 3.3 V/5 V 26, 38, 51, 88 28, 40, 53, 90 1, 41, 61, 81, 121, 141
GND 100, 21, 31, 44, 62, 69, 2, 23, 33, 46, 64, 71, 20, 31, 40, 51, 70, 80,
75, 84 77, 86 99, 100, 110, 120, 127,
137, 160
No Connects – – –

R
XC95144 -7 PQ 160 C

Number of Pins
Speed Package Type

-10 10 ns pin-to-pin delay TQ100 100-Pin Very Thin Quad Flat Pack (TQFP)
Temperature Options
C Commercial 0°C to 70°C
I Industrial –40°C to 85°C
5
Pins 100 160
PQFP TQFP PQFP
Code PQ100 TQ100 PQ160
–15 C,I C,I C,I
XC95144 –10 C,I C,I C,I
–7 C C C
C = Commercial = 0°C to +70°C I = Industrial = –40°C to +G85°C
Revision Control
Date Revision
11/30/98 V 3.1 - Update AC characteristics and internal parameters.

0

CPLD

• 10 ns pin-to-pin logic delays on all pins Power dissipation can be reduced in the XC95216 by con-
• 216 macrocells with 4800 usable gates operation. Unused macrocells are turned off to minimize
• 5 V in-system programmable Operating current for each design can be approximated for
• Flexible 36V18 Function Block Where:
5
and reset signals
macrocell
device.
600
nce
erforma
XC9500 concurrently High P
• Available in 160-pin PQFP, 352-pin BGA, and 208-pin 400
Typical ICC (mA)
(360)
HQFP packages (340)
Low Power
Description
The XC95216 is a high-performance CPLD providing 200
general purpose logic integration. It is comprised of twelve
X5918
Figure 1: Typical ICC vs. Frequency For XC95216

R
3
JTAG
Controller
36
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O
36

I/O Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
36
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 36
1 Function
I/O/GSR 18 Block 4
2 Macrocells
I/O/GTS 1 to 18
36
Function
18 Block 12
Macrocells
1 to 18
X5917

Function Block outputs (indicated by the bold line) drive the I/O Blocks directly

R

Recommended Operating Conditions 1

(4.5) (5.5)

R

VCC = Min
VCC = Min
VCC = Min
VCC = Min
VIN = GND or VCC
VIN = GND or VCC
f = 1.0 MHz
ICC Operating Supply Current VI = GND, No load 200 (typ) ma
AC Characteristics
XC95216-10 XC95216-15 XC95216-20

R
VTEST
R1
3.3 V 3.3 V 260 Ω 360 Ω 35 pF
R2 CL
X5906

5
XC95216-10 XC95216-15 XC95216-20
Buffer Delays
Feedback Delays
tLF Function Block local feeback delay 3.5 3.5 5.0 ns
Time Adders

R
XC95216 I/O Pins

Macrocell PQ160 HQ208 BG352 Notes Macrocell PQ160 HQ208 BG352 Notes
1 1 – – – 645 3 1 – – – 537
1 2 18 22 M25 642 3 2 32 43 AA26 534
1 3 19 23 M26 639 3 3 33 44 Y24 531 [1]
1 4 – 28 N26 636 3 4 – 39 U23 528
1 5 21 25 N25 633 3 5 34 45 AB25 525
1 6 22 30 P23 630 3 6 35 46 AA24 522 [1]
1 7 – – – 627 3 7 – – – 519
1 8 23 31 P24 624 3 8 36 47 Y23 516
1 9 24 32 R26 621 3 9 37 49 AA23 513
1 10 – 12 G26 618 3 10 – 67 AD18 510
1 11 25 33 R24 615 3 11 38 50 AB24 507
1 12 26 34 T26 612 3 12 39 51 AD25 504
1 13 – – – 609 3 13 – – – 501
1 14 27 35 T25 606 3 14 42 55 AD23 498 [1]
1 15 28 36 T23 603 3 15 43 56 AF24 495
1 16 29 37 V26 600 3 16 – 80 AE12 492
1 17 30 38 U24 597 3 17 44 57 AE23 489
1 18 – – – 594 3 18 – – – 486
2 1 – – – 591 4 1 – – – 483
2 2 6 7 E25 588 [1] 4 2 152 198 D18 480
2 3 7 8 G24 585 4 3 153 199 A21 477
2 4 – 29 P25 582 4 4 – 196 B19 474
2 5 8 9 F26 579 [1] 4 5 154 200 B20 471
2 6 9 10 H23 576 4 6 155 201 C20 468
2 7 – – – 573 4 7 – – – 465
2 8 11 15 K23 570 4 8 156 202 B22 462
2 9 12 16 K24 567 4 9 158 205 B24 459
2 10 – - – 564 4 10 – - – 456
2 11 13 17 J25 561 4 11 159 206 C23 453 [1]
2 12 14 18 L24 558 4 12 2 3 E23 450 [1]
2 13 – – – 555 4 13 – – – 447
2 14 15 19 K25 552 4 14 3 4 C26 444
2 15 16 20 L26 549 4 15 4 5 E24 441 [1]
2 16 – 14 H25 546 4 16 – 203 D20 438
2 17 17 21 M24 543 4 17 5 6 F24 435
2 18 – – – 540 4 18 – – – 432

R

5 1 – – – 429 7 1 – – – 321
5 2 45 58 AE22 426 7 2 58 76 AE13 318
5 3 47 60 AE21 423 7 3 59 77 AC13 315
5 4 – 41 W25 420 7 4 – 54 AE24 312
5 5 48 61 AF21 417 7 5 60 78 AD13 309
5 6 49 63 AD19 414 7 6 62 82 AD12 306
5 7 – – – 411 7 7 – – – 303
5 8 50 64 AE20 408 7 8 63 83 AC12 300
5 9 52 70 AF18 405 7 9 64 84 AF11 297
5 10 – 109 AD1 402 7 10 – 91 AD8 294
5 11 53 71 AE17 399 7 11 65 85 AE11 291
5 12 54 72 AE16 396 7 12 66 86 AE9 288
5 13 – – – 393 7 13 – – – 285
5 14 55 73 AF16 390 7 14 67 87 AD9 282
5 15 56 74 AE14 387 7 15 68 88 AC10 279
5 16 – 40 Y26 384 7 16 – 48 AC26 276 5
5 17 57 75 AF14 381 7 17 69 89 AF7 273
5 18 – – – 378 7 18 – – – 270
6 1 – – – 375 8 1 – – – 267
6 2 140 180 A12 372 8 2 126 162 B5 264
6 3 142 182 A13 369 8 3 128 164 B6 261
6 4 – 208 D22 366 8 4 – 143 J1 258
6 5 143 185 C14 363 8 5 129 166 D8 255
6 6 144 186 A15 360 8 6 130 167 B7 252
6 7 – – – 357 8 7 – – – 249
6 8 145 187 B15 354 8 8 131 170 C10 246
6 9 146 188 C15 351 8 9 132 171 B9 243
6 10 – 183 B14 348 8 10 – 195 A20 240
6 11 147 191 A16 345 8 11 133 173 A9 237
6 12 148 192 C16 342 8 12 134 174 D11 234
6 13 – – – 339 8 13 – – – 231
6 14 149 193 C17 336 8 14 135 175 B11 228
6 15 150 194 B18 333 8 15 138 178 C12 225
6 16 – 169 D9 330 8 16 – 189 D15 222
6 17 151 197 C19 327 8 17 139 179 B12 219
6 18 – – – 324 8 18 – – – 216

R

9 1 – – – 213 11 1 – – – 105
9 2 72 95 AD7 210 11 2 87 115 Y1 102
9 3 74 97 AE5 207 11 3 88 116 V4 99
9 4 – 101 AD4 204 11 4 – 119 U4 96
9 5 76 99 AC7 201 11 5 89 117 V3 93
9 6 77 100 AE3 198 11 6 90 118 W2 90
9 7 – – – 195 11 7 – – – 87
9 8 78 102 AC5 192 11 8 91 121 V2 84
9 9 79 103 AD3 189 11 9 92 122 U2 81
9 10 – 90 AE8 186 11 10 – 107 AC3 78
9 11 82 110 AA4 183 11 11 93 123 T2 75
9 12 83 111 AB2 180 11 12 95 125 R4 72
9 13 – – – 177 11 13 – – – 69
9 14 84 112 AC1 174 11 14 96 126 R3 66
9 15 85 113 AA2 171 11 15 97 127 R2 63
9 16 – 62 AC19 168 11 16 – 120 U3 60
9 17 86 114 AA1 165 11 17 98 128 R1 57
9 18 – – – 162 11 18 – – – 54
10 1 – – – 159 12 1 – – – 51
10 2 113 147 H3 156 12 2 101 131 P1 48
10 3 114 148 J4 153 12 3 102 133 N2 45
10 4 – 144 K3 150 12 4 – 106 AD2 42
10 5 115 149 G2 147 12 5 103 134 N4 39
10 6 116 150 G3 144 12 6 104 135 N3 36
10 7 – – – 141 12 7 – – – 33
10 8 117 152 E2 138 12 8 105 136 M1 30
10 9 118 154 D2 135 12 9 106 137 M3 27
10 10 – 168 A7 132 12 10 – 151 F2 24
10 11 119 155 F4 129 12 11 107 138 M4 21
10 12 122 158 B3 126 12 12 108 139 L1 18
10 13 – – – 123 12 13 – – – 15
10 14 123 159 A3 120 12 14 109 140 L2 12
10 15 124 160 D6 117 12 15 111 145 G1 9
10 16 – 165 A6 114 12 16 – 142 L3 6
10 17 125 161 C6 111 12 17 112 146 H2 3
10 18 – – – 108 12 18 – – – 0

R

Pin Type PQ160 HQ208 BG352
I/O/GCK1 33 44 Y24
I/O/GCK2 35 46 AA24
I/O/GCK3 42 55 AD23
I/O/GTS1 6 7 E25
I/O/GTS2 8 9 F26
I/O/GTS3 2 3 E23
I/O/GTS4 4 5 E24
I/O/GSR 159 206 C23
TCK 75 98 AD6
TDI 71 94 AF6
TDO 136 176 D12
TMS 73 96 AE6
VCCINT 5 V 10,46,94,157 11, 59, 124, 153, 204 H24, AF23, T1, G4, C22
VCCIO 3.3 V/5 V 1,41,61,81,121,141 1, 26, 53, 65, 79, 92, 105, 132, A10, A17, B2, B25, D7, D13,
157, 172, 181, 184 D19, G23, H4, K1, K26, N23, P4,
U1, U26, W23, Y4, AC8, AC14,
AC20, AE25, AF10, AF17 5
GND 20, 31, 40, 51, 70, 80, 99, 100, 2, 13, 24, 27, 42, 52, 66, 68, 69, A1, A2, A5, A8, A14, A19, A22,
110, 120, 127, 137, 160 81, 93, 104, 108, 129, 130, 141, A25, A26, B1, B26, C7, E1, E26,
156, 163, 177, 190, 207 H1, H26, N1, P3, P26, V23, W1,
W26, AB1, AB4, AB26, AC9,
AC17, AE1, AE26, AF1, AF2,
AF5, AF8, AF13, AF19, AF20,
AF22, AF25, AF26
No Connects – – A4, A11, A18, A23, A24, B4, B8,
B10, B13, B16, B17, B21, B23,
C1, C2, C3, C4, C5, C8, C9, C11,
C13, C18, C21, C24, C25, D1,
D3, D4, D5, D10, D14, D16, D17,
D21, D23, D24, D25, D26, E3,
E4, F1, F3, F23, F25, G25, J2,
J3, J23, J24, J26, K2, K4, L4,
L23, L25, M2, M23, N24, P2,
R23, R25, T3, T4, T24, U25, V1,
V24, V25, W3, W4, W24, Y2, Y3,
Y25, AA3, AA25, AB3, AB23,
AC2, AC4, AC6, AC11, AC15,
AC16, AC18, AC21, AC22,
AC23, AC24, AC25, AD5, AD10,
AD11, AD14, AD15, AD16,
AD17, AD20, AD21, AD22,
AD24, AD26, AE2, AE4, AE7,
AE10, AE15, AE18, AE19, AF3,
AF4, AF9, AF12, AF15

R
XC95216 -10 HQ 208 C

Number of Pins
Speed Package Type

- 20 20 ns pin-to-pin delay PQ160 160-Pin Plastic Quad Flat Pack (PQFP)
-15 15 ns pin-to-pin delay HQ208 208-Pin Heat Sink Quad Flat Pack (HQFP)
-10 10 ns pin-to-pin delay BG352 352-Pin Ball Grid Array (BGA)
Temperature Options
Pins 160 208 352
Plastic Power Plastic
Type
PQFP QFP BGA
Code PQ160 HQ208 BG352
–20 C(I) C(I) C(I)
XC95216 –15 C(I) C(I) C(I)
–10 C(I) C(I) C(I)
C = Commercial = 0°C to +70°C I = Industrial = –40°C to +85°C
Revision Control
Date Revision
11/30/98 Update AC Characteristics and Internal Parameters

0

CPLD

• 15 ns pin-to-pin logic delays on all pins Power dissipation can be reduced in the XC95288 by con-
• 288 macrocells with 6,400 usable gates operation. Unused macrocells are turned off to minimize
• 5 V in-system programmable Operating current for each design can be approximated for
Where:
5
and reset signals
macrocell
device.
• High-drive 24 mA outputs 900
ance
XC9500 concurrently erform
High P
600
Typical ICC (mA)
• Available in 352-pin BGA and 208-pin HQFP packages

(500) (500)
Description Low Power

300
general purpose logic integration. It is comprised of sixteen
X7131
Figure 1: Typical ICC vs. Frequency For XC95288

R
3
JTAG
Controller
36
Function
I/O 18 Block 1
Macrocells
I/O 1 to 18
I/O
36

I/O Function
18 Block 2
Macrocells
I/O 1 to 18
Blocks
I/O
36
Function
I/O
18 Block 3
I/O Macrocells
1 to 18
I/O
3
I/O/GCK 36
1 Function
I/O/GSR 18 Block 4
2 Macrocells
I/O/GTS 1 to 18
36
Function
18 Block 16
Macrocells
1 to 18
X5924


R


(4.5) (5.5)

R

VCC = Min
VCC = Min
VCC = Min
VCC = Min
VIN = GND or VCC
VIN = GND or VCC
CIN I/O capacitance VIN = GND ±10.0 pF
f = 1.0 MHz
AC Characteristics
XC95288-15 XC95288-20
Min Max Min Max
tPD I/O to output valid 15.0 20.0 ns
tSU I/O setup time before GCK 8.0 10.0 ns
tH I/O hold time after GCK 0.0 0.0 ns
tCO GCK to output valid 8.0 10.0 ns
fCNT1 16-bit counter frequency 95.2 83.3 MHz
fSYSTEM 2 Multiple FB internal operating frequency 55.6 50.0 MHz
tPSU I/O setup time before p-term clock input 4.0 4.0 ns
tPH I/O hold time after p-term clock input 4.0 6.0 ns
tPCO P-term clock to output valid 12.0 16.0 ns
tOE GTS to output valid 11.0 16.0 ns
tOD GTS to output disable 11.0 16.0 ns
tPOE Product term OE to output enabled 14.0 18.0 ns
tPOD Product term OE to output disabled 14.0 18.0 ns
tWLH GCK pulse width (High or Low) 5.5 5.5 ns

R
VTEST
R1
3.3 V 3.3 V 260 Ω 360 Ω 35 pF
R2 CL
X5906

XC95288-15 XC95288-20
Min Max Min Max 5
Buffer Delays
tIN Input buffer delay 4.5 6.5 ns
tGCK GCK buffer delay 3.0 3.0 ns
tGSR GSR buffer delay 7.5 9.5 ns
tGTS GTS buffer delay 11.0 16.0 ns
tOUT Output buffer delay 4.5 6.5 ns
tEN Output buffer enable/disable delay 0.0 0.0 ns
tPTCK Product term clock delay 2.5 2.5 ns
tPTSR Product term set/reset delay 3.0 3.0 ns
tPTTS Product term 3-state delay 5.0 5.0 ns
tPDI Combinatorial logic propagation delay 3.0 4.0 ns
tSUI Register setup time 3.5 3.5 ns
tHI Register hold time 4.5 6.5 ns
tCOI Register clock to output valid time 0.5 0.5 ns
tAOI Register async. S/R to output delay 8.0 8.0 ns
tRAI Register async. S/R recovery before clock 10.0 10.0 ns
tLOGI Internal logic delay 3.0 3.0 ns
tLOGILP Internal low power logic delay 11.5 11.5 ns
Feedback Delays
tF FastCONNECT matrix feedback delay 11.0 13.0 ns
tLF Function Block local feeback delay 3.5 5.0 ns
Time Adders
tPTA3 Incremental Product Term Allocator delay 1.0 1.5 ns
tSLEW Slew-rate limited delay 5.0 5.5 ns

R
XC95288 I/O Pins

Macrocell HQ208 BG352 Notes Macrocell HQ208 BG352 Notes
1 1 – – 861 3 1 – – 753
1 2 28 N26 858 3 2 38 U24 750
1 3 29 P25 855 3 3 39 U23 747
1 4 – – 852 3 4 – – 744
1 5 30 P23 849 3 5 40 Y26 741
1 6 31 P24 846 3 6 41 W25 738
1 7 – – 843 3 7 – – 735
1 8 32 R26 840 3 8 43 AA26 732
1 9 – R25 837 3 9 – Y25 729
1 10 33 R24 834 3 10 44 Y24 726 [1]
1 11 – R23 831 3 11 – AA25 723
1 12 34 T26 828 3 12 45 AB25 720
1 13 – – 825 3 13 – – 717
1 14 35 T25 822 3 14 46 AA24 714 [1]
1 15 36 T23 819 3 15 47 Y23 711
1 16 – – 816 3 16 – – 708
1 17 37 V26 813 3 17 48 AC26 705
1 18 – – 810 3 18 – – 702
2 1 – – 807 4 1 – – 699
2 2 15 K23 804 4 2 3 E23 696 [1]
2 3 16 K24 801 4 3 4 C26 693
2 4 – – 798 4 4 – – 690
2 5 17 J25 795 4 5 5 E24 687 [1]
2 6 18 L24 792 4 6 6 F24 684
2 7 – – 789 4 7 – – 681
2 8 19 K25 786 4 8 7 E25 678 [1]
2 9 – L25 783 4 9 – D26 675
2 10 20 L26 780 4 10 8 G24 672
2 11 – M23 777 4 11 – F25 669
2 12 21 M24 774 4 12 9 F26 666 [1]
2 13 – – 771 4 13 – – 663
2 14 22 M25 768 4 14 10 H23 660
2 15 23 M26 765 4 15 12 G26 657
2 16 – – 762 4 16 – – 654
2 17 25 N25 759 4 17 14 H25 651
2 18 – – 756 4 18 – – 648

Macrocell outputs to package pins subject to change, contact factory for latest information. Power, GND, JTAG and Global
Signals are fixed.

R

5 1 – – 645 7 1 – – 537
5 2 49 AA23 642 7 2 62 AC19 534
5 3 50 AB24 639 7 3 63 AD19 531
5 4 – – 636 7 4 – – 528
5 5 51 AD25 633 7 5 64 AE20 525
5 6 54 AE24 630 7 6 66 AC18 522
5 7 – – 627 7 7 – – 519
5 8 55 AD23 624 [1] 7 8 67 AD18 516
5 9 – AC22 621 7 9 – AE19 513
5 10 56 AF24 618 7 10 69 AD17 510
5 11 – AD22 615 7 11 – AE18 507
5 12 57 AE23 612 7 12 70 AF18 504
5 13 – – 609 7 13 – – 501
5 14 58 AE22 606 7 14 71 AE17 498
5 15 60 AE21 603 7 15 72 AE16 495
5 16 – – 600 7 16 – – 492 5
5 17 61 AF21 597 7 17 73 AF16 489
5 18 – – 594 7 18 – – 486
6 1 – – 591 8 1 – – 483
6 2 197 C19 588 8 2 186 A15 480
6 3 198 D18 585 8 3 187 B15 477
6 4 – – 582 8 4 – – 474
6 5 199 A21 579 8 5 188 C15 471
6 6 200 B20 576 8 6 189 D15 468
6 7 – – 573 8 7 – – 465
6 8 201 C20 570 8 8 191 A16 462
6 9 – B21 567 8 9 – B16 459
6 10 202 B22 564 8 10 192 C16 456
6 11 – C21 561 8 11 – B17 453
6 12 203 D20 558 8 12 193 C17 450
6 13 – – 555 8 13 – – 447
6 14 205 B24 552 8 14 194 B18 444
6 15 206 C23 549 [1] 8 15 195 A20 441
6 16 – – 546 8 16 – – 438
6 17 208 D22 543 8 17 196 B19 435
6 18 – – 540 8 18 – – 432
Note: [1] Global control pin

R

9 1 – – 429 11 1 – – 321
9 2 74 AE14 426 11 2 87 AD9 318
9 3 75 AF14 423 11 3 88 AC10 315
9 4 – – 420 11 4 – – 312
9 5 76 AE13 417 11 5 89 AF7 309
9 6 77 AC13 414 11 6 90 AE8 306
9 7 – – 411 11 7 – – 303
9 8 78 AD13 408 11 8 91 AD8 300
9 9 – AF12 405 11 9 – AE7 297
9 10 80 AE12 402 11 10 95 AD7 294
9 11 82 AD12 399 11 11 97 AE5 291
9 12 83 AC12 396 11 12 99 AC7 288
9 13 – – 393 11 13 – – 285
9 14 84 AF11 390 11 14 100 AE3 282
9 15 85 AE11 387 11 15 101 AD4 279
9 16 – – 384 11 16 – – 276
9 17 86 AE9 381 11 17 102 AC5 273
9 18 – – 378 11 18 – – 270
10 1 – – 375 12 1 – – 267
10 2 170 C10 372 12 2 158 B3 264
10 3 171 B9 369 12 3 159 A3 261
10 4 – – 366 12 4 – – 258
10 5 173 A9 363 12 5 160 D6 255
10 6 174 D11 360 12 6 161 C6 252
10 7 – – 357 12 7 – – 249
10 8 175 B11 354 12 8 162 B5 246
10 9 – A11 351 12 9 – A4 243
10 10 178 C12 348 12 10 164 B6 240
10 11 179 B12 345 12 11 165 A6 237
10 12 180 A12 342 12 12 166 D8 234
10 13 – – 339 12 13 – – 231
10 14 182 A13 336 12 14 167 B7 228
10 15 183 B14 333 12 15 168 A7 225
10 16 – – 330 12 16 – – 222
10 17 185 C14 327 12 17 169 D9 219
10 18 – – 324 12 18 – – 216

R

13 1 – – 213 15 1 – – 105
13 2 103 AD3 210 15 2 117 V3 102
13 3 106 AD2 207 15 3 118 W2 99
13 4 – – 204 15 4 – – 96
13 5 107 AC3 201 15 5 119 U4 93
13 6 109 AD1 198 15 6 120 U3 90
13 7 – – 195 15 7 – – 87
13 8 110 AA4 192 15 8 121 V2 84
13 9 – AA3 189 15 9 – V1 81
13 10 111 AB2 186 15 10 122 U2 78
13 11 112 AC1 183 15 11 123 T2 75
13 12 113 AA2 180 15 12 125 R4 72
13 13 – – 177 15 13 – – 69
13 14 114 AA1 174 15 14 126 R3 66
13 15 115 Y1 171 15 15 127 R2 63
13 16 – – 168 15 16 – – 60 5
13 17 116 V4 165 15 17 128 R1 57
13 18 – – 162 15 18 – – 54
14 1 – – 159 16 1 – – 51
14 2 144 K3 156 16 2 131 P1 48
14 3 145 G1 153 16 3 133 N2 45
14 4 – – 150 16 4 – – 42
14 5 146 H2 147 16 5 134 N4 39
14 6 147 H3 144 16 6 135 N3 36
14 7 – – 141 16 7 – – 33
14 8 148 J4 138 16 8 136 M1 30
14 9 – F1 135 16 9 – M2 27
14 10 149 G2 132 16 10 137 M3 24
14 11 150 G3 129 16 11 138 M4 21
14 12 151 F2 126 16 12 139 L1 18
14 13 – – 123 16 13 – – 15
14 14 152 E2 120 16 14 140 L2 12
14 15 154 D2 117 16 15 142 L3 9
14 16 – – 114 16 16 – – 6
14 17 155 F4 111 16 17 143 J1 3
14 18 – – 108 16 18 – – 0

R

Pin Type HQ208 BG352
I/O/GCK1 44 Y24
I/O/GCK2 46 AA24
I/O/GCK3 55 AD23
I/O/GTS1 7 E25
I/O/GTS2 9 F26
I/O/GTS3 3 E23
I/O/GTS4 5 E24
I/O/GSR 206 C23
TCK 98 AD6
TDI 94 AF6
TDO 176 D12
TMS 96 AE6
VCCINT 5 V 11, 59, 124, 153, 204 J23, V24, AF23, AC15, AF15,
AD11, AD5, Y3, T1, J3, G4, D5,
D10, B13, D17, C22, H24
VCCIO 3.3 V/5 V 1, 26, 53, 65, 79, 92, 105, 132, A10, A17, B2, B25, D7, D13,
157, 172, 181, 184 D19, G23, H4, K1, K26, N23, P4,
U1, U26, W23, Y4, AC8, AC14,
AC20, AE25, AF10, AF17
GND 2, 13, 24, 27, 42, 52, 68, 81, 93, A1, A2, A5, A8, A14, A19, A22,
104,1 08, 129, 130, 141, 156, A25, A26, B1, B26, C7, C9, C13,
163, 177, 190, 207 C18, D24, E1, E26, H1, H26, K4,
N1, N24, P3, P26, V23, W1, W4,
W26, AB1, AB4, AB26, AC9,
AD10, AD14, AD15, AD20, AE1,
AE26, AF1, AF2, AF5, AF8,
AF13, AF19, AF22, AF25, AF26
No Connects A18, A23, A24, B4, B8, B10, B23,
C1, C2, C3, C4, C5, C8, C11,
C24, C25, D1, D3, D4, D14, D16,
D21, D23, D25, E3, E4, F3, F23,
G25, J2, J24, J26, K2, L4, L23,
P2, T3, T4, T24, U25, V25, W3,
W24, Y2, AB3, AB23, AC2, AC4,
AC6, AC11, AC16, AC17, AC21,
AC23, AC24, AC25, AD16,
AD21, AD24, AD26, AE2, AE4,
AE10, AE15, AF3, AF4, AF9,
AF20

R
XC95288 -15 HQ 208 C

Package Type

- 20 20 ns pin-to-pin delay HQ208 208-Pin Heat Sink Quad Flat Pack (HQFP)
-15 15 ns pin-to-pin delay BG352 352-Pin Plastic Ball Grid Array (BGA)
Temperature Options
Pins 208 352
Plastic Plastic
Type
HQFP BGA
Code HQ BG
–20 C(I) C(I)
XC95288
–15 C(I) C(I)
Revision Control
Date Revision
11/30/98 V 3.1 - Update AC Characteristics and Internal Parameters

R
XC4000 Products
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

0
R
XC4000E and XC4000X Series
Table of Contents
0 6*
XC4000E and XC4000X Series Field Programmable Gate Arrays

XC4000E and XC4000X Series Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
Low-Voltage Versions Available . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
Additional XC4000X Series Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
Taking Advantage of Reconfiguration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
XC4000E and XC4000X Series Compared to the XC4000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-9
Improvements in XC4000E and XC4000X . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-9
Additional Improvements in XC4000X Only . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10
Detailed Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-11
Basic Building Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-11
Function Generators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-11
Flip-Flops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
Latches (XC4000X only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
Clock Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12 6
Clock Enable. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
Set/Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13
Global Set/Reset. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13
Data Inputs and Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13
Control Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13
Using FPGA Flip-Flops and Latches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13
Using Function Generators as RAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13
Fast Carry Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-20
Input/Output Blocks (IOBs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-23
IOB Input Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-23
IOB Output Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-26
Other IOB Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-28
Three-State Buffers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-29
Three-State Buffer Modes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-29
Three-State Buffer Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-29
Wide Edge Decoders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-30
On-Chip Oscillator. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-30
Programmable Interconnect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-31
Interconnect Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-31
CLB Routing Connections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-31
Programmable Switch Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-32
Single-Length Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-32
Double-Length Lines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-34
Quad Lines (XC4000X only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-34
Longlines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-34
Direct Interconnect (XC4000X only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-35
I/O Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-35
Octal I/O Routing (XC4000X only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-35
Global Nets and Buffers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-38
Global Nets and Buffers (XC4000E only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-38
Global Nets and Buffers (XC4000X only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-40
Power Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-42
6-3
XC4000E and XC4000X Series Table of Contents

Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-45
Data Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-46
Instruction Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-47
Bit Sequence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-47
Including Boundary Scan in a Schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-47
Avoiding Inadvertent Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-48
Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-48
Special Purpose Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-48
Master Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-49
Additional Address lines in XC4000 devices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-49
Peripheral Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-49
Slave Serial Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-49
Cyclic Redundancy Check (CRC) for Configuration and Readback . . . . . . . . . . . . . . . . . . . . . . 6-52
Configuration Memory Clear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-53
Initialization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-54
Start-Up. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-54
DONE Goes High to Signal End of Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-57
Release of User I/O After DONE Goes High . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-57
Release of Global Set/Reset After DONE Goes High. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-57
Configuration Complete After DONE Goes High . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-57
Readback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-58
Readback Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-59
Read Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-59
Read Abort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-59
Clock Select . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-59
Violating the Maximum High and Low Time Specification for the Readback Clock . . . . . . . . . . 6-59
Readback with the XChecker Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-59
XC4000E/EX/XL Program Readback Switching Characteristic Guidelines . . . . . . . . . . . . . . . . 6-60
Configuration Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-63
Master Parallel Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-65
Additional Address lines in XC4000 devices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-65
Synchronous Peripheral Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-67
Asynchronous Peripheral Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-69
Write to FPGA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-69
Status Read . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-69
Configuration Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-71
Master Modes (XC4000E/EX). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-71
Master Modes (XC4000XL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-71
Slave and Peripheral Modes(All) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-71
XC4000XL Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-72
Additional Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-72
Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-72
Recommended Operating Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-72
DC Characteristics Over Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . 6-73
XC4000XL Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 6-74
6-4
XC4000XL Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 6-75
XC4000XL CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-76
XC4000XL Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-79
Capacitive Load Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-79
XC4000XL Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-80
XC4000XL Pin-to-Pin Input Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-81
XC4000XL Global Low Skew Clock, Set-Up and Hold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-81
XC4000XL BUFGE #s 1, 2, 5, and 6 Global Early Clock, Set-up and Hold for IFF and FCL . . . 6-82
XC4000XL BUFGE #s 3, 4, 7, and 8 Global Early Clock, Set-up and Hold for IFF and FCL . . . 6-83
XC4000XL IOB Input Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-84
XC4000XL IOB Input Switching Characteristic Guidelines (Cont) . . . . . . . . . . . . . . . . . . . . . . . 6-85
XC4000XL IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-86
XC4000EX Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-87
XC4000EX Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-87
XC4000EX Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-87
XC4000EX DC Characteristics Over Recommended Operating Conditions . . . . . . . . . . . . . . . 6-88
XC4000EX Longline and Wide Decoder Timing Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-89
XC4000EX CLB Switching Characteristic Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-90
XC4000EX CLB RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . . . . . . . . 6-92
XC4000EX CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . 6-92
XC4000EX CLB RAM Asynchronous (Level-Sensitive) Write and Read Operation Guidelines . 6-93
XC4000EX CLB RAM Asynchronous (Level-Sensitive) Timing Characteristics . . . . . . . . . . . . . 6-94
XC4000EX Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-95 6
XC4000EX Output MUX, Clock to Out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-95
XC4000EX Output Level and Slew Rate Adjustments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-95
XC4000EX Pin-to-Pin Input Parameter Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-96
XC4000EX Global Early Clock, Set-Up and Hold for IFF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-96
XC4000EX Global Early Clock, Set-Up and Hold for FCL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-96
XC4000EX Input Threshold Adjustments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-96
XC4000EX IOB Input Switching Characteristic Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-97
XC4000EX IOB Input Switching Characteristic Guidelines (Continued) . . . . . . . . . . . . . . . . . . . 6-98
XC4000EX IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-99
XC4000E Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-100
XC4000E Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-100
XC4000E Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-100
XC4000E DC Characteristics Over Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-101
XC4000E Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-101
XC4000E Horizontal Longline Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . 6-102
XC4000E Wide Decoder Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 6-103
XC4000E CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-104
XC4000E CLB RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . . . . . . . . . 6-107
XC4000E CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . 6-107
XC4000E CLB Level-Sensitive RAM Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . 6-108
XC4000E CLB Level-Sensitive RAM Timing Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-109
XC4000E Guaranteed Input and Output Parameters (Pin-to-Pin, TTL I/O) . . . . . . . . . . . . . . . . 6-110
XC4000E IOB Input Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-110
XC4000E IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-112
XC4000E Boundary Scan (JTAG) Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . 6-114
XC4002XL Device Pinout Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-115
Additional XC4002XL Package Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-115
Pin Locations for XC4003E Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-116
Additional XC4003E Package Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-116
Pin Locations for XC4005E/XL Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-117
6-5
XC4000E and XC4000X Series Table of Contents
Additional XC4005E/XL Package Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-118

Pin Locations for XC4025E, XC4028EX/XL Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-128
Additional XC4025E, XC4028EX/XL Package Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-130
Pin Locations for XC4036EX/XL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-131
Additional XC4036EX/XL Package Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-134
Pin Locations for XC4044XL Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-134
User I/O Per Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-156
6-6
book
1
R
XC4000E and XC4000X Series Field
January 29, 1999 (Version 1.5) 1 6* Product Specification
XC4000E and XC4000X Series Low-Voltage Versions Available

Features • Low-Voltage Devices Function at 3.0 - 3.6 Volts
Note: XC4000 Series devices described in this data sheet • XC4000XL: High Performance Low-Voltage Versions of
include the XC4000E family and XC4000X Series. XC4000EX devices
XC4000X Series devices described in this data sheet
include the XC4000EX and XC4000XL families. Separate Additional XC4000X Series Features
data sheets are available for two other Families in the • Highest Performance — 3.3 V XC4000XL
XC4000X series, the XC4000XLT and XC4000XV. This • Highest Capacity — Over 180,000 Usable Gates
information does not apply to the older Xilinx families: • 5V tolerant I/Os on XC4000XL
XC4000, XC4000A, XC4000D, XC4000H, or XC4000L. For • 0.35µ SRAM process for XC4000XL
information on these devices, see the Xilinx WEBLINX at • Additional Routing Over XC4000E
http://www.xilinx.com. - almost twice the routing capacity for high-density
• System featured Field-Programmable Gate Arrays designs
- Select-RAMTM memory: on-chip ultra-fast RAM with • Buffered Interconnect for Maximum Speed
- synchronous write option • New Latch Capability in Configurable Logic Blocks
- dual-port RAM option • Improved VersaRingTM I/O Interconnect for Better Fixed 6
- Fully PCI compliant (speed grades -2 and faster) Pinout Flexibility
- Abundant flip-flops • 12-mA Sink Current Per XC4000X Output
- Flexible function generators • Flexible New High-Speed Clock Network
- Dedicated high-speed carry logic - 8 additional Early Buffers for shorter clock delays
- Wide edge decoders on each edge - Virtually unlimited number of clock signals
- Hierarchy of interconnect lines • Optional Multiplexer or 2-input Function Generator on
- Internal 3-state bus capability Device Outputs
- 8 global low-skew clock or signal distribution • 4 Additional Address Bits in Master Parallel
networks Configuration Mode
• System Performance beyond 80 MHz • XC4000XLT devices, optimized for PCI applications, are
• Flexible Array Architecture available.
• Low Power Segmented Routing Architecture • The XC4000XV Family offers the highest density with
• Systems-Oriented Features 0.25 micron 2.5 volt technology.
- IEEE 1149.1-compatible boundary scan logic
support Introduction
- Individually programmable output slew rate XC4000 Series high-performance, high-capacity Field Pro-
- Programmable input pull-up or pull-down resistors grammable Gate Arrays (FPGAs) provide the benefits of
- 12-mA sink current per XC4000E output custom CMOS VLSI, while avoiding the initial cost, long
• Configured by Loading Binary File development cycle, and inherent risk of a conventional
- Unlimited reprogrammability masked gate array.
• Readback Capability
The result of thirteen years of FPGA design experience and
- Program verification
feedback from thousands of customers, these FPGAs com-
- Internal node observability
bine architectural versatility, on-chip Select-RAM memory
• Backward Compatible with XC4000 Devices
with edge-triggered and dual-port modes, increased
• Development System runs on most common computer
speed, abundant routing resources, and new, sophisticated
platforms
software to achieve fully automated implementation of
- Interfaces to popular design environments
complex, high-density, high-performance designs.
- Fully automatic mapping, placement and routing
- Interactive design editor for design optimization The XC4000E and XC4000X Series currently have 20
members, as shown in Table 2.

R
XC4000E and XC4000X Series Field Programmable Gate
Table 2: XC4000E and XC4000X Series Field Programmable Gate Arrays
Max Logic Max. RAM Typical Number

Logic Gates Bits Gate Range CLB Total of Max.
Device Cells (No RAM) (No Logic) (Logic and RAM)* Matrix CLBs Flip-Flops User I/O
XC4002XL 152 1,600 2,048 1,000 - 3,000 8x8 64 256 64
XC4003E 238 3,000 3,200 2,000 - 5,000 10 x 10 100 360 80
XC4005E/XL 466 5,000 6,272 3,000 - 9,000 14 x 14 196 616 112
XC4006E 608 6,000 8,192 4,000 - 12,000 16 x 16 256 768 128
XC4008E 770 8,000 10,368 6,000 - 15,000 18 x 18 324 936 144
XC4010E/XL 950 10,000 12,800 7,000 - 20,000 20 x 20 400 1,120 160
XC4013E/XL 1368 13,000 18,432 10,000 - 30,000 24 x 24 576 1,536 192
XC4020E/XL 1862 20,000 25,088 13,000 - 40,000 28 x 28 784 2,016 224
XC4025E 2432 25,000 32,768 15,000 - 45,000 32 x 32 1,024 2,560 256
XC4028EX/XL 2432 28,000 32,768 18,000 - 50,000 32 x 32 1,024 2,560 256
XC4036EX/XL 3078 36,000 41,472 22,000 - 65,000 36 x 36 1,296 3,168 288
XC4044XL 3800 44,000 51,200 27,000 - 80,000 40 x 40 1,600 3,840 320
XC4052XL 4598 52,000 61,952 33,000 - 100,000 44 x 44 1,936 4,576 352
XC4062XL 5472 62,000 73,728 40,000 - 130,000 48 x 48 2,304 5,376 384
XC4085XL 7448 85,000 100,352 55,000 - 180,000 56 x 56 3,136 7,168 448
* Max values of Typical Gate Range include 20-30% of CLBs used as RAM.
Note: All functionality in low-voltage families is the same as where hardware is changed dynamically, or where hard-
in the corresponding 5-Volt family, except where numerical ware must be adapted to different user applications.
references are made to timing or power. FPGAs are ideal for shortening design and development
cycles, and also offer a cost-effective solution for produc-
Description tion rates well beyond 5,000 systems per month. For lowest
XC4000 Series devices are implemented with a regular, high-volume unit cost, a design can first be implemented in
flexible, programmable architecture of Configurable Logic the XC4000E or XC4000X, then migrated to one of Xilinx’
Blocks (CLBs), interconnected by a powerful hierarchy of compatible HardWire mask-programmed devices.
versatile routing resources, and surrounded by a perimeter
of programmable Input/Output Blocks (IOBs). They have
Taking Advantage of Reconfiguration
generous routing resources to accommodate the most FPGA devices can be reconfigured to change logic function
complex interconnect patterns. while resident in the system. This capability gives the sys-
tem designer a new degree of freedom not available with
The devices are customized by loading configuration data
any other type of logic.
into internal memory cells. The FPGA can either actively
read its configuration data from an external serial or Hardware can be changed as easily as software. Design
byte-parallel PROM (master modes), or the configuration updates or modifications are easy, and can be made to
data can be written into the FPGA from an external device products already in the field. An FPGA can even be recon-
(slave and peripheral modes). figured dynamically to perform different functions at differ-
ent times.
XC4000 Series FPGAs are supported by powerful and
sophisticated software, covering every aspect of design Reconfigurable logic can be used to implement system
from schematic or behavioral entry, floorplanning, simula- self-diagnostics, create systems capable of being reconfig-
tion, automatic block placement and routing of intercon- ured for different environments or operations, or implement
nects, to the creation, downloading, and readback of the multi-purpose hardware for a given application. As an
configuration bit stream. added benefit, using reconfigurable FPGA devices simpli-
fies hardware design and debugging and shortens product
Because Xilinx FPGAs can be reprogrammed an unlimited
time-to-market.
number of times, they can be used in innovative designs

R
XC4000E and XC4000X Series much as 50% from XC4000 values. See “Fast Carry Logic”
on page 20 for more information.
Compared to the XC4000
For readers already familiar with the XC4000 family of Xil- Select-RAM Memory: Edge-Triggered, Synchronous
inx Field Programmable Gate Arrays, the major new fea- RAM Modes
tures in the XC4000 Series devices are listed in this The RAM in any CLB can be configured for synchronous,
section. The biggest advantages of XC4000E and edge-triggered, write operation. The read operation is not
XC4000X devices are significantly increased system affected by this change to an edge-triggered write.
speed, greater capacity, and new architectural features,
particularly Select-RAM memory. The XC4000X devices Dual-Port RAM
also offer many new routing features, including special A separate option converts the 16x2 RAM in any CLB into a
high-speed clock buffers that can be used to capture input 16x1 dual-port RAM with simultaneous Read/Write.
data with minimal delay.
The function generators in each CLB can be configured as
Any XC4000E device is pinout- and bitstream-compatible either level-sensitive (asynchronous) single-port RAM,
with the corresponding XC4000 device. An existing edge-triggered (synchronous) single-port RAM, edge-trig-
XC4000 bitstream can be used to program an XC4000E gered (synchronous) dual-port RAM, or as combinatorial
device. However, since the XC4000E includes many new logic.
features, an XC4000E bitstream cannot be loaded into an
XC4000 device. Configurable RAM Content
XC4000X Series devices are not bitstream-compatible with The RAM content can now be loaded at configuration time,
equivalent array size devices in the XC4000 or XC4000E so that the RAM starts up with user-defined data.
families. However, equivalent array size devices, such as
H Function Generator
the XC4025, XC4025E, XC4028EX, and XC4028XL, are
6
pinout-compatible. In current XC4000 Series devices, the H function generator
is more versatile than in the original XC4000. Its inputs can
Improvements in XC4000E and XC4000X come not only from the F and G function generators but
also from up to three of the four control input lines. The H
Increased System Speed function generator can thus be totally or partially indepen-
XC4000E and XC4000X devices can run at synchronous dent of the other two function generators, increasing the
system clock rates of up to 80 MHz, and internal perfor- maximum capacity of the device.
mance can exceed 150 MHz. This increase in performance
over the previous families stems from improvements in both IOB Clock Enable
device processing and system architecture. XC4000 The two flip-flops in each IOB have a common clock enable
Series devices use a sub-micron multi-layer metal process. input, which through configuration can be activated individ-
In addition, many architectural improvements have been ually for the input or output flip-flop or both. This clock
made, as described below. enable operates exactly like the EC pin on the XC4000
The XC4000XL family is a high performance 3.3V family CLB. This new feature makes the IOBs more versatile, and
based on 0.35µ SRAM technology and supports system avoids the need for clock gating.
speeds to 80 MHz.
Output Drivers
PCI Compliance The output pull-up structure defaults to a TTL-like
XC4000 Series -2 and faster speed grades are fully PCI totem-pole. This driver is an n-channel pull-up transistor,
compliant. XC4000E and XC4000X devices can be used to pulling to a voltage one transistor threshold below Vcc, just
implement a one-chip PCI solution. like the XC4000 family outputs. Alternatively, XC4000
Series devices can be globally configured with CMOS out-
Carry Logic puts, with p-channel pull-up transistors pulling to Vcc. Also,
the configurable pull-up resistor in the XC4000 Series is a
The speed of the carry logic chain has increased dramati-
p-channel transistor that pulls to Vcc, whereas in the origi-
cally. Some parameters, such as the delay on the carry
nal XC4000 family it is an n-channel transistor that pulls to
chain through a single CLB (TBYP), have improved by as
a voltage one transistor threshold below Vcc.

R
Input Thresholds Additional Improvements in XC4000X Only

The input thresholds of 5V devices can be globally config-
Increased Routing
ured for either TTL (1.2 V threshold) or CMOS (2.5 V
threshold), just like XC2000 and XC3000 inputs. The two New interconnect in the XC4000X includes twenty-two
global adjustments of input threshold and output level are additional vertical lines in each column of CLBs and twelve
independent of each other. The XC4000XL family has an new horizontal lines in each row of CLBs. The twelve “Quad
input threshold of 1.6V, compatible with both 3.3V CMOS Lines” in each CLB row and column include optional repow-
and TTL levels. ering buffers for maximum speed. Additional high-perfor-
mance routing near the IOBs enhances pin flexibility.
Global Signal Access to Logic
Faster Input and Output
There is additional access from global clocks to the F and
G function generator inputs. A fast, dedicated early clock sourced by global clock buffers
is available for the IOBs. To ensure synchronization with the
Configuration Pin Pull-Up Resistors regular global clocks, a Fast Capture latch driven by the
During configuration, the three mode pins, M0, M1, and early clock is available. The input data can be initially
M2, have weak pull-up resistors. For the most popular con- loaded into the Fast Capture latch with the early clock, then
figuration mode, Slave Serial, the mode pins can thus be transferred to the input flip-flop or latch with the low-skew
left unconnected. global clock. A programmable delay on the input can be
used to avoid hold-time requirements. See “IOB Input Sig-
The three mode inputs can be individually configured with
nals” on page 23 for more information.
or without weak pull-up or pull-down resistors after configu-
ration. Latch Capability in CLBs
The PROGRAM input pin has a permanent weak pull-up. Storage elements in the XC4000X CLB can be configured
as either flip-flops or latches. This capability makes the
Soft Start-up
FPGA highly synthesis-compatible.
Like the XC3000A, XC4000 Series devices have “Soft
Start-up.” When the configuration process is finished and IOB Output MUX From Output Clock
the device starts up, the first activation of the outputs is A multiplexer in the IOB allows the output clock to select
automatically slew-rate limited. This feature avoids poten- either the output data or the IOB clock enable as the output
tial ground bounce when all outputs are turned on simulta- to the pad. Thus, two different data signals can share a sin-
neously. Immediately after start-up, the slew rate of the gle output pad, effectively doubling the number of device
individual outputs is, as in the XC4000 family, determined outputs without requiring a larger, more expensive pack-
by the individual configuration option. age. This multiplexer can also be configured as an
AND-gate to implement a very fast pin-to-pin path. See
XC4000 and XC4000A Compatibility
“IOB Output Signals” on page 26 for more information.
Existing XC4000 bitstreams can be used to configure an
XC4000E device. XC4000A bitstreams must be recompiled Additional Address Bits
for use with the XC4000E due to improved routing Larger devices require more bits of configuration data. A
resources, although the devices are pin-for-pin compatible. daisy chain of several large XC4000X devices may require
a PROM that cannot be addressed by the eighteen address
bits supported in the XC4000E. The XC4000X Series
therefore extends the addressing in Master Parallel config-
uration mode to 22 bits.

R
Detailed Functional Description Each CLB contains two storage elements that can be used
to store the function generator outputs. However, the stor-
XC4000 Series devices achieve high speed through age elements and function generators can also be used
advanced semiconductor technology and improved archi- independently. These storage elements can be configured
tecture. The XC4000E and XC4000X support system clock as flip-flops in both XC4000E and XC4000X devices; in the
rates of up to 80 MHz and internal performance in excess XC4000X they can optionally be configured as latches. DIN
of 150 MHz. Compared to older Xilinx FPGA families, can be used as a direct input to either of the two storage
XC4000 Series devices are more powerful. They offer elements. H1 can drive the other through the H function
on-chip edge-triggered and dual-port RAM, clock enables generator. Function generator outputs can also drive two
on I/O flip-flops, and wide-input decoders. They are more outputs independent of the storage element outputs. This
versatile in many applications, especially those involving versatility increases logic capacity and simplifies routing.
RAM. Design cycles are faster due to a combination of
increased routing resources and more sophisticated soft- Thirteen CLB inputs and four CLB outputs provide access
ware. to the function generators and storage elements. These
inputs and outputs connect to the programmable intercon-
Basic Building Blocks nect resources outside the block.
Xilinx user-programmable gate arrays include two major Function Generators

configurable elements: configurable logic blocks (CLBs)
Four independent inputs are provided to each of two func-
and input/output blocks (IOBs).
tion generators (F1 - F4 and G1 - G4). These function gen-
• CLBs provide the functional elements for constructing erators, with outputs labeled F’ and G’, are each capable of
the user’s logic. implementing any arbitrarily defined Boolean function of
• IOBs provide the interface between the package pins four inputs. The function generators are implemented as
and internal signal lines. memory look-up tables. The propagation delay is therefore
independent of the function implemented. 6
Three other types of circuits are also available:
• 3-State buffers (TBUFs) driving horizontal longlines are A third function generator, labeled H’, can implement any
associated with each CLB. Boolean function of its three inputs. Two of these inputs can
• Wide edge decoders are available around the periphery optionally be the F’ and G’ functional generator outputs.
of each device. Alternatively, one or both of these inputs can come from
• An on-chip oscillator is provided. outside the CLB (H2, H0). The third input must come from
outside the block (H1).
Programmable interconnect resources provide routing
paths to connect the inputs and outputs of these config- Signals from the function generators can exit the CLB on
urable elements to the appropriate networks. two outputs. F’ or H’ can be connected to the X output. G’ or
H’ can be connected to the Y output.
The functionality of each circuit block is customized during
configuration by programming internal static memory cells. A CLB can be used to implement any of the following func-
The values stored in these memory cells determine the tions:
logic functions and interconnections implemented in the • any function of up to four variables, plus any second
FPGA. Each of these available circuits is described in this function of up to four unrelated variables, plus any third
section. function of up to three unrelated variables1
• any single function of five variables
Configurable Logic Blocks (CLBs) • any function of four variables together with some
Configurable Logic Blocks implement most of the logic in functions of six variables
an FPGA. The principal CLB elements are shown in • some functions of up to nine variables.
Figure 2. Two 4-input function generators (F and G) offer Implementing wide functions in a single block reduces both
unrestricted versatility. Most combinatorial logic functions the number of blocks required and the delay in the signal
need four or fewer inputs. However, a third function gener- path, achieving both increased capacity and speed.
ator (H) is provided. The H function generator has three
The versatility of the CLB function generators significantly
inputs. Either zero, one, or two of these inputs can be the
outputs of F and G; the other input(s) are from outside the improves system speed. In addition, the design-software
CLB. The CLB can, therefore, implement certain functions tools can deal with each function generator independently.
of up to nine variables, like parity check or expand- This flexibility improves cell usage.
able-identity comparison of two sets of four inputs.
1. When three separate functions are generated, one of the function outputs must be captured in a flip-flop internal to the CLB. Only two
unregistered function generator outputs are available from the CLB.

R
4
C1 • • • C 4
H1 D IN /H 2 SR/H 0 EC
G4 S/R Bypass
CONTROL
DIN YQ
G3 LOGIC SD
F' D
FUNCTION G' Q
G'
OF
H'
G2 G1-G4
G1
LOGIC
EC
FUNCTION RD
G'
OF H' H'
F', G', 1
AND Y
H1
F4 Bypass
S/R
CONTROL
DIN XQ
F3 LOGIC SD
F'
FUNCTION F' D Q
G'
OF
H'
F2 F1-F4
F1
EC
RD
K
(CLOCK) 1
H'
X
F'
X6692
Figure 2: Simplified Block Diagram of XC4000 Series CLB (RAM and Carry Logic functions not shown)
Flip-Flops Clock Enable

The CLB can pass the combinatorial output(s) to the inter- The clock enable signal (EC) is active High. The EC pin is
connect network, but can also store the combinatorial shared by both storage elements. If left unconnected for
results or other incoming data in one or two flip-flops, and either, the clock enable for that storage element defaults to
connect their outputs to the interconnect network as well. the active state. EC is not invertible within the CLB.
The two edge-triggered D-type flip-flops have common
clock (K) and clock enable (EC) inputs. Either or both clock Table 3: CLB Storage Element Functionality
inputs can also be permanently enabled. Storage element (active rising edge is shown)
functionality is described in Table 3.
Mode K EC SR D Q
Latches (XC4000X only) Power-Up or
X X X X SR
GSR
The CLB storage elements can also be configured as
latches. The two latches have common clock (K) and clock X X 1 X SR
enable (EC) inputs. Storage element functionality is Flip-Flop __/ 1* 0* D D
described in Table 3. 0 X 0* X Q
1 1* 0* X Q
Clock Input Latch
0 1* 0* D D
Each flip-flop can be triggered on either the rising or falling Both X 0 0* X Q
clock edge. The clock pin is shared by both storage ele- Legend:
ments. However, the clock is individually invertible for each X Don’t care
storage element. Any inverter placed on the clock input is __/ Rising edge
automatically absorbed into the CLB. SR Set or Reset value. Reset is default.

R
Set/Reset Two fast feed-through paths are available, as shown in

Figure 2. A two-to-one multiplexer on each of the XQ and
An asynchronous storage element input (SR) can be con-
YQ outputs selects between a storage element output and
figured as either set or reset. This configuration option
any of the control inputs. This bypass is sometimes used by
determines the state in which each flip-flop becomes oper-
the automated router to repower internal signals.
ational after configuration. It also determines the effect of a
Global Set/Reset pulse during normal operation, and the Control Signals
effect of a pulse on the SR pin of the CLB. All three
Multiplexers in the CLB map the four control inputs (C1 - C4
set/reset functions for any single flip-flop are controlled by
in Figure 2) into the four internal control signals (H1,
the same configuration data bit.
DIN/H2, SR/H0, and EC). Any of these inputs can drive any
The set/reset state can be independently specified for each of the four internal control signals.
flip-flop. This input can also be independently disabled for
When the logic function is enabled, the four inputs are:
either flip-flop.
• EC — Enable Clock
The set/reset state is specified by using the INIT attribute,
• SR/H0 — Asynchronous Set/Reset or H function
or by placing the appropriate set or reset flip-flop library
generator Input 0
symbol.
• DIN/H2 — Direct In or H function generator Input 2
SR is active High. It is not invertible within the CLB. • H1 — H function generator Input 1.
Global Set/Reset When the memory function is enabled, the four inputs are:
A separate Global Set/Reset line (not shown in Figure 2) • EC — Enable Clock
sets or clears each storage element during power-up, • WE — Write Enable
reconfiguration, or when a dedicated Reset net is driven • D0 — Data Input to F and/or G function generator
active. This global net (GSR) does not compete with other • D1 — Data input to G function generator (16x1 and
16x2 modes) or 5th Address bit (32x1 mode).
6
routing resources; it uses a dedicated distribution network.
Each flip-flop is configured as either globally set or reset in Using FPGA Flip-Flops and Latches
the same way that the local set/reset (SR) is specified.
The abundance of flip-flops in the XC4000 Series invites
Therefore, if a flip-flop is set by SR, it is also set by GSR.
pipelined designs. This is a powerful way of increasing per-
Similarly, a reset flip-flop is reset by both SR and GSR.
formance by breaking the function into smaller subfunc-
tions and executing them in parallel, passing on the results
STARTUP
through pipeline flip-flops. This method should be seriously
PAD GSR Q2 considered wherever throughput is more important than
GTS Q3 latency.
IBUF
Q1Q4
To include a CLB flip-flop, place the appropriate library
CLK DONEIN
symbol. For example, FDCE is a D-type flip-flop with clock
X5260 enable and asynchronous clear. The corresponding latch
Figure 3: Schematic Symbols for Global Set/Reset symbol (for the XC4000X only) is called LDCE.
In XC4000 Series devices, the flip flops can be used as reg-
GSR can be driven from any user-programmable pin as a isters or shift registers without blocking the function gener-
global reset input. To use this global net, place an input pad ators from performing a different, perhaps unrelated task.
and input buffer in the schematic or HDL code, driving the This ability increases the functional capacity of the devices.
GSR pin of the STARTUP symbol. (See Figure 3.) A spe- The CLB setup time is specified between the function gen-
cific pin location can be assigned to this input using a LOC erator inputs and the clock input K. Therefore, the specified
attribute or property, just as with any other user-program- CLB flip-flop setup time includes the delay through the
mable pad. An inverter can optionally be inserted after the function generator.
input buffer to invert the sense of the Global Set/Reset sig-
nal. Using Function Generators as RAM
Alternatively, GSR can be driven from any internal node. Optional modes for each CLB make the memory look-up
tables in the F’ and G’ function generators usable as an
Data Inputs and Outputs array of Read/Write memory cells. Available modes are
The source of a storage element data input is programma- level-sensitive (similar to the XC4000/A/H families),
ble. It is driven by any of the functions F’, G’, and H’, or by edge-triggered, and dual-port edge-triggered. Depending
the Direct In (DIN) block input. The flip-flops or latches drive on the selected mode, a single CLB can be configured as
the XQ and YQ CLB outputs. either a 16x2, 32x1, or 16x1 bit array.

R
Supported CLB memory configurations and timing modes The selected timing mode applies to both function genera-
for single- and dual-port modes are shown in Table 4. tors within a CLB when both are configured as RAM.
XC4000 Series devices are the first programmable logic The number of read ports is also programmable:
devices with edge-triggered (synchronous) and dual-port
• Single Port: each function generator has a common
RAM accessible to the user. Edge-triggered RAM simpli-
read and write port
fies system timing. Dual-port RAM doubles the effective
• Dual Port: both function generators are configured
throughput of FIFO applications. These features can be
together as a single 16x1 dual-port RAM with one write
individually programmed in any XC4000 Series CLB.
port and two read ports. Simultaneous read and write
Advantages of On-Chip and Edge-Triggered RAM operations to the same or different addresses are
supported.
The on-chip RAM is extremely fast. The read access time is
the same as the logic delay. The write access time is RAM configuration options are selected by placing the
slightly slower. Both access times are much faster than appropriate library symbol.
any off-chip solution, because they avoid I/O delays.
Choosing a RAM Configuration Mode
Edge-triggered RAM, also called synchronous RAM, is a
The appropriate choice of RAM mode for a given design
feature never before available in a Field Programmable
should be based on timing and resource requirements,
Gate Array. The simplicity of designing with edge-triggered
desired functionality, and the simplicity of the design pro-
RAM, and the markedly higher achievable performance,
cess. Recommended usage is shown in Table 5.
add up to a significant improvement over existing devices
with on-chip RAM. The difference between level-sensitive, edge-triggered,
and dual-port RAM is only in the write operation. Read
Three application notes are available from Xilinx that dis-
operation and timing is identical for all modes of operation.
cuss edge-triggered RAM: “XC4000E Edge-Triggered and
Dual-Port RAM Capability,” “Implementing FIFOs in
Table 5: RAM Mode Selection
XC4000E RAM,” and “Synchronous and Asynchronous
FIFO Designs.” All three application notes apply to both Dual-Port
XC4000E and XC4000X RAM. Level-Sens Edge-Trigg Edge-Trigg
itive ered ered
Table 4: Supported RAM Modes Use for New
No Yes Yes
Designs?
16 16 32 Edge- Level-
x x x Triggered Sensitive Size (16x1,
1/2 CLB 1/2 CLB 1 CLB
1 2 1 Timing Timing Registered)
Single-Port √ √ √ √ √ Simultaneous
No No Yes
Read/Write
Dual-Port √ √
Relative 2X (4X
X 2X
Performance effective)
RAM Configuration Options
The function generators in any CLB can be configured as RAM Inputs and Outputs
RAM arrays in the following sizes:
The F1-F4 and G1-G4 inputs to the function generators act
• Two 16x1 RAMs: two data inputs and two data outputs as address lines, selecting a particular memory cell in each
with identical or, if preferred, different addressing for look-up table.
each RAM
• One 32x1 RAM: one data input and one data output. The functionality of the CLB control signals changes when
the function generators are configured as RAM. The
One F or G function generator can be configured as a 16x1 DIN/H2, H1, and SR/H0 lines become the two data inputs
RAM while the other function generators are used to imple- (D0, D1) and the Write Enable (WE) input for the 16x2
ment any function of up to 5 inputs. memory. When the 32x1 configuration is selected, D1 acts
Additionally, the XC4000 Series RAM may have either of as the fifth address bit and D0 is the data input.
two timing modes: The contents of the memory cell(s) being addressed are
• Edge-Triggered (Synchronous): data written by the available at the F’ and G’ function-generator outputs. They
designated edge of the CLB clock. WE acts as a true can exit the CLB through its X and Y outputs, or can be cap-
clock enable. tured in the CLB flip-flop(s).
• Level-Sensitive (Asynchronous): an external WE signal Configuring the CLB function generators as Read/Write
acts as the write strobe. memory does not affect the functionality of the other por-

R
tions of the CLB, with the exception of the redefinition of the nals. An internal write pulse is generated that performs the
control signals. In 16x2 and 16x1 modes, the H’ function write. See Figure 5 and Figure 6 for block diagrams of a
generator can be used to implement Boolean functions of CLB configured as 16x2 and 32x1 edge-triggered, sin-
F’, G’, and D1, and the D flip-flops can latch the F’, G’, H’, or gle-port RAM.
D0 signals. The relationships between CLB pins and RAM inputs and
Single-Port Edge-Triggered Mode outputs for single-port, edge-triggered mode are shown in
Table 6.
Edge-triggered (synchronous) RAM simplifies timing
requirements. XC4000 Series edge-triggered RAM timing The Write Clock input (WCLK) can be configured as active
operates like writing to a data register. Data and address on either the rising edge (default) or the falling edge. It uses
are presented. The register is enabled for writing by a logic the same CLB pin (K) used to clock the CLB flip-flops, but it
High on the write enable input, WE. Then a rising or falling can be independently inverted. Consequently, the RAM
clock edge loads the data into the register, as shown in output can optionally be registered within the same CLB
Figure 4. either by the same clock edge as the RAM, or by the oppo-
site edge of this clock. The sense of WCLK applies to both
TWPS function generators in the CLB when both are configured
WCLK (K) as RAM.
TWSS TWHS The WE pin is active-High and is not invertible within the
CLB.
WE
Note: The pulse following the active edge of WCLK (TWPS
TDSS TDHS in Figure 4) must be less than one millisecond wide. For
most applications, this requirement is not overly restrictive;
DATA IN however, it must not be forgotten. Stopping WCLK at this 6
point in the write cycle could result in excessive current and
TASS TAHS
even damage to the larger devices if many CLBs are con-
figured as edge-triggered RAM.
ADDRESS
Table 6: Single-Port Edge-Triggered RAM Signals
TILO TILO
TWOS RAM Signal CLB Pin Function
D D0 or D1 (16x2, Data In
DATA OUT OLD NEW 16x1), D0 (32x1)
X6461
A[3:0] F1-F4 or G1-G4 Address
A[4] D1 (32x1) Address
Figure 4: Edge-Triggered RAM Write Timing
WE WE Write Enable
Complex timing relationships between address, data, and WCLK K Clock
write enable signals are not required, and the external write SPO F’ or G’ Single Port Out
enable pulse becomes a simple clock enable. The active (Data Out) (Data Out)
edge of WCLK latches the address, input data, and WE sig-

R
4
C1 • • • C 4
WE D1 D0 EC
DIN
WRITE 16-LATCH
DECODER ARRAY MUX G'
4 4
G1 • • • G 4
1 of 16
LATCH
ENABLE
READ
WRITE PULSE ADDRESS
DIN
WRITE 16-LATCH
DECODER ARRAY MUX F'
4 4
F1 • • • F 4
1 of 16
LATCH
ENABLE
K
(CLOCK) READ
WRITE PULSE ADDRESS
X6752
Figure 5: 16x2 (or 16x1) Edge-Triggered Single-Port RAM
4
C1 • • • C4
EC
WE D1/A4 D0 EC
DIN
WRITE 16-LATCH
G1 • • • G4 4 4
F1 • • • F4 1 of 16
LATCH
ENABLE
READ
WRITE PULSE ADDRESS
H'
DIN
WRITE 16-LATCH
4 4
1 of 16
LATCH
ENABLE
K
(CLOCK) READ
WRITE PULSE ADDRESS
X6754
Figure 6: 32x1 Edge-Triggered Single-Port RAM (F and G addresses are identical)

R
Dual-Port Edge-Triggered Mode Table 7: Dual-Port Edge-Triggered RAM Signals

In dual-port mode, both the F and G function generators RAM Signal CLB Pin Function
are used to create a single 16x1 RAM array with one write D D0 Data In
port and two read ports. The resulting RAM array can be A[3:0] F1-F4 Read Address for F,
read and written simultaneously at two independent Write Address for F and G
addresses. Simultaneous read and write operations at the DPRA[3:0] G1-G4 Read Address for G
same address are also supported. WE WE Write Enable
WCLK K Clock
Dual-port mode always has edge-triggered write timing, as
SPO F’ Single Port Out
shown in Figure 4.
(addressed by A[3:0])
Figure 7 shows a simple model of an XC4000 Series CLB DPO G’ Dual Port Out
configured as dual-port RAM. One address port, labeled (addressed by DPRA[3:0])
A[3:0], supplies both the read and write address for the F
Note: The pulse following the active edge of WCLK (TWPS
function generator. This function generator behaves the
in Figure 4) must be less than one millisecond wide. For
same as a 16x1 single-port edge-triggered RAM array. The
most applications, this requirement is not overly restrictive;
RAM output, Single Port Out (SPO), appears at the F func-
however, it must not be forgotten. Stopping WCLK at this
tion generator output. SPO, therefore, reflects the data at
point in the write cycle could result in excessive current and
address A[3:0].
even damage to the larger devices if many CLBs are con-
The other address port, labeled DPRA[3:0] for Dual Port figured as edge-triggered RAM.
Read Address, supplies the read address for the G function
generator. The write address for the G function generator, Single-Port Level-Sensitive Timing Mode
however, comes from the address A[3:0]. The output from Note: Edge-triggered mode is recommended for all new
this 16x1 RAM array, Dual Port Out (DPO), appears at the designs. Level-sensitive mode, also called asynchronous
G function generator output. DPO, therefore, reflects the 6
mode, is still supported for XC4000 Series backward-com-
data at address DPRA[3:0]. patibility with the XC4000 family.
Therefore, by using A[3:0] for the write address and Level-sensitive RAM timing is simple in concept but can be
DPRA[3:0] for the read address, and reading only the DPO complicated in execution. Data and address signals are
output, a FIFO that can read and write simultaneously is presented, then a positive pulse on the write enable pin
easily generated. Simultaneous access doubles the effec- (WE) performs a write into the RAM at the designated
tive throughput of the FIFO. address. As indicated by the “level-sensitive” label, this
The relationships between CLB pins and RAM inputs and RAM acts like a latch. During the WE High pulse, changing
outputs for dual-port, edge-triggered mode are shown in the data lines results in new data written to the old address.
Table 7. See Figure 8 on page 18 for a block diagram of a Changing the address lines while WE is High results in spu-
CLB configured in this mode. rious data written to the new address—and possibly at
other addresses as well, as the address lines inevitably do
not all change simultaneously.
RAM16X1D Primitive
DPO (Dual Port Out)
The user must generate a carefully timed WE signal. The
WE WE
Registered DPO
delay on the WE signal and the address lines must be care-
D D D Q
DPRA[3:0] AR[3:0]
fully verified to ensure that WE does not become active
AW[3:0] until after the address lines have settled, and that WE goes
inactive before the address lines change again. The data
G Function Generator
must be stable before and after the falling edge of WE.
SPO (Single Port Out)
WE
In practical terms, WE is usually generated by a 2X clock. If
D D Q Registered SPO a 2X clock is not available, the falling edge of the system
A[3:0] AR[3:0]
clock can be used. However, there are inherent risks in this
AW[3:0]
approach, since the WE pulse must be guaranteed inactive
F Function Generator
before the next rising edge of the system clock. Several
WCLK
older application notes are available from Xilinx that dis-
cuss the design of level-sensitive RAMs. These application
X6755
notes include XAPP031, “Using the XC4000 RAM Capabil-
Figure 7: XC4000 Series Dual-Port RAM, Simple ity,” and XAPP042, “High-Speed RAM Design in XC4000.”
Model However, the edge-triggered RAM available in the XC4000
Series is superior to level-sensitive RAM for almost every
application.

R
4
C1 • • • C 4
WE D1 D0 EC
DIN
WRITE 16-LATCH
4
1 of 16
LATCH
ENABLE
READ
4 WRITE PULSE ADDRESS
G1 • • • G 4
DIN
WRITE 16-LATCH
4 4
F1 • • • F 4 1 of 16
LATCH
ENABLE
K
READ
(CLOCK) WRITE PULSE ADDRESS
X6748
Figure 8: 16x1 Edge-Triggered Dual-Port RAM
Figure 9 shows the write timing for level-sensitive, sin- attached to the RAM or ROM symbol, as described in the
gle-port RAM. schematic library guide. If not defined, all RAM contents
are initialized to all zeros, by default.
The relationships between CLB pins and RAM inputs and
outputs for single-port level-sensitive mode are shown in RAM initialization occurs only during configuration. The
Table 8. RAM content is not affected by Global Set/Reset.
Figure 10 and Figure 11 show block diagrams of a CLB Table 8: Single-Port Level-Sensitive RAM Signals
configured as 16x2 and 32x1 level-sensitive, single-port
RAM Signal CLB Pin Function
RAM.
D D0 or D1 Data In
Initializing RAM at Configuration A[3:0] F1-F4 or G1-G4 Address
WE WE Write Enable
Both RAM and ROM implementations of the XC4000 O F’ or G’ Data Out
Series devices are initialized during configuration. The ini-
tial contents are defined via an INIT attribute or property

R
T WC
ADDRESS
TAS T WP T AH
WRITE ENABLE
T DS T DH
DATA IN REQUIRED
X6462
Figure 9: Level-Sensitive RAM Write Timing
4
C1 • • • C 4
WE D1 D0 EC
DIN
Enable
WRITE 16-LATCH
4
G1 • • • G 4
1 of 16 6
4
READ ADDRESS
DIN
Enable
WRITE 16-LATCH
4
F1 • • • F 4
1 of 16
4
X6746 READ ADDRESS
Figure 10: 16x2 (or 16x1) Level-Sensitive Single-Port RAM

R
4
C1 • • • C4
WE D1/A4 D0 EC
DIN
Enable
WRITE 16-LATCH
G1 • • • G4 4
F1 • • • F4 1 of 16
4
READ ADDRESS
H'
DIN
Enable
WRITE 16-LATCH
4
1 of 16
4
READ ADDRESS X6749
Figure 11: 32x1 Level-Sensitive Single-Port RAM (F and G addresses are identical)
Fast Carry Logic The carry chain in XC4000E devices can run either up or
down. At the top and bottom of the columns where there
Each CLB F and G function generator contains dedicated
are no CLBs above or below, the carry is propagated to the
arithmetic logic for the fast generation of carry and borrow
right. (See Figure 12.) In order to improve speed in the
signals. This extra output is passed on to the function gen-
high-capacity XC4000X devices, which can potentially
erator in the adjacent CLB. The carry chain is independent
have very long carry chains, the carry chain travels upward
of normal routing resources.
only, as shown in Figure 13. Additionally, standard intercon-
Dedicated fast carry logic greatly increases the efficiency nect can be used to route a carry signal in the downward
and performance of adders, subtractors, accumulators, direction.
comparators and counters. It also opens the door to many
Figure 14 on page 22 shows an XC4000E CLB with dedi-
new applications involving arithmetic operation, where the
cated fast carry logic. The carry logic in the XC4000X is
previous generations of FPGAs were not fast enough or too
similar, except that COUT exits at the top only, and the sig-
inefficient. High-speed address offset calculations in micro-
nal CINDOWN does not exist. As shown in Figure 14, the
processor or graphics systems, and high-speed addition in
carry logic shares operand and control inputs with the func-
digital signal processing are two typical applications.
tion generators. The carry outputs connect to the function
The two 4-input function generators can be configured as a generators, where they are combined with the operands to
2-bit adder with built-in hidden carry that can be expanded form the sums.
to any length. This dedicated carry circuitry is so fast and
Figure 15 on page 23 shows the details of the carry logic
efficient that conventional speed-up methods like carry
for the XC4000E. This diagram shows the contents of the
generate/propagate are meaningless even at the 16-bit
box labeled “CARRY LOGIC” in Figure 14. The XC4000X
level, and of marginal benefit at the 32-bit level.
carry logic is very similar, but a multiplexer on the
This fast carry logic is one of the more significant features pass-through carry chain has been eliminated to reduce
of the XC4000 Series, speeding up arithmetic and counting delay. Additionally, in the XC4000X the multiplexer on the
into the 70 MHz range. G4 path has a memory-programmable 0 input, which per-

R
mits G4 to directly connect to COUT. G4 thus becomes an

additional high-speed initialization path for carry-in.
The dedicated carry logic is discussed in detail in Xilinx CLB CLB CLB CLB
document XAPP 013: “Using the Dedicated Carry Logic in
XC4000.” This discussion also applies to XC4000E
devices, and to XC4000X devices when the minor logic
changes are taken into account.
CLB CLB CLB CLB
The fast carry logic can be accessed by placing special
library symbols, or by using Xilinx Relationally Placed Mac-
ros (RPMs) that already include these symbols.
CLB CLB CLB CLB

CLB CLB CLB CLB
CLB CLB CLB CLB

CLB CLB CLB CLB
X6610
Figure 13: Available XC4000X Carry Propagation

Paths (dotted lines use general interconnect)
CLB CLB CLB CLB
6
CLB CLB CLB CLB
X6687
Figure 12: Available XC4000E Carry Propagation

Paths

R
C OUT C IN DOWN D IN
CARRY
LOGIC
Y
G H
CARRY
G4
G3
G
G2 DIN
H S/R
G D Q YQ
F
G1
EC
COUT0
H1 H
DIN
F H S/R
CARRY G D Q XQ
F
EC
F4
F3
F
F2
H
F1
X
F
CIN UP C OUT K S/R EC
X6699
Figure 14: Fast Carry Logic in XC4000E CLB (shaded area not present in XC4000X)

R
C OUT
M
G1
M
1
0 1 G2
I 0
G4
G3
C OUT0
TO
M FUNCTION
GENERATORS
F2
M
1
0 1
F1
M 0
F4
M 0 1
M 3
F3 1 M
M 0
M 1 0
C IN UP 6
X2000
C IN DOWN
Figure 15: Detail of XC4000E Dedicated Carry Logic
Input/Output Blocks (IOBs) The choice is made by placing the appropriate library sym-
bol. For example, IFD is the basic input flip-flop (rising edge
User-configurable input/output blocks (IOBs) provide the triggered), and ILD is the basic input latch (transpar-
interface between external package pins and the internal
ent-High). Variations with inverted clocks are available, and
logic. Each IOB controls one package pin and can be con-
some combinations of latches and flip-flops can be imple-
figured for input, output, or bidirectional signals. mented in a single IOB, as described in the XACT Libraries
Figure 16 shows a simplified block diagram of the Guide.
XC4000E IOB. A more complete diagram which includes The XC4000E inputs can be globally configured for either
the boundary scan logic of the XC4000E IOB can be found TTL (1.2V) or 5.0 volt CMOS thresholds, using an option in
in Figure 41 on page 46, in the “Boundary Scan” section. the bitstream generation software. There is a slight input
The XC4000X IOB contains some special features not hysteresis of about 300mV. The XC4000E output levels are
included in the XC4000E IOB. These features are high- also configurable; the two global adjustments of input
lighted in a simplified block diagram found in Figure 17, and threshold and output level are independent.
discussed throughout this section. When XC4000X special Inputs on the XC4000XL are TTL compatible and 3.3V
features are discussed, they are clearly identified in the
CMOS compatible. Outputs on the XC4000XL are pulled to
text. Any feature not so identified is present in both
the 3.3V positive supply.
XC4000E and XC4000X devices.
The inputs of XC4000 Series 5-Volt devices can be driven
IOB Input Signals by the outputs of any 3.3-Volt device, if the 5-Volt inputs are
Two paths, labeled I1 and I2 in Figure 16 and Figure 17, in TTL mode.
bring input signals into the array. Inputs also connect to an Supported sources for XC4000 Series device inputs are
input register that can be programmed as either an shown in Table 9.
edge-triggered flip-flop or a level-sensitive latch.

R
Passive
Slew Rate Pull-Up/
Control Pull-Down
T
Flip-Flop
D Q
Out Output
CE Buffer
Pad
Output
Clock
I1
Flip- Input
Flop/ Buffer
Latch
I2
Q D
Delay
Clock
Enable CE
Input
Clock
X6704
Figure 16: Simplified Block Diagram of XC4000E IOB
Slew Rate Passive

Pull-Up/
Control Pull-Down
T
Output MUX
0
1
Flip-Flop
Out D Q
Output
CE Buffer
Pad
Output Clock Input

Buffer
I1
Flip-Flop/
Latch
I2 Delay Delay
Q D
Q D
Latch
Clock Enable CE Fast G
Capture
Latch
Input Clock
X5984
Figure 17: Simplified Block Diagram of XC4000X IOB (shaded areas indicate differences from XC4000E)

R
Table 9: Supported Sources for XC4000 Series Device Optional Delay Guarantees Zero Hold Time
Inputs
The data input to the register can optionally be delayed by
XC4000E/EX XC4000XL several nanoseconds. With the delay enabled, the setup
Series Inputs Series Inputs time of the input flip-flop is increased so that normal clock
Source routing does not result in a positive hold-time requirement.
5 V, 5 V, 3.3 V
TTL CMOS CMOS A positive hold time requirement can lead to unreliable,
temperature- or processing-dependent operation.
Any device, Vcc = 3.3 V,
√ √ The input flip-flop setup time is defined between the data
CMOS outputs
Unreli measured at the device I/O pin and the clock input at the
XC4000 Series, Vcc = 5 V,
√ -able √ IOB (not at the clock pin). Any routing delay from the device
TTL outputs
Data clock pin to the clock input of the IOB must, therefore, be
Any device, Vcc = 5 V,
√ √ subtracted from this setup time to arrive at the real setup
TTL outputs (Voh ≤ 3.7 V)
time requirement relative to the device pins. A short speci-
Any device, Vcc = 5 V,
√ √ √ fied setup time might, therefore, result in a negative setup
CMOS outputs
time at the device pins, i.e., a positive hold-time require-
XC4000XL 5-Volt Tolerant I/Os ment.
The I/Os on the XC4000XL are fully 5-volt tolerant even When a delay is inserted on the data line, more clock delay
though the VCC is 3.3 volts. This allows 5 V signals to can be tolerated without causing a positive hold-time
directly connect to the XC4000XL inputs without damage, requirement. Sufficient delay eliminates the possibility of a
as shown in Table 9. In addition, the 3.3 volt VCC can be data hold-time requirement at the external pin. The maxi-
applied before or after 5 volt signals are applied to the I/Os. mum delay is therefore inserted as the default.
This makes the XC4000XL immune to power supply The XC4000E IOB has a one-tap delay element: either the
sequencing problems. delay is inserted (default), or it is not. The delay guarantees 6
a zero hold time with respect to clocks routed through any
Registered Inputs
of the XC4000E global clock buffers. (See “Global Nets and
The I1 and I2 signals that exit the block can each carry Buffers (XC4000E only)” on page 38 for a description of the
either the direct or registered input signal. global clock buffers in the XC4000E.) For a shorter input
The input and output storage elements in each IOB have a register setup time, with non-zero hold, attach a NODELAY
common clock enable input, which, through configuration, attribute or property to the flip-flop.
can be activated individually for the input or output flip-flop, The XC4000X IOB has a two-tap delay element, with
or both. This clock enable operates exactly like the EC pin choices of a full delay, a partial delay, or no delay. The
on the XC4000 Series CLB. It cannot be inverted within the attributes or properties used to select the desired delay are
IOB. shown in Table 11. The choices are no added attribute,
The storage element behavior is shown in Table 10. MEDDELAY, and NODELAY. The default setting, with no
added attribute, ensures no hold time with respect to any of
Table 10: Input Register Functionality the XC4000X clock buffers, including the Global Low-Skew
(active rising edge is shown) buffers. MEDDELAY ensures no hold time with respect to
Clock the Global Early buffers. Inputs with NODELAY may have a
Mode Clock D Q positive hold time with respect to all clock buffers. For a
Enable
description of each of these buffers, see “Global Nets and
Power-Up or X X X SR
Buffers (XC4000X only)” on page 40.
GSR
Flip-Flop __/ 1* D D Table 11: XC4000X IOB Input Delay Element
0 X X Q Value When to Use
Latch 1 1* X Q full delay Zero Hold with respect to Global
0 1* D D (default, no Low-Skew Buffer, Global Early Buffer
Both X 0 X Q attribute added)
Legend: MEDDELAY Zero Hold with respect to Global Early
X Don’t care Buffer
__/ Rising edge
SR Set or Reset value. Reset is default. NODELAY Short Setup, positive Hold time

R
Additional Input Latch for Fast Capture (XC4000X only) the desired delay based on the discussion in the previous
subsection.
The XC4000X IOB has an additional optional latch on the
input. This latch, as shown in Figure 17, is clocked by the IOB Output Signals
output clock — the clock used for the output flip-flop —
rather than the input clock. Therefore, two different clocks Output signals can be optionally inverted within the IOB,
can be used to clock the two input storage elements. This and can pass directly to the pad or be stored in an
additional latch allows the very fast capture of input data, edge-triggered flip-flop. The functionality of this flip-flop is
which is then synchronized to the internal clock by the IOB shown in Table 12.
flip-flop or latch. An active-High 3-state signal can be used to place the out-
To use this Fast Capture technique, drive the output clock put buffer in a high-impedance state, implementing 3-state
pin (the Fast Capture latching signal) from the output of one outputs or bidirectional I/O. Under configuration control, the
of the Global Early buffers supplied in the XC4000X. The output (OUT) and output 3-state (T) signals can be
second storage element should be clocked by a Global inverted. The polarity of these signals is independently con-
Low-Skew buffer, to synchronize the incoming data to the figured for each IOB.
internal logic. (See Figure 18.) These special buffers are The 4-mA maximum output current specification of many
described in “Global Nets and Buffers (XC4000X only)” on FPGAs often forces the user to add external buffers, which
page 40. are especially cumbersome on bidirectional I/O lines. The
The Fast Capture latch (FCL) is designed primarily for use XC4000E and XC4000EX/XL devices solve many of these
with a Global Early buffer. For Fast Capture, a single clock problems by providing a guaranteed output sink current of
signal is routed through both a Global Early buffer and a 12 mA. Two adjacent outputs can be interconnected exter-
Global Low-Skew buffer. (The two buffers share an input nally to sink up to 24 mA. The XC4000E and XC4000EX/XL
pad.) The Fast Capture latch is clocked by the Global Early FPGAs can thus directly drive buses on a printed circuit
buffer, and the standard IOB flip-flop or latch is clocked by board.
the Global Low-Skew buffer. This mode is the safest way to By default, the output pull-up structure is configured as a
use the Fast Capture latch, because the clock buffers on TTL-like totem-pole. The High driver is an n-channel pull-up
both storage elements are driven by the same pad. There is transistor, pulling to a voltage one transistor threshold
no external skew between clock pads to create potential below Vcc. Alternatively, the outputs can be globally config-
problems. ured as CMOS drivers, with p-channel pull-up transistors
To place the Fast Capture latch in a design, use one of the pulling to Vcc. This option, applied using the bitstream gen-
special library symbols, ILFFX or ILFLX. ILFFX is a trans- eration software, applies to all outputs on the device. It is
parent-Low Fast Capture latch followed by an active-High not individually programmable. In the XC4000XL, all out-
input flip-flop. ILFLX is a transparent-Low Fast Capture puts are pulled to the positive supply rail.
latch followed by a transparent-High input latch. Any of the
clock inputs can be inverted before driving the library ele- Table 12: Output Flip-Flop Functionality (active rising
ment, and the inverter is absorbed into the IOB. If a single edge is shown)
BUFG output is used to drive both clock inputs, the soft-
Clock
ware automatically runs the clock through both a Global
Mode Clock Enable T D Q
Low-Skew buffer and a Global Early buffer, and clocks the
Fast Capture latch appropriately. Power-Up X X 0* X SR
or GSR
Figure 17 on page 24 also shows a two-tap delay on the
X 0 0* X Q
input. By default, if the Fast Capture latch is used, the Xilinx
Flip-Flop __/ 1* 0* D D
software assumes a Global Early buffer is driving the clock,
and selects MEDDELAY to ensure a zero hold time. Select X X 1 X Z
0 X 0* X Q
ILFFX Legend:
X Don’t care
IPAD D Q to internal __/ Rising edge
logic
SR Set or Reset value. Reset is default.
GF 0* Input is Low or unconnected (default value)
BUFGE
CE Z 3-state
C
IPAD
BUFGLS
X9013
Figure 18: Examples Using XC4000X FCL

R
Any XC4000 Series 5-Volt device with its outputs config- Power/Ground pin pairs are connected to special Power
ured in TTL mode can drive the inputs of any typical and Ground planes within the packages, to reduce ground
3.3-Volt device. (For a detailed discussion of how to inter- bounce. Therefore, the maximum total capacitive load is
face between 5 V and 3.3 V devices, see the 3V Products 300 pF between each external Power/Ground pin pair.
section of The Programmable Logic Data Book.) Maximum loading may vary for the low-voltage devices.
Supported destinations for XC4000 Series device outputs For slew-rate limited outputs this total is two times larger for
are shown in Table 13. each device type: 400 pF for XC4000E devices and 600 pF
An output can be configured as open-drain (open-collector) for XC4000X devices. This maximum capacitive load
by placing an OBUFT symbol in a schematic or HDL code, should not be exceeded, as it can result in ground bounce
of greater than 1.5 V amplitude and more than 5 ns dura-
then tying the 3-state pin (T) to the output signal, and the
tion. This level of ground bounce may cause undesired
input pin (I) to Ground. (See Figure 19.)
transient behavior on an output, or in the internal logic. This
Table 13: Supported Destinations for XC4000 Series restriction is common to all high-speed digital ICs, and is
Outputs not particular to Xilinx or the XC4000 Series.
XC4000 Series XC4000 Series devices have a feature called “Soft
Outputs Start-up,” designed to reduce ground bounce when all out-
Destination 3.3 V, 5 V, 5 V, puts are turned on simultaneously at the end of configura-
CMOS TTL CMOS tion. When the configuration process is finished and the
Any typical device, Vcc = 3.3 V, √ √ some1 device starts up, the first activation of the outputs is auto-
CMOS-threshold inputs matically slew-rate limited. Immediately following the initial
activation of the I/O, the slew rate of the individual outputs
Any device, Vcc = 5 V, √ √ √
is determined by the individual configuration option for each
TTL-threshold inputs
IOB.
Any device, Vcc = 5 V, Unreliable √ 6
CMOS-threshold inputs Data Global Three-State
1. Only if destination device has 5-V tolerant inputs A separate Global 3-State line (not shown in Figure 16 or
Figure 17) forces all FPGA outputs to the high-impedance
state, unless boundary scan is enabled and is executing an
EXTEST instruction. This global net (GTS) does not com-
OPAD pete with other routing resources; it uses a dedicated distri-
OBUFT bution network.
X6702
Figure 19: Open-Drain Output GTS can be driven from any user-programmable pin as a
global 3-state input. To use this global net, place an input
Output Slew Rate pad and input buffer in the schematic or HDL code, driving
the GTS pin of the STARTUP symbol. A specific pin loca-
The slew rate of each output buffer is, by default, reduced, tion can be assigned to this input using a LOC attribute or
to minimize power bus transients when switching non-criti- property, just as with any other user-programmable pad. An
cal signals. For critical signals, attach a FAST attribute or inverter can optionally be inserted after the input buffer to
property to the output buffer or flip-flop. invert the sense of the Global 3-State signal. Using GTS is
For XC4000E devices, maximum total capacitive load for similar to GSR. See Figure 3 on page 13 for details.
simultaneous fast mode switching in the same direction is Alternatively, GTS can be driven from any internal node.
200 pF for all package pins between each Power/Ground
pin pair. For XC4000X devices, additional internal

R
Output Multiplexer/2-Input Function Generator Other IOB Options

(XC4000X only)
There are a number of other programmable options in the
As shown in Figure 17 on page 24, the output path in the XC4000 Series IOB.
XC4000X IOB contains an additional multiplexer not avail-
able in the XC4000E IOB. The multiplexer can also be con- Pull-up and Pull-down Resistors
figured as a 2-input function generator, implementing a Programmable pull-up and pull-down resistors are useful
pass-gate, AND-gate, OR-gate, or XOR-gate, with 0, 1, or 2 for tying unused pins to Vcc or Ground to minimize power
inverted inputs. The logic used to implement these func- consumption and reduce noise sensitivity. The configurable
tions is shown in the upper gray area of Figure 17. pull-up resistor is a p-channel transistor that pulls to Vcc.
When configured as a multiplexer, this feature allows two The configurable pull-down resistor is an n-channel transis-
output signals to time-share the same output pad; effec- tor that pulls to Ground.
tively doubling the number of device outputs without requir- The value of these resistors is 50 kΩ − 100 kΩ. This high
ing a larger, more expensive package. value makes them unsuitable as wired-AND pull-up resis-
When the MUX is configured as a 2-input function generators.
tor, logic can be implemented within the IOB itself. Com- The pull-up resistors for most user-programmable IOBs are
bined with a Global Early buffer, this arrangement allows active during the configuration process. See Table 23 on
very high-speed gating of a single signal. For example, a page 61 for a list of pins with pull-ups active before and dur-
wide decoder can be implemented in CLBs, and its output ing configuration.
gated with a Read or Write Strobe Driven by a BUFGE
After configuration, voltage levels of unused pads, bonded
buffer, as shown in Figure 20. The critical-path pin-to-pin
delay of this circuit is less than 6 nanoseconds. or unbonded, must be valid logic levels, to reduce noise
sensitivity and avoid excess current. Therefore, by default,
As shown in Figure 17, the IOB input pins Out, Output unused pads are configured with the internal pull-up resis-
Clock, and Clock Enable have different delays and different tor active. Alternatively, they can be individually configured
flexibilities regarding polarity. Additionally, Output Clock with the pull-down resistor, or as a driven output, or to be
sources are more limited than the other inputs. Therefore, driven by an external source. To activate the internal
the Xilinx software does not move logic into the IOB func- pull-up, attach the PULLUP library component to the net
tion generators unless explicitly directed to do so. attached to the pad. To activate the internal pull-down,
The user can specify that the IOB function generator be attach the PULLDOWN library component to the net
used, by placing special library symbols beginning with the attached to the pad.
letter “O.” For example, a 2-input AND-gate in the IOB func-
Independent Clocks
tion generator is called OAND2. Use the symbol input pin
labelled “F” for the signal on the critical path. This signal is Separate clock signals are provided for the input and output
placed on the OK pin — the IOB input with the shortest flip-flops. The clock can be independently inverted for each
delay to the function generator. Two examples are shown in flip-flop within the IOB, generating either falling-edge or ris-
Figure 21. ing-edge triggered flip-flops. The clock inputs for each IOB
are independent, except that in the XC4000X, the Fast
Capture latch shares an IOB input with the output clock pin.
IPAD
BUFGE Early Clock for IOBs (XC4000X only)
F Special early clocks are available for IOBs. These clocks
OPAD
from are sourced by the same sources as the Global Low-Skew
internal FAST
OAND2
buffers, but are separately buffered. They have fewer loads
logic
and therefore less delay. The early clock can drive either
X9019
the IOB output clock or the IOB input clock, or both. The
Figure 20: Fast Pin-to-Pin Path in XC4000X early clock allows fast capture of input data, and fast
clock-to-output on output data. The Global Early buffers
that drive these clocks are described in “Global Nets and
OMUX2 Buffers (XC4000X only)” on page 40.
F D0
O
D1 Global Set/Reset
OAND2 S0 As with the CLB registers, the Global Set/Reset signal
X6598
X6599 (GSR) can be used to set or clear the input and output reg-
Figure 21: AND & MUX Symbols in XC4000X IOB isters, depending on the value of the INIT attribute or prop-
erty. The two flip-flops can be individually configured to set

R
or clear on reset and after configuration. Other than the glo- Standard 3-State Buffer
bal GSR net, no user-controlled set/reset signal is available
All three pins are used. Place the library element BUFT.
to the I/O flip-flops. The choice of set or clear applies to
Connect the input to the I pin and the output to the O pin.
both the initial state of the flip-flop and the response to the The T pin is an active-High 3-state (i.e. an active-Low
Global Set/Reset pulse. See “Global Set/Reset” on enable). Tie the T pin to Ground to implement a standard
page 13 for a description of how to use GSR. buffer.
JTAG Support Wired-AND with Input on the I Pin
Embedded logic attached to the IOBs contains test struc- The buffer can be used as a Wired-AND. Use the WAND1
tures compatible with IEEE Standard 1149.1 for boundary library symbol, which is essentially an open-drain buffer.
scan testing, permitting easy chip and board-level testing. WAND4, WAND8, and WAND16 are also available. See the
More information is provided in “Boundary Scan” on
XACT Libraries Guide for further information.
page 45.
The T pin is internally tied to the I pin. Connect the input to
Three-State Buffers the I pin and the output to the O pin. Connect the outputs of
all the WAND1s together and attach a PULLUP symbol.
A pair of 3-state buffers is associated with each CLB in the
array. (See Figure 28 on page 33.) These 3-state buffers Wired OR-AND
can be used to drive signals onto the nearest horizontal
longlines above and below the CLB. They can therefore be The buffer can be configured as a Wired OR-AND. A High
used to implement multiplexed or bidirectional buses on the level on either input turns off the output. Use the
WOR2AND library symbol, which is essentially an
horizontal longlines, saving logic resources. Programmable
open-drain 2-input OR gate. The two input pins are func-
pull-up resistors attached to these longlines help to imple-
ment a wide wired-AND function. tionally equivalent. Attach the two inputs to the I0 and I1
pins and tie the output to the O pin. Tie the outputs of all the 6
The buffer enable is an active-High 3-state (i.e. an WOR2ANDs together and attach a PULLUP symbol.
active-Low enable), as shown in Table 14.
Three-State Buffer Examples
Another 3-state buffer with similar access is located near
each I/O block along the right and left edges of the array. Figure 22 shows how to use the 3-state buffers to imple-
(See Figure 34 on page 37.) ment a wired-AND function. When all the buffer inputs are
High, the pull-up resistor(s) provide the High output.
The horizontal longlines driven by the 3-state buffers have
a weak keeper at each end. This circuit prevents undefined Figure 23 shows how to use the 3-state buffers to imple-
floating levels. However, it is overridden by any driver, even ment a multiplexer. The selection is accomplished by the
a pull-up resistor. buffer 3-state signal.
Special longlines running along the perimeter of the array Pay particular attention to the polarity of the T pin when
can be used to wire-AND signals coming from nearby IOBs using these buffers in a design. Active-High 3-state (T) is
or from internal longlines. These longlines form the wide identical to an active-Low output enable, as shown in
edge decoders discussed in “Wide Edge Decoders” on Table 14.
page 30. Table 14: Three-State Buffer Functionality
Three-State Buffer Modes IN T OUT
The 3-state buffers can be configured in three modes: X 1 Z
• Standard 3-state buffer IN 0 IN
• Wired-AND with input on the I pin
• Wired OR-AND
P
Z=D ●D ● (D +D ) ● (D +D ) U
A B C D E F U
L P
L
D D
C E
D D D D
A B D F
WAND1 WAND1
WOR2AND WOR2AND
X6465
Figure 22: Open-Drain Buffers Implement a Wired-AND Function

R
Z = DA • A + DB • B + DC • C + DN • N
~100 kΩ
DA DB DC DN
BUFT BUFT BUFT BUFT
A B C N
X6466
"Weak Keeper"
Wide Edge Decoders

Dedicated decoder circuitry boosts the performance of LUP symbol. Location attributes or properties such as L
wide decoding functions. When the address or data field is (left edge) or TR (right half of top edge) should also be used
wider than the function generator inputs, FPGAs need to ensure the correct placement of the decoder inputs.
multi-level decoding and are thus slower than PALs.
XC4000 Series CLBs have nine inputs. Any decoder of up INTERCONNECT
to nine inputs is, therefore, compact and fast. However,
there is also a need for much wider decoders, especially for IOB IOB
address decoding in large microprocessor systems. .I1 .I1
An XC4000 Series FPGA has four programmable decoders A C B
located on each edge of the device. The inputs to each

decoder are any of the IOB I1 signals on that edge plus one
local interconnect per CLB row or column. Each row or col-
umn of CLBs provides up to three variables or their compli- ( C) .....
ments., as shown in Figure 24. Each decoder generates a (A • B • C) .....
High output (resistor pull-up) when the AND condition of (A • B • C) .....
the selected inputs, or their complements, is true. This is (A • B • C) .....
analogous to a product term in typical PAL devices.
X2627
Each of these wired-AND gates is capable of accepting up
Figure 24: XC4000 Series Edge Decoding Example
to 42 inputs on the XC4005E and 72 on the XC4013E.
There are up to 96 inputs for each decoder on the
XC4028X and 132 on the XC4052X. The decoders may
also be split in two when a larger number of narrower
OSC4
decoders are required, for a maximum of 32 decoders per F8M
device. F500K
F16K
The decoder outputs can drive CLB inputs, so they can be
F490
combined with other logic to form a PAL-like AND/OR struc-
ture. The decoder outputs can also be routed directly to the F15
chip outputs. For fastest speed, the output should be on the X6703
same chip edge as the decoder. Very large PALs can be

Figure 25: XC4000 Series Oscillator Symbol
emulated by ORing the decoder outputs in a CLB. This
decoding feature covers what has long been considered a
weakness of older FPGAs. Users often resorted to external On-Chip Oscillator
PALs for simple but fast decoding functions. Now, the dedi- XC4000 Series devices include an internal oscillator. This
cated decoders in the XC4000 Series device can imple- oscillator is used to clock the power-on time-out, for config-
ment these functions fast and efficiently. uration memory clearing, and as the source of CCLK in
To use the wide edge decoders, place one or more of the Master configuration modes. The oscillator runs at a nomi-
WAND library symbols (WAND1, WAND4, WAND8, nal 8 MHz frequency that varies with process, Vcc, and
WAND16). Attach a DECODE attribute or property to each temperature. The output frequency falls between 4 and 10
WAND symbol. Tie the outputs together and attach a PUL- MHz.

R
The oscillator output is optionally available after configura- • Global routing consists of dedicated networks primarily
tion. Any two of four resynchronized taps of a built-in divider designed to distribute clocks throughout the device with
are also available. These taps are at the fourth, ninth, four- minimum delay and skew. Global routing can also be
teenth and nineteenth bits of the divider. Therefore, if the used for other high-fanout signals.
primary oscillator output is running at the nominal 8 MHz, Five interconnect types are distinguished by the relative
the user has access to an 8 MHz clock, plus any two of 500 length of their segments: single-length lines, double-length
kHz, 16kHz, 490Hz and 15Hz (up to 10% lower for low-volt- lines, quad and octal lines (XC4000X only), and longlines.
age devices). These frequencies can vary by as much as
In the XC4000X, direct connects allow fast data flow
-50% or +25%.
between adjacent CLBs, and between IOBs and CLBs.
These signals can be accessed by placing the OSC4
Extra routing is included in the IOB pad ring. The XC4000X
library element in a schematic or in HDL code (see
also includes a ring of octal interconnect lines near the
Figure 25). IOBs to improve pin-swapping and routing to locked pins.
The oscillator is automatically disabled after configuration if
XC4000E/X devices include two types of global buffers.
the OSC4 symbol is not used in the design. These global buffers have different properties, and are
intended for different purposes. They are discussed in
Programmable Interconnect detail later in this section.
All internal connections are composed of metal segments
with programmable switching points and switching matrices CLB Routing Connections
to implement the desired routing. A structured, hierarchical A high-level diagram of the routing resources associated
matrix of routing resources is provided to achieve efficient with one CLB is shown in Figure 26. The shaded arrows
automated routing. represent routing present only in XC4000X devices.
The XC4000E and XC4000X share a basic interconnect Table 15 shows how much routing of each type is available
structure. XC4000X devices, however, have additional rout- 6
in XC4000E and XC4000X CLB arrays. Clearly, very large
ing not available in the XC4000E. The extra routing designs, or designs with a great deal of interconnect, will
resources allow high utilization in high-capacity devices. All route more easily in the XC4000X. Smaller XC4000E
XC4000X-specific routing resources are clearly identified designs, typically requiring significantly less interconnect,
throughout this section. Any resources not identified as do not require the additional routing.
XC4000X-specific are present in all XC4000 Series
devices. Figure 28 on page 33 is a detailed diagram of both the
XC4000E and the XC4000X CLB, with associated routing.
This section describes the varied routing resources avail- The shaded square is the programmable switch matrix,
able in XC4000 Series devices. The implementation soft- present in both the XC4000E and the XC4000X. The
ware automatically assigns the appropriate resources L-shaped shaded area is present only in XC4000X devices.
based on the density and timing requirements of the As shown in the figure, the XC4000X block is essentially an
design. XC4000E block with additional routing.
Interconnect Overview CLB inputs and outputs are distributed on all four sides,
providing maximum routing flexibility. In general, the entire
There are several types of interconnect.
architecture is symmetrical and regular. It is well suited to
• CLB routing is associated with each row and column of established placement and routing algorithms. Inputs, out-
the CLB array. puts, and function generators can freely swap positions
• IOB routing forms a ring (called a VersaRing) around within a CLB to avoid routing congestion during the place-
the outside of the CLB array. It connects the I/O with the ment and routing operation.
internal logic blocks.

R
Quad
Single
Double
Long
Direct
CLB Connect
Long
Quad Long Global Long Double Single Global Carry Direct

Clock Clock Chain Connect
x5994
Figure 26: High-Level Routing Diagram of XC4000 Series CLB (shaded arrows indicate XC4000X only)
Table 15: Routing per CLB in XC4000 Series Devices
e
es
e
bl
bl
gl
ou
XC4000E XC4000X
ou
in
D
S
D
Vertical Horizontal Vertical Horizontal
Singles 8 8 8 8
Double
Doubles 4 4 4 4
Quads 0 0 12 12
Singles
Longlines 6 6 10 6
Six Pass Transistors
Direct 0 0 2 2 Per Switch Matrix
Interconnect Point
Connects Double
Globals 4 0 8 0
X6600
Carry Logic 2 0 1 0
Total 24 18 45 32 Figure 27: Programmable Switch Matrix (PSM)
Programmable Switch Matrices Single-Length Lines

The horizontal and vertical single- and double-length lines Single-length lines provide the greatest interconnect flexi-
intersect at a box called a programmable switch matrix bility and offer fast routing between adjacent blocks. There
(PSM). Each switch matrix consists of programmable pass are eight vertical and eight horizontal single-length lines
transistors used to establish connections between the lines associated with each CLB. These lines connect the switch-
(see Figure 27). ing matrices that are located in every row and a column of
CLBs.
For example, a single-length signal entering on the right
side of the switch matrix can be routed to a single-length Single-length lines are connected by way of the program-
line on the top, left, or bottom sides, or any combination mable switch matrices, as shown in Figure 29. Routing
thereof, if multiple branches are required. Similarly, a dou- connectivity is shown in Figure 28.
ble-length signal can be routed to a double-length line on Single-length lines incur a delay whenever they go through
any or all of the other three edges of the programmable a switching matrix. Therefore, they are not suitable for rout-
switch matrix. ing signals for long distances. They are normally used to
conduct signals within a localized area and to provide the
branching for nets with fanout greater than one.

R
QUAD
DOUBLE
SINGLE
DOUBLE
6
LONG
F4 C4 G4
YQ
Y DIRECT
G1
C1
F1
CLB G3
C3 FEEDBACK
F3
K
X
XQ
F2 C2 G2
LONG
LO G LO D D LO G D
LO O O LO IR
Q N N U SI U N BA EC FE
U G BA G BL N BL G
AD L E G E L T ED
LE BA
C
K
Common to XC4000E and XC4000X
XC4000X only
Programmable Switch Matrix
Figure 28: Detail of Programmable Interconnect Associated with XC4000 Series CLB

R
CLB CLB CLB
Doubles
CLB CLB CLB
PSM PSM Singles
Doubles
CLB CLB CLB

CLB CLB CLB
PSM PSM
CLB CLB CLB CLB CLB CLB
X6601
Figure 29: Single- and Double-Length Lines, with X9014
Programmable Switch Matrices (PSMs) Figure 30: Quad Lines (XC4000X only)
Double-Length Lines and up to two independent outputs. Only one of the inde-
The double-length lines consist of a grid of metal segments, pendent inputs can be buffered.
each twice as long as the single-length lines: they run past The place and route software automatically uses the timing
two CLBs before entering a switch matrix. Double-length requirements of the design to determine whether or not a
lines are grouped in pairs with the switch matrices stag- quad line signal should be buffered. A heavily loaded signal
gered, so that each line goes through a switch matrix at is typically buffered, while a lightly loaded one is not. One
every other row or column of CLBs (see Figure 29). scenario is to alternate buffers and pass transistors. This
There are four vertical and four horizontal double-length allows both vertical and horizontal quad lines to be buffered
lines associated with each CLB. These lines provide faster at alternating buffered switch matrices.
signal routing over intermediate distances, while retaining Due to the buffered switch matrices, quad lines are very
routing flexibility. Double-length lines are connected by way fast. They provide the fastest available method of routing
of the programmable switch matrices. Routing connectivity heavily loaded signals for long distances across the device.
is shown in Figure 28.
Longlines
Quad Lines (XC4000X only)
Longlines form a grid of metal interconnect segments that
XC4000X devices also include twelve vertical and twelve run the entire length or width of the array. Longlines are
horizontal quad lines per CLB row and column. Quad lines intended for high fan-out, time-critical signal nets, or nets
are four times as long as the single-length lines. They are that are distributed over long distances. In XC4000X
interconnected via buffered switch matrices (shown as dia- devices, quad lines are preferred for critical nets, because
monds in Figure 28 on page 33). Quad lines run past four the buffered switch matrices make them faster for high
CLBs before entering a buffered switch matrix. They are fan-out nets.
grouped in fours, with the buffered switch matrices stag-
gered, so that each line goes through a buffered switch Two horizontal longlines per CLB can be driven by 3-state
matrix at every fourth CLB location in that row or column. or open-drain drivers (TBUFs). They can therefore imple-
(See Figure 30.) ment unidirectional or bidirectional buses, wide multiplex-
ers, or wired-AND functions. (See “Three-State Buffers” on
The buffered switch matrixes have four pins, one on each page 29 for more details.)
edge. All of the pins are bidirectional. Any pin can drive any
or all of the other pins. Each horizontal longline driven by TBUFs has either two
(XC4000E) or eight (XC4000X) pull-up resistors. To acti-
Each buffered switch matrix contains one buffer and six vate these resistors, attach a PULLUP symbol to the
pass transistors. It resembles the programmable switch long-line net. The software automatically activates the
matrix shown in Figure 27, with the addition of a program- appropriate number of pull-ups. There is also a weak
mable buffer. There can be up to two independent inputs keeper at each end of these two horizontal longlines. This

R
circuit prevents undefined floating levels. However, it is I/O Routing

overridden by any driver, even a pull-up resistor.
XC4000 Series devices have additional routing around the
Each XC4000E longline has a programmable splitter switch IOB ring. This routing is called a VersaRing. The VersaRing
at its center, as does each XC4000X longline driven by facilitates pin-swapping and redesign without affecting
TBUFs. This switch can separate the line into two indepen- board layout. Included are eight double-length lines span-
dent routing channels, each running half the width or height ning two CLBs (four IOBs), and four longlines. Global lines
of the array. and Wide Edge Decoder lines are provided. XC4000X
Each XC4000X longline not driven by TBUFs has a buff- devices also include eight octal lines.
ered programmable splitter switch at the 1/4, 1/2, and 3/4 A high-level diagram of the VersaRing is shown in
points of the array. Due to the buffering, XC4000X longline Figure 32. The shaded arrows represent routing present
performance does not deteriorate with the larger array only in XC4000X devices.
sizes. If the longline is split, the resulting partial longlines
Figure 34 on page 37 is a detailed diagram of the XC4000E
are independent.
and XC4000X VersaRing. The area shown includes two
Routing connectivity of the longlines is shown in Figure 28 IOBs. There are two IOBs per CLB row or column, there-
on page 33. fore this diagram corresponds to the CLB routing diagram
shown in Figure 28 on page 33. The shaded areas repre-
Direct Interconnect (XC4000X only)
sent routing and routing connections present only in
The XC4000X offers two direct, efficient and fast connec- XC4000X devices.
tions between adjacent CLBs. These nets facilitate a data
flow from the left to the right side of the device, or from the Octal I/O Routing (XC4000X only)
top to the bottom, as shown in Figure 31. Signals routed on Between the XC4000X CLB array and the pad ring, eight
the direct interconnect exhibit minimum interconnect prop- interconnect tracks provide for versatility in pin assignment
agation delay and use no general routing resources. and fixed pinout flexibility. (See Figure 33 on page 36.) 6
The direct interconnect is also present between CLBs and These routing tracks are called octals, because they can be
adjacent IOBs. Each IOB on the left and top device edges broken every eight CLBs (sixteen IOBs) by a programma-
has a direct path to the nearest CLB. Each CLB on the right ble buffer that also functions as a splitter switch. The buffers
and bottom edges of the array has a direct path to the near- are staggered, so each line goes through a buffer at every
est two IOBs, since there are two IOBs for each row or col- eighth CLB location around the device edge.
umn of CLBs.
The octal lines bend around the corners of the device. The
The place and route software uses direct interconnect lines cross at the corners in such a way that the segment
whenever possible, to maximize routing resources and min- most recently buffered before the turn has the farthest dis-
imize interconnect delays. tance to travel before the next buffer, as shown in
Figure 33.
IOB
IOB
IOB
IOB
IOB
IOB
~ ~
~
IOB IOB
CLB CLB CLB
~
IOB IOB
~ ~
~ ~ ~ ~
~ ~ ~ ~
~ ~
~ ~
~
IOB IOB
CLB CLB CLB
~
IOB IOB
IOB
IOB
IOB
IOB
IOB
IOB
X6603
Figure 31: XC4000X Direct Interconnect

R
WED
IOB
Quad
WED
Single
Double
INTERCONNECT
Long
Direct
Connect
Long
IOB
WED
Direct Edge Double Long Global Octal

Connect Decode Clock
X5995
Figure 32: High-Level Routing Diagram of XC4000 Series VersaRing (Left Edge)
WED = Wide Edge Decoder, IOB = I/O Block (shaded arrows indicate XC4000X only)
IOB IOB
IOB IOB
Segment with nearest buffer

connects to segment with furthest buffer
X9015
Figure 33: XC4000X Octal I/O Routing

R
QUAD
T
O
DOUBLE
C
SINGLE L
B
DOUBLE
6
LONG
A
R
R
IOB
A
DECODER
I1 I2
IK
OK
T
CE
O
DIRECT
Y
DECODER
IOB
T O
OK CE
DECODER
IK
I1 I2
LONG
G
LO
ED EC
LO
D
N
G OD
D
O
BA
G
E E
O
C
U
TA
L
BL
L
E
Common to XC4000E and XC4000X
XC4000X only
Figure 34: Detail of Programmable Interconnect Associated with XC4000 Series IOB (Left Edge)

R
IOB inputs and outputs interface with the octal lines via the Two different types of clock buffers are available in the
single-length interconnect lines. Single-length lines are XC4000E:
also used for communication between the octals and dou-
• Primary Global Buffers (BUFGP)
ble-length lines, quads, and longlines within the CLB array.
• Secondary Global Buffers (BUFGS)
Segmentation into buffered octals was found to be optimal
Four Primary Global buffers offer the shortest delay and
for distributing signals over long distances around the
negligible skew. Four Secondary Global buffers have
device.
slightly longer delay and slightly more skew due to poten-
tially heavier loading, but offer greater flexibility when used
Global Nets and Buffers
to drive non-clock CLB inputs.
Both the XC4000E and the XC4000X have dedicated glo-
The Primary Global buffers must be driven by the
bal networks. These networks are designed to distribute
semi-dedicated pads. The Secondary Global buffers can
clocks and other high fanout control signals throughout the
be sourced by either semi-dedicated pads or internal nets.
devices with minimal skew. The global buffers are
described in detail in the following sections. The text Each CLB column has four dedicated vertical Global lines.
descriptions and diagrams are summarized in Table 16. Each of these lines can be accessed by one particular Pri-
The table shows which CLB and IOB clock pins can be mary Global buffer, or by any of the Secondary Global buff-
sourced by which global buffers. ers, as shown in Figure 35. Each corner of the device has
one Primary buffer and one Secondary buffer.
In both XC4000E and XC4000X devices, placement of a
library symbol called BUFG results in the software choos- IOBs along the left and right edges have four vertical global
ing the appropriate clock buffer, based on the timing longlines. Top and bottom IOBs can be clocked from the
requirements of the design. The detailed information in global lines in the adjacent CLB column.
these sections is included only for reference. A global buffer should be specified for all timing-sensitive
Global Nets and Buffers (XC4000E only) global signal distribution. To use a global buffer, place a
BUFGP (primary buffer), BUFGS (secondary buffer), or
Four vertical longlines in each CLB column are driven BUFG (either primary or secondary buffer) element in a
exclusively by special global buffers. These longlines are schematic or in HDL code. If desired, attach a LOC
in addition to the vertical longlines used for standard inter- attribute or property to direct placement to the designated
connect. The four global lines can be driven by either of two location. For example, attach a LOC=L attribute or property
types of global buffers. The clock pins of every CLB and to a BUFGS symbol to direct that a buffer be placed in one
IOB can also be sourced from local interconnect. of the two Secondary Global buffers on the left edge of the
device, or a LOC=BL to indicate the Secondary Global
buffer on the bottom edge of the device, on the left.
Table 16: Clock Pin Access
XC4000E XC4000X Local

L&R T&B Inter-
BUFGP BUFGS BUFGLS
BUFGE BUFGE connect
All CLBs in Quadrant √ √ √ √ √ √
All CLBs in Device √ √ √ √
IOBs on Adjacent Vertical √ √ √ √ √ √
Half Edge
IOBs on Adjacent Vertical √ √ √ √ √
Full Edge
IOBs on Adjacent Horizontal √ √
Half Edge (Direct)
IOBs on Adjacent Horizontal √ √ √ √ √ √
Half Edge (through CLB globals)
IOBs on Adjacent Horizontal √ √ √ √
Full Edge (through CLB globals)
L = Left, R = Right, T = Top, B = Bottom

R
IOB IOB IOB IOB
locals
locals
locals
locals
BUFGS BUFGP
PGCK1 SGCK4
SGCK1 PGCK4
4
4
BUFGP BUFGS
4
4 locals locals
CLB CLB
IOB IOB
locals locals
X4 Any BUFGS X4 X4 Any BUFGS X4
locals locals
One BUFGP One BUFGP
IOB per Global Line per Global Line IOB
locals CLB CLB locals
BUFGS BUFGP
PGCK2 SGCK3
SGCK2 PGCK3
locals
locals
locals
locals
BUFGP BUFGS
IOB IOB IOB IOB X6604
Figure 35: XC4000E Global Net Distribution 6
BUFGLS IOB IOB IOB IOB BUFGLS
GCK1 GCK8 GCK7 GCK6
BUFGE BUFGE
locals
locals
locals
locals
BUFGLS BUFGE BUFGE BUFGLS
CLB CLB
X4 BUFGLS 8 X8 X8 8 BUFGLS X8
BUFGLS 8 locals locals 8 BUFGLS
locals locals
4 8
8 8
IOB CLB CLOCKS CLB CLOCKS IOB

IOB IOB
(PER COLUMN) (PER COLUMN)
CLOCKS CLOCKS
locals locals
locals locals
IOB CLB CLOCKS CLB CLOCKS IOB
IOB CLOCKS (PER COLUMN) (PER COLUMN) CLOCKS IOB
8 8
locals 4 8 locals
BUFGLS 8 locals locals 8 BUFGLS
BUFGLS 8 8 BUFGLS
X4 X8 X8 X8
CLB CLB
locals
locals
locals
locals
BUFGLS BUFGE BUFGE BUFGLS
BUFGE BUFGE
GCK2 GCK3 GCK4 GCK5
BUFGLS IOB IOB IOB IOB BUFGLS X9018
Figure 36: XC4000X Global Net Distribution

R
Global Nets and Buffers (XC4000X only) Choosing an XC4000X Clock Buffer
Eight vertical longlines in each CLB column are driven by The clocking structure of the XC4000X provides a large
special global buffers. These longlines are in addition to the variety of features. However, it can be simple to use, with-
vertical longlines used for standard interconnect. The glo- out understanding all the details. The software automati-
bal lines are broken in the center of the array, to allow faster cally handles clocks, along with all other routing, when the
distribution and to minimize skew across the whole array. appropriate clock buffer is placed in the design. In fact, if a
Each half-column global line has its own buffered multi- buffer symbol called BUFG is placed, rather than a specific
plexer, as shown in Figure 36. The top and bottom global type of buffer, the software even chooses the buffer most
lines cannot be connected across the center of the device, appropriate for the design. The detailed information in this
as this connection might introduce unacceptable skew. The section is provided for those users who want a finer level of
top and bottom halves of the global lines must be sepa- control over their designs.
rately driven — although they can be driven by the same If fine control is desired, use the following summary and
global buffer. Table 16 on page 38 to choose an appropriate clock buffer.
The eight global lines in each CLB column can be driven by • The simplest thing to do is to use a Global Low-Skew
either of two types of global buffers. They can also be buffer.
driven by internal logic, because they can be accessed by • If a faster clock path is needed, try a BUFG. The
single, double, and quad lines at the top, bottom, half, and software will first try to use a Global Low-Skew Buffer. If
quarter points. Consequently, the number of different timing requirements are not met, a faster buffer will
clocks that can be used simultaneously in an XC4000X automatically be used.
device is very large. • If a single quadrant of the chip is sufficient for the
There are four global lines feeding the IOBs at the left edge clocked logic, and the timing requires a faster clock than
of the device. IOBs along the right edge have eight global the Global Low-Skew buffer, use a Global Early buffer.
lines. There is a single global line along the top and bottom
Global Low-Skew Buffers
edges with access to the IOBs. All IOB global lines are bro-
ken at the center. They cannot be connected across the Each corner of the XC4000X device has two Global
center of the device, as this connection might introduce Low-Skew buffers. Any of the eight Global Low-Skew buff-
unacceptable skew. ers can drive any of the eight vertical Global lines in a col-
umn of CLBs. In addition, any of the buffers can drive any of
IOB global lines can be driven from two types of global buff-
the four vertical lines accessing the IOBs on the left edge of
ers, or from local interconnect. Alternatively, top and bottom
the device, and any of the eight vertical lines accessing the
IOBs can be clocked from the global lines in the adjacent
IOBs on the right edge of the device. (See Figure 37 on
CLB column.
page 41.)
Two different types of clock buffers are available in the
IOBs at the top and bottom edges of the device are
XC4000X:
accessed through the vertical Global lines in the CLB array,
• Global Low-Skew Buffers (BUFGLS) as in the XC4000E. Any Global Low-Skew buffer can,
• Global Early Buffers (BUFGE) therefore, access every IOB and CLB in the device.
Global Low-Skew Buffers are the standard clock buffers. The Global Low-Skew buffers can be driven by either
They should be used for most internal clocking, whenever a semi-dedicated pads or internal logic.
large portion of the device must be driven.
To use a Global Low-Skew buffer, instantiate a BUFGLS
Global Early Buffers are designed to provide a faster clock element in a schematic or in HDL code. If desired, attach a
access, but CLB access is limited to one-fourth of the LOC attribute or property to direct placement to the desig-
device. They also facilitate a faster I/O interface. nated location. For example, attach a LOC=T attribute or
Figure 36 is a conceptual diagram of the global net struc- property to direct that a BUFGLS be placed in one of the
ture in the XC4000X. two Global Low-Skew buffers on the top edge of the device,
or a LOC=TR to indicate the Global Low-Skew buffer on the
Global Early buffers and Global Low-Skew buffers share a top edge of the device, on the right.
single pad. Therefore, the same IPAD symbol can drive one
buffer of each type, in parallel. This configuration is particu-
larly useful when using the Fast Capture latches, as
described in “IOB Input Signals” on page 23. Paired Global
Early and Global Low-Skew buffers share a common input;
they cannot be driven by two different signals.

R
8 7 8 7
IOB IOB IOB IOB
1 6 1 6
I I I I
O CLB CLB O O CLB CLB O
B B B B
I I I I
O CLB CLB O O CLB CLB O
B B B B
2 5 2 5
IOB IOB IOB IOB
3 4 3 4
X6751
X6753
Figure 37: Any BUFGLS (GCK1 - GCK8) Can Figure 38: Left and Right BUFGEs Can Drive Any or
Drive Any or All Clock Inputs on the Device All Clock Inputs in Same Quadrant or Edge (GCK1 is
shown. GCK2, GCK5 and GCK6 are similar.)
Global Early Buffers
The left-side Global Early buffers can each drive two of the
Each corner of the XC4000X device has two Global Early
four vertical lines accessing the IOBs on the entire left edge
buffers. The primary purpose of the Global Early buffers is 6
of the device. The right-side Global Early buffers can each
to provide an earlier clock access than the potentially
drive two of the eight vertical lines accessing the IOBs on
heavily-loaded Global Low-Skew buffers. A clock source
the entire right edge of the device. (See Figure 38.)
applied to both buffers will result in the Global Early clock
edge occurring several nanoseconds earlier than the Glo- Each left and right Global Early buffer can also drive half of
bal Low-Skew buffer clock edge, due to the lighter loading. the IOBs along either the top or bottom edge of the device,
using a dedicated line that can only be accessed through
Global Early buffers also facilitate the fast capture of device
the Global Early buffers.
inputs, using the Fast Capture latches described in “IOB
Input Signals” on page 23. For Fast Capture, take a single The top and bottom Global Early buffers can drive half of
clock signal, and route it through both a Global Early buffer the IOBs along either the left or right edge of the device, as
and a Global Low-Skew buffer. (The two buffers share an shown in Figure 39. They can only access the top and bot-
input pad.) Use the Global Early buffer to clock the Fast tom IOBs via the CLB global lines.
Capture latch, and the Global Low-Skew buffer to clock the
normal input flip-flop or latch, as shown in Figure 18 on 8 7
page 26. IOB IOB
1 6
The Global Early buffers can also be used to provide a fast
Clock-to-Out on device output pins. However, an early clock I I
in the output flip-flop IOB must be taken into consideration O CLB CLB O
when calculating the internal clock speed for the design. B B
The Global Early buffers at the left and right edges of the
chip have slightly different capabilities than the ones at the
top and bottom. Refer to Figure 38, Figure 39, and I I
Figure 36 on page 39 while reading the following explana- O CLB CLB O
B B
tion.
Each Global Early buffer can access the eight vertical Glo- 2 5
bal lines for all CLBs in the quadrant. Therefore, only IOB IOB
one-fourth of the CLB clock pins can be accessed. This 3 4
X6747
restriction is in large part responsible for the faster speed of
the buffers, relative to the Global Low-Skew buffers. Figure 39: Top and Bottom BUFGEs Can Drive Any
or All Clock Inputs in Same Quadrant (GCK8 is
shown. GCK3, GCK4 and GCK7 are similar.)

R
The top and bottom Global Early buffers are about 1 ns

slower clock to out than the left and right Global Early buff- GND
ers.
Ground and
Vcc Ring for
The Global Early buffers can be driven by either semi-ded- I/O Drivers
icated pads or internal logic. They share pads with the Glo-
bal Low-Skew buffers, so a single net can drive both global
buffers, as described above.
To use a Global Early buffer, place a BUFGE element in a Vcc Vcc
schematic or in HDL code. If desired, attach a LOC
attribute or property to direct placement to the designated Logic
location. For example, attach a LOC=T attribute or property Power Grid
to direct that a BUFGE be placed in one of the two Global
Early buffers on the top edge of the device, or a LOC=TR to
indicate the Global Early buffer on the top edge of the
GND
device, on the right. X5422
Figure 40: XC4000 Series Power Distribution

Power Distribution
Power for the FPGA is distributed through a grid to achieve Pin Descriptions
high noise immunity and isolation between logic and I/O.
Inside the FPGA, a dedicated Vcc and Ground ring sur- There are three types of pins in the XC4000 Series
rounding the logic array provides power to the I/O drivers, devices:
as shown in Figure 40. An independent matrix of Vcc and • Permanently dedicated pins
Ground lines supplies the interior logic of the device. • User I/O pins that can have special functions
This power distribution grid provides a stable supply and • Unrestricted user-programmable I/O pins.
ground for all internal logic, providing the external package Before and during configuration, all outputs not used for the
power pins are all connected and appropriately decoupled. configuration process are 3-stated with a 50 kΩ - 100 kΩ
Typically, a 0.1 µF capacitor connected between each Vcc pull-up resistor.
pin and the board’s Ground plane will provide adequate
After configuration, if an IOB is unused it is configured as
decoupling.
an input with a 50 kΩ - 100 kΩ pull-up resistor.
Output buffers capable of driving/sinking the specified 12
XC4000 Series devices have no dedicated Reset input.
mA loads under specified worst-case conditions may be
Any user I/O can be configured to drive the Global
capable of driving/sinking up to 10 times as much current
Set/Reset net, GSR. See “Global Set/Reset” on page 13
under best case conditions.
for more information on GSR.
Noise can be reduced by minimizing external load capaci-
XC4000 Series devices have no Powerdown control input,
tance and reducing simultaneous output transitions in the
as the XC3000 and XC2000 families do. The
same direction. It may also be beneficial to locate heavily
XC3000/XC2000 Powerdown control also 3-stated all of the
loaded output buffers near the Ground pads. The I/O Block
device
output buffers have a slew-rate limited mode (default) which
I/O pins. For XC4000 Series devices, use the global 3-state
should be used where output rise and fall times are not
net, GTS, instead. This net 3-states all outputs, but does
speed-critical.
not place the device in low-power mode. See “IOB Output
Signals” on page 26 for more information on GTS.
Device pins for XC4000 Series devices are described in
Table 17. Pin functions during configuration for each of the
seven configuration modes are summarized in Table 23 on
page 61, in the “Configuration Timing” section.

R
I/O I/O
During After
Eight or more (depending on package) connections to the nominal +5 V supply voltage
VCC I I (+3.3 V for low-voltage devices). All must be connected, and each must be decoupled
with a 0.01 - 0.1 µF capacitor to Ground.
Eight or more (depending on package type) connections to Ground. All must be con-
GND I I
nected.
During configuration, Configuration Clock (CCLK) is an output in Master modes or Asyn-
chronous Peripheral mode, but is an input in Slave mode and Synchronous Peripheral
mode. After configuration, CCLK has a weak pull-up resistor and can be selected as the
CCLK I or O I
Readback Clock. There is no CCLK High or Low time restriction on XC4000 Series de-
vices, except during Readback. See “Violating the Maximum High and Low Time Spec-
ification for the Readback Clock” on page 59 for an explanation of this exception.
indicates the completion of the configuration process. As an input, a Low level on DONE
DONE I/O O can be configured to delay the global logic initialization and the enabling of outputs.
The optional pull-up resistor is selected as an option in the XACTstep program that cre-
ates the configuration bitstream. The resistor is included by default.
PROGRAM is an active Low input that forces the FPGA to clear its configuration mem- 6
finishes the current clear cycle and executes another complete clear cycle, before it
PROGRAM I I
goes into a WAIT state and releases INIT.
The PROGRAM pin has a permanent weak pull-up, so it need not be externally pulled
up to Vcc.
During Peripheral mode configuration, this pin indicates when it is appropriate to write
another byte of data into the FPGA. The same status is also available on D7 in Asyn-
RDY/BUSY O I/O chronous Peripheral mode, if a read operation is performed when the device is selected.
After configuration, RDY/BUSY is a user-programmable I/O pin.
RDY/BUSY is pulled High with a high-impedance pull-up prior to INIT going High.
During Master Parallel configuration, each change on the A0-A17 outputs (A0 - A21 for
XC4000X) is preceded by a rising edge on RCLK, a redundant output signal. RCLK is
RCLK O I/O
useful for clocked PROMs. It is rarely used during configuration. After configuration,
RCLK is a user-programmable I/O pin.
As Mode inputs, these pins are sampled after INIT goes High to determine the configu-
ration mode to be used. After configuration, M0 and M2 can be used as inputs, and M1
can be used as a 3-state output. These three pins have no associated input or output
registers.
I (M0), During configuration, these pins have weak pull-up resistors. For the most popular con-
M0, M1, M2 I O (M1), figuration mode, Slave Serial, the mode pins can thus be left unconnected. The three
I (M2) mode inputs can be individually configured with or without weak pull-up or pull-down re-
sistors. A pull-down resistor value of 4.7 kΩ is recommended.
These pins can only be used as inputs or outputs when called out by special schematic
definitions. To use these pins, place the library components MD0, MD1, and MD2 in-
stead of the usual pad symbols. Input or output buffers must still be used.
this pin is a 3-state output without a register, after configuration is completed.
TDO O O This pin can be user output only when called out by special schematic definitions. To
use this pin, place the library component TDO instead of the usual pad symbol. An out-
put buffer must still be used.

R
I/O I/O
During After
I/O
I or I
(JTAG)
In this case, they must be called out by special schematic definitions. To use these pins,
High During Configuration (HDC) is driven High until the I/O go active. It is available as
INIT I/O I/O
occurred. After the I/O go active, INIT is a user-programmable I/O pin.
Four Primary Global inputs each drive a dedicated internal global net with short delay
PGCK1 - and minimal skew. If not used to drive a global buffer, any of these pins is a user-pro-
PGCK4 Weak grammable I/O.
I or I/O
(XC4000E Pull-up The PGCK1-PGCK4 pins drive the four Primary Global Buffers. Any input pad symbol
only) connected directly to the input of a BUFGP symbol is automatically placed on one of
these pins.
Four Secondary Global inputs each drive a dedicated internal global net with short delay
SGCK1 - and minimal skew. These internal global nets can also be driven from internal logic. If
SGCK4 Weak not used to drive a global net, any of these pins is a user-programmable I/O pin.
I or I/O
(XC4000E Pull-up The SGCK1-SGCK4 pins provide the shortest path to the four Secondary Global Buff-
only) ers. Any input pad symbol connected directly to the input of a BUFGS symbol is auto-
matically placed on one of these pins.
Eight inputs can each drive a Global Low-Skew buffer. In addition, each can drive a Glo-
GCK1 - bal Early buffer. Each pair of global buffers can also be driven from internal logic, but
GCK8 Weak must share an input signal. If not used to drive a global buffer, any of these pins is a
I or I/O
(XC4000X Pull-up user-programmable I/O.
only) Any input pad symbol connected directly to the input of a BUFGLS or BUFGE symbol
is automatically placed on one of these pins.
FCLK1 - Four inputs can each drive a Fast Clock (FCLK) buffer which can deliver a clock signal
FCLK4 to any IOB clock input in the octant of the die served by the Fast Clock buffer. Two Fast
(XC4000XLA Weak Clock buffers serve the two IOB octants on the left side of the die and the other two Fast
I or I/O
and Pull-up Clock buffers serve the two IOB octants on the rigth side of the die. On each side of the
XC4000XV die, one Fast Clock buffer serves the upper octant and the other serves the lower octant.
only) If not used to drive a Fast Clock buffer, any of these pins is a user-programmable I/O.

R
I/O I/O
During After
These four inputs are used in Asynchronous Peripheral mode. The chip is selected
when CS0 is Low and CS1 is High. While the chip is selected, a Low on Write Strobe
(WS) loads the data present on the D0 - D7 inputs into the internal data buffer. A Low
CS0, CS1, on Read Strobe (RS) changes D7 into a status output — High if Ready, Low if Busy —
I I/O
WS, RS and drives D0 - D6 High.
In Express mode, CS1 is used as a serial-enable signal for daisy-chaining.
WS and RS should be mutually exclusive, but if both are Low simultaneously, the Write
Strobe overrides. After configuration, these are user-programmable I/O pins.
During Master Parallel configuration, these 18 output pins address the configuration
A0 - A17 O I/O
EPROM. After configuration, they are user-programmable I/O pins.
A18 - A21 During Master Parallel configuration with an XC4000X master, these 4 output pins add
(XC4000X O I/O 4 more bits to address the configuration EPROM. After configuration, they are user-pro-
only) grammable I/O pins. (See Master Parallel Configuration section for additional details.)
During Master Parallel and Peripheral configuration, these eight input pins receive con-
D0 - D7 I I/O
figuration data. After configuration, they are user-programmable I/O pins.
DIN I I/O input receiving data on the rising edge of CCLK. During Parallel configuration, DIN is
the D0 input. After configuration, DIN is a user-programmable I/O pin.
During configuration in any mode but Express mode, DOUT is the serial configuration 6
data output that can drive the DIN of daisy-chained slave FPGAs. DOUT data changes
on the falling edge of CCLK, one-and-a-half CCLK periods after it was received at the
DOUT O I/O DIN input.
In Express mode, DOUT is the status output that can drive the CS1 of daisy-chained
FPGAs, to enable and disable downstream devices.
Weak
Pull-up
tor (25 kΩ - 100 kΩ) that defines the logic level as High.
Boundary Scan of how to enable this circuitry are covered later in this sec-
tion.
The ‘bed of nails’ has been the traditional method of testing
electronic assemblies. This approach has become less By exercising these input signals, the user can serially load
appropriate, due to closer pin spacing and more sophisti- commands and data into these devices to control the driv-
cated assembly methods like surface-mount technology ing of their outputs and to examine their inputs. This
and multi-layer boards. The IEEE Boundary Scan Standard method is an improvement over bed-of-nails testing. It
1149.1 was developed to facilitate board-level testing of avoids the need to over-drive device outputs, and it reduces
electronic assemblies. Design and test engineers can the user interface to four pins. An optional fifth pin, a reset
imbed a standard test logic structure in their device to for the control logic, is described in the standard but is not
achieve high fault coverage for I/O and internal logic. This implemented in Xilinx devices.
structure is easily implemented with a four-pin interface on The dedicated on-chip logic implementing the IEEE 1149.1
any boundary scan-compatible IC. IEEE 1149.1-compati- functions includes a 16-state machine, an instruction regis-
ble devices may be serial daisy-chained together, con- ter and a number of data registers. The functional details
nected in parallel, or a combination of the two. can be found in the IEEE 1149.1 specification and are also
The XC4000 Series implements IEEE 1149.1-compatible discussed in the Xilinx application note XAPP 017: “Bound-
BYPASS, PRELOAD/SAMPLE and EXTEST boundary ary Scan in XC4000 Devices.”
scan instructions. When the boundary scan configuration Figure 41 on page 46 shows a simplified block diagram of
option is selected, three normal user I/O pins become ded- the XC4000E Input/Output Block with boundary scan
icated inputs for these functions. Another user output pin implemented. XC4000X boundary scan logic is identical.
becomes the dedicated boundary scan output. The details

R
Figure 42 on page 47 is a diagram of the XC4000 Series data register, respectively, and BSCANT.UPD, which is
boundary scan logic. It includes three bits of Data Register always the last bit of the data register. These three bound-
per IOB, the IEEE 1149.1 Test Access Port controller, and ary scan bits are special-purpose Xilinx test signals.
the Instruction Register with decodes.
The other standard data register is the single flip-flop
XC4000 Series devices can also be configured through the BYPASS register. It synchronizes data being passed
boundary scan logic. See “Readback” on page 58. through the FPGA to the next downstream boundary scan
device.
Data Registers
The FPGA provides two additional data registers that can
The primary data register is the boundary scan register. For be specified using the BSCAN macro. The FPGA provides
each IOB pin in the FPGA, bonded or not, it includes three two user pins (BSCAN.SEL1 and BSCAN.SEL2) which are
bits for In, Out and 3-State Control. Non-IOB pins have the decodes of two user instructions. For these instructions,
appropriate partial bit population for In or Out only. PRO- two corresponding pins (BSCAN.TDO1 and
GRAM, CCLK and DONE are not included in the boundary BSCAN.TDO2) allow user scan data to be shifted out on
scan register. Each EXTEST CAPTURE-DR state captures TDO. The data register clock (BSCAN.DRCK) is available
all In, Out, and 3-state pins. for control of test logic which the user may wish to imple-
The data register also includes the following non-pin bits: ment with CLBs. The NAND of TCK and RUN-TEST-IDLE
TDO.T, and TDO.O, which are always bits 0 and 1 of the is also provided (BSCAN.IDLE).
EXTEST SLEW PULL PULL

TS INV M RATE DOWN UP
TS/OE
3-State TS
TS - capture VCC
Boundary
Scan
TS - update
OUTPUT
INVERT
OUTPUT
M
sd
D Q
Ouput Data O
EC
M INVERT
Ouput Clock OK M PAD

rd OUT
M S/R SEL
O - capture
Clock Enable Boundary Q - capture
Scan
O - update
M
I - capture
Boundary
Scan
Input Data 1 I1
I - update
M M
sd
Q M M
D
EC Input Data 2 I2
DELAY QL
M INVERT
M
FLIP-FLOP/LATCH
Input Clock IK
rd
M S/R
INPUT
GLOBAL
S/R X5792
Figure 41: Block Diagram of XC4000E IOB with Boundary Scan (some details not shown).
XC4000X Boundary Scan Logic is Identical.

R
DATA IN
IOB.T 0
1 sd
D Q D Q 1
0
IOB IOB sd
1
D Q D Q
0
IOB IOB
LE
IOB IOB
1
IOB.I
0
IOB IOB
1 sd
IOB IOB D Q D Q
0
LE
IOB IOB
1
0
IOB IOB IOB.Q
BYPASS
REGISTER
IOB.T 0
M TDO
U 1 sd
TDI
X 0
LE
1 sd
D Q D Q
0
LE
IOB.I
1 6
0

SHIFT/ CLOCK DATA
CAPTURE REGISTER
X9016
Figure 42: XC4000 Series Boundary Scan Logic
Instruction Set Including Boundary Scan in a Schematic

The XC4000 Series boundary scan instruction set also If boundary scan is only to be used during configuration, no
includes instructions to configure the device and read back special schematic elements need be included in the sche-
the configuration data. The instruction set is coded as matic or HDL code. In this case, the special boundary scan
shown in Table 18. pins TDI, TMS, TCK and TDO can be used for user func-
tions after configuration.
Bit Sequence
To indicate that boundary scan remain enabled after config-
The bit sequence within each IOB is: In, Out, 3-State. The uration, place the BSCAN library symbol and connect the
input-only M0 and M2 mode pins contribute only the In bit TDI, TMS, TCK and TDO pad symbols to the appropriate
to the boundary scan I/O data register, while the out- pins, as shown in Figure 44.
put-only M1 pin contributes all three bits.
Even if the boundary scan symbol is used in a schematic,
The first two bits in the I/O data register are TDO.T and the input pins TMS, TCK, and TDI can still be used as
TDO.O, which can be used for the capture of internal sig- inputs to be routed to internal logic. Care must be taken not
nals. The final bit is BSCANT.UPD, which can be used to to force the chip into an undesired boundary scan state by
drive an internal net. These locations are primarily used by inadvertently applying boundary scan input patterns to
Xilinx for internal testing. these pins. The simplest way to prevent this is to keep TMS
From a cavity-up view of the chip (as shown in XDE or High, and then apply whatever signal is desired to TDI and
Epic), starting in the upper right chip corner, the boundary TCK.
scan data-register bits are ordered as shown in Figure 43.
The device-specific pinout tables for the XC4000 Series
include the boundary scan locations for each IOB pin.
BSDL (Boundary Scan Description Language) files for
XC4000 Series devices are available on the Xilinx FTP site.

R
Table 18: Boundary Scan Instructions

Optional To User
Instruction Test I/O Data Logic
TDO Source IBUF
I2 I1 I0 Selected Source
0 0 0 EXTEST DR DR BSCAN
TDI TDI TDO TDO
0 0 1 SAMPLE/PR DR Pin/Logic
TMS TMS DRCK
ELOAD
TCK TCK IDLE
0 1 0 USER 1 BSCAN. User Logic To User
TDO1 SEL1 Logic
TDO1 From
User Logic TDO2 SEL2
0 1 1 USER 2 BSCAN. User Logic X2675
TDO2
1 0 0 READBACK Readback Pin/Logic Figure 44: Boundary Scan Schematic Example
Data
1 0 1 CONFIGURE DOUT Disabled
Configuration
1 1 0 Reserved — — Configuration is the process of loading design-specific pro-
1 1 1 BYPASS Bypass — gramming data into one or more FPGAs to define the func-
Register tional operation of the internal blocks and their
mand registers of a programmable peripheral chip. XC4000
Bit 0 ( TDO end) TDO.T Series devices use several hundred bits of configuration
Bit 1 TDO.O
Bit 2
data per CLB and its associated interconnects. Each con-
Top-edge IOBs (Right to Left) figuration bit defines the state of a static memory cell that
controls either a function look-up table bit, a multiplexer
Left-edge IOBs (Top to Bottom) input, or an interconnect pass transistor. The XACTstep
MD1.T development system translates the design into a netlist file.
MD1.O It automatically partitions, places and routes the logic and
MD1.I
MD0.I generates the configuration data in PROM format.
MD2.I
Bottom-edge IOBs (Left to Right)

Special Purpose Pins
Three configuration mode pins (M2, M1, M0) are sampled
Right-edge IOBs (Bottom to Top) prior to configuration to determine the configuration mode.
(TDI end) B SCANT.UPD After configuration, these pins can be used as auxiliary
connections. M2 and M0 can be used as inputs, and M1
X6075 can be used as an output. The XACTstep development sys-
tem does not use these resources unless they are explicitly
Figure 43: Boundary Scan Bit Sequence
specified in the design entry. This is done by placing a spe-
cial pad symbol called MD2, MD1, or MD0 instead of the
Avoiding Inadvertent Boundary Scan
input or output pad symbol.
If TMS or TCK is used as user I/O, care must be taken to
In XC4000 Series devices, the mode pins have weak
ensure that at least one of these pins is held constant dur-
pull-up resistors during configuration. With all three mode
ing configuration. In some applications, a situation may
pins High, Slave Serial mode is selected, which is the most
occur where TMS or TCK is driven during configuration.
popular configuration mode. Therefore, for the most com-
This may cause the device to go into boundary scan mode
mon configuration mode, the mode pins can be left uncon-
and disrupt the configuration process.
nected. (Note, however, that the internal pull-up resistor
To prevent activation of boundary scan during configura- value can be as high as 100 kΩ.) After configuration, these
tion, do either of the following: pins can individually have weak pull-up or pull-down resis-
• TMS: Tie High to put the Test Access Port controller tors, as specified in the design. A pull-down resistor value
in a benign RESET state of 4.7 kΩ is recommended.
• TCK: Tie High or Low—don't toggle this clock input. These pins are located in the lower left chip corner and are
For more information regarding boundary scan, refer to the near the readback nets. This location allows convenient
Xilinx Application Note XAPP 017.001, “Boundary Scan in routing if compatibility with the XC2000 and XC3000 family
XC4000E Devices.“ conventions of M0/RT, M1/RD is desired.

R
Configuration Modes Additional Address lines in XC4000 devices

XC4000E devices have six configuration modes. XC4000X The XC4000X devices have additional address lines
devices have the same six modes, plus an additional con- (A18-A21) allowing the additional address space required
figuration mode. These modes are selected by a 3-bit input to daisy-chain several large devices.
code applied to the M2, M1, and M0 inputs. There are three The extra address lines are programmable in XC4000EX
self-loading Master modes, two Peripheral modes, and a devices. By default these address lines are not activated. In
Serial Slave mode, which is used primarily for the default mode, the devices are compatible with existing
daisy-chained devices. The coding for mode selection is XC4000 and XC4000E products. If desired, the extra
shown in Table 19. address lines can be used by specifying the address lines
Table 19: Configuration Modes option in bitgen as 22 (bitgen -g AddressLines:22). The
lines (A18-A21) are driven when a master device detects,
Mode M2 M1 M0 CCLK Data via the bitstream, that it should be using all 22 address
Master Serial 0 0 0 output Bit-Serial lines. Because these pins will initially be pulled high by
Slave Serial 1 1 1 input Bit-Serial internal pull-ups, designers using Master Parallel Up mode
Master 1 0 0 output Byte-Wide, should use external pull down resistors on pins A18-A21. If
Parallel Up increment Master Parallel Down mode is used external resistors are
from 00000 not necessary.
Master 1 1 0 output Byte-Wide, All 22 address lines are always active in Master Parallel
Parallel Down decrement modes with XC4000XL devices. The additional address
from 3FFFF lines behave identically to the lower order address lines. If
Peripheral 0 1 1 input Byte-Wide the Address Lines option in bitgen is set to 18, it will be
Synchronous* ignored by the XC4000XL device.
Peripheral 1 0 1 output Byte-Wide The additional address lines (A18-A21) are not available in
6
Asynchronous the PC84 package.
Reserved 0 1 0 — —
Peripheral Modes
Reserved 0 0 1 — —
The two Peripheral modes accept byte-wide data from a
Note: * Peripheral Synchronous can be considered bus. A RDY/BUSY status is available as a handshake sig-
byte-wide Slave Parallel
nal. In Asynchronous Peripheral mode, the internal oscilla-
A detailed description of each configuration mode, with tim- tor generates a CCLK burst signal that serializes the
ing information, is included later in this data sheet. During byte-wide data. CCLK can also drive slave devices. In the
configuration, some of the I/O pins are used temporarily for synchronous mode, an externally supplied clock input to
the configuration process. All pins used during configura- CCLK serializes the data.
tion are shown in Table 23 on page 61.
Slave Serial Mode
Master Modes
In Slave Serial mode, the FPGA receives serial configura-
The three Master modes use an internal oscillator to gener- tion data on the rising edge of CCLK and, after loading its
ate a Configuration Clock (CCLK) for driving potential slave configuration, passes additional data out, resynchronized
devices. They also generate address and timing for exter- on the next falling edge of CCLK.
nal PROM(s) containing the configuration data.
Multiple slave devices with identical configurations can be
Master Parallel (Up or Down) modes generate the CCLK wired with parallel DIN inputs. In this way, multiple devices
signal and PROM addresses and receive byte parallel data. can be configured simultaneously.
The data is internally serialized into the FPGA data-frame
format. The up and down selection generates starting Serial Daisy Chain
addresses at either zero or 3FFFF (3FFFFF when 22 Multiple devices with different configurations can be con-
address lines are used), for compatibility with different nected together in a “daisy chain,” and a single combined
microprocessor addressing conventions. The Master Serial bitstream used to configure the chain of slave devices.
mode generates CCLK and receives the configuration data
in serial form from a Xilinx serial-configuration PROM. To configure a daisy chain of devices, wire the CCLK pins
of all devices in parallel, as shown in Figure 52 on page
CCLK speed is selectable as either 1 MHz (default) or 8 63. Connect the DOUT of each device to the DIN of the
MHz. Configuration always starts at the default slow fre- next. The lead or master FPGA and following slaves each
quency, then can switch to the higher frequency during the passes resynchronized configuration data coming from a
first frame. Frequency tolerance is -50% to +25%. single source. The header data, including the length count,

R
is passed through and is captured by each FPGA when it tiated and most boundary scan instructions cannot be
recognizes the 0010 preamble. Following the length-count used.
data, each FPGA outputs a High on DOUT until it has
The user has some control over the relative timing of these
received its required number of data frames. events and can, therefore, make sure that they occur at the
After an FPGA has received its configuration data, it proper time and the finish point F is reached. Timing is con-
passes on any additional frame start bits and configuration trolled using options in the bitstream generation software.
clocks applied after memory initialization equals the value XC3000 Master with an XC4000 Series Slave
of the 24-bit length count, the FPGAs begin the start-up Some designers want to use an inexpensive lead device in
sequence and become operational together. FPGA I/O are peripheral mode and have the more precious I/O pins of the
normally released two CCLK cycles after the last configura- XC4000 Series devices all available for user I/O. Figure 45
tion bit is received. Figure 48 on page 56 shows the provides a solution for that case.
start-up timing for an XC4000 Series device.
This solution requires one CLB, one IOB and pin, and an
The daisy-chained bitstream is not simply a concatenation internal oscillator with a frequency of up to 5 MHz as a
of the individual bitstreams. The PROM file formatter must clock source. The XC3000 master device must be config-
be used to combine the bitstreams for a daisy-chained con- ured with late Internal Reset, which is the default option.
figuration.
One CLB and one IOB in the lead XC3000-family device
Multi-Family Daisy Chain are used to generate the additional CCLK pulse required by
the XC4000 Series devices. When the lead device removes
All Xilinx FPGAs of the XC2000, XC3000, and XC4000 the internal RESET signal, the 2-bit shift register responds
Series use a compatible bitstream format and can, there- to its clock input and generates an active Low output signal
fore, be connected in a daisy chain in an arbitrary
for the duration of the subsequent clock period. An external
sequence. There is, however, one limitation. The lead
connection between this output and CCLK thus creates the
device must belong to the highest family in the chain. If the extra CCLK pulse.
chain contains XC4000 Series devices, the master nor-
mally cannot be an XC2000 or XC3000 device.
The reason for this rule is shown in Figure 48 on page 56.
Since all devices in the chain store the same length count
value and generate or receive one common sequence of
CCLK pulses, they all recognize length-count match on the
same CCLK edge, as indicated on the left edge of OE/T
Output
Figure 48. The master device then generates additional Connected
Reset to CCLK
CCLK pulses until it reaches its finish point F. The different 0 0
families generate or require different numbers of additional 1
1
0
1
Active Low Output
Active High Output
CCLK pulses until they reach F. Not reaching F means that 0 1
0 1
the device does not really finish its configuration, although . etc .
. . X5223
DONE may have gone High, the outputs became active,
and the internal reset was released. For the XC4000 Series Figure 45: CCLK Generation for XC3000 Master
device, not reaching F means that readback cannot be ini- Driving an XC4000 Series Slave

R
Setting CCLK Frequency Data Stream Format

For Master modes, CCLK can be generated in either of two The data stream (“bitstream”) format is identical for all con-
frequencies. In the default slow mode, the frequency figuration modes.
ranges from 0.5 MHz to 1.25 MHz for XC4000E and
The data stream formats are shown in Table 20. Bit-serial
XC4000EX devices and from 0.6 MHz to 1.8 MHz for
data is read from left to right, and byte-parallel data is effec-
XC4000XL devices. In fast CCLK mode, the frequency
tively assembled from this serial bitstream, with the first bit
ranges from 4 MHz to 10 MHz for XC4000EX devices and
in each byte assigned to D0.
from 5 MHz to 15 MHz for XC4000XL devices. The fre-
quency is selected by an option when running the bitstream The configuration data stream begins with a string of eight
generation software. If an XC4000 Series Master is driving ones, a preamble code, followed by a 24-bit length count
an XC3000- or XC2000-family slave, slow CCLK mode and a separator field of ones. This header is followed by the
must be used. In addition, an XC4000XL device driving a actual configuration data in frames. The length and number
XC4000E or XC4000EX should use slow mode. Slow mode of frames depends on the device type (see Table 21 and
is the default. Table 22). Each frame begins with a start field and ends
with an error check. A postamble code is required to signal
Table 20: XC4000 Series Data Stream Formats
the end of data for a single device. In all cases, additional
All Other start-up bytes of data are required to provide four clocks for
Data Type the startup sequence at the end of configuration. Long
Modes (D0...)
Fill Byte 11111111b daisy chains require additional startup bytes to shift the last
data through the chain. All startup bytes are don’t-cares;
Preamble Code 0010b
these bytes are not included in bitstreams created by the
Length Count COUNT(23:0) Xilinx software.
Fill Bits 1111b
Start Field 0b 6
by the bitstream generation software. The non-CRC error
Data Frame DATA(n-1:0) checking tests for a designated end-of-frame field for each
CRC or Constant xxxx (CRC) frame. For CRC error checking, the software calculates a
Field Check or 0110b running CRC and inserts a unique four-bit partial check at
Extend Write Cycle — the end of each frame. The 11-bit CRC check of the last
Postamble 01111111b frame of an FPGA includes the last seven data bits.
Start-Up Bytes xxh Detection of an error results in the suspension of data load-
ing and the pulling down of the INIT pin. In Master modes,
LEGEND: CCLK and address signals continue to operate externally.
Unshaded Once per bitstream The user must detect INIT and initialize a new configuration
by pulsing the PROGRAM pin Low or cycling Vcc.
Light Once per data frame
Dark Once per device

R
Table 21: XC4000E Program Data
Device XC4003E XC4005E XC4006E XC4008E XC4010E XC4013E XC4020E XC4025E

Max Logic Gates 3,000 5,000 6,000 8,000 10,000 13,000 20,000 25,000
CLBs 100 196 256 324 400 576 784 1,024
(Row x Col.) (10 x 10) (14 x 14) (16 x 16) (18 x 18) (20 x 20) (24 x 24) (28 x 28) (32 x 32)
IOBs 80 112 128 144 160 192 224 256
Flip-Flops 360 616 768 936 1,120 1,536 2,016 2,560
Bits per Frame 126 166 186 206 226 266 306 346
Frames 428 572 644 716 788 932 1,076 1,220
Program Data 53,936 94,960 119,792 147,504 178,096 247,920 329,264 422,128
PROM Size 53,984 95,008 119,840 147,552 178,144 247,968 329,312 422,176
(bits)
Notes: 1. Bits per Frame = (10 x number of rows) + 7 for the top + 13 for the bottom + 1 + 1 start bit + 4 error check bits
Number of Frames = (36 x number of columns) + 26 for the left edge + 41 for the right edge + 1
Program Data = (Bits per Frame x Number of Frames) + 8 postamble bits
PROM Size = Program Data + 40 (header) + 8
2. The user can add more “one” bits as leading dummy bits in the header, or, if CRC = off, as trailing dummy bits at the end of
any frame, following the four error check bits. However, the Length Count value must be adjusted for all such extra “one”
bits, even for extra leading ones at the beginning of the header.
Table 22: XC4000EX/XL Program Data

Device XC4002XL XC4005 XC4010 XC4013 XC4020 XC4028 XC4036 XC4044 XC4052 XC4062 XC4085
Max Logic 5,000 10,000 13,000 20,000 28,000 36,000 44,000 52,000 62,000 85,000
Gates
CLBs 196 400 576 784 1,024 1,296 1,600 1,936 2,304 3,136
(Row x (14 x 14) (20 x 20) (24 x 24) (28 x 28) (32 x 32) (36 x 36) (40 x 40) (44 x 44) (48 x 48) (56 x 56)
Column)
IOBs 112 160 192 224 256 288 320 352 384 448
Flip-Flops 616 1,120 1,536 2,016 2,560 3,168 3,840 4,576 5,376 7,168
Bits per 205 277 325 373 421 469 517 565 613 709
Frame
Frames 741 1,023 1,211 1,399 1,587 1,775 1,963 2,151 2,339 2,715
Program Data 151,910 283,376 393,580 521,832 668,132 832,480 1,014,876 1,215,320 1,433,812 1,924,940
PROM Size 151,960 283,424 393,632 521,880 668,184 832,528 1,014,928 1,215,368 1,433,864 1,924,992
(bits)
Notes: 1. Bits per frame = (12 x number of rows) + 8 for the top + 16 for the bottom + 8 + 1 start bit + 4 error check bits.
Frames = (47 x number of columns) + 27 for the left edge + 52 for the right edge + 4.
Program data = (bits per frame x number of frames) + 5 postamble bits.
PROM size = (program data + 40 header bits + 8 start bits) rounded up to the nearest byte.
2. The user can add more “one” bits as leading dummy bits in the header, or, if CRC = off, as trailing dummy bits at the end of
any frame, following the four error check bits. However, the Length Count value must be adjusted for all such extra “one” bits,
even for extra leading “ones” at the beginning of the header.t
Cyclic Redundancy Check (CRC) for error is detected during the loading of the FPGA, the con-
Configuration and Readback figuration process with a potentially corrupted bitstream is
terminated. The FPGA pulls the INIT pin Low and goes into
The Cyclic Redundancy Check is a method of error detec- a Wait state.
tion in data transmission applications. Generally, the trans-
During Readback, 11 bits of the 16-bit checksum are added
mitting system performs a calculation on the serial
to the end of the Readback data stream. The checksum is
bitstream. The result of this calculation is tagged onto the
data stream as additional check bits. The receiving system computed using the CRC-16 CCITT polynomial, as shown
performs an identical calculation on the bitstream and com- in Figure 46. The checksum consists of the 11 most signif-
pares the result with the received checksum. icant bits of the 16-bit code. A change in the checksum indi-
cates a change in the Readback bitstream. A comparison
Each data frame of the configuration bitstream has four to a previous checksum is meaningful only if the readback
error bits at the end, as shown in Table 20. If a frame data data is independent of the current device state. CLB out-

R
puts should not be included (Read Capture option not

used), and if RAM is present, the RAM content must be
unchanged. VCC No
Boundary Scan >3.5 V
Statistically, one error out of 2048 might go undetected. Instructions
Available:
Yes
There are four major steps in the XC4000 Series power-up Test M0 Generate
PROGRAM
One Time-Out Pulse
configuration sequence. of 16 or 64 ms = Low
Yes
• Configuration Memory Clear
• Initialization Keep Clearing
• Configuration
• Start-Up
EXTEST*
The full process is illustrated in Figure 47. SAMPLE/PRELOAD Completely Clear
BYPASS Configuration Memory ~1.3 µs per Frame
CONFIGURE* Once More
Configuration Memory Clear (* if PROGRAM = High)
When power is first applied or is reapplied to an FPGA, an

internal circuit forces initialization of the configuration logic. INIT No
High? if
When Vcc reaches an operational level, and the circuit Master
passes the write and read test of a sample pair of configu- Yes Master Waits 50 to 250 µs
Before Sampling Mode Lines
ration bits, a time delay is started. This time delay is nomi-
nally 16 ms, and up to 10% longer in the low-voltage Sample
Mode Lines
devices. The delay is four times as long when in Master
Modes (M0 Low), to allow ample time for all slaves to reach Master CCLK 6
Goes Active
a stable Vcc. When all INIT pins are tied together, as rec-

Load One
ommended, the longest delay takes precedence. There- Configuration
fore, devices with different time delays can easily be mixed Data Frame
and matched in a daisy chain.

This delay is applied only on power-up. It is not applied Yes
Frame Pull INIT Low
when reconfiguring an FPGA by pulsing the PROGRAM pin Error and Stop
No
X2 X15
X16 SAMPLE/PRELOAD Config-
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 BYPASS uration No
memory
Full
SERIAL DATA IN Yes
Polynomial: X16 + X15 + X2 + 1 Pass

Configuration
Data to DOUT
1 1 1 1 1 0 15 14 13 12 11 10 9 8 7 6 5
START BIT
LAST DATA FRAME CRC – CHECKSUM CCLK

Count Equals No
Length
Count
X1789 Yes

Start-Up
Sequence
F
I/O Active
Operational
EXTEST
SAMPLE PRELOAD
BYPASS
USER 1 If Boundary Scan
USER 2 is Selected
CONFIGURE
READBACK
X6076
Figure 47: Power-up Configuration Sequence

R
Low. During this time delay, or as long as the PROGRAM rise time is excessive or poorly defined. As long as PRO-
input is asserted, the configuration logic is held in a Config- GRAM is Low, the FPGA keeps clearing its configuration
uration Memory Clear state. The configuration-memory memory. When PROGRAM goes High, the configuration
frames are consecutively initialized, using the internal oscil- memory is cleared one more time, followed by the begin-
lator. ning of configuration, provided the INIT input is not exter-
nally held Low. Note that a Low on the PROGRAM input
At the end of each complete pass through the frame
automatically forces a Low on the INIT output. The XC4000
addressing, the power-on time-out delay circuitry and the
Series PROGRAM pin has a permanent weak pull-up.
level of the PROGRAM pin are tested. If neither is asserted,
the logic initiates one additional clearing of the configura- Using an open-collector or open-drain driver to hold INIT
tion frames and then tests the INIT input. Low before the beginning of configuration causes the
FPGA to wait after completing the configuration memory
Initialization clear operation. When INIT is no longer held Low exter-
During initialization and configuration, user pins HDC, LDC, nally, the device determines its configuration mode by cap-
INIT and DONE provide status outputs for the system inter- turing its mode pins, and is ready to start the configuration
face. The outputs LDC, INIT and DONE are held Low and process. A master device waits up to an additional 250 µs
HDC is held High starting at the initial application of power. to make sure that any slaves in the optional daisy chain
have seen that INIT is High.
The open drain INIT pin is released after the final initializa-
tion pass through the frame addresses. There is a deliber- Start-Up
ate delay of 50 to 250 µs (up to 10% longer for low-voltage
Start-up is the transition from the configuration process to
devices) before a Master-mode device recognizes an inac-
the intended user operation. This transition involves a
tive INIT. Two internal clocks after the INIT pin is recognized
change from one clock source to another, and a change
as High, the FPGA samples the three mode lines to deter-
from interfacing parallel or serial configuration data where
mine the configuration mode. The appropriate interface
most outputs are 3-stated, to normal operation with I/O pins
lines become active and the configuration preamble and
active in the user-system. Start-up must make sure that the
data can be loaded.Configuration
user-logic ‘wakes up’ gracefully, that the outputs become
The 0010 preamble code indicates that the following 24 bits active without causing contention with the configuration sig-
represent the length count. The length count is the total nals, and that the internal flip-flops are released from the
number of configuration clocks needed to load the com- global Reset or Set at the right time.
plete configuration data. (Four additional configuration
Figure 48 describes start-up timing for the three Xilinx fam-
clocks are required to complete the configuration process,
ilies in detail. The configuration modes can use any of the
as discussed below.) After the preamble and the length
four timing sequences.
count have been passed through to all devices in the daisy
chain, DOUT is held High to prevent frame start bits from To access the internal start-up signals, place the STARTUP
reaching any daisy-chained devices. library symbol.
A specific configuration bit, early in the first frame of a mas- Start-up Timing
ter device, controls the configuration-clock rate and can
increase it by a factor of eight. Therefore, if a fast configu- Different FPGA families have different start-up sequences.
ration clock is selected by the bitstream, the slower clock The XC2000 family goes through a fixed sequence. DONE
rate is used until this configuration bit is detected. goes High and the internal global Reset is de-activated one
Each frame has a start field followed by the frame-configu- CCLK period after the I/O become active.
ration data bits and a frame error field. If a frame data error The XC3000A family offers some flexibility. DONE can be
is detected, the FPGA halts loading, and signals the error programmed to go High one CCLK period before or after
by pulling the open-drain INIT pin Low. After all configura- the I/O become active. Independent of DONE, the internal
tion frames have been loaded into an FPGA, DOUT again global Reset is de-activated one CCLK period before or
follows the input data so that the remaining data is passed after the I/O become active.
on to the next device.
The XC4000 Series offers additional flexibility. The three
Delaying Configuration After Power-Up events — DONE going High, the internal Set/Reset being
de-activated, and the user I/O going active — can all occur
There are two methods of delaying configuration after
in any arbitrary sequence. Each of them can occur one
power-up: put a logic Low on the PROGRAM input, or pull
CCLK period before or after, or simultaneous with, any of
the bidirectional INIT pin Low, using an open-collector
the others. This relative timing is selected by means of soft-
(open-drain) driver. (See Figure 47 on page 53.)
ware options in the bitstream generation software.
A Low on the PROGRAM input is the more radical
approach, and is recommended when the power-supply

R
The default option, and the most practical one, is for DONE received since INIT went High equals the loaded value of
to go High first, disconnecting the configuration data source the length count.
and avoiding any contention when the I/Os become active
The next rising clock edge sets a flip-flop Q0, shown in
one clock later. Reset/Set is then released another clock
Figure 49. Q0 is the leading bit of a 5-bit shift register. The
period later to make sure that user-operation starts from
outputs of this register can be programmed to control three
stable internal conditions. This is the most common
events.
sequence, shown with heavy lines in Figure 48, but the
designer can modify it to meet particular requirements. • The release of the open-drain DONE output
• The change of configuration-related pins to the user
Normally, the start-up sequence is controlled by the internal
function, activating all IOBs.
device oscillator output (CCLK), which is asynchronous to
• The termination of the global Set/Reset initialization of
the system clock.
all CLB and IOB storage elements.
XC4000 Series offers another start-up clocking option,
The DONE pin can also be wire-ANDed with DONE pins of
UCLK_NOSYNC. The three events described above need
other FPGAs or with other external signals, and can then
not be triggered by CCLK. They can, as a configuration be used as input to bit Q3 of the start-up register. This is
option, be triggered by a user clock. This means that the called “Start-up Timing Synchronous to Done In” and is
device can wake up in synchronism with the user system. selected by either CCLK_SYNC or UCLK_SYNC.
When the UCLK_SYNC option is enabled, the user can
When DONE is not used as an input, the operation is called
externally hold the open-drain DONE output Low, and thus
“Start-up Timing Not Synchronous to DONE In,” and is
stall all further progress in the start-up sequence until
selected by either CCLK_NOSYNC or UCLK_NOSYNC.
DONE is released and has gone High. This option can be
used to force synchronization of several FPGAs to a com- As a configuration option, the start-up control register
mon user clock, or to guarantee that all devices are suc- beyond Q0 can be clocked either by subsequent CCLK
cessfully configured before any I/Os go active. pulses or from an on-chip user net called STARTUP.CLK. 6
These signals can be accessed by placing the STARTUP
If either of these two options is selected, and no user clock
library symbol.
is specified in the design or attached to the device, the chip
could reach a point where the configuration of the device is Start-up from CCLK
complete and the Done pin is asserted, but the outputs do
If CCLK is used to drive the start-up, Q0 through Q3 pro-
not become active. The solution is either to recreate the bit-
vide the timing. Heavy lines in Figure 48 show the default
stream specifying the start-up clock as CCLK, or to supply
timing, which is compatible with XC2000 and XC3000
the appropriate user clock.
devices using early DONE and late Reset. The thin lines
Start-up Sequence indicate all other possible timing options.
The Start-up sequence begins when the configuration

memory is full, and the total number of configuration clocks

R
Length Count Match

CCLK Period
CCLK
F
DONE
I/O
XC2000
Global Reset
F = Finished, no more
F
configuration clocks needed
DONE Daisy-chain lead device
XC3000 must have latest F
I/O
Heavy lines describe
default timing
Global Reset
F
DONE
C1 C2 C3 C4
XC4000E/X I/O
CCLK_NOSYNC
C2 C3 C4
GSR Active
C2 C3 C4
DONE IN
F
DONE
C1, C2 or C3
XC4000E/X I/O
CCLK_SYNC
Di Di+1
GSR Active
Di Di+1
F
DONE
C1 U2 U3 U4
I/O
XC4000E/X
UCLK_NOSYNC U2 U3 U4
GSR Active
U2 U3 U4
DONE IN
F
DONE
C1 U2
I/O
XC4000E/X
UCLK_SYNC Di Di+1 Di+2
GSR Active
Di Di+1 Di+2
Synchronization
Uncertainty UCLK Period
X9024
Figure 48: Start-up Timing

R
Start-up from a User Clock (STARTUP.CLK) Release of User I/O After DONE Goes High
When, instead of CCLK, a user-supplied start-up clock is By default, the user I/O are released one CCLK cycle after
selected, Q1 is used to bridge the unknown phase relation- the DONE pin goes High. If CCLK is not clocked after
ship between CCLK and the user clock. This arbitration DONE goes High, the outputs remain in their initial state —
causes an unavoidable one-cycle uncertainty in the timing 3-stated, with a 50 kΩ - 100 kΩ pull-up. The delay from
of the rest of the start-up sequence. DONE High to active user I/O is controlled by an option to
the bitstream generation software.
DONE Goes High to Signal End of Configuration
XC4000 Series devices read the expected length count Release of Global Set/Reset After DONE Goes
from the bitstream and store it in an internal register. The High
length count varies according to the number of devices and By default, Global Set/Reset (GSR) is released two CCLK
the composition of the daisy chain. Each device also counts cycles after the DONE pin goes High. If CCLK is not
the number of CCLKs during configuration. clocked twice after DONE goes High, all flip-flops are held
Two conditions have to be met in order for the DONE pin to in their initial set or reset state. The delay from DONE High
go high: to GSR inactive is controlled by an option to the bitstream
generation software.
• the chip's internal memory must be full, and
• the configuration length count must be met, exactly. Configuration Complete After DONE Goes High
This is important because the counter that determines Three full CCLK cycles are required after the DONE pin
when the length count is met begins with the very first goes High, as shown in Figure 48 on page 56. If CCLK is
CCLK, not the first one after the preamble. not clocked three times after DONE goes High, readback
Therefore, if a stray bit is inserted before the preamble, or cannot be initiated and most boundary scan instructions
the data source is not ready at the time of the first CCLK, cannot be used.
6
the internal counter that holds the number of CCLKs will be
one ahead of the actual number of data bits read. At the
Configuration Through the Boundary Scan
end of configuration, the configuration memory will be full, Pins
but the number of bits in the internal counter will not match XC4000 Series devices can be configured through the
the expected length count. boundary scan pins. The basic procedure is as follows:
As a consequence, a Master mode device will continue to • Power up the FPGA with INIT held Low (or drive the
send out CCLKs until the internal counter turns over to PROGRAM pin Low for more than 300 ns followed by a
zero, and then reaches the correct length count a second High while holding INIT Low). Holding INIT Low allows
time. This will take several seconds [224 ∗ CCLK period] — enough time to issue the CONFIG command to the
which is sometimes interpreted as the device not configur- FPGA. The pin can be used as I/O after configuration if
ing at all. a resistor is used to hold INIT Low.
If it is not possible to have the data ready at the time of the • Issue the CONFIG command to the TMS input
first CCLK, the problem can be avoided by increasing the • Wait for INIT to go High
number in the length count by the appropriate value. The • Sequence the boundary scan Test Access Port to the
XACT User Guide includes detailed information about man- SHIFT-DR state
ually altering the length count. • Toggle TCK to clock data into TDI pin.
Note that DONE is an open-drain output and does not go The user must account for all TCK clock cycles after INIT
High unless an internal pull-up is activated or an external goes High, as all of these cycles affect the Length Count
pull-up is attached. The internal pull-up is activated as the compare.
default by the bitstream generation software. For more detailed information, refer to the Xilinx application
note XAPP017, “Boundary Scan in XC4000 Devices.” This
application note also applies to XC4000E and XC4000X
devices.

R
Q3 Q1/Q4
STARTUP DONE
Q2
IN
* IOBs OPERATIONAL PER CONFIGURATION
* GLOBAL SET/RESET OF
ALL CLB AND IOB FLIP-FLOP
1
0
GSR ENABLE
GSR INVERT
STARTUP.GSR CONTROLLED BY STARTUP SYMBOL
IN THE USER SCHEMATIC (SEE
STARTUP.GTS LIBRARIES GUIDE)
GTS INVERT
GTS ENABLE
0
GLOBAL 3-STATE OF ALL IOBs
1
Q S
* DONE
" FINISHED "

1 1 ENABLES BOUNDARY
0 0 SCAN, READBACK AND
CONTROLS THE OSCILLATOR
Q0 Q1 Q2 Q3 Q4
FULL 1
S Q D Q D Q D Q D Q
LENGTH COUNT 0
K K K * K K
CLEAR MEMORY
CCLK 0
STARTUP.CLK 1
USER NET
M
CONFIGURATION BIT OPTIONS SELECTED BY USER IN "MAKEBITS"
* * X1528
Figure 49: Start-up Logic
Readback BACK library symbol and attach the appropriate pad sym-
bols, as shown in Figure 50.
The user can read back the content of configuration mem-
ory and the level of certain internal nodes without interfer- After Readback has been initiated by a Low-to-High transi-
ing with the normal operation of the device. tion on RDBK.TRIG, the RDBK.RIP (Read In Progress)
output goes High on the next rising edge of RDBK.CLK.
Readback not only reports the downloaded configuration Subsequent rising edges of this clock shift out Readback
bits, but can also include the present state of the device, data on the RDBK.DATA net.
represented by the content of all flip-flops and latches in
CLBs and IOBs, as well as the content of function genera- Readback data does not include the preamble, but starts
tors used as RAMs. with five dummy bits (all High) followed by the Start bit
(Low) of the first frame. The first two data bits of the first
Note that in XC4000 Series devices, configuration data is frame are always High.
not inverted with respect to configuration as it is in XC2000
and XC3000 families. Each frame ends with four error check bits. They are read
XC4000 Series Readback does not use any dedicated read back as High. An additional Start bit (Low) and an
pins, but uses four internal nets (RDBK.TRIG, RDBK.DATA, 11-bit Cyclic Redundancy Check (CRC) signature follow,
RDBK.RIP and RDBK.CLK) that can be routed to any IOB. before RDBK.RIP returns Low.
To access the internal Readback signals, place the READ-

R
IF UNCONNECTED,
DEFAULT IS CCLK
CLK DATA READ_DATA

MD1
READBACK OBUF
READ_TRIGGER TRIG RIP
MD0
IBUF X1786
Figure 50: Readback Schematic Example
Readback Options
I/O I/O
Readback options are: Read Capture, Read Abort, and
Clock Select. They are set with the bitstream generation PROGRAMMABLE
INTERCONNECT
software.
Read Capture
DATA
TRIG
RIP
When the Read Capture option is selected, the readback
I
data stream includes sampled values of CLB and IOB sig- rdbk I/O I/O I/O rdclk
nals. The rising edge of RDBK.TRIG latches the inverted

values of the four CLB outputs, the IOB output flip-flops and X1787
the input signals I1 and I2. Note that while the bits describ- Figure 51: READBACK Symbol in Graphical Editor
ing configuration (interconnect, function generators, and
RAM content) are not inverted, the CLB and IOB output sig- Violating the Maximum High and Low Time
nals are inverted. Specification for the Readback Clock 6
When the Read Capture option is not selected, the values The readback clock has a maximum High and Low time
of the capture bits reflect the configuration data originally specification. In some cases, this specification cannot be
written to those memory locations. met. For example, if a processor is controlling readback, an
If the RAM capability of the CLBs is used, RAM data are interrupt may force it to stop in the middle of a readback.
available in readback, since they directly overwrite the F This necessitates stopping the clock, and thus violating the
and G function-table configuration of the CLB. specification.
RDBK.TRIG is located in the lower-left corner of the device, The specification is mandatory only on clocking data at the
as shown in Figure 51. end of a frame prior to the next start bit. The transfer mech-
anism will load the data to a shift register during the last six
Read Abort clock cycles of the frame, prior to the start bit of the follow-
ing frame. This loading process is dynamic, and is the
When the Read Abort option is selected, a High-to-Low
source of the maximum High and Low time requirements.
transition on RDBK.TRIG terminates the readback opera-
tion and prepares the logic to accept another trigger. Therefore, the specification only applies to the six clock
cycles prior to and including any start bit, including the
After an aborted readback, additional clocks (up to one
clocks before the first start bit in the readback data stream.
readback clock per configuration frame) may be required to
re-initialize the control logic. The status of readback is indi-
the register is not dynamic. Thus, it can be shifted out just
cated by the output control net RDBK.RIP. RDBK.RIP is
like a regular shift register.
High whenever a readback is in progress.
The user must precisely calculate the location of the read-
Clock Select back data relative to the frame. The system must keep
CCLK is the default clock. However, the user can insert track of the position within a data frame, and disable inter-
another clock on RDBK.CLK. Readback control and data rupts before frame boundaries. Frame lengths and data
are clocked on rising edges of RDBK.CLK. If readback formats are listed in Table 20, Table 21 and Table 22.
must be inhibited for security reasons, the readback control
nets are simply not connected. Readback with the XChecker Cable
RDBK.CLK is located in the lower right chip corner, as Logic Probe uses the readback feature for bitstream verifi-
shown in Figure 51. cation. It can also display selected internal signals on the
PC or workstation screen, functioning as a low-cost in-cir-
cuit emulator.

R
XC4000E/EX/XL Program Readback Switching Characteristic Guidelines

Finished
Internal Net
rdbk.TRIG
TRCRT TRCRT 2
1 TRTRC 2 1 TRTRC
rdclk.I
4 TRCL TRCH 5
rdbk.RIP
6
TRCRR
TRCRD
7 X1790
E/EX
XL

R
Table 23: Pin Functions During Configuration
CONFIGURATION MODE <M2:M1:M0>

SLAVE MASTER SYNCH. ASYNCH. MASTER MASTER
USER
SERIAL SERIAL PERIPHERAL PERIPHERAL PARALLEL DOWN PARALLEL UP
OPERATION
<1:1:1> <0:0:0> <0:1:1> <1:0:1> <1:1:0> <1:0:0>
M2(HIGH) (I) M2(LOW) (I) M2(LOW) (I) M2(HIGH) (I) M2(HIGH) (I) M2(HIGH) (I) (I)
M1(HIGH) (I) M1(LOW) (I) M1(HIGH) (I) M1(LOW) (I) M1(HIGH) (I) M1(LOW) (I) (O)
M0(HIGH) (I) M0(LOW) (I) M0(HIGH) (I) M0(HIGH) (I) M0(LOW) (I) M0(LOW) (I) (I)
HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) I/O
LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) I/O
INIT INIT INIT INIT INIT INIT I/O
DONE DONE DONE DONE DONE DONE DONE
PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM
CCLK (I) CCLK (O) CCLK (I) CCLK (O) CCLK (O) CCLK (O) CCLK (I)
RDY/BUSY (O) RDY/BUSY (O) RCLK (O) RCLK (O) I/O
RS (I) I/O
CS0 (I) I/O
DATA 7 (I) DATA 7 (I) DATA 7 (I) DATA 7 (I) I/O
DATA 1 (I) DATA 1 (I) DATA 1 (I) DATA 1 (I) I/O 6
DIN (I) DIN (I) DATA 0 (I) DATA 0 (I) DATA 0 (I) DATA 0 (I) I/O
DOUT DOUT DOUT DOUT DOUT DOUT SGCK4-GCK5-I/O
TDI TDI TDI TDI TDI TDI TDI-I/O
TCK TCK TCK TCK TCK TCK TCK-I/O
TMS TMS TMS TMS TMS TMS TMS-I/O
TDO TDO TDO TDO TDO TDO TDO-(O)
WS (I) A0 A0 I/O
A1 A1 PGCK4-GCK6-I/O
CS1 A2 A2 I/O
A3 A3 I/O
A4 A4 I/O
A5 A5 I/O
A6 A6 I/O
A7 A7 I/O
A8 A8 I/O
A9 A9 I/O
A10 A10 I/O
A11 A11 I/O
A12 A12 I/O
A13 A13 I/O
A14 A14 I/O
A15 A15 SGCK1-GCK7-I/O
A17 A17 I/O
A18* A18* I/O
A19* A19* I/O
A20* A20* I/O
A21* A21* I/O
ALL OTHERS

R
Table 24: Pin Functions During Configuration
CONFIGURATION MODE <M2:M1:M0>

SLAVE MASTER SYNCH. ASYNCH. MASTER MASTER
USER
SERIAL SERIAL PERIPHERAL PERIPHERAL PARALLEL DOWN PARALLEL UP
OPERATION
<1:1:1> <0:0:0> <0:1:1> <1:0:1> <1:1:0> <1:0:0>
M2(HIGH) (I) M2(LOW) (I) M2(LOW) (I) M2(HIGH) (I) M2(HIGH) (I) M2(HIGH) (I) (I)
M1(HIGH) (I) M1(LOW) (I) M1(HIGH) (I) M1(LOW) (I) M1(HIGH) (I) M1(LOW) (I) (O)
M0(HIGH) (I) M0(LOW) (I) M0(HIGH) (I) M0(HIGH) (I) M0(LOW) (I) M0(LOW) (I) (I)
HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) I/O
LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) I/O
INIT INIT INIT INIT INIT INIT I/O
DONE DONE DONE DONE DONE DONE DONE
PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM
CCLK (I) CCLK (O) CCLK (I) CCLK (O) CCLK (O) CCLK (O) CCLK (I)
RDY/BUSY (O) RDY/BUSY (O) RCLK (O) RCLK (O) I/O
RS (I) I/O
CS0 (I) I/O
DIN (I) DIN (I) DATA 0 (I) DATA 0 (I) DATA 0 (I) DATA 0 (I) I/O
DOUT DOUT DOUT DOUT DOUT DOUT SGCK4-GCK5-I/O
TDI TDI TDI TDI TDI TDI TDI-I/O
TCK TCK TCK TCK TCK TCK TCK-I/O
TMS TMS TMS TMS TMS TMS TMS-I/O
TDO TDO TDO TDO TDO TDO TDO-(O)
WS (I) A0 A0 I/O
CS1 A2 A2 I/O
A3 A3 I/O
A4 A4 I/O
A5 A5 I/O
A6 A6 I/O
A7 A7 I/O
A8 A8 I/O
A9 A9 I/O
A10 A10 I/O
A11 A11 I/O
A12 A12 I/O
A13 A13 I/O
A14 A14 I/O
A15 A15 SGCK1-GCK7-I/O
A17 A17 I/O
A18* A18* I/O
A19* A19* I/O
A20* A20* I/O
A21* A21* I/O
ALL OTHERS
* XC4000X only
Notes 1. A shaded table cell represents a 50 kΩ - 100 kΩ pull-up before and during configuration.

R
Configuration Timing There is an internal delay of 0.5 CCLK periods, which

The seven configuration modes are discussed in detail in the next FPGA in the daisy chain accepts data on the sub-
this section. Timing specifications are included. sequent rising CCLK edge.
Slave Serial Mode Figure 52 shows a full master/slave system. An XC4000
Series device in Slave Serial mode should be connected as
In Slave Serial mode, an external signal drives the CCLK shown in the third device from the left.
input of the FPGA. The serial configuration bitstream must
be available at the DIN input of the lead FPGA a short Slave Serial mode is selected by a <111> on the mode pins
setup time before each rising CCLK edge. (M2, M1, M0). Slave Serial is the default mode if the mode
pins are left unconnected, as they have weak pull-up resis-
tors during configuration.
NOTE: NOTE:
M2, M1, M0 can be shorted M2, M1, M0 can be shorted
to Ground if not used as I/O to VCC if not used as I/O
VCC
N/C
4.7 KΩ 4.7 KΩ 4.7 KΩ 4.7 KΩ 4.7 KΩ
4.7 KΩ
M0 M1 M0 M1 M0 M1 PWRDN
M2 N/C M2 M2
CCLK CCLK
XC4000E/X VCC
MASTER XC1700D +5 V XC4000E/X, XC3100A
4.7 KΩ
SERIAL XC5200 SLAVE
SLAVE
CCLK
DIN
CLK
DATA
VPP
6
PROGRAM X9025
DIN Bit n Bit n + 1
CCLK
4 TCCH 3 TCCO
DOUT
Bit n - 1 Bit n
(Output)
X5379

CCLK

R
Master Serial Mode frame, increases the CCLK frequency by a factor of eight.
For actual timing values please refer to “Configuration
In Master Serial mode, the CCLK output of the lead FPGA Switching Characteristics” on page 71. Be sure that the
drives a Xilinx Serial PROM that feeds the FPGA DIN input. serial PROM and slaves are fast enough to support this
Each rising edge of the CCLK output increments the Serial
data rate. XC2000, XC3000/A, and XC3100A devices do
PROM internal address counter. The next data bit is put on
not support the Fast ConfigRate option.
the SPROM data output, connected to the FPGA DIN pin.
The lead FPGA accepts this data on the subsequent rising The SPROM CE input can be driven from either LDC or
CCLK edge. DONE. Using LDC avoids potential contention on the DIN
pin, if this pin is configured as user-I/O, but LDC is then
The lead FPGA then presents the preamble data—and all restricted to be a permanently High user output after con-
data that overflows the lead device—on its DOUT pin. figuration. Using DONE can also avoid contention on DIN,
There is an internal pipeline delay of 1.5 CCLK periods, provided the early DONE option is invoked.
which means that DOUT changes on the falling CCLK
edge, and the next FPGA in the daisy chain accepts data Figure 52 on page 63 shows a full master/slave system.
on the subsequent rising CCLK edge. The leftmost device is in Master Serial mode.
In the bitstream generation software, the user can specify Master Serial mode is selected by a <000> on the mode
Fast ConfigRate, which, starting several bits into the first pins (M2, M1, M0).
CCLK
(Output)
2 TCKDS
1 TDSCK

(Output)
X3223

CCLK

R
Master Parallel Modes Master Parallel Down mode is selected by a <110> on the
mode pins. The EPROM addresses start at 3FFFF and
In the two Master Parallel modes, the lead FPGA directly decrement.
addresses an industry-standard byte-wide EPROM, and
accepts eight data bits just before incrementing or decre- Additional Address lines in XC4000 devices
menting the address outputs.
The XC4000X devices have additional address lines
The eight data bits are serialized in the lead FPGA, which (A18-A21) allowing the additional address space required
then presents the preamble data—and all data that over- to daisy-chain several large devices.
flows the lead device—on its DOUT pin. There is an inter-
The extra address lines are programmable in XC4000EX
nal delay of 1.5 CCLK periods, after the rising CCLK edge
devices. By default these address lines are not activated. In
that accepts a byte of data (and also changes the EPROM
the default mode, the devices are compatible with existing
address) until the falling CCLK edge that makes the LSB
XC4000 and XC4000E products. If desired, the extra
(D0) of this byte appear at DOUT. This means that DOUT
address lines can be used by specifying the address lines
changes on the falling CCLK edge, and the next FPGA in
option in bitgen as 22 (bitgen -g AddressLines:22). The
the daisy chain accepts data on the subsequent rising
lines (A18-A21) are driven when a master device detects,
CCLK edge.
via the bitstream, that it should be using all 22 address
The PROM address pins can be incremented or decre- lines. Because these pins will initially be pulled high by
mented, depending on the mode pin settings. This option internal pull-ups, designers using Master Parallel Up mode
allows the FPGA to share the PROM with a wide variety of should use external pull down resistors on pins A18-A21. If
microprocessors and microcontrollers. Some processors Master Parallel Down mode is used external resistors are
must boot from the bottom of memory (all zeros) while oth- not necessary.
ers must boot from the top. The FPGA is flexible and can
All 22 address lines are always active in Master Parallel
load its configuration bitstream from either end of the mem-
modes with XC4000XL devices. The additional address 6
ory.
lines behave identically to the lower order address lines. If
Master Parallel Up mode is selected by a <100> on the the Address Lines option in bitgen is set to 18, it will be
mode pins (M2, M1, M0). The EPROM addresses start at ignored by the XC4000XL device.
00000 and increment.
The additional address lines (A18-A21) are not available in
the PC84 package.
HIGH TO DIN OF OPTIONAL

or DAISY-CHAINED FPGAS
4.7KΩ N/C
LOW
N/C
M0 M1 M2 TO CCLK OF OPTIONAL
DAISY-CHAINED FPGAS
CCLK
DOUT
NOTE:M0 can be shorted
to Ground if not used A17 ... M0 M1 M2
as I/O.
A16 ... DIN DOUT
VCC
A15 ... EPROM
(8K x 8) CCLK
A14 ...
4.7KΩ (OR LARGER)
USER CONTROL OF HIGHER
INIT A13 ...
ORDER PROM ADDRESS BITS XC4000E/X
CAN BE USED TO SELECT BETWEEN SLAVE
A12 A12
ALTERNATIVE CONFIGURATIONS
A11 A11
PROGRAM
A10 A10
PROGRAM A9 A9
DONE INIT
D7 A8 A8
D6 A7 A7 D7
D5 A6 A6 D6
D4 A5 A5 D5
D3 A4 A4 D4
D2 A3 A3 D3
D1 A2 A2 D2
D0 A1 A1 D1
A0 A0 D0
DONE OE
CE
DATA BUS 8
PROGRAM
X9026
Figure 55: Master Parallel Mode Circuit Diagram

R
A0-A17
(output) Address for Byte n Address for Byte n + 1
1 TRAC
D0-D7
Byte
2 TDRC 3 TRCD
RCLK
(output)
7 CCLKs CCLK
CCLK
(output)
DOUT
(output) D6 D7
Byte n - 1 X6078

Delay to Address valid 1 TRAC 0 200 ns
RCLK Data setup time 2 TDRC 60 ns
Data hold time 3 TRCD 0 ns
2. The first Data byte is loaded and CCLK starts at the end of the first RCLK active cycle (rising edge).
This timing diagram shows that the EPROM requirements are extremely relaxed. EPROM access time can be longer than
500 ns. EPROM data output has no hold-time requirements.
Figure 56: Master Parallel Mode Programming Switching Characteristics

R
Synchronous Peripheral Mode The lead FPGA serializes the data and presents the pre-
amble data (and all data that overflows the lead device) on
Synchronous Peripheral mode can also be considered its DOUT pin. There is an internal delay of 1.5 CCLK peri-
Slave Parallel mode. An external signal drives the CCLK ods, which means that DOUT changes on the falling CCLK
input(s) of the FPGA(s). The first byte of parallel configura- edge, and the next FPGA in the daisy chain accepts data
tion data must be available at the Data inputs of the lead
on the subsequent rising CCLK edge.
FPGA a short setup time before the rising CCLK edge.
Subsequent data bytes are clocked in on every eighth con- In order to complete the serial shift operation, 10 additional
secutive rising CCLK edge. CCLK rising edges are required after the last data byte has
been loaded, plus one more CCLK cycle for each
The same CCLK edge that accepts data, also causes the daisy-chained device.
RDY/BUSY output to go High for one CCLK period. The pin
name is a misnomer. In Synchronous Peripheral mode it is Synchronous Peripheral mode is selected by a <011> on
really an ACKNOWLEDGE signal. Synchronous operation the mode pins (M2, M1, M0).
does not require this response, but it is a meaningful signal
for test purposes. Note that RDY/BUSY is pulled High with
a high-impedance pullup prior to INIT going High.
NOTE:
M2 can be shorted to Ground
if not used as I/O
N/C 4.7 kΩ N/C
M0 M1 M2 M0 M1 M2
CLOCK CCLK CCLK 6
OPTIONAL
8 DAISY-CHAINED
DATA BUS D0-7 FPGAs
DOUT DIN DOUT
VCC XC4000E/X XC4000E/X

SYNCHRO- SLAVE
NOUS
4.7 kΩ
PERIPHERAL
CONTROL RDY/BUSY
SIGNALS INIT DONE INIT DONE
4.7 kΩ
X9027
Figure 57: Synchronous Peripheral Mode Circuit Diagram

R
CCLK
INIT
BYTE BYTE
0 1
BYTE 0 OUT BYTE 1 OUT
DOUT 0 1 2 3 4 5 6 7 0 1
RDY/BUSY
X6096

INIT (High) setup time TIC 5 µs
D0 - D7 setup time TDC 60 ns
D0 - D7 hold time TCD 0 ns
CCLK
Notes: 1. Peripheral Synchronous mode can be considered Slave Parallel mode. An external CCLK provides timing, clocking in the
first data byte on the second rising edge of CCLK after INIT goes High. Subsequent data bytes are clocked in on every
eighth consecutive rising edge of CCLK.
2. The RDY/BUSY line goes High for one CCLK period after data has been clocked in, although synchronous operation does
not require such a response.
3. The pin name RDY/BUSY is a misnomer. In Synchronous Peripheral mode this is really an ACKNOWLEDGE signal.
4. Note that data starts to shift out serially on the DOUT pin 0.5 CCLK periods after it was loaded in parallel. Therefore,
additional CCLK pulses are clearly required after the last byte has been loaded.
Figure 58: Synchronous Peripheral Mode Programming Switching Characteristics

R
Asynchronous Peripheral Mode The READY/BUSY handshake can be ignored if the delay
from any one Write to the end of the next Write is guaran-
Write to FPGA teed to be longer than 10 CCLK periods.
Asynchronous Peripheral mode uses the trailing edge of Status Read
the logic AND condition of WS and CS0 being Low and RS
and CS1 being High to accept byte-wide data from a micro- The logic AND condition of the CS0, CS1and RS inputs
processor bus. In the lead FPGA, this data is loaded into a puts the device status on the Data bus.
double-buffered UART-like parallel-to-serial converter and • D7 High indicates Ready
is serially shifted into the internal logic. • D7 Low indicates Busy
The lead FPGA presents the preamble data (and all data • D0 through D6 go unconditionally High
that overflows the lead device) on its DOUT pin. The It is mandatory that the whole start-up sequence be started
RDY/BUSY output from the lead FPGA acts as a hand- and completed by one byte-wide input. Otherwise, the pins
shake signal to the microprocessor. RDY/BUSY goes Low used as Write Strobe or Chip Enable might become active
when a byte has been received, and goes High again when outputs and interfere with the final byte transfer. If this
the byte-wide input buffer has transferred its information transfer does not occur, the start-up sequence is not com-
into the shift register, and the buffer is ready to receive new pleted all the way to the finish (point F in Figure 48 on page
data. A new write may be started immediately, as soon as 56).
the RDY/BUSY output has gone Low, acknowledging
In this case, at worst, the internal reset is not released. At
receipt of the previous data. Write may not be terminated
best, Readback and Boundary Scan are inhibited. The
until RDY/BUSY is High again for one CCLK period. Note
that RDY/BUSY is pulled High with a high-impedance length-count value, as generated by the XACTstep soft-
pull-up prior to INIT going High. ware, ensures that these problems never occur.
The length of the BUSY signal depends on the activity in Although RDY/BUSY is brought out as a separate signal,
microprocessors can more easily read this information on
6
the UART. If the shift register was empty when the new
one of the data lines. For this purpose, D7 represents the
byte was received, the BUSY signal lasts for only two CCLK
periods. If the shift register was still full when the new byte RDY/BUSY status when RS is Low, WS is High, and the
two chip select lines are both active.
was received, the BUSY signal can be as long as nine
CCLK periods. Asynchronous Peripheral mode is selected by a <101> on
Note that after the last byte has been entered, only seven of the mode pins (M2, M1, M0).
its bits are shifted out. CCLK remains High with DOUT
equal to bit 6 (the next-to-last bit) of the last byte entered.
N/C
N/C N/C
4.7 kΩ
M0 M1 M2 M0 M1 M2
DATA 8
D0–7 CCLK CCLK
BUS
OPTIONAL
DAISY-CHAINED
FPGAs
DOUT DIN DOUT
VCC ADDRESS CS0
ADDRESS DECODE XC4000E/X
...
BUS LOGIC
ASYNCHRO- XC4000E/X
NOUS SLAVE
4.7 kΩ
PERIPHERAL
4.7 kΩ CS1
RS
WS
CONTROL RDY/BUSY
SIGNALS
INIT INIT
DONE DONE
REPROGRAM
PROGRAM PROGRAM
4.7 kΩ
X9028
Figure 59: Asynchronous Peripheral Mode Circuit Diagram

R
Write to LCA Read Status
WS/CS0 RS, CS0
RS, CS1 WS, CS1

1 TCA
3 TCD 7 4
2 TDC
READY D7
D0-D7
BUSY
CCLK
TWTRB 4
6 TBUSY
RDY/BUSY
DOUT Previous Byte D6 D7 D0 D1 D2
X6097

Effective Write time 1 TCA 100 ns
(CS0, WS=Low; RS, CS1=High)
Write
DIN setup time 2 TDC 60 ns
DIN hold time 3 TCD 0 ns
RDY/BUSY delay after end of 4 TWTRB 60 ns
Write or Read
RDY RDY/BUSY active after beginning 7 60 ns
of Read
RDY/BUSY Low output (Note 4) 6 TBUSY 2 9 CCLK
periods
Notes: 1. Configuration must be delayed until the INIT pins of all daisy-chained FPGAs are High.
2. The time from the end of WS to CCLK cycle for the new byte of data depends on the completion of previous byte processing
and the phase of the internal timing generator for CCLK.
3. CCLK and DOUT timing is tested in slave mode.
4. TBUSY indicates that the double-buffered parallel-to-serial converter is not yet ready to receive new data. The shortest
TBUSY occurs when a byte is loaded into an empty parallel-to-serial converter. The longest TBUSY occurs when a new word
is loaded into the input register before the second-level buffer has started shifting out data.
This timing diagram shows very relaxed requirements. Data need not be held beyond the rising edge of WS. RDY/BUSY will
go active within 60 ns after the end of WS. A new write may be asserted immediately after RDY/BUSY goes Low, but write
may not be terminated until RDY/BUSY has been High for one CCLK period.
Figure 60: Asynchronous Peripheral Mode Programming Switching Characteristics

R
Vcc T POR
RE-PROGRAM
>300 ns
PROGRAM
T PI
INIT
T ICCK TCCLK
<300 ns
M0, M1, M2
VALID DONE RESPONSE
(Required)
X1532
<300 ns
I/O
Master Modes (XC4000E/EX)

Description Symbol Min Max Units 6
M0 = High TPOR 10 40 ms
Power-On Reset M0 = Low TPOR 40 130 ms
CLB column
Master Modes (XC4000XL)

M0 = High TPOR 10 40 ms
Power-On Reset M0 = Low TPOR 40 130 ms
CLB column
Slave and Peripheral Modes(All)

CLB column
CCLK (input) Period (required) TCCLK 100 ns

XC4000E and XC4000X Series Field Programmable Gate Arrays January 29, 1999 (Version 1.5) 6
XC4000XL Switching Characteristics

Definition of Terms
follows:
Unmarked: Specifications not identified as either Advance or Preliminary are to be considered final.
All specifications subject to change without notice.
Additional Specifications
temperature conditions. The parameters included are common to popular designs and typical applications. For design
considerations requiring more detailed timing information, see the appropriate family a.c. supplements available on the
Xilinx WEBLINX at http://www.xilinx.com.

VCCt Longest Supply Voltage Rise Time from 1 V to 3V 50 ms
TJ
Note 1: Maximum DC overshoot or undershoot above Vcc or below GND must be limited to either 0.5 V or 10 mA, whichever is easier to
achieve. During transitions, the device pins may undershoot to -2.0 V or overshoot to + 7.0 V, provided this over- or undershoot lasts
less than 10 ns and with the forcing current being limited to 200 mA.
Note 2: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress rat-
ings only, and functional operation of the device at these or any other conditions beyond those listed under Operating Conditions is
not implied. Exposure to Absolute Maximum Ratings conditions for extended periods of time may affect device reliability.
Recommended Operating Conditions

Unit
Symbol Description Min Max
s
Supply voltage relative to GND, TJ = 0 °C to +85°C Commercial 3.0 3.6 V
VCC Supply voltage relative to GND, TJ = -40°C to Industrial 3.0 3.6 V
+100°C
Note 1: At junction temperatures above those listed as Operating Conditions, all delay parameters increase by 0.35% per °C.
Note 2: Input and output measurement threshold is ~50% of VCC.

R

VOH
VOL Low-level output voltage @ IOL = 12.0 mA, VCC min (LVTTL) (Note 1) 0.4 V
Input capacitance (sample tested) BGA, SBGA, PQ, HQ, MQ 10 pF
CIN packages
PGA packages 16 pF
IRPU Pad pull-up (when selected) @ Vin = 0 V (sample tested) 0.02 0.25 mA
IRPD Pad pull-down (when selected) @ Vin = 3.6 V (sample tested) 0.02 0.15 mA
IRLL Horizontal Longline pull-up (when selected) @ logic Low 0.3 2.0 mA
Note 1: With up to 64 pins simultaneously sinking 12 mA.
Note 2: With no output current loads, no active input or Longline pull-up resistors, all I/O pins Tri-stated and floating.
6

R

VOH
VOL
Input capacitance (sample tested) BGA, SBGA, PQ, HQ, MQ 10 pF
CIN packages
PGA packages 16 pF
XC4000XL Global Buffer Switching Characteristic Guidelines

temperature. Values apply to all XC4000XL devices and are expressed in nanoseconds unless otherwise noted.
XC4000XL Global Low Skew Buffer to Clock K
Speed Grade All -3 -2 -1 -09 -08
Units
Description Symbol Device Min Max Max Max Max Max
Delay from pad through GLS buffer to TGLS XC4002XL 0.3 2.1 1.8 1.6 1.5 ns
any clock input, K XC4005XL 0.4 2.7 2.3 2.0 1.9 ns
XC4010XL 0.5 3.2 2.8 2.4 2.3 ns
XC4013XL 0.6 3.6 3.1 2.7 2.6 2.3 ns
XC4020XL 0.7 4.0 3.5 3.0 2.9 ns
XC4028XL 0.9 4.4 3.8 3.3 3.2 ns
XC4036XL 1.1 4.8 4.2 3.6 3.5 3.1 ns
XC4044XL 1.2 5.3 4.6 4.0 3.9 ns
XC4052XL 1.3 5.7 5.0 4.5 4.4 ns
XC4062XL 1.4 6.3 5.4 4.7 4.6 4.0 ns
XC4085XL 1.6 7.2 6.2 5.7 5.5 ns

R
XC4000XL Global Buffer Switching Characteristic Guidelines

temperature. Values apply to all XC4000XL devices and are expressed in nanoseconds unless otherwise noted.
XC4000XL Global Early BUFGE #s 1, 2, 5, and 6 to IOB Clock
Units
Delay from pad through GE buffer to any TGE XC4002XL 0.1 1.6 1.4 1.3 1.2 ns
IOB clock input. XC4005XL 0.3 1.9 1.8 1.7 1.6 ns
XC4010XL 0.3 2.2 1.9 1.7 1.7 ns
XC4013XL 0.4 2.4 2.1 1.8 1.7 1.5 ns
XC4020XL 0.4 2.6 2.2 2.1 2.0 ns
XC4028XL 0.3 2.8 2.4 2.1 2.0 ns
XC4036XL 0.3 3.1 2.7 2.3 2.2 1.9 ns
XC4044XL 0.2 3.5 3.0 2.6 2.4 ns
6
XC4052XL 0.3 4.0 3.5 3.0 3.0 ns
XC4062XL 0.3 4.9 4.3 3.7 3.4 3.0 ns
XC4085XL 0.4 5.8 5.1 4.7 4.3 ns
XC4000XL Global Early BUFGE #s 3, 4, 7, and 8 to IOB Clock

Units
Delay from pad through GE buffer to any TGE XC4002XL 0.5 2.8 2.5 2.1 1.7 ns
IOB clock input. XC4005XL 0.7 3.1 2.8 2.7 2.5 ns
XC4010XL 0.7 3.5 3.1 2.8 2.7 ns
XC4013XL 0.7 3.8 3.3 2.9 2.8 2.4 ns
XC4020XL 0.8 4.1 3.6 3.4 3.2 ns
XC4028XL 0.9 4.4 3.9 3.4 3.3 ns
XC4036XL 0.9 4.7 4.2 3.7 3.6 3.1 ns
XC4044XL 1.0 5.1 4.5 4.0 3.7 ns
XC4052XL 1.1 5.5 4.8 4.3 4.3 ns
XC4062XL 1.2 5.9 5.2 4.8 4.5 4.0 ns
XC4085XL 1.3 6.8 6.0 5.5 5.2 ns

R
XC4000XL CLB Switching Characteristic Guidelines

XC4000XL devices and are expressed in nanoseconds unless otherwise noted.
Speed Grade -3 -2 -1 -09 -08
Description Symbol Min Max Min Max Min Max Min Max Min Max
F/G inputs to X/Y outputs TILO 1.6 1.5 1.3 1.2 1.1
F/G inputs via H’ to X/Y outputs TIHO 2.7 2.4 2.2 2.0 1.9
F/G inputs via transparent latch to Q outputs TITO 2.9 2.6 2.2 2.0 1.8
C inputs via SR/H0 via H to X/Y outputs THH0O 2.5 2.2 2.0 1.8 1.8
C inputs via H1 via H to X/Y outputs THH1O 2.4 2.1 1.9 1.6 1.5
C inputs via DIN/H2 via H to X/Y outputs THH2O 2.5 2.2 2.0 1.8 1.8
C inputs via EC, DIN/H2 to YQ, XQ output (bypass) TCBYP 1.5 1.3 1.1 1.0 0.9
Operand inputs (F1, F2, G1, G4) to COUT TOPCY 2.7 2.3 2.0 1.6 1.6
Add/Subtract input (F3) to COUT TASCY 3.3 2.9 2.5 1.8 1.8
Initialization inputs (F1, F3) to COUT TINCY 2.0 1.8 1.5 1.0 0.9
CIN through function generators to X/Y outputs TSUM 2.8 2.6 2.4 1.7 1.5
CIN to COUT, bypass function generators TBYP 0.26 0.23 0.20 0.14 0.14
Carry Net Delay, COUT to CIN TNET 0.32 0.28 0.25 0.24 0.24
Sequential Delays
Clock K to Flip-Flop outputs Q TCKO 2.1 1.9 1.6 1.5 1.4
Clock K to Latch outputs Q TCKLO 2.1 1.9 1.6 1.5 1.4
F/G inputs TICK 1.1 1.0 0.9 0.8 0.8
F/G inputs via H TIHCK 2.2 1.9 1.7 1.6 1.5
C inputs via H0 through H THH0CK 2.0 1.7 1.6 1.4 1.4
C inputs via DIN TDICK 0.9 0.8 0.7 0.6 0.6
C inputs via EC TECCK 1.0 0.9 0.8 0.7 0.7
C inputs via S/R, going Low (inactive) TRCK 0.6 0.5 0.5 0.4 0.4
CIN input via F/G TCCK 2.3 2.1 1.9 1.3 1.2
CIN input via F/G and H TCHCK 3.4 3.0 2.7 2.1 2.0
F/G inputs TCKI 0 0 0 0 0
F/G inputs via H TCKIH 0 0 0 0 0
C inputs via SR/H0 through H TCKHH0 0 0 0 0 0
C inputs via H1 through H TCKHH1 0 0 0 0 0
C inputs via DIN/H2 through H TCKHH2 0 0 0 0 0
C inputs via DIN/H2 TCKDI 0 0 0 0 0
C inputs via EC TCKEC 0 0 0 0 0
C inputs via SR, going Low (inactive) TCKR 0 0 0 0 0
Clock
Clock High time TCH 3.0 2.8 2.5 2.3 2.1
Clock Low time TCL 3.0 2.8 2.5 2.3 2.1
Set/Reset Direct
Width (High) TRPW 3.0 2.8 2.5 2.3 2.3
Delay from C inputs via S/R, going High to Q TRIO 3.7 3.2 2.8 2.7 2.6
Global Set/Reset
Minimum GSR Pulse Width TMRW 19.8 17.3 15.0 14.0 14.0
Toggle Frequency (MHz) (for export control) FTOG (MHz) 166 179 200 217 238

R
XC4000XL RAM Synchronous (Edge-Triggered) Write Operation Guidelines

Speed Grade -3 -2 -1 -09 -08

Single Port RAM
Size Symbol Min Max Min Max Min Max Min Max Min Max
Write Operation
Address write cycle time (clock K period) 16x2 TWCS 9.0 8.4 7.7 7.4 7.4
32x1 TWCTS 9.0 8.4 7.7 7.4 7.4
Clock K pulse width (active edge) 16x2 TWPS 4.5 4.2 3.9 3.7 3.7
32x1 TWPTS 4.5 4.2 3.9 3.7 3.7
Address setup time before clock K 16x2 TASS 2.2 2.0 1.7 1.7 1.6
32x1 TASTS 2.2 2.0 1.7 1.7 1.7
Address hold time after clock K 16x2 TAHS 0 0 0 0 0
32x1 TAHTS 0 0 0 0 0
DIN setup time before clock K 16x2 TDSS 2.0 1.9 1.7 1.7 1.7 6
32x1 TDSTS 2.5 2.3 2.1 2.1 2.1
DIN hold time after clock K 16x2 TDHS 0 0 0 0 0
32x1 TDHTS 0 0 0 0 0
WE setup time before clock K 16x2 TWSS 2.0 1.8 1.6 1.6 1.6
32x1 TWSTS 1.8 1.7 1.5 1.5 1.5
WE hold time after clock K 16x2 TWHS 0 0 0 0 0
32x1 TWHTS 0 0 0 0 0
Data valid after clock K 16x2 TWOS 6.8 6.3 5.8 5.8 5.7
32x1 TWOTS 8.1 7.5 6.9 6.7 6.7
Read Operation
Address read cycle time 16x2 TRC 4.5 3.1 2.6 2.6 2.6
32x1 TRCT 6.5 5.5 3.8 3.8 3.8
Data Valid after address change (no 16x2 TILO 1.6 1.5 1.3 1.2 1.1
Write Enable) 32x1 TIHO 2.7 2.4 2.2 2.0 1.9
Address setup time before clock K 16x2 TICK 1.1 1.0 0.9 0.8 0.8
32x1 TIHCK 2.2 1.9 1.7 1.6 1.5

R
XC4000XL CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines

Speed Grade -3 -2 --1 -09 -08

Dual Port RAM
Size Symbol Min Max Min Max Min Max Min Max Min Max
Address write cycle time (clock K period) 16x1 TWCDS 9.0 8.4 7.7 7.4 7.4
Clock K pulse width (active edge) 16x1 TWPDS 4.5 4.2 3.9 3.7 3.7
Address setup time before clock K 16x1 TASDS 2.5 2.0 1.7 1.7 1.6
Address hold time after clock K 16x1 TAHDS 0 0 0 0 0
DIN setup time before clock K 16x1 TDSDS 2.5 2.3 2.0 2.0 2.0
DIN hold time after clock K 16x1 TDHDS 0 0 0 0 0
WE setup time before clock K 16x1 TWSDS 1.8 1.7 1.6 1.6 1.6
WE hold time after clock K 16x1 TWHDS 0 0 0 0 0
Data valid after clock K 16x1 TWODS 7.8 7.3 6.7 6.7 6.6
XC4000XL CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS TWPDS
WCLK (K) WCLK (K)
WE WE
TDSS TDHS TDHDS

TDSDS
DATA IN DATA IN
ADDRESS
ADDRESS
TILO TILO TILO TILO

TWOS
TWODS
X6461 X6474

R
XC4000XL Pin-to-Pin Output Parameter Guidelines

the static timing analyzer report. Values are expressed in nanoseconds unless otherwise noted.
XC4000XL Output Flip-Flop, Clock to Out
Units
Global Low Skew Clock to Output us- TICKOF XC4002XL 1.2 7.1 6.1 5.4 5.1 ns
ing Output Flip Flop XC4005XL 1.3 7.7 6.6 5.8 5.4 ns
XC4010XL 1.4 8.2 7.1 6.2 5.8 ns
XC4013XL 1.5 8.6 7.4 6.5 6.1 5.6 ns
XC4020XL 1.6 9.0 7.8 6.8 6.4 ns
XC4028XL 1.8 9.4 8.1 7.1 6.7 ns
XC4036XL 2.0 9.8 8.5 7.4 7.0 6.4 ns
XC4044XL 2.1 10.3 8.9 7.8 7.4 ns
XC4052XL 2.2 10.7 9.3 8.3 7.9 ns
XC4062XL 2.3 11.3 9.7 8.5 8.1 7.3 ns
XC4085XL 2.5 12.2 10.5 9.5 9.0 ns 6
For output SLOW option add TSLOW All Devices 0.5 3.0 2.5 2.0 1.7 1.6 ns
Notes: Clock-to-out minimum delay is measured with the fastest route and the lightest load, Clock-to-out maximum delay is
measured using the farthest distance and a reference load of one clock pin (IK or OK) per IOB as well as driving all
accessible CLB flip-flops. For designs with a smaller number of clock loads, the pad-to-IOB clock pin delay as determined
by the static timing analyzer (TRCE) can be added to the AC parameter Tokpof and used as a worst-case pin-to-pin
clock-to-out delay for clocked outputs for FAST mode configurations.
Output timing is measured at ~50% VCC threshold with 50 pF external capacitive load. For different loads, see Figure 1.

2
Delta Delay (ns)
and load capacitance. It allows a user to adjust the speci-

1
120 pF, add 2.5 ns to the specified delay. If the load capac- 0
delay. -1
Figure 61 is usable over the specified operating conditions
of voltage and temperature and is independent of the out- -2
0 20 40 60 80 100 120 140
put slew rate control.
Capacitance (pF)
X8257

R
XC4000XL Pin-to-Pin Output Parameter Guidelines

XC4000XL Output Flip-Flop, Clock to Out, BUFGE #s 1, 2, 5, and 6
Units
Global Early Clock to Output using TICKEOF XC4002XL 1.0 6.6 5.7 5.1 4.8 ns
Output Flip Flop. Values are for BUF- XC4005XL 1.2 6.9 6.1 5.5 5.2 ns
GE #s 1, 2, 5, and 6. XC4010XL 1.2 7.2 6.2 5.5 5.3 ns
XC4013XL 1.3 7.4 6.4 5.6 5.3 4.8 ns
XC4020XL 1.3 7.6 6.5 5.9 5.6 ns
XC4028XL 1.2 7.8 6.7 5.9 5.6 ns
XC4036XL 1.2 8.1 7.0 6.1 5.8 5.2 ns
XC4044XL 1.1 8.5 7.3 6.4 6.0 ns
XC4052XL 1.2 9.0 7.8 6.8 6.6 ns
XC4062XL 1.2 9.9 8.6 7.5 7.0 6.3 ns
XC4085XL 1.3 10.8 9.4 8.5 7.9 ns
XC4000XL Output Flip-Flop, Clock to Out, BUFGE #s 3, 4, 7, and 8

Units
Global Early Clock to Output using TICKEOF XC4002XL 1.3 7.8 6.8 5.9 5.3 ns
Output Flip Flop. Values are for BUF- XC4005XL 1.5 8.1 7.1 6.5 6.1 ns
GE #s 3, 4, 7, and 8. XC4010XL 1.6 8.5 7.4 6.6 6.3 ns
XC4013XL 1.6 8.8 7.6 6.7 6.4 5.7 ns
XC4020XL 1.7 9.1 7.9 7.2 6.8 ns
XC4028XL 1.7 9.4 8.2 7.2 6.9 ns
XC4036XL 1.8 9.7 8.5 7.5 7.2 6.4 ns
XC4044XL 1.9 10.1 8.8 7.8 7.3 ns
XC4052XL 2.0 10.5 9.1 8.1 7.9 ns
XC4062XL 2.0 10.9 9.5 8.6 8.1 7.3 ns
XC4085XL 2.2 11.8 10.3 9.3 8.8 ns

R
XC4000XL Pin-to-Pin Input Parameter Guidelines

the static timing analyzer report. Values are expressed in nanoseconds unless otherwise noted
XC4000XL Global Low Skew Clock, Set-Up and Hold

Speed Grade -3 -2 -1 -09 -08
Units
Description Symbol Device Min Min Min Min Min
Input Setup and Hold Times
No Delay TPSN/TPHN XC4002XL 2.5 / 1.5 2.2 / 1.3 1.9 / 1.2 1.7 / 1.0 ns
Global Low Skew Clock and IFF XC4005XL 1.2 / 2.6 1.1 / 2.2 0.9 / 2.0 0.8 / 1.7 ns
Global Low Skew Clock and FCL XC4010XL 1.2 / 3.0 1.1 / 2.6 0.9 / 2.3 0.8 / 2.0 ns
XC4013XL 1.2 / 3.2 1.1 / 2.8 0.9 / 2.4 0.8 / 2.1 0.8 / 2.1 ns
XC4020XL 1.2 / 3.7 1.1 / 3.2 0.9 / 2.8 0.8 / 2.4 ns
XC4028XL 1.2 / 4.4 1.1 / 3.8 0.9 / 3.3 0.8 / 2.9 ns
XC4036XL 1.2 / 5.5 1.1 / 4.8 0.9 / 4.1 0.8 / 3.6 0.8 / 3.6 ns
XC4044XL 1.2 / 5.8 1.1 / 5.0 0.9 / 4.4 0.8 / 3.8 ns
XC4052XL 1.2 / 7.1 1.1 / 6.2 0.9 / 5.4 0.8 / 4.7 ns
XC4062XL 1.2 / 7.0 1.1 / 6.1 0.9 / 5.3 0.8 / 4.6 0.8 / 4.6 ns 6
XC4085XL 1.2 / 9.4 1.1 / 8.2 0.9 / 7.1 0.8 / 6.2 ns
Partial Delay TPSP/TPHP XC4002XL 8.4 / 0.0 7.3 / 0.0 6.3 / 0.0 5.5 / 0.0 ns
Global Low Skew Clock and IFF XC4005XL 10. 5 / 0.0 9.1 / 0.0 7.9 / 0.0 6.9 / 0.0 ns
Global Low Skew Clock and FCL XC4010XL 11.1 / 0.0 9.7 / 0.0 8.4 / 0.0 7.3 / 0.0 ns
XC4013XL* 6.1 / 1.0 5.3 / 1.0 4.6 / 1.0 4.0 / 1.0 3.7 / 0.5 ns
XC4020XL 11.9 / 1.0 10.3 / 1.0 9.0 / 1.0 7.8 / 1.0 ns
XC4028XL 12.3 / 1.0 10.7 / 1.0 9.3 / 1.0 8.1 / 1.0 ns
XC4036XL* 6.4 / 1.0 5.6 / 1.0 4.8 / 1.0 4.2 / 1.0 4.0/ 0.8 ns
XC4044XL 13.1 / 1.0 11.4 / 1.0 9.9 / 1.0 8.6 / 1.0 ns
XC4052XL 11.9 / 1.0 10.3 / 1.0 9.0 / 1.0 7.8 / 1.0 ns
XC4062XL* 6.7 / 1.2 5.8 / 1.2 5.1 / 1.2 4.4 / 1.2 4.2/ 1.0 ns
XC4085XL 12.9 / 1.2 11.2 / 1.2 9.8 / 1.2 8.5 / 1.2 ns
Full Delay TPSD/TPHD XC4002XL 6.8 / 0.0 6.0 / 0.0 5.2 / 0.0 4.5 / 0.0 ns
Global Low Skew Clock and IFF XC4005XL 8.8 / 0.0 7.6 / 0.0 6.6 / 0.0 5.6 / 0.0 ns
XC4010XL 9.0 / 0.0 7.8 / 0.0 6.8 / 0.0 5.8 / 0.0 ns
XC4013XL* 6.4 / 0.0 6.0 / 0.0 5.6 / 0.0 4.8 / 0.0 4.8 / 0.0 ns
XC4020XL 8.8 / 0.0 7.6 / 0.0 6.6 / 0.0 6.2 / 0.0 ns
XC4028XL 9.3 / 0.0 8.1 / 0.0 7.0 / 0.0 6.4 / 0.0 ns
XC4036XL* 6.6 / 0.0 6.2 / 0.0 5.8 / 0.0 5.3 / 0.0 5.3 / 0.0 ns
XC4044XL 10.6 / 0.0 9.2 / 0.0 8.0 / 0.0 6.8 / 0.0 ns
XC4052XL 11.2 / 0.0 9.7 / 0.0 8.4 / 0.0 7.0 / 0.0 ns
XC4062XL* 6.8 / 0.0 6.4 / 0.0 6.0 / 0.0 5.5 / 0.0 5.5 / 0.0 ns
XC4085XL 12.7 / 0.0 11.0 / 0.0 9.6 / 0.0 8.4 / 0.0 ns
Notes: Input setup time is measured with the fastest route and the lightest load.
Input hold time is measured using the furthest distance and a reference load of one clock pin per IOB as well as driving all
by the static timing analyzer (TRCE) can be used as a worst-case pin-to-pin no-delay input hold specification.
* The XC4013XL, XC4036XL, and 4062XL have significantly faster partial and full delay setup times than other devices.

R
XC4000XL BUFGE #s 1, 2, 5, and 6 Global Early Clock, Set-up and Hold for IFF and FCL
Speed Grade -3 -2 -1 -09 -08

Units
Input Setup and Hold
Times
No Delay XC4002XL 2.8 / 1.5 2.5 / 1.3 2.2 / 1.2 1.9 / 1.0 ns
Global Early Clock and IFF TPSEN/TPHEN XC4005XL 1.2 / 4.1 1.1 / 3.6 0.9 / 3.1 0.8 / 2.7 ns
Global Early Clock and TPFSEN/TPFHEN XC4010XL 1.2 / 4.4 1.1 / 3.8 0.9 / 3.3 0.8 / 2.9 ns
FCL XC4013XL 1.2 / 4.7 1.1 / 4.1 0.9 / 3.6 0.8 / 3.1 0.5 / 2.7 ns
XC4020XL 1.2 / 4.6 1.1 / 4.0 0.9 / 3.5 0.8 / 3.0 ns
XC4028XL 1.2 / 5.3 1.1 / 4.6 0.9 / 4.0 0.8 / 3.5 ns
XC4036XL 1.2 / 6.7 1.1 / 5.8 0.9 / 5.1 0.8 / 4.4 0.5 / 3.7 ns
XC4044XL 1.2 / 6.5 1.1 / 5.7 0.9 / 4.9 0.8 / 4.3 ns
XC4052XL 1.2 / 6.7 1.1 / 5.8 0.9 / 5.1 0.8 / 4.4 ns
XC4062XL 1.2 / 8.4 1.1 / 7.3 0.9 / 6.3 0.8 / 5.5 0.5 / 4.7 ns
XC4085XL 1.2 / 8.7 1.1 / 7.5 0.9 / 6.6 0.8 / 5.7 ns
Partial Delay XC4002XL 8.1 / 0.9 7.0 / 0.8 6.1 / 0.7 5.3 / 0.6 ns
Global Early Clock and IFF TPSEP/TPHEP XC4005XL 9.0 / 0.0 8.5 / 0.0 8.0 / 0.0 7.5 / 0.0 ns
Global Early Clock and TPFSEP/TPFHEP XC4010XL 11.9 / 0.0 10.4 / 0.0 9.0 / 0.0 8.0 / 0.0 ns
FCL XC4013XL* 6.4 / 0.0 5.9 / 0.0 5.4 / 0.0 4.9 / 0.0 4.4 / 0.0 ns
XC4020XL 10.8 / 0.0 10.3 / 0.0 9.8 / 0.0 9.0 / 0.0 ns
XC4028XL 14.0 / 0.0 12.2 / 0.0 10.6 / 0.0 9.8 / 0.0 ns
XC4036XL* 7.0 / 0.0 6.6 / 0.0 6.2 / 0.0 5.2 / 0.0 4.7 / 0.0 ns
XC4044XL 14.6 / 0.0 12.7 / 0.0 11.0 / 0.0 10.8 / 0.0 ns
XC4052XL 16.4 / 0.0 14.3 / 0.0 12.4 / 0.0 11.4 / 0.0 ns
XC4062XL* 9.0 / 0.8 8.6 / 0.8 8.2 / 0.8 7.0 / 0.8 6.3 / 0.5 ns
XC4085XL 16.7 / 0.0 14.5 / 0.0 12.6 / 0.0 11.6 / 0.0 ns
Full Delay XC4002XL 6.7 / 0.0 5.8 / 0.0 5.1 / 0.0 4.4 / 0.0 ns
Global Early Clock and IFF TPSED/TPHED XC4005XL 10.8 / 0.0 9.4 / 0.0 8.2 / 0.0 7.1 / 0.0 ns
XC4010XL 10.3 / 0.0 9.0 / 0.0 7.8 / 0.0 6.8 / 0.0 ns
XC4013XL* 10.0 / 0.0 8.7 / 0.0 7.6 / 0.0 6.6 / 0.0 6.0 / 0.0 ns
XC4020XL 12.0 / 0.0 10.4 / 0.0 9.1 / 0.0 7.9 / 0.0 ns
XC4028XL 12.6 / 0.0 11.0 / 0.0 9.5 / 0.0 8.3 / 0.0 ns
XC4036XL* 12.2 / 0.0 10.6 / 0.0 9.2 / 0.0 8.0 / 0.0 7.2 / 0.0 ns
XC4044XL 13.8 / 0.0 12.0 / 0.0 10.5 / 0.0 9.1 / 0.0 ns
XC4052XL 14.1 / 0.0 12.3 / 0.0 10.7 / 0.0 9.3 / 0.0 ns
XC4062XL* 13.1 / 0.0 11.4 / 0.0 9.9 / 0.0 8.6 / 0.0 7.8 / 0.0 ns
XC4085XL 17.9 / 0.0 15.6 / 0.0 13.6 / 0.0 11.8 / 0.0 ns
IFF = Input Flip-Flop or Latch, FCL = Fast Capture Latch

R
XC4000XL BUFGE #s 3, 4, 7, and 8 Global Early Clock, Set-up and Hold for IFF and FCL
Speed Grade -3 -2 -1 -09 -08

Units
Input Setup & Hold Times
No Delay XC4002XL 3.0 / 2.0 2.6 / 1.7 2.3 / 1.5 2.0 / 1.3 ns
Global Early Clock and TPSEN/TPHEN XC4005XL 1.2 / 4.1 1.1 / 3.6 0.9 / 3.1 0.8 / 2.7 ns
IFF TPFSEN/TPFHEN XC4010XL 1.2 / 4.4 1.1 / 3.8 0.9 / 3.3 0.8 / 2.9 ns
Global Early Clock and XC4013XL 1.2 / 4.7 1.1 / 4.1 0.9 / 3.6 0.8 / 3.1 0.5 / 2.7 ns
FCL XC4020XL 1.2 / 4.6 1.1 / 4.0 0.9 / 3.5 0.8 / 3.0 ns
XC4028XL 1.2 / 5.3 1.1 / 4.6 0.9 / 4.0 0.8 / 3.5 ns
XC4036XL 1.2 / 6.7 1.1 / 5.8 0.9 / 5.1 0.8 / 4.4 0.5 / 3.7 ns
XC4044XL 1.2 / 6.5 1.1 / 5.7 0.9 / 4.9 0.8 / 4.3 ns
XC4052XL 1.2 / 6.7 1.1 / 5.8 0.9 / 5.1 0.8 / 4.4 ns
XC4062XL 1.2 / 8.4 1.1 / 7.3 0.9 / 6.3 0.8 / 5.5 0.5 / 4.7 ns
XC4085XL 1.2 / 8.7 1.1 / 7.5 0.9 / 6.6 0.8 / 5.7 ns
Partial Delay XC4002XL 7.3 / 1.5 6.4 / 1.3 5.5 / 1.2 4.8 / 1.0 ns 6
Global Early Clock and TPSEP/TPHEP XC4005XL 8.4 / 0.0 7.9 / 0.0 7.4 / 0.0 7.2 / 0.0 ns
IFF TPFSEP/TPFHEP XC4010XL 10.3 / 0.0 9.0 / 0.0 7.8 / 0.0 7.4 / 0.0 ns
Global Early Clock and XC4013XL* 5.4 / 0.0 4.9 / 0.0 4.4 / 0.0 4.3 / 0.0 4.0 / 0.0 ns
FCL XC4020XL 9.8 / 0.0 9.3 / 0.0 8.8 / 0.0 8.5 / 0.0 ns
XC4028XL 12.7 / 0.0 11.0 / 0.0 9.6 / 0.0 9.3 / 0.0 ns
XC4036XL* 6.4 / 0.8 5.9 / 0.8 5.4 / 0.8 5.0 / 0.8 4.6 / 0.2 ns
XC4044XL 13.8 / 0.0 12.0 / 0.0 10.4 / 0.0 10.2 / 0.0 ns
XC4052XL 14.5 / 0.0 12.7 / 0.0 11.0 / 0.0 10.7 / 0.0 ns
XC4062XL* 8.4 / 1.5 7.9 / 1.5 7.4 / 1.5 6.8 / 1.5 6.2 / 0.0 ns
XC4085XL 14.5 / 0.0 12.7 / 0.0 11.0 / 0.0 10.8 / 0.0 ns
Full Delay XC4002XL 5.9 / 0.0 5.2 / 0.0 4.5 / 0.0 3.9 / 0.0 ns
Global Early Clock and TPSED/TPHED XC4005XL 10.8 / 0.0 9.4 / 0.0 8.2 / 0.0 7.1 / 0.0 ns
IFF XC4010XL 10.3 / 0.0 9.0 / 0.0 7.8 / 0.0 6.8 / 0.0 ns
XC4013XL* 10.0 / 0.0 8.7 / 0.0 7.6 / 0.0 6.6 / 0.0 6.0 / 0.0 ns
XC4020XL 12.0 / 0.0 10.4 / 0.0 9.1 / 0.0 7.9 / 0.0 ns
XC4028XL 12.6 / 0.0 11.0 / 0.0 9.5 / 0.0 8.3 / 0.0 ns
XC4036XL* 12.2 / 0.0 10.6 / 0.0 9.2 / 0.0 8.0 / 0.0 7.2 / 0.0 ns
XC4044XL 13.8 / 0.0 12.0 / 0.0 10.5 / 0.0 9.1 / 0.0 ns
XC4052XL 14.1 / 0.0 12.3 / 0.0 10.7 / 0.0 9.3 / 0.0 ns
XC4062XL* 13.1 / 0.0 11.4 / 0.0 9.9 / 0.0 8.6 / 0.0 7.8 / 0.0 ns
XC4085XL 17.9 / 0.0 15.6 / 0.0 13.6 / 0.0 11.8 / 0.0 ns
IFF = Input Flip Flop or Latch. FCL = Fast Capture Latch

R
XC4000XL IOB Input Switching Characteristic Guidelines

Speed Grade -3 -2 -1 -09 -08

Units
Clocks
Clock Enable (EC) to Clock (IK) TECIK All devices 0.1 0.1 0.1 0.1 0.1 ns
Delay from FCL enable (OK) active TOKIK XC4002XL 3.0 2.7 2.3 2.3 ns
edge to IFF clock (IK) active edge XC4013, 36, 62XL 2.2 1.9 1.6 1.6 1.6 ns
Balance of Family 2.2 1.9 1.6 1.6 ns
Setup Times
Pad to Clock (IK), no delay TPICK XC4002XL 2.6 2.3 2.0 2.0 ns
XC4013, 36, 62XL 1.7 1.5 1.3 1.3 1.2 ns
Pad to Clock (IK), via transparent Fast TPICKF XC4002XL 3.2 2.9 2.5 2.4 ns
Capture Latch, no delay XC4013, 36, 62XL 2.3 2.0 1.8 1.7 1.6 ns
Pad to Fast Capture Latch Enable (OK), TPOCK XC4013, 36, 62XL 1.2 1.0 0.9 0.9 0.9 ns
no delay Balance of Family 1.2 1.0 0.9 0.9 ns
Hold Times
All Hold Times All Devices 0 0 0 0 0
Global Set/Reset
Minimum GSR Pulse Width TMRW All devices 19.8 17.3 15.0 14.0 14.0 ns
Global Set/Reset Max Max Max Max Max
Delay from GSR input to any Q TRRI* XC4002XL 9.8 8.5 7.4 7.0 ns
XC4005XL 11.3 9.8 8.5 8.1 ns
XC4010XL 13.9 12.1 10.5 10.0 ns
XC4013XL 15.9 13.8 12.0 11.4 10.9 ns
XC4020XL 18.6 16.1 14.0 13.3 ns
XC4028XL 20.5 17.9 15.5 14.3 ns
XC4036XL 22.5 19.6 17.0 16.2 16.2 ns
XC4044XL 25.1 21.9 19.0 18.1 ns
XC4052XL 27.2 23.6 20.5 19.5 ns
XC4062XL 29.1 25.3 22.0 20.9 20.4 ns
XC4085XL 34.4 29.9 26.0 24.7 ns
* Indicates Minimum Amount of Time to Assure Valid Data. IFF = Input Flip-Flop or Latch, FCL = Fast Capture Latch.

R
XC4000XL IOB Input Switching Characteristic Guidelines (Cont)

Speed Grade -3 -2 -1 -09 -08

Units
Description Symbol Device Max Max Max Max Max
Propagation Delays
Pad to I1, I2 TPID All devices 1.6 1.4 1.2 1.1 1.0 ns
Pad to I1, I2 via transparent input latch, TPLI XC4002XL 4.7 4.2 3.6 3.5 ns
no delay XC4013, 36, 62XL 3.1 2.7 2.4 2.2 2.1 ns
Pad to I1, I2 via transparent FCL and in- TPFLI X4002XL 5.4 4.7 4.1 3.9 ns
put latch, no delay XC4013, 36, 62XL 3.7 3.3 2.8 2.7 2.5 ns
Clock (IK) to I1, I2 (flip-flop) TIKRI All devices 1.7 1.5 1.3 1.2 1.2 ns
Clock (IK) to I1, I2 (latch enable, active TIKLI All devices 1.8 1.6 1.4 1.3 1.3 ns
Low) ns
FCL Enable (OK) active edge to I1, I2 TOKLI XC4002XL 5.2 4.6 4.0 3.8 ns
(via transparent standard input latch) XC4013, 36, 62XL 3.6 3.1 2.7 2.6 2.5 ns
6
Balance of Family 3.6 3.1 2.7 2.6

R
XC4000XL IOB Output Switching Characteristic Guidelines

worst-case operating conditions (supply voltage and junction temperature). For Propagation Delays, slew-rate = fast unless
otherwise noted. Values are expressed in nanoseconds unless otherwise noted.
-3 -2 -1 -09 -08
Description Symbol Min Max Min Max Min Max Min Max Min Max
Clocks
Clock High TCH 3.0 2.8 2.5 2.3 2.1
Clock Low TCL 3.0 2.8 2.5 2.3 2.1
Propagation Delays
Clock (OK) to Pad TOKPOF 5.0 4.3 3.8 3.5 3.3
Output (O) to Pad TOPF 4.1 3.6 3.1 3.0 2.8
3-state to Pad hi-Z (slew-rate independent) TTSHZ 4.0 3.5 3.0 2.9 2.9
3-state to Pad active and valid TTSONF 4.4 3.8 3.3 3.3 3.3
Output (O) to Pad via Fast Output MUX TOFPF 5.5 4.8 4.2 4.0 3.7
Select (OK) to Pad via Fast MUX TOKFPF 5.1 4.5 3.9 3.7 3.4
Output (O) to clock (OK) setup time TOOK 0.5 0.4 0.3 0.3 0.3
Output (O) to clock (OK) hold time TOKO 0.0 0.0 0.0 0.0 0.0
Clock Enable (EC) to clock (OK) setup time TECOK 0.0 0.0 0.0 0.0 0.0
Clock Enable (EC) to clock (OK) hold time TOKEC 0.3 0.2 0.1 0.0 0.0
Global Set/Reset
Minimum GSR pulse width TMRW 19.8 17.3 15.0 14.0 14.0
Delay from GSR input to any Pad TRPO*
XC4002XL 14.3 12.5 10.9 10.3
XC4005XL 15.9 13.8 12.0 11.4
XC4010XL 18.5 16.1 14.0 13.3
XC4013XL 20.5 17.8 15.5 14.7 14.0
XC4020XL 23.2 20.1 17.5 16.6
XC4028XL 25.1 21.9 19.0 17.6
XC4036XL 27.1 23.6 20.5 19.4 19.3
XC4044XL 29.7 25.9 22.5 21.4
XC4052XL 31.7 27.6 24.0 22.8
XC4062XL 33.7 29.3 25.5 24.2 23.5
XC4085XL 39.0 33.9 29.5 28.0
For output SLOW option add TSLOW 3.0 2.5 2.0 1.7 1.6
Note: Output timing is measured at ~50% VCC threshold, with 50 pF external capacitive loads.
* Indicates Minimum Amount of Time to Assure Valid Data.

R
XC4000EX Switching Characteristics

Definition of Terms
follows:
families. Values are subject to change. Use as estimates, not for production.
XC4000EX Absolute Maximum Ratings

VIN Input voltage relative to GND (Note 1) -0.5 to VCC +0.5 V
VTS Voltage applied to 3-state output (Note 1) -0.5 to VCC +0.5 V
VCCt Longest Supply Voltage Rise Time from 1 V to 4 V 50 ms
Junction temperature Ceramic packages +150 °C 6
TJ
achieve. During transitions, the device pins may undershoot to -2.0 V or overshoot to Vcc + 2.0 V, provided this over- or undershoot
lasts less than 20 ns.
XC4000EX Recommended Operating Conditions

VCC
High-level input voltage TTL inputs 2.0 VCC V
VIH
Low-level input voltage TTL inputs 0 0.8 V
VIL
Note 2: Input and output measurement thresholds for TTL are 1.5 V. Input and output measurement thresholds for CMOS are 2.5 V.
Note 3: All timing parameters are specified for Commercial temperature range only.

R
XC4000EX DC Characteristics Over Recommended Operating Conditions

High-level output voltage @ IOH = -4.0 mA, VCC min TTL outputs 2.4 V
VOH
High-level output voltage @ IOH = -1.0 mA CMOS outputs VCC-0.5 V
Low-level output voltage @ IOL = 12.0 mA, VCC min TTL outputs 0.4 V
VOL (Note 1) CMOS outputs 0.4 V
Input capacitance (sample tested) BGA, SBGA, PQ, 10 pF
CIN HQ, MQ packages
PGA packages 16 pF
Note 2: With no output current loads, no active input or Longline pull-up resistors, all package pins at Vcc or GND.
XC4000EX Global Buffer Switching Characteristic Guidelines

temperature).
Speed Grade -4 -3 -2
Units
Description Symbol Device Max Max Max
From pad through Global Low Skew buffer, TGLS XC4028EX 9.2 7.5 6.4 ns
to any clock K XC4036EX 9.8 7.9 7.1 ns
From pad through Global Early buffer, TGE XC4028EX 5.7 4.4 4.2 ns
to any clock K in same quadrant XC4036EX 5.9 4.6 4.4 ns

R
XC4000EX Longline and Wide Decoder Timing Guidelines

worst-case operating conditions (supply voltage and junction temperature). Values apply to all XC4000EX devices unless
otherwise noted. Fewer than the specified number of pullup resistors can be used, if desired. Using fewer pullups reduces
power consumption but increases delays. Use the static timing analyzer to determine delays if fewer pullups are used.
XC4000EX Horizontal Longline Switching Characteristic Guidelines
Units
TBUF driving a Horizontal Longline
I going High or Low to Horizontal Longline going High TIO1 XC4028EX 13.7 11.3 10.9 ns
or Low, while T is Low. Buffer is constantly active. XC4036EX 16.5 13.6 13.2 ns
T going Low to Horizontal Longline going from resis- TON XC4028EX 14.7 12.1 11.7 ns
tive pull-up or floating High to active Low. TBUF con- XC4036EX 17.4 14.4 14.0 ns
figured as open-drain or active buffer with I = Low.
TBUF driving Half a Horizontal Longline
I going High or Low to half of a Horizontal Longline go- THIO1 XC4028EX 6.3 5.6 4.6 ns
ing High or Low, while T is Low. Buffer is constantly XC4036EX 7.3 6.0 5.7 ns 6
active.
T going Low to half of a Horizontal Longline going from THON XC4028EX 7.2 6.4 5.4 ns
resistive pull-up or floating High to active Low. TBUF XC4036EX 8.2 6.8 6.5 ns
configured as open-drain or active buffer with I = Low.
Note 1: These values include a minimum load of one output, spaced as far as possible from the activated pullup(s). Use the statictiming ana-
lyzer to determine the delay for each destination.

R
XC4000EX CLB Switching Characteristic Guidelines

XC4000EX devicees unless otherwise noted.
Units
Description Symbol Min Max Min Max Min Max
F/G inputs to X/Y outputs TILO 2.2 1.8 1.5 ns
F/G inputs via H’ to X/Y outputs TIHO 3.8 3.2 2.7 ns
F/G inputs via transparent latch to Q outputs TITO 3.2 2.7 2.5 ns
C inputs via SR/H0 via H’ to X/Y outputs THH0O 3.6 3.0 2.5 ns
C inputs via H1 via H’ to X/Y outputs THH1O 3.0 2.5 2.3 ns
C inputs via DIN/H2 via H’ to X/Y outputs THH2O 3.6 3.0 2.5 ns
C inputs via EC, DIN/H2 to YQ, XQ output (bypass) TCBYP 2.0 1.6 1.4 ns
Operand inputs (F1, F2, G1, G4) to COUT TOPCY 2.5 2.2 1.9 ns
Add/Subtract input (F3) to COUT TASCY 4.1 3.6 3.1 ns
Initialization inputs (F1, F3) to COUT TINCY 1.9 1.6 1.4 ns
CIN through function generators to X/Y outputs TSUM 3.0 2.6 2.2 ns
CIN to COUT, bypass function generators TBYP 0.60 0.50 0.40 ns
Carry Net Delay, COUT to CIN TNET 0.18 0.15 0.15 ns
Sequential Delays
Clock K to Flip-Flop outputs Q TCKO 2.2 1.9 1.7 ns
Clock K to Latch outputs Q TCKLO 2.2 1.9 1.7 ns
F/G inputs TICK 1.3 1.1 1.1 ns
F/G inputs via H’ TIHCK 3.0 2.5 2.2 ns
C inputs via H0 through H’ THH0CK 2.8 2.3 2.0 ns
C inputs via DIN TDICK 1.2 0.9 0.9 ns
C inputs via EC TECCK 1.2 1.0 0.9 ns
C inputs via S/R, going Low (inactive) TRCK 0.8 0.7 0.6 ns
CIN input via F’/G’ TCCK 2.2 1.8 2.1 ns
CIN input via F’/G’ and H’ TCHCK 3.9 3.2 3.2 ns
F/G inputs TCKI 0 0 0 ns
F/G inputs via H’ TCKIH 0 0 0 ns
C inputs via SR/H0 through H’ TCKHH0 0 0 0 ns
C inputs via H1 through H’ TCKHH1 0 0 0 ns
C inputs via DIN/H2 through H’ TCKHH2 0 0 0 ns
C inputs via DIN/H2 TCKDI 0 0 0 ns
C inputs via EC TCKEC 0 0 0 ns
C inputs via SR, going Low (inactive) TCKR 0 0 0 ns
Clock
Clock High time TCH 3.5 3.0 3.0 ns
Clock Low time TCL 3.5 3.0 3.0 ns
Set/Reset Direct
Width (High) TRPW 3.5 3.0 3.0 ns
Delay from C inputs via S/R, going High to Q TRIO 4.5 3.8 3.6 ns
Global Set/Reset
Minimum GSR Pulse Width TMRW 13.0 11.5 11.5 ns
Delay from GSR input to any Q (XC4028EX) TMRQ 22.8 19.0 19.0 ns
Delay from GSR input to any Q (XC4036EX) TMRQ 24.0 21.0 21.0 ns
Toggle Frequency ) (for export control purposes) FTOG 143 166 166 MHz

R
XC4000EX CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines

Testing of switching parameters is modeled after testing methods specified by MIL-M-38510/605. All devices are 100% functionally tested.
Internal timing parameters are derived from measuring internal test patterns. Listed below are representative values. For more specific, more
precise, and worst-case guaranteed data, use the values reported by the static timing analyzer (TRCE in the Xilinx Development System) and
back-annotated to the simulation netlist. All timing parameters assume worst-case operating conditions (supply voltage and junction
temperature). Values apply to all XC4000EX devices unless otherwise noted.
Size Symbol Min Max Min Max Min Max
Write Operation
Address write cycle time (clock K period) 16x2 TWCS 11.0 9.0 9.0 ns
32x1 TWCTS 11.0 9.0 9.0 ns
Clock K pulse width (active edge) 16x2 TWPS 5.5 4.5 4.5 ns
32x1 TWPTS 5.5 4.5 4.5 ns
Address setup time before clock K 16x2 TASS 2.7 2.3 2.2 ns
32x1 TASTS 2.6 2.2 2.2 ns
Address hold time after clock K 16x2 TAHS 0 0 0 ns
32x1 TAHTS 0 0 0 ns
DIN setup time before clock K 16x2 TDSS 2.4 2.0 2.0 ns
32x1 TDSTS 2.9 2.5 2.5 ns 6
DIN hold time after clock K 16x2 TDHS 0 0 0 ns
32x1 TDHTS 0 0 0 ns
WE setup time before clock K 16x2 TWSS 2.3 2.0 2.0 ns
32x1 TWSTS 2.1 1.8 1.8 ns
WE hold time after clock K 16x2 TWHS 0 0 0 ns
32x1 TWHTS 0 0 0 ns
Data valid after clock K 16x2 TWOS 8.2 6.8 6.8 ns
32x1 TWOTS 10.1 8.4 8.2 ns
Note 1: Timing for the 16x1 RAM option is identical to 16x2 RAM timing.
Note 2: Applicable Read timing specifications are identical to Level-Sensitive Read timing.
Dual-Port RAM Units
Write Operation
Address write cycle time (clock K period) 16x1 TWCDS 11.0 9.0 9.0 ns
Clock K pulse width (active edge) 16x1 TWPDS 5.5 4.5 4.5 ns
Address setup time before clock K 16x1 TASDS 3.1 2.6 2.5 ns
Address hold time after clock K 16x1 TAHDS 0 0 0 ns
DIN setup time before clock K 16x1 TDSDS 2.9 2.5 2.5 ns
DIN hold time after clock K 16x1 TDHDS 0 0 0 ns
WE setup time before clock K 16x1 TWSDS 2.1 1.8 1.8 ns
WE hold time after clock K 16x1 TWHDS 0 0 0 ns
Data valid after clock K 16x1 TWODS 9.4 7.8 7.8 ns

R
XC4000EX CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS
WCLK (K)
TWSS TWHS
WE
TDSS TDHS
DATA IN
TASS TAHS
ADDRESS
TILO TILO
TWOS
DATA OUT OLD NEW
X6461
XC4000EX CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing
TWPDS
WCLK (K)
TWSDS TWHDS
WE
TDSDS TDHDS
DATA IN
TASDS TAHDS
ADDRESS
TILO TILO
TWODS
DATA OUT OLD NEW
X6474

R
XC4000EX CLB RAM Asynchronous (Level-Sensitive) Write and Read Operation Guidelines
XC4000EX devices unless otherwise noted.
Units
Description Size Symbol Min Max Min Max Min Max
Write Operation
Address write cycle time 16x2 TWC 10.6 9.2 8.0 ns

32x1 TWCT 10.6 9.2 8.0 ns
Write Enable pulse width (High) 16x2 TWP 5.3 4.6 4.0 ns
32x1 TWPT 5.3 4.6 4.0 ns
Address setup time before WE 16x2 TAS 2.8 2.4 2.0 ns

32x1 TAST 2.9 2.5 2.0 ns
Address hold time after end of WE 16x2 TAH 1.7 1.4 1.4 ns
32x1 TAHT 1.7 1.4 1.4 ns
DIN setup time before end of WE 16x2 TDS 1.1 0.9 0.8 ns 6
32x1 TDST 1.1 0.9 0.8 ns
DIN hold time after end of WE 16x2 TDH 6.6 5.7 5.0 ns
32x1 TDHT 6.6 5.7 5.0 ns
Read Operation
Address read cycle time 16x2 TRC 4.5 3.1 3.1 ns

32x1 TRCT 6.5 5.5 5.5 ns
Data valid after address change 16x2 TILO 2.2 1.8 1.5 ns
(no Write Enable) 32x1 TIHO 3.8 3.2 2.7 ns
Read Operation, Clocking Data into Flip-Flop
Address setup time before clock K 16x2 TICK 1.5 1.2 1.2 ns
32x1 TIHCK 3.2 2.6 2.6 ns
Read During Write
Data valid after WE goes active 16x2 TWO 6.5 5.7 4.9 ns
(DIN stable before WE) 32x1 TWOT 7.4 6.5 5.6 ns
Data valid after DIN 16x2 TDO 7.7 6.7 5.8 ns

(DIN changes during WE) 32x1 TDOT 8.2 7.2 6.2 ns
Read During Write, Clocking Data into Flip-Flop
WE setup time before clock K 16x2 TWCK 7.1 6.2 5.5 ns

32x1 TWCKT 9.2 8.1 7.0 ns
Data setup time before clock K 16x2 TDCK 5.9 5.2 4.6 ns
32x1 TDCKT 8.4 7.4 6.4 ns

R
XC4000EX CLB RAM Asynchronous (Level-Sensitive) Timing Characteristics

T WC
ADDRESS
WRITE
TAS T WP T AH
WRITE ENABLE
T DS T DH
DATA IN REQUIRED
READ WITHOUT WRITE

T ILO
X,Y OUTPUTS VALID VALID
READ, CLOCKING DATA INTO FLIP-FLOP

T ICK T CH
CLOCK
T CKO
VALID VALID
XQ, YQ OUTPUTS
(OLD) (NEW)
READ DURING WRITE

T WP
WRITE ENABLE
T DH
DATA IN
(stable during WE)
T WO
X, Y OUTPUTS VALID VALID
DATA IN
(changing during WE) OLD NEW
T WO T DO
VALID VALID VALID

X, Y OUTPUTS
(PREVIOUS) (OLD) (NEW)
READ DURING WRITE, CLOCKING DATA INTO FLIP-FLOP

T WP
WRITE ENABLE
T WCK
T DCK
DATA IN
CLOCK
T CKO
XQ, YQ OUTPUTS
X2640

R
XC4000EX Pin-to-Pin Output Parameter Guidelines

the static timing analyzer report. Values apply to all XC4000EX devices unless otherwise noted.
XC4000EX Output Flip-Flop, Clock to Out
Units
Global Low Skew Clock to TTL TICKOF XC4028EX 16.6 13.7 12.4 ns
Output (fast) using OFF XC4036EX 17.2 14.1 13.1 ns
Global Early Clock to TTL Output (fast) using TICKEOF XC4028EX 13.1 10.6 10.2 ns
OFF XC4036EX 13.3 10.8 10.4 ns
OFF = Output Flip Flop
XC4000EX Output MUX, Clock to Out

Units
Description Symbol Device Max Max Max 6
Global Low Skew Clock to TTL TPFPF XC4028EX 15.9 13.1 11.8 ns
Output (fast) using OMUX XC4036EX 16.5 13.5 12.5 ns
Global Early Clock to TTL Output (fast) us- TPEFPF XC4028EX 12.4 10.0 9.6 ns
ing OMUX XC4036EX 12.6 10.2 9.8 ns
OMUX = Output MUX
Note 1: Listed above are representative values where one global clock input drives one vertical clock line in each accessible column, and
where all accessible IOB and CLB flip-flops are clocked by the global clock net.
Note 2: Output timing is measured at TTL threshold with 50 pF external capacitive load.
Note 3: Set-up time is measured with the fastest route and the lightest load. Hold time is measured using the farthest distance and a refer-
ence load of one clock pin per two IOBs. Use the static timing analyzer to determine the setup and hold times under given design
conditions.
XC4000EX Output Level and Slew Rate Adjustments

The following table must be used to adjust output parameters and output switching characteristics.
Description Symbol Device Max Max Max Units
For TTL output FAST add TTTLOF All Devices 0 0 0 ns
For TTL output SLOW add TTTLO All Devices 2.9 2.4 2.4 ns
For CMOS FAST output add TCMOSOF All Devices 1.0 0.8 0.8 ns
For CMOS SLOW output add TCMOSO All Devices 3.6 3.0 3.0 ns

R
XC4000EX Pin-to-Pin Input Parameter Guidelines

the static timing analyzer report. Values apply to all XC4000EX devices unless otherwise noted
XC4000EX Global Low Skew Clock, Set-Up and Hold

Units
Description Symbol Device Min Min Min
Input Setup Time, using Global Low Skew TPSD XC4028EX 8.0 6.8 6.8 ns
clock and IFF (full delay) XC4036EX 8.0 6.8 6.8 ns
Input Hold Time, using Global Low Skew TPHD XC4028EX 0 0 0 ns
clock and IFF (full delay) XC4036EX 0 0 0 ns
IFF = Flip-Flop or Latch
XC4000EX Global Early Clock, Set-Up and Hold for IFF

Units
Input Setup Time, using Global Early clock TPSEP XC4028EX 6.5 5.4 5.4 ns
and IFF (partial delay) XC4036EX 6.5 5.4 5.4 ns
Input Hold Time, using Global Early clock TPHEP XC4028EX 0 0 0 ns
and IFF (partial delay) XC4036EX 0 0 0 ns
Note 1: Set-up parameters are for BUFGE #s 3, 4, 7 and 8. Add 1.6 ns for BUFGE #s 1, 2, 5 and 6.
XC4000EX Global Early Clock, Set-Up and Hold for FCL

Units
Input Setup Time, using Global Early clock TPFSEP XC4028EX 3.4 3.4 3.4 ns
and FCL (partial delay) XC4036EX 4.4 4.2 4.2 ns
Input Hold Time, using Global Early clock TPFHEP XC4028EX 0 0 0 ns
and FCL (partial delay) XC4036EX 0 0 0 ns
FCL = Fast Capture Latch
Note 1: For CMOS input levels, see the “XC4000EX Input Threshold Adjustments” on page 96.
Setup time is measured with the fastest route and the lightest load. Use the static timing analyzer to determine the setup time
Note 2: under given design conditions. Hold time is measured using the farthest distance and a reference load of one clock pin per two IOBs.
Use the static timing analyzer to determine the setup and hold times under given design conditions.
XC4000EX Input Threshold Adjustments

The following table must be used to adjust input parameters and input switching characteristics.
Description Symbol Device Max Max Max Units
For TTL input add TTTLI All Devices 0 0 0 ns
For CMOS input add TCMOSI All Devices 0.3 0.2 0.2 ns

R
XC4000EX IOB Input Switching Characteristic Guidelines

otherwise noted.
Units
Clocks
Delay from FCL enable (OK) active edge to IFF TOKIK All devices 3.2 2.6 2.6 ns
clock (IK) active edge
Propagation Delays Max Max Max
Pad to I1, I2 TPID All devices 2.2 1.9 1.8 ns
Pad to I1, I2 via transparent input latch, no delay TPLI All devices 3.8 3.2 3.0 ns
Pad to I1, I2 via transparent input latch, TPPLI XC4028EX 13.3 11.1 10.9 ns
partial delay XC4036EX 14.5 12.1 11.9 ns
Pad to I1, I2 via transparent input latch, full delay TPDLI XC4028EX 18.2 15.2 14.9 ns
XC4036EX 19.4 16.2 15.9 ns
Pad to I1, I2 via transparent FCL and input latch, TPFLI All devices 5.3 4.4 4.2 ns
no delay 6
Pad to I1, I2 via transparent FCL and input latch, TPPFLI XC4028EX 13.6 11.3 11.1 ns
Propagation Delays
Clock (IK) to I1, I2 (flip-flop) TIKRI All devices 3.0 2.5 2.4 ns
Clock (IK) to I1, I2 (latch enable, active Low) TIKLI All devices 3.2 2.7 2.6 ns
FCL Enable (OK) active edge to I1, I2 TOKLI All devices 6.2 5.2 5.0 ns
(via transparent standard input latch)
Global Set/Reset
Minimum GSR Pulse Width TMRW All devices 13.0 11.5 11.5 ns
Delay from GSR input to any Q TRRI XC4028EX 22.8 19.0 19.0 ns
Delay from GSR input to any Q TRRI XC4036EX 24.0 21.0 21.0 ns
FCL = Fast Capture Latch, IFF = Input Flip-Flop or Latch
Note 2: For setup and hold times with respect to the clock input pin, see the Global Low Skew Clock and Global Early Clock Set-up and Hold
tables on page 96.
.

R
XC4000EX IOB Input Switching Characteristic Guidelines (Continued)

otherwise noted.
Units
Setup Times
Pad to Clock (IK), no delay TPICK All devices 2.5 2.0 2.0 ns
Pad to Clock (IK), partial delay TPICKP XC4028EX 10.8 9.0 9.0 ns
XC4036EX 12.0 10.0 10.0 ns
Pad to Clock (IK), full delay TPICKD XC4028EX 15.7 13.1 13.1 ns
XC4036EX 16.9 14.1 14.1 ns
Pad to Clock (IK), via transparent Fast TPICKF All devices 3.9 3.3 3.3 ns
Capture Latch, no delay
Pad to Clock (IK), via transparent Fast TPICKFP XC4028EX 12.3 10.2 10.2 ns
Capture Latch, partial delay XC4036EX 13.5 11.2 11.2 ns
Pad to Fast Capture Latch Enable (OK), TPOCK All devices 0.8 0.7 0.7 ns
no delay
Pad to Fast Capture Latch Enable (OK), TPOCKP XC4028EX 9.1 7.6 7.6 ns
Setup Times (TTL or CMOS Inputs)
Clock Enable (EC) to Clock (IK) TECIK All devices 0.3 0.2 0.2 ns
Hold Times
Pad to Clock (IK),
no delay TIKPI All devices 0 0 0 ns
partial delay TIKPIP All devices 0 0 0 ns
full delay TIKPID All devices 0 0 0 ns
Pad to Clock (IK) via transparent Fast
Capture Latch,
no delay TIKFPI All devices 0 0 0 ns
partial delay TIKFPIP All devices 0 0 0 ns
full delay TIKFPID All devices 0 0 0 ns
Clock Enable (EC) to Clock (IK),
no delay TIKEC All devices 0 0 0 ns
partial delay TIKECP All devices 0 0 0 ns
full delay TIKECD All devices 0 0 0 ns
Pad to Fast Capture Latch Enable (OK),
no delay TOKPI All devices 0 0 0 ns
partial delay TOKPIP All devices 0 0 0 ns
tables on page 96.

R
XC4000EX IOB Output Switching Characteristic Guidelines

otherwise noted. Values apply to all XC4000EX devices unless otherwise noted.
Units
Propagation Delays
Clock (OK) to Pad TOKPOF 7.4 6.2 6.0 ns
Output (O) to Pad TOPF 6.2 5.2 5.0 ns
3-state to Pad hi-Z (slew-rate independent) TTSHZ 4.9 4.1 4.1 ns
3-state to Pad active and valid TTSONF 6.2 5.2 5.0 ns
Output MUX Select (OK) to Pad TOKFPF 6.7 5.6 5.4 ns
Fast Path Output MUX Input (EC) to Pad TCEFPF 6.2 5.1 5.0 ns
Slowest Path Output MUX Input (O) to Pad TOFPF 7.3 6.0 5.9 ns
Output (O) to clock (OK) setup time TOOK 0.6 0.5 0.5 ns
Output (O) to clock (OK) hold time TOKO 0 0 0 ns 6
Clock Enable (EC) to clock (OK) setup TECOK 0 0 0 ns
Clock Enable (EC) to clock (OK) hold TOKEC 0 0 0 ns
Clock
Clock High TCH 3.5 3.0 3.0 ns
Clock Low TCL 3.5 3.0 3.0 ns
Global Set/Reset
Minimum GSR pulse width TMRW 13.0 11.5 11.5 ns
Delay from GSR input to any Pad (XC4028EX) TRPO 30.2 25.2 25.0 ns
Delay from GSR input to any Pad (XC4036EX) TRPO 31.4 27.2 27.0 ns
Note 1: Output timing is measured at TTL threshold, with 35pF external capacitive loads.
Note 2: For CMOS output levels, see the “XC4000EX Output Level and Slew Rate Adjustments” on page 95.

R
XC4000E Switching Characteristics

Definition of Terms
follows:
families. Use as estimates, not for production.
Unmarked: Specifications not identified as either Advance or Preliminary are to be considered Final.1
XC4000E Absolute Maximum Ratings

TJ Junction temperature Ceramic packages +150 °C
XC4000E Recommended Operating Conditions

VCC Supply voltage relative to GND, TJ = -0 °C to +85°C Commercial 4.75 5.25 V
Supply voltage relative to GND, TC = -55°C to +125°C Military 4.5 5.5 V
Note 1: At junction temperatures above those listed as Recommended Operating Conditions, all delay parameters increase by 0.35% per °C.
Note 2: Input and output Measurement thresholds are: 1.5V for TTL and 2.5V for CMOS.
1. Notwithstanding the definition of the above terms, all specifications are subject to change without notice.
6-100 January 29, 1999 (Version 1.5)

R
XC4000E DC Characteristics Over Operating Conditions

High-level output voltage @ IOH = -1.0mA, VCC min CMOS outputs VCC-0.5 V
VOL Low-level output voltage @ IOL = 12.0mA, VCC min TTL outputs 0.4 V
(Note 1) CMOS outputs 0.4 V
ICCO Quiescent FPGA supply current (Note 2) Commercial 3.0 mA
Industrial 6.0 mA
Military 6.0 mA
CIN Input capacitance (sample tested) PQFP and MQFP 10 pF
packages
Other packages 16 pF
IRIN* Pad pull-up (when selected) @ VIN = 0V (sample tested) -0.02 -0.25 mA
IRLL* Horizontal Longline pull-up (when selected) @ logic Low 0.2 2.5 mA
Note 2: With no output current loads, no active input or Longline pull-up resistors, all package pins at Vcc or GND, and the FPGA configured
with a Development system Tie option.
Note 3: *Characterized Only.
XC4000E Global Buffer Switching Characteristic Guidelines

6
temperature)
Speed Grade -4 -3 -2 -1
Description Symbol Device Max Max Max Max Units
From pad through TPG XC4003E 7.0 4.7 4.0 3.5 ns
Primary buffer, XC4005E 7.0 4.7 4.3 3.8 ns
to any clock K XC4006E 7.5 5.3 5.2 4.6 ns
XC4008E 8.0 6.1 5.2 4.6 ns
XC4010E 11.0 6.3 5.4 4.8 ns
XC4013E 11.5 6.8 5.8 5.2 ns
XC4020E 12.0 7.0 6.4 6.0 ns
XC4025E 12.5 7.2 6.9 – ns
From pad through TSG XC4003E 7.5 5.2 4.4 4.0 ns
Secondary buffer, XC4005E 7.5 5.2 4.7 4.3 ns
to any clock K XC4006E 8.0 5.8 5.6 5.1 ns
XC4008E 8.5 6.6 5.6 5.1 ns
XC4010E 11.5 6.8 5.8 5.3 ns
XC4013E 12.0 7.3 6.2 5.7 ns
XC4020E 12.5 7.5 6.7 6.5 ns
XC4025E 13.0 7.7 7.2 – ns
January 29, 1999 (Version 1.5) 6-101

R
XC4000E Horizontal Longline Switching Characteristic Guidelines

worst-case operating conditions (supply voltage and junction temperature). Values apply to all XC4000E devices unless
otherwise noted.
TBUF driving a Horizontal Longline (LL):
I going High or Low to LL going High or TIO1 XC4003E 5.0 4.2 3.4 2.9 ns
Low, while T is Low. XC4005E 5.0 5.0 4.0 3.4 ns
XC4006E 6.0 5.9 4.7 4.0 ns
Buffer is constantly active. XC4008E 7.0 6.3 5.0 4.3 ns
(Note1) XC4010E 8.0 6.4 5.1 4.4 ns
XC4013E 9.0 7.2 5.7 4.9 ns
XC4020E 10.0 8.2 7.3 5.6 ns
XC4025E 11.0 9.1 7.3 – ns
I going Low to LL going from resistive TIO2 XC4003E 5.0 4.2 3.6 3.1 ns
pull-up High to active Low. XC4005E 6.0 5.3 4.5 3.8 ns
XC4006E 7.8 6.4 5.4 4.6 ns
TBUF configured as open-drain. XC4008E 8.1 6.8 5.8 4.9 ns
XC4010E 10.5 6.9 5.9 5.0 ns
(Note1) XC4013E 11.0 7.7 6.5 5.5 ns
XC4020E 12.0 8.7 8.7 7.4 ns
XC4025E 12.0 9.6 9.6 – ns
T going Low to LL going from resistive TON XC4003E 5.5 4.6 3.9 3.5 ns
pull-up or floating High to active Low. XC4005E 7.0 6.0 5.7 4.7 ns
XC4006E 7.5 6.7 5.7 4.9 ns
TBUF configured as open-drain or active XC4008E 8.0 7.1 6.0 5.2 ns
buffer with I = Low. XC4010E 8.5 7.3 6.2 5.4 ns
XC4013E 8.7 7.5 7.0 6.2 ns
(Note1) XC4020E 11.0 8.4 7.1 6.3 ns
XC4025E 11.0 8.4 7.1 – ns
T going High to TBUF going inactive, TOFF All devices 1.8 1.5 1.3 1.1 ns
not driving LL
T going High to LL going from Low to TPUS XC4003E 20.0 14.0 14.0 12.0 ns
High, pulled up by a single resistor. XC4005E 23.0 16.0 16.0 14.0 ns
XC4006E 25.0 18.0 18.0 16.0 ns
XC4008E 27.0 20.0 20.0 16.0 ns
(Note 1) XC4010E 29.0 22.0 22.0 18.0 ns
XC4013E 32.0 26.0 26.0 21.0 ns
XC4020E 35.0 32.5 32.5 26.0 ns
XC4025E 42.0 39.1 39.1 – ns
T going High to LL going from Low to TPUF XC4003E 9.0 7.0 6.0 5.4 ns
High, pulled up by two resistors. XC4005E 10.0 8.0 6.8 5.8 ns
XC4006E 11.5 9.0 7.7 6.5 ns
XC4008E 12.5 10.0 8.5 7.5 ns
(Note1) XC4010E 13.5 11.0 9.4 8.0 ns
XC4013E 15.0 13.0 11.7 9.4 ns
XC4020E 16.0 14.8 14.8 10.5 ns
XC4025E 18.0 16.5 16.5 – ns
Note 1: These values include a minimum load. Use the static timing analyzer to determine the delay for each destination.
6-102 January 29, 1999 (Version 1.5)

R
XC4000E Wide Decoder Switching Characteristic Guidelines

otherwise noted.
Full length, both pull-ups, TWAF XC4003E 9.2 5.0 5.0 4.3 ns
inputs from IOB I-pins XC4005E 9.5 6.0 6.0 5.1 ns
XC4006E 12.0 7.0 7.0 6.0 ns
XC4008E 12.5 8.0 8.0 6.5 ns
XC4010E 15.0 9.0 9.0 7.5 ns
XC4013E 16.0 11.0 11.0 8.6 ns
XC4020E 17.0 13.9 13.9 10.1 ns
XC4025E 18.0 16.9 16.9 – ns
Full length, both pull-ups, TWAFL XC4003E 12.0 7.0 7.0 5.5 ns
inputs from internal logic XC4005E 12.5 8.0 8.0 6.4 ns
XC4006E 14.0 9.0 9.0 7.0 ns 6
XC4008E 16.0 10.0 10.0 7.5 ns
XC4010E 18.0 11.0 11.0 8.5 ns
XC4013E 19.0 13.0 13.0 10.0 ns
XC4020E 20.0 15.5 15.5 11.8 ns
XC4025E 21.0 18.9 18.9 – ns
Half length, one pull-up, TWAO XC4003E 10.5 6.0 6.0 5.1 ns
inputs from IOB I-pins XC4005E 10.5 7.0 7.0 6.0 ns
XC4006E 13.5 8.0 8.0 6.5 ns
XC4008E 14.0 9.0 9.0 7.0 ns
XC4010E 16.0 10.0 10.0 7.5 ns
XC4013E 17.0 12.0 12.0 10.0 ns
XC4020E 18.0 15.0 15.0 11.8 ns
XC4025E 19.0 17.6 17.6 – ns
Half length, one pull-up, TWAOL XC4003E 12.0 8.0 8.0 6.0 ns
inputs from internal logic XC4005E 12.5 9.0 9.0 7.0 ns
XC4006E 14.0 10.0 10.0 7.6 ns
XC4008E 16.0 11.0 11.0 8.4 ns
XC4010E 18.0 12.0 12.0 9.2 ns
XC4013E 19.0 14.0 14.0 10.8 ns
XC4020E 20.0 16.8 16.8 12.6 ns
XC4025E 21.0 19.6 19.6 – ns
Note 1: These delays are specified from the decoder input to the decoder output.
Note 2: Fewer than the specified number of pullup resistors can be used, if desired. Using fewer pullups reduces power consumption but
increases delays. Use the static timing analyzer to determine delays if fewer pullups are used.
January 29, 1999 (Version 1.5) 6-103

R
XC4000E CLB Switching Characteristic Guidelines

otherwise noted.
Units
Description Symbol Min Max Min Max Min Max Min Max
F/G inputs to X/Y outputs TILO 2.7 2.0 1.6 1.3 ns
F/G inputs via H to X/Y outputs TIHO 4.7 4.3 2.7 2.2 ns
C inputs via SR through H to X/Y outputs THH0O 4.1 3.3 2.4 1.9 ns
C inputs via H to X/Y outputs THH1O 3.7 3.6 2.2 1.6 ns
C inputs via DIN through H to X/Y outputs THH2O 4.5 3.6 2.6 1.9 ns
Operand inputs (F1, F2, G1, G4) to COUT TOPCY 3.2 2.6 2.1 1.7 ns
Add/Subtract input (F3) to COUT TASCY 5.5 4.4 3.7 2.5 ns
Initialization inputs (F1, F3) to COUT TINCY 1.7 1.7 1.4 1.2 ns
CIN through function generators to TSUM 3.8 3.3 2.6 1.8 ns
X/Y outputs
CIN to COUT, bypass function generators TBYP 1.0 0.7 0.6 0.5 ns
Sequential Delays
Clock K to outputs Q TCKO 3.7 2.8 2.8 1.9 ns
F/G inputs TICK 4.0 3.0 2.4 1.8 ns
F/G inputs via H TIHCK 6.1 4.6 3.9 2.8 ns
C inputs via H0 through H THH0CK 4.5 3.6 3.5 2.4 ns
C inputs via DIN TDICK 3.0 2.4 2.0 1.0 ns
C inputs via EC TECCK 4.0 3.0 2.6 2.0 ns
C inputs via S/R, going Low (inactive) TRCK 4.2 4.0 4.0 1.5 ns
CIN input via F/G TCCK 2.5 2.1 ns
CIN input via F/G and H TCHCK 4.2 3.5 ns
6-104 January 29, 1999 (Version 1.5)

R
XC4000E CLB Switching Characteristic Guidelines (continued)

otherwise noted.
Units
F/G inputs TCKI 0 0 0 0 ns
F/G inputs via H TCKIH 0 0 0 0 ns
C inputs via H0 through H TCKHH0 0 0 0 0 ns
C inputs via DIN TCKDI 0 0 0 0 ns
C inputs via EC TCKEC 0 0 0 0 ns
C inputs via SR, going Low (inactive) TCKR 0 0 0 0 ns
Clock
Clock High time TCH 4.5 4.0 4.0 3.0 ns
Clock Low time TCL 4.5 4.0 4.0 3.0 ns 6
Set/Reset Direct
Width (High) TRPW 5.5 4.0 4.0 3.0 ns
Delay from C inputs via S/R, TRIO 6.5 4.0 4.0 3.0 ns
going High to Q
Master Set/Reset (Note 1)
Width (High or Low) TMRW 13.0 11.5 11.5 10.0 ns
Delay from Global Set/Reset net to Q TMRQ 23.0 18.7 17.4 15.0 ns
Global Set/Reset inactive to first TMRK
active clock K edge
Toggle Frequency (Note 2) FTOG 111 125 125 166 MHz
Note 1: Timing is based on the XC4005E. For other devices see the static timing analyzer.
Note 2: Export Control Max. flip-flop toggle rate.
January 29, 1999 (Version 1.5) 6-105

R
XC4000E CLB Edge-Triggered (Synchronous) RAM Switching Characteristic Guidelines

temperature). Values apply to all XC4000E devices unless otherwise noted.
Size Symbol Min Max Min Max Min Max Min Max
Write Operation
Address write cycle time (clock K period) 16x2 TWCS 15.0 14.4 11.6 8.0 ns
32x1 TWCTS 15.0 14.4 11.6 8.0 ns
Clock K pulse width (active edge) 16x2 TWPS 7.5 1 ms 7.2 1 ms 5.8 1 ms 4.0 ns
32x1 TWPTS 7.5 1 ms 7.2 1 ms 5.8 1 ms 4.0 ns
Address setup time before clock K 16x2 TASS 2.8 2.4 2.0 1.5 ns
32x1 TASTS 2.8 2.4 2.0 1.5 ns
Address hold time after clock K 16x2 TAHS 0 0 0 0 ns
32x1 TAHTS 0 0 0 0 ns
DIN setup time before clock K 16x2 TDSS 3.5 3.2 2.7 1.5 ns
32x1 TDSTS 2.5 1.9 1.7 1.5 ns
DIN hold time after clock K 16x2 TDHS 0 0 0 0 ns
32x1 TDHTS 0 0 0 0 ns
WE setup time before clock K 16x2 TWSS 2.2 2.0 1.6 1.5 ns
32x1 TWSTS 2.2 2.0 1.6 1.5 ns
WE hold time after clock K 16x2 TWHS 0 0 0 0 ns
32x1 TWHTS 0 0 0 0 ns
Data valid after clock K 16x2 TWOS 10.3 8.8 7.9 6.5 ns
32x1 TWOTS 11.6 10.3 9.3 7.0 ns
Dual-Port RAM Units
Size Symbol Min Max Min Max Min Max Min Max
Write Operation
Address write cycle time (clock K period) 16x1 TWCDS 15.0 14.4 11.6 8.0 ns
Clock K pulse width (active edge) 16x1 TWPDS 7.5 1 ms 7.2 1 ms 5.8 1 ms 4.0 ns
Address setup time before clock K 16x1 TASDS 7.5 2.5 2.1 1.5 ns
Address hold time after clock K 16x1 TAHDS 2.8 0 0 0 ns
DIN setup time before clock K 16x1 TDSDS 0 2.5 1.6 1.5 ns
DIN hold time after clock K 16x1 TDHDS 2.2 0 0 0 ns
WE setup time before clock K 16x1 TWSDS 0 1.8 1.6 1.5 ns
WE hold time after clock K 16x1 TWHDS 2.2 0 0 0 ns
Data valid after clock K 16x1 TWODS 0.3 10.0 7.8 7.0 6.5 ns
6-106 January 29, 1999 (Version 1.5)

R
XC4000E CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS
WCLK (K)
TWSS TWHS
WE
TDSS TDHS
DATA IN
TASS TAHS
ADDRESS
TILO TILO
TWOS
DATA OUT OLD NEW
X6461
6
XC4000E CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing
TWPDS
WCLK (K)
TWSDS TWHDS
WE
TDSDS TDHDS
DATA IN
TASDS TAHDS
ADDRESS
TILO TILO
TWODS
DATA OUT OLD NEW
X6474
January 29, 1999 (Version 1.5) 6-107

R
XC4000E CLB Level-Sensitive RAM Switching Characteristic Guidelines

XC4000E devices unless otherwise noted.
Units
Description Size Symbol Min Max Min Max Min Max Min Max
Write Operation
Address write cycle time 16x2 TWC 8.0 8.0 8.0 8.0 ns
32x1 TWCT 8.0 8.0 8.0 8.0 ns
Write Enable pulse width (High) 16x2 TWP 4.0 4.0 4.0 4.0 ns
32x1 TWPT 4.0 4.0 4.0 4.0 ns
Address setup time before WE 16x2 TAS 2.0 2.0 2.0 2.0 ns
32x1 TAST 2.0 2.0 2.0 2.0 ns
Address hold time after end of WE 16x2 TAH 2.5 2.0 2.0 2.0 ns
32x1 TAHT 2.0 2.0 2.0 2.0 ns
DIN setup time before end of WE 16x2 TDS 4.0 2.2 0.8 0.8 ns
32x1 TDST 5.0 2.2 0.8 0.8 ns
DIN hold time after end of WE 16x2 TDH 2.0 2.0 2.0 2.0 ns
32x1 TDHT 2.0 2.0 2.0 2.0 ns
Read Operation
Address read cycle time 16x2 TRC 4.5 3.1 2.6 2.6 ns
32x1 TRCT 6.5 5.5 3.8 3.8 ns
Data valid after address change 16x2 TILO 2.7 1.8 1.6 1.6 ns
(no Write Enable) 32x1 TIHO 4.7 3.2 2.7 2.7 ns
Address setup time before clock K 16x2 TICK 4.0 3.0 2.4 2.4 ns
32x1 TIHCK 6.1 4.6 3.9 3.9 ns
Read During Write
Data valid after WE goes active (DIN 16x2 TWO 10.0 6.0 4.9 4.9 ns
stable before WE) 32x1 TWOT 12.0 7.3 5.6 5.6 ns
Data valid after DIN 16x2 TDO 9.0 6.6 5.8 5.8 ns
(DIN changes during WE) 32x1 TDOT 11.0 7.6 6.2 6.2 ns
WE setup time before clock K 16x2 TWCK 8.0 6.0 5.1 5.1 ns
32x1 TWCKT 9.6 6.8 5.8 5.8 ns
Data setup time before clock K 16x2 TDCK 7.0 5.2 4.4 4.4 ns
32x1 TDCKT 8.0 6.2 5.3 5.3 ns
Preliminary
6-108 January 29, 1999 (Version 1.5)

R
XC4000E CLB Level-Sensitive RAM Timing Characteristics

T WC
ADDRESS
WRITE
TAS T WP T AH
WRITE ENABLE
T DS T DH
DATA IN REQUIRED
READ WITHOUT WRITE

T ILO

T ICK T CH
CLOCK
T CKO 6
VALID VALID
XQ, YQ OUTPUTS
(OLD) (NEW)
READ DURING WRITE

T WP
WRITE ENABLE
T DH
DATA IN
(stable during WE)
T WO
DATA IN
T WO T DO
VALID VALID VALID

X, Y OUTPUTS

T WP
WRITE ENABLE
T WCK
T DCK
DATA IN
CLOCK
T CKO
XQ, YQ OUTPUTS
X2640
January 29, 1999 (Version 1.5) 6-109

R
XC4000E Guaranteed Input and Output Parameters (Pin-to-Pin, TTL I/O)

the static timing analyzer report. Values apply to all XC4000E devices unless otherwise noted.
Units
Description Symbol Device
Global Clock to Output TICKOF XC4003E 12.5 10.2 8.7 5.8 ns
(fast) using OFF XC4005E 14.0 10.7 9.1 6.2 ns
XC4006E 14.5 10.7 9.1 6.4 ns
XC4008E 15.0 10.8 9.2 6.6 ns
TPG OFF
.
.
(Max) XC4010E 16.0 10.9 9.3 6.8 ns
. XC4013E 16.5 11.0 9.4 7.2 ns
.
Global Clock-to-Output Delay
. XC4020E 17.0 11.0 10.2 7.4 ns
X3202 XC4025E 17.0 12.6 10.8 – ns
Global Clock to Output TICKO XC4003E 16.5 14.0 11.5 7.8 ns
(slew-limited) using OFF XC4005E 18.0 14.7 12.0 8.2 ns
XC4006E 18.5 14.7 12.0 8.4 ns
XC4008E 19.0 14.8 12.1 8.6 ns
. (Max) XC4010E 20.0 14.9 12.2 8.8 ns
TPG OFF .
. XC4013E 20.5 15.0 12.8 9.2 ns
.
. XC4020E 21.0 15.1 12.8 9.4 ns
X3202 XC4025E 21.0 15.3 13.0 – ns
Input Setup Time, using IFF TPSUF XC4003E 2.5 2.3 2.3 1.5 ns
(no delay) XC4005E 2.0 1.2 1.2 0.8 ns
XC4006E 1.9 1.0 1.0 0.6 ns
D XC4008E 1.4 0.6 0.6 0.2 ns
Input (Min) XC4010E 1.0 0.2 0.2 0 ns
Set - Up IFF
& TPG XC4013E 0.5 0 0 0 ns
Hold
Time XC4020E 0 0 0 0 ns
X3201
XC4025E 0 0 0 – ns
Input Hold Time, using IFF TPHF XC4003E 4.0 4.0 4.0 1.5 ns
(no delay) XC4005E 4.6 4.5 4.5 2.0 ns
XC4006E 5.0 4.7 4.7 2.0 ns
D XC4008E 6.0 5.1 5.1 2.5 ns
Input (Min) XC4010E 6.0 5.5 5.5 2.5 ns
Set - Up IFF
& TPG XC4013E 7.0 6.5 5.5 3.0 ns
Hold
Time XC4020E 7.5 6.7 5.7 3.5 ns
X3201
XC4025E 8.0 7.0 5.9 – ns
Input Setup Time, using IFF TPSU XC4003E 8.5 7.0 6.0 5.0 ns
(with delay) XC4005E 8.5 7.0 6.0 5.0 ns
XC4006E 8.5 7.0 6.0 5.0 ns
D XC4008E 8.5 7.0 6.0 5.0 ns
Input (Min) XC4010E 8.5 7.0 6.0 5.0 ns
Set - Up IFF
& TPG XC4013E 8.5 7.0 6.0 5.0 ns
Hold
Time XC4020E 9.5 7.0 6.8 5.0 ns
X3201
XC4025E 9.5 7.6 6.8 – ns
Input Hold Time, using IFF TPH XC4003E 0 0 0 0 ns

(with delay) XC4005E 0 0 0 0 ns
XC4006E 0 0 0 0 ns
D XC4008E 0 0 0 0 ns
Input (Min) XC4010E 0 0 0 0 ns
Set - Up IFF
& TPG XC4013E 0 0 0 0 ns
Hold
Time XC4020E 0 0 0 0 ns
X3201
XC4025E 0 0 0 – ns
OFF = Output Flip-Flop IFF = Input Flip-Flop or Latch Preliminary
6-110 January 29, 1999 (Version 1.5)

R
XC4000E IOB Input Switching Characteristic Guidelines (continued)

otherwise noted.
Units
Description Symbol Device Min Max Min Max Min Max Min Max
Setup Times (TTL Inputs)
Pad to Clock (IK), no delay TPICK All devices 4.0 2.6 2.0 1.5 ns
with delay TPICKD XC4003E 10.9 8.2 6.0 4.8 ns
XC4005E 10.9 8.7 6.1 5.1 ns
XC4006E 10.9 9.2 6.2 5.8 ns
XC4008E 11.1 9.6 6.3 5.8 ns
XC4010E 11.3 9.8 6.4 6.0 ns
XC4013E 11.8 10.2 7.9 7.6 ns
XC4020E 14.0 11.4 9.4 8.2 ns
XC4025E 14.0 11.4 10.0 – ns
Setup Time (CMOS Inputs)
Pad to Clock (IK), no delay TPICKC All devices 6.0 3.3 2.4 2.4 ns 6
with delay TPICKDC XC4003E 12.0 8.8 6.9 5.3 ns
XC4005E 12.0 9.7 8.0 5.6 ns
XC4006E 12.3 9.9 8.1 6.3 ns
XC4008E 12.8 10.3 8.2 6.3 ns
XC4010E 13.0 10.5 8.3 6.5 ns
XC4013E 13.5 10.9 10.0 7.9 ns
XC4020E 16.0 12.1 12.1 8.1 ns
XC4025E 16.0 12.1 12.1 – ns
(TTL or CMOS)
Clock Enable (EC) to Clock
(IK), no delay TECIK All devices 3.5 2.5 2.1 1.5 ns
with delay TECIKD XC4003E 10.4 8.1 4.3 4.3 ns
XC4005E 10.4 8.5 5.6 5.0 ns
XC4006E 10.4 9.1 6.7 6.0 ns
XC4008E 10.4 9.5 6.9 6.0 ns
XC4010E 10.7 9.7 7.1 6.5 ns
XC4013E 11.1 10.1 9.0 8.0 ns
XC4020E 14.0 11.3 10.6 9.0 ns
XC4025E 14.0 11.3 11.0 – ns
Global Set/Reset (Note 3)
Delay from GSR net TRRI 12.0 7.8 6.8 6.8 ns
through Q to I1, I2
GSR width TMRW 13.0 11.5 11.5 10.0 ns
GSR inactive to first active TMRI
Clock (IK) edge
Preliminary
Note 1: Input pad setup and hold times are specified with respect to the internal clock (IK). For setup and hold times with respect to
the clock input pin, see the pin-to-pin parameters in the Guaranteed Input and Output Parameters table.
Note 3: Timing is based on the XC4005E. For other devices see the XACT timing calculator.
January 29, 1999 (Version 1.5) 6-111

R
XC4000E IOB Output Switching Characteristic Guidelines

otherwise noted.
Units
Propagation Delays
(TTL Output Levels)
Clock (OK) to Pad, fast TOKPOF 7.5 6.5 4.5 3.0 ns
slew-rate limited TOKPOS 11.5 9.5 7.0 5.0 ns
Output (O) to Pad, fast TOPF 8.0 5.5 4.8 3.2 ns
slew-rate limited TOPS 12.0 8.5 7.3 5.2 ns
3-state to Pad hi-Z TTSHZ 5.0 4.2 3.8 3.0 ns
(slew-rate independent)
3-state to Pad active
and valid, fast TTSONF 9.7 8.1 7.3 6.8 ns
slew-rate limited TTSONS 13.7 11.1 9.8 8.8 ns
Propagation Delays
(CMOS Output Levels)
Clock (OK) to Pad, fast TOKPOFC 9.5 7.8 7.0 4.0 ns
slew-rate limited TOKPOSC 13.5 11.6 10.4 7.0 ns
Output (O) to Pad, fast TOPFC 10.0 9.7 8.7 4.0 ns
slew-rate limited TOPSC 14.0 13.4 12.1 6.0 ns
3-state to Pad hi-Z TTSHZC 5.2 4.3 3.9 3.9 ns
and valid, fast TTSONFC 9.1 7.6 6.8 6.8 ns
slew-rate limited TTSONSC 13.1 11.4 10.2 8.8 ns
Preliminary
Note 1: Output timing is measured at pin threshold, with 50pF external capacitive loads (incl. test fixture). Slew-rate limited output
rise/fall times are approximately two times longer than fast output rise/fall times. For the effect of capacitive loads on ground
bounce, see the “Additional XC4000 Data” section of the Programmable Logic Data Book.
6-112 January 29, 1999 (Version 1.5)

R
XC4000E IOB Output Switching Characteristic Guidelines (continued)

otherwise noted. Values apply to all XC4000E devices unless otherwise noted.
Units
Setup and Hold
Output (O) to clock (OK) TOOK 5.0 4.6 3.8 2.3 ns
setup time
Output (O) to clock (OK) TOKO 0 0 0 0 ns
hold time
Clock Enable (EC) to TECOK 4.8 3.5 2.7 2.0 ns
clock (OK) setup
Clock Enable (EC) to TOKEC 1.2 1.2 0.5 0 ns
clock (OK) hold
Clock
Clock High TCH 4.5 4.0 4.0 3.0 ns
Clock Low TCL 4.5 4.0 4.0 3.0 ns 6
Delay from GSR net to Pad TRPO 15.0 11.8 8.7 7.0 ns
GSR width TMRW 13.0 11.5 11.5 ns
GSR inactive to first active TMRO
clock (OK) edge
Preliminary
Note 3: Timing is based on the XC4005E. For other devices see the XACT timing calculator.
January 29, 1999 (Version 1.5) 6-113

R
XC4000E Boundary Scan (JTAG) Switching Characteristic Guidelines

that are taken at device introduction, prior to any process improvements. For more detailed, more precise, and more
up-to-date information, use the values provided by the XACT timing calculator and used in the simulator. These values can
be printed in tabular format by running LCA2XNF -S.
The following guidelines reflect worst-case values over the recommended operating conditions. They are expressed in units
of nanoseconds and apply to all XC4000E devices unless otherwise noted.
Units
Setup and Hold
Input (TDI) to clock (TCK) TTDITCK 30.0 30.0 30.0 20.0 ns
setup time
Input (TDI) to clock (TCK) TTCKTDI 0 0 0 0 ns
hold time
Input (TMS) to clock (TCK) TTMSTCK 15.0 15.0 15.0 10.0 ns
setup time
Input (TMS) to clock (TCK) TTCKTMS 0 0 0 0 ns
hold time
Propagation Delay
Clock (TCK) to Pad (TDO) TTCKPO 30.0 30.0 30.0 20.0 ns
Clock
Clock (TCK) High TTCKH 5.0 5.0 5.0 4.0 ns
Clock (TCK) Low TTCKL 5.0 5.0 5.0 4.0 ns
FMAX (MHz)
FMAX 15.0 15.0 15.0 25.0 ns
Preliminary
6-114 January 29, 1999 (Version 1.5)


Device-specific tables include all packages for each XC4000 and XC4000X Series device. They follow the pad locations
around the die, and include boundary scan register locations..
XC4002XL Device Pinout Tables XC4002XL

PC84 PQ100 VQ100 PG120
Boundary
Pad Names Scan
XC4002XL Boundary I/O (D6) P58 P58 P55 M10 157

PC84 PQ100 VQ100 PG120 I/O - P59 P56 N11 160
Pad Names Scan
VCC P2 P92 P89 G3 I/O (D5) P59 P60 P57 M9 163
I/O (A8) P3 P93 P90 G1 26 I/O (CS0) P60 P61 P58 N10 166
I/O (A9) P4 P94 P91 F1 29 I/O (D4) P61 P64 P61 M8 169
I/O (A10) P5 P97 P94 F3 32 I/O P62 P65 P62 N8 172
I/O (A11) P6 P98 P95 D1 35 VCC P63 P66 P63 M7
I/O (A12) P7 P99 P96 C1 38 GND P64 P67 P64 L7
I/O (A13) P8 P100 P97 D2 41 I/O (D3) P65 P68 P65 N7 175
I/O (A14) P9 P1 P98 C2 44 I/O (RS) P66 P69 P66 N6 178
I/O,GCK8 (A15) P10 P2 P99 D3 47 I/O (D2) P67 P71 P68 L6 181
VCC P11 P3 P100 C3 I/O P68 P72 P69 N4 184
GND P12 P4 P1 C4 I/O (D1) P69 P73 P70 M5 187
I/O, GCK1(A16) P13 P5 P2 B2 50 I/O (RCLK, P70 P74 P71 N3 190
I/O (A17) P14 P6 P3 B3 53 RDY/BUSY)
I/O, TDI P15 P7 P4 C5 56 I/O (D0, DIN) P71 P75 P72 N2 193
I/O, TCK P16 P8 P5 B4 59 I/O, GCK6 (DOUT) P72 P76 P73 M3 196
I/O, TMS P17 P9 P6 B5 62 CCLK P73 P77 P74 L4
I/O P18 P10 P7 A4 65 VCC P74 P78 P75 L3
I/O P19 P12 P9 B6 68 O, TDO P75 P79 P76 M2
I/O P20 P13 P10 A6 71 GND P76 P80 P77 K3
GND P21 P14 P11 B7 I/O (A0, WS) P77 P81 P78 L2 2
VCC P22 P15 P12 C7 I/O, GCK7(A1) P78 P82 P79 N1 5 6
I/O P23 P16 P13 A7 74 I/O (CS1, A2) P79 P83 P80 K2 8
I/O P24 P17 P14 A8 77 I/O (A3) P80 P84 P81 L1 11
I/O P25 P19 P16 C8 80 I/O (A4) P81 P85 P82 J2 14
I/O P26 P20 P17 A10 83 I/O (A5) P82 P86 P83 K1 17
I/O P27 P21 P18 B9 86 I/O (A6) P83 P89 P86 H2 20
I/O - P22 P19 A11 89 I/O (A7) P84 P90 P87 H1 23
I/O P28 P23 P20 C9 92 GND P1 P91 P88 G2
I/O, GCK2 P29 P24 P21 A12 95 1/22/99
O (M1) P30 P25 P22 B11 98
GND P31 P26 P23 C10
I (M0) P32 P27 P24 C11 101
VCC P33 P28 P25 D11
Additional XC4002XL Package Pins
I (M2) P34 P29 P26 B12 102 PG120
I/O, GCK3 P35 P30 P27 C12 103 N.C. Pins
I/O (HDC) P36 P31 P28 A13 106 E1 F2 E2 E3 B1 A1
I/O - P32 P29 D12 109 A2 A3 C6 A5 A9 B8
I/O (LDC) P37 P33 P30 C13 112 B10 B13 E11 E13 J13 H12
I/O P38 P34 P31 E12 115 K12 J11 N12 L9 L8 N9
I/O P39 P35 P32 D13 118 N5 M6 M4 L5 M1 J3
I/O P40 P38 P35 F12 121 H3 J1
I/O (INIT) P41 P39 P36 F13 124 1/22/99
VCC P42 P40 P37 G12
GND P43 P41 P38 G11 •
I/O P44 P42 P39 G13 127 PQ100
I/O P45 P43 P40 H13 130 N.C. Pins
I/O P46 P46 P43 H11 133 P11 P18 P36 P37 P44 P45
I/O P47 P47 P44 K13 136 P62 P63 P70 P87 P88 P95
I/O P48 P48 P45 J12 139 P96
I/O P49 P49 P46 L13 142 1/22//99
I/O P50 P50 P47 M13 145
I/O, GCK4 P51 P51 P48 L12 148 •
GND P52 P52 P49 K11 VQ100
DONE P53 P53 P50 L11 N.C. Pins
VCC P54 P54 P51 L10 P8 P15 P33 P34 P41 P42
PROGRAM P55 P55 P52 M12 P59 P60 P67 P84 P85 P92
I/O (D7) P56 P56 P53 M11 151 P93
I/O, GCK5 P57 P57 P54 N13 154 1/22//99
January 29, 1999 (Version 1.5) 6-115

R
Pin Locations for XC4003E Devices XC4003E

PC84 PQ100 VQ100 PG120 Bndry Scan
Pad Name
XC4003E I/O P46 P46 P43 H11 169
PC84 PQ100 VQ100 PG120 Bndry Scan I/O P47 P47 P44 K13 172
Pad Name
VCC P2 P92 P89 G3 - I/O P48 P48 P45 J12 175
I/O (A8) P3 P93 P90 G1 32 I/O P49 P49 P46 L13 178
I/O (A9) P4 P94 P91 F1 35 I/O P50 P50 P47 M13 181
I/O - P95 P92 E1 38 I/O, SGCK3 P51 P51 P48 L12 184
I/O - P96 P93 F2 41 GND P52 P52 P49 K11 -
I/O (A10) P5 P97 P94 F3 44 DONE P53 P53 P50 L11 -
I/O (A11) P6 P98 P95 D1 47 VCC P54 P54 P51 L10 -
I/O (A12) P7 P99 P96 C1 50 PROGRAM P55 P55 P52 M12 -
I/O (A13) P8 P100 P97 D2 53 I/O (D7) P56 P56 P53 M11 187
I/O (A14) P9 P1 P98 C2 56 I/O, PGCK3 P57 P57 P54 N13 190
I/O, SGCK1 (A15) P10 P2 P99 D3 59 I/O (D6) P58 P58 P55 M10 193
VCC P11 P3 P100 C3 - I/O - P59 P56 N11 196
GND P12 P4 P1 C4 - I/O (D5) P59 P60 P57 M9 199
I/O, PGCK1 (A16) P13 P5 P2 B2 62 I/O (CS0) P60 P61 P58 N10 202
I/O (A17) P14 P6 P3 B3 65 I/O - P62 P59 L8 205
I/O, TDI P15 P7 P4 C5 68 I/O - P63 P60 N9 208
I/O, TCK P16 P8 P5 B4 71 I/O (D4) P61 P64 P61 M8 211
I/O, TMS P17 P9 P6 B5 74 I/O P62 P65 P62 N8 214
I/O P18 P10 P7 A4 77 VCC P63 P66 P63 M7 -
I/O - - - C6 80 GND P64 P67 P64 L7 -
I/O - P11 P8 A5 83 I/O (D3) P65 P68 P65 N7 217
I/O P19 P12 P9 B6 86 I/O (RS) P66 P69 P66 N6 220
I/O P20 P13 P10 A6 89 I/O - P70 P67 N5 223
GND P21 P14 P11 B7 - I/O - - - M6 226
VCC P22 P15 P12 C7 - I/O (D2) P67 P71 P68 L6 229
I/O P23 P16 P13 A7 92 I/O P68 P72 P69 N4 232
I/O P24 P17 P14 A8 95 I/O (D1) P69 P73 P70 M5 235
I/O - P18 P15 A9 98 I/O (RCLK, P70 P74 P71 N3 238
I/O - - - B8 101 RDY/BUSY)
I/O P25 P19 P16 C8 104 I/O (D0, DIN) P71 P75 P72 N2 241
I/O P26 P20 P17 A10 107 I/O, SGCK4 (DOUT) P72 P76 P73 M3 244
I/O P27 P21 P18 B9 110 CCLK P73 P77 P74 L4 -
I/O - P22 P19 A11 113 VCC P74 P78 P75 L3 -
I/O P28 P23 P20 C9 116 O, TDO P75 P79 P76 M2 0
I/O, SGCK2 P29 P24 P21 A12 119 GND P76 P80 P77 K3 -
O (M1) P30 P25 P22 B11 122 I/O (A0, WS) P77 P81 P78 L2 2
GND P31 P26 P23 C10 - I/O, PGCK4 (A1) P78 P82 P79 N1 5
I (M0) P32 P27 P24 C11 125 I/O (CS1, A2) P79 P83 P80 K2 8
VCC P33 P28 P25 D11 - I/O (A3) P80 P84 P81 L1 11
I (M2) P34 P29 P26 B12 126 I/O (A4) P81 P85 P82 J2 14
I/O, PGCK2 P35 P30 P27 C12 127 I/O (A5) P82 P86 P83 K1 17
I/O (HDC) P36 P31 P28 A13 130 I/O - P87 P84 H3 20
I/O - P32 P29 D12 133 I/O - P88 P85 J1 23
I/O (LDC) P37 P33 P30 C13 136 I/O (A6) P83 P89 P86 H2 26
I/O P38 P34 P31 E12 139 I/O (A7) P84 P90 P87 H1 29
I/O P39 P35 P32 D13 142 GND P1 P91 P88 G2 -
I/O - P36 P33 F11 145 5/5/97
I/O - P37 P34 E13 148
I/O P40 P38 P35 F12 151 Additional XC4003E Package Pins
I/O (INIT) P41 P39 P36 F13 154 PG120
VCC P42 P40 P37 G12 - Not Connected Pins
GND P43 P41 P38 G11 - A1 A2 A3 B1 B10 B13
I/O P44 P42 P39 G13 157 E2 E3 E11 J3 J11 K12
I/O P45 P43 P40 H13 160 L5 L9 M1 M4 N12 -
I/O - P44 P41 J13 163 5/5/97
I/O - P45 P42 H12 166
6-116 January 29, 1999 (Version 1.5)

R
Pin Locations for XC4005E/XL Devices

The following table may contain pinout information for unsupported device/package combinations. Please see the availability
charts elsewhere in the XC4000 Series data sheet for availability information.
XC4005E/XL PC PQ VQ TQ PG PQ PQ Bndry XC4005E/XL PC PQ VQ TQ PG PQ PQ Bndry

Pad Name 84 100 100†† 144 156† 160 208 Scan Pad Name 84 100 100†† 144 156† 160 208 Scan
VCC P2 P92 P89 P128 H3 P142 P183 - GND - - - P45 F14 P51 P67 -
I/O (A8) P3 P93 P90 P129 H1 P143 P184 44 I/O - - - P46 F15 P52 P68 193
I/O (A9) P4 P94 P91 P130 G1 P144 P185 47 I/O - - - P47 E16 P53 P69 196
I/O (A19) †† - P95 P92 P131 G2 P145 P186 50 I/O P38 P34 P31 P48 F16 P54 P70 199
I/O (A18) †† - P96 P93 P132 G3 P146 P187 53 I/O P39 P35 P32 P49 G14 P55 P71 202
I/O (A10) P5 P97 P94 P133 F1 P147 P190 56 I/O - P36 P33 P50 G15 P56 P74 205
I/O (A11) P6 P98 P95 P134 F2 P148 P191 59 I/O - P37 P34 P51 G16 P57 P75 208
I/O - - - P135 E1 P149 P192 62 I/O P40 P38 P35 P52 H16 P58 P76 211
I/O - - - P136 E2 P150 P193 65 I/O (INIT) P41 P39 P36 P53 H15 P59 P77 214
GND - - - P137 F3 P151 P194 - VCC P42 P40 P37 P54 H14 P60 P78 -
I/O (A12) P7 P99 P96 P138 E3 P154 P199 68 GND P43 P41 P38 P55 J14 P61 P79 -
I/O (A13) P8 P100 P97 P139 C1 P155 P200 71 I/O P44 P42 P39 P56 J15 P62 P80 217
I/O - - - P140 C2 P156 P201 74 I/O P45 P43 P40 P57 J16 P63 P81 220
I/O - - - P141 D3 P157 P202 77 I/O - P44 P41 P58 K16 P64 P82 223
I/O (A14) P9 P1 P98 P142 B1 P158 P203 80 I/O - P45 P42 P59 K15 P65 P83 226
I/O, SGCK1 †, P10 P2 P99 P143 B2 P159 P204 83 I/O P46 P46 P43 P60 K14 P66 P86 229
GCK8 †† (A15) I/O P47 P47 P44 P61 L16 P67 P87 232
VCC P11 P3 P100 P144 C3 P160 P205 - I/O - - - P62 M16 P68 P88 235
GND P12 P4 P1 P1 C4 P1 P2 - I/O - - - P63 L15 P69 P89 238
I/O, PGCK1†, P13 P5 P2 P2 B3 P2 P4 86 GND - - - P64 L14 P70 P90 -
GCK1†† (A16) I/O P48 P48 P45 P65 P16 P73 P95 241
I/O (A17) P14 P6 P3 P3 A1 P3 P5 89 I/O P49 P49 P46 P66 M14 P74 P96 244
I/O - - - P4 A2 P4 P6 92 I/O - - - P67 N15 P75 P97 247
I/O - - - P5 C5 P5 P7 95 I/O - - - P68 P15 P76 P98 250
6
I/O, TDI P15 P7 P4 P6 B4 P6 P8 98 I/O P50 P50 P47 P69 N14 P77 P99 253
I/O, TCK P16 P8 P5 P7 A3 P7 P9 101 I/O, SGCK3 †, P51 P51 P48 P70 R16 P78 P100 256
GND - - - P8 C6 P10 P14 - GCK4 ††
I/O - - - P9 B5 P11 P15 104 GND P52 P52 P49 P71 P14 P79 P101 -
I/O - - - P10 B6 P12 P16 107 DONE P53 P53 P50 P72 R15 P80 P103 -
I/O, TMS P17 P9 P6 P11 A5 P13 P17 110 VCC P54 P54 P51 P73 P13 P81 P106 -
I/O P18 P10 P7 P12 C7 P14 P18 113 PROGRAM P55 P55 P52 P74 R14 P82 P108 -
I/O - - - P13 B7 P15 P21 116 I/O (D7) P56 P56 P53 P75 T16 P83 P109 259
I/O - P11 P8 P14 A6 P16 P22 119 I/O, PGCK3†, P57 P57 P54 P76 T15 P84 P110 262
I/O P19 P12 P9 P15 A7 P17 P23 122 GCK5††
I/O P20 P13 P10 P16 A8 P18 P24 125 I/O - - - P77 R13 P85 P111 265
GND P21 P14 P11 P17 C8 P19 P25 - I/O - - - P78 P12 P86 P112 268
VCC P22 P15 P12 P18 B8 P20 P26 - I/O (D6) P58 P58 P55 P79 T14 P87 P113 271
I/O P23 P16 P13 P19 C9 P21 P27 128 I/O - P59 P56 P80 T13 P88 P114 274
I/O P24 P17 P14 P20 B9 P22 P28 131 GND - - - P81 P11 P91 P119 -
I/O - P18 P15 P21 A9 P23 P29 134 I/O - - - P82 R11 P92 P120 277
I/O - - - P22 B10 P24 P30 137 I/O - - - P83 T11 P93 P121 280
I/O P25 P19 P16 P23 C10 P25 P33 140 I/O (D5) P59 P60 P57 P84 T10 P94 P122 283
I/O P26 P20 P17 P24 A10 P26 P34 143 I/O (CS0) P60 P61 P58 P85 P10 P95 P123 286
I/O - - - P25 A11 P27 P35 146 I/O - P62 P59 P86 R10 P96 P126 289
I/O - - - P26 B11 P28 P36 149 I/O - P63 P60 P87 T9 P97 P127 292
GND - - - P27 C11 P29 P37 - I/O (D4) P61 P64 P61 P88 R9 P98 P128 295
I/O P27 P21 P18 P28 B12 P32 P42 152 I/O P62 P65 P62 P89 P9 P99 P129 298
I/O - P22 P19 P29 A13 P33 P43 155 VCC P63 P66 P63 P90 R8 P100 P130 -
I/O - - - P30 A14 P34 P44 158 GND P64 P67 P64 P91 P8 P101 P131 -
I/O - - - P31 C12 P35 P45 161 I/O (D3) P65 P68 P65 P92 T8 P102 P132 301
I/O P28 P23 P20 P32 B13 P36 P46 164 I/O (RS) P66 P69 P66 P93 T7 P103 P133 304
I/O, SGCK2 †, P29 P24 P21 P33 B14 P37 P47 167 I/O - P70 P67 P94 T6 P104 P134 307
GCK2 †† I/O - - - P95 R7 P105 P135 310
O (M1) P30 P25 P22 P34 A15 P38 P48 170 I/O (D2) P67 P71 P68 P96 P7 P106 P138 313
GND P31 P26 P23 P35 C13 P39 P49 - I/O P68 P72 P69 P97 T5 P107 P139 316
I (M0) P32 P27 P24 P36 A16 P40 P50 173 I/O - - - P98 R6 P108 P140 319
VCC P33 P28 P25 P37 C14 P41 P55 - I/O - - - P99 T4 P109 P141 322
I (M2) P34 P29 P26 P38 B15 P42 P56 174 GND - - - P100 P6 P110 P142 -
I/O, PGCK2 †, P35 P30 P27 P39 B16 P43 P57 175 I/O (D1) P69 P73 P70 P101 T3 P113 P147 325
GCK3 †† I/O (RCLK, P70 P74 P71 P102 P5 P114 P148 328
I/O (HDC) P36 P31 P28 P40 D14 P44 P58 178 RDY/BUSY)
I/O - - - P41 C15 P45 P59 181 I/O - - - P103 R4 P115 P149 331
I/O - - - P42 D15 P46 P60 184 I/O - - - P104 R3 P116 P150 334
I/O - P32 P29 P43 E14 P47 P61 187 I/O (D0, DIN) P71 P75 P72 P105 P4 P117 P151 337
I/O (LDC) P37 P33 P30 P44 C16 P48 P62 190
January 29, 1999 (Version 1.5) 6-117

R
XC4005E/XL PC PQ VQ TQ PG PQ PQ Bndry PG156

Pad Name 84 100 100†† 144 156† 160 208 Scan Not Connected Pins
I/O, SGCK4 †, P72 P76 P73 P106 T2 P118 P152 340 A4 A12 D1 D2 D16 E15
GCK6 †† (DOUT) M1 M2 M15 N16 R5 R12
CCLK P73 P77 P74 P107 R2 P119 P153 - T12 - - - - -
VCC P74 P78 P75 P108 P3 P120 P154 - 5/5/97
O, TDO P75 P79 P76 P109 T1 P121 P159 0
GND P76 P80 P77 P110 N3 P122 P160 -
I/O (A0, WS) P77 P81 P78 P111 R1 P123 P161 2 PQ160
I/O, PGCK4 †, P78 P82 P79 P112 P2 P124 P162 5 Not Connected Pins
GCK7 †† (A1) P8 P9 P30 P31 P49 P50
I/O - - - P113 N2 P125 P163 8 P71 P72 P89 P90 P111 P112
I/O - - - P114 M3 P126 P164 11 P129 P130 P136 P152 P153 -
I/O (CS1, A2) P79 P83 P80 P115 P1 P127 P165 14 6/16/97
I/O (A3) P80 P84 P81 P116 N1 P128 P166 17
GND - - - P118 L3 P131 P171 -
I/O - - - P119 L2 P132 P172 20
I/O - - - P120 L1 P133 P173 23 PQ208
I/O (A4) P81 P85 P82 P121 K3 P134 P174 26 Not Connected Pins
I/O (A5) P82 P86 P83 P122 K2 P135 P175 29 P1 P3 P10 P11 P12 P13
I/O (A21) †† - P87 P84 P123 K1 P137 P178 32 P19 P20 P31 P32 P38 P39
I/O (A20) †† - P88 P85 P124 J1 P138 P179 35 P40 P41 P51 P52 P53 P54
I/O (A6) P83 P89 P86 P125 J2 P139 P180 38 P63 P64 P65 P66 P72 P73
I/O (A7) P84 P90 P87 P126 J3 P140 P181 41 P84 P85 P91 P92 P93 P94
GND P1 P91 P88 P127 H2 P141 P182 - P102 P104 P105 P107 P115 P116
6/10/97 P117 P118 P124 P125 P136 P137
P143 P144 P145 P146 P155 P156
† = E only P157 P158 P167 P168 P169 P170
†† = XL only P176 P177 P188 P189 P195 P196
P197 P198 P206 P207 P208 -
Additional XC4005E/XL Package Pins 6/5/97
TQ144
Not Connected Pins
P117 - - - - -
5/5/97
Pin Locations for XC4006E Devices

XC4006E PC TQ PG PQ PQ Bndry XC4006E PC TQ PG PQ PQ Bndry
Pad Name 84 144 156 160 208 Scan Pad Name 84 144 156 160 208 Scan
VCC P2 P128 H3 P142 P183 - I/O - P9 B5 P11 P15 122
I/O (A8) P3 P129 H1 P143 P184 50 I/O - P10 B6 P12 P16 125
I/O (A9) P4 P130 G1 P144 P185 53 I/O, TMS P17 P11 A5 P13 P17 128
I/O - P131 G2 P145 P186 56 I/O P18 P12 C7 P14 P18 131
I/O - P132 G3 P146 P187 59 I/O - P13 B7 P15 P21 134
I/O (A10) P5 P133 F1 P147 P190 62 I/O - P14 A6 P16 P22 137
I/O (A11) P6 P134 F2 P148 P191 65 I/O P19 P15 A7 P17 P23 140
I/O - P135 E1 P149 P192 68 I/O P20 P16 A8 P18 P24 143
I/O - P136 E2 P150 P193 71 GND P21 P17 C8 P19 P25 -
GND - P137 F3 P151 P194 - VCC P22 P18 B8 P20 P26 -
I/O - - D1 P152 P197 74 I/O P23 P19 C9 P21 P27 146
I/O - - D2 P153 P198 77 I/O P24 P20 B9 P22 P28 149
I/O (A12) P7 P138 E3 P154 P199 80 I/O - P21 A9 P23 P29 152
I/O (A13) P8 P139 C1 P155 P200 83 I/O - P22 B10 P24 P30 155
I/O - P140 C2 P156 P201 86 I/O P25 P23 C10 P25 P33 158
I/O - P141 D3 P157 P202 89 I/O P26 P24 A10 P26 P34 161
I/O (A14) P9 P142 B1 P158 P203 92 I/O - P25 A11 P27 P35 164
I/O, SGCK1 (A15) P10 P143 B2 P159 P204 95 I/O - P26 B11 P28 P36 167
VCC P11 P144 C3 P160 P205 - GND - P27 C11 P29 P37 -
GND P12 P1 C4 P1 P2 - I/O - - A12 P30 P40 170
I/O, PGCK1 (A16) P13 P2 B3 P2 P4 98 I/O - - - P31 P41 173
I/O (A17) P14 P3 A1 P3 P5 101 I/O P27 P28 B12 P32 P42 176
I/O - P4 A2 P4 P6 104 I/O - P29 A13 P33 P43 179
I/O - P5 C5 P5 P7 107 I/O - P30 A14 P34 P44 182
I/O, TDI P15 P6 B4 P6 P8 110 I/O - P31 C12 P35 P45 185
I/O, TCK P16 P7 A3 P7 P9 113 I/O P28 P32 B13 P36 P46 188
I/O - - A4 P8 P10 116 I/O, SGCK2 P29 P33 B14 P37 P47 191
I/O - - - P9 P11 119 O (M1) P30 P34 A15 P38 P48 194
GND - P8 C6 P10 P14 - GND P31 P35 C13 P39 P49 -
6-118 January 29, 1999 (Version 1.5)

R
XC4006E PC TQ PG PQ PQ Bndry XC4006E PC TQ PG PQ PQ Bndry

Pad Name 84 144 156 160 208 Scan Pad Name 84 144 156 160 208 Scan
I (M0) P32 P36 A16 P40 P50 197 I/O - P94 T6 P104 P134 349
VCC P33 P37 C14 P41 P55 - I/O - P95 R7 P105 P135 352
I (M2) P34 P38 B15 P42 P56 198 I/O (D2) P67 P96 P7 P106 P138 355
I/O, PGCK2 P35 P39 B16 P43 P57 199 I/O P68 P97 T5 P107 P139 358
I/O (HDC) P36 P40 D14 P44 P58 202 I/O - P98 R6 P108 P140 361
I/O - P41 C15 P45 P59 205 I/O - P99 T4 P109 P141 364
I/O - P42 D15 P46 P60 208 GND - P100 P6 P110 P142 -
I/O - P43 E14 P47 P61 211 I/O - - R5 P111 P145 367
I/O (LDC) P37 P44 C16 P48 P62 214 I/O - - - P112 P146 370
I/O - - E15 P49 P63 217 I/O (D1) P69 P101 T3 P113 P147 373
I/O - - D16 P50 P64 220 I/O (RCLK, P70 P102 P5 P114 P148 376
GND - P45 F14 P51 P67 - RDY/BUSY)
I/O - P46 F15 P52 P68 223 I/O - P103 R4 P115 P149 379
I/O - P47 E16 P53 P69 226 I/O - P104 R3 P116 P150 382
I/O P38 P48 F16 P54 P70 229 I/O (D0, DIN) P71 P105 P4 P117 P151 385
I/O P39 P49 G14 P55 P71 232 I/O, SGCK4 (DOUT) P72 P106 T2 P118 P152 388
I/O - P50 G15 P56 P74 235 CCLK P73 P107 R2 P119 P153 -
I/O - P51 G16 P57 P75 238 VCC P74 P108 P3 P120 P154 -
I/O P40 P52 H16 P58 P76 241 O, TDO P75 P109 T1 P121 P159 0
I/O (INIT) P41 P53 H15 P59 P77 244 GND P76 P110 N3 P122 P160 -
VCC P42 P54 H14 P60 P78 - I/O (A0, WS) P77 P111 R1 P123 P161 2
GND P43 P55 J14 P61 P79 - I/O, PGCK4 (A1) P78 P112 P2 P124 P162 5
I/O P44 P56 J15 P62 P80 247 I/O - P113 N2 P125 P163 8
I/O P45 P57 J16 P63 P81 250 I/O - P114 M3 P126 P164 11
I/O - P58 K16 P64 P82 253 I/O (CS1, A2) P79 P115 P1 P127 P165 14
I/O - P59 K15 P65 P83 256 I/O (A3) P80 P116 N1 P128 P166 17
I/O P46 P60 K14 P66 P86 259 I/O - P117 M2 P129 P167 20
I/O P47 P61 L16 P67 P87 262 I/O - - M1 P130 P168 23
I/O - P62 M16 P68 P88 265 GND - P118 L3 P131 P171 - 6
I/O - P63 L15 P69 P89 268 I/O - P119 L2 P132 P172 26
GND - P64 L14 P70 P90 - I/O - P120 L1 P133 P173 29
I/O - - N16 P71 P93 271 I/O (A4) P81 P121 K3 P134 P174 32
I/O - - M15 P72 P94 274 I/O (A5) P82 P122 K2 P135 P175 35
I/O P48 P65 P16 P73 P95 277 I/O - P123 K1 P137 P178 38
I/O P49 P66 M14 P74 P96 280 I/O - P124 J1 P138 P179 41
I/O - P67 N15 P75 P97 283 I/O (A6) P83 P125 J2 P139 P180 44
I/O - P68 P15 P76 P98 286 I/O (A7) P84 P126 J3 P140 P181 47
I/O P50 P69 N14 P77 P99 289 GND P1 P127 H2 P141 P182 -
I/O, SGCK3 P51 P70 R16 P78 P100 292 5/5/97
GND P52 P71 P14 P79 P101 -
DONE P53 P72 R15 P80 P103 -
VCC P54 P73 P13 P81 P106 -
PROGRAM P55 P74 R14 P82 P108 - Additional XC4006E Package Pins
I/O (D7) P56 P75 T16 P83 P109 295 PQ160
I/O, PGCK3 P57 P76 T15 P84 P110 298 Not Connected Pins
I/O - P77 R13 P85 P111 301 P136 - - - -
I/O - P78 P12 P86 P112 304 5/5/97
I/O (D6) P58 P79 T14 P87 P113 307
I/O - P80 T13 P88 P114 310
I/O - - R12 P89 P115 313 PQ208
I/O - - T12 P90 P116 316 Not Connected Pins
GND - P81 P11 P91 P119 - P1 P3 P12 P13 P19
I/O - P82 R11 P92 P120 319 P20 P31 P32 P38 P39
I/O - P83 T11 P93 P121 322 P51 P52 P53 P54 P65
I/O (D5) P59 P84 T10 P94 P122 325 P66 P72 P73 P84 P85
I/O (CS0) P60 P85 P10 P95 P123 328 P91 P92 P102 P104 P105
I/O - P86 R10 P96 P126 331 P107 P117 P118 P124 P125
I/O - P87 T9 P97 P127 334 P136 P137 P143 P144 P155
I/O (D4) P61 P88 R9 P98 P128 337 P156 P157 P158 P169 P170
I/O P62 P89 P9 P99 P129 340 P176 P177 P188 P189 P195
VCC P63 P90 R8 P100 P130 - P196 P206 P207 P208 -
GND P64 P91 P8 P101 P131 - 6/5/97
I/O (D3) P65 P92 T8 P102 P132 343
I/O (RS) P66 P93 T7 P103 P133 346
January 29, 1999 (Version 1.5) 6-119

R

XC4008E Pad Name PC84 PQ160 PG191 PQ208 Bndry Scan XC4008E Pad Name PC84 PQ160 PG191 PQ208 Bndry Scan
VCC P2 P142 J4 P183 - I/O (HDC) P36 P44 E16 P58 226
I/O (A8) P3 P143 J3 P184 56 I/O - P45 C17 P59 229
I/O (A9) P4 P144 J2 P185 59 I/O - P46 D17 P60 232
I/O - P145 J1 P186 62 I/O - P47 B18 P61 235
I/O - P146 H1 P187 65 I/O (LDC) P37 P48 E17 P62 238
I/O - - H2 P188 68 I/O - P49 F16 P63 241
I/O - - H3 P189 71 I/O - P50 C18 P64 244
I/O (A10) P5 P147 G1 P190 74 GND - P51 G16 P67 -
I/O (A11) P6 P148 G2 P191 77 I/O - P52 E18 P68 247
I/O - P149 F1 P192 80 I/O - P53 F18 P69 250
I/O - P150 E1 P193 83 I/O P38 P54 G17 P70 253
GND - P151 G3 P194 - I/O P39 P55 G18 P71 256
I/O - P152 C1 P197 86 I/O - - H16 P72 259
I/O - P153 E2 P198 89 I/O - - H17 P73 262
I/O (A12) P7 P154 F3 P199 92 I/O - P56 H18 P74 265
I/O (A13) P8 P155 D2 P200 95 I/O - P57 J18 P75 268
I/O - P156 B1 P201 98 I/O P40 P58 J17 P76 271
I/O - P157 E3 P202 101 I/O (INIT) P41 P59 J16 P77 274
I/O (A14) P9 P158 C2 P203 104 VCC P42 P60 J15 P78 -
I/O, SGCK1 (A15) P10 P159 B2 P204 107 GND P43 P61 K15 P79 -
VCC P11 P160 D3 P205 - I/O P44 P62 K16 P80 277
GND P12 P1 D4 P2 - I/O P45 P63 K17 P81 280
I/O, PGCK1 (A16) P13 P2 C3 P4 110 I/O - P64 K18 P82 283
I/O (A17) P14 P3 C4 P5 113 I/O - P65 L18 P83 286
I/O - P4 B3 P6 116 I/O - - L17 P84 289
I/O - P5 C5 P7 119 I/O - - L16 P85 292
I/O, TDI P15 P6 A2 P8 122 I/O P46 P66 M18 P86 295
I/O, TCK P16 P7 B4 P9 125 I/O P47 P67 M17 P87 298
I/O - P8 C6 P10 128 I/O - P68 N18 P88 301
I/O - P9 A3 P11 131 I/O - P69 P18 P89 304
GND - P10 C7 P14 - GND - P70 M16 P90 -
I/O - P11 A4 P15 134 I/O - P71 T18 P93 307
I/O - P12 A5 P16 137 I/O - P72 P17 P94 310
I/O, TMS P17 P13 B7 P17 140 I/O P48 P73 N16 P95 313
I/O P18 P14 A6 P18 143 I/O P49 P74 T17 P96 316
I/O - - C8 P19 146 I/O - P75 R17 P97 319
I/O - - A7 P20 149 I/O - P76 P16 P98 322
I/O - P15 B8 P21 152 I/O P50 P77 U18 P99 325
I/O - P16 A8 P22 155 I/O, SGCK3 P51 P78 T16 P100 328
I/O P19 P17 B9 P23 158 GND P52 P79 R16 P101 -
I/O P20 P18 C9 P24 161 DONE P53 P80 U17 P103 -
GND P21 P19 D9 P25 - VCC P54 P81 R15 P106 -
VCC P22 P20 D10 P26 - PROGRAM P55 P82 V18 P108 -
I/O P23 P21 C10 P27 164 I/O (D7) P56 P83 T15 P109 331
I/O P24 P22 B10 P28 167 I/O, PGCK3 P57 P84 U16 P110 334
I/O - P23 A9 P29 170 I/O - P85 T14 P111 337
I/O - P24 A10 P30 173 I/O - P86 U15 P112 340
I/O - - A11 P31 176 I/O (D6) P58 P87 V17 P113 343
I/O - - C11 P32 179 I/O - P88 V16 P114 346
I/O P25 P25 B11 P33 182 I/O - P89 T13 P115 349
I/O P26 P26 A12 P34 185 I/O - P90 U14 P116 352
I/O - P27 B12 P35 188 GND - P91 T12 P119 -
I/O - P28 A13 P36 191 I/O - P92 U13 P120 355
GND - P29 C12 P37 - I/O - P93 V13 P121 358
I/O - P30 A15 P40 194 I/O (D5) P59 P94 U12 P122 361
I/O - P31 C13 P41 197 I/O (CS0) P60 P95 V12 P123 364
I/O P27 P32 B14 P42 200 I/O - - T11 P124 367
I/O - P33 A16 P43 203 I/O - - U11 P125 370
I/O - P34 B15 P44 206 I/O - P96 V11 P126 373
I/O - P35 C14 P45 209 I/O - P97 V10 P127 376
I/O P28 P36 A17 P46 212 I/O (D4) P61 P98 U10 P128 379
I/O, SGCK2 P29 P37 B16 P47 215 I/O P62 P99 T10 P129 382
O (M1) P30 P38 C15 P48 218 VCC P63 P100 R10 P130 -
GND P31 P39 D15 P49 - GND P64 P101 R9 P131 -
I (M0) P32 P40 A18 P50 221 I/O (D3) P65 P102 T9 P132 385
VCC P33 P41 D16 P55 - I/O (RS) P66 P103 U9 P133 388
I (M2) P34 P42 C16 P56 222 I/O - P104 V9 P134 391
I/O, PGCK2 P35 P43 B17 P57 223 I/O - P105 V8 P135 394
I/O - - U8 P136 397
6-120 January 29, 1999 (Version 1.5)

R
XC4008E Pad Name PC84 PQ160 PG191 PQ208 Bndry Scan XC4008E Pad Name PC84 PQ160 PG191 PQ208 Bndry Scan
I/O - - T8 P137 400 I/O - - L3 P176 38
I/O (D2) P67 P106 V7 P138 403 I/O - P136 L2 P177 41
I/O P68 P107 U7 P139 406 I/O - P137 L1 P178 44
I/O - P108 V6 P140 409 I/O - P138 K1 P179 47
I/O - P109 U6 P141 412 I/O (A6) P83 P139 K2 P180 50
GND - P110 T7 P142 - I/O (A7) P84 P140 K3 P181 53
I/O - P111 U5 P145 415 GND P1 P141 K4 P182 -
I/O - P112 T6 P146 418 5/5/97
I/O (D1) P69 P113 V3 P147 421
I/O (RCLK, RDY/BUSY) P70 P114 V2 P148 424
I/O - P115 U4 P149 427
I/O - P116 T5 P150 430 Additional XC4008E Package Pins
I/O (D0, DIN) P71 P117 U3 P151 433 PG191
I/O, SGCK4 (DOUT) P72 P118 T4 P152 436 Not Connected Pins
CCLK P73 P119 V1 P153 - A14 B5 B6 B13 D1 D18
VCC P74 P120 R4 P154 - F2 F17 N2 N17 R1 R18
O, TDO P75 P121 U2 P159 0 V4 V5 V14 V15 - -
GND P76 P122 R3 P160 - 6/3/97
I/O (A0, WS) P77 P123 T3 P161 2
I/O, PGCK4 (A1) P78 P124 U1 P162 5
I/O - P125 P3 P163 8
PQ208
I/O - P126 R2 P164 11
I/O (CS1, A2) P79 P127 T2 P165 14 Not Connected Pins
I/O (A3) P80 P128 N3 P166 17 P1 P3 P12 P13 P38 P39
I/O - P129 P2 P167 20 P51 P52 P53 P54 P65 P66
P91 P92 P102 P104 P105 P107
I/O - P130 T1 P168 23
GND - P131 M3 P171 - P117 P118 P143 P144 P155 P156
I/O - P132 P1 P172 26 P157 P158 P169 P170 P195 P196
I/O - P133 N1 P173 29 P206 P207 P208 - - -
I/O (A4) P81 P134 M2 P174 32
6/3/97 6
I/O (A5) P82 P135 M1 P175 35

PQ/H PQ/H
XC4010E/XL PC PQ TQ PQ TQ PG BG BG Bndry XC4010E/XL PC PQ TQ PQ TQ PG BG BG Bndry
Q Q
Pad Name 84 100†† 144†† 160 176†† 191† 225† 256†† Scan Pad Name 84 100†† 144†† 160 176†† 191† 225† 256†† Scan
208 208
VCC P2 P92 P128 P142 P155 VCC* P183 VCC* VCC* - I/O - - P4 P4 P4 B3 P6 C2 D2 128
I/O (A8) P3 P93 P129 P143 P156 J3 P184 E8 C10 62 I/O - - P5 P5 P5 C5 P7 E5 D3 131
I/O (A9) P4 P94 P130 P144 P157 J2 P185 B7 D10 65 I/O, TDI P15 P7 P6 P6 P6 A2 P8 D3 E4 134
I/O (19) - P95 P131 P145 P158 J1 P186 A7 A9 68 I/O, TCK P16 P8 P7 P7 P7 B4 P9 C1 C1 137
I/O (18) - P96 P132 P146 P159 H1 P187 C7 B9 71 I/O - - - P8 P8 C6 P10 D2 D1 140
I/O - - - - P160 H2 P188 D7 C9 74 I/O - - - P9 P9 A3 P11 G6 E3 143
I/O - - - - P161 H3 P189 E7 D9 77 I/O - - - - - B5 P12 E4 E2 146
I/O (A10) P5 P97 P133 P147 P162 G1 P190 A6 A8 80 I/O - - - - - B6 P13 D1 E1 149
I/O (A11) P6 P98 P134 P148 P163 G2 P191 B6 B8 83 GND - - P8 P10 P10 GND* P14 GND* GND* -
VCC - - - - - VCC* - VCC* VCC* - I/O - - P9 P11 P11 A4 P15 F5 G3 152
I/O - - P135 P149 P164 F1 P192 A5 B6 86 I/O - - P10 P12 P12 A5 P16 E1 G2 155
I/O - - P136 P150 P165 E1 P193 B5 A5 89 I/O, TMS P17 P9 P11 P13 P13 B7 P17 F4 G1 158
GND - - P137 P151 P166 GND* P194 GND* GND* - I/O P18 P10 P12 P14 P14 A6 P18 F3 H3 161
I/O - - - - - F2 P195 D6 C6 92 VCC - - - - - VCC* - VCC* VCC* -
I/O - - - - P167 D1 P196 C5 B5 95 I/O - - - - P15 C8 P19 G4 J2 164
I/O - - - P152 P168 C1 P197 A4 A4 98 I/O - - - - P16 A7 P20 G3 J1 167
I/O - - - P153 P169 E2 P198 E6 C5 101 I/O - - P13 P15 P17 B8 P21 G2 K2 170
I/O (A12) P7 P99 P138 P154 P170 F3 P199 B4 B4 104 I/O - P11 P14 P16 P18 A8 P22 G1 K3 173
I/O (A13) P8 P100 P139 P155 P171 D2 P200 D5 A3 107 I/O P19 P12 P15 P17 P19 B9 P23 G5 K1 176
I/O - - P140 P156 P172 B1 P201 B3 B3 110 I/O P20 P13 P16 P18 P20 C9 P24 H3 L1 179
I/O - - P141 P157 P173 E3 P202 F6 B2 113 GND P21 P14 P17 P19 P21 GND* P25 GND* GND* -
I/O (A14) P9 P1 P142 P158 P174 C2 P203 A2 A2 116 VCC P22 P15 P18 P20 P22 VCC* P26 VCC* VCC* -
I/O, SGCK1 †, P10 P2 P143 P159 P175 B2 P204 C3 C3 119 I/O P23 P16 P19 P21 P23 C10 P27 H4 L2 182
GCK8 †† I/O P24 P17 P20 P22 P24 B10 P28 H5 L3 185
(A15) I/O - P18 P21 P23 P25 A9 P29 J2 L4 188
VCC P11 P3 P144 P160 P176 VCC* P205 VCC* VCC* - I/O - - P22 P24 P26 A10 P30 J1 M1 191
GND P12 P4 P1 P1 P1 GND* P2 GND* GND* - I/O - - - - P27 A11 P31 J3 M2 194
I/O, PGCK1†, P13 P5 P2 P2 P2 C3 P4 D4 B1 122 I/O - - - - P28 C11 P32 J4 M3 197
GCK1†† VCC - - - - - VCC* - VCC* VCC* -
(A16) I/O P25 P19 P23 P25 P29 B11 P33 K2 P1 200
I/O (A17) P14 P6 P3 P3 P3 C4 P5 B1 C2 125 I/O P26 P20 P24 P26 P30 A12 P34 K3 P2 203
January 29, 1999 (Version 1.5) 6-121

R
PQ/H PQ/H
XC4010E/XL PC PQ TQ PQ TQ PG BG BG Bndry XC4010E/XL PC PQ TQ PQ TQ PG BG BG Bndry
Q Q
Pad Name 84 100†† 144†† 160 176†† 191† 225† 256†† Scan Pad Name 84 100†† 144†† 160 176†† 191† 225† 256†† Scan
208 208
I/O - - P25 P27 P31 B12 P35 J6 R1 206 I/O - - P77 P85 P93 T14 P111 L11 T17 373
I/O - - P26 P28 P32 A13 P36 L1 P3 209 I/O - - P78 P86 P94 U15 P112 M13 V20 376
GND - - P27 P29 P33 GND* P37 GND* GND* - I/O (D6) P58 P58 P79 P87 P95 V17 P113 J10 T19 379
I/O - - - - - B13 P38 L3 T2 212 I/O - P59 P80 P88 P96 V16 P114 L12 T20 382
I/O - - - - - A14 P39 M1 U1 215 I/O - - - P89 P97 T13 P115 M15 R18 385
I/O - - - P30 P34 A15 P40 K5 T3 218 I/O - - - P90 P98 U14 P116 L13 R19 388
I/O - - - P31 P35 C13 P41 M2 U2 221 I/O - - - - - V15 P117 L14 R20 391
I/O P27 P21 P28 P32 P36 B14 P42 L4 V1 224 I/O - - - - - V14 P118 K11 P18 394
I/O - P22 P29 P33 P37 A16 P43 N1 T4 227 GND - - P81 P91 P99 GND* P119 GND* GND* -
I/O - - P30 P34 P38 B15 P44 M3 U3 230 I/O - - P82 P92 P100 U13 P120 K13 N19 397
I/O - - P31 P35 P39 C14 P45 N2 V2 233 I/O - - P83 P93 P101 V13 P121 K14 N20 400
I/O P28 P23 P32 P36 P40 A17 P46 K6 W1 236 VCC - - - - - VCC* - VCC* VCC* -
I/O, SGCK2 †, P29 P24 P33 P37 P41 B16 P47 P1 V3 239 I/O (D5) P59 P60 P84 P94 P102 U12 P122 K15 M17 403
GCK2 †† I/O (CS0) P60 P61 P85 P95 P103 V12 P123 J12 M18 406
O (M1) P30 P25 P34 P38 P42 C15 P48 N3 W2 242 I/O - - - - P104 T11 P124 J13 M20 409
GND P31 P26 P35 P39 P43 GND* P49 GND* GND* - I/O - - - - P105 U11 P125 J14 L19 412
I (M0) P32 P27 P36 P40 P44 A18 P50 P2 Y1 245 I/O - P62 P86 P96 P106 V11 P126 J15 L18 415
VCC P33 P28 P37 P41 P45 VCC* P55 VCC* VCC* - I/O - P63 P87 P97 P107 V10 P127 J11 L20 418
I (M2) P34 P29 P38 P42 P46 C16 P56 M4 W3 246 I/O (D4) P61 P64 P88 P98 P108 U10 P128 H13 K20 421
I/O, PGCK2 †, P35 P30 P39 P43 P47 B17 P57 R2 Y2 247 I/O P62 P65 P89 P99 P109 T10 P129 H14 K19 424
GCK3 †† VCC P63 P66 P90 P100 P110 VCC* P130 VCC* VCC* -
I/O (HDC) P36 P31 P40 P44 P48 E16 P58 P3 W4 250 GND P64 P67 P91 P101 P111 GND* P131 GND* GND* -
I/O - - P41 P45 P49 C17 P59 L5 V4 253 I/O (D3) P65 P68 P92 P102 P112 T9 P132 H12 K18 427
I/O - - P42 P46 P50 D17 P60 N4 U5 256 I/O (RS) P66 P69 P93 P103 P113 U9 P133 H11 K17 430
I/O - P32 P43 P47 P51 B18 P61 R3 Y3 259 I/O - P70 P94 P104 P114 V9 P134 G14 J20 433
I/O (LDC) P37 P33 P44 P48 P52 E17 P62 P4 Y4 262 I/O - - P95 P105 P115 V8 P135 G15 J19 436
I/O - - - P49 P53 F16 P63 K7 V5 265 I/O - - - - P116 U8 P136 G13 J18 439
I/O - - - P50 P54 C18 P64 M5 W5 268 I/O - - - - P117 T8 P137 G12 J17 442
I/O - - - - - D18 P65 R4 Y5 271 I/O (D2) P67 P71 P96 P106 P118 V7 P138 G11 H19 445
I/O - - - - - F17 P66 N5 V6 274 I/O P68 P72 P97 P107 P119 U7 P139 F15 H18 448
GND - - P45 P51 P55 GND* P67 GND* GND* - VCC - - - - - VCC* - VCC* VCC* -
I/O - - P46 P52 P56 E18 P68 R5 W7 277 I/O - - P98 P108 P120 V6 P140 F14 G19 451
I/O - - P47 P53 P57 F18 P69 M6 Y7 280 I/O - - P99 P109 P121 U6 P141 F13 F20 454
I/O P38 P34 P48 P54 P58 G17 P70 N6 V8 283 GND - - P100 P110 P122 GND* P142 GND* GND* -
I/O P39 P35 P49 P55 P59 G18 P71 P6 W8 286 I/O - - - - - V5 P143 E13 D20 457
VCC - - - - - VCC* - VCC* VCC* - I/O - - - - - V4 P144 D15 E18 460
I/O - - - - P60 H16 P72 R6 Y8 289 I/O - - - P111 P123 U5 P145 F11 D19 463
I/O - - - - P61 H17 P73 M7 U9 292 I/O - - - P112 P124 T6 P146 D14 C20 466
I/O - P36 P50 P56 P62 H18 P74 R7 V10 295 I/O (D1) P69 P73 P101 P113 P125 V3 P147 E12 E17 469
I/O - P37 P51 P57 P63 J18 P75 L7 Y10 298 I/O (RCLK, P70 P74 P102 P114 P126 V2 P148 C15 D18 472
I/O P40 P38 P52 P58 P64 J17 P76 N8 Y11 301 RDY/BUSY)
I/O (INIT) P41 P39 P53 P59 P65 J16 P77 P8 W11 304 I/O - - P103 P115 P127 U4 P149 D13 C19 475
VCC P42 P40 P54 P60 P66 VCC* P78 VCC* VCC* - I/O - - P104 P116 P128 T5 P150 C14 B20 478
GND P43 P41 P55 P61 P67 GND* P79 GND* GND* - I/O (D0, DIN) P71 P75 P105 P117 P129 U3 P151 F10 C18 481
I/O P44 P42 P56 P62 P68 K16 P80 L8 V11 307 I/O, SGCK4 †, P72 P76 P106 P118 P130 T4 P152 B15 B19 484
I/O P45 P43 P57 P63 P69 K17 P81 P9 U11 310 GCK6 ††
I/O - P44 P58 P64 P70 K18 P82 R9 Y12 313 (DOUT)
I/O - P45 P59 P65 P71 L18 P83 N9 W12 316 CCLK P73 P77 P107 P119 P131 V1 P153 C13 A20 -
I/O - - - - P72 L17 P84 M9 V12 319 VCC P74 P78 P108 P120 P132 VCC* P154 VCC* VCC* -
I/O - - - - P73 L16 P85 L9 U12 322 O, TDO P75 P79 P109 P121 P133 U2 P159 A15 A19 0
VCC - - - - - VCC* - VCC* VCC* - GND P76 P80 P110 P122 P134 GND* P160 GND* GND* -
I/O P46 P46 P60 P66 P74 M18 P86 N10 Y15 325 I/O (A0, WS) P77 P81 P111 P123 P135 T3 P161 A14 B18 2
I/O P47 P47 P61 P67 P75 M17 P87 K9 V14 328 I/O, PGCK4 †, P78 P82 P112 P124 P136 U1 P162 B13 B17 5
I/O - - P62 P68 P76 N18 P88 R11 W15 331 GCK7 †† (A1)
I/O - - P63 P69 P77 P18 P89 P11 Y16 334 I/O - - P113 P125 P137 P3 P163 E11 C17 8
GND - - P64 P70 P78 GND* P90 GND* GND* - I/O - - P114 P126 P138 R2 P164 C12 D16 11
I/O - - - - - N17 P91 R12 Y17 337 I/O (CS1, A2) P79 P83 P115 P127 P139 T2 P165 A13 A18 14
I/O - - - - - R18 P92 L10 V16 340 I/O (A3) P80 P84 P116 P128 P140 N3 P166 B12 A17 17
I/O - - - P71 P79 T18 P93 P12 W17 343 I/O - - P117 P129 P141 P2 P167 A12 A16 20
I/O - - - P72 P80 P17 P94 M11 Y18 346 I/O - - - P130 P142 T1 P168 C11 C15 23
I/O P48 P48 P65 P73 P81 N16 P95 R13 U16 349 I/O - - - - - R1 P169 B11 B15 26
I/O P49 P49 P66 P74 P82 T17 P96 N12 V17 352 I/O - - - - - N2 P170 E10 A15 29
I/O - - P67 P75 P83 R17 P97 P13 W18 355 GND - - P118 P131 P143 GND* P171 GND* GND* -
I/O - - P68 P76 P84 P16 P98 K10 Y19 358 I/O - - P119 P132 P144 P1 P172 A11 B14 32
I/O P50 P50 P69 P77 P85 U18 P99 R14 V18 361 I/O - - P120 P133 P145 N1 P173 D10 A14 35
I/O, SGCK3 †, P51 P51 P70 P78 P86 T16 P100 N13 W19 364 VCC - - - - - VCC* - VCC* VCC* -
GCK4 †† I/O (A4) P81 P85 P121 P134 P146 M2 P174 A10 C12 38
GND P52 P52 P71 P79 P87 GND* P101 GND* GND* - I/O (A5) P82 P86 P122 P135 P147 M1 P175 D9 B12 41
DONE P53 P53 P72 P80 P88 U17 P103 P14 Y20 - I/O - - - - P148 L3 P176 C9 A12 44
VCC P54 P54 P73 P81 P89 VCC* P106 VCC* VCC* - I/O - - - P136 P149 L2 P177 B9 B11 47
PROGRAM P55 P55 P74 P82 P90 V18 P108 M12 V19 - I/O (A21)†† - P87 P123 P137 P150 L1 P178 A9 C11 50
I/O (D7) P56 P56 P75 P83 P91 T15 P109 P15 U19 367 I/O (A20)†† - P88 P124 P138 P151 K1 P179 E9 A11 53
I/O, PGCK3 †, P57 P57 P76 P84 P92 U16 P110 N14 U18 370 I/O (A6) P83 P89 P125 P139 P152 K2 P180 C8 A10 56
GCK5 †† I/O (A7) P84 P90 P126 P140 P153 K3 P181 B8 B10 59
6-122 January 29, 1999 (Version 1.5)

R
XC4010E/XL PC PQ TQ PQ TQ PG
PQ/H
BG BG Bndry BG225
Q
Pad Name 84 100†† 144†† 160 176†† 191† 225† 256†† Scan VCC Pins
208
GND P1 P91 P127 P141 P154 GND* P182 GND* GND* -
B2 B14 D8 H1 H15 R1 R8
6/19/97 R15 - - - - - -
GND Pins
* Pads labelled GND* or VCC* are internally bonded to Ground or
A1 A8 D12 F8 G7 G8 G9
VCC planes within the package. They have no direct connection to
H2 H6 H7 H8 H9 H10 J7
any specific package pin.
J8 J9 K8 M8 - - -
† = E only Not Connected Pins
†† = XL only A3 B10 C4 C6 C10 D11 E2
E3 E14 E15 F1 F2 F7 F9
Additional XC4010E/XL Package Pins F12 G10 J5 K1 K4 K12 L2
PQ/HQ208 L6 L15 M10 M14 N7 N11 N15
Not Connected Pins P5 P7 P10 R10 - - -
P1 P3 P51 P52 P53 P54 P102 6/16/97
P104 P105 P107 P155 P156 P157 P158
P206 P207 P208 - - - -
5/27/97 BG256
VCC Pins
C14 D6 D7 D11 D14 D15 E20
F1 F4 F17 G4 G17 K4 L17
PG191
P4 P17 P19 R2 R4 R17 U6
VCC Pins
U7 U10 U14 U15 V7 W20 -
D3 D10 D16 J4 J15 R4 R10
GND Pins
R15 - - - - - -
A1 B7 D4 D8 D13 D17 G20
GND Pins
H4 H17 N3 N4 N17 U4 U8
C7 C12 D4 D9 D15 G3 G16
U13 U17 W14 - - - -
K4 K15 M3 M16 R3 R9 R16
Not Connected Pins
T7 T12 - - - - -
A6 A7 A13 B13 B16 C4 C7
5/27/97
C8 C13 C16 D5 D12 E19 F2
F3 F18 F19 G18 H1 H2 H20
6
J3 J4 M4 M19 N1 N2 N18
P20 R3 T1 T18 U20 V9 V13
V15 W6 W9 W10 W13 W16 Y6
Y9 Y13 Y14 - - - -
5/27/97

XC4013E PQ/H XC4013E PQ/H

HT PQ HT PQ/HQ PG BG BG Bndry HT PQ HT PQ/HQ PG BG BG Bndry
/XL Q /XL Q
144†† 160 176†† 208 223† 225† 256†† Scan 144†† 160 176†† 208 223† 225† 256†† Scan
Pad Name 240 Pad Name 240
VCC P128 P142 P155 P183 VCC* VCC* P212 VCC* - I/O (A14) P142 P158 P174 P203 C2 A2 P238 A2 140
I/O (A8) P129 P143 P156 P184 J3 E8 P213 C10 74 I/O, P143 P159 P175 P204 B2 C3 P239 C3 143
SGCK1 †,
I/O (A9) P130 P144 P157 P185 J2 B7 P214 D10 77
GCK8 ††
I/O P131 P145 P158 P186 J1 A7 P215 A9 80
(A15)
(A19) ††
VCC P144 P160 P176 P205 VCC* VCC* P240 VCC* -
I/O P132 P146 P159 P187 H1 C7 P216 B9 83
GND P1 P1 P1 P2 GND* GND* P1 GND* -
(A18) ††
I/O, P2 P2 P2 P4 C3 D4 P2 B1 146
I/O - - P160 P188 H2 D7 P217 C9 86
PGCK1 †,
I/O - - P161 P189 H3 E7 P218 D9 89
GCK1 ††
I/O (A10) P133 P147 P162 P190 G1 A6 P220 A8 92 (A16)
I/O (A11) P134 P148 P163 P191 G2 B6 P221 B8 95 I/O (A17) P3 P3 P3 P5 C4 B1 P3 C2 149
VCC - - - - VCC* VCC* P222 VCC* - I/O P4 P4 P4 P6 B3 C2 P4 D2 152
I/O - - - - H4 C6 P223 A6 98 I/O P5 P5 P5 P7 C5 E5 P5 D3 155
I/O - - - - G4 F7 P224 C7 101 I/O, TDI P6 P6 P6 P8 A2 D3 P6 E4 158
I/O P135 P149 P164 P192 F1 A5 P225 B6 104 I/O, TCK P7 P7 P7 P9 B4 C1 P7 C1 161
I/O P136 P150 P165 P193 E1 B5 P226 A5 107 I/O - P8 P8 P10 C6 D2 P8 D1 164
GND P137 P151 P166 P194 GND* GND* P227 GND* - I/O - P9 P9 P11 A3 G6 P9 E3 167
I/O - - - P195 F2 D6 P228 C6 110 I/O - - - P12 B5 E4 P10 E2 170
I/O - - P167 P196 D1 C5 P229 B5 113 I/O - - - P13 B6 D1 P11 E1 173
I/O - P152 P168 P197 C1 A4 P230 A4 116 I/O - - - - D5 E3 P12 F3 176
I/O - P153 P169 P198 E2 E6 P231 C5 119 I/O - - - - D6 E2 P13 F2 179
I/O (A12) P138 P154 P170 P199 F3 B4 P232 B4 122 GND P8 P10 P10 P14 GND* GND* P14 GND* -
I/O (A13) P139 P155 P171 P200 D2 D5 P233 A3 125 I/O P9 P11 P11 P15 A4 F5 P15 G3 182
I/O - - - - F4 A3 P234 D5 128 I/O P10 P12 P12 P16 A5 E1 P16 G2 185
I/O - - - - E4 C4 P235 C4 131 I/O, TMS P11 P13 P13 P17 B7 F4 P17 G1 188
I/O P140 P156 P172 P201 B1 B3 P236 B3 134 I/O P12 P14 P14 P18 A6 F3 P18 H3 191
I/O P141 P157 P173 P202 E3 F6 P237 B2 137 VCC - - - - VCC* VCC* P19 VCC* -
January 29, 1999 (Version 1.5) 6-123

R
XC4013E PQ/H XC4013E PQ/H

HT PQ HT PQ/HQ PG BG BG Bndry HT PQ HT PQ/HQ PG BG BG Bndry
/XL Q /XL Q
144†† 160 176†† 208 223† 225† 256†† Scan 144†† 160 176†† 208 223† 225† 256†† Scan
Pad Name 240 Pad Name 240
I/O - - - - D7 F2 P20 H2 194 I/O P57 P63 P69 P81 K17 P9 P93 U11 370
I/O - - - - D8 F1 P21 H1 197 I/O P58 P64 P70 P82 K18 R9 P94 Y12 373
I/O - - P15 P19 C8 G4 P23 J2 200 I/O P59 P65 P71 P83 L18 N9 P95 W12 376
I/O - - P16 P20 A7 G3 P24 J1 203 I/O - - P72 P84 L17 M9 P96 V12 379
I/O P13 P15 P17 P21 B8 G2 P25 K2 206 I/O - - P73 P85 L16 L9 P97 U12 382
I/O P14 P16 P18 P22 A8 G1 P26 K3 209 I/O - - - - L15 R10 P99 V13 385
I/O P15 P17 P19 P23 B9 G5 P27 K1 212 I/O - - - - M15 P10 P100 Y14 388
I/O P16 P18 P20 P24 C9 H3 P28 L1 215 VCC - - - - VCC* VCC* P101 VCC* -
GND P17 P19 P21 P25 GND* GND* P29 GND* - I/O P60 P66 P74 P86 M18 N10 P102 Y15 391
VCC P18 P20 P22 P26 VCC* VCC* P30 VCC* - I/O P61 P67 P75 P87 M17 K9 P103 V14 394
I/O P19 P21 P23 P27 C10 H4 P31 L2 218 I/O P62 P68 P76 P88 N18 R11 P104 W15 397
I/O P20 P22 P24 P28 B10 H5 P32 L3 221 I/O P63 P69 P77 P89 P18 P11 P105 Y16 400
I/O P21 P23 P25 P29 A9 J2 P33 L4 224 GND P64 P70 P78 P90 GND* GND* P106 GND* -
I/O P22 P24 P26 P30 A10 J1 P34 M1 227 I/O - - - - N15 M10 P107 V15 403
I/O - - P27 P31 A11 J3 P35 M2 230 I/O - - - - P15 N11 P108 W16 406
I/O - - P28 P32 C11 J4 P36 M3 233 I/O - - - P91 N17 R12 P109 Y17 409
I/O - - - - D11 J5 P38 N1 236 I/O - - - P92 R18 L10 P110 V16 412
I/O - - - - D12 K1 P39 N2 239 I/O - P71 P79 P93 T18 P12 P111 W17 415
VCC - - - - VCC* VCC* P40 VCC* - I/O - P72 P80 P94 P17 M11 P112 Y18 418
I/O P23 P25 P29 P33 B11 K2 P41 P1 242 I/O P65 P73 P81 P95 N16 R13 P113 U16 421
I/O P24 P26 P30 P34 A12 K3 P42 P2 245 I/O P66 P74 P82 P96 T17 N12 P114 V17 424
I/O P25 P27 P31 P35 B12 J6 P43 R1 248 I/O P67 P75 P83 P97 R17 P13 P115 W18 427
I/O P26 P28 P32 P36 A13 L1 P44 P3 251 I/O P68 P76 P84 P98 P16 K10 P116 Y19 430
GND P27 P29 P33 P37 GND* GND* P45 GND* - I/O P69 P77 P85 P99 U18 R14 P117 V18 433
I/O - - - - D13 L2 P46 T1 254 I/O, P70 P78 P86 P100 T16 N13 P118 W19 436
I/O - - - - D14 K4 P47 R3 257 SGCK3 †,
I/O - - - P38 B13 L3 P48 T2 260 GCK4 ††
I/O - - - P39 A14 M1 P49 U1 263 GND P71 P79 P87 P101 GND* GND* P119 GND* -
I/O - P30 P34 P40 A15 K5 P50 T3 266 DONE P72 P80 P88 P103 U17 P14 P120 Y20 -
I/O - P31 P35 P41 C13 M2 P51 U2 269 VCC P73 P81 P89 P106 VCC* VCC* P121 VCC* -
I/O P28 P32 P36 P42 B14 L4 P52 V1 272 PRO- P74 P82 P90 P108 V18 M12 P122 V19 -
I/O P29 P33 P37 P43 A16 N1 P53 T4 275 GRAM
I/O P30 P34 P38 P44 B15 M3 P54 U3 278 I/O (D7) P75 P83 P91 P109 T15 P15 P123 U19 439
I/O P31 P35 P39 P45 C14 N2 P55 V2 281 I/O, P76 P84 P92 P110 U16 N14 P124 U18 442
PGCK3 †,
I/O P32 P36 P40 P46 A17 K6 P56 W1 284
GCK5 ††
I/O, P33 P37 P41 P47 B16 P1 P57 V3 287
I/O P77 P85 P93 P111 T14 L11 P125 T17 445
SGCK2 †,
GCK2 †† I/O P78 P86 P94 P112 U15 M13 P126 V20 448
O (M1) P34 P38 P42 P48 C15 N3 P58 W2 290 I/O - - - - R14 N15 P127 U20 451
GND P35 P39 P43 P49 GND* GND* P59 GND* - I/O - - - - R13 M14 P128 T18 454
I (M0) P36 P40 P44 P50 A18 P2 P60 Y1 293 I/O (D6) P79 P87 P95 P113 V17 J10 P129 T19 457
VCC P37 P41 P45 P55 VCC* VCC* P61 VCC* - I/O P80 P88 P96 P114 V16 L12 P130 T20 460
I (M2) P38 P42 P46 P56 C16 M4 P62 W3 294 I/O - P89 P97 P115 T13 M15 P131 R18 463
I/O, P39 P43 P47 P57 B17 R2 P63 Y2 295 I/O - P90 P98 P116 U14 L13 P132 R19 466
PGCK2 †, I/O - - - P117 V15 L14 P133 R20 469
GCK3 †† I/O - - - P118 V14 K11 P134 P18 472
I/O (HDC) P40 P44 P48 P58 E16 P3 P64 W4 298 GND P81 P91 P99 P119 GND* GND* P135 GND* -
I/O P41 P45 P49 P59 C17 L5 P65 V4 301 I/O - - - - R12 L15 P136 P20 475
I/O P42 P46 P50 P60 D17 N4 P66 U5 304 I/O - - - - R11 K12 P137 N18 478
I/O P43 P47 P51 P61 B18 R3 P67 Y3 307 I/O P82 P92 P100 P120 U13 K13 P138 N19 481
I/O (LDC) P44 P48 P52 P62 E17 P4 P68 Y4 310 I/O P83 P93 P101 P121 V13 K14 P139 N20 484
I/O - P49 P53 P63 F16 K7 P69 V5 313 VCC - - - - VCC* VCC* P140 VCC* -
I/O - P50 P54 P64 C18 M5 P70 W5 316 I/O (D5) P84 P94 P102 P122 U12 K15 P141 M17 487
I/O - - - P65 D18 R4 P71 Y5 319 I/O (CS0) P85 P95 P103 P123 V12 J12 P142 M18 490
I/O - - - P66 F17 N5 P72 V6 322 I/O - - P104 P124 T11 J13 P144 M20 493
I/O - - - - E15 P5 P73 W6 325 I/O - - P105 P125 U11 J14 P145 L19 496
I/O - - - - F15 L6 P74 Y6 328 I/O P86 P96 P106 P126 V11 J15 P146 L18 499
GND P45 P51 P55 P67 GND* GND* P75 GND* - I/O P87 P97 P107 P127 V10 J11 P147 L20 502
I/O P46 P52 P56 P68 E18 R5 P76 W7 331 I/O (D4) P88 P98 P108 P128 U10 H13 P148 K20 505
I/O P47 P53 P57 P69 F18 M6 P77 Y7 334 I/O P89 P99 P109 P129 T10 H14 P149 K19 508
I/O P48 P54 P58 P70 G17 N6 P78 V8 337 VCC P90 P100 P110 P130 VCC* VCC* P150 VCC* -
I/O P49 P55 P59 P71 G18 P6 P79 W8 340 GND P91 P101 P111 P131 GND* GND* P151 GND* -
VCC - - - - VCC* VCC* P80 VCC* - I/O (D3) P92 P102 P112 P132 T9 H12 P152 K18 511
I/O - - P60 P72 H16 R6 P81 Y8 343 I/O (RS) P93 P103 P113 P133 U9 H11 P153 K17 514
I/O - - P61 P73 H17 M7 P82 U9 346 I/O P94 P104 P114 P134 V9 G14 P154 J20 517
I/O - - - - G15 N7 P84 Y9 349 I/O P95 P105 P115 P135 V8 G15 P155 J19 520
I/O - - - - H15 P7 P85 W10 352 I/O - - P116 P136 U8 G13 P156 J18 523
I/O P50 P56 P62 P74 H18 R7 P86 V10 355 I/O - - P117 P137 T8 G12 P157 J17 526
I/O P51 P57 P63 P75 J18 L7 P87 Y10 358 I/O (D2) P96 P106 P118 P138 V7 G11 P159 H19 529
I/O P52 P58 P64 P76 J17 N8 P88 Y11 361 I/O P97 P107 P119 P139 U7 F15 P160 H18 532
I/O (INIT) P53 P59 P65 P77 J16 P8 P89 W11 364 VCC - - - - VCC* VCC* P161 VCC* -
VCC P54 P60 P66 P78 VCC* VCC* P90 VCC* - I/O P98 P108 P120 P140 V6 F14 P162 G19 535
GND P55 P61 P67 P79 GND* GND* P91 GND* - I/O P99 P109 P121 P141 U6 F13 P163 F20 538
I/O P56 P62 P68 P80 K16 L8 P92 V11 367 I/O - - - - R8 G10 P164 G18 541
6-124 January 29, 1999 (Version 1.5)

R
XC4013E
HT PQ HT PQ/HQ PG BG
PQ/H
BG Bndry Additional XC4013E/XL Package Pins
/XL Q
144†† 160 176†† 208 223† 225† 256†† Scan
Pad Name 240 PQ/HQ208
I/O - - - - R7 E15 P165 F19 544 Not Connected Pins
GND P100 P110 P122 P142 GND* GND* P166 GND* - P1 P3 P51 P52 P53 P54
I/O - - - - R6 E14 P167 F18 547 P102 P104 P105 P107 P155 P156
I/O - - - - R5 F12 P168 E19 550 P157 P158 P206 P207 P208 -
I/O - - - P143 V5 E13 P169 D20 553
5/5/97
I/O - - - P144 V4 D15 P170 E18 556
I/O - P111 P123 P145 U5 F11 P171 D19 559
I/O - P112 P124 P146 T6 D14 P172 C20 562
I/O (D1) P101 P113 P125 P147 V3 E12 P173 E17 565 PG223
I/O (RCLK, P102 P114 P126 P148 V2 C15 P174 D18 568 VCC Pins
RDY/BUS D3 D10 D16 J4 J15 R4
Y) R10 R15 - - - -
I/O P103 P115 P127 P149 U4 D13 P175 C19 571
GND Pins
I/O P104 P116 P128 P150 T5 C14 P176 B20 574
C7 C12 D4 D9 D15 G3
I/O (D0, P105 P117 P129 P151 U3 F10 P177 C18 577
DIN) G16 K4 K15 M3 M16 R3
I/O, P106 P118 P130 P152 T4 B15 P178 B19 580 R9 R16 T7 T12 - -
SGCK4 †, 5/5/97
GCK6 ††
(DOUT)
CCLK P107 P119 P131 P153 V1 C13 P179 A20 -
VCC P108 P120 P132 P154 VCC* VCC* P180 VCC* - BG225
O, TDO P109 P121 P133 P159 U2 A15 P181 A19 0 VCC Pins
GND P110 P122 P134 P160 GND* GND* P182 GND* - B2 B14 D8 H1 H15
I/O (A0, P111 P123 P135 P161 T3 A14 P183 B18 2 R1 R8 R15 - -
WS) GND Pins
I/O, P112 P124 P136 P162 U1 B13 P184 B17 5 A1 A8 D12 F8 G7
PGCK4 †, G8 G9 H2 H6 H7
GCK7 ††
H8 H9 H10 J7 J8
(A1)
I/O P113 P125 P137 P163 P3 E11 P185 C17 8 J9 K8 M8 - - 6
I/O P114 P126 P138 P164 R2 C12 P186 D16 11 5/5/97
I/O (CS1, P115 P127 P139 P165 T2 A13 P187 A18 14
A2)
The BG225 package pins in this table are bonded to an internal
I/O (A3) P116 P128 P140 P166 N3 B12 P188 A17 17 Ground plane on the XC4013E die. They must all be externally con-
I/O - - - - P4 F9 P189 C16 20 nected to Ground.
I/O - - - - N4 D11 P190 B16 23
I/O P117 P129 P141 P167 P2 A12 P191 A16 26 PQ/HQ240
I/O - P130 P142 P168 T1 C11 P192 C15 29 GND Pins
I/O - - - P169 R1 B11 P193 B15 32 P22‡ P37‡ P83‡ P98‡ P143‡ P158‡
I/O - - - P170 N2 E10 P194 A15 35 P204‡ P219‡ - - - -
GND P118 P131 P143 P171 GND* GND* P196 GND* - Not Connected Pins
I/O P119 P132 P144 P172 P1 A11 P197 B14 38 P195 - - - - -
I/O P120 P133 P145 P173 N1 D10 P198 A14 41 6/9/97
I/O - - - - M4 C10 P199 C13 44
I/O - - - - L4 B10 P200 B13 47 ‡ Pins marked with this symbol are used for Ground connections on
VCC - - - - VCC* VCC* P201 VCC* - some revisions of the device. These pins may not physically con-
I/O (A4) P121 P134 P146 P174 M2 A10 P202 C12 50 nect to anything on the current device revision. However, they
I/O (A5) P122 P135 P147 P175 M1 D9 P203 B12 53 should be externally connected to Ground, if possible.
I/O - - P148 P176 L3 C9 P205 A12 56
I/O - P136 P149 P177 L2 B9 P206 B11 59 BG256
I/O P123 P137 P150 P178 L1 A9 P207 C11 62 VCC Pins
(A21) ††
C14 D6 D7 D11 D14 D15
I/O P124 P138 P151 P179 K1 E9 P208 A11 65
(A20) †† E20 F1 F4 F17 G4 G17
I/O (A6) P125 P139 P152 P180 K2 C8 P209 A10 68 K4 L17 P4 P17 P19 R2
I/O (A7) P126 P140 P153 P181 K3 B8 P210 B10 71 R4 R17 U6 U7 U10 U14
GND P127 P141 P154 P182 GND* GND* P211 GND* - U15 V7 W20 - - -
6/9/97 GND Pins
* Pads labelled GND* or VCC* are internally bonded to Ground or A1 B7 D4 D8 D13 D17
VCC planes within the package. They have no direct connection to G20 H4 H17 N3 N4 N17
any specific package pin. U4 U8 U13 U17 W14 -
† = E only, †† = XL only Not Connected Pins
A7 A13 C8 D12 H20 J3
J4 M4 M19 V9 W9 W13
Y13 - - - - -
6/4/97
January 29, 1999 (Version 1.5) 6-125

R

XC4020E/XL HT PQ HT HQ208† PG HQ240† BG Bndry XC4020E/XL HT PQ HT HQ208† PG HQ240† BG Bndry
Pad Name 144†† 160†† 176†† PQ208†† 223† PQ240†† 256†† Scan Pad Name 144†† 160†† 176†† PQ208†† 223† PQ240†† 256†† Scan
VCC P128 P142 P155 P183 VCC* P212 VCC* - I/O P24 P26 P30 P34 A12 P42 P2 287
I/O (A8) P129 P143 P156 P184 J3 P213 C10 86 I/O P25 P27 P31 P35 B12 P43 R1 290
I/O (A9) P130 P144 P157 P185 J2 P214 D10 89 I/O P26 P28 P32 P36 A13 P44 P3 293
I/O (A19) †† P131 P145 P158 P186 J1 P215 A9 92 GND P27 P29 P33 P37 GND* P45 GND* -
I/O (A18) †† P132 P146 P159 P187 H1 P216 B9 95 I/O - - - - D13 P46 T1 296
I/O - - P160 P188 H2 P217 C9 98 I/O - - - - D14 P47 R3 299
I/O - - P161 P189 H3 P218 D9 101 I/O - - - P38 B13 P48 T2 302
I/O (A10) P133 P147 P162 P190 G1 P220 A8 104 I/O - - - P39 A14 P49 U1 305
I/O (A11) P134 P148 P163 P191 G2 P221 B8 107 I/O - P30 P34 P40 A15 P50 T3 308
I/O - - - - - - C8 110 I/O - P31 P35 P41 C13 P51 U2 311
I/O - - - - - - A7 113 I/O P28 P32 P36 P42 B14 P52 V1 320
VCC - - - - VCC* P222 VCC* - I/O P29 P33 P37 P43 A16 P53 T4 323
I/O - - - - H4 P223 A6 116 I/O P30 P34 P38 P44 B15 P54 U3 326
I/O - - - - G4 P224 C7 119 I/O P31 P35 P39 P45 C14 P55 V2 329
I/O P135 P149 P164 P192 F1 P225 B6 122 I/O P32 P36 P40 P46 A17 P56 W1 332
I/O P136 P150 P165 P193 E1 P226 A5 125 I/O, SGCK2 †, P33 P37 P41 P47 B16 P57 V3 335
GCK2 ††
GND P137 P151 P166 P194 GND* P227 GND* -
O (M1) P34 P38 P42 P48 C15 P58 W2 338
I/O - - - P195 F2 P228 C6 128
GND P35 P39 P43 P49 GND* P59 GND* -
I/O - - P167 P196 D1 P229 B5 131
I (M0) P36 P40 P44 P50 A18 P60 Y1 341
I/O - P152 P168 P197 C1 P230 A4 134
VCC P37 P41 P45 P55 VCC* P61 VCC* -
I/O - P153 P169 P198 E2 P231 C5 137
I (M2) P38 P42 P46 P56 C16 P62 W3 342
I/O (A12) P138 P154 P170 P199 F3 P232 B4 140
I/O PGCK2 †, P39 P43 P47 P57 B17 P63 Y2 343
I/O (A13) P139 P155 P171 P200 D2 P233 A3 143
GCK3 ††
I/O - - - - F4 P234 D5 152
I/O (HDC) P40 P44 P48 P58 E16 P64 W4 346
I/O - - - - E4 P235 C4 155
I/O P41 P45 P49 P59 C17 P65 V4 349
I/O P140 P156 P172 P201 B1 P236 B3 158
I/O P42 P46 P50 P60 D17 P66 U5 352
I/O P141 P157 P173 P202 E3 P237 B2 161
I/O P43 P47 P51 P61 B18 P67 Y3 355
I/O (A14) P142 P158 P174 P203 C2 P238 A2 164
I/O (LDC) P44 P48 P52 P62 E17 P68 Y4 358
I/O, SGCK1 †, P143 P159 P175 P204 B2 P239 C3 167
I/O - P49 P53 P63 F16 P69 V5 367
GCK8 †† (A15)
I/O - P50 P54 P64 C18 P70 W5 370
VCC P144 P160 P176 P205 VCC* P240 VCC* -
I/O - - - P65 D18 P71 Y5 373
GND P1 P1 P1 P2 GND* P1 GND* -
I/O - - - P66 F17 P72 V6 376
I/O, PGCK1 †, P2 P2 P2 P4 C3 P2 B1 170
GCK1 †† (A16) I/O - - - - E15 P73 W6 379
I/O (A17) P3 P3 P3 P5 C4 P3 C2 173 I/O - - - - F15 P74 Y6 382
I/O P4 P4 P4 P6 B3 P4 D2 176 GND P45 P51 P55 P67 GND* P75 GND* -
I/O P5 P5 P5 P7 C5 P5 D3 179 I/O P46 P52 P56 P68 E18 P76 W7 385
I/O, TDI P6 P6 P6 P8 A2 P6 E4 182 I/O P47 P53 P57 P69 F18 P77 Y7 388
I/O, TCK P7 P7 P7 P9 B4 P7 C1 185 I/O P48 P54 P58 P70 G17 P78 V8 391
I/O - P8 P8 P10 C6 P8 D1 194 I/O P49 P55 P59 P71 G18 P79 W8 394
I/O - P9 P9 P11 A3 P9 E3 197 VCC - - - - VCC* P80 VCC* -
I/O - - - P12 B5 P10 E2 200 I/O - - P60 P72 H16 P81 Y8 397
I/O - - - P13 B6 P11 E1 203 I/O - - P61 P73 H17 P82 U9 400
I/O - - - - D5 P12 F3 206 I/O - - - - - - V9 403
I/O - - - - D6 P13 F2 209 I/O - - - - - - W9 406
GND P8 P10 P10 P14 GND* P14 GND* - I/O - - - - G15 P84 Y9 409
I/O P9 P11 P11 P15 A4 P15 G3 212 I/O - - - - H15 P85 W10 412
I/O P10 P12 P12 P16 A5 P16 G2 215 I/O P50 P56 P62 P74 H18 P86 V10 415
I/O, TMS P11 P13 P13 P17 B7 P17 G1 218 I/O P51 P57 P63 P75 J18 P87 Y10 418
I/O P12 P14 P14 P18 A6 P18 H3 221 I/O P52 P58 P64 P76 J17 P88 Y11 421
VCC - - - - VCC* P19 VCC* - I/O (INIT) P53 P59 P65 P77 J16 P89 W11 424
I/O - - - - D7 P20 H2 224 VCC P54 P60 P66 P78 VCC* P90 VCC* -
I/O - - - - D8 P21 H1 227 GND P55 P61 P67 P79 GND* P91 GND* -
I/O - - - - - - J4 230 I/O P56 P62 P68 P80 K16 P92 V11 427
I/O - - - - - - J3 233 I/O P57 P63 P69 P81 K17 P93 U11 430
I/O - - P15 P19 C8 P23 J2 236 I/O P58 P64 P70 P82 K18 P94 Y12 433
I/O - - P16 P20 A7 P24 J1 239 I/O P59 P65 P71 P83 L18 P95 W12 436
I/O P13 P15 P17 P21 B8 P25 K2 242 I/O - - P72 P84 L17 P96 V12 439
I/O P14 P16 P18 P22 A8 P26 K3 245 I/O - - P73 P85 L16 P97 U12 442
I/O P15 P17 P19 P23 B9 P27 K1 248 I/O - - - - - - Y13 445
I/O P16 P18 P20 P24 C9 P28 L1 251 I/O - - - - - - W13 448
GND P17 P19 P21 P25 GND* P29 GND* - I/O - - - - L15 P99 V13 451
VCC P18 P20 P22 P26 VCC* P30 VCC* - I/O - - - - M15 P100 Y14 454
I/O P19 P21 P23 P27 C10 P31 L2 254 VCC - - - - VCC* P101 VCC* -
I/O P20 P22 P24 P28 B10 P32 L3 257 I/O P60 P66 P74 P86 M18 P102 Y15 457
I/O P21 P23 P25 P29 A9 P33 L4 260 I/O P61 P67 P75 P87 M17 P103 V14 460
I/O P22 P24 P26 P30 A10 P34 M1 263 I/O P62 P68 P76 P88 N18 P104 W15 463
I/O - - P27 P31 A11 P35 M2 266 I/O P63 P69 P77 P89 P18 P105 Y16 466
I/O - - P28 P32 C11 P36 M3 269 GND P64 P70 P78 P90 GND* P106 GND* -
I/O - - - - - - M4 272 I/O - - - - N15 P107 V15 469
I/O - - - - D11 P38 N1 278 I/O - - - - P15 P108 W16 472
I/O - - - - D12 P39 N2 281 I/O - - - P91 N17 P109 Y17 475
VCC - - - - VCC* P40 VCC* - I/O - - - P92 R18 P110 V16 478
I/O P23 P25 P29 P33 B11 P41 P1 284 I/O - P71 P79 P93 T18 P111 W17 481
I/O - P72 P80 P94 P17 P112 Y18 484
6-126 January 29, 1999 (Version 1.5)

R
XC4020E/XL HT PQ HT HQ208† PG HQ240† BG Bndry XC4020E/XL HT PQ HT HQ208† PG HQ240† BG Bndry

Pad Name 144†† 160†† 176†† PQ208†† 223† PQ240†† 256†† Scan Pad Name 144†† 160†† 176†† PQ208†† 223† PQ240†† 256†† Scan
I/O P65 P73 P81 P95 N16 P113 U16 493 I/O - - - - P4 P189 C16 26
I/O P66 P74 P82 P96 T17 P114 V17 496 I/O - - - - N4 P190 B16 29
I/O P67 P75 P83 P97 R17 P115 W18 499 I/O P117 P129 P141 P167 P2 P191 A16 32
I/O P68 P76 P84 P98 P16 P116 Y19 502 I/O - P130 P142 P168 T1 P192 C15 35
I/O P69 P77 P85 P99 U18 P117 V18 505 I/O - - - P169 R1 P193 B15 38
I/O, SGCK3 †, P70 P78 P86 P100 T16 P118 W19 508 I/O - - - P170 N2 P194 A15 41
GCK4 †† GND P118 P131 P143 P171 GND* P196 GND* -
GND P71 P79 P87 P101 GND* P119 GND* - I/O P119 P132 P144 P172 P1 P197 B14 44
DONE P72 P80 P88 P103 U17 P120 Y20 - I/O P120 P133 P145 P173 N1 P198 A14 47
VCC P73 P81 P89 P106 VCC* P121 VCC* - I/O - - - - M4 P199 C13 50
PROGRAM P74 P82 P90 P108 V18 P122 V19 - I/O - - - - L4 P200 B13 53
I/O (D7) P75 P83 P91 P109 T15 P123 U19 511 VCC - - - - VCC* P201 VCC* -
I/O, PGCK3 †, P76 P84 P92 P110 U16 P124 U18 514 I/O - - - - - - A13 56
GCK5 †† I/O - - - - - - D12 59
I/O P77 P85 P93 P111 T14 P125 T17 517 I/O (A4) P121 P134 P146 P174 M2 P202 C12 62
I/O P78 P86 P94 P112 U15 P126 V20 520 I/O (A5) P122 P135 P147 P175 M1 P203 B12 65
I/O - - - - R14 P127 U20 523 I/O - - P148 P176 L3 P205 A12 68
I/O - - - - R13 P128 T18 526 I/O - P136 P149 P177 L2 P206 B11 71
I/O (D6) P79 P87 P95 P113 V17 P129 T19 535 I/O (A21) †† P123 P137 P150 P178 L1 P207 C11 74
I/O P80 P88 P96 P114 V16 P130 T20 538 I/O (A20) †† P124 P138 P151 P179 K1 P208 A11 77
I/O - P89 P97 P115 T13 P131 R18 541 I/O (A6) P125 P139 P152 P180 K2 P209 A10 80
I/O - P90 P98 P116 U14 P132 R19 544 I/O (A7) P126 P140 P153 P181 K3 P210 B10 83
I/O - - - P117 V15 P133 R20 547 GND P127 P141 P154 P182 GND* P211 GND* -
I/O - - - P118 V14 P134 P18 550 6/24/97
GND P81 P91 P99 P119 GND* P135 GND* -
I/O - - - - R12 P136 P20 553 † = E only
I/O - - - - R11 P137 N18 556 †† = XL only
I/O P82 P92 P100 P120 U13 P138 N19 559
I/O
VCC
P83
-
P93
-
P101
-
P121
-
V13
VCC*
P139
P140
N20
VCC*
562
-
Additional XC4020E/XL Package Pins
I/O (D5) P84 P94 P102 P122 U12 P141 M17 565 PQ/HQ208
I/O (CS0) P85 P95 P103 P123 V12 P142 M18 568 Not Connected Pins
I/O
I/O
-
-
-
-
-
P104
-
P124
-
T11
-
P144
M19
M20
574
577
P1 P3 P51 P52 P53 P54 6
I/O - - P105 P125 U11 P145 L19 580
P102 P104 P105 P107 P155 P156
I/O P86 P96 P106 P126 V11 P146 L18 583 P157 P158 P206 P207 P208 -
I/O P87 P97 P107 P127 V10 P147 L20 586 5/5/97
I/O (D4) P88 P98 P108 P128 U10 P148 K20 589
I/O P89 P99 P109 P129 T10 P149 K19 592
VCC P90 P100 P110 P130 VCC* P150 VCC* -
GND P91 P101 P111 P131 GND* P151 GND* - PG223
I/O (D3) P92 P102 P112 P132 T9 P152 K18 595 VCC Pins
I/O (RS) P93 P103 P113 P133 U9 P153 K17 598 D3 D10 D16 J4 J15 R4
I/O P94 P104 P114 P134 V9 P154 J20 601 R10 R15 - - - -
I/O P95 P105 P115 P135 V8 P155 J19 604 GND Pins
I/O - - P116 P136 U8 P156 J18 607
C7 C12 D4 D9 D15 G3
I/O - - P117 P137 T8 P157 J17 610
G16 K4 K15 M3 M16 R3
I/O - - - - - - H20 613
I/O (D2) P96 P106 P118 P138 V7 P159 H19 619 R9 R16 T7 T12 - -
I/O P97 P107 P119 P139 U7 P160 H18 622 5/5/97
VCC - - - VCC* P161 VCC* -
I/O P98 P108 P120 P140 V6 P162 G19 625
I/O P99 P109 P121 P141 U6 P163 F20 628 PQ/HQ240
I/O - - - - R8 P164 G18 631 GND Pins
I/O - - - - R7 P165 F19 634 P22‡ P37‡ P83‡ P98‡ P143‡ P158‡
GND P100 P110 P122 P142 GND* P166 GND* - P204‡ P219‡ - - - -
I/O - - - - R6 P167 F18 637
Not Connected Pins
I/O - - - - R5 P168 E19 640
I/O - - - P143 V5 P169 D20 643
P195 - - - - -
I/O - - - P144 V4 P170 E18 646 6/9/97
I/O - P111 P123 P145 U5 P171 D19 649 ‡ Pins marked with this symbol are used for Ground connections on
I/O - P112 P124 P146 T6 P172 C20 652
I/O (D1) P101 P113 P125 P147 V3 P173 E17 655
some revisions of the device. These pins may not physically con-
I/O (RCLK, P102 P114 P126 P148 V2 P174 D18 658 nect to anything on the current device revision. However, they
RDY/BUSY) should be externally connected to Ground, if possible.
I/O P103 P115 P127 P149 U4 P175 C19 667
I/O P104 P116 P128 P150 T5 P176 B20 670 BG256
I/O (D0, DIN) P105 P117 P129 P151 U3 P177 C18 673
VCC Pins
I/O, SGCK4 †, P106 P118 P130 P152 T4 P178 B19 676
GCK6 †† (DOUT) C14 D6 D7 D11 D14 D15
CCLK P107 P119 P131 P153 V1 P179 A20 - E20 F1 F4 F17 G4 G17
VCC P108 P120 P132 P154 VCC* P180 VCC* - K4 L17 P4 P17 P19 R2
O, TDO P109 P121 P133 P159 U2 P181 A19 0 R4 R17 U6 U7 U10 U14
GND P110 P122 P134 P160 GND* P182 GND* - U15 V7 W20 - - -
I/O (A0, WS) P111 P123 P135 P161 T3 P183 B18 2
GND Pins
I/O, PGCK4 †, P112 P124 P136 P162 U1 P184 B17 5
GCK7 †† (A1) A1 B7 D4 D8 D13 D17
I/O P113 P125 P137 P163 P3 P185 C17 8 G20 H4 H17 N3 N4 N17
I/O P114 P126 P138 P164 R2 P186 D16 11 U4 U8 U13 U17 W14 -
I/O (CS1, A2) P115 P127 P139 P165 T2 P187 A18 14 6/17/97
I/O (A3) P116 P128 P140 P166 N3 P188 A17 17
January 29, 1999 (Version 1.5) 6-127

R
Pin Locations for XC4025E, XC4028EX/XL Devices

XC4025E, XC4025E,
XC4028 HQ HQ PG HQ BG PG HQ BG Bndry XC4028 HQ HQ PG HQ BG PG HQ BG Bndry
EX/XL 160†† 208‡ 223† 240 256†† 299 304 352‡ Scan EX/XL 160†† 208‡ 223† 240 256†† 299 304 352‡ Scan
Pad Name Pad Name
VCC P142 P183 VCC* P212 VCC* VCC* P38 VCC* - I/O - - - - - B5 P288 G26 239
I/O (A8) P143 P184 J3 P213 C10 K2 P37 D14 98 GND P10 P14 GND* P14 GND* GND* P287 GND* -
I/O (A9) P144 P185 J2 P214 D10 K3 P36 C14 101 I/O P11 P15 A4 P15 G3 B6 P286 J23 242
I/O (A19) ‡ P145 P186 J1 P215 A9 K5 P35 A15 104 I/O P12 P16 A5 P16 G2 D8 P285 J24 245
I/O (A18) ‡ P146 P187 H1 P216 B9 K4 P34 B15 107 I/O, TMS P13 P17 B7 P17 G1 C7 P284 H25 248
I/O - P188 H2 P217 C9 J1 P33 C15 110 I/O P14 P18 A6 P18 H3 B7 P283 K23 251
I/O - P189 H3 P218 D9 J2 P32 D15 113 VCC - - VCC* P19 VCC* VCC* P282 VCC* -
I/O (A10) P147 P190 G1 P220 A8 H1 P31 A16 116 I/O - - D7 P20 H2 C8 P280 K24 254
I/O (A11) P148 P191 G2 P221 B8 J3 P30 B16 119 I/O - - D8 P21 H1 E9 P279 J25 257
GND - - - - GND* GND* - GND* - I/O - - - - - A7 P278 L24 260
I/O - - - - - J4 P29 C16 122 I/O - - - - - D9 P277 K25 263
I/O - - - - - J5 P28 B17 125 GND - - - P22 GND* GND* - GND* -
I/O - - - - C8 H2 P27 C17 128 I/O - - - - J4 B8 P276 L25 266
I/O - - - - A7 G1 P26 B18 131 I/O - - - - J3 A8 P275 L26 269
VCC - - VCC* P222 VCC* VCC* P25 VCC* - I/O - P19 C8 P23 J2 C9 P274 M23 272
I/O - - H4 P223 A6 H3 P23 C18 134 I/O - P20 A7 P24 J1 B9 P273 M24 275
I/O - - G4 P224 C7 G2 P22 D17 137 I/O P15 P21 B8 P25 K2 E10 P272 M25 278
I/O P149 P192 F1 P225 B6 H4 P21 A20 140 I/O P16 P22 A8 P26 K3 A9 P271 M26 281
I/O P150 P193 E1 P226 A5 F2 P20 B19 143 I/O P17 P23 B9 P27 K1 D10 P270 N24 284
GND P151 P194 GND* P227 GND* GND* P19 GND* - I/O P18 P24 C9 P28 L1 C10 P269 N25 287
I/O - - - - - H5 P18 C19 146 GND P19 P25 GND* P29 GND* GND* P268 GND* -
I/O - - - - - G3 P17 D18 149 VCC P20 P26 VCC* P30 VCC* VCC* P267 VCC* -
I/O - P195 F2 P228 C6 D1 P16 A21 152 I/O P21 P27 C10 P31 L2 B10 P266 N26 290
I/O - P196 D1 P229 B5 G4 P15 B20 155 I/O P22 P28 B10 P32 L3 B11 P265 P25 293
I/O P152 P197 C1 P230 A4 E2 P14 C20 158 I/O P23 P29 A9 P33 L4 C11 P264 P23 296
I/O P153 P198 E2 P231 C5 F3 P13 B21 161 I/O P24 P30 A10 P34 M1 E11 P263 P24 299
I/O (A12) P154 P199 F3 P232 B4 G5 P12 B22 164 I/O - P31 A11 P35 M2 D11 P262 R26 302
I/O (A13) P155 P200 D2 P233 A3 C1 P10 C21 167 I/O - P32 C11 P36 M3 A12 P261 R25 305
GND - - - - GND* GND* - GND* - I/O - - - - M4 B12 P260 R24 308
VCC - - - - VCC* VCC* - VCC* - I/O - - - - - A13 P259 R23 311
I/O - - - - - F4 P9 D20 170 GND - - - P37 GND* GND* - GND* -
I/O - - - - - E3 P8 A23 173 I/O - - - - - C12 P258 T26 314
I/O - - F4 P234 D5 D2 P7 D21 176 I/O - - - - - D12 P257 T25 317
I/O - - E4 P235 C4 C2 P6 C22 179 I/O - - D11 P38 N1 E12 P256 T23 320
I/O P156 P201 B1 P236 B3 F5 P5 B24 182 I/O - - D12 P39 N2 B13 P255 V26 323
I/O P157 P202 E3 P237 B2 E4 P4 C23 185 VCC - - VCC* P40 VCC* VCC* P253 VCC* -
I/O (A14) P158 P203 C2 P238 A2 D3 P3 D22 188 I/O P25 P33 B11 P41 P1 A14 P252 U24 326
I/O, P159 P204 B2 P239 C3 C3 P2 C24 191 I/O P26 P34 A12 P42 P2 C13 P251 V25 329
SGCK1 †, I/O P27 P35 B12 P43 R1 B14 P250 V24 332
GCK8 ‡ I/O P28 P36 A13 P44 P3 D13 P249 U23 335
(A15) GND P29 P37 GND* P45 GND* GND* P248 GND* -
VCC P160 P205 VCC* P240 VCC* VCC* P1 VCC* - I/O - - - - - B15 P247 Y26 338
GND P1 P2 GND* P1 GND* GND* P304 GND* - I/O - - - - - E13 P246 W25 341
I/O, P2 P4 C3 P2 B1 D4 P303 D23 194 I/O - - D13 P46 T1 C14 P245 W24 344
PGCK1 †, I/O - - D14 P47 R3 A17 P244 V23 347
GCK1 ‡ I/O - P38 B13 P48 T2 D14 P243 AA26 350
(A16)
I/O - P39 A14 P49 U1 B16 P242 Y25 353
I/O (A17) P3 P5 C4 P3 C2 B2 P302 C25 197
I/O P30 P40 A15 P50 T3 C15 P241 Y24 356
I/O P4 P6 B3 P4 D2 B3 P301 D24 200
I/O P31 P41 C13 P51 U2 E14 P240 AA25 359
I/O P5 P7 C5 P5 D3 E6 P300 E23 203
GND - - - - GND* GND* - GND* -
I/O, TDI P6 P8 A2 P6 E4 D5 P299 C26 206
VCC - - - - VCC* VCC* - VCC* -
I/O, TCK P7 P9 B4 P7 C1 C4 P298 E24 209
I/O - - - - - A18 P239 AB25 362
I/O - - - - - A3 P297 F24 212
I/O - - - - - D15 P238 AA24 365
I/O - - - - - D6 P296 E25 215
I/O P32 P42 B14 P52 V1 C16 P237 Y23 368
I/O P33 P43 A16 P53 T4 B17 P236 AC26 371
I/O P34 P44 B15 P54 U3 B18 P235 AA23 374
I/O P8 P10 C6 P8 D1 E7 P295 D26 218
I/O P35 P45 C14 P55 V2 E15 P234 AB24 377
I/O P9 P11 A3 P9 E3 B4 P294 G24 221
I/O P36 P46 A17 P56 W1 D16 P233 AD25 380
I/O - P12 B5 P10 E2 C5 P293 F25 224
I/O, P37 P47 B16 P57 V3 C17 P232 AC24 383
I/O - P13 B6 P11 E1 A4 P292 F26 227 SGCK2 †,
I/O - - D5 P12 F3 D7 P291 H23 230 GCK2 ‡
I/O - - D6 P13 F2 C6 P290 H24 233 O (M1) P38 P48 C15 P58 W2 A20 P231 AB23 386
I/O - - - - - E8 P289 G25 236 GND P39 P49 GND* P59 GND* GND* P230 GND* -
6-128 January 29, 1999 (Version 1.5)

R
XC4025E, XC4025E,
Pad Name Pad Name
I (M0) P40 P50 A18 P60 Y1 C18 P229 AD24 389 I/O - P92 R18 P110 V16 R18 P165 AD7 550
VCC P41 P55 VCC* P61 VCC* VCC* P228 VCC* - I/O P71 P93 T18 P111 W17 P16 P164 AE6 553
I (M2) P42 P56 C16 P62 W3 D17 P227 AC23 390 I/O P72 P94 P17 P112 Y18 V20 P163 AE5 556
I/O, P43 P57 B17 P63 Y2 B19 P226 AE24 391 GND - - - - GND* GND* - GND* -
PGCK2 †, VCC - - - - VCC* VCC* - VCC* -
GCK3 ‡ I/O - - - - - R17 P162 AD6 559
I/O (HDC) P44 P58 E16 P64 W4 C19 P225 AD23 394 I/O - - - - - T18 P161 AC7 562
I/O P45 P59 C17 P65 V4 F16 P224 AC22 397 I/O P73 P95 N16 P113 U16 U19 P160 AF4 565
I/O P46 P60 D17 P66 U5 E17 P223 AF24 400 I/O P74 P96 T17 P114 V17 V19 P159 AF3 568
I/O P47 P61 B18 P67 Y3 D18 P222 AD22 403 I/O P75 P97 R17 P115 W18 R16 P158 AD5 571
I/O (LDC) P48 P62 E17 P68 Y4 C20 P221 AE23 406 I/O P76 P98 P16 P116 Y19 T17 P157 AE3 574
I/O - - - - - F17 P220 AE22 409 I/O P77 P99 U18 P117 V18 U18 P156 AD4 577
I/O - - - - - G16 P219 AF23 412 I/O, P78 P100 T16 P118 W19 X20 P155 AC5 580
VCC - - - - VCC* VCC* - VCC* - SGCK3 †,
GND - - - - GND* GND* - GND* - GCK4 ‡
I/O P49 P63 F16 P69 V5 D19 P218 AD20 415 GND P79 P101 GND* P119 GND* GND* P154 GND* -
I/O P50 P64 C18 P70 W5 E18 P217 AE21 418 DONE P80 P103 U17 P120 Y20 V18 P153 AD3 -
I/O - P65 D18 P71 Y5 D20 P216 AF21 421 VCC P81 P106 VCC* P121 VCC* VCC* P152 VCC* -
I/O - P66 F17 P72 V6 G17 P215 AC19 424 PRO- P82 P108 V18 P122 V19 U17 P151 AC4 -
I/O - - E15 P73 W6 F18 P214 AD19 427 GRAM
I/O - - F15 P74 Y6 H16 P213 AE20 430 I/O (D7) P83 P109 T15 P123 U19 W19 P150 AD2 583
I/O - - - - - E19 P212 AF20 433 I/O, P84 P110 U16 P124 U18 W18 P149 AC3 586
I/O - - - - - F19 P211 AC18 436 PGCK3 †,
GND P51 P67 GND* P75 GND* GND* P210 GND* - GCK5 ‡
I/O P52 P68 E18 P76 W7 H17 P209 AD18 439 I/O P85 P111 T14 P125 T17 T15 P148 AB4 589
I/O P53 P69 F18 P77 Y7 G18 P208 AE19 442 I/O P86 P112 U15 P126 V20 U16 P147 AD1 592
I/O - - R14 P127 U20 V17 P146 AA4 595
I/O P54 P70 G17 P78 V8 G19 P207 AC17 445
I/O - - R13 P128 T18 X18 P145 AA3 598
6
I/O P55 P71 G18 P79 W8 H18 P206 AD17 448
VCC - - VCC* P80 VCC* VCC* P204 VCC* - I/O - - - - - U15 P144 AB2 601
I/O - P72 H16 P81 Y8 J16 P203 AE18 451 I/O - - - - - T14 P143 AC1 604
I/O - P73 H17 P82 U9 G20 P202 AF18 454 VCC - - - - VCC* VCC* - VCC* -
I/O - - - - - J17 P201 AE17 457 GND - - - - GND* GND* - GND* -
I/O - - - - - H19 P200 AE16 460 I/O (D6) P87 P113 V17 P129 T19 W17 P142 Y3 607
GND - - - P83 GND* GND* - GND* - I/O P88 P114 V16 P130 T20 V16 P141 AA2 610
I/O - - - - V9 H20 P199 AF16 463 I/O P89 P115 T13 P131 R18 X17 P140 AA1 613
I/O - - - - W9 J18 P198 AC15 466 I/O P90 P116 U14 P132 R19 U14 P139 W4 616
I/O - - G15 P84 Y9 J19 P197 AD15 469 I/O - P117 V15 P133 R20 V15 P138 W3 619
I/O - - H15 P85 W10 K16 P196 AE15 472 I/O - P118 V14 P134 P18 T13 P137 Y2 622
I/O P56 P74 H18 P86 V10 J20 P195 AF15 475 I/O - - - - - W16 P136 Y1 625
I/O P57 P75 J18 P87 Y10 K17 P194 AD14 478 I/O - - - - - W15 P135 V4 628
I/O P58 P76 J17 P88 Y11 K18 P193 AE14 481 GND P91 P119 GND* P135 GND* GND* P134 GND* -
I/O (INIT) P59 P77 J16 P89 W11 K19 P192 AF14 484 I/O - - R12 P136 P20 U13 P133 V3 631
VCC P60 P78 VCC* P90 VCC* VCC* P191 VCC* - I/O - - R11 P137 N18 V14 P132 W2 634
GND P61 P79 GND* P91 GND* GND* P190 GND* - I/O P92 P120 U13 P138 N19 W14 P131 U4 637
I/O P62 P80 K16 P92 V11 L19 P189 AE13 487 I/O P93 P121 V13 P139 N20 V13 P130 U3 640
I/O P63 P81 K17 P93 U11 L18 P188 AC13 490 VCC - - VCC* P140 VCC* VCC* P129 VCC* -
I/O P64 P82 K18 P94 Y12 L16 P187 AD13 493 I/O (D5) P94 P122 U12 P141 M17 T12 P127 V2 643
I/O P65 P83 L18 P95 W12 L17 P186 AF12 496 I/O (CS0) P95 P123 V12 P142 M18 X14 P126 V1 646
I/O - P84 L17 P96 V12 M20 P185 AE12 499 I/O - - - - - U12 P125 U2 649
I/O - P85 L16 P97 U12 M19 P184 AD12 502 I/O - - - - - W13 P124 T2 652
I/O - - - - Y13 N20 P183 AC12 505 GND - - - P143 GND* GND* - GND* -
I/O - - - - W13 M18 P182 AF11 508 I/O - - - - - X13 P123 T1 655
GND - - - P98 GND* GND* - GND* - I/O - - - - M19 V12 P122 R4 658
I/O - - - - - M17 P181 AE11 511 I/O - P124 T11 P144 M20 W12 P121 R3 661
I/O - - - - - M16 P180 AD11 514 I/O - P125 U11 P145 L19 T11 P120 R2 664
I/O - - L15 P99 V13 N19 P179 AF9 517 I/O P96 P126 V11 P146 L18 X12 P119 R1 667
I/O - - M15 P100 Y14 P20 P178 AD10 520 I/O P97 P127 V10 P147 L20 U11 P118 P3 670
VCC - - VCC* P101 VCC* VCC* P177 VCC* - I/O (D4) P98 P128 U10 P148 K20 V11 P117 P2 673
I/O P66 P86 M18 P102 Y15 N18 P175 AE9 523 I/O P99 P129 T10 P149 K19 W11 P116 P1 676
I/O P67 P87 M17 P103 V14 P19 P174 AD9 526 VCC P100 P130 VCC* P150 VCC* VCC* P115 VCC* -
I/O P68 P88 N18 P104 W15 N17 P173 AC10 529 GND P101 P131 GND* P151 GND* GND* P114 GND* -
I/O P69 P89 P18 P105 Y16 R19 P172 AF7 532 I/O (D3) P102 P132 T9 P152 K18 W10 P113 N2 679
GND P70 P90 GND* P106 GND* GND* P171 GND* - I/O (RS) P103 P133 U9 P153 K17 V10 P112 N4 682
I/O - - - - - N16 P170 AE8 535 I/O P104 P134 V9 P154 J20 T10 P111 N3 685
I/O - - - - - P18 P169 AD8 538 I/O P105 P135 V8 P155 J19 U10 P110 M1 688
I/O - - N15 P107 V15 U20 P168 AC9 541 I/O - P136 U8 P156 J18 X9 P109 M2 691
I/O - - P15 P108 W16 P17 P167 AF6 544 I/O - P137 T8 P157 J17 W9 P108 M3 694
I/O - P91 N17 P109 Y17 T19 P166 AE7 547 I/O - - - - H20 X8 P107 M4 697
January 29, 1999 (Version 1.5) 6-129

R
XC4025E, XC4025E,
Pad Name Pad Name
I/O - - - - - V9 P106 L1 700 I/O - - - - A13 M5 P51 A9 62
GND - - - P158 GND* GND* - GND* - I/O - - - - D12 P1 P50 D11 65
I/O - - - - - U9 P105 L2 703 I/O - - - - - M4 P49 B11 68
I/O - - - - - T9 P104 L3 706 I/O - - - - - N2 P48 A11 71
I/O (D2) P106 P138 V7 P159 H19 W8 P103 J1 709 GND - - - - GND* GND* - GND* -
I/O P107 P139 U7 P160 H18 X7 P102 K3 712 I/O (A4) P134 P174 M2 P202 C12 N1 P47 D12 74
VCC - - VCC* P161 VCC* VCC* P101 VCC* - I/O (A5) P135 P175 M1 P203 B12 M3 P46 C12 77
I/O P108 P140 V6 P162 G19 V8 P99 J2 715 I/O - P176 L3 P205 A12 M2 P45 B12 80
I/O P109 P141 U6 P163 F20 W7 P98 J3 718 I/O P136 P177 L2 P206 B11 L5 P44 A12 83
I/O - - R8 P164 G18 U8 P97 K4 721 I/O (A21) ‡ P137 P178 L1 P207 C11 M1 P43 C13 86
I/O - - R7 P165 F19 W6 P96 G1 724 I/O (A20) ‡ P138 P179 K1 P208 A11 L4 P42 B13 89
GND P110 P142 GND* P166 GND* GND* P95 GND* - I/O (A6) P139 P180 K2 P209 A10 L3 P41 A13 92
I/O - - - - - T8 P94 H2 727 I/O (A7) P140 P181 K3 P210 B10 L2 P40 B14 95
I/O - - - - - V7 P93 H3 730 GND P141 P182 GND* P211 GND* GND* P39 GND* -
I/O - - R6 P167 F18 X4 P92 J4 733 6/19/97
I/O - - R5 P168 E19 U7 P91 F1 736 * Pads labelled GND* or VCC* are internally bonded to Ground or
I/O - P143 V5 P169 D20 W5 P90 G2 739 VCC planes within the associated package. They have no direct
I/O - P144 V4 P170 E18 V6 P89 G3 742
connection to any specific package pin.
I/O P111 P145 U5 P171 D19 T7 P88 F2 745
I/O P112 P146 T6 P172 C20 X3 P87 E2 748 † = E only
GND - - - - GND* GND* - GND* - †† = XL only
VCC - - - - VCC* VCC* - VCC* - ‡ = EX, XL only
I/O (D1) P113 P147 V3 P173 E17 U6 P86 F3 751
I/O (RCLK, P114 P148 V2 P174 D18 V5 P85 G4 754 Additional XC4025E, XC4028EX/XL Package
RDY/BUS
Y) Pins
I/O - - - - - W4 P84 D2 757
HQ208
I/O - - - - - W3 P83 F4 760
Not Connected Pins
I/O P115 P149 U4 P175 C19 T6 P82 E3 763 P1 P52 P102 P107 P157 P207
I/O P116 P150 T5 P176 B20 U5 P81 C2 766 P3 P53 P104 P155 P158 P208
I/O (D0, P117 P151 U3 P177 C18 V4 P80 D3 769 P51 P54 P105 P156 P206
DIN)
5/9/97
I/O, P118 P152 T4 P178 B19 X1 P79 E4 772
SGCK4 †,
GCK6 ‡
(DOUT) PG223
CCLK P119 P153 V1 P179 A20 V3 P78 C3 - VCC Pins
VCC P120 P154 VCC* P180 VCC* VCC* P77 VCC* - D3 D10 D16 J4
O, TDO P121 P159 U2 P181 A19 U4 P76 D4 0 J15 R4 R10 R15
GND P122 P160 GND* P182 GND* GND* P75 GND* - GND Pins
I/O (A0, P123 P161 T3 P183 B18 W2 P74 B3 2 C7 C12 D4 D9
WS) D15 G3 G16 K4
I/O, P124 P162 U1 P184 B17 V2 P73 C4 5 K15 M3 M16 R3
PGCK4 †, R9 R16 T7 T12
GCK7 ‡ 5/9/97
(A1)
I/O P125 P163 P3 P185 C17 R5 P72 D5 8
I/O P126 P164 R2 P186 D16 T4 P71 A3 11
I/O (CS1, P127 P165 T2 P187 A18 U3 P70 D6 14 HQ240
A2) GND Pins
I/O (A3) P128 P166 N3 P188 A17 V1 P69 C6 17 P204 P219
I/O - - - - - R4 P68 B5 20 5/9/97
I/O - - - - - P5 P67 A4 23
Note: These pins may be Not Connected for this device revision,
however for compatability with other devices in this package, these
pins should be tied to GND.
I/O - - P4 P189 C16 U2 P66 C7 26
I/O - - N4 P190 B16 T3 P65 B6 29
I/O P129 P167 P2 P191 A16 U1 P64 A6 32
I/O P130 P168 T1 P192 C15 P4 P63 D8 35
I/O - P169 R1 P193 B15 R3 P62 B7 38
I/O - P170 N2 P194 A15 N5 P61 A7 41
I/O - - - P195 - T2 P60 D9 44
I/O - - - - - R2 P59 C9 47
GND P131 P171 GND* P196 GND* GND* P58 GND* -
I/O P132 P172 P1 P197 B14 N4 P57 B8 50
I/O P133 P173 N1 P198 A14 P3 P56 D10 53
I/O - - M4 P199 C13 P2 P55 C10 56
I/O - - L4 P200 B13 N3 P54 B9 59
VCC - - VCC* P201 VCC* VCC* P52 VCC* -
6-130 January 29, 1999 (Version 1.5)

R
BG256 HQ304
VCC Pins Not Connected Pins
C14 D6 D7 D11 P11 P53 P128 P205 P281
D14 D15 E20 F1 P24 P100 P176 P254 -
F4 F17 G4 G17 5/15/97
K4 L17 P4 P17 Note: In XC4025 (no extension) devices in the HQ304 package,
P19 R2 R4 R17 P101 is a No Connect (N.C.) pin. P101 is Vcc in XC4025E and
U6 U7 U10 U14 XC4028EX/XL devices. Where necessary for compatibility, this pin
U15 V7 W20 - can be left unconnected.
GND Pins
A1 B7 D4 D8
D13 D17 G20 H4
BG352
H17 N3 N4 N17
VCC Pins
U4 U8 U13 U17
A10 A17 B2 B25 D7 D13
W14 - - -
D19 G23 H4 K1 K26 N23
5/9/97
P4 U1 U26 W23 Y4 AC8
AC14 AC20 AE2 AE25 AF10 AF17
GND Pins
PG299
A1 A2 A5 A8 A14 A19
VCC Pins
A22 A25 A26 B1 B26 E1
A2 A6 A11 A16
E26 H1 H26 N1 P26 W1
B20 E1 E5 F20
W26 AB1 AB26 AE1 AE26 AF1
K1 L20 R1 T16
AF2 AF5 AF8 AF13 AF19 AF22
T20 W1 X5 X10
AF25 AF26 - - - -
X15 X19 - -
Not Connected Pins
GND Pins
A18 A24 B4 B10 B23 C1
A5 A10 A15 A19
C5 C8 C11 D1 D16 D25
B1 E16 E20 F1
F23 J26 K2 L4 L23 T3
K20 L1 R20 T1
T4 T24 U25 AB3 AC2 AC6
T5 W20 X2 X6
X11 X16 - -
AC11 AC16 AC21 AC25 AD16 AD21 6
AD26 AE4 AE10 - - -
6/18/97
5/9/97
Pin Locations for XC4036EX/XL

XC4036EX/XL PQ HQ HQ HQ BG PG BG Bndry XC4036EX/XL PQ HQ HQ HQ BG PG BG Bndry

Pad Name 160†† 208†† 240 304 352 411 432 Scan Pad Name 160†† 208†† 240 304 352 411 432 Scan
VCC P142 P183 P212 P38 VCC* VCC* VCC* - I/O (A13) P155 P200 P233 P10 C21 G3 B26 185
I/O (A8) P143 P184 P213 P37 D14 W3 D17 110 GND - - - - GND* GND* GND* -
I/O (A9) P144 P185 P214 P36 C14 Y2 A17 113 VCC - - - - VCC* VCC* VCC* -
I/O (A19) P145 P186 P215 P35 A15 V4 C18 116 I/O - - - P9 D20 K6 A27 188
I/O (A18) P146 P187 P216 P34 B15 T2 D18 119 I/O - - - P8 A23 G1 D25 191
I/O - P188 P217 P33 C15 U1 B18 122 I/O - - - - A24 E1 C26 194
I/O - P189 P218 P32 D15 V6 A19 125 I/O - - - - B23 E3 B27 197
I/O (A10) P147 P190 P220 P31 A16 U3 B19 128 I/O - - P234 P7 D21 J7 C27 200
I/O (A11) P148 P191 P221 P30 B16 R1 C19 131 I/O - - P235 P6 C22 H6 B28 203
VCC - - - - VCC* VCC* VCC* - I/O P156 P201 P236 P5 B24 C3 D27 206
GND - - - - GND* GND* GND* - I/O P157 P202 P237 P4 C23 D2 B29 209
I/O - - - P29 C16 U5 D19 134 I/O (A14) P158 P203 P238 P3 D22 E5 C28 212
I/O - - - P28 B17 T4 A20 137 I/O, GCK8 (A15) P159 P204 P239 P2 C24 G7 D28 215
I/O - - - - D16 P2 B20 140 VCC P160 P205 P240 P1 VCC* VCC* VCC* -
I/O - - - - A18 N1 C20 143 GND P1 P2 P1 P304 GND* GND* GND* -
I/O - - - P27 C17 R5 C21 146 I/O, GCK1 (A16) P2 P4 P2 P303 D23 H8 D29 218
I/O - - - P26 B18 M2 A22 149 I/O (A17) P3 P5 P3 P302 C25 F6 C30 221
VCC - - P222 P25 VCC* VCC* VCC* - I/O P4 P6 P4 P301 D24 B4 E28 224
I/O - - P223 P23 C18 L3 B22 152 I/O P5 P7 P5 P300 E23 D4 E29 227
I/O - - P224 P22 D17 T6 C22 155 I/O, TDI P6 P8 P6 P299 C26 B2 D30 230
I/O P149 P192 P225 P21 A20 N5 B23 158 I/O, TCK P7 P9 P7 P298 E24 G9 D31 233
I/O P150 P193 P226 P20 B19 M4 A24 161 I/O - - - - D25 F8 E30 236
GND P151 P194 P227 P19 GND* GND* GND* - I/O - - - - F23 C5 E31 239
I/O - - - P18 C19 K2 D22 164 I/O - - - P297 F24 A7 G28 242
I/O - - - P17 D18 K4 C23 167 I/O - - - P296 E25 A5 G29 245
I/O - P195 P228 P16 A21 P6 B24 170 VCC - - - - VCC* VCC* VCC* -
I/O - P196 P229 P15 B20 M6 C24 173 GND - - - - GND* GND* GND* -
I/O P152 P197 P230 P14 C20 J3 A26 176 I/O P8 P10 P8 P295 D26 B8 H28 248
I/O P153 P198 P231 P13 B21 H2 C25 179 I/O P9 P11 P9 P294 G24 C9 H29 251
I/O (A12) P154 P199 P232 P12 B22 H4 D24 182 I/O - P12 P10 P293 F25 E9 G30 254
January 29, 1999 (Version 1.5) 6-131

R

I/O - P13 P11 P292 F26 F12 H30 257 I (M2) P42 P56 P62 P227 AC23 G33 AJ28 438
I/O - - P12 P291 H23 D10 J28 260 I/O, GCK3 P43 P57 P63 P226 AE24 D36 AK29 439
I/O - - P13 P290 H24 B10 J29 263 I/O (HDC) P44 P58 P64 P225 AD23 C37 AH27 442
I/O - - - P289 G25 F10 H31 266 I/O P45 P59 P65 P224 AC22 F34 AK28 445
I/O - - - P288 G26 F14 J30 269 I/O P46 P60 P66 P223 AF24 J33 AJ27 448
GND P10 P14 P14 P287 GND* GND* GND* - I/O P47 P61 P67 P222 AD22 D38 AL28 451
I/O P11 P15 P15 P286 J23 C11 K28 272 I/O (LDC) P48 P62 P68 P221 AE23 G35 AH26 454
I/O P12 P16 P16 P285 J24 B12 K29 275 I/O - - - - AC21 E39 AL27 457
I/O, TMS P13 P17 P17 P284 H25 E11 K30 278 I/O - - - - AD21 K34 AH25 460
I/O P14 P18 P18 P283 K23 E15 K31 281 I/O - - - P220 AE22 F38 AK26 463
VCC - - P19 P282 VCC* VCC* VCC* - I/O - - - P219 AF23 G37 AL26 466
I/O - - P20 P280 K24 F16 L29 284 VCC - - - - VCC* VCC* VCC* -
I/O - - P21 P279 J25 C13 L30 287 GND - - - - GND* GND* GND* -
I/O - - - - J26 B14 M29 290 I/O P49 P63 P69 P218 AD20 H38 AH24 469
I/O - - - - L23 E17 M31 293 I/O P50 P64 P70 P217 AE21 J37 AJ25 472
I/O - - - P278 L24 E13 N31 296 I/O - P65 P71 P216 AF21 G39 AK25 475
I/O - - - P277 K25 A15 N28 299 I/O - P66 P72 P215 AC19 M34 AJ24 478
GND - - P22 - GND* GND* GND* - I/O - - P73 P214 AD19 N35 AL24 481
VCC - - - - VCC* VCC* VCC* - I/O - - P74 P213 AE20 P34 AH22 484
I/O - - - P276 L25 B16 P30 302 I/O - - - P212 AF20 J35 AJ23 487
I/O - - - P275 L26 D16 P28 305 I/O - - - P211 AC18 L37 AK23 490
I/O - P19 P23 P274 M23 D18 P29 308 GND P51 P67 P75 P210 GND* GND* GND* -
I/O - P20 P24 P273 M24 A17 R31 311 I/O P52 P68 P76 P209 AD18 M38 AJ22 493
I/O P15 P21 P25 P272 M25 E19 R30 314 I/O P53 P69 P77 P208 AE19 R35 AK22 496
I/O P16 P22 P26 P271 M26 B18 R28 317 I/O P54 P70 P78 P207 AC17 H36 AL22 499
I/O P17 P23 P27 P270 N24 C17 R29 320 I/O P55 P71 P79 P206 AD17 T34 AJ21 502
I/O P18 P24 P28 P269 N25 C19 T31 323 VCC - - P80 P204 VCC* VCC* VCC* -
GND P19 P25 P29 P268 GND* GND* GND* - I/O - P72 P81 P203 AE18 N37 AH20 505
VCC P20 P26 P30 P267 VCC* VCC* VCC* - I/O - P73 P82 P202 AF18 N39 AK21 508
I/O P21 P27 P31 P266 N26 F20 T30 326 I/O - - - - AC16 U35 AK20 511
I/O P22 P28 P32 P265 P25 B20 T29 329 I/O - - - - AD16 R39 AJ19 514
I/O P23 P29 P33 P264 P23 C21 U31 332 I/O - - - P201 AE17 M36 AL20 517
I/O P24 P30 P34 P263 P24 B22 U30 335 I/O - - - P200 AE16 V34 AH18 520
I/O - P31 P35 P262 R26 E21 U28 338 GND - - P83 - GND* GND* GND* -
I/O - P32 P36 P261 R25 D22 U29 341 VCC - - - - VCC* VCC* VCC* -
I/O - - - P260 R24 A23 V30 344 I/O - - - P199 AF16 R37 AK19 523
I/O - - - P259 R23 B24 V29 347 I/O - - - P198 AC15 T38 AJ18 526
VCC - - - - VCC* VCC* VCC* - I/O - - P84 P197 AD15 T36 AL19 529
GND - - P37 - GND* GND* GND* - I/O - - P85 P196 AE15 V36 AK18 532
I/O - - - P258 T26 A25 W30 350 I/O P56 P74 P86 P195 AF15 U37 AH17 535
I/O - - - P257 T25 D24 W29 353 I/O P57 P75 P87 P194 AD14 U39 AJ17 538
I/O - - - - T24 B26 Y30 356 I/O P58 P76 P88 P193 AE14 V38 AJ16 541
I/O - - - - U25 A27 Y29 359 I/O (INIT) P59 P77 P89 P192 AF14 W37 AK16 544
I/O - - P38 P256 T23 C27 Y28 362 VCC P60 P78 P90 P191 VCC* VCC* VCC* -
I/O - - P39 P255 V26 F24 AA30 365 GND P61 P79 P91 P190 GND* GND* GND* -
VCC - - P40 P253 VCC* VCC* VCC* - I/O P62 P80 P92 P189 AE13 Y34 AL16 547
I/O P25 P33 P41 P252 U24 E25 AA29 368 I/O P63 P81 P93 P188 AC13 AC37 AH15 550
I/O P26 P34 P42 P251 V25 E27 AB31 371 I/O P64 P82 P94 P187 AD13 AB38 AK15 553
I/O P27 P35 P43 P250 V24 B28 AB30 374 I/O P65 P83 P95 P186 AF12 AD36 AJ14 556
I/O P28 P36 P44 P249 U23 C29 AB29 377 I/O - P84 P96 P185 AE12 AA35 AH14 559
GND P29 P37 P45 P248 GND* GND* GND* - I/O - P85 P97 P184 AD12 AE37 AK14 562
I/O - - - P247 Y26 F26 AB28 380 I/O - - - P183 AC12 AB36 AL13 565
I/O - - - P246 W25 D28 AC30 383 I/O - - - P182 AF11 AD38 AK13 568
I/O - - P46 P245 W24 B30 AC29 386 VCC - - - - VCC* VCC* VCC* -
I/O - - P47 P244 V23 E29 AC28 389 GND - - P98 - GND* GND* GND* -
I/O - P38 P48 P243 AA26 F28 AD29 392 I/O - - - P181 AE11 AB34 AJ13 571
I/O - P39 P49 P242 Y25 F30 AD28 395 I/O - - - P180 AD11 AE39 AH13 574
I/O P30 P40 P50 P241 Y24 C31 AE30 398 I/O - - - - AE10 AM36 AL12 577
I/O P31 P41 P51 P240 AA25 E31 AE29 401 I/O - - - - AC11 AC35 AK12 580
GND - - - - GND* GND* GND* - I/O - - P99 P179 AF9 AG39 AH12 583
VCC - - - - VCC* VCC* VCC* - I/O - - P100 P178 AD10 AG37 AJ11 586
I/O - - - P239 AB25 B32 AF31 404 VCC - - P101 P177 VCC* VCC* VCC* -
I/O - - - P238 AA24 A33 AE28 407 I/O P66 P86 P102 P175 AE9 AD34 AL10 589
I/O P32 P42 P52 P237 Y23 A35 AG31 410 I/O P67 P87 P103 P174 AD9 AN39 AK10 592
I/O P33 P43 P53 P236 AC26 F32 AF28 413 I/O P68 P88 P104 P173 AC10 AE35 AJ10 595
I/O - - - - AD26 C35 AG30 416 I/O P69 P89 P105 P172 AF7 AH38 AK9 598
I/O - - - - AC25 B38 AG29 419 GND P70 P90 P106 P171 GND* GND* GND* -
I/O P34 P44 P54 P235 AA23 E33 AH31 422 I/O - - - P170 AE8 AJ37 AL8 601
I/O P35 P45 P55 P234 AB24 G31 AG28 425 I/O - - - P169 AD8 AG35 AH10 604
I/O P36 P46 P56 P233 AD25 H32 AH30 428 I/O - - P107 P168 AC9 AF34 AJ9 607
I/O, GCK2 P37 P47 P57 P232 AC24 B36 AJ30 431 I/O - - P108 P167 AF6 AH36 AK8 610
O (M1) P38 P48 P58 P231 AB23 A39 AH29 434 I/O - P91 P109 P166 AE7 AK36 AK7 613
GND P39 P49 P59 P230 GND* GND* GND* - I/O - P92 P110 P165 AD7 AM34 AL6 616
I (M0) P40 P50 P60 P229 AD24 E35 AH28 437 I/O P71 P93 P111 P164 AE6 AH34 AJ7 619
VCC P41 P55 P61 P228 VCC* VCC* VCC* - I/O P72 P94 P112 P163 AE5 AJ35 AH8 622
6-132 January 29, 1999 (Version 1.5)

R

GND - - - - GND* GND* GND* - I/O P107 P139 P160 P102 K3 AP16 L3 802
VCC - - - - VCC* VCC* VCC* - VCC - - P161 P101 VCC* VCC* VCC* -
I/O - - - P162 AD6 AL37 AK6 625 I/O P108 P140 P162 P99 J2 AV12 K1 805
I/O - - - P161 AC7 AT38 AL5 628 I/O P109 P141 P163 P98 J3 AR13 K2 808
I/O P73 P95 P113 P160 AF4 AM38 AH7 631 I/O - - P164 P97 K4 AU11 K3 811
I/O P74 P96 P114 P159 AF3 AN37 AJ6 634 I/O - - P165 P96 G1 AT12 K4 814
I/O - - - - AE4 AK34 AK5 637 GND P110 P142 P166 P95 GND* GND* GND* -
I/O - - - - AC6 AR39 AL4 640 I/O - - - P94 H2 AP14 J2 817
I/O P75 P97 P115 P158 AD5 AN35 AK4 643 I/O - - - P93 H3 AR11 J3 820
I/O P76 P98 P116 P157 AE3 AL33 AH5 646 I/O - - P167 P92 J4 AV10 J4 823
I/O P77 P99 P117 P156 AD4 AV38 AK3 649 I/O - - P168 P91 F1 AT8 H1 826
I/O, GCK4 P78 P100 P118 P155 AC5 AT36 AJ4 652 I/O - P143 P169 P90 G2 AT10 H2 829
GND P79 P101 P119 P154 GND* GND* GND* - I/O - P144 P170 P89 G3 AP10 H3 832
DONE P80 P103 P120 P153 AD3 AR35 AH4 - I/O P111 P145 P171 P88 F2 AP12 H4 835
VCC P81 P106 P121 P152 VCC* VCC* VCC* - I/O P112 P146 P172 P87 E2 AR9 G2 838
PROGRAM P82 P108 P122 P151 AC4 AN33 AH3 - GND - - - - GND* GND* GND* -
I/O (D7) P83 P109 P123 P150 AD2 AM32 AJ2 655 VCC - - - - VCC* VCC* VCC* -
I/O, GCK5 P84 P110 P124 P149 AC3 AP34 AG4 658 I/O (D1) P113 P147 P173 P86 F3 AU7 G4 841
I/O P85 P111 P125 P148 AB4 AW39 AG3 661 I/O (RCLK, P114 P148 P174 P85 G4 AW7 F2 844
I/O P86 P112 P126 P147 AD1 AN31 AH2 664 RDY/BUSY)
I/O - - - - AB3 AV36 AH1 667 I/O - - - - D1 AW5 F3 847
I/O - - - - AC2 AR33 AF4 670 I/O - - - - C1 AV6 E1 850
I/O - - P127 P146 AA4 AP32 AF3 673 I/O - - - P84 D2 AR7 E3 853
I/O - - P128 P145 AA3 AU35 AG2 676 I/O - - - P83 F4 AV4 D1 856
I/O - - - P144 AB2 AW33 AE3 679 I/O P115 P149 P175 P82 E3 AN9 E4 859
I/O - - - P143 AC1 AU33 AF2 682 I/O P116 P150 P176 P81 C2 AW1 D2 862
VCC - - - - VCC* VCC* VCC* - I/O (D0, DIN) P117 P151 P177 P80 D3 AP6 C2 865
GND - - - - GND* GND* GND* - I/O, GCK6 P118 P152 P178 P79 E4 AU3 D3 868
I/O (D6) P87 P113 P129 P142 Y3 AV32 AF1 685 (DOUT)
I/O P88 P114 P130 P141 AA2 AU31 AD4 688 CCLK P119 P153 P179 P78 C3 AR5 D4 -
I/O P89 P115 P131 P140 AA1 AR31 AD3 691 VCC P120 P154 P180 P77 VCC* VCC* VCC* - 6
I/O P90 P116 P132 P139 W4 AP28 AE2 694 O, TDO P121 P159 P181 P76 D4 AN7 C4 0
I/O - P117 P133 P138 W3 AT32 AC3 697 GND P122 P160 P182 P75 GND* GND* GND* -
I/O - P118 P134 P137 Y2 AV30 AD1 700 I/O (A0, WS) P123 P161 P183 P74 B3 AT4 B3 2
I/O - - - P136 Y1 AR29 AC2 703 I/O, GCK7 (A1) P124 P162 P184 P73 C4 AV2 D5 5
I/O - - - P135 V4 AP26 AB4 706 I/O P125 P163 P185 P72 D5 AM8 B4 8
GND P91 P119 P135 P134 GND* GND* GND* - I/O P126 P164 P186 P71 A3 AL7 C5 11
I/O - - P136 P133 V3 AU29 AB3 709 I/O - - - - C5 AR3 B5 14
I/O - - P137 P132 W2 AV28 AB2 712 I/O - - - - B4 AR1 C6 17
I/O P92 P120 P138 P131 U4 AT28 AB1 715 I/O (CS1, A2) P127 P165 P187 P70 D6 AK6 A5 20
I/O P93 P121 P139 P130 U3 AR25 AA3 718 I/O (A3) P128 P166 P188 P69 C6 AN3 D7 23
VCC - - P140 P129 VCC* VCC* VCC* - I/O - - - P68 B5 AM6 B6 26
I/O (D5) P94 P122 P141 P127 V2 AP24 AA2 721 I/O - - - P67 A4 AM2 A6 29
I/O (CS0) P95 P123 P142 P126 V1 AU27 Y2 724 VCC - - - - VCC* VCC* VCC* -
I/O - - - - T4 AR27 Y4 727 GND - - - - GND* GND* GND* -
I/O - - - - T3 AW27 Y3 730 I/O - - P189 P66 C7 AL3 D8 32
I/O - - - P125 U2 AT24 W4 733 I/O - - P190 P65 B6 AH6 C7 35
I/O - - - P124 T2 AR23 W3 736 I/O P129 P167 P191 P64 A6 AP2 B7 38
GND - - P143 - GND* GND* GND* - I/O P130 P168 P192 P63 D8 AK4 D9 41
VCC - - - - VCC* VCC* VCC* - I/O - P169 P193 P62 B7 AG5 D10 44
I/O - - - P123 T1 AP22 V4 739 I/O - P170 P194 P61 A7 AF6 C9 47
I/O - - - P122 R4 AV24 V3 742 I/O - - P195 P60 D9 AL5 B9 50
I/O - P124 P144 P121 R3 AU23 U1 745 I/O - - - P59 C9 AJ3 C10 53
I/O - P125 P145 P120 R2 AT22 U2 748 GND P131 P171 P196 P58 GND* GND* GND* -
I/O P96 P126 P146 P119 R1 AR21 U4 751 I/O P132 P172 P197 P57 B8 AH2 B10 56
I/O P97 P127 P147 P118 P3 AV22 U3 754 I/O P133 P173 P198 P56 D10 AE5 A10 59
I/O (D4) P98 P128 P148 P117 P2 AP20 T1 757 I/O - - P199 P55 C10 AM4 C11 62
I/O P99 P129 P149 P116 P1 AU21 T2 760 I/O - - P200 P54 B9 AD6 D12 65
VCC P100 P130 P150 P115 VCC* VCC* VCC* - VCC - - P201 P52 VCC* VCC* VCC* -
GND P101 P131 P151 P114 GND* GND* GND* - I/O - - - P51 A9 AG3 B11 68
I/O (D3) P102 P132 P152 P113 N2 AU19 T3 763 I/O - - - P50 D11 AG1 C12 71
I/O (RS) P103 P133 P153 P112 N4 AV20 R1 766 I/O - - - - C11 AC5 C13 74
I/O P104 P134 P154 P111 N3 AV18 R2 769 I/O - - - - B10 AE1 A12 77
I/O P105 P135 P155 P110 M1 AR19 R4 772 I/O - - - P49 B11 AH4 D14 80
I/O - P136 P156 P109 M2 AT18 R3 775 I/O - - - P48 A11 AB6 B13 83
I/O - P137 P157 P108 M3 AW17 P2 778 GND - - - - GND* GND* GND* -
I/O - - - P107 M4 AV16 P3 781 VCC - - - - VCC* VCC* VCC* -
I/O - - - P106 L1 AP18 P4 784 I/O (A4) P134 P174 P202 P47 D12 AD2 C14 86
VCC - - - - VCC* VCC* VCC* - I/O (A5) P135 P175 P203 P46 C12 AB4 A13 89
GND - - P158 - GND* GND* GND* - I/O - P176 P205 P45 B12 AE3 B14 92
I/O - - - P105 L2 AR17 N3 787 I/O P136 P177 P206 P44 A12 AC1 D15 95
I/O - - - P104 L3 AT16 N4 790 I/O (A21) P137 P178 P207 P43 C13 AD4 C15 98
I/O - - - - K2 AV14 M1 793 I/O (A20) P138 P179 P208 P42 B13 AA5 B15 101
I/O - - - - L4 AW13 M2 796 I/O (A6) P139 P180 P209 P41 A13 AA3 B16 104
I/O (D2) P106 P138 P159 P103 J1 AR15 L2 799 I/O (A7) P140 P181 P210 P40 B14 Y6 A16 107
January 29, 1999 (Version 1.5) 6-133

R
XC4036EX/XL PQ HQ HQ HQ BG PG BG Bndry PG411

Pad Name 160†† 208†† 240 304 352 411 432 Scan VCC Pins
GND P141 P182 P211 P39 GND* GND* GND* - A3 A11 A21 A31 C39 D6
6/17/97 F36 J1 L39 W1 AA39 AJ1
* Pads labelled GND* or VCC* are internally bonded to Ground or VCC AL39 AP4 AT34 AU1 AW9 AW19
planes within the associated package. They have no direct connection to AW29 AW37 - - - -
any specific package pin. GND Pins
A9 A19 A29 A37 C1 D14
†† = XL only D20 D26 D34 F4 J39 L1
P4 P36 W39 Y4 Y36 AA1
AF4 AF36 AJ39 AL1 AP36 AT6
Additional XC4036EX/XL Package Pins AT14 AT20 AT26 AU39 AW3 AW11
AW21 AW31 - - - -
HQ208 Not Connected Pins
Not Connected Pins A13 B6 B34 C7 C15 C23
P1 P3 P51 P52 P53 C25 C33 D8 D12 D30 D32
P54 P102 P104 P105 P107 E7 E23 E37 F2 F18 F22
P155 P156 P157 P158 P206 G5 H34 J5 K36 K38 L5
P207 P208 - - - L35 N3 P38 R3 V2 W5
5/15/97 W35 Y38 AA37 AB2 AC3 AC39
AF2 AF38 AJ5 AK2 AK38 AL35
AN1 AN5 AP8 AP30 AP38 AR37
AT2 AT30 AU5 AU9 AU13 AU15
HQ240
AU17 AU25 AU37 AV8 AV26 AV34
GND Pins
AW15 AW23 AW25 AW35 - -
P204 P219 - - -
6/16/97
6/17/97
The Ground (GND) package pins in the above table should be

externally connected to Ground if possible; however, they can be BG432
left unconnected if necessary for compatibility with other devices. VCC Pins
A1 A11 A21 A31 C3 C29
D11 D21 L1 L4 L28 L31
AA1 AA4 AA28 AA31 AH11 AH21
HQ304
AJ3 AJ29 AL1 AL11 AL21 AL31
Not Connected Pins
GND Pins
P11 P24 P53 P100 P128
A2 A3 A7 A9 A14 A18
P176 P205 P254 P281 -
A23 A25 A29 A30 B1 B2
5/15/97 B30 B31 C1 C31 D16 G1
G31 J1 J31 P1 P31 T4
T28 V1 V31 AC1 AC31 AE1
BG352 AE31 AH16 AJ1 AJ31 AK1 AK2
VCC Pins AK30 AK31 AL2 AL3 AL7 AL9
A10 A17 B2 B25 D7 D13 AL14 AL18 AL23 AL25 AL29 AL30
D19 G23 H4 K1 K26 N23 Not Connected Pins
P4 U1 U26 W23 Y4 AC8 A4 A8 A15 A28 B8 B12
AC14 AC20 AE2 AE25 AF10 AF17 B17 B21 B25 C8 C16 C17
GND Pins D6 D13 D20 D23 D26 E2
A1 A2 A5 A8 A14 A19 F1 F4 F28 F29 F30 F31
A22 A25 A26 B1 B26 E1 G3 M3 M4 M28 M30 N1
E26 H1 H26 N1 P26 W1 N2 N29 N30 V2 V28 W1
W26 AB1 AB26 AE1 AE26 AF1 W2 W28 W31 Y1 Y31 AC4
AF2 AF5 AF8 AF13 AF19 AF22 AD2 AD30 AD31 AE4 AF29 AF30
AF25 AF26 - - - - AG1 AH6 AH9 AH19 AH23 AJ5
Not Connected Pins AJ8 AJ12 AJ15 AJ20 AJ26 AK11
C8 - - - - AK17 AK24 AK27 AL15 AL17 -
6/16/97 5/15/97
Pin Locations for XC4044XL Devices

(Note: XC4044XL is also available in the HQ304 package. XC4044XL HQ HQ HQ BG PG BG
Pad Name 160 208 240 352 411 432
The pinout is identical to the XC4036XL in the HQ304. ) I/O (A18) P146 P187 P216 B15 T2 D18
I/O - P188 P217 C15 U1 B18
XC4044XL HQ HQ HQ BG PG BG I/O - P189 P218 D15 V6 A19
Pad Name 160 208 240 352 411 432
I/O (A10) P147 P190 P220 A16 U3 B19
VCC P142 P183 P212 VCC* VCC* VCC*
I/O (A11) P148 P191 P221 B16 R1 C19
I/O (A8) P143 P184 P213 D14 W3 D17
VCC - - - VCC* VCC* VCC*
I/O (A9) P144 P185 P214 C14 Y2 A17
GND - - - GND* GND* GND*
I/O - - - - V2 C17
I/O - - - C16 U5 D19
I/O - - - - W5 B17
I/O - - - B17 T4 A20
I/O (A19) P145 P186 P215 A15 V4 C18
6-134 January 29, 1999 (Version 1.5)

R
XC4044XL HQ HQ HQ BG PG BG XC4044XL HQ HQ HQ BG PG BG
Pad Name 160 208 240 352 411 432 Pad Name 160 208 240 352 411 432
I/O - - - D16 P2 B20 I/O - P20 P24 M24 A17 R31
I/O - - - A18 N1 C20 I/O P15 P21 P25 M25 E19 R30
I/O - - - C17 R5 C21 I/O P16 P22 P26 M26 B18 R28
I/O - - - B18 M2 A22 I/O P17 P23 P27 N24 C17 R29
VCC - - P222 VCC* VCC* VCC* I/O P18 P24 P28 N25 C19 T31
I/O - - P223 C18 L3 B22 GND P19 P25 P29 GND* GND* GND*
I/O - - P224 D17 T6 C22 VCC P20 P26 P30 VCC* VCC* VCC*
I/O P149 P192 P225 A20 N5 B23 I/O P21 P27 P31 N26 F20 T30
I/O P150 P193 P226 B19 M4 A24 I/O P22 P28 P32 P25 B20 T29
GND P151 P194 P227 GND* GND* GND* I/O P23 P29 P33 P23 C21 U31
I/O - - - C19 K2 D22 I/O P24 P30 P34 P24 B22 U30
I/O - - - D18 K4 C23 I/O - P31 P35 R26 E21 U28
I/O - P195 P228 A21 P6 B24 I/O - P32 P36 R25 D22 U29
I/O - P196 P229 B20 M6 C24 I/O - - - R24 A23 V30
I/O - - - - L5 D23 I/O - - - R23 B24 V29
I/O - - - - J5 B25 I/O - - - - C23 V28
I/O P152 P197 P230 C20 J3 A26 I/O - - - - F22 W31
I/O P153 P198 P231 B21 H2 C25 VCC - - - VCC* VCC* VCC*
I/O (A12) P154 P199 P232 B22 H4 D24 GND - - P37 GND* GND* GND*
I/O (A13) P155 P200 P233 C21 G3 B26 I/O - - - T26 A25 W30
GND - - - GND* GND* GND* I/O - - - T25 D24 W29
VCC - - - VCC* VCC* VCC* I/O - - - T24 B26 Y30
I/O - - - D20 K6 A27 I/O - - - U25 A27 Y29
I/O - - - A23 G1 D25 I/O - - P38 T23 C27 Y28
I/O - - - A24 E1 C26 I/O - - P39 V26 F24 AA30
I/O - - - B23 E3 B27 VCC - - P40 VCC* VCC* VCC*
I/O - - P234 D21 J7 C27 I/O P25 P33 P41 U24 E25 AA29
I/O - - P235 C22 H6 B28 I/O P26 P34 P42 V25 E27 AB31
I/O P156 P201 P236 B24 C3 D27 I/O P27 P35 P43 V24 B28 AB30
I/O P157 P202 P237 C23 D2 B29 I/O P28 P36 P44 U23 C29 AB29
I/O (A14) P158 P203 P238 D22 E5 C28 GND P29 P37 P45 GND* GND* GND* 6
I/O, GCK8 (A15) P159 P204 P239 C24 G7 D28 I/O - - - Y26 F26 AB28
VCC P160 P205 P240 VCC* VCC* VCC* I/O - - - W25 D28 AC30
GND P1 P2 P1 GND* GND* GND* I/O - - P46 W24 B30 AC29
I/O, GCK1 (A16) P2 P4 P2 D23 H8 D29 I/O - - P47 V23 E29 AC28
I/O (A17) P3 P5 P3 C25 F6 C30 I/O - - - - D30 AD31
I/O P4 P6 P4 D24 B4 E28 I/O - - - - D32 AD30
I/O P5 P7 P5 E23 D4 E29 I/O - P38 P48 AA26 F28 AD29
I/O, TDI P6 P8 P6 C26 B2 D30 I/O - P39 P49 Y25 F30 AD28
I/O, TCK P7 P9 P7 E24 G9 D31 I/O P30 P40 P50 Y24 C31 AE30
I/O - - - D25 F8 E30 I/O P31 P41 P51 AA25 E31 AE29
I/O - - - F23 C5 E31 GND - - - GND* GND* GND*
I/O - - - F24 A7 G28 VCC - - - VCC* VCC* VCC*
I/O - - - E25 A5 G29 I/O - - - AB25 B32 AF31
VCC - - - VCC* VCC* VCC* I/O - - - AA24 A33 AE28
GND - - - GND* GND* GND* I/O P32 P42 P52 Y23 A35 AG31
I/O - - - - C7 F30 I/O P33 P43 P53 AC26 F32 AF28
I/O - - - - D8 F31 I/O - - - AD26 C35 AG30
I/O P8 P10 P8 D26 B8 H28 I/O - - - AC25 B38 AG29
I/O P9 P11 P9 G24 C9 H29 I/O P34 P44 P54 AA23 E33 AH31
I/O - P12 P10 F25 E9 G30 I/O P35 P45 P55 AB24 G31 AG28
I/O - P13 P11 F26 F12 H30 I/O P36 P46 P56 AD25 H32 AH30
I/O - - P12 H23 D10 J28 I/O, GCK2 P37 P47 P57 AC24 B36 AJ30
I/O - - P13 H24 B10 J29 O (M1) P38 P48 P58 AB23 A39 AH29
I/O - - - G25 F10 H31 GND P39 P49 P59 GND* GND* GND*
I/O - - - G26 F14 J30 I (M0) P40 P50 P60 AD24 E35 AH28
GND P10 P14 P14 GND* GND* GND* VCC P41 P55 P61 VCC* VCC* VCC*
I/O P11 P15 P15 J23 C11 K28 I (M2) P42 P56 P62 AC23 G33 AJ28
I/O P12 P16 P16 J24 B12 K29 I/O, GCK3 P43 P57 P63 AE24 D36 AK29
I/O, TMS P13 P17 P17 H25 E11 K30 I/O (HDC) P44 P58 P64 AD23 C37 AH27
I/O P14 P18 P18 K23 E15 K31 I/O P45 P59 P65 AC22 F34 AK28
VCC - - P19 VCC* VCC* VCC* I/O P46 P60 P66 AF24 J33 AJ27
I/O - - P20 K24 F16 L29 I/O P47 P61 P67 AD22 D38 AL28
I/O - - P21 J25 C13 L30 I/O (LDC) P48 P62 P68 AE23 G35 AH26
I/O - - - J26 B14 M29 I/O - - - AC21 E39 AL27
I/O - - - L23 E17 M31 I/O - - - AD21 K34 AH25
I/O - - - L24 E13 N31 I/O - - - AE22 F38 AK26
I/O - - - K25 A15 N28 I/O - - - AF23 G37 AL26
GND - - P22 GND* GND* GND* VCC - - - VCC* VCC* VCC*
VCC - - - VCC* VCC* VCC* GND - - - GND* GND* GND*
I/O - - - - F18 N29 I/O P49 P63 P69 AD20 H38 AH24
I/O - - - - C15 N30 I/O P50 P64 P70 AE21 J37 AJ25
I/O - - - L25 B16 P30 I/O - P65 P71 AF21 G39 AK25
I/O - - - L26 D16 P28 I/O - P66 P72 AC19 M34 AJ24
I/O - P19 P23 M23 D18 P29 I/O - - - - K36 AH23
January 29, 1999 (Version 1.5) 6-135

R
Pad Name 160 208 240 352 411 432 Pad Name 160 208 240 352 411 432
I/O - - - - K38 AK24 I/O P77 P99 P117 AD4 AV38 AK3
I/O - - P73 AD19 N35 AL24 I/O, GCK4 P78 P100 P118 AC5 AT36 AJ4
I/O - - P74 AE20 P34 AH22 GND P79 P101 P119 GND* GND* GND*
I/O - - - AF20 J35 AJ23 DONE P80 P103 P120 AD3 AR35 AH4
I/O - - - AC18 L37 AK23 VCC P81 P106 P121 VCC* VCC* VCC*
GND P51 P67 P75 GND* GND* GND* PROGRAM P82 P108 P122 AC4 AN33 AH3
I/O P52 P68 P76 AD18 M38 AJ22 I/O (D7) P83 P109 P123 AD2 AM32 AJ2
I/O P53 P69 P77 AE19 R35 AK22 I/O, GCK5 P84 P110 P124 AC3 AP34 AG4
I/O P54 P70 P78 AC17 H36 AL22 I/O P85 P111 P125 AB4 AW39 AG3
I/O P55 P71 P79 AD17 T34 AJ21 I/O P86 P112 P126 AD1 AN31 AH2
VCC - - P80 VCC* VCC* VCC* I/O - - - AB3 AV36 AH1
I/O - P72 P81 AE18 N37 AH20 I/O - - - AC2 AR33 AF4
I/O - P73 P82 AF18 N39 AK21 I/O - - P127 AA4 AP32 AF3
I/O - - - AC16 U35 AK20 I/O - - P128 AA3 AU35 AG2
I/O - - - AD16 R39 AJ19 I/O - - - AB2 AW33 AE3
I/O - - - AE17 M36 AL20 I/O - - - AC1 AU33 AF2
I/O - - - AE16 V34 AH18 VCC - - - VCC* VCC* VCC*
GND - - P83 GND* GND* GND* GND - - - GND* GND* GND*
VCC - - - VCC* VCC* VCC* I/O (D6) P87 P113 P129 Y3 AV32 AF1
I/O - - - AF16 R37 AK19 I/O P88 P114 P130 AA2 AU31 AD4
I/O - - - AC15 T38 AJ18 I/O P89 P115 P131 AA1 AR31 AD3
I/O - - P84 AD15 T36 AL19 I/O P90 P116 P132 W4 AP28 AE2
I/O - - P85 AE15 V36 AK18 I/O - - - - AP30 AD2
I/O P56 P74 P86 AF15 U37 AH17 I/O - - - - AT30 AC4
I/O P57 P75 P87 AD14 U39 AJ17 I/O - P117 P133 W3 AT32 AC3
I/O - - - - W35 AK17 I/O - P118 P134 Y2 AV30 AD1
I/O - - - - AC39 AL17 I/O - - - Y1 AR29 AC2
I/O P58 P76 P88 AE14 V38 AJ16 I/O - - - V4 AP26 AB4
I/O (INIT) P59 P77 P89 AF14 W37 AK16 GND P91 P119 P135 GND* GND* GND*
VCC P60 P78 P90 VCC* VCC* VCC* I/O - - P136 V3 AU29 AB3
GND P61 P79 P91 GND* GND* GND* I/O - - P137 W2 AV28 AB2
I/O P62 P80 P92 AE13 Y34 AL16 I/O P92 P120 P138 U4 AT28 AB1
I/O P63 P81 P93 AC13 AC37 AH15 I/O P93 P121 P139 U3 AR25 AA3
I/O - - - - Y38 AL15 VCC - - P140 VCC* VCC* VCC*
I/O - - - - AA37 AJ15 I/O (D5) P94 P122 P141 V2 AP24 AA2
I/O P64 P82 P94 AD13 AB38 AK15 I/O (CS0) P95 P123 P142 V1 AU27 Y2
I/O P65 P83 P95 AF12 AD36 AJ14 I/O - - - T4 AR27 Y4
I/O - P84 P96 AE12 AA35 AH14 I/O - - - T3 AW27 Y3
I/O - P85 P97 AD12 AE37 AK14 I/O - - - U2 AT24 W4
I/O - - - AC12 AB36 AL13 I/O - - - T2 AR23 W3
I/O - - - AF11 AD38 AK13 GND - - P143 GND* GND* GND*
VCC - - - VCC* VCC* VCC* VCC - - - VCC* VCC* VCC*
GND - - P98 GND* GND* GND* I/O - - - - AW25 W2
I/O - - - AE11 AB34 AJ13 I/O - - - - AW23 V2
I/O - - - AD11 AE39 AH13 I/O - - - T1 AP22 V4
I/O - - - AE10 AM36 AL12 I/O - - - R4 AV24 V3
I/O - - - AC11 AC35 AK12 I/O - P124 P144 R3 AU23 U1
I/O - - P99 AF9 AG39 AH12 I/O - P125 P145 R2 AT22 U2
I/O - - P100 AD10 AG37 AJ11 I/O P96 P126 P146 R1 AR21 U4
VCC - - P101 VCC* VCC* VCC* I/O P97 P127 P147 P3 AV22 U3
I/O P66 P86 P102 AE9 AD34 AL10 I/O (D4) P98 P128 P148 P2 AP20 T1
I/O P67 P87 P103 AD9 AN39 AK10 I/O P99 P129 P149 P1 AU21 T2
I/O P68 P88 P104 AC10 AE35 AJ10 VCC P100 P130 P150 VCC* VCC* VCC*
I/O P69 P89 P105 AF7 AH38 AK9 GND P101 P131 P151 GND* GND* GND*
GND P70 P90 P106 GND* GND* GND* I/O (D3) P102 P132 P152 N2 AU19 T3
I/O - - - AE8 AJ37 AL8 I/O (RS) P103 P133 P153 N4 AV20 R1
I/O - - - AD8 AG35 AH10 I/O P104 P134 P154 N3 AV18 R2
I/O - - P107 AC9 AF34 AJ9 I/O P105 P135 P155 M1 AR19 R4
I/O - - P108 AF6 AH36 AK8 I/O - P136 P156 M2 AT18 R3
I/O - - - - AK38 AJ8 I/O - P137 P157 M3 AW17 P2
I/O - - - - AP38 AH9 I/O - - - M4 AV16 P3
I/O - P91 P109 AE7 AK36 AK7 I/O - - - L1 AP18 P4
I/O - P92 P110 AD7 AM34 AL6 I/O - - - - AU17 N1
I/O P71 P93 P111 AE6 AH34 AJ7 I/O - - - - AW15 N2
I/O P72 P94 P112 AE5 AJ35 AH8 VCC - - - VCC* VCC* VCC*
GND - - - GND* GND* GND* GND - - P158 GND* GND* GND*
VCC - - - VCC* VCC* VCC* I/O - - - L2 AR17 N3
I/O - - - AD6 AL37 AK6 I/O - - - L3 AT16 N4
I/O - - - AC7 AT38 AL5 I/O - - - K2 AV14 M1
I/O P73 P95 P113 AF4 AM38 AH7 I/O - - - L4 AW13 M2
I/O P74 P96 P114 AF3 AN37 AJ6 I/O (D2) P106 P138 P159 J1 AR15 L2
I/O - - - AE4 AK34 AK5 I/O P107 P139 P160 K3 AP16 L3
I/O - - - AC6 AR39 AL4 VCC - - P161 VCC* VCC* VCC*
I/O P75 P97 P115 AD5 AN35 AK4 I/O P108 P140 P162 J2 AV12 K1
I/O P76 P98 P116 AE3 AL33 AH5 I/O P109 P141 P163 J3 AR13 K2
6-136 January 29, 1999 (Version 1.5)

R
Pad Name 160 208 240 352 411 432 Pad Name 160 208 240 352 411 432
I/O - - P164 K4 AU11 K3 GND P141 P182 P211 GND* GND* GND*
I/O - - P165 G1 AT12 K4 6/18//97
GND P110 P142 P166 GND* GND* GND* * Pads labelled GND* or VCC* are internally bonded to Ground or
I/O - - - H2 AP14 J2
VCC planes within the associated package. They have no direct
I/O - - - H3 AR11 J3
connection to any specific package pin.
I/O - - P167 J4 AV10 J4
I/O - - P168 F1 AT8 H1
I/O - P143 P169 G2 AT10 H2
I/O - P144 P170 G3 AP10 H3 Additional XC4044XL Package Pins
I/O P111 P145 P171 F2 AP12 H4
HQ208
I/O P112 P146 P172 E2 AR9 G2
Not Connected Pins
I/O - - - - AU9 G3
P1 P3 P51 P52 P53 P54 P102
I/O - - - - AV8 F1
P104 P105 P107 P155 P156 P157 P158
P206 P207 P208 - - - -
5/29/97
I/O (D1) P113 P147 P173 F3 AU7 G4
I/O (RCLK, RDY/BUSY) P114 P148 P174 G4 AW7 F2
I/O - - - D1 AW5 F3
I/O - - - C1 AV6 E1 HQ240
I/O - - - D2 AR7 E3 GND Pins
I/O - - - F4 AV4 D1 P204 P219 - - - - -
I/O P115 P149 P175 E3 AN9 E4 5/29/97
I/O P116 P150 P176 C2 AW1 D2
I/O (D0, DIN) P117 P151 P177 D3 AP6 C2 Note: These pins may be Not Connected for this device revision,
I/O, GCK6 (DOUT) P118 P152 P178 E4 AU3 D3 however for compatability with other devices in this package, these
CCLK P119 P153 P179 C3 AR5 D4 pins should be tied to GND.
VCC P120 P154 P180 VCC* VCC* VCC*
O, TDO P121 P159 P181 D4 AN7 C4
GND P122 P160 P182 GND* GND* GND*
BG352
I/O (A0, WS) P123 P161 P183 B3 AT4 B3
I/O, GCK7 (A1) P124 P162 P184 C4 AV2 D5
VCC Pins 6
A10 A17 B2 B25 D7 D13 D19
I/O P125 P163 P185 D5 AM8 B4
G23 H4 K1 K26 N23 P4 U1
I/O P126 P164 P186 A3 AL7 C5
U26 W23 Y4 AC8 AC14 AC20 AE2
I/O - - - C5 AR3 B5
AE25 AF10 AF17 - - - -
I/O - - - B4 AR1 C6
GND Pins
I/O (CS1,A2) P127 P165 P187 D6 AK6 A5
A1 A2 A5 A8 A14 A19 A22
I/O (A3) P128 P166 P188 C6 AN3 D7
A25 A26 B1 B26 E1 E26 H1
I/O - - - B5 AM6 B6
H26 N1 P26 W1 W26 AB1 AB26
I/O - - - A4 AM2 A6
AE1 AE26 AF1 AF2 AF5 AF8 AF13
AF19 AF22 AF25 AF26 - - -
6/13/97
I/O - - P189 C7 AL3 D8
I/O - - P190 B6 AH6 C7
I/O P129 P167 P191 A6 AP2 B7
I/O P130 P168 P192 D8 AK4 D9 PG411
I/O - - - C8 AN1 B8 VCC Pins
I/O - - - - AK2 A8 A3 A11 A21 A31 C39 D6 F36
I/O - P169 P193 B7 AG5 D10 J1 L39 W1 AA39 AJ1 AL39 AP4
I/O - P170 P194 A7 AF6 C9 AT34 AU1 AW9 AW19 AW29 AW37 -
I/O - - P195 D9 AL5 B9 GND Pins
A9 A19 A29 A37 C1 D14 D20
I/O - - - C9 AJ3 C10
D26 D34 F4 J39 L1 P4 P36
GND P131 P171 P196 GND* GND* GND* W39 Y4 Y36 AA1 AF4 AF36 AJ39
I/O P132 P172 P197 B8 AH2 B10 AL1 AP36 AT6 AT14 AT20 AT26 AU39
I/O P133 P173 P198 D10 AE5 A10 AW3 AW11 AW21 AW31 - - -
I/O - - P199 C10 AM4 C11 Not Connected Pins
I/O - - P200 B9 AD6 D12 A13 B6 B34 C25 C33 D12 E7
VCC - - P201 VCC* VCC* VCC* E23 E37 F2 G5 H34 L35 N3
I/O - - - A9 AG3 B11 P38 R3 AF2 AF38 AJ5 AL35 AN5
AP8 AR37 AT2 AU5 AU13 AU15 AU25
I/O - - - D11 AG1 C12
AU37 AV26 AV34 AW35 - - -
I/O - - - C11 AC5 C13 6/2/97
I/O - - - B10 AE1 A12
I/O - - - B11 AH4 D14
I/O - - - A11 AB6 B13
I/O (A4) P134 P174 P202 D12 AD2 C14
I/O (A5) P135 P175 P203 C12 AB4 A13
I/O - P176 P205 B12 AE3 B14
I/O P136 P177 P206 A12 AC1 D15
I/O (A21) P137 P178 P207 C13 AD4 C15
I/O (A20) P138 P179 P208 B13 AA5 B15
I/O - - - - AB2 A15
I/O - - - - AC3 C16
I/O (A6) P139 P180 P209 A13 AA3 B16
I/O (A7) P140 P181 P210 B14 Y6 A16
January 29, 1999 (Version 1.5) 6-137

R
BG432
VCC Pins
A1 A11 A21 A31 C3 C29 D11
D21 L1 L4 L28 L31 AA1 AA4
AA28 AA31 AH11 AH21 AJ3 AJ29 AL1
AL11 AL21 AL31 - - - -
GND Pins
A2 A3 A7 A9 A14 A18 A23
A25 A29 A30 B1 B2 B30 B31
C1 C31 D16 G1 G31 J1 J31
P1 P31 T4 T28 V1 V31 AC1
AC31 AE1 AE31 AH16 AJ1 AJ31 AK1
AK2 AK30 AK31 AL2 AL3 AL7 AL9
AL14 AL18 AL23 AL25 AL29 AL30 -
Not Connected Pins
A4 A28 B12 B21 C8 D6 D13
D20 D26 E2 F4 F28 F29 M3
M4 M28 M30 W1 W28 Y1 Y31
AE4 AF29 AF30 AG1 AH6 AH19 AJ5
AJ12 AJ20 AJ26 AK11 AK27 - -
5/29/97

(Note: XC4052XL is also available in the HQ304 package. XC4052XL HQ PG BG BG
The pinout is identical to the XC4036XL in HQ304.) Pad Name 240 411 432 560
GND - GND* GND* GND*
VCC - VCC* VCC* VCC*
I/O - K6 A27 C28
XC4052XL HQ PG BG BG I/O - G1 D25 D27
Pad Name 240 411 432 560 I/O - E1 C26 B30
VCC P212 VCC* VCC* VCC* I/O - E3 B27 C29
I/O (A8) P213 W3 D17 A17 I/O - F2 A28 E27
I/O (A9) P214 Y2 A17 B18 I/O - G5 D26 A31
I/O - V2 C17 C18 GND - GND* GND* GND*
I/O - W5 B17 E18 I/O P234 J7 C27 D28
GND - GND* GND* GND* I/O P235 H6 B28 C30
I/O (A19) P215 V4 C18 C19 I/O P236 C3 D27 D29
I/O (A18) P216 T2 D18 D19 I/O P237 D2 B29 E28
I/O P217 U1 B18 E19 I/O (A14) P238 E5 C28 D30
I/O P218 V6 A19 B20 I/O, GCK8 (A15) P239 G7 D28 E29
I/O (A10) P220 U3 B19 C20 VCC P240 VCC* VCC* VCC*
I/O (A11) P221 R1 C19 D20 GND P1 GND* GND* GND*
VCC - VCC* VCC* VCC* I/O, GCK1 (A16) P2 H8 D29 B33
GND - GND* GND* GND* I/O (A17) P3 F6 C30 F29
I/O - U5 D19 A21 I/O P4 B4 E28 E30
I/O - T4 A20 E20 I/O P5 D4 E29 D31
I/O - P2 B20 B21 I/O, TDI P6 B2 D30 F30
I/O - N1 C20 C21 I/O, TCK P7 G9 D31 C33
I/O - R3 B21 D21 GND - GND* GND* GND*
I/O - N3 D20 B22 I/O - E7 F28 G29
GND - GND* GND* GND* I/O - B6 F29 E31
I/O - R5 C21 C23 I/O - F8 E30 D32
I/O - M2 A22 E22 I/O - C5 E31 G30
VCC P222 VCC* VCC* VCC* I/O - A7 G28 F31
I/O P223 L3 B22 B24 I/O - A5 G29 H29
I/O P224 T6 C22 D23 VCC - VCC* VCC* VCC*
I/O P225 N5 B23 C24 GND - GND* GND* GND*
I/O P226 M4 A24 A25 I/O - C7 F30 H30
GND P227 GND* GND* GND* I/O - D8 F31 G31
I/O - K2 D22 E23 I/O P8 B8 H28 J29
I/O - K4 C23 B25 I/O P9 C9 H29 F33
I/O P228 P6 B24 D24 I/O P10 E9 G30 G32
I/O P229 M6 C24 C25 I/O P11 F12 H30 J30
GND - GND* GND* GND* GND - GND* GND* GND*
I/O - L5 D23 E25 I/O P12 D10 J28 K30
I/O - J5 B25 C27 I/O P13 B10 J29 H33
I/O P230 J3 A26 D26 I/O - F10 H31 L29
I/O P231 H2 C25 B28 I/O - F14 J30 K31
I/O (A12) P232 H4 D24 B29 GND P14 GND* GND* GND*
I/O (A13) P233 G3 B26 E26 I/O P15 C11 K28 L30
6-138 January 29, 1999 (Version 1.5)

R
XC4052XL HQ PG BG BG XC4052XL HQ PG BG BG
Pad Name 240 411 432 560 Pad Name 240 411 432 560
I/O P16 B12 K29 K32 I/O - A33 AE28 AF29
I/O, TMS P17 E11 K30 J33 I/O - C33 AF30 AH31
I/O P18 E15 K31 M29 I/O - B34 AF29 AG30
VCC P19 VCC* VCC* VCC* I/O P52 A35 AG31 AK32
I/O P20 F16 L29 L32 I/O P53 F32 AF28 AJ31
I/O P21 C13 L30 M31 GND - GND* GND* GND*
GND - GND* GND* GND* I/O - C35 AG30 AG29
I/O - A13 M30 N29 I/O - B38 AG29 AL33
I/O - D12 M28 L33 I/O P54 E33 AH31 AH30
I/O - B14 M29 M32 I/O P55 G31 AG28 AK31
I/O - E17 M31 P29 I/O P56 H32 AH30 AJ30
I/O - E13 N31 P30 I/O, GCK2 P57 B36 AJ30 AH29
I/O - A15 N28 N33 O (M1) P58 A39 AH29 AK30
GND P22 GND* GND* GND* GND P59 GND* GND* GND*
VCC - VCC* VCC* VCC* I (M0) P60 E35 AH28 AJ29
I/O - F18 N29 P31 VCC P61 VCC* VCC* VCC*
I/O - C15 N30 P32 I (M2) P62 G33 AJ28 AN32
I/O - B16 P30 R29 I/O, GCK3 P63 D36 AK29 AJ28
I/O - D16 P28 R30 I/O (HDC) P64 C37 AH27 AK29
I/O P23 D18 P29 R31 I/O P65 F34 AK28 AL30
I/O P24 A17 R31 R33 I/O P66 J33 AJ27 AK28
GND - GND* GND* GND* I/O P67 D38 AL28 AM31
I/O P25 E19 R30 T31 I/O (LDC) P68 G35 AH26 AJ27
I/O P26 B18 R28 T29 GND - GND* GND* GND*
I/O P27 C17 R29 U32 I/O - E37 AK27 AN31
I/O P28 C19 T31 U31 I/O - H34 AJ26 AL29
GND P29 GND* GND* GND* I/O - E39 AL27 AK27
VCC P30 VCC* VCC* VCC* I/O - K34 AH25 AL28
I/O P31 F20 T30 U29 I/O - F38 AK26 AJ26 6
I/O P32 B20 T29 U30 I/O - G37 AL26 AM30
I/O P33 C21 U31 V31 VCC - VCC* VCC* VCC*
I/O P34 B22 U30 V29 GND - GND* GND* GND*
GND - GND* GND* GND* I/O P69 H38 AH24 AM29
I/O P35 E21 U28 V30 I/O P70 J37 AJ25 AK26
I/O P36 D22 U29 W33 I/O P71 G39 AK25 AL27
I/O - A23 V30 W31 I/O P72 M34 AJ24 AJ25
I/O - B24 V29 W30 I/O - K36 AH23 AN29
I/O - C23 V28 W29 I/O - K38 AK24 AN28
I/O - F22 W31 Y32 GND - GND* GND* GND*
VCC - VCC* VCC* VCC* I/O P73 N35 AL24 AL25
GND P37 GND* GND* GND* I/O P74 P34 AH22 AJ23
I/O - A25 W30 Y31 I/O - J35 AJ23 AN26
I/O - D24 W29 Y30 I/O - L37 AK23 AL24
I/O - E23 W28 AA32 GND P75 GND* GND* GND*
I/O - C25 Y31 AA31 I/O P76 M38 AJ22 AK23
I/O - B26 Y30 AA30 I/O P77 R35 AK22 AN25
I/O - A27 Y29 AB32 I/O P78 H36 AL22 AJ22
GND - GND* GND* GND* I/O P79 T34 AJ21 AL23
I/O P38 C27 Y28 AA29 VCC P80 VCC* VCC* VCC*
I/O P39 F24 AA30 AB31 I/O P81 N37 AH20 AM24
VCC P40 VCC* VCC* VCC* I/O P82 N39 AK21 AK22
I/O P41 E25 AA29 AC31 GND - GND* GND* GND*
I/O P42 E27 AB31 AB29 I/O - P38 AJ20 AK21
I/O P43 B28 AB30 AD32 I/O - L35 AH19 AM22
I/O P44 C29 AB29 AC30 I/O - U35 AK20 AJ20
GND P45 GND* GND* GND* I/O - R39 AJ19 AL21
I/O - F26 AB28 AD31 I/O - M36 AL20 AN21
I/O - D28 AC30 AE33 I/O - V34 AH18 AK20
I/O P46 B30 AC29 AC29 GND P83 GND* GND* GND*
I/O P47 E29 AC28 AE32 VCC - VCC* VCC* VCC*
GND - GND* GND* GND* I/O - R37 AK19 AL20
I/O - D30 AD31 AG33 I/O - T38 AJ18 AJ19
I/O - D32 AD30 AH33 I/O P84 T36 AL19 AM20
I/O P48 F28 AD29 AE29 I/O P85 V36 AK18 AK19
I/O P49 F30 AD28 AG31 I/O P86 U37 AH17 AL19
I/O P50 C31 AE30 AF30 I/O P87 U39 AJ17 AN19
I/O P51 E31 AE29 AH32 GND - GND* GND* GND*
GND - GND* GND* GND* I/O - W35 AK17 AL18
VCC - VCC* VCC* VCC* I/O - AC39 AL17 AM18
I/O - B32 AF31 AJ32 I/O P88 V38 AJ16 AK17
January 29, 1999 (Version 1.5) 6-139

R
Pad Name 240 411 432 560 Pad Name 240 411 432 560
I/O (INIT) P89 W37 AK16 AJ17 I/O - AV34 AG1 AG4
VCC P90 VCC* VCC* VCC* I/O - AW35 AE4 AH3
GND P91 GND* GND* GND* I/O - AW33 AE3 AF5
I/O P92 Y34 AL16 AL17 I/O - AU33 AF2 AJ2
I/O P93 AC37 AH15 AM17 VCC - VCC* VCC* VCC*
I/O - Y38 AL15 AN17 GND - GND* GND* GND*
I/O - AA37 AJ15 AK16 I/O (D6) P129 AV32 AF1 AJ1
GND - GND* GND* GND* I/O P130 AU31 AD4 AF4
I/O P94 AB38 AK15 AM16 I/O P131 AR31 AD3 AG3
I/O P95 AD36 AJ14 AL15 I/O P132 AP28 AE2 AE5
I/O P96 AA35 AH14 AK15 I/O - AP30 AD2 AH1
I/O P97 AE37 AK14 AJ15 I/O - AT30 AC4 AF3
I/O - AB36 AL13 AN15 GND - GND* GND* GND*
I/O - AD38 AK13 AM14 I/O P133 AT32 AC3 AE3
VCC - VCC* VCC* VCC* I/O P134 AV30 AD1 AC5
GND P98 GND* GND* GND* I/O - AR29 AC2 AE1
I/O - AB34 AJ13 AL14 I/O - AP26 AB4 AD3
I/O - AE39 AH13 AK14 GND P135 GND* GND* GND*
I/O - AM36 AL12 AJ14 I/O P136 AU29 AB3 AC4
I/O - AC35 AK12 AN13 I/O P137 AV28 AB2 AD2
I/O - AL35 AJ12 AM13 I/O P138 AT28 AB1 AB5
I/O - AF38 AK11 AL13 I/O P139 AR25 AA3 AC3
GND - GND* GND* GND* VCC P140 VCC* VCC* VCC*
I/O P99 AG39 AH12 AK12 I/O (D5) P141 AP24 AA2 AA5
I/O P100 AG37 AJ11 AN11 I/O (CS0) P142 AU27 Y2 AB3
VCC P101 VCC* VCC* VCC* GND - GND* GND* GND*
I/O P102 AD34 AL10 AJ12 I/O - AR27 Y4 AB2
I/O P103 AN39 AK10 AL11 I/O - AW27 Y3 AA4
I/O P104 AE35 AJ10 AK11 I/O - AU25 Y1 AA3
I/O P105 AH38 AK9 AM10 I/O - AV26 W1 Y5
GND P106 GND* GND* GND* I/O - AT24 W4 Y3
I/O - AJ37 AL8 AL10 I/O - AR23 W3 Y2
I/O - AG35 AH10 AJ11 GND P143 GND* GND* GND*
I/O P107 AF34 AJ9 AN9 VCC - VCC* VCC* VCC*
I/O P108 AH36 AK8 AK10 I/O - AW25 W2 W5
GND - GND* GND* GND* I/O - AW23 V2 W4
I/O - AK38 AJ8 AN7 I/O - AP22 V4 W3
I/O - AP38 AH9 AJ9 I/O - AV24 V3 W1
I/O P109 AK36 AK7 AL7 I/O P144 AU23 U1 V3
I/O P110 AM34 AL6 AK8 I/O P145 AT22 U2 V5
I/O P111 AH34 AJ7 AN6 GND - GND* GND* GND*
I/O P112 AJ35 AH8 AM6 I/O P146 AR21 U4 V4
GND - GND* GND* GND* I/O P147 AV22 U3 V2
VCC - VCC* VCC* VCC* I/O (D4) P148 AP20 T1 U5
I/O - AL37 AK6 AJ8 I/O P149 AU21 T2 U4
I/O - AT38 AL5 AL6 VCC P150 VCC* VCC* VCC*
I/O P113 AM38 AH7 AK7 GND P151 GND* GND* GND*
I/O P114 AN37 AJ6 AM5 I/O (D3) P152 AU19 T3 U3
I/O - AK34 AK5 AM4 I/O (RS) P153 AV20 R1 T2
I/O - AR39 AL4 AJ7 I/O P154 AV18 R2 T4
GND - GND* GND* GND* I/O P155 AR19 R4 R1
I/O - AR37 AH6 AL5 GND - GND* GND* GND*
I/O - AU37 AJ5 AK6 I/O P156 AT18 R3 R3
I/O P115 AN35 AK4 AN3 I/O P157 AW17 P2 R4
I/O P116 AL33 AH5 AK5 I/O - AV16 P3 R5
I/O P117 AV38 AK3 AJ6 I/O - AP18 P4 P2
I/O, GCK4 P118 AT36 AJ4 AL4 I/O - AU17 N1 P3
GND P119 GND* GND* GND* I/O - AW15 N2 P4
DONE P120 AR35 AH4 AJ5 VCC - VCC* VCC* VCC*
VCC P121 VCC* VCC* VCC* GND P158 GND* GND* GND*
PROGRAM P122 AN33 AH3 AM1 I/O - AR17 N3 N1
I/O (D7) P123 AM32 AJ2 AH5 I/O - AT16 N4 P5
I/O, GCK5 P124 AP34 AG4 AJ4 I/O - AV14 M1 N2
I/O P125 AW39 AG3 AK3 I/O - AW13 M2 N3
I/O P126 AN31 AH2 AH4 I/O - AU15 M3 N5
I/O - AV36 AH1 AL1 I/O - AU13 M4 M3
I/O - AR33 AF4 AG5 GND - GND* GND* GND*
GND - GND* GND* GND* I/O (D2) P159 AR15 L2 M4
I/O P127 AP32 AF3 AJ3 I/O P160 AP16 L3 L1
I/O P128 AU35 AG2 AK2 VCC P161 VCC* VCC* VCC*
6-140 January 29, 1999 (Version 1.5)

R
Pad Name 240 411 432 560 Pad Name 240 411 432 560
I/O P162 AV12 K1 K2 I/O - AF2 D13 C13
I/O P163 AR13 K2 L4 I/O - AJ5 B12 E14
I/O P164 AU11 K3 J1 I/O - AC5 C13 A13
I/O P165 AT12 K4 K3 I/O - AE1 A12 D14
GND P166 GND* GND* GND* I/O - AH4 D14 C14
I/O - AP14 J2 L5 I/O - AB6 B13 B14
I/O - AR11 J3 J2 GND - GND* GND* GND*
I/O P167 AV10 J4 K4 VCC - VCC* VCC* VCC*
I/O P168 AT8 H1 J3 I/O (A4) P202 AD2 C14 E15
GND - GND* GND* GND* I/O (A5) P203 AB4 A13 D15
I/O P169 AT10 H2 G1 I/O P205 AE3 B14 C15
I/O P170 AP10 H3 F1 I/O P206 AC1 D15 A15
I/O P171 AP12 H4 J5 I/O (A21) P207 AD4 C15 C16
I/O P172 AR9 G2 G3 I/O (A20) P208 AA5 B15 E16
I/O - AU9 G3 H4 GND - GND* GND* GND*
I/O - AV8 F1 F2 I/O - AB2 A15 B17
GND - GND* GND* GND* I/O - AC3 C16 C17
VCC - VCC* VCC* VCC* I/O (A6) P209 AA3 B16 E17
I/O (D1) P173 AU7 G4 F3 I/O (A7) P210 Y6 A16 D17
I/O (RCLK, RDY/BUSY) P174 AW7 F2 G4 GND P211 GND* GND* GND*
I/O - AW5 F3 D2 6/20/97
I/O - AV6 E1 E3 * Pads labelled GND* or VCC* are internally bonded to Ground or
I/O - AU5 F4 G5 VCC planes within the associated package. They have no direct
I/O - AP8 E2 C1 connection to any specific package pin.
I/O - AR7 E3 F4
I/O - AV4 D1 D3
I/O P175 AN9 E4 B3 HQ240
I/O P176 AW1 D2 F5 GND Pins 6
I/O (D0, DIN) P177 AP6 C2 E4 P204 P219 - - - - -
I/O, GCK6 (DOUT) P178 AU3 D3 D4 6/3/97
CCLK P179 AR5 D4 C4 Note: These pins may be Not Connected for this device revision,
VCC P180 VCC* VCC* VCC* however for compatability with other devices in this package, these
O, TDO P181 AN7 C4 E6 pins should be tied to GND.
GND P182 GND* GND* GND*
I/O (A0, WS) P183 AT4 B3 D5
PG411
I/O, GCK7 (A1) P184 AV2 D5 A2 VCC Pins
I/O P185 AM8 B4 D6 A3 A11 A21 A31 C39 D6 F36
I/O P186 AL7 C5 A3 J1 L39 W1 AA39 AJ1 AL39 AP4
I/O - AT2 A4 E7 AT34 AU1 AW9 AW19 AW29 AW37 -
I/O - AN5 D6 C5 GND Pins
GND - GND* GND* GND* A9 A19 A29 A37 C1 D14 D20
I/O - AR3 B5 B4 D26 D34 F4 J39 LI P4 P36
I/O - AR1 C6 D7 W39 Y4 Y36 AA1 AF4 AF36 AJ39
I/O (CS1, A2) P187 AK6 A5 C6 AL1 AP36 AT6 AT14 AT20 AT26 AU39
I/O (A3) P188 AN3 D7 E8 AW3 AW11 AW21 AW31 - - -
I/O - AM6 B6 B5 6/3/97
I/O - AM2 A6 A5
VCC - VCC* VCC* VCC*
BG432
VCC Pins
I/O P189 AL3 D8 D8
A1 A11 A21 A31 C3 C29 D11
I/O P190 AH6 C7 C7
D21 L1 L4 L28 L31 AA1 AA4
I/O P191 AP2 B7 E9
AA28 AA31 AH11 AH21 AJ3 AJ29 AL1
I/O P192 AK4 D9 A6
AL11 AL21 AL31 - - - -
I/O - AN1 B8 B7
GND Pins
I/O - AK2 A8 D9
A2 A3 A7 A9 A14 A18 A23
A25 A29 A30 B1 B2 B30 B31
I/O P193 AG5 D10 E11
C1 C31 D16 G1 G31 J1 J31
I/O P194 AF6 C9 A9
P1 P31 T4 T28 V1 V31 AC1
I/O P195 AL5 B9 C10
AC31 AE1 AE31 AH16 AJ1 AJ31 AK1
I/O - AJ3 C10 D11
AK2 AK30 AK31 AL2 AL3 AL7 AL9
GND P196 GND* GND* GND*
AL14 AL18 AL23 AL25 AL29 AL30 -
I/O P197 AH2 B10 B10
Not Connected Pins
I/O P198 AE5 A10 E12
C8 - - - - - -
I/O P199 AM4 C11 C11
6/3/97
I/O P200 AD6 D12 B11
VCC P201 VCC* VCC* VCC*
I/O - AG3 B11 D12
I/O - AG1 C12 A11
January 29, 1999 (Version 1.5) 6-141

R
PG560
VCC Pins
A4 A10 A16 A22 A26 A30 B2
B13 B19 B32 C3 C31 C32 D1
D33 E5 H1 K33 M1 N32 R2
T33 V1 W32 AA2 AB33 AD1 AF33
AK1 AK4 AK33 AL2 AL3 AL31 AM2
AM15 AM21 AM32 AN4 AN8 AN12 AN18
AN24 AN30 - - - - -
GND Pins
A7 A12 A14 A18 A20 A24 A29
A32 B1 B6 B9 B15 B23 B27
B31 C2 E1 F32 G2 G33 J32
K1 L2 M33 P1 P33 R32 T1
V33 W2 Y1 Y33 AB1 AC32 AD33
AE2 AG1 AG32 AH2 AJ33 AL32 AM3
AM7 AM11 AM19 AM25 AM28 AM33 AN2
AN5 AN10 AN14 AN16 AN20 AN22 AN27
Not Connected Pins
A1 A8 A19 A23 A27 A28 A33
B8 B12 B16 B26 C8 C9 C12
C22 C26 D10 D13 D16 D18 D22
D25 E2 E10 E13 E21 E24 E32
E33 H2 H3 H5 H31 H32 J4
J31 K5 K29 L3 L31 M2 M5
M30 N4 N30 N31 T3 T5 T30
T32 U1 U2 U33 V32 Y4 Y29
AA1 AA33 AB4 AB30 AC1 AC2 AC33
AD4 AD5 AD29 AD30 AE4 AE30 AE31
AF1 AF2 AF31 AF32 AG2 AJ10 AJ13
AJ16 AJ18 AJ21 AJ24 AK9 AK13 AK18
AK24 AK25 AL8 AL9 AL12 AL16 AL22
AL26 AM8 AM9 AM12 AM23 AM26 AM27
AN1 AN23 AN33 - - - -
6/20/97

(Note: XC4062XL is also available in the HQ304 package. XC4062XL
HQ240 BG432 PG475 BG560
The pinout is identical to the XC4036XL in HQ304.) Pad Name
I/O P224 C22 P4 D23
I/O P225 B23 R7 C24
XC4062XL
HQ240 BG432 PG475 BG560 I/O P226 A24 M2 A25
Pad Name
VCC P212 VCC* VCC* VCC* GND P227 GND* GND* GND*
I/O (A8) P213 D17 Y2 A17 I/O - D22 M4 E23
I/O (A9) P214 A17 Y4 B18 I/O - C23 L3 B25
I/O - C17 W5 C18 I/O P228 B24 N5 D24
I/O - B17 Y6 E18 I/O P229 C24 K2 C25
I/O - - U3 D18 I/O - - L5 B26
I/O - - W3 A19 I/O - - J1 E24
I/O (A19) P215 C18 W1 C19 I/O - D23 M6 E25
I/O (A18) P216 D18 U5 D19 I/O - B25 K4 C27
I/O P217 B18 W7 E19 I/O P230 A26 J3 D26
I/O P218 A19 U7 B20 I/O P231 C25 J5 B28
I/O (A10) P220 B19 V2 C20 I/O (A12) P232 D24 H2 B29
I/O (A11) P221 C19 V4 D20 I/O (A13) P233 B26 G1 E26
VCC - VCC* VCC* VCC* GND - GND* GND* GND*
GND - GND* GND* GND* VCC - VCC* VCC* VCC*
I/O - D19 V6 A21 I/O - A27 L7 C28
I/O - A20 R1 E20 I/O - D25 K6 D27
I/O - B20 T6 B21 I/O - C26 E1 B30
I/O - C20 R3 C21 I/O - B27 H4 C29
I/O - B21 R5 D21 I/O - A28 G5 E27
I/O - D20 T4 B22 I/O - D26 F2 A31
I/O - C21 P2 C23 I/O P234 C27 H6 D28
I/O - A22 N1 E22 I/O P235 B28 C3 C30
VCC P222 VCC* VCC* VCC* I/O P236 D27 F4 D29
I/O P223 B22 N3 B24 I/O P237 B29 C5 E28
6-142 January 29, 1999 (Version 1.5)

R
XC4062XL XC4062XL
HQ240 BG432 PG475 BG560 HQ240 BG432 PG475 BG560
Pad Name Pad Name
I/O (A14) P238 C28 E3 D30 GND - GND* GND* GND*
I/O GCK8 (A15) P239 D28 E5 E29 I/O P35 U28 C23 V30
VCC P240 VCC* VCC* VCC* I/O P36 U29 F24 W33
GND P1 GND* GND* GND* I/O - V30 A23 W31
I/O, GCK1 (A16) P2 D29 G7 B33 I/O - V29 E25 W30
I/O (A17) P3 C30 D4 F29 I/O - V28 G23 W29
I/O P4 E28 A5 E30 I/O - W31 B24 Y32
I/O P5 E29 B4 D31 VCC - VCC* VCC* VCC*
I/O, TDI P6 D30 D6 F30 GND P37 GND* GND* GND*
I/O, TCK P7 D31 F8 C33 I/O - W30 D24 Y31
GND - GND* GND* GND* I/O - W29 C25 Y30
I/O - F28 B6 G29 I/O - W28 D28 AA32
I/O - F29 E7 E31 I/O - Y31 A27 AA31
I/O - E30 D8 D32 I/O - Y30 E29 AA30
I/O - E31 G9 G30 I/O - Y29 C27 AB32
I/O - G28 E9 F31 GND - GND* GND* GND*
I/O - G29 A7 H29 I/O P38 Y28 G25 AA29
VCC - VCC* VCC* VCC* I/O P39 AA30 D26 AB31
I/O - F30 B8 H30 I/O P41 AA29 F26 AC31
I/O - F31 C9 G31 I/O P42 AB31 B28 AB29
I/O P8 H28 G11 J29 I/O P43 AB30 D30 AD32
I/O P9 H29 D10 F33 I/O P44 AB29 A29 AC30
I/O P10 G30 E11 G32 GND P45 GND* GND* GND*
I/O P11 H30 A9 J30 I/O - AB28 C29 AD31
GND - GND* GND* GND* I/O - AC30 G27 AE33
I/O - - B10 H32 I/O P46 AC29 F30 AC29
I/O - - C11 J31 I/O P47 AC28 B30 AE32
I/O P12 J28 F12 K30 I/O - - E31 AD30 6
I/O P13 J29 D12 H33 I/O - - C31 AE31
I/O - H31 A11 L29 GND - GND* GND* GND*
I/O - J30 G15 K31 I/O - AD31 F28 AG33
GND P14 GND* GND* GND* I/O - AD30 D32 AH33
I/O P15 K28 B12 L30 I/O P48 AD29 B32 AE29
I/O P16 K29 E13 K32 I/O P49 AD28 G31 AG31
I/O, TMS P17 K30 C13 J33 I/O P50 AE30 A33 AF30
I/O P18 K31 A13 M29 I/O P51 AE29 C33 AH32
I/O P20 L29 B14 L32 VCC - VCC* VCC* VCC*
I/O P21 L30 C15 M31 I/O - AF31 B34 AJ32
GND - GND* GND* GND* I/O - AE28 A35 AF29
I/O - M30 G17 N29 I/O - AF30 E33 AH31
I/O - M28 F14 L33 I/O - AF29 D34 AG30
I/O - M29 D16 M32 I/O P52 AG31 D36 AK32
I/O - M31 D14 P29 I/O P53 AF28 B36 AJ31
I/O - N31 A15 P30 GND - GND* GND* GND*
I/O - N28 C17 N33 I/O - AG30 F34 AG29
GND P22 GND* GND* GND* I/O - AG29 D38 AL33
VCC - VCC* VCC* VCC* I/O P54 AH31 C37 AH30
I/O - N29 D18 P31 I/O P55 AG28 G37 AK31
I/O - N30 B18 P32 I/O P56 AH30 B38 AJ30
I/O - P30 F16 R29 I/O, GCK2 P57 AJ30 F38 AH29
I/O - P28 G19 R30 O (M1) P58 AH29 A39 AK30
I/O P23 P29 E17 R31 GND P59 GND* GND* GND*
I/O P24 R31 E19 R33 I (M0) P60 AH28 E35 AJ29
I/O P25 R30 A19 T31 I (M2) P62 AJ28 G33 AN32
I/O P26 R28 F18 T29 I/O, GCK3 P63 AK29 J37 AJ28
I/O - - C19 T30 I/O (HDC) P64 AH27 G35 AK29
I/O - - D20 T32 I/O P65 AK28 K36 AL30
I/O P27 R29 F20 U32 I/O P66 AJ27 C39 AK28
I/O P28 T31 B20 U31 I/O P67 AL28 K38 AM31
GND P29 GND* GND* GND* I/O (LDC) P68 AH26 C41 AJ27
I/O P31 T30 C21 U29 I/O - AK27 D40 AN31
I/O P32 T29 A21 U30 I/O - AJ26 L37 AL29
I/O - - D22 U33 I/O - AL27 H36 AK27
I/O - - B22 V32 I/O - AH25 M36 AL28
I/O P33 U31 E23 V31 I/O - AK26 J35 AJ26
I/O P34 U30 F22 V29 I/O - AL26 E41 AM30
January 29, 1999 (Version 1.5) 6-143

R
XC4062XL XC4062XL
Pad Name Pad Name
VCC - VCC* VCC* VCC* I/O P100 AJ11 AH40 AN11
I/O P69 AH24 F40 AM29 I/O P102 AL10 AJ41 AJ12
I/O P70 AJ25 H38 AK26 I/O P103 AK10 AJ39 AL11
I/O P71 AK25 N37 AL27 I/O P104 AJ10 AJ37 AK11
I/O P72 AJ24 L35 AJ25 I/O P105 AK9 AG35 AM10
I/O - AH23 R35 AN29 GND P106 GND* GND* GND*
I/O - AK24 G41 AN28 I/O - AL8 AK40 AL10
GND - GND* GND* GND* I/O - AH10 AK38 AJ11
I/O - - H40 AM26 I/O P107 AJ9 AL37 AN9
I/O - - P38 AK24 I/O P108 AK8 AL39 AK10
I/O P73 AL24 J39 AL25 I/O - - AM38 AM9
I/O P74 AH22 R37 AJ23 I/O - - AM40 AL9
I/O - AJ23 J41 AN26 GND - GND* GND* GND*
I/O - AK23 K40 AL24 I/O - AJ8 AN41 AN7
GND P75 GND* GND* GND* I/O - AH9 AM36 AJ9
I/O P76 AJ22 L39 AK23 I/O P109 AK7 AK36 AL7
I/O P77 AK22 M38 AN25 I/O P110 AL6 AU41 AK8
I/O P78 AL22 T36 AJ22 I/O P111 AJ7 AN39 AN6
I/O P79 AJ21 M40 AL23 I/O P112 AH8 AP40 AM6
I/O P81 AH20 N39 AM24 VCC - VCC* VCC* VCC*
I/O P82 AK21 N41 AK22 I/O - AK6 AR41 AJ8
GND - GND* GND* GND* I/O - AL5 AL35 AL6
I/O - AJ20 P40 AK21 I/O P113 AH7 AV40 AK7
I/O - AH19 T38 AM22 I/O P114 AJ6 AN37 AM5
I/O - AK20 U35 AJ20 I/O - AK5 AT38 AM4
I/O - AJ19 U37 AL21 I/O - AL4 AP38 AJ7
I/O - AL20 R39 AN21 GND - GND* GND* GND*
I/O - AH18 R41 AK20 I/O - AH6 AT40 AL5
GND P83 GND* GND* GND* I/O - AJ5 AW39 AK6
VCC - VCC* VCC* VCC* I/O P115 AK4 AP36 AN3
I/O - AK19 V36 AL20 I/O P116 AH5 AU37 AK5
I//O - AJ18 U39 AJ19 I/O P117 AK3 AR37 AJ6
I/O P84 AL19 V38 AM20 I/O, GCK4 P118 AJ4 AU39 AL4
I/O P85 AK18 V40 AK19 GND P119 GND* GND* GND*
I/O P86 AH17 W37 AL19 DONE P120 AH4 AR35 AJ5
I/O P87 AJ17 W35 AN19 VCC P121 VCC* VCC* VCC*
GND - GND* GND* GND* PROGRAM P122 AH3 AN35 AM1
I/O - - W41 AJ18 I/O (D7) P123 AJ2 AU35 AH5
I/O - - Y36 AK18 I/O, GCK5 P124 AG4 AV38 AJ4
I/O - AK17 W39 AL18 I/O P125 AG3 AT34 AK3
I/O - AL17 AB36 AM18 I/O P126 AH2 BA39 AH4
I/O P88 AJ16 Y40 AK17 I/O - AH1 AU33 AL1
I/O (INIT) P89 AK16 Y38 AJ17 I/O - AF4 AY38 AG5
GND P91 GND* GND* GND* I/O P127 AF3 AV36 AJ3
I/O P92 AL16 AA39 AL17 I/O P128 AG2 AR31 AK2
I/O P93 AH15 AB38 AM17 I/O - AG1 AR33 AG4
I/O - AL15 AB40 AN17 I/O - AE4 AV32 AH3
I/O - AJ15 AC37 AK16 I/O - AE3 BA37 AF5
I/O - - AC39 AJ16 I/O - AF2 AY36 AJ2
I/O - - AC41 AL16 VCC - VCC* VCC* VCC*
I/O P94 AK15 AD36 AM16 I/O (D6) P129 AF1 AV34 AJ1
I/O P95 AJ14 AC35 AL15 I/O P130 AD4 BA35 AF4
I/O P96 AH14 AE37 AK15 I/O P131 AD3 AU31 AG3
I/O P97 AK14 AD40 AJ15 I/O P132 AE2 AY34 AE5
I/O - AL13 AD38 AN15 I/O - AD2 AT30 AH1
I/O - AK13 AE39 AM14 I/O - AC4 AW33 AF3
GND P98 GND* GND* GND* I/O - - BA33 AF1
I/O - AJ13 AG41 AL14 I/O - - AV30 AD4
I/O - AH13 AG39 AK14 I/O P133 AC3 AY32 AE3
I/O - AL12 AG37 AJ14 I/O P134 AD1 AU29 AC5
I/O - AK12 AE35 AN13 I/O - AC2 AW31 AE1
I/O - AJ12 AH38 AM13 I/O - AB4 BA31 AD3
I/O - AK11 AF38 AL13 GND P135 GND* GND* GND*
GND - GND* GND* GND* I/O P136 AB3 AR27 AC4
I/O P99 AH12 AF36 AK12 I/O P137 AB2 AT28 AD2
6-144 January 29, 1999 (Version 1.5)

R
XC4062XL XC4062XL
Pad Name Pad Name
I/O P138 AB1 AY30 AB5 I/O - G3 AY8 H4
I/O P139 AA3 AW29 AC3 I/O - F1 BA7 F2
I/O (D5) P141 AA2 BA29 AA5 VCC - VCC* VCC* VCC*
I/O (CS0) P142 Y2 AY28 AB3 I/O (D1) P173 G4 AV8 F3
GND P143 GND* GND* GND* I/O (RCLK, RDY/BUSY) P174 F2 AY6 G4
I/O - Y4 AR25 AB2 I/O - F3 AR11 D2
I/O - Y3 AV28 AA4 I/O - E1 AT8 E3
I/O - Y1 AW27 AA3 I/O - F4 AU9 G5
I/O - W1 AT26 Y5 I/O - E2 AW5 C1
I/O - W4 AV26 Y3 GND - GND* GND* GND*
I/O - W3 BA27 Y2 I/O - E3 AY4 F4
GND - GND* GND* GND* I/O - D1 BA5 D3
VCC - VCC* VCC* VCC* I/O P175 E4 AV4 B3
I/O - W2 AW25 W5 I/O P176 D2 AR9 F5
I/O - V2 AV24 W4 I/O (D0, DIN) P177 C2 AU5 E4
I/O - V4 AU25 W3 I/O, GCK6 (DOUT) P178 D3 AV6 D4
I/O - V3 AR23 W1 CCLK P179 D4 AR5 C4
I/O P144 U1 AT24 V3 VCC P180 VCC* VCC* VCC*
I/O P145 U2 AY24 V5 O, TDO P181 C4 AN7 E6
GND - GND* GND* GND* GND P182 GND* GND* GND*
I/O P146 U4 BA23 V4 I/O (A0, WS) P183 B3 AR7 D5
I/O P147 U3 AU23 V2 I/O, GCK7 (A1) P184 D5 AW3 A2
I/O - - AW23 U2 I/O P185 B4 AU3 D6
I/O - - AV20 U1 I/O P186 C5 AW1 A3
I/O (D4) P148 T1 AY22 U5 I/O - A4 AP6 E7
I/O P149 T2 AV22 U4 I/O - D6 AV2 C5
GND P151 GND* GND* GND* I/O - B5 AT4 B4 6
I/O (D3) P152 T3 AW21 U3 I/O - C6 AN5 D7
I/O (RS) P153 R1 BA21 T2 I/O (CS1, A2) P187 A5 AU1 C6
I/O - - AU19 T3 I/O (A3) P188 D7 AM6 E8
I/O - - AY20 T5 I/O - B6 AT2 B5
I/O P154 R2 AU17 T4 I/O - A6 AL7 A5
I/O P155 R4 AW19 R1 VCC - VCC* VCC* VCC*
I/O P156 R3 BA19 R3 I/O P189 D8 AR1 D8
I/O P157 P2 AT16 R4 I/O P190 C7 AP2 C7
I/O - P3 AR19 R5 I/O P191 B7 AM4 E9
I/O - P4 AV14 P2 I/O P192 D9 AN3 A6
I/O - N1 AY18 P3 I/O - B8 AL5 B7
I/O - N2 AV18 P4 I/O - A8 AK6 D9
GND P158 GND* GND* GND* I/O - - AN1 D10
I/O - N3 AT18 N1 I/O - - AJ5 C9
I/O - N4 AW17 P5 I/O P193 D10 AM2 E11
I/O - M1 AR15 N2 I/O P194 C9 AH4 A9
I/O - M2 BA15 N3 I/O P195 B9 AL3 C10
I/O - M3 AT14 N5 I/O - C10 AK4 D11
I/O - M4 AR17 M3 GND P196 GND* GND* GND*
GND - GND* GND* GND* I/O P197 B10 AG7 B10
I/O (D2) P159 L2 AW15 M4 I/O P198 A10 AG5 E12
I/O P160 L3 AV16 L1 I/O P199 C11 AK2 C11
VCC P161 VCC* VCC* VCC* I/O P200 D12 AJ3 B11
I/O P162 K1 AY14 K2 VCC P201 VCC* VCC* VCC*
I/O P163 K2 BA13 L4 I/O - B11 AJ1 D12
I/O P164 K3 AU13 J1 I/O - C12 AF6 A11
I/O P165 K4 AW13 K3 GND - GND* GND* GND*
GND P166 GND* GND* GND* I/O - D13 AH2 C13
I/O - J2 AY12 L5 I/O - B12 AF4 E14
I/O - J3 BA11 J2 I/O - C13 AE7 A13
I/O P167 J4 AV12 K4 I/O - A12 AE5 D14
I/O P168 H1 AT12 J3 I/O - D14 AG3 C14
I/O - - AW11 H2 I/O - B13 AG1 B14
I/O - - AY10 K5 GND - GND* GND* GND*
GND - GND* GND* GND* VCC - VCC* VCC* VCC*
I/O P169 H2 BA9 G1 I/O (A4) P202 C14 AD6 E15
I/O P170 H3 AU11 F1 I/O (A5) P203 A13 AD4 D15
I/O P171 H4 AW9 J5 I/O P205 B14 AE3 C15
I/O P172 G2 AV10 G3 I/O P206 D15 AC5 A15
January 29, 1999 (Version 1.5) 6-145

R
XC4062XL PG475
HQ240 BG432 PG475 BG560
Pad Name VCC Pins
I/O (A21) P207 C15 AD2 C16 A37 B2 B16 B26 B40 D2
I/O (A20) P208 B15 AC7 E16 E21 F6 F36 G13 G29 N7
GND - GND* GND* GND* N35 T2 T40 AA1 AA5 AA37
I/O - - AC1 D16 AA41 AF2 AF40 AJ7 AJ35 AR13
I/O - - AC3 B16 AR29 AT6 AT22 AT36 AU21 AW37
I/O - A15 AB6 B17 AW41 AY2 AY16 AY26 AY40 BA3
I/O - C16 AB2 C17 GND Pins
I/O (A6) P209 B16 AB4 E17 A3 C1 C7 G3 L1 P6
I/O (A7) P210 A16 AA3 D17 U1 A17 A25 A41 AA7 AE1
GND P211 GND* GND* GND* AH6 AL1 AR3 AW7 BA1 C35
6/16/97 E15 E27 F10 F32 G21 G39
* Pads labelled GND* or VCC* are internally bonded to Ground or L41 P36 U41 AA35 AE41 AH36
VCC planes within the package. They have no direct connection to AL41 AR21 AR39 AT10 AT20 AT32
any specific package pin. AU15 AU27 AW35 BA17 BA25 BA41
E37 E39 A31 J7 AP4 AU7
Additional XC4062XL Package Pins 5/5/97
HQ240
GND Pins
P204 P219 - - - - BG560
5/5/97 VCC Pins
A4 A10 A16 A22 A26 A30 B2
Note: These pins may be Not Connected for this device revision, B13 B19 B32 C3 C31 C32 D1
however for compatability with other devices in this package, these D33 E5 H1 K33 M1 N32 R2
pins should be tied to GND. T33 V1 W32 AA2 AB33 AD1 AF33
BG432 AN24 AN30 - - - - -
VCC Pins GND Pins
A1 A11 A21 A31 C3 C29 D11 A7 A12 A14 A18 A20 A24 A29
D21 L1 L4 L28 L31 AA1 AA4 A32 B1 B6 B9 B15 B23 B27
AA28 AA31 AH11 AH21 AJ3 AJ29 AL1 B31 C2 E1 F32 G2 G33 J32
AL11 AL21 AL31 - - - - K1 L2 M33 P1 P33 R32 T1
GND Pins V33 W2 Y1 Y33 AB1 AC32 AD33
A2 A3 A7 A9 A14 A18 A23 AE2 AG1 AG32 AH2 AJ33 AL32 AM3
A25 A29 A30 B1 B2 B30 B31 AM11 AM19 AM25 AM28 AM33 AM7 AN2
C1 C31 D16 G1 G31 J1 J31 AN5 AN10 AN14 AN16 AN20 AN22 AN27
P1 P31 T4 T28 V1 V31 AC1 Not Connected Pins
AC31 AE1 AE31 AH16 AJ1 AJ31 AK1 A1 A8 A23 A27 A28 A33 B8
AK2 AK30 AK31 AL2 AL3 AL7 AL9 B12 C8 C12 C22 C26 D13 D22
AL14 AL18 AL23 AL25 AL29 AL30 - D25 E2 E10 E13 E21 E32 E33
Not Connected Pins H3 H5 H31 J4 K29 L3 L31
C8 - - - - - - M2 M5 M30 N4 N30 N31 Y4
5/5/97 Y29 AA1 AA33 AB4 AB30 AC1 AC2
AC33 AD5 AD29 AE4 AE30 AF2 AF31
AF32 AG2 AJ10 AJ13 AJ21 AJ24 AK9
AK13 AK25 AL8 AL12 AL22 AL26 AM8
AM12 AM23 AM27 AN1 AN23 AN33 -
5/5/97

XC4085XL XC4085XL
BG432 BG560 PG559 BG432 BG560 PG559
Pad Name Pad Name
VCC VCC VCC* VCC* I/O B18 E19 Y6
I/O (A8) D17 A17 AB6 I/O A19 B20 AA1
I/O (A9) A17 B18 AB4 I/O (A10) B19 C20 Y4
I/O C17 C18 AA7 I/O (A11) C19 D20 W7
I/O B17 E18 AC1 VCC VCC* VCC* VCC*
I/O - D18 AA5 GND GND* GND* GND*
I/O - A19 AA3 I/O D19 A21 W5
GND GND* GND* GND* I/O A20 E20 V6
I/O (A19) C18 C19 Y8 I/O B20 B21 V4
I/O (A18) D18 D19 AB2 I/O C20 C21 Y2
I/O B21 D21 U3
6-146 January 29, 1999 (Version 1.5)

R
XC4085XL XC4085XL
BG432 BG560 PG559 BG432 BG560 PG559
Pad Name Pad Name
I/O D20 B22 U7 I/O G28 F31 H10
I/O - E21 V2 I/O G29 H29 G9
I/O - C22 U5 VCC VCC* VCC* VCC*
GND GND* GND* GND* GND GND* GND* GND*
I/O - D22 T4 I/O - E32 F8
I/O - A23 U1 I/O - E33 D8
I/O C21 C23 R3 I/O F30 H30 B6
I/O A22 E22 R5 I/O F31 G31 E9
VCC VCC VCC* VCC* I/O H28 J29 A7
I/O B22 B24 T8 I/O H29 F33 G11
I/O C22 D23 T2 I/O G30 G32 H14
I/O B23 C24 P4 I/O H30 J30 F12
I/O A24 A25 R7 VCC VCC* VCC* VCC*
I/O D22 E23 N3 I/O - H31 G13
I/O C23 B25 R1 I/O - K29 E11
I/O B24 D24 N5 I/O - H32 B8
I/O C24 C25 P2 I/O - J31 D10
I/O - B26 M4 I/O J28 K30 A9
I/O - E24 L1 I/O J29 H33 G15
I/O - C26 L3 I/O H31 L29 B10
I/O - D25 P8 I/O J30 K31 H16
VCC VCC* VCC* VCC* I/O K28 L30 C9
I/O - A27 N7 I/O K29 K32 E13 6
I/O - A28 K2 I/O (TMS) K30 J33 A11
I/O D23 E25 M6 I/O K31 M29 D12
I/O B25 C27 J1 VCC VCC* VCC* VCC*
I/O A26 D26 L5 I/O - L31 C11
I/O C25 B28 H2 I/O - M30 B14
I/O (A12) D24 B29 K4 I/O L29 L32 G17
I/O (A13) B26 E26 J3 I/O L30 M31 E15
VCC VCC* VCC* VCC* I/O M30 N29 D14
I/O A27 C28 L7 I/O M28 L33 A15
I/O D25 D27 J5 I/O - N30 C13
I/O C26 B30 G1 I/O - N31 B16
I/O B27 C29 H4 I/O M29 M32 E17
I/O A28 E27 F2 I/O M31 P29 F18
I/O D26 A31 G5 I/O N31 P30 A17
GND GND* GND* GND* I/O N28 N33 G19
I/O C27 D28 H6 GND GND* GND* GND*
I/O B28 C30 K8 VCC VCC* VCC* VCC*
I/O D27 D29 D2 I/O N29 P31 D16
I/O B29 E28 J7 I/O N30 P32 C15
I/O (A14) C28 D30 F4 I/O P30 R29 B18
I/O, GCK8 D28 E29 E3 I/O P28 R30 H20
(A15) I/O P29 R31 B20
VCC VCC* VCC* VCC* I/O R31 R33 E19
I/O, GCK1 D29 B33 C1 I/O R30 T31 D18
(A16) I/O R28 T29 F20
I/O (A17) C30 F29 C3 I/O - T30 G21
I/O E28 E30 F6 I/O - T32 C17
I/O E29 D31 A3 I/O R29 U32 D20
I/O (TDI) D30 F30 H8 I/O T31 U31 E21
I/O (TCK) D31 C33 D4 GND GND* GND* GND*
GND GND* GND* GND* VCC VCC* VCC* VCC*
I/O F28 G29 D6 I/O T30 U29 C21
I/O F29 E31 C5 I/O T29 U30 F22
I/O E30 D32 E7 I/O - U33 A21
I/O E31 G30 B4
January 29, 1999 (Version 1.5) 6-147

R
XC4085XL XC4085XL
BG432 BG560 PG559 BG432 BG560 PG559
Pad Name Pad Name
I/O - V32 D22 I/O AG28 AK31 H34
I/O U31 V31 B22 I/O AH30 AJ30 B40
I/O U30 V29 G23 I/O, GCK2 AJ30 AH29 E37
GND GND* GND* GND* O (M1) AH29 AK30 D38
I/O U28 V30 E23 GND GND* GND* GND*
I/O U29 W33 C23 I (M0) AH28 AJ29 C39
I/O V30 W31 A23 VCC VCC* VCC* VCC*
I/O V29 W30 D24 I (M2) AJ28 AN32 H36
I/O V28 W29 B24 I/O, GCK3 AK29 AJ28 F38
I/O W31 Y32 H24 I/O (HDC) AH27 AK29 C41
VCC VCC* VCC* VCC* I/O AK28 AL30 D40
GND GND* GND* GND* I/O AJ27 AK28 B42
I/O W30 Y31 F24 I/O AL28 AM31 J37
I/O W29 Y30 E25 I/O (LDC) AH26 AJ27 K36
I/O - AA33 B26 GND GND* GND* GND*
I/O - Y29 D26 I/O AK27 AN31 H38
I/O W28 AA32 A27 I/O AJ26 AL29 D42
I/O Y31 AA31 G25 I/O AL27 AK27 G39
I/O Y30 AA30 B28 I/O AH25 AL28 C43
I/O Y29 AB32 C27 I/O AK26 AJ26 F40
GND GND* GND* GND* I/O AL26 AM30 E41
I/O Y28 AA29 F26 VCC VCC* VCC* VCC*
I/O AA30 AB31 E27 GND GND* GND* GND*
I/O - AB30 A29 I/O AH24 AM29 L37
I/O - AC33 D28 I/O AJ25 AK26 J39
VCC VCC* VCC* VCC* I/O AK25 AL27 F42
I/O AA29 AC31 G27 I/O AJ24 AJ25 H40
I/O AB31 AB29 B30 I/O AH23 AN29 G43
I/O AB30 AD32 C29 I/O AK24 AN28 J41
I/O AB29 AC30 E29 I/O - AK25 H42
GND GND* GND* GND* I/O - AL26 N37
I/O AB28 AD31 D30 VCC VCC* VCC* VCC*
I/O AC30 AE33 A33 GND GND* GND* GND*
I/O AC29 AC29 C31 I/O - AJ24 P36
I/O AC28 AE32 B34 I/O - AM27 M38
I/O - AD30 H28 I/O - AM26 J43
I/O - AE31 A35 I/O - AK24 L39
I/O - AF32 G29 I/O AL24 AL25 K42
I/O - AD29 E31 I/O AH22 AJ23 K40
GND GND* GND* GND* I/O AJ23 AN26 L43
VCC VCC* VCC* VCC* I/O AK23 AL24 L41
I/O - AF31 D32 GND GND* GND* GND*
I/O - AE30 C35 I/O AJ22 AK23 R37
I/O AD31 AG33 C33 I/O AK22 AN25 P42
I/O AD30 AH33 B36 I/O AL22 AJ22 T36
I/O AD29 AE29 H30 I/O AJ21 AL23 N39
I/O AD28 AG31 A37 VCC VCC* VCC* VCC*
I/O AE30 AF30 G31 I/O AH20 AM24 M40
I/O AE29 AH32 F32 I/O AK21 AK22 R43
GND GND* GND* GND* I/O - AM23 N41
VCC VCC* VCC* VCC* I/O - AJ21 R39
I/O AF31 AJ32 E33 GND GND* GND* GND*
I/O AE28 AF29 D34 I/O - AL22 U37
I/O AF30 AH31 B38 I/O - AN23 T42
I/O AF29 AG30 G33 I/O AJ20 AK21 P40
I/O AG31 AK32 A41 I/O AH19 AM22 U43
I/O AF28 AJ31 E35 I/O AK20 AJ20 R41
GND GND* GND* GND* I/O AJ19 AL21 V42
I/O AG30 AG29 D36 I/O AL20 AN21 U39
I/O AG29 AL33 F36 I/O AH18 AK20 V38
I/O AH31 AH30 G35 GND GND* GND* GND*
6-148 January 29, 1999 (Version 1.5)

R
XC4085XL XC4085XL
BG432 BG560 PG559 BG432 BG560 PG559
Pad Name Pad Name
VCC VCC* VCC* VCC* I/O - AL8 AN43
I/O AK19 AL20 W37 I/O AJ8 AN7 AM38
I/O AJ18 AJ19 T40 I/O AH9 AJ9 AP42
I/O AL19 AM20 Y42 I/O AK7 AL7 AN39
I/O AK18 AK19 U41 I/O AL6 AK8 AR43
I/O AH17 AL19 Y36 I/O AJ7 AN6 AP40
I/O AJ17 AN19 V40 I/O AH8 AM6 AT40
I/O - AJ18 W39 VCC VCC* VCC* VCC*
I/O - AK18 AA43 I/O AK6 AJ8 AN37
I/O AK17 AL18 Y38 I.O AL5 AL6 AR39
I/O AL17 AM18 Y40 I/O AH7 AK7 AT42
I/O AJ16 AK17 AA37 I/O AJ6 AM5 BA43
I/O (INIT) AK16 AJ17 AA39 I/O AK5 AM4 AU43
VCC VCC* VCC* VCC* I/O AL4 AJ7 AU39
I/O AL16 AL17 AA41 I/O AH6 AL5 AT38
I/O AH15 AM17 AB38 I/O AJ5 AK6 AP36
I/O AL15 AN17 AB42 I/O AK4 AN3 AR37
I/O AJ15 AK16 AB40 I/O AH5 AK5 AV42
I/O - AJ16 AC37 I/O AK3 AJ6 AV40
I/O - AL16 AC39 I/O, GCK4 AJ4 AL4 AW41
I/O AK15 AM16 AD36 DONE AH4 AJ5 AY42
I/O AJ14 AL15 AC41 VCC VCC* VCC* VCC* 6
I/O AH14 AK15 AD38 PROGRAM AH3 AM1 BB42
I/O AK14 AJ15 AC43 I/O (D7) AJ2 AH5 BC41
I/O AL13 AN15 AD40 I/O, GCK5 AG4 AJ4 AV38
I/O AK13 AM14 AE39 I/O AG3 AK3 BA39
VCC VCC* VCC* VCC* I/O AH2 AH4 AT36
GND GND* GND* GND* I/O AH1 AL1 BB40
I/O AJ13 AL14 AE37 I/O AF4 AG5 AY40
I/O AH13 AK14 AF40 GND GND* GND* GND*
I/O AL12 AJ14 AD42 I/O AF3 AJ3 BA41
I/O AK12 AN13 AF42 I/O AG2 AK2 BB38
I/O AJ12 AM13 AF38 I/O AG1 AG4 AY38
I/O AK11 AL13 AG39 I/O AE4 AH3 BC37
I/O - AK13 AG43 I/O AE3 AF5 AW37
I/O - AJ13 AG37 I/O AF2 AJ2 AT34
I/O - AM12 AH40 GND GND* GND* GND*
I/O - AL12 AJ41 I/O (D6) AF1 AJ1 AU35
I/O AH12 AK12 AG41 I/O AD4 AF4 AV36
I/O AJ11 AN11 AK40 I/O AD3 AG3 BB36
VCC VCC* VCC* VCC* I/O AE2 AE5 AY36
I/O AL10 AJ12 AJ39 I/O AD2 AH1 BC35
I/O AK10 AL11 AH42 I/O AC4 AF3 AW35
I/O AJ10 AK11 AH36 I/O - AE4 AU33
I/O AK9 AM10 AL39 I/O - AG2 AT30
I/O AL8 AL10 AJ37 GND GND* GND* GND*
I/O AH10 AJ11 AJ43 I/O - AD5 AV32
I/O AJ9 AN9 AM40 I/O - AF2 AU31
I/O AK8 AK10 AK42 I/O - AF1 AW33
I/O - AM9 AN41 I/O - AD4 BB34
I/O - AL9 AL41 I/O AC3 AE3 AY34
I/O - AJ10 AR41 I/O AD1 AC5 BC33
I/O - AM8 AK36 I/O AC2 AE1 AU29
GND GND* GND* GND* I/O AB4 AD3 AT28
VCC VCC* VCC* VCC* GND GND* GND* GND*
I/O - AK9 AL37 I/O AB3 AC4 BA35
January 29, 1999 (Version 1.5) 6-149

R
XC4085XL XC4085XL
BG432 BG560 PG559 BG432 BG560 PG559
Pad Name Pad Name
I/O AB2 AD2 BB30 I/O - M5 BA15
I/O AB1 AB5 AW31 VCC VCC* VCC* VCC*
I/O AA3 AC3 AY32 I/O K1 K2 AW15
VCC VCC* VCC* VCC* I/O K2 L4 BC15
I/O - AB4 BA33 I/O K3 J1 AY14
I/O - AC1 AU27 I/O K4 K3 BA13
I/O (D5) AA2 AA5 BC29 GND GND* GND* GND*
I/O (CS0) Y2 AB3 AW29 I/O J2 L5 AT16
GND GND* GND* GND* I/O J3 J2 BB14
I/O Y4 AB2 AY30 I/O J4 K4 AU15
I/O Y3 AA4 BA31 I/O H1 J3 BC11
I/O Y1 AA3 BB28 I/O - H2 AW13
I/O W1 Y5 AW27 I/O - K5 BB10
I/O - AA1 BC27 I/O - H3 AY12
I/O - Y4 AV26 I/O - J4 BA11
I/O W4 Y3 AU25 GND GND* GND* GND*
I/O W3 Y2 AY28 VCC VCC* VCC* VCC*
GND GND* GND* GND* I/O H2 G1 AT14
VCC VCC* VCC* VCC* I/O H3 F1 AU13
I/O W2 W5 BA29 I/O H4 J5 AV12
I/O V2 W4 AT24 I/O G2 G3 BC9
I/O V4 W3 BB26 I/O G3 H4 AW11
I/O V3 W1 AW25 I/O F1 F2 BB8
I/O U1 V3 BB24 I/O - E2 AY10
I/O U2 V5 AY26 I/O - H5 AU11
I/O U4 V4 AV24 VCC VCC* VCC* VCC*
I/O U3 V2 AU23 I/O (D1) G4 F3 BA9
I/O - U2 BA27 I/O (RCLK F2 G4 AW9
I/O - U1 BC23 RDY/BUSY)
I/O (D4) T1 U5 AY24 I/O F3 D2 BC7
I/O T2 U4 AW23 I/O E1 E3 AY8
VCC VCC* VCC* VCC* I/O F4 G5 AV8
GND GND* GND* GND* I/O E2 C1 AT10
I/O (D3) T3 U3 BA23 GND GND* GND* GND*
I/O (RS) R1 T2 AV22 I/O E3 F4 AU9
I/O - T3 AY22 I/O D1 D3 BB6
I/O - T5 BB22 I/O E4 B3 AW7
I/O R2 T4 AU21 I/O D2 F5 BC3
I/O R4 R1 AW21 I/O (D0, DIN) C2 E4 AY6
GND GND* GND* GND* I/O, GCK6 D3 D4 BB4
I/O R3 R3 BA21 (DOUT)
I/O P2 R4 BC21 CCLK D4 C4 BA5
I/O P3 R5 AY20 VCC VCC* VCC* VCC*
I/O P4 P2 BB20 O, TDO C4 E6 BA3
I/O N1 P3 AT20 GND GND* GND* GND*
I/O N2 P4 AV20 I/O (A0, WS) B3 D5 AT8
VCC VCC* VCC* VCC* I/O, GCK7 D5 A2 AV6
GND GND* GND* GND* (A1)
I/O N3 N1 AW19 I/O B4 D6 BB2
I/O N4 P5 AY18 I/O C5 A3 AY4
I/O M1 N2 BB18 I/O A4 E7 AR7
I/O M2 N3 AU19 I/O D6 C5 AP8
I/O - N4 BC17 GND GND* GND* GND*
I/O - M2 BA17 I/O B5 B4 AT6
I/O M3 N5 AV18 I/O C6 D7 AY2
I/O M4 M3 AW17 I/O (CS1, A2) A5 C6 AU5
GND GND* GND* GND* I/O (A3) D7 E8 BA1
I/O (D2) L2 M4 AY16 I/O B6 B5 AV4
I/O L3 L1 BB16 I/O A6 A5 AW3
I/O - L3 AU17 VCC VCC* VCC* VCC*
GND GND* GND* GND*
6-150 January 29, 1999 (Version 1.5)

R
XC4085XL
BG432 BG560 PG559
Pad Name
BG560
I/O D8 D8 AN7
VCC Pins
I/O C7 C7 AR5
A4 A10 A16 A22 A26 A30 B2
I/O B7 E9 AV2
B13 B19 B32 C3 C31 C32 D1
I/O D9 A6 AT4
D33 E5 H1 K33 M1 N32 R2
I/O B8 B7 AU1
T33 V1 W32 AA2 AB33 AD1 AF33
I/O A8 D9 AR3
I/O - C8 AT2
I/O - E10 AL7
AN24 AN30 - - - - -
VCC VCC* VCC* VCC*
GND Pins
GND GND* GND* GND*
A7 A12 A14 A18 A20 A24 A29
I/O - B8 AK8
A32 B1 B6 B9 B15 B23 B27
I/O - A8 AM6
B31 C2 E1 F32 G2 G33 J32
I/O - D10 AN5
K1 L2 M33 P1 P33 R32 T1
I/O - C9 AR1
V33 W2 Y1 Y33 AB1 AC32 AD33
I/O D10 E11 AP4
AE2 AG1 AG32 AH2 AJ33 AL32 AM3
I/O C9 A9 AN3
AM11 AM19 AM25 AM28 AM33 AM7 AN2
I/O B9 C10 AP2
AN5 AN10 AN14 AN16 AN20 AN22 AN27
I/O C10 D11 AJ7
Not Connected Pins
GND GND* GND* GND*
A1 A33 AC2 AN1 AN33 - -
I/O B10 B10 AH8
6/4/97
I/O A10 E12 AL5
I/O C11 C11 AN1
I/O D12 B11 AM4
VCC VCC* VCC* VCC*
I/O B11 D12 AL3 6
I/O C12 A11 AJ5
I/O - E13 AK2
I/O - C12 AG7
GND GND* GND* GND*
I/O - B12 AK4
I/O - D13 AJ3
I/O D13 C13 AG5
I/O B12 E14 AJ1
I/O C13 A13 AF6
I/O A12 D14 AH2
I/O D14 C14 AE7
I/O B13 B14 AH4
GND GND* GND* GND*
VCC VCC* VCC* VCC*
I/O (A4) C14 E15 AG3
I/O (A5) A13 D15 AD8
I/O B14 C15 AG1
I/O D15 A15 AF4
I/O (A21) C15 C16 AE5
I/O (A20) B15 E16 AD6
GND GND* GND* GND*
I/O - D16 AD4
I/O - B16 AF2
I/O A15 B17 AC7
I/O C16 C17 AD2
I/O (A6) B16 E17 AC5
I/O (A7) A16 D17 AC3
GND GND* GND* GND*
6/13/97
*Pads labelled GND* or VCC* are internally bonded to Ground or

VCC planes within the package. They have no direct connection to
any specific package pin.
January 29, 1999 (Version 1.5) 6-151

R
PG559 BG432
VCC Pins VCC Pins
A13 A31 A43 B2 C7 C19 C25 A1 A11 A21 A31 D11 D21 L1
C37 F14 F30 G3 G7 G37 G41 L4 L28 L31 AA1 AA4 AA28 AA31
H12 H18 H26 H32 M8 M36 N1 AH11 AH21 AL1 AL11 AL21 AL31 C3
N43 P6 P38 V8 V36 W3 W41 C29 AJ3 AJ29
AE3 AE41 AF8 AF36 AK6 AK38 AL1
AL43 AM8 AM36 AT12 AT18 AT26 AT32
AU3 AU7 AU37 AU41 AV14 AV30 BA7
BA19 BA25 BA37 BC1 BC13 BC31 BC43
GND Pins GND Pins
A5 A19 A25 A39 B12 B32 E1 A2 A3 A7 A9 A14 A18 A23
E5 E39 E43 F10 F16 F28 F34 A25 A29 A30 B1 B2 B30 B31
H22 K6 K38 M2 M42 T6 T38 C1 C31 D16 G1 G31 J1 J31
W1 W43 AB8 AB36 AE1 AE43 AH6 P1 P31 T4 T28 V1 V31 AC1
AH38 AM2 AM42 AP6 AP38 AT22 AV10 AC31 AE1 AE31 AH16 AJ1 AJ31 AK1
AV16 AV28 AV34 AW1 AW5 AW39 AW43 AK2 AK30 AK31 AL2 AL3 AL7 AL9
BB12 BB32 BC5 BC19 BC25 BC39 - AL14 AL18 AL23 AL25 AL29 AL30
5/8/97 Not Connected Pins
C8
3/3/98
6-152 January 29, 1999 (Version 1.5)

R
January 29, 1999 (Version 1.5) 6-153

Table 25 - Table 27 show the planned packages and speed grades for XC4000-Series devices. Call your local sales office for
the latest availability information, or see the Xilinx WEBLINX at http://www.xilinx.com for the latest revision of the
specifications.
Table 25: Component Availability Chart for XC4000XL FPGAs
PINS 84 100 100 144 144 160 160 176 176 208 208 240 240 256 299 304 352 411 432 475 559 560
High-Perf.
High-Perf.
High-Perf.
High-Perf.
High-Perf.
High-Perf.
Ceram.
Ceram.
Ceram.
Ceram.
PQFP
VQFP
PQFP
PQFP
PQFP
TQFP
TQFP
TQFP
TQFP
PLCC
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
BGA
PGA
BGA
PGA
BGA
PGA
PGA
BGA
QFP
QFP
QFP
QFP
TYPE
HQ160
HQ208
HQ240
HQ304
PQ100
VQ100
PQ160
PQ208
PQ240
BG256
PG299
BG352
PG411
BG432
PG475
PG559
BG560
TQ144
TQ176
HT144
HT176
PC84
CODE
-3 CI CI CI
-2 CI CI CI
XC4002XL -1 CI CI CI
-09C C C C
-3 CI CI CI CI CI CI
-2 CI C CI CI CI CI
XC4005XL
-09C C C C C C C
-3 CI CI CI CI CI CI CI
XC4010XL -1 CI CI CI CI CI CI CI
-09C C C C C C C C
XC4013XL -1 CI CI CI CI CI CI
-09C C C C C C C
-08C C C C C C C
XC4020XL -1 CI CI CI CI CI CI
-09C C C C C C C
-09C C C C C C C C
-2 CI CI C CI CI CI CI
-09C C C C C C C C
-08C C C C C C C C
-09C C C C C C C C
-3 CI CI CI CI CI
-2 CI CI CI CI CI
XC4052XL -1 CI CI CI CI CI
-09C C C C C C
-3 CI CI CI CI CI
-2 CI CI CI CI CI
XC4062XL -1 CI CI CI CI CI
-09C C C C C C
-08C C C C C C
-3 CI CI CI
-2 CI CI CI
XC4085XL -1 CI CI CI
-09C C C C
1/29/99
C = Commercial TJ = 0° to +85°C
I= Industrial TJ = -40°C to +100°C
6-154 January 29, 1999 (Version 1.5)

R
Table 26: Component Availability Chart for XC4000E FPGAs
PINS 84 100 100 120 144 156 160 191 208 208 223 225 240 240 299 304
High-Perf.
High-Perf.
High-Perf.
Ceram.
Ceram.
Ceram.
Ceram.
Ceram.
PQFP
VQFP
PQFP
PQFP
PQFP
TQFP
PLCC
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
PGA
PGA
PGA
PGA
BGA
PGA
QFP
QFP
QF
TYPE
HQ208
HQ240
HQ304
PQ100
VQ100
PG120
PG156
PQ160
PG191
PQ208
PG223
BG225
PQ240
PG299
TQ144
PC84
CODE
-4 CI CI CI CI
-3 CI CI CI CI
XC4003E
-2 CI CI CI CI
-1 C C C C
XC4005E -2 CI CI CI CI CI CI
-1 C C C C C C
-4 CI CI CI CI CI
-3 CI CI CI CI CI
XC4006E -2 CI CI CI CI CI
-1 C C C C C
-4 CI CI CI CI
-3 CI CI CI CI
XC4008E -2 CI CI CI CI
-1 C C C C
XC4010E -2 CI CI CI CI CI CI 6
-1 C C C C C C
XC4013E -2 CI CI CI CI CI CI CI
-1 C C C C C C C
-4 CI CI CI
-3 CI CI CI
XC4020E -2 CI CI CI
-1 C C C
-4 CI CI CI CI
XC4025E -3 CI CI CI CI
-2 C C C C
1/29/99
Table 27: Component Availability Chart for XC4000EX FPGAs
PINS 208 240 299 304 352 411 432
High-Perf. High-Perf. Ceram. High-Perf. Plast. Ceram. Plast.

TYPE
QFP QFP PGA QFP BGA PGA BGA
CODE HQ208 HQ240 PG299 HQ304 BG352 PG411 BG432
-4 CI CI CI CI CI
XC4028EX -3 CI CI CI CI CI
-2 C C C C C
-4 CI CI CI CI CI
XC4036EX -3 CI CI CI CI CI
-2 C C C C C
1/29/99
January 29, 1999 (Version 1.5) 6-155

R
User I/O Per Package

Table 28 - Table 30 show the number of user I/Os available in each package for XC4000-Series devices. Call your local sales
office for the latest availability information, or see the Xilinx WEBLINX at http://www.xilinx.com for the latest revision of the
specifications.
Table 28: User I/O Chart for XC4000XL FPGAs
Maximum User Accessible I/O by Package Type
HQ160
HQ208
HQ240
HQ304
PQ100
VQ100
PQ160
PQ208
PQ240
BG256
PG299
BG352
PG411
BG432
PG475
PG559
BG560
TQ144
TQ176
HT144
HT176
PC84
Max
Device I/O
XC4002XL 64 61 64 64
XC4005XL 112 61 77 77 112 112 112
XC4010XL 160 61 77 113 129 145 160 160
XC4013XL 192 113 129 145 160 192 192
XC4020XL 224 113 129 145 160 193 205
XC4028XL 256 129 160 193 205 256 256 256
XC4036XL 288 129 160 193 256 288 288 288
XC4044XL 320 129 160 193 256 289 320 320
XC4052XL 352 193 256 352 352 352
XC4062XL 384 193 256 352 384 384
XC4085XL 448 352 448 448
1/29/99
Table 29: User I/O Chart for XC4000E FPGAs
Maximum User I/O by Package Type

HQ208
HQ240
HQ304
PQ100
VQ100
PG120
PG156
PQ160
PG191
PQ208
PG223
BG225
PQ240
PG299
TQ144
PC84
Max
Device I/O
XC4003E 80 61 77 77 80
XC4005E 112 61 77 112 112 112 112
XC4006E 128 61 113 125 128 128
XC4008E 144 61 129 144 144
XC4010E 160 61 129 160 160 160 160
XC4013E 192 129 160 160 192 192 192 192
XC4020E 224 160 192 193
XC4025E 256 192 193 256 256
1/29/99
Table 30: User I/O Chart for XC4000EX FPGAs
Max Maximum User Accessible I/O by Package Type

Device I/O HQ208 HQ240 PG299 HQ304 BG352 PG411 BG432
XC4028EX 256 160 193 256 256 256
XC4036EX 288 193 256 288 288 288
1/29/99
6-156 January 29, 1999 (Version 1.5)

R
Example: XC4013E-3HQ240C
Device Type
Temperature Range
Speed Grade C = Commercial (TJ = 0 to +85°C)
-6 I = Industrial (TJ = -40 to +100°C)
-5 M = Military (TC = -55 to+125°C)
-4
-3 Number of Pins
-2
-1
Package Type
PC = Plastic Lead Chip Carrier BG = Ball Grid Array
PQ = Plastic Quad Flat Pack PG = Ceramic Pin Grid Array
VQ = Very Thin Quad Flat Pack HQ = High Heat Dissipation Quad Flat Pack
TQ = Thin Quad Flat Pack MQ = Metal Quad Flat Pack
CB = Top Brazed Ceramic Quad Flat Pack
X9020
Revision Control
Version Description
March 30, 1998 (1.5) Updated XC4000XL timing and added XC4002XL
1/29/99 Updated pin diagrams.
January 29, 1999 (Version 1.5) 6-157

R
6-158 January 29, 1999 (Version 1.5)

0
XC4000XLA/XV Field
R
Table of Contents
0 6*
XC4000XLA/XV Field Programmable Gate Arrays

XC4000XLA/XV Family FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-161
XC4000XLA/XV Electrical Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-161
Additional XC4000XLA/XV Family Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-161
Xilinx SRAM XC4000 Series FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-162
XC4000XLA/XV FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-162
XC4000XV FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-162
Differences Between the XC4000XLA/XV and XC4000XL FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-162
IOB Enhancements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-162
Three-State Register. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-163
FastCLK Clock Buffers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-163
XLA, XV Power Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-163
XLA, XV Logic Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-163
Using Fast I/O CLKS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-164
JTAG Enhancements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-165
Differences between the XC4000XV and XC4000XLA FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-165
XC4000XLA Socket Compatibility with XL FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-165 6
XC4000XV Socket Compatibility with XL/XLA FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-165
XC4000XLA/XV I/O Signalling Standards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-167
VCC Clamping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-167
Overshoot and Undershoot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-167
Express Configuration Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-167
Pseudo Daisy Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-168
Serial PROM Recommendation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-171
XC4000XLA Product Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-173
XC4000XV Product Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-174
XC4000XLA D.C. and Switching Performance Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-175
XC4000XLA Clock Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . 6-177
XC4000XLA Global Low Skew Clock Input to Clock K . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-177
XC4000XLA FastCLK Input to IOB Clock Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-177
XC4000XLA Clock Buffer Switching Characteristic Guidelines (Cont) . . . . . . . . . . . . . . . . . . . . 6-178
XC4000XLA Global Early Clock Input to IOB Clock Input; BUFGE #s 1, 2, 5, and 6. . . . . . . . . 6-178
XC4000XLA Global Early Clock Input to IOB Clock Input; BUFGE #s 3, 4, 7, and 8. . . . . . . . . 6-178
XC4000XLA CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-179
XC4000XLA RAM Synchronous (Edge-Triggered) Write Operation Guidelines. . . . . . . . . . . . . 6-180
XC4000XLA CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines. . . . . . . . . 6-181
XC4000XLA CLB RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . . . . . . . 6-181
XC4000XLA Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-182
XC4000XLA Global Clock Input to Output Delay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-182
XC4000XLA FastCLK Input to Output Delay for BUFNW, BUFSW, BUFNE, & BUFSE . . . . . . 6-182
XC4000XLA Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-183
6-159
XC4000XLA/XV Field Programmable Gate Arrays Table of Contents
XC4000XLA Global Early Clock Input to Output Delay for BUFGE #s 1, 2, 5, and 6 . . . . . . . . . 6-183
XC4000XLA Global Early Clock Input to Output Delay for BUFGE #s 3, 4, 7, and 8 . . . . . . . . . 6-183
XC4000XLA Pin-to-Pin Input Parameter Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-185
XC4000XLA Global Low Skew Clock, Set-Up and Hold. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-185
XC4000XLA Pin-to-Pin Input Parameter Guidelines (Cont.). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-186
XC4000XLA FastCLK Input Set-Up and Hold for BUFNW, BUFSW, BUFNE, & BUFSE. . . . . . 6-186
XC4000XLA BUFGE #s 1, 2, 5, and 6 Global Early Clock, Set-up and Hold for IFF and FCL . . 6-187
XC4000XLA BUFGE #s 3, 4, 7, and 8 Global Early Clock, Set-up and Hold for IFF and FCL . . 6-188
XC4000XLA IOB Input Switching Characteristic Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-189
XC4000XLA IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 6-190
XC4000XV D.C. and Switching Performance Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-191
All specifications subject to change without notice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-191
XC4000XV Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 6-193
XC4000XV CLB Switching Characteristic Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-194
XC4000XV RAM Synchronous (Edge-Triggered) Write Operation Guidelines. . . . . . . . . . . . . . 6-195
XC4000XV CLB RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . . . . . . . . 6-196
XC4000XV Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-197
XC4000XV Global Clock Input to Output Delay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-197
XC4000XV FastCLK Input to Output Delay for BUFNW, BUFSW, BUFNE, and BUFSE. . . . . . 6-197
XC4000XV Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-198
XC4000XV Global Early Clock Input to Output Delay for BUFGE #s 1, 2, 5, and 6 . . . . . . . . . . 6-198
XC4000XV Global Early Clock Input to Output Delay for BUFGE #s 3, 4, 7, and 8 . . . . . . . . . . 6-198
XC4000XV Pin-to-Pin Input Parameter Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-199
XC4000XV Global Low Skew Clock, Set-Up and Hold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-199
XC4000XV FastCLK Input Set-Up and Hold for BUFNW, BUFSW, BUFNE, & BUFSE. . . . . . . 6-199
XC4000XV BUFGE #s 1, 2, 5, & 6 Global Early Clock, Set-Up and Hold for IFF and FCL . . . . 6-200
XC4000XV IOB Input Switching Characteristic Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-201
XC4000XV IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-202
XC4013XLA Pinout Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-203
XC4020XLA Pinout Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-207
XC40110XV Pinout Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-246
XC40150XV Pinout Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-255
6-160
0
R
XC4000XLA/XV
February 1, 1999 (Version 1.0) 0 6* Product Specification
XC4000XLA/XV Family FPGAs XC4000XLA/XV Electrical Features

Note: XC4000XLA devices are improved versions of • XLA devices require 3.0 - 3.6 Volts (VCC)
XC4000XL devices. The XC4000XV devices have the • XV devices require 2.3- 2.7 Volts (VCCINT)
same features as XLA devices, incorporate additional inter- and 3.0-3.6V (VCCIO)
connect resources and extend gate capacity to 500,000 • 5.0 V TTL compatible I/O
system gates. The XC4000XV devices require a separate • 3.3 V LVTTL, LVCMOS compliant I/O
2.5V power supply for internal logic but maintain 5V I/O • 5.0 V and 3.0 V PCI compliant I/O
compatibility via a separate 3.3V I/O power supply. A gen- • 12-mA or 24-mA current sink capability
eral description of the XC4000XLA/XV device architecture • Safe under all power-up sequences
in included in the XC4000E and XC4000X Series (Descrip- • XLA devices consume 40% less power than do XL
tion only) file, v1.4 (11/97) on the Xilinx WEBLINX at: devices
http://www.xilinx.com. • XV devices consume 65% less power than do XL
• System-featured Field-Programmable Gate Arrays devices
- Select-RAMTM memory: on-chip ultra-fast RAM with • Optional input clamping to VCC (XLA) or VCCIO (XV)
- Synchronous write option
- Dual-port RAM option
Additional XC4000XLA/XV Family 6
- Flexible function generators and abundant flip-flops Features
- Dedicated high-speed carry logic • Footprint Compatible with XC4000XL FPGAs Lower
- Internal 3-state bus capability cost with improved performance and lower power.
- 8 global low-skew clock or signal distribution • Advanced Technology — 5 layer metal, 0.25µ CMOS
networks process (XV) or 0.35µ CMOS process (XLA)
• Flexible array architecture • Highest Performance — System performance beyond
• Low power segmented routing architecture 100 MHz
• Systems-oriented features • High Capacity — Up to 500,000 system gates and
- IEEE 1149.1-compatible boundary scan 270,000 synchronous SRAM bits
- Individually programmable output slew rate • Low Power — 3.3V/2.5V technology plus segmented
- Programmable input pull-up or pull-down resistors routing architecture
- Unlimited reprogrammability • Safe and Easy to Use — Interfaces to any combination
• Readback capability of 3.3 V and 5.0 V TTL compatible devices.
- Program verification and internal node observability
Table 1: XC4000XLA Series Field Programmable Gate Arrays
*
Max Logic Max. RAM Typical Number Required

Logic Gates Bits Gate Range CLB Total of Max. Configur-
Device Cells (No RAM) (No Logic) (Logic and RAM)* Matrix CLBs Flip-Flops User I/O ation Bits
XC4013XLA 1,368 13,000 18,432 10,000 - 30,000 24 x 24 576 1,536 192 393,632
XC4020XLA 1,862 20,000 25,088 13,000 - 40,000 28 x 28 784 2,016 224 521,880
XC4028XLA 2,432 28,000 32,768 18,000 - 50,000 32 x 32 1,024 2,560 256 668,184
XC4036XLA 3,078 36,000 41,472 22,000 - 65,000 36 x 36 1,296 3,168 288 832,528
XC4044XLA 3,800 44,000 51,200 27,000 - 80,000 40 x 40 1,600 3,840 320 1,014,928
XC4052XLA 4,598 52,000 61,952 33,000 - 100,000 44 x 44 1,936 4,576 352 1,215,368
XC4062XLA 5,472 62,000 73,728 40,000 - 130,000 48 x 48 2,304 5,376 384 1,433,864
XC4085XLA 7,448 85,000 100,352 55,000 - 180,000 56 x 56 3,136 7,168 448 1,924,992
XC40110XV 9,728 110,000 131,072 75,000 - 235,000 64 x 64 4,096 9,216 448 2,686,136
XC40150XV 12,312 150,000 165,888 100,000 - 300,000 72 x 72 5,184 11,520 448 3,373,448
XC40200XV 16,758 200,000 225,792 130,000 - 400,000 84 x 84 7,056 15,456 448 4,551,056
XC40250XV 20,102 250,000 270,848 180,000 - 500,000 92 x 92 8,464 18,400 448 5,433,888
* Maximum values of gate range assume 20-30% of CLBs used as RAM
February 1, 1999 (Version 1.0) 6-161

R
Xilinx SRAM XC4000 Series FPGAs The XV devices also incorporate additional routing
resources in the form of 8 octal-length segmented routing
XC4000 Series high-performance, high-capacity Field Pro- channels vertically and horizontally per row and column.
grammable Gate Arrays (FPGAs) provide the benefits of
custom CMOS VLSI, while avoiding the initial cost, long Differences Between the
development cycle, and inherent risk of a conventional
masked gate array.
XC4000XLA/XV and XC4000XL
The result of fifteen years of FPGA design experience and
FPGAs
feedback from thousands of customers, these FPGAs com- The XC4000XLA/XV families of FPGAs are logically identi-
bine architectural versatility, increased speed, abundant cal to XC4000EX and XC4000XL FPGAs, however I/O,
routing resources, and new, sophisticated software to configuration logic, JTAG functionality, and performance
achieve fully automated implementation of complex, have been enhanced. In addition, they deliver:
high-density, high-performance designs. • Improved Performance
XLA/XV devices benefit from advance processing
technology and a reduction in interconnect capacitance
which improves performance over XL devices by more
than 30%.
• Lower Power
XLA/XV devices have reduced power requirements
compared to equivalent XL devices.
• Shorter routing delays
The smaller die of XLA/XV devices directly reduces
clock delays and the delay of high-fanout signals. The
reduction in clock delay allows improved pin-to-pin I/O
specifications.
• Lower Cost
XLA/XV device cost is directly related to the die size
and has been reduced significantly from that of
equivalent XL devices.
• Express mode configuration
Express mode configuration is available on the XLA and
Figure 1: Cross Section of Xilinx 0.25 micron, 5 layer XV devices.
metal XC4000XV FPGA. Visible features are five layers of
metallization, tungsten plug vias and trench isolation. The IOB Enhancements
small gaps above the lowest layer are 0.25 micron • 12/24 mA Output Drive
polysilicon MOSFET gates. The excellent planarity of each The XLA/XV family of FPGAs allow individual IOBs to
metal layer is due to the use of “chemical-mechanical be configured as high drive outputs. Each output can be
polishing” or CMP. In effect, each layer is ground flat before configured to have 24 mA drive strength as opposed to
a new layer is added. the standard default strength of 12 mA.
• VCC Clamping Diode
XLA and XV FPGAs have an optional clamping diode
XC4000XLA/XV FPGAs connected from each output to VCC (VCCIO for XV).
XC4000XLA/XV FPGAs use 5 layer metal silicon technol- When enabled they clamp ringing transients back to the
ogy to improve performance while reducing device cost and 3.3V supply rail. This clamping action is required in
power. In addition, IOB enhancements provide full PCI 3.3V PCI applications. VCC clamping is a global option
compliance and the JTAG functionality is expanded. affecting all I/O pins. If enabled, TTL I/O compatibility is
maintained, but full 5.0 Volt I/O tolerance is sacrificed.
XC4000XV FPGAs • Enhanced ESD protection
An improved ESD structure allows XV devices to safely
XC4000XV FPGAs incorporate all the features of the XLA
pass the stringent 5V PCI (4.2.1.3) ringing test. This
devices but require a separate 2.5V power supply for inter-
test applies an 11V pulse to each IOB for 11 ns via a 55
nal logic. I/O pads are still driven from a 3.3V power supply.
ohm resistor.
The 2.5V logic supply is named VCCINT and the 3.3 V IO
supply is named VCCIO.
6-162 February 1, 1999 (Version 1.0)

R
• Full 3.3V and 5.0V PCI compliance F = Frequency in MHz

The addition of 12/24 mA drive, optional 3.3V clamping
Table 2: K-Factor and Relative Power.
and improved ESD provides full compliance with either
3.3V or 5.0V PCI specifications. Power Power
Relative Relative
Three-State Register FPGA Family K-Factor To XL To XLA
XC4000XLA/XV devices incorporate an optional register XC4000XL 28 1.00 1.65
controlling the three-state enable in the IOBs.The use of XC4000XLA 17 0.60 1.00
the three-state control register can significantly improve XC4000XV 13 0.35 0.58
output enable and disable time.
XLA, XV Logic Performance
FastCLK Clock Buffers
XC4000XLA/XV devices feature 30% faster device speed
The XLA/XV devices incorporate FastCLK clock buffers. than XL devices, and consistent performance is achieved
Two FastCLK buffers are available on each of the right and across all family members. Table 3 illustrates the perfor-
left edges of the die. Each FastCLK buffer can provide a mance of the XLA devices. For details regarding the imple-
fast clock signal (typically < 1.5 ns clock delay) to all the mentation of these benchmarks refer to XBRF15 “Speed
IOBs within the IOB octant containing the buffer. The Fast- Metrics for High Performance FPGAs”.
CLK buffers can be instantiated by use of the BUFFCLK
Table 3: XLA/XV Estimated Benchmark Performance
symbols. (In addition to FastCLK buffers, the Global Early
BUFGE clock buffers #1, #2, #5, and #6 can also provide Register - Register Maximum
fast clock signals (typically < 1.5 ns clock delay) to IOBs on Size
Benchmarks Frequency
the top and bottom of the die. 8-Bit 172 MHz
XLA, XV Power Requirements Adder 16-Bit 144 MHz 6
32-Bit 108 MHz
XC4000XLA devices require 40% less power per CLB than
2 Cascaded Adders 16-Bit 94 MHz
equivalent XL devices. XC4000XV devices require 42%
less power per CLB than equivalent XLA devices and 65% 4 Cascaded Adders 16-Bit 57 MHz
less power than XL devices The representative K-Factor for 1 Level 314 MHz
the following families can be found in Table 2. The K-Factor 2 Level 193 MHz
Cascaded 4LUTs
predicts device current for typical user designs and is 4 Level 108 MHz
based on filling the FPGA with active 16-Bit counters and 6 Level 75 MHz
measuring the device current at 1 MHz. This technique is 1 CLBs 325 MHz
described in XBRF14 “A Simple Method of Estimating
4 CLBs 260 MHz
Power in XC4000XL/EX/E FPGAs”. To predict device Interconnect
power (P) using the K-Factor use the following formula: 16 CLBs 185 MHz
(Manhattan Distance)
64 CLBs 108 MHz
128 CLBs 81 MHz
P=V*K*N*F; where:
Dual Port RAM 8-Bits by 16 172 MHz
P= Device Power (Pipelined) 8-Bits by 256 172 MHz
V= Power supply voltage
K= the Device K-Factor
N = number of active registers

R
Using Fast I/O CLKS • BUFGE (I,O) - The Global Early Buffer
• BUFGLS (I,O)- The Global Low Skew Buffer
There are several issues associated with implementing fast • BUFFCLK (I,O) - The FastCLK Buffer
I/O clocks by using multiple FastCLK and BUFGE clock • ILFFX (D, GF, CE, C, Q) - The Fast Capture Latch
buffers for I/O transfers and a BUFGLS clock buffer for Macro
internal logic.
Locating I/O elements - It is necessary to connect these
Reduced Clock to Out Period - When transferring data elements to a particular I/O pad in order to select which
from a BUFGLS clocked register to an IOB output register buffer or fast capture latch will be used.
which is clocked with a fast I/O clock, the total amount of
time available for the transfer is reduced. Restricted Clock Loading - Because the input hold
requirement is a function of internal clock delay, it may be
Using Fast Capture Latch in IOB input - It is necessary to necessary to restrict the routing of BUFGE to IOBs along
transfer data captured with the fast I/O clock edge to a the top and bottom of the die to obtain sub-ns clock delays.
delayed BUFGLS clock without error. The use of the Fast
Capture Latch in the IOBs provides this functionality. BUFGE 1 BUFGE 6
Driving multiple clock inputs - Since each FastCLK input

can only reach one octant of IOBs it will usually be neces- FCLK 1 FCLK 4
sary to drive multiple FastCLK and BUFGE input pads with
a copy of the system clock. Xilinx recommends that sys-
tems which use multiple FastCLK and BUFGE input buffers
use a “Zero Delay” clock buffer such as the Cypress
CY2308 to drive up to 8 input pins. These devices contain a
Phase locked loop to eliminate clock delay, and specify less
FCLK 2 FCLK 3
than 250ps output jitter.
PCB layout - The recommended layout is to place the PLL BUFGLS 2
BUFGE 5
underneath the FPGA on the reverse side of the PCB. All 8 BUFGE 2
clock lines should be of equal length. This arrangement will
allow all the clock line to be less than 2 cm in length which
Figure 2: Location of FastCLK, BUFGE and BUFGLS
will generally eliminate the need for clock termination.
Clock Buffers in XC4000XLA/XV FPGAs
Advancing the FPGAs clock - An additional advantage to
using a PLL-equipped clock buffer is that it can advance the
FPGA clocks relative to the system clock by incorporating
additional board delay in the feedback path. Approximately PLL
6 inches of trace length are necessary to delay the signal Clock O0 BUFGE1 XC4000XLA
BUFGE2
by 1 ns. Buffer O1
O2
XC4000XV
BUFGE5
O3 BUFGE6
Advancing the FPGA’s clock directly reduces input hold O4 FCLK1
requirements and improves clock to out delay. FPGA clocks O5 FCLK2
SysClk FB O6 FCLK3
should not be advanced more than the guaranteed mini- Ref O7 FCLK4
mum Output Hold Time (minus any associated clock jitter)
or the outputs may change state before the system clock
edge. For XLA and XV FPGAs the Output Hold Time is
specified as a minimum Clock to Output Delay in the tables Figure 3: Diagram of XC4000XLA/XV FPGA
on pages 183, 182, 197, and 198. The maximum recom- Connected to PLL Clock Buffer Driving 4 BUFGE and
mended clock advance equals this value minus any clock 4 FastCLK Clock Buffers.
jitter.
Instantiating I/O elements- Depending on the design
environment, it may be necessary to instantiate the fast I/O
elements. They are found in the libraries as:

R
JTAG Enhancements • Bypass FF - Bypass FF and IOB is modified to provide

DRCLOCK only during BYPASS for the bypass flip-flop
XC4000XLA/XV devices have improved JTAG functionality and during EXTEST or SAMPLE/PRELOAD for the IOB
and performance in the following areas: register.
• IDCODE - The IDCODE register in JTAG is now
supported. All future Xilinx FPGAs will support the Differences between the XC4000XV
IDCODE register. By using the IDCODE, the device and XC4000XLA FPGAs
connected to the JTAG port can be determined. The
The high density of the XC4000XV family FPGAs is
use of the IDCODE enables selective configuration
achieved by using advanced 0.25 micron silicon technol-
dependent upon the FPGA found. The IDCODE register
ogy. A 2.5 Volt power supply (VCCINT) is necessary to pro-
has the following binary format:
vide the reduced supply voltage required by 0.25 micron
vvvv:ffff:fffa:aaaa:aaaa:cccc:cccc:ccc1
internal logic, however to maintain TTL compatibility a 3.3V
Where:
power supply (VCCIO) is required by the I/O.
c = the company code;
a = the array dimension in CLBs; To accommodate the higher gate capacity of XV devices,
f = the Family code; additional interconnect has been added. These differences
v = the die version number are detailed below.
Family Codes = 01 for XLA; • VCCINT (2.5 Volt) Power Supply Pins
= 02 for SpartanXL; The XV family of FPGAs requires a 2.5V power supply
for internal logic, which is named VCCINT. The pins
= 03 for Virtex;
assigned to the VCCINT supply are named in the pinout
= 07 for XV. guide for the XC4000XV FPGAs and in Table 5 on page
166.
Xilinx company code = 49 (hex) 6
• VCCIO (3.3 Volt) Power Supply Pins
Table 4: IDCODEs assigned to XC4000XLA/XV FPGAs Both the XV and XLA FPGAs use a 3.3V power supply
to power the I/O pins. The I/O supply is named VCCIO
FPGA IDCODE
in the XV family.
XC4013XLA 0x00218093
• Octal-Length Interconnect Channels
XC4020XLA 0x0021c093 The XC40110XV, XC40150XV, XC40200XV, and
XC4028XLA 0x00220093 XC40250XV have enhanced routing. Eight routing
XC4036XLA 0x00224093 channels of octal length have been added to each CLB
XC4044XLA 0x00228093 in both vertical and horizontal dimensions.
XC4052XLA 0x0022c093
XC4062XLA 0x00230093
XC4000XLA Socket Compatibility
XC4085XLA 0x00238093 with XL FPGAs
XC40110XV 0x00e40093 The XC4000XLA devices are generally available in the
XC40150XV 0x00e48093 same packages as equivalent XL devices, however the
XC40200XV 0x00e54093 range of packages available for the XC4085XLA has been
extended to include smaller packages such as the HQ240.
XC40250XV 0x00e5c093
• Configuration State - The configuration state is XC4000XV Socket Compatibility

available to JTAG controllers. with XL/XLA FPGAs
• Configure Disable - The JTAG port can be prevented
XC4000XV devices are available in five package options,
from reconfiguring the FPGA
pin-grid PG599 and ball-grid BG560, BG432, and BG352
• TCK Startup - TCK can now be used to clock the
and quad-flatpack HQ240. With the exception of the
start-up block in addition to other user clocks.
VCCINT power pins, XC4000XV FPGAs are compatible
• CCLK holdoff - Changed the requirement for Boundary
with XL and XLA devices in these packages if the following
Scan Configure or EXTEST to be issued prior to the
guidelines are followed:
release of INIT pin and CCLK cycling.
• Reissue configure - The Boundary Scan Configure • Lay out the PCB for the XV pinout.
can be reissued to recover from an unfinished attempt • When an XL or XLA device is installed disconnect the
to configure the device. VCCINT (2.5 V) supply. For the PG599, VCCINT should
be connected to 3.3V. For BG560, BG432 and BG352
and HQ240 packages, the VCCINT voltage source
should be left unconnected. The unused I/O pins in the

R
XL/XLA devices connected to VCCINT will be pulled up Table 5: VCCINT (2.5 V) Pins in XV Packages
to 3.3V. Care must be taken to insure that these pins
are not driven when the XL/XLA device is operative. HQ240 BG352 BG432 BG560 PG559
• When an XC4000XV is installed, the VCCINT pins must P198 D10 A10 E12 H12
be connected to a 2.5V power supply. P185 D5 AB2 AD2 H18
The differences between the XL and XV packages are P164 K4 AB30 AD32 H26
detailed below: P154 N3 AG28 AK31 H32
P137 W2 AH15 AM17 M8
PG559 - XLA and XL devices in the PG599 package have
56 VCC pins.The XC4000XV devices allocate 16 of these P116 AE3 AH5 AK5 M36
I/O pins to VCCINT (2.5V). P104 AC10 AJ10 AK11 V8
BG560 - XLA and XL devices in the BG560 package have P93 AC13 AK22 AN25 V36
448 I/O pins.The XC4000XV devices allocate 16 of these P77 AE19 B23 C24 AF8
I/O pins to VCCINT (2.5V). P55 AB24 B4 D6 AF36
BG432- XLA and XL devices in the BG432 package have P43 V24 C16 C17 AM8
352 I/O pins. The XC4000XV devices allocate 16 of these P27 N24 E28 E30 AM36
I/O pins to VCCINT (2.5V). P16 J24 K29 K32 AT12
BG352 - XLA and XL devices in the BG352 package have P4 D24 K3 J1 AT18
289 I/O pins.The XC4000XV devices allocate 15 of these P225 A20 R2 T3 AT26
I/O pins to VCCINT (2.5V). - - R29 U32 AT32
HQ240- XLA and XL devices in the HQ240 package have
193 I/O pins.The XC4000XV devices allocate 15 of these
I/O pins to VCCINT (2.5V).

R
XC4000XLA/XV I/O Signalling XLA/XV devices maintain LVTTL I/O compatibility when
VCC clamping is enabled, however full 5.0V TTL I/O com-
Standards patibility is sacrificed.
XLA and XV devices are compatible with TTL, LVTTL, PCI
3V, PCI 5V and LVCMOS signalling. The various standards Overshoot and Undershoot
are illustrated in Table 6 and the signaling environment is Ringing wave forms are allowed on XLA/XV inputs as long
illustrated in Figure 4. as undershoot is limited to -2.0V and overshoot is limited to
+7.0V and current is limited to 100 mA for less than 10 ns.
VCC Clamping If VCC clamping is enabled then overshoot will begin to be
XLA/XV devices are fully 5V TTL I/O compatible if VCC clamped at VCC/VCCIO plus one diode voltage drop and
clamping is not enabled. The I/O pins can withstand input undershoot will be clamped to ground minus one diode volt-
voltages up to 7V. With VCC clamping enabled, the XLA/XV age drop. In either case the current must be limited to 100
devices will begin to clamp input voltages to one diode volt- mA per pin for less than 10 ns.
age drop above VCC. In both cases negative voltage is
clamped to one diode voltage drop below ground.
Table 6: I/O Standards supported by XC4000XLA and XV FPGAs
Signaling VCC
Standard Clamping Output Drive VIH_MAX VIH MIN VIL MAX VOH MIN VOL MAX
TTL Not allowed 12/24 mA 5.5 2.0 0.8 2.4 0.4
LVTTL OK 12/24 mA 3.6 2.0 0.8 2.4 0.4
6
PCI5V Not allowed 24 mA 5.5 2.0 0.8 2.4 0.4
PCI3V Required 12 mA 3.6 50% of 30% of 90% of 10% of
VCC/VCCIO VCC/VCCIO VCC/VCCIO VCC/VCCIO
LVCMOS 3V OK 12/24 mA 3.6 50% of 30% of 90% of 10% of
VCC/VCCIO VCC/VCCIO VCC/VCCIO VCC/VCCIO
5.0 V Power
3.3 V Power
2.5 V Power
VCC (5 V) VCCIO VCCINT VCC (3.3 V)
TTL LVTTL
5 Volt Device XC4000XV 3.3 Volt Device
LVTTL
Ground
X7147
Figure 4: The Signalling Environment for XLA/XV FPGAS. For XLA devices the VCCIO and VCCINT supplies are
replaced by a single 3.3 Volt VCC supply, however, all indicated I/O signalling is still supported.
Express Configuration Mode

Express configuration mode is similar to Slave Serial con- loaded directly into the configuration data shift registers
figuration mode, except that data is processed one byte per (Figure 5). A CCLK frequency of 10 MHz is equivalent to a
CCLK cycle instead of one bit per CCLK cycle. An external 80 MHz serial rate, because eight bits of configuration data
source is used to drive CCLK, while byte-wide data is are loaded per CCLK cycle. Express mode does not sup-

R
port CRC error checking, but does support constant-field vents all devices in the chain from going High until the last
error checking. A length count is not used in Express mode. device in the chain has completed its configuration cycle. If
the DONE pin of a device is left unconnected, the device
Express mode must be specified as an option to the BitGen
program, which generates the bitstream. The Express becomes active as soon as that device has been config-
mode bitstream is not compatible with the other configura- ured.
tion modes. Express mode is selected by a <010> on the Table 7: Pin Functions During Configuration
mode pins (M2, M1, M0). (4000XLA/XV Express mode only)
The first byte of parallel configuration data must be avail- CONFIGURATION MODE USER
able at the D inputs of the FPGA a short setup time before <M2:M1:M0> OPERATION
the second rising CCLK edge. Subsequent data bytes are EXPRESS MODE
PIN FUNCTION
clocked in on each consecutive rising CCLK edge <0:1:0>
(Figure 6). M2(LOW) (I) M2
M1(HIGH) (I) M1
Pseudo Daisy Chain M0(LOW) (I) M0
HDC (HIGH) I/O
As illustrated in Figures 5 and 6, multiple devices with dif- LDC (LOW) I/O
ferent configurations can be configured in a pseudo daisy INIT I/O
chain provided that all of the devices are in Express mode. DONE DONE
A single combined byte-wide data stream is used to config- PROGRAM (I) PROGRAM
ure the chain of Express mode devices. CCLK pins are tied CCLK (I) CCLK (I)
together and D0-D7 pins are tied together as a data buss DATA 7 (I) I/O
for all devices along the chain. A status signal is passed DATA 6 (I) I/O
from DOUT of each device to the CS1 input of the device DATA 5 (I) I/O
which follows it in the chain. Frame data is accepted only DATA 4 (I) I/O
when CS1 is High and the device’s configuration memory is DATA 3 (I) I/O
not already full. The lead device in the chain has its CS1 DATA 2 (I) I/O
input tied High (or floating, since there is an internal pullup). DATA 1 (I) I/O
The status pin DOUT is initially High for all devices in the DATA 0 (I) I/O
chain until the data stream header of seven bytes is loaded. DOUT SGCK4-I/O
This allows header data to be loaded into all devices in the TDI TDI-I/O
TCK TCK-I/O
chain simultaneously. After the header is loaded in all
TMS TMS-I/O
devices, their DOUT pins are pulled Low disabling configu-
TDO TDO-(O)
ration of all devices in the chain except the first device. As
CS1 I/O
each device in the chain is filled, its DOUT goes High driv-
Notes 1. A shaded table cell represents the internal
ing High the CS1 input of the next device, thereby enabling pull-up used before and during
configuration of the next device in the pseudo daisy chain. configuration.
2. (I) represents an input; (O) represents an
The requirement that all DONE pins in a daisy chain be output.
wired together applies only to Express mode, and only if all 3. INIT is an open-drain output during
devices in the chain are to become active simultaneously. configuration.
All 4000XLA/XV devices in Express mode are synchro- Because only XC4000XLA/XV, SpartanXL, and XC5200
nized to the DONE pin. User I/O for each device becomes devices support Express mode, only these devices can be
active after the DONE pin for that device goes High (The used to form an Express mode pseudo daisy chain.
exact timing is determined by BitGen options.)
Since the DONE pin is open-drain and does not drive a
High value, tying the DONE pins of all devices together pre-

R
VCC
To Additional
M0 M1 M2 M0 M1 M2 Optional
Daisy-Chained
Devices
CS1 DOUT CS1 DOUT
8 8
Optional
VCC Daisy-Chained
4000XLA/XV 4000XLA/XV
4.7K
CCLK CCLK
To Additional
Optional
Daisy-Chained
6
CCLK
Devices
99010800
Table 8: Express Mode Programming Switching Characteristic

INIT (High) setup time TIC 5 µs
D0 - D7 setup time TDC 20 ns
D0 - D7 hold time TCD 0 ns
CCLK
Preliminary

R
CCLK
1 TIC
INIT
TCD 3
2 T
DC
BYTE BYTE BYTE BYTE BYTE BYTE BYTE

D0-D7 5 6
0 1 2 3 4
Header
DOUT
Header Loaded FPGA Filled
CS1
99010700
Note: CS1 must remain High throughout loading of the configuration data stream. In the pseudo daisy chain of Figure 5, the 7 byte
data stream header is loaded into all devices simultaneously. Each device’s data frames are then loaded in turn when its
CS1 pin is driven High by the DOUT of the preceding device in the chain.
Figure 6: Express Mode Programming Switching Waveforms
Data Stream Format Table 9: 4000XLA/XV Express Mode Data Stream

Format
The data stream (“bitstream”) format is identical for all
serial configuration modes, but different for the Express Mode
4000XLA/XV Express mode. In Express mode, the device Data Type (D0-D7)
becomes active when DONE goes High, therefore no (4000XLA only)
length count is required. Additionally, CRC error checking is Fill Byte FFFFh
not supported in Express mode. The data stream format is Preamble Code 11110010b
shown in Table 9. Express mode data is shown with D0 at
Fill Byte FFFFFFh
the left and D7 at the right.
End-of-Header 11010010b
The configuration data stream begins with two bytes of Field Check Byte
eight ones each, a preamble code of one byte, followed by Start Field 11111110b
three bytes of eight ones each, and finally an end-of-
Data Frame DATA(n-1:0)
header field check byte. This header of seven bytes is fol-
lowed by the actual configuration data in frames. The End-of-Frame 11010010b
length and number of frames depends on the device type. Field Check Byte
Each frame begins with a start field and ends with an Extend Write Cycle FFD2FFFFFFh
end-of-frame field check byte. In all cases, additional Start-Up Bytes FFFFFFFFFFFFh
start-up bytes of data are required to provide six, or more,
LEGEND:
clocks for the start-up sequence at the end of configuration.
Long daisy chains require additional startup bytes to shift Unshaded Once per data stream
the last data through the chain. All startup bytes are Light Once per data frame
don’t-cares; these bytes are not included in bitstreams cre-
ated by the Xilinx software. Detection of an error results in the suspension of data load-
ing and the pulling down of the INIT pin. The user must
detect INIT and initialize a new configuration by pulsing the
by the bitstream generation software. The 4000XLA
PROGRAM pin Low or cycling VCC.
Express mode only supports non-CRC error checking. The
non-CRC error checking tests for a designated
end-of-frame field check byte for each frame. non-CRC
error checking tests for a designated end-of-frame field
check byte for each frame.

R
Serial PROM Recommendation

Table 10 shows the physical characteristics of each FPGA family member and the recommended Xilinx Serial PROM
recommended for use as configuration storage.
Table 10: Physical Characteristics and Recommended Serial PROM
Number Max. RAM Required

Max. CLB Total Logic
Device of Bits Configur- Serial PROM
User I/O Matrix CLBs Cells
Flip-Flops (No Logic) ation Bits
XC4013XLA 192 24 x 24 576 1,368 1,536 18,432 393,632 XC17512L
XC4020XLA 224 28 x 28 784 1,862 2,016 25,088 521,880 XC17512L
XC4028XLA 256 32 x 32 1,024 2,432 2,560 32,768 668,184 XC1701L
XC4036XLA 288 36 x 36 1,296 3,078 3,168 41,472 832,528 XC1701L
XC4044XLA 320 40 x 40 1,600 3,800 3,840 51,200 1,014,928 XC1701L
XC4052XLA 352 44 x 44 1,936 4,598 4,576 61,952 1,215,368 XC1702L
XC4062XLA 384 48 x 48 2,304 5,472 5,376 73,728 1,433,864 XC1702L
XC4085XLA 448 56 x 56 3,136 7,448 7,168 100,352 1,924,992 XC1702L
XC40110XV 448 64 x 64 4,096 9,728 9,216 131,072 2,686,136 XC1704L
XC40150XV 448 72 x 72 5,184 12,312 11,520 165,888 3,373,448 XC1704L
XC40200XV 448 84 x 84 7,056 16,758 15,456 225,792 4,551,056 XC1704L+XC17512L
XC40250XV 448 92 x 92 8,464 20,102 18,400 270,848 5,433,888 XC1704L+XC1702L

R

Table 11 shows the number of user I/Os available in each package for XC4000XLA/XV-Series devices. Call your local sales
office for the latest availability information.
Table 11: User I/O Pins Available by Device and Package
Maximum I/O Acessible per Package
HQ160
HQ208
HQ240
HQ304
PQ160
PQ208
PQ240
BG256
BG352
BG432
PG559
BG560
Max
Device I/O
XC4013XLA 192 129 160 192 192
XC4020XLA 224 129 160 193 205
XC4028XLA 256 129 160 193 205 256 256
XC4036XLA 288 129 160 193 256 288 288
XC4044XLA 320 129 160 193 256 289 320
XC4052XLA 352 129 160 193 256 289 352 352
XC4062XLA 384 129 160 193 256 289 352 384
XC4085XLA 448 129 160 193 256 289 352 448
XC40110XV 448 178 274 336 432
XC40150XV 448 178 274 336 448 432
XC40200XV 448 336 432
XC40250XV 448 336 448 432

R
XC4000XLA Product Availability
Table 12 shows the planned packages and speed grades for XC4000XLA-Series devices. Call your local sales office for the
latest availability information, or see the Xilinx WEBLINX at http://www.xilinx.com for the latest revision of the specifications.
Table 12: Component Availability Chart for XC4000XLA FPGAs
PINS 84 100 100 144 144 160 160 176 176 208 208 240 240 256 299 304 352 411 432 475 559 560
High-Perf.
High-Perf.
High-Perf.
High-Perf.
High-Perf.
High-Perf.
Ceram.
Ceram.
Ceram.
Ceram.
PQFP
VQFP
PQFP
PQFP
PQFP
TQFP
TQFP
TQFP
TQFP
PLCC
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
BGA
PGA
BGA
PGA
BGA
PGA
PGA
BGA
QFP
QFP
QFP
QFP
TYPE
HQ160
HQ208
HQ240
HQ304
PQ100
VQ100
PQ160
PQ208
PQ240
BG256
PG299
BG352
PG411
BG432
PG475
PG559
BG560
TQ144
TQ176
HT144
HT176
PC84
CODE
-09 CI CI CI CI
XC4013XLA -08 CI CI CI CI
-07 C C C C
-09 CI CI CI CI
XC4020XLA -08 CI CI CI CI
-07 C C C C
-09 CI CI CI CI CI
XC4028XLA -08 CI CI CI CI CI
-07 C C C C C
-09 CI CI CI CI CI
XC4036XLA -08 CI CI CI CI CI 6
-07 C C C C C
XC4044XLA -08 CI CI CI CI CI CI
-07 C C C C C C
XC4052XLA -08 CI CI CI CI CI CI CI
-07 C C C C C C C
-07 C C C C C C C
-07 C C C C C C C
1/25/99

R
XC4000XV Product Availability

Figure 13 shows the planned packages and speed grades for XC4000XV-Series devices. Call your local sales office for the
latest availability information, or see the Xilinx WEBLINX at http://www.xilinx.com for the latest revision of the specifications.
Table 13: Component Availability Chart for XC4000XV FPGAs
PINS 84 100 100 144 144 160 160 176 176 208 208 240 240 256 299 304 352 411 432 475 559 560
High-Perf.
High-Perf.
High-Perf.
High-Perf.
High-Perf.
High-Perf.
Ceram.
Ceram.
Ceram.
Ceram.
PQFP
VQFP
PQFP
PQFP
PQFP
PLCC
TQFP
TQFP
TQFP
TQFP
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
BGA
PGA
BGA
PGA
BGA
PGA
PGA
BGA
QFP
QFP
QFP
QFP
TYPE
HQ160
HQ208
HQ240
HQ304
PQ100
VQ100
PQ160
PQ208
PQ240
BG256
PG299
BG352
PG411
BG432
PG475
PG559
BG560
TQ144
TQ176
HT144
HT176
PC84
CODE
-09 CI CI CI CI
XC40110XV -08 CI CI CI CI
-07 C C C C
-09 CI CI CI CI CI
XC40150XV -08 CI CI CI CI CI
-07 C C C C C
-09 CI CI
XC40200XV -08 CI CI
-07 C C
-09 CI CI CI
XC40250XV -08 CI CI CI
-07 C C C
11/2498

R
XC4000XLA D.C. and Switching Performance Characteristics

Definition of Terms
follows:
Occasionally, values are of mixed classification. These are highlighted in bold face and the tables involved include a note
explaining the nature of the bold face entries. All specifications are subject to change without notice.
Except for pin-to-pin input and output parameters, the AC parameter delay specifications included in this document are

Symbol Description Values Units
6
TJ
Notes: 1. Maximum DC overshoot or undershoot above Vcc or below GND must be limited to either 0.5 V or 10 mA, whichever is
easier to achieve. During transitions, the device pins may undershoot to -2.0 V or overshoot to + 7.0 V, provided this over-
or undershoot lasts less than 10 ns and with the forcing current being limited to 200 mA.
2. Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are

VCC
Note: At junction temperatures above those listed as Operating Conditions, all delay parameters increase by 0.35% per °C.

R

VOH
Input capacitance (sample tested) BGA, SBGA, PQ, HQ, MQ packages 10 pF
CIN
PGA packages 16 pF
Notes: 1. With up to 64 pins simultaneously sinking 24 mA
2. With no output current loads, no active input or Longline pull-up resistors, all I/O pins Tri-stated and floating

R
XC4000XLA Clock Buffer Switching Characteristic Guidelines

temperature. Values apply to all XC4000XLA devices and expressed in nanoseconds unless otherwise noted.
XC4000XLA Global Low Skew Clock Input to Clock K

Speed Grade All -09 -08 -07
Units
Description Symbol Device Min Max Max Max
Delay from pad through Global Low Skew (GLS) clock TGLS XC4013XLA 2.4 2.1 1.9 ns
buffer to any clock input, K. XC4020XLA 2.6 2.3 2.1 ns
XC4028XLA 2.9 2.6 2.3 ns
XC4036XLA 3.2 2.8 2.5 ns
XC4044XLA 3.6 3.1 2.8 ns
XC4052XLA 3.9 3.4 3.1 ns
6
XC4062XLA 4.2 3.7 3.3 ns
XC4085XLA 5.0 4.4 3.9 ns
Preliminary
XC4000XLA FastCLK Input to IOB Clock Input

Units
Delay from pad through FastCLK buffer to any IOB clock TFCLK XC4013XLA 1.5 1.3 1.1 ns
input. XC4020XLA 1.5 1.3 1.2 ns
XC4028XLA 1.6 1.4 1.3 ns
XC4036XLA 1.7 1.5 1.4 ns
XC4044XLA 1.8 1.6 1.4 ns
XC4052XLA 1.9 1.7 1.5 ns
XC4062XLA 2.0 1.8 1.6 ns
XC4085XLA 2.3 2.0 1.8 ns
Note: Values in bold face are preliminary, all other values are advance. Preliminary

R
XC4000XLA Clock Buffer Switching Characteristic Guidelines (Cont)

XC4000XLA Global Early Clock Input to IOB Clock Input; BUFGE #s 1, 2, 5, and 6.
Units
Delay from pad through Global Early (GE) clock buffer to TGE XC4013XLA 1.7 1.5 1.3 ns
any IOB clock input for BUFGE #1, 2, 5, and 6. XC4020XLA 1.9 1.7 1.5 ns
XC4028XLA 2.1 1.9 1.7 ns
XC4036XLA 2.4 2.2 1.9 ns
XC4044XLA 2.7 2.4 2.2 ns
XC4052XLA 3.0 2.7 2.4 ns
XC4062XLA 3.3 3.0 2.7 ns
XC4085XLA 3.7 3.3 3.0 ns
Preliminary
XC4000XLA Global Early Clock Input to IOB Clock Input; BUFGE #s 3, 4, 7, and 8.
Units
Delay from pad through Global Early (GE) clock buffer to TGE XC4013XLA 2.4 2.1 1.9 ns
any IOB clock input for BUFGE 3, 4, 7, and 8. XC4020XLA 2.6 2.3 2.1 ns
XC4028XLA 2.9 2.6 2.3 ns
XC4036XLA 3.2 2.8 2.5 ns
XC4044XLA 3.6 3.1 2.8 ns
XC4052XLA 3.9 3.4 3.1 ns
XC4062XLA 4.2 3.7 3.3 ns
XC4085XLA 5.0 4.4 3.9 ns
Preliminary

R
XC4000XLA CLB Switching Characteristic Guidelines

XC4000XLA devices and expressed in nanoseconds unless otherwise noted.
Units
C inputs via SR/H0 via H to X/Y outputs THH0O 1.7 1.6 1.4 ns
C inputs via H1 via H to X/Y outputs THH1O 1.6 1.4 1.3 ns
C inputs via DIN/H2 via H to X/Y outputs THH2O 1.7 1.6 1.4 ns
C inputs via EC, DIN/H2 to YQ, XQ output (bypass) TCBYP 1.1 1.0 0.9 ns
Sequential Delays 6
F/G inputs via H TIHCK 1.4 1.3 1.2 ns
C inputs via H0 through H THH0CK 1.3 1.2 1.1 ns
CIN input via F/G TCCK 1.2 1.1 1.0 ns
CIN input via F/G and H TCHCK 2.0 1.7 1.6 ns
All Hold Times 0.0 0.0 0.0 ns
Clock
Set/Reset Direct
Global Set/Reset
Toggle Frequency (MHz) (for export control) FTOG 227 263 294 MHz
Preliminary

R
XC4000XLA RAM Synchronous (Edge-Triggered) Write Operation Guidelines

XC4000XLA devices and are expressed in nanoseconds unless otherwise noted.

Write Operation
32x1 TWCTS 6.7 5.9 5.3 ns
32x1 TWPTS 3.4 3.0 2.7 ns
32x1 TASTS 1.5 1.3 1.2 ns
Address hold time after clock K 16x2 TAHS 0.0 0.0 0.0 ns
32x1 TAHTS 0.0 0.0 0.0 ns
32x1 TDSTS 2.11 1.6 1.5 ns
.8
DIN hold time after clock K 16x2 TDHS 0.0 0.0 0.0 ns
32x1 TDHTS 0.0 0.0 0.0 ns
32x1 TWSTS 1.3 1.2 1.1 ns
WE hold time after clock K 16x2 TWHS 0.0 0.0 0.0 ns
32x1 TWHTS 0.0 0.0 0.0 ns
32x1 TWOTS 5.8 5.2 4.7 ns
Read Operation
32x1 TRCT 3.8 3.8 3.8 ns
Data Valid after address change (no Write En- 16x2 TILO 1.1 1.0 0.9 ns
able) 32x1 TIHO 1.9 1.7 1.5 ns
32x1 TIHCK 1.4 1.3 1.2 ns
Advance

R
XC4000XLA CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines

XC4000XLA devices and are expressed in nanoseconds unless otherwise noted.
Speed Grade --09 -08 -07

Dual Port RAM Units
Address hold time after clock K 16x1 TAHDS 0.0 0.0 0.0 ns
DIN hold time after clock K 16x1 TDHDS 0.0 0.0 0.0 ns
WE hold time after clock K 16x1 TWHDS 0.0 0.0 0.0 ns
Preliminary
6
XC4000XLA CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS I1
WCLK (K)
ILD
TWSS TWHS IPAD D Q I2
G
WE
IK
TDSS TDHS
Internal Write Performed
DATA IN
IK
TASS TAHS
ADDRESS
IPAD, I1 Address Data
TILO TILO
TWOS
I2 Address
DATA OUT OLD NEW
Address Latched
X6461 X6774
Single Port RAM Dual Port RAM

R
XC4000XLA Pin-to-Pin Output Parameter Guidelines

XC4000XLA Global Clock Input to Output Delay

Units
Global Low Skew (GLS) Clock Input to Output Delay us- TICKOF XC4013XLA 5.6 5.0 4.5 ns
ing Output Flip-Flop XC4020XLA 5.8 5.2 4.7 ns
XC4028XLA 6.1 5.5 4.9 ns
XC4036XLA 6.4 5.7 5.1 ns
XC4044XLA 6.8 6.0 5.4 ns
XC4052XLA 7.1 6.3 5.7 ns
XC4062XLA 7.4 6.6 5.9 ns
XC4085XLA 8.2 7.3 6.5 ns
For output SLOW option add TSLOW All Devices 1.7 1.6 1.4 ns
Preliminary
Notes: Listed above are representative values where one global clock input drives one vertical clock line in each accessible column,
Note: Values in bold face are preliminary, all other values are advance.
XC4000XLA FastCLK Input to Output Delay for BUFNW, BUFSW, BUFNE, & BUFSE
Units
FastCLK Input to Output Delay using Output Flip-Flop TICKFOF XC4013XLA 4.6 4.1 3.7 ns
for FastCLK buffers BUFNW, BUFSW, BUFNE, and XC4020XLA 4.7 4.2 3.7 ns
BUFSE.
XC4028XLA 4.8 4.3 3.8 ns
XC4036XLA 4.9 4.4 3.9 ns
XC4044XLA 5.0 4.4 4.0 ns
XC4052XLA 5.1 4.5 4.1 ns
XC4062XLA 5.2 4.6 4.1 ns
XC4085XLA 5.4 4.8 4.3 ns
Preliminary
Note: Values in bold face are preliminary, all other values are advance.

R
XC4000XLA Pin-to-Pin Output Parameter Guidelines

XC4000XLA Global Early Clock Input to Output Delay for BUFGE #s 1, 2, 5, and 6
Units
Global Clock Signal Input to Output Delay using TICKEOF XC4013XLA 4.9 4.4 3.9 ns
Global Early (GE) clock buffer to clock Output XC4020XLA 5.1 4.6 4.1 ns
Flip-Flop for BUFGE #s 1, 2, 5, & 6. XC4028XLA 5.3 4.8 4.3 ns
XC4036XLA 5.6 5.1 4.5 ns
XC4044XLA 5.9 5.3 4.8 ns
XC4052XLA 6.2 5.6 5.0 ns
XC4062XLA 6.5 5.9 5.3 ns
XC4085XLA 6.9 6.2 5.6 ns 6
Preliminary
XC4000XLA Global Early Clock Input to Output Delay for BUFGE #s 3, 4, 7, and 8
Units
Global Clock Signal Input to Output Delay using TICKEOF XC4013XLA 5.7 5.1 4.5 ns
Global Early (GE) clock buffer to clock Output XC4020XLA 5.9 5.3 4.7 ns
Flip-Flop for BUFGE #s 3, 4, 7, & 8. XC4028XLA 6.1 5.4 4.9 ns
XC4036XLA 6.3 5.6 5.0 ns
XC4044XLA 6.5 5.8 5.2 ns
XC4052XLA 6.8 6.0 5.4 ns
XC4062XLA 7.0 6.3 5.6 ns
XC4085XLA 7.5 6.7 6.0 ns
Preliminary

R

2
Delta Delay (ns)

1
delay. -1
Figure 7 is usable over the specified operating conditions of
voltage and temperature and is independent of the output -2
0 20 40 60 80 100 120 140
slew rate control.
Capacitance (pF)
X8257

R
XC4000XLA Pin-to-Pin Input Parameter Guidelines

XC4000XLA Global Low Skew Clock, Set-Up and Hold

Units
Input Setup and Hold Time Relative to Global Clock Input Signal
No Delay XC4013XLA 1.0 / 3.0 0.8 / 2.6 0.2 / 2.5 ns
Global Low Skew Clock and IFF TPSN/TPHN XC4020XLA 0.9 / 3.2 0.7 / 2.9 0.1 / 2.7 ns
XC4028XLA 0.8 / 3.8 0.6 / 3.3 0.0 / 3.0 ns
XC4036XLA 0.6 / 4.0 0.4 / 3.5 0.0 / 3.3 ns
XC4044XLA 0.4 / 4.4 0.2 / 3.9 0.0 / 3.6 ns
XC4052XLA 0.3 / 4.6 0.2 / 4.1 0.0 / 3.9 ns
XC4062XLA 0.2 / 5.0 0.1 / 4.5 0.0 / 4.2 ns
XC4085XLA 0.0 / 5.4 0.0 / 4.8 0.0 / 4.5 ns
Partial Delay XC4013XLA 4.4 / 0.5 4.1 / 0.3 3.7/ 0.0 ns 6
Global Low Skew Clock and IFF TPSP/TPHP XC4020XLA 4.5 / 0.6 4.1 / 0.3 3.7 / 0.0 ns
XC4028XLA 4.6 / 0.7 4.2 / 0.4 3.7/ 0.0 ns
XC4036XLA 4.6 / 0.8 4.2 / 0.4 3.7/ 0.0 ns
XC4044XLA 4.7 / 0.9 4.3 / 0.5 3.8 / 0.0 ns
XC4052XLA 4.8 / 1.0 4.3 / 0.6 3.8 / 0.2 ns
XC4062XLA 5.0 / 1.0 4.4 / 0.7 3.8 / 0.4 ns
XC4085XLA 5.5 / 1.2 4.7 / 0.9 3.8 / 0.5 ns
Full Delay XC4013XLA 4.4 / 0.0 4.1 / 0.0 3.7 / 0.0 ns
Global Low Skew Clock and IFF TPSD/TPHD XC4020XLA 4.6 / 0.0 4.2 / 0.0 3.8 / 0.0 ns
XC4028XLA 4.8 / 0.0 4.4 / 0.0 3.9 / 0.0 ns
XC4036XLA 4.9 / 0.0 4.5 / 0.0 4.0 / 0.0 ns
XC4044XLA 5.0 / 0.0 4.6 / 0.0 4.1 / 0.0 ns
XC4052XLA 5.2 / 0.0 4.7 / 0.0 4.2 / 0.0 ns
XC4062XLA 5.5 / 0.0 4.9 / 0.0 4.3 / 0.0 ns
XC4085XLA 6.0 / 0.0 5.2/ 0.0 4.4 / 0.0 ns
Preliminary
Notes: Setup time is measured relative to the Global Clock input signal with the fastest route and the lightest load. Hold time is
measured relative to the Global Clock input signal using the furthest distance and a reference load of one clock pin per two
IOBs. Use the static timing analyzer (TRCE) to determine the setup and hold times under given design conditions.

R
XC4000XLA Pin-to-Pin Input Parameter Guidelines (Cont.)

XC4000XLA FastCLK Input Set-Up and Hold for BUFNW, BUFSW, BUFNE, & BUFSE
Units
Input Setup and Hold Time
Relative to FastCLK Input Signal
No Delay XC4013XLA 0.0 / 3.2 0.0 / 2.9 0.0 / 2.6 ns
FastCLK and IFF TPSFN/TPHFN XC4020XLA 0.0 / 3.3 0.0 / 3.0 0.0 / 2.7 ns
XC4028XLA 0.0 / 3.4 0.0 / 3.1 0.0 / 2.8 ns
XC4036XLA 0.0 / 3.5 0.0 / 3.2 0.0 / 2.9 ns
XC4044XLA 0.0 / 3.6 0.0 / 3.3 0.0 / 3.0 ns
XC4052XLA 0.0 / 3.7 0.0 / 3.4 0.0 / 3.1 ns
XC4062XLA 0.0 / 3.8 0.0 / 3.5 0.0 / 3.2 ns
XC4085XLA 0.0 / 3.9 0.0 / 3.6 0.0 / 3.3 ns
Partial Delay XC4013XLA 3.5 / 0.6 3.2 / 0.3 2.9 / 0.0 ns
FastCLK and IFF TPSFPTPHFP XC4020XLA 3.7 / 0.4 3.4 / 0.2 3.1 / 0.0 ns
XC4028XLA 3.9 / 0.2 3.6 / 0.1 3.3 / 0.0 ns
XC4036XLA 4.1 / 0.0 3.8 / 0.0 3.5 / 0.0 ns
XC4044XLA 4.3 / 0.0 4.0 / 0.0 3.7 / 0.0 ns
XC4052XLA 4.5 / 0.0 4.2 / 0.0 3.9 / 0.0 ns
XC4062XLA 4.7 / 0.0 4.4 / 0.0 4.1 / 0.0 ns
XC4085XLA 5.1 / 0.0 4.8 / 0.0 4.5 / 0.0 ns
FastCLK and IFF TPSFD/TPHFD XC4020XLA 3.8 / 0.4 3.5 / 0.2 3.2 / 0.0 ns
XC4028XLA 4.0 / 0.2 3.7 / 0.1 3.4 / 0.0 ns
XC4036XLA 4.3 / 0.0 4.0 / 0.0 3.7 / 0.0 ns
XC4044XLA 4.6 / 0.0 4.3 / 0.0 4.0 / 0.0 ns
XC4052XLA 4.9 / 0.0 4.6 / 0.0 4.3 / 0.0 ns
XC4062XLA 5.3 / 0.0 5.0 / 0.0 4.7 / 0.0 ns
XC4085XLA 6.1 / 0.0 5.8 / 0.0 5.5 / 0.0 ns
Preliminary
IFF = Input Flip-Flop
Notes: Setup time is measured with the fastest route and the lightest load. Hold time is measured using the furthest distance and a
reference load of one clock pin per two IOBs. Use the static timing analyzer(TRCE) to determine the setup and hold times
under given design conditions.

R
XC4000XLA BUFGE #s 1, 2, 5, and 6 Global Early Clock, Set-up and Hold for IFF and FCL

Units
No Delay XC4013XLA 1.0 / 3.2 0.8 / 2.6 0.5 / 1.8 ns
Global Early Clock and IFF TPSEN/TPHEN XC4020XLA 1.0 / 3.4 0.8 / 2.8 0.5 / 2.0 ns
Global Early Clock and FCL TPFSEN/TPFHEN XC4028XLA 1.0 / 3.5 0.8 / 3.0 0.5 / 2.2 ns
XC4036XLA 1.0 / 3.6 0.8 / 3.1 0.5 / 2.4 ns
XC4044XLA 1.0 / 3.8 0.8 / 3.3 0.5 / 2.6 ns
XC4052XLA 1.0 / 4.0 0.8 / 3.5 0.5 / 2.8 ns
XC4062XLA 1.0 / 4.2 0.8 / 3.7 0.5 / 3.0 ns
XC4085XLA 1.0 / 4.6 0.8 / 4.0 0.5 / 3.2 ns
Global Early Clock and IFF TPSEP/TPHEP XC4020XLA 4.8 / 0.1 4.4 / 0.1 4.1 / 0.0 ns
Global Early Clock and FCL TPFSEP/TPFHEP XC4028XLA 4.9 / 0.1 4.6 / 0.1 4.4 / 0.0 ns 6
XC4036XLA 5.0 / 0.2 4.7 / 0.1 4.5 / 0.0 ns
XC4044XLA 5.5 / 0.3 5.1 / 0.2 4.8 / 0.0 ns
XC4052XLA 5.8 / 0.3 5.3 / 0.2 5.0 / 0.0 ns
XC4062XLA 6.2 / 0.4 5.6 / 0.2 5.2 / 0.0 ns
XC4085XLA 6.5 / 0.5 5.9 / 0.3 5.4 / 0.0 ns
Global Early Clock and IFF TPSED/TPHED XC4020XLA 4.9 / 0.0 4.5 / 0.0 4.1 / 0.0 ns
XC4028XLA 5.1 / 0.0 4.7 / 0.0 4.4 / 0.0 ns
XC4036XLA 5.3 / 0.0 4.9 / 0.0 4.5 / 0.0 ns
XC4044XLA 5.8 / 0.0 5.3 / 0.0 5.0 / 0.0 ns
XC4052XLA 6.2 / 0.0 5.7 / 0.0 5.3 / 0.0 ns
XC4062XLA 6.7 / 0.0 6.1 / 0.0 5.6 / 0.0 ns
XC4085XLA 7.0 / 0.0 6.4 / 0.0 6.0 / 0.0 ns
Preliminary

R
XC4000XLA BUFGE #s 3, 4, 7, and 8 Global Early Clock, Set-up and Hold for IFF and FCL

Units
No Delay XC4013XLA 0.8 / 3.2 0.6 / 2.6 0.4 / 2.0 ns
Global Early Clock and IFF TPSEN/TPHEN XC4020XLA 0.8 / 3.4 0.6 / 2.8 0.4 / 2.2 ns
Global Early Clock and FCL TPFSEN/TPFHEN XC4028XLA 0.8 / 3.5 0.6 / 3.0 0.4 / 2.4 ns
XC4036XLA 0.8 / 3.6 0.6 / 3.1 0.4 / 2.6 ns
XC4044XLA 0.8 / 3.8 0.6 / 3.3 0.4 / 2.8 ns
XC4052XLA 0.8 / 4.0 0.6 / 3.5 0.4 / 3.0 ns
XC4062XLA 0.8 / 4.2 0.6 / 3.7 0.4 / 3.2 ns
XC4085XLA 0.8 / 4.6 0.6 / 4.0 0.4 / 3.4 ns
Global Early Clock and IFF TPSEP/TPHEP XC4020XLA 4.6 / 0.1 4.2 / 0.1 3.8 / 0.0 ns
Global Early Clock and FCL TPFSEP/TPFHEP XC4028XLA 4.7 / 0.1 4.4 / 0.1 4.1 / 0.0 ns
XC4036XLA 4.8 / 0.2 4.5 / 0.2 4.2 / 0.0 ns
XC4044XLA 5.2 / 0.3 4.8 / 0.3 4.4 / 0.0 ns
XC4052XLA 5.6 / 0.3 5.1 / 0.3 4.6 / 0.0 ns
XC4062XLA 6.0 / 0.4 5.4 / 0.4 4.8 / 0.0 ns
XC4085XLA 6.3 / 0.5 5.7 / 0.5 5.0 / 0.0 ns
Global Early Clock and IFF TPSED/TPHED XC4020XLA 4.7 / 0.0 4.3 / 0.0 3.8 / 0.0 ns
XC4028XLA 4.9 / 0.0 4.5 / 0.0 4.1 / 0.0 ns
XC4036XLA 5.1 / 0.0 4.7 / 0.0 4.2 / 0.0 ns
XC4044XLA 5.6 / 0.0 5.1 / 0.0 4.6 / 0.0 ns
XC4052XLA 6.0 / 0.0 5.5 / 0.0 4.9 / 0.0 ns
XC4062XLA 6.5 / 0.0 5.9 / 0.0 5.2 / 0.0 ns
XC4085XLA 6.8 / 0.0 6.2 / 0.0 5.6 / 0.0 ns
Preliminary
, FCL = Fast Capture LatchIFF = Input Flip-Flop or Latch, FCL = Fast Capture Latch

R
XC4000XLA IOB Input Switching Characteristic Guidelines


Units
Description Symbol Device Min Max Min Max Min Max
Clocks
Clock Enable (EC) to Clock (IK) TECIK All devices 0.0 0.0 0.0 ns
Delay from FCL enable (OK) active edge to IFF TOKIK All Devices 1.4 1.3 1.2 ns
clock (IK) active edge
Setup Times
Pad to Clock (IK), no delay TPICK All Devices 1.2 1.0 0.9 ns
Pad to Clock (IK), via transparent Fast Capture TPICKF All Devices 1.6 1.4 1.3 ns
Latch, no delay
Pad to Fast Capture Latch Enable (OK), no de- TPOCK All Devices 0.8 0.7 0.6 ns
lay
Hold Times
All Hold Times All Devices 0.0 0.0 0.0 ns 6
Global Set/Reset
Minimum GSR Pulse Width TMRW All devices 12.8 11.4 10.2 ns
Global Set/Reset
Delay from GSR input to any Q TRRI* XC4013XLA 11.4 10.2 9.1 ns
XC4020XLA 13.3 11.9 10.6 ns
XC4028XLA 14.3 12.8 11.4 ns
XC4036XLA 16.2 14.5 12.9 ns
XC4044XLA 18.1 16.2 14.4 ns
XC4052XLA 19.5 17.4 15.6 ns
XC4062XLA 20.9 18.7 16.7 ns
XC4085XLA 24.7 22.1 19.7 ns
Propagation Delays
Pad to I1, I2 via transparent input latch, no de- TPLI All devices 2.1 1.9 1.7 ns
lay
Pad to I1, I2 via transparent FCL and input TPFLI All devices 2.5 2.2 2.0 ns
latch, no delay
Clock (IK) to I1, I2 (flip-flop) TIKRI All devices 1.1 1.0 0.9 ns
Clock (IK) to I1, I2 (latch enable, active Low) TIKLI All devices 1.2 1.1 1.0 ns
FCL Enable (OK) active edge to I1, I2 TOKLI All devices 2.4 2.1 1.9 ns
Preliminary
Notes: IFF = Input Flip-Flop or Latch, FCL = Fast Capture Latch

R
XC4000XLA IOB Output Switching Characteristic Guidelines

-09 -08 -07

Units
Clocks
Propagation Delays
Output (O) to Pad via Fast Output MUX TOFPF 3.6 3.2 2.9 ns
Select (OK) to Pad via Fast MUX TOKFPF 3.3 3.0 2.6 ns
Output (O) to clock (OK) hold time TOKO 0.0 0.0 0.0 ns
Clock Enable (EC) to clock (OK) setup time TECOK 0.0 0.0 0.0 ns
Clock Enable (EC) to clock (OK) hold time TOKEC 0.0 0.0 0.0 ns
Global Set/Reset
Delay from GSR input to any Pad TRPO* XC4013XLA 14.4 12.8 11.5
XC4020XLA 16.3 14.5 13.0
XC4028XLA 17.3 15.4 13.8
XC4036XLA 19.1 17.1 15.3
XC4044XLA 21.0 18.8 16.8
XC4052XLA 22.5 20.1 17.9
XC4062XLA 23.9 21.3 19.0
XC4085XLA 27.7 24.7 22.1
For output SLOW option add TSLOW 1.7 1.6 1.4 ns
Preliminary
* Indicates Minimum Amount of Time to Assure Valid Data

R
XC4000XV D.C. and Switching Performance Characteristics

Definition of Terms
follows:
derived from measuring internal test patterns.All specifications are representative of worst- case supply voltage and junction

VCCINT Supply voltage relative to GND -0.5 to 3.0 V
VCCIO Supply voltage relative to GND -0.5 to 4.0 V 6
VCC Longest Supply Voltage Rise Time from 1 V to 3V 50 ms
Ceramic packages +150 °C
TJ Junction temperature
Notes: Maximum DC overshoot or undershoot above Vcc or below GND must be limited to either 0.5 V or 10 mA, whichever is easier to
achieve. During transitions, the device pins may undershoot to -2.0 V or overshoot to + 7.0 V, provided this over- or
undershoot lasts less than 10 ns and with the forcing current being limited to 200 mA.
Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress
reliability.

R

VCCINT Supply voltage relative to GND, TJ = -40°C to Industrial 2.3 2.7 V
+100°C
VCCIO Supply voltage relative to GND, TJ = -40°C to Industrial 3.0 3.6 V
+100°C
Notes 1: At junction temperatures above those listed as Operating Conditions, all delay parameters increase by 0.35% per °C.

VOH
VOL
VCCINT Data Retention Supply Voltage (below which configuration data may be 2.1 V
VDRINT
lost)
VCCIO Data Retention Supply Voltage (below which configuration data may be 2.5 V
VDRIO
lost)
BGA, SBGA, PQ, HQ, & MQ packages 10 pF
CIN Input capacitance (sample tested)
PGA packages 16 pF

R
XC4000XV Global Buffer Switching Characteristic Guidelines

temperature).

Units
Delay from pad through Global Low Skew (GLS) TGLS XC40110XV 7.7 6.7 5.8 ns
clock buffer to any clock input, K. XC40150XV 7.8 6.8 5.9 ns
XC40200XV 9.3 8.1 7.0 ns
XC40250XV 9.4 8.2 7.1 ns
Delay from pad through Global Early (GE) clock TGE_1256 XC40110XV 5.6 4.9 4.2 ns
buffer to any IOB clock input for BUFGE #s 1, 2, XC40150XV 5.7 5.0 4.3 ns
5, and 6.
XC40200XV 6.1 5.3 4.6 ns 6
XC40250XV 6.2 5.4 4.7 ns
Delay from pad through Global Early (GE) clock TGE_3478 XC40110XV 5.6 4.9 4.2 ns
buffer to any IOB clock input for BUFGE #s 3, 4, XC40150XV 5.7 5.0 4.3 ns
7, and 8.
XC40200XV 6.1 5.3 4.6 ns
XC40250XV 6.2 5.4 4.7 ns
Delay from pad through FastCLK buffer to any TFCLK XC40110XV 3.0 2.6 2.2 ns
IOB clock input XC40150XV 3.1 2.7 2.3 ns
XC40200XV 3.4 3.0 2.6 ns
XC40250XV 3.5 3.1 2.7 ns
Advance

R
XC4000XV CLB Switching Characteristic Guidelines

parameters assume worst-case operating conditions (supply voltage and junction temperature). Values are expressed in
nanoseconds unless otherwise noted.
Units
C inputs via SR/H0 via H to X/Y outputs THH0O 1.5 1.3 1.1 ns
C inputs via H1 via H to X/Y outputs THH1O 1.3 1.1 1.0 ns
C inputs via DIN/H2 via H to X/Y outputs THH2O 1.5 1.3 1.1 ns
C inputs via EC, DIN/H2 to YQ, XQ output (bypass0 TCBYP 0.8 0.7 0.6 ns
Sequential Delays
F/G inputs via H TIHCK 2.6 2.2 1.9 ns
CIN input via F/G TCCK 2.8 2.4 2.1 ns
CIN input via F/G and H TCHC 3.8 3.3 2.9 ns
All Hold Times 0.0 0.0 0.0 ns
Clocks
Set/Reset Direct
Global Set/Reset
Delay from GSR input to any Q TMRQ See page 201 for TRRI values per device
Toggle Frequency (MHz) (for export control purposes) FTOG 188 217 250 MHz
Advance

R
XC4000XV RAM Synchronous (Edge-Triggered) Write Operation Guidelines

XC4000XV devices and are expressed in nanoseconds unless otherwise noted.

Write Operation
32x1 TWCTS 9.6 8.4 7.3 ns
32x1 TWPTS 4.8 4.2 3.7 ns
32x1 TASTS 2.0 1.8 1.5 ns
Address hold time after clock K 16x2 TAHS 0.0 0.0 0.0 ns
32x1 TAHTS 0.0 0.0 0.0 ns 6
32x1 TDSTS 2.5 2.2 1.9 ns
DIN hold time after clock K 16x2 TDHS 0.0 0.0 0.0 ns
32x1 TDHTS 0.0 0.0 0.0 ns
32x1 TWSTS 1.8 1.5 1.3 ns
WE hold time after clock K 16x2 TWHS 0.0 0.0 0.0 ns
32x1 TWHTS 0.0 0.0 0.0 ns
32x1 TWOTS 7.5 6.5 5.7 ns
Read Operation
32x1 TRCT 6.5 5.5 5.5 ns
Data Valid after address change (no Write Enable) 16x2 TILO 1.2 1.1 0.9 ns
32x1 TIHO 2.3 2.0 1.7 ns
32x1 TIHCK 2.6 2.2 1.9 ns
Advance

R
XC4000XV CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines


Dual Port RAM Units
Address hold time after clock K 16x1 TAHDS 0.0 0.0 0.0 ns
DIN hold time after clock K 16x1 TDHDS 0.0 0.0 0.0 ns
WE hold time after clock K 16x1 TWHDS 0.0 0.0 0.0 ns
Advance
Note: Timing for 16x1 option is identical to 16x2 RAM.
XC4000XV CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS TWPDS
WCLK (K) WCLK (K)
WE WE
DATA IN DATA IN
ADDRESS ADDRESS
TILO TILO
TILO TILO
TWOS TWODS
DATA OUT OLD NEW

DATA OUT OLD NEW
X6474
X6461
Single Port RAM Dual Port RAM

R
XC4000XV Pin-to-Pin Output Parameter Guidelines

XC4000XV Global Clock Input to Output Delay

Units
Global Low Skew (GLS) Clock Input to Output TICKOF XC40110XV 10.6 9.2 8.0 ns
Delay using Output Flip-Flop. XC40150XV 10.7 9.3 8.1 ns
XC40200XV 12.2 10.6 9.2 ns
XC40250XV 12.3 10.7 9.3 ns
Advance
Notes: Listed above are representative values where one global clock input drives one vertical clock line in each accessible
column, and where all accessible IOB and CLB flip-flops are clocked by the global clock net.
Clock-to-out minimum delay is measured with the fastest route and the lightest load, Clock-to-out maximum delay is 6
accessible CLB flip-flops. For designs with fewer number of clock loads, the pad-to-IOB clock pin delay as determined
clock-to-out delay for clocked outputs in FAST mode specification.
XC4000XV FastCLK Input to Output Delay for BUFNW, BUFSW, BUFNE, and BUFSE
Units
FastCLK Input to Output Delay using Output TICKFOF XC40110XV 5.9 5.1 4.4 ns
Flip-Flop for FastCLK buffers BUFNW, BUFSW, XC40150XV 6.0 5.2 4.5 ns
BUFNE, and BUFSE. XC40200XV 6.3 5.5 4.8 ns
XC40250XV 6.4 5.6 4.9 ns
Advance
Notes: Listed above are representative values where one FastCLK input drives one vertical clock line in each accessible column,
and where all accessible IOB and CLB flip-flops are clocked by the FastCLK net.
Clock-to-out minimum delay is measured with the fastest route and the lightest load, Clock-to-out maximum delay is
accessible CLB flip-flops. For designs with fewer number of clock loads, the pad-to-IOB clock pin delay as determined
clock-to-out delay for clocked outputs in FAST slew mode specification.

R
XC4000XV Pin-to-Pin Output Parameter Guidelines

XC4000XV Global Early Clock Input to Output Delay for BUFGE #s 1, 2, 5, and 6
Units
Global Clock Signal Input to Output Delay using TICKEOF_1256 XC40110XV 8.5 7.4 6.4 ns
Global Early (GE) clock buffer to clock Output XC40150XV 8.6 7.5 6.5 ns
Flip-Flop for BUFGE #s 1, 2, 5, and 6.
XC40200XV 9.0 7.8 6.8 ns
XC40250XV 9.1 7.9 6.9 ns
Advance
accessible CLB flip-flops. For designs with fewer number of clock loads, the pad-to-IOB clock pin delay as determined by
the static timing analyzer (TRCE) can be added to the AC parameter Tokpof and used as a worst-case pin-to-pin
XC4000XV Global Early Clock Input to Output Delay for BUFGE #s 3, 4, 7, and 8
Units
Global Clock Signal Input to Output Delay using TICKEOF_3478 XC40110XV 8.5 7.4 6.4 ns
Global Early (GE) clock buffer to clock Output XC40150XV 8.6 7.5 6.5 ns
Flip-Flop for BUFGE #s 3, 4, 7, and 8.
XC40200XV 9.0 7.8 6.8 ns
XC40250XV 9.1 7.9 6.9 ns
Advance
accessible CLB flip-flops. For designs with fewer number of clock loads, the pad-to-IOB clock pin delay as determined by
the static timing analyzer (TRCE) can be added to the AC parameter Tokpof and used as a worst-case pin-to-pin

2
Delta Delay (ns)

delay. -1
voltage and temperature and is independent of the output -2
slew rate control. 0 20 40 60 80 100 120 140
Capacitance (pF)
X8257

R
XC4000XV Pin-to-Pin Input Parameter Guidelines

XC4000XV Global Low Skew Clock, Set-Up and Hold

Units
No Delay XC40110XV 0.0 / 6.8 0.0 / 5.9 0.0 / 5.1 ns
Global Low Skew Clock and IFF TPSN/TPHN XC40150XV 0.0 / 6.9 0.0 / 6.0 0.0 / 5.2 ns
Global Low Skew Clock and FCL XC40200XV 0.0 / 7.9 0.0 / 6.9 0.0 / 6.0 ns
XC40250XV 0.0 / 8.7 0.0 / 7.6 0.0 / 6.6 ns
Partial Delay XC40110XV 5.7 / 1.0 5.0 / 0.9 4.3 / 0.8 ns
Global Low Skew Clock and IFF TPSP/TPHP XC40150XV 5.8 / 1.0 5.1 / 0.9 4.4 / 0.8 ns
Global Low Skew Clock and FCL XC40200XV 6.2 / 1.0 5.4 / 0.9 4.7 / 0.8 ns
XC40250XV 6.6 / 1.0 5.8 / 0.9 5.0 / 0.8 ns
Full Delay XC40110XV 6.5 / 0.0 5.7 / 0.0 4.9 / 0.0 ns 6
Global Low Skew Clock and IFF TPSD/TPHD XC40150XV 6.6 / 0.0 5.8 / 0.0 5.0 / 0.0 ns
XC40200XV 7.3 / 0.0 6.3 / 0.0 5.5 / 0.0 ns
XC40250XV 7.9 / 0.0 6.9 / 0.0 6.0 / 0.0 ns
Advance
XC4000XV FastCLK Input Set-Up and Hold for BUFNW, BUFSW, BUFNE, & BUFSE
Units
Input Setup and Hold Time Relative to FastCLK Input Signal
No Delay XC40110XV 3.3 / 2.6 2.9 / 2.2 2.5 / 1.9 ns
FastCLK and IFF TPSFN/TPHFN XC40150XV 3.3 / 2.7 2.9 / 2.3 2.5 / 2.0 ns
XC40200XV 3.3 / 2.9 2.9 / 2.5 2.5 / 2.2 ns
XC40250XV 3.3 / 3.2 2.9 / 2.8 2.5 / 2.4 ns
Advance
Notes: Setup time is measured relative to the FastCLK input signal with the fastest route and the lightest load. Hold time is measured
relative to the FastCLK input signal using the furthest distance and a reference load of one clock pin per two IOBs. Use the
static timing analyzer(TRCE) to determine the setup and hold times under given design conditions.

R
XC4000XV BUFGE #s 1, 2, 5, & 6 Global Early Clock, Set-Up and Hold for IFF and FCL
Units
No Delay XC40110XV 0.7 / 5.2 0.6 / 4.5 0.5 / 3.9 ns
Global Early Clock and IFF TPSEN/TPHEN XC40150XV 0.7 / 5.3 0.6 / 4.6 0.5 / 4.0 ns
Global Early Clock and FCL TPFSEN/TPFHEN XC40200XV 0.7 / 5.6 0.6 / 4.8 0.5 / 4.2 ns
XC40250XV 0.7 / 5.8 0.6 / 5.1 0.5 / 4.4 ns
Global Early Clock andIFF TPSEP/TPHEP XC40150XV 9.0 / 0.0 7.8 / 0.0 6.8 / 0.0 ns
Global Early Clock and FCL TPFSEP/TPFHEP XC40200XV 9.7 / 0.0 8.4 / 0.0 7.3 / 0.0 ns
XC40250XV 10.3 / 0.0 9.0 / 0.0 7.8 / 0.0 ns
FullDelay XC40110XV 9.4 / 0.0 8.2 / 0.0 7.1 / 0.0 ns
Global Early Clock and IFF TPSED/TPHED XC40150XV 9.5 / 0.0 8.3 / 0.0 7.2 / 0.0 ns
XC40200XV 10.7 / 0.0 9.3 / 0.0 8.1 / 0.0 ns
XC40250XV 11.9 / 0.0 10.4 / 0.0 9.0 / 0.0 ns
Advance
IOBs. Use the static timing analyzer(TRCE) to determine the setup and hold times under given design conditions.
XC4000XV BUFGE #s 3, 4, 7, & 8 Global Early Clock, Set-Up and Hold for IFF and FCL
Units
No Delay XC40110XV 0.7 / 5.2 0.6 / 4.5 0.5 / 3.9 ns
Global Early Clock and IFF TPSEN/TPHEN XC40150XV 0.7 / 5.3 0.6 / 4.6 0.5 / 4.0 ns
Global Early Clock and FCL TPFSEN/TPFHEN XC40200XV 0.7 / 5.6 0.6 / 4.8 0.5 / 4.2 ns
XC40250XV 0.7 / 5.8 0.6 / 5.1 0.5 / 4.4 ns
Global Early Clock and IFF TPSEP/TPHEP XC40150XV 9.0 / 0.0 7.8 / 0.0 6.8 / 0.0 ns
Global Early Clock and FCL TPFSEP/TPFHEP XC40200XV 9.7 / 0.0 8.4 / 0.0 7.3 / 0.0 ns
XC40250XV 10.3 / 0.0 9.0 / 0.0 7.8 / 0.0 ns
FullDelay XC40110XV 9.4 / 0.0 8.2 / 0.0 7.1 / 0.0 ns
Global Early Clock and IFF TPSED/TPHED XC40150XV 9.5 / 0.0 8.3 / 0.0 7.2 / 0.0 ns
XC40200XV 10.7 / 0.0 9.3 / 0.0 8.1 / 0.0 ns
XC40250XV 11.9 / 0.0 10.4 / 0.0 9.0 / 0.0 ns
Advance
IOBs. Use the static timing analyzer(TRCE) to determine the setup and hold times under given design conditions.

R
XC4000XV IOB Input Switching Characteristic Guidelines


Units
Clocks
Clock Enable (EC) to Clock (IK) TECIK All Devices 0.0 0.0 0.0 ns
Delay from Fast Capture Latch enable (OK) active TOKIK All Devices 2.0 1.8 1.5 ns
edge to IFF clock (IK) active edge
Setup Times
Pad to Clock (IK), no delay TPICK All Devices 0.7 0.6 0.5 ns
Pad to Clock (IK), via transparent Fast Capture TPICKF Al Devices 1.2 1.1 0.9 ns
Latch, no delay
Pad to Fast Capture Latch Enable (OK), no delay TPOCK All Devices 0.2 0.2 0.1 ns
Hold Times
All Hold Times All Devices 0.0 0.0 0.0 ns
Global Set/Reset
Minimum GSR Pulse Width TMRW All Devices 18.4 16.0 13.9 ns 6
Delay from GSR input to any Q TRRI* XC40110XV 35.0 30.5 26.5 ns
XC40150XV 38.5 33.5 29.1 ns
XC40200XV 42.4 36.9 32.1 ns
XC40250XV 46.6 40.5 35.3 ns
Propagation Delays
Pad to I1, I2 via transparent input latch, no delay TPLI All devices 1.5 1.3 1.1 ns
Pad to I1, I2 via transparent FCL and input latch, TPFLI All Devices 2.0 1.8 1.6 ns
no delay
Clock (IK) to I1, I2 (flip-flop) TIKRI All Devices 1.1 0.9 0.8 ns
Clock (IK) to I1, I2 (latch enable, active Low) TIKLI All Devices 1.2 1.1 0.9 ns
FCL Enable (OK) active edge to I1, I2 (via trans- TOKLI All Devices 2.8 2.5 2.1 ns
parent standard input latch)
Advance
IFF = Input Flip-Flop or Latch, FCL = Fast Capture Latch.

R
XC4000XV IOB Output Switching Characteristic Guidelines


Units
Clocks
Propagation Delays (See Note 1)
Output (O) to Pad via Fast Output MUX TOFPF 2.9 2.5 2.2 ns
Select (OK) to Pad via Fast MUX TOKFPF 2.5 2.2 1.9 ns
Output (O) to clock (OK) hold time TOKO 0.0 0.0 0.0 ns
Clock Enable (EC) to clock (OK) setup time TECOK 0.0 0.0 0.0 ns
Clock Enable (EC) to clock (OK) hold time TOKEC 0.3 0.2 0.1 ns
Global Set/Reset
Delay from GSR input to any Pad TRPO* XC40110XV 37.1 32.2 28.0 ns
XC40150XV 40.5 35.3 30.7 ns
XC40200XV 44.4 38.6 33.6 ns
XC40250XV 48.6 42.3 36.8 ns
For output SLOW option add TSLOW 1.7 1.6 1.4 ns
Advance
Note: Output timing is measured at ~50% VCC threshold, with 50 pF external capacitive loads.

XC4000XLA/XV Field Programmable Gate Arrays February 1, 1999 (Version 1.0) 6
XC4013XLA Pinout Table XC4013XLA Pinout Table (Continued)
XC4013XLA Pinout Table PAD NAME PQ160 PQ208 PQ240 BG256

I/O – – P13 F2
PAD NAME PQ160 PQ208 PQ240 BG256
GND P10 P14 P14 GND*
VCC P142 P183 P212 VCC*
I/O, FCLK1 P11 P15 P15 G3
I/O (A8) P143 P184 P213 C10
I/O P12 P16 P16 G2
I/O (A9) P144 P185 P214 D10
I/O (TMS) P13 P17 P17 G1
I/O (A19) P145 P186 P215 A9
I/O P14 P18 P18 H3
I/O (A18) P146 P187 P216 B9
VCC – – P19 VCC*
I/O – P188 P217 C9
I/O – – P20 H2
I/O – P189 P218 D9
I/O – – P21 H1
GND – – P219 GND*
I/O (A10) P147 P190 P220 A8
I/O – P19 P23 J2
I/O (A11) P148 P191 P221 B8
I/O – P20 P24 J1
I/O P15 P21 P25 K2
I/O – – P223 A6
I/O P16 P22 P26 K3
I/O – – P224 C7
I/O P17 P23 P27 K1
I/O P149 P192 P225 B6
I/O P18 P24 P28 L1
I/O P150 P193 P226 A5
GND P151 P194 P227 GND*
VCC P20 P26 P30 VCC*
I/O – P195 P228 C6
I/O P21 P27 P31 L2 6
I/O – P196 P229 B5
I/O P22 P28 P32 L3
I/O P152 P197 P230 A4
I/O P23 P29 P33 L4
I/O P153 P198 P231 C5
I/O P24 P30 P34 M1
I/O (A12) P154 P199 P232 B4
I/O – P31 P35 M2
I/O (A13) P155 P200 P233 A3
I/O – P32 P36 M3
GND – – – GND*
VCC – – – VCC*
I/O – – P38 N1
I/O – – P234 D5
I/O – – P39 N2
I/O – – P235 C4
I/O P156 P201 P236 B3
I/O P25 P33 P41 P1
I/O P157 P202 P237 B2
I/O P26 P34 P42 P2
I/O (A14) P158 P203 P238 A2
I/O P27 P35 P43 R1
I/O, GCK8 (A15) P159 P204 P239 C3
I/O, FCLK2 P28 P36 P44 P3
VCC P160 P205 P240 VCC*
GND P1 P2 P1 GND*
I/O – – P46 T1
I/O, GCK1 (A16) P2 P4 P2 B1
I/O – – P47 R3
I/O (A17) P3 P5 P3 C2
I/O – P38 P48 T2
I/O P4 P6 P4 D2
I/O P39 P49 U1
I/O P5 P7 P5 D3
I/O P30 P40 P50 T3
I/O (TDI) P6 P8 P6 E4
I/O P31 P41 P51 U2
I/O (TCK) P7 P9 P7 C1
I/O P32 P42 P52 V1
I/O P8 P10 P8 D1
I/O P33 P43 P53 T4
I/O P9 P11 P9 E3
I/O P34 P44 P54 U3
I/O – P12 P10 E2
I/O P35 P45 P55 V2
I/O – P13 P11 E1
I/O P36 P46 P56 W1
I/O – – P12 F3

R
XC4013XLA Pinout Table (Continued) XC4013XLA Pinout Table (Continued)
PAD NAME PQ160 PQ208 PQ240 BG256 PAD NAME PQ160 PQ208 PQ240 BG256
I/O, GCK2 P37 P47 P57 V3 VCC – – P101 VCC*
O (M1) P38 P48 P58 W2 I/O P66 P86 P102 Y15
GND P39 P49 P59 GND* I/O P67 P87 P103 V14
I (M0) P40 P50 P60 Y1 I/O P68 P88 P104 W15
VCC P41 P55 P61 VCC* I/O P69 P89 P105 Y16
I (M2) P42 P56 P62 W3 GND P70 P90 P106 GND*
I/O, GCK3 P43 P57 P63 Y2 I/O – – P107 V15
I/O (HDC) P44 P58 P64 W4 I/O – – P108 W16
I/O P45 P59 P65 V4 I/O – P91 P109 Y17
I/O P46 P60 P66 U5 I/O – P92 P110 V16
I/O P47 P61 P67 Y3 I/O P71 P93 P111 W17
I/O (/LDC) P48 P62 P68 Y4 I/O P72 P94 P112 Y18
VCC – – – VCC* GND – – – GND*
GND – – – GND* VCC – – – VCC*
I/O P49 P63 P69 V5 I/O P73 P95 P113 U16
I/O P50 P64 P70 W5 I/O P74 P96 P114 V17
I/O – P65 P71 Y5 I/O P75 P97 P115 W18
I/O – P66 P72 V6 I/O P76 P98 P116 Y19
I/O – – P73 W6 I/O P77 P99 P117 V18
I/O – – P74 Y6 I/O, GCK4 P78 P100 P118 W19
GND P51 P67 P75 GND* GND P79 P101 P119 GND*
I/O P52 P68 P76 W7 DONE P80 P103 P120 Y20
I/O P53 P69 P77 Y7 VCC P81 P106 P121 VCC*
I/O P54 P70 P78 V8 /PROGRAM P82 P108 P122 V19
I/O P55 P71 P79 W8 I/O (D7) P83 P109 P123 U19
VCC – – P80 VCC* I/O, GCK5 P84 P110 P124 U18
I/O – P72 P81 Y8 I/O P85 P111 P125 T17
I/O – P73 P82 U9 I/O P86 P112 P126 V20
GND – – P83 GND* I/O – – P127 U20
I/O – – P84 Y9 I/O – – P128 T18
I/O – – P85 W10 VCC – – – VCC*
I/O P56 P74 P86 V10 GND – – – GND*
I/O P57 P75 P87 Y10 I/O (D6) P87 P113 P129 T19
I/O P58 P76 P88 Y11 I/O P88 P114 P130 T20
I/O (/INIT) P59 P77 P89 W11 I/O P89 P115 P131 R18
VCC P60 P78 P90 VCC* I/O P90 P116 P132 R19
GND P61 P79 P91 GND* I/O – P117 P133 R20
I/O P62 P80 P92 V11 I/O – P118 P134 P18
I/O P63 P81 P93 U11 GND P91 P119 P135 GND*
I/O P64 P82 P94 Y12 I/O – – P136 P20
I/O P65 P83 P95 W12 I/O – – P137 N18
I/O – P84 P96 V12 I/O, FCLK3 P92 P120 P138 N19
I/O – P85 P97 U12 I/O P93 P121 P139 N20
GND – – P98 GND* VCC – – P140 VCC*
I/O – – P99 V13 I/O (D5) P94 P122 P141 M17
I/O – – P100 Y14 I/O (/CS0) P95 P123 P142 M18

R
GND – – P143 GND* I/O, (CS1, A2) P127 P165 P187 A18
I/O – P124 P144 M20 I/O (A3) P128 P166 P188 A17
I/O – P125 P145 L19 VCC – – – VCC*
I/O P96 P126 P146 L18 GND – – – GND*
I/O P97 P127 P147 L20 I/O – – P189 C16
I/O (D4) P98 P128 P148 K20 I/O – – P190 B16
I/O P99 P129 P149 K19 I/O P129 P167 P191 A16
VCC P100 P130 P150 VCC* I/O P130 P168 P192 C15
GND P101 P131 P151 GND* I/O – P169 P193 B15
I/O (D3) P102 P132 P152 K18 I/O – P170 P194 A15
I/O (/RS) P103 P133 P153 K17 GND P131 P171 P196 GND*
I/O P104 P134 P154 J20 I/O P132 P172 P197 B14
I/O P105 P135 P155 J19 I/O P133 P173 P198 A14
I/O – P136 P156 J18 I/O – – P199 C13
I/O – P137 P157 J17 I/O – – P200 B13
GND – – P158 GND* VCC – – P201 VCC*
I/O (D2) P106 P138 P159 H19 I/O (A4) P134 P174 P202 C12
I/O P107 P139 P160 H18 I/O (A5) P135 P175 P203 B12
VCC – – P161 VCC* GND – – P204 GND* 6
I/O P108 P140 P162 G19 I/O – P176 P205 A12
I/O, FCLK4 P109 P141 P163 F20 I/O P136 P177 P206 B11
I/O – – P164 G18 I/O (A21) P137 P178 P207 C11
I/O –– – P165 F19 I/O (A20) P138 P179 P208 A11
GND P110 P142 P166 GND* I/O (A6) P139 P180 P209 A10
I/O – – P167 F18 I/O (A7) P140 P181 P210 B10
I/O – – P168 E19 GND P141 P182 P211 GND*
I/O – P143 P169 D20
I/O – P144 P170 E18 VCC – – – D6
I/O P111 P145 P171 D19 VCC – – – D11
I/O P112 P146 P172 C20 VCC – – – D15
GND – – – GND* VCC – – – F4
VCC – – – VCC* VCC – – – F17
I/O (D1) P113 P147 P173 E17 VCC – – – K4
I/O (/RCK, P114 P148 P174 D18 VCC – – – L17
RDY_/BUSY) VCC – – – R4
I/O P115 P149 P175 C19 VCC – – – R17
I/O P116 P150 P176 B20 VCC – – – U6
I/O (D0, DIN) P117 P151 P177 C18 VCC – – – U10
I/O, GCK6 (DOUT) P118 P152 P178 B19 VCC – – – U15
CCLK P119 P153 P179 A20 VCC – – – C14
VCC P120 P154 P180 VCC* VCC – – – F1
O, TDO P121 P159 P181 A19 VCC – – – R2
GND P122 P160 P182 GND* VCC – – – E20
I/O (A0, /WS) P123 P161 P183 B18 VCC – – – P19
I/O, GCK7 (A1) P124 P162 P184 B17 VCC – – – V7
I/O P125 P163 P185 C17 VCC – – – D7
I/O P126 P164 P186 D16

R
VCC – – – D14 GND – – – W14
VCC – – – G17 GND – – – G20
VCC – – – P17
VCC – – – W20 NC – P1 P195 A7
VCC – – – U14 NC – P3 – C8
VCC – – – U7 NC – P51 – D12
VCC – – – P4 NC – P52 – A13
VCC – – – G4 NC – P53 – H20
NC – P54 – Y13
GND – – – A1 NC – P102 – W13
GND – – – D4 NC – P104 – M19
GND – – – D8 NC – P105 – W9
GND – – – D13 NC – P107 – V9
GND – – – D17 NC – P155 – M4
GND – – – H4 NC – P156 – J3
GND – – – H17 NC – P157 – J4
GND – – – N4 NC – P158 – –
GND – – – N17 NC – P206 – –
GND – – – U4 NC – P207 – –
GND – – – U8 NC – P208 – –
GND – – – U13 12/18/98
GND – – – U17
GND – – – B7
GND – – – N3

XC4020XLA Pinout Table PAD NAME PQ160 PQ208 PQ240 BG256

I/O – – P21 H1
PAD NAME PQ160 PQ208 PQ240 BG256
VCC P142 P183 P212 VCC*
I/O – – – J4
I/O (A8) P143 P184 P213 C10
I/O – – – J3
I/O (A9) P144 P185 P214 D10
I/O – P19 P23 J2
I/O (A19) P145 P186 P215 A9
I/O – P20 P24 J1
I/O (A18) P146 P187 P216 B9
I/O P15 P21 P25 K2
I/O – P188 P217 C9
I/O P16 P22 P26 K3
I/O – P189 P218 D9
I/O P17 P23 P27 K1
I/O P18 P24 P28 L1
I/O (A10) P147 P190 P220 A8 GND P19 P25 P29 GND*
I/O (A11) P148 P191 P221 B8 VCC P20 P26 P30 VCC*
I/O – – – C8 I/O P21 P27 P31 L2
I/O – – – A7 I/O P22 P28 P32 L3
VCC – – P222 VCC* I/O P23 P29 P33 L4
I/O – – P223 A6
I/O P24 P30 P34 M1
I/O – – P224 C7
I/O – P31 P35 M2
I/O P149 P192 P225 B6
I/O – P32 P36 M3
I/O P150 P193 P226 A5
I/O – – – M4
GND P151 P194 P227 GND*
I/O – P195 P228 C6
I/O – – P38 N1
I/O – P196 P229 B5
I/O – – P39 N2
I/O P152 P197 P230 A4
VCC – – P40 VCC* 6
I/O P153 P198 P231 C5
I/O P25 P33 P41 P1
I/O (A12) P154 P199 P232 B4
I/O P26 P34 P42 P2
I/O (A13) P155 P200 P233 A3
I/O P27 P35 P43 R1
I/O, FCLK2 P28 P36 P44 P3
I/O – – P234 D5
I/O – – P46 T1
I/O – – P235 C4
I/O – – P47 R3
I/O P156 P201 P236 B3
I/O – P38 P48 T2
I/O P157 P202 P237 B2 I/O – P39 P49 U1
I/O (A14) P158 P203 P238 A2 I/O P30 P40 P50 T3
I/O, GCK8 (A15)” P159 P204 P239 C3 I/O P31 P41 P51 U2
VCC P160 P205 P240 VCC GND – – – GND*
GND P1 P2 P1 GND* VCC – – – VCC*
I/O, GCK1 (A16) P2 P4 P2 B1
I/O P32 P42 P52 V1
I/O (A17) P3 P5 P3 C2
I/O P33 P43 P53 T4
I/O P4 P6 P4 D2
I/O P34 P44 P54 U3
I/O P5 P7 P5 D3
I/O P35 P45 P55 V2
I/O (TDI) P6 P8 P6 E4
I/O P36 P46 P56 W1
I/O (TCK) P7 P9 P7 C1
I/O, GCK2 P37 P47 P57 V3
O (M1) P38 P48 P58 W2
I/O P8 P10 P8 D1
I (M0) P40 P50 P60 Y1
I/O P9 P11 P9 E3
VCC P41 P55 P61 VCC*
I/O – P12 P10 E2
I (M2) P42 P56 P62 W3
I/O – P13 P11 E1
I/O, GCK3 P43 P57 P63 Y2
I/O – – P12 F3
I/O (HDC) P44 P58 P64 W4
I/O – – P13 F2
I/O P45 P59 P65 V4
I/O P46 P60 P66 U5
I/O, FCLK1 P11 P15 P15 G3 I/O P47 P61 P67 Y3
I/O P12 P16 P16 G2 I/O (/LDC) P48 P62 P68 Y4
I/O (TMS) P13 P17 P17 G1 VCC – – – VCC*
I/O P14 P18 P18 H3 GND – – – GND*
VCC – – P19 VCC* I/O P49 P63 P69 V5
I/O – – P20 H2
I/O P50 P64 P70 W5

R
I/O – P65 P71 Y5 /PROGRAM P82 P108 P122 V19
I/O – P66 P72 V6 I/O (D7) P83 P109 P123 U19
I/O – – P73 W6 I/O, GCK5 P84 P110 P124 U18
I/O – – P74 Y6 I/O P85 P111 P125 T17
GND P51 P67 P75 GND* I/O P86 P112 P126 V20
I/O P52 P68 P76 W7 I/O – – P127 U20
I/O P53 P69 P77 Y7 I/O – – P128 T18
I/O P54 P70 P78 V8 VCC – – – VCC*
I/O P55 P71 P79 W8 GND – – – GND*
VCC – P80 VCC* I/O (D6) P87 P113 P129 T19
I/O – P72 P81 Y8 I/O P88 P114 P130 T20
I/O – P73 P82 U9 I/O P89 P115 P131 R18
GND – – P83 GND* I/O P90 P116 P132 R19
I/O – – – V9 I/O – P117 P133 R20
I/O – – – W9 I/O – P118 P134 P18
I/O – – P84 Y9 GND P91 P119 P135 GND*
I/O – – P85 W10 I/O – – P136 P20
I/O P56 P74 P86 V10 I/O – – P137 N18
I/O P57 P75 P87 Y10 I/O, FCLK3 P92 P120 P138 N19
I/O P58 P76 P88 Y11 I/O P93 P121 P139 N20
I/O (/INIT) P59 P77 P89 W11 VCC – – P140 VCC*
VCC P60 P78 P90 VCC* I/O (D5) P94 P122 P141 M17
GND P61 P79 P91 GND* I/O (/CS0) P95 P123 P142 M18
I/O P62 P80 P92 V11 GND – – P143 GND*
I/O P63 P81 P93 U11 I/O – – – M19
I/O P64 P82 P94 Y12 I/O – P124 P144 M20
I/O P65 P83 P95 W12 I/O – P125 P145 L19
I/O – P84 P96 V12 I/O P96 P126 P146 L18
I/O – P85 P97 U12 I/O P97 P127 P147 L20
I/O – – – Y13 I/O (D4) P98 P128 P148 K20
I/O – – – W13 I/O P99 P129 P149 K19
GND – – P98 GND* VCC P100 P130 P150 VCC*
I/O – – P99 V13 GND P101 P131 P151 GND*
I/O – – P100 Y14 I/O (D3) P102 P132 P152 K18
VCC – – P101 VCC* I/O (/RS) P103 P133 P153 K17
I/O P66 P86 P102 Y15 I/O P104 P134 P154 J20
I/O P67 P87 P103 V14 I/O P105 P135 P155 J19
I/O P68 P88 P104 W15 I/O – P136 P156 J18
I/O P69 P89 P105 Y16 I/O – P137 P157 J17
GND P70 P90 P106 GND* I/O – – – H20
I/O – – P107 V15 GND – – P158 GND*
I/O – – P108 W16 I/O (D2) P106 P138 P159 H19
I/O – P91 P109 Y17 I/O P107 P139 P160 H18
I/O – P92 P110 V16 VCC – – P161 VCC*
I/O P71 P93 P111 W17 I/O P108 P140 P162 G19
I/O P72 P94 P112 Y18 I/O, FCLK4 P109 P141 P163 F20
GND – – – GND* I/O – – P164 G18
VCC – – – VCC* I/O – – P165 F19
I/O P73 P95 P113 U16 GND P110 P142 P166 GND*
I/O P74 P96 P114 V17 I/O – – P167 F18
I/O P75 P97 P115 W18 I/O – – P168 E19
I/O P76 P98 P116 Y19 I/O – P143 P169 D20
I/O P77 P99 P117 V18 I/O – P144 P170 E18
I/O, GCK4 P78 P100 P118 W19 I/O P111 P145 P171 D19
GND P79 P101 P119 GND* I/O P112 P146 P172 C20
DONE P80 P103 P120 Y20 GND – – – GND*
VCC P81 P106 P121 VCC* VCC – – – VCC*

R
I/O (D1) P113 P147 P173 E17 VCC – – – R2
I/O (/RCK, RDY_/BUSY) P114 P148 P174 D18 VCC – – – E20
I/O P115 P149 P175 C19 VCC – – – P19
I/O P116 P150 P176 B20 VCC – – – V7
I/O (D0, DIN) P117 P151 P177 C18 VCC – – – D7-CORE
I/O, GCK6 (DOUT) P118 P152 P178 B19 VCC – – – D14-CORE
CCLK P119 P153 P179 A20 VCC – – – G17
VCC P120 P154 P180 VCC* VCC – – – P17
O, TDO P121 P159 P181 A19 VCC – – – W20
GND P122 P160 P182 GND* VCC – – – U14-CORE
I/O (A0, /WS) P123 P161 P183 B18 VCC – – – U7-CORE
I/O, GCK7 (A1)” P124 P162 P184 B17 VCC – – – P4
I/O P125 P163 P185 C17 VCC – – – G4-CORE
I/O P126 P164 P186 D16
I/O, (CS1, A2) P127 P165 P187 A18 GND – – – A1
I/O (A3) P128 P166 P188 A17 GND – – – D4
VCC – – – VCC* GND – – – D8
GND – – – GND* GND – – – D13
I/O – – P189 C16 GND – – – D17
I/O – – P190 B16 GND – – – H4
I/O P129 P167 P191 A16 GND – – – H17
I/O P130 P168 P192 C15 GND – – – N4
I/O – P169 P193 B15 GND – – – N17 6
I/O – P170 P194 A15 GND – – – U4
GND P131 P171 P196 GND* GND – – – U8
I/O P132 P172 P197 B14 GND – – – U13
I/O P133 P173 P198 A14 GND – – – U17
I/O – – P199 C13 GND – – – B7
I/O – – P200 B13 GND – – – N3
VCC – – P201 VCC* GND – – – W14
I/O – – – A13 GND – – – G20
I/O – – – D12
I/O (A4) P134 P174 P202 C12 NC – P1 P195 –
I/O (A5) P135 P175 P203 B12 NC – P3 – –
GND – – P204 GND* NC – P51 – –
I/O – P176 P205 A12 NC – P52 – –
I/O P136 P177 P206 B11 NC – P53 – –
I/O (A21) P137 P178 P207 C11 NC – P54 – –
I/O (A20) P138 P179 P208 A11 NC – P102 – –
I/O (A6) P139 P180 P209 A10 NC – P104 – –
I/O (A7) P140 P181 P210 B10 NC – P105 – –
GND P141 P182 P211 GND* NC – P107 – –
NC – P155 – –
VCC – – – D6 NC – P156 – –
VCC – – – D11 NC – P157 – –
VCC – – – D15 NC – P158 – –
VCC – – – F4 NC – P206 – –
VCC – – – F17 NC – P207 – –
VCC – – – K4 NC – P208 – –
VCC – – – L17 12/18/98
VCC – – – R4
VCC – – – R17
VCC – – – U6
VCC – – – U10
VCC – – – U15
VCC – – – C14
VCC – – – F1

R

XC4028XLA Pinout Table PAD NAME HQ160 HQ208 HQ240 BG256 HQ304 BG352
I/O – – P13 F2 P290 H24
PAD NAME HQ160 HQ208 HQ240 BG256 HQ304 BG352
I/O – – – – P289 G25
VCC P142 P183 P212 VCC* P38 VCC*
I/O – – – – P288 G26
I/O (A8) P143 P184 P213 C10 P37 D14
GND P10 P14 P14 GND* P287 GND*
I/O (A9) P144 P185 P214 D10 P36 C14
I/O, FCLK1 P11 P15 P15 G3 P286 J23
I/O (A19) P145 P186 P215 A9 P35 A15
I/O P12 P16 P16 G2 P285 J24
I/O (A18) P146 P187 P216 B9 P34 B15
I/O (TMS) P13 P17 P17 G1 P284 H25
I/O – P188 P217 C9 P33 C15
I/O P14 P18 P18 H3 P283 K23
I/O – P189 P218 D9 P32 D15
VCC – – P19 VCC* P282 VCC*
I/O (A10) P147 P190 P220 A8 P31 A16
I/O – – P20 H2 P280 K24
I/O (A11) P148 P191 P221 B8 P30 B16 I/O – – P21 H1 P279 J25
GND – – – GND* – GND* I/O – – – – P278 L24
I/O – – – – P29 C16 I/O – – – – P277 K25
I/O – – – – P28 B17 GND – – P22 GND* GND*
I/O – – – C8 P27 C17 I/O – – – J4 P276 L25
I/O – – – A7 P26 B18
I/O – – – J3 P275 L26
I/O – P19 P23 J2 P274 M23
I/O – – P223 A6 P23 C18
I/O – P20 P24 J1 P273 M24
I/O – – P224 C7 P22 D17
I/O P15 P21 P25 K2 P272 M25
I/O P149 P192 P225 B6 P21 A20
I/O P16 P22 P26 K3 P271 M26
I/O P150 P193 P226 A5 P20 B19
I/O P17 P23 P27 K1 P270 N24
I/O P18 P24 P28 L1 P269 N25
I/O – – – – P18 C19
GND P19 P25 P29 GND* P268 GND* 6
I/O – – – – P17 D18
I/O – P195 P228 C6 P16 A21
I/O P21 P27 P31 L2 P266 N26
I/O – P196 P229 B5 P15 B20
I/O P22 P28 P32 L3 P265 P25
I/O P152 P197 P230 A4 P14 C20
I/O P23 P29 P33 L4 P264 P23
I/O P153 P198 P231 C5 P13 B21
I/O P24 P30 P34 M1 P263 P24
I/O (A12) P154 P199 P232 B4 P12 B22
I/O – P31 P35 M2 P262 R26
I/O (A13) P155 P200 P233 A3 P10 C21
I/O – P32 P36 M3 P261 R25
GND – – – GND* – GND*
I/O – – – M4 P260 R24
VCC – – – VCC* – VCC* I/O – – – P259 R23
I/O – – – – P9 D20 GND – – P37 GND* – GND*
I/O – – – – P8 A23 I/O – – – – P258 T26
I/O – – P234 D5 P7 D21 I/O – – – – P257 T25
I/O – – P235 C4 P6 C22 I/O – – P38 N1 P256 T23
I/O P156 P201 P236 B3 P5 B24
I/O – – P39 N2 P255 V26
I/O P157 P202 P237 B2 P4 C23
I/O (A14) P158 P203 P238 A2 P3 D22
I/O P25 P33 P41 P1 P252 U24
I/O, GCK8 (A15) P159 P204 P239 C3 P2 C24
I/O P26 P34 P42 P2 P251 V25
I/O P27 P35 P43 R1 P250 V24
I/O, FCLK2 P28 P36 P44 P3 P249 U23
I/O, GCK1 (A16) P2 P4 P2 B1 P303 D23
I/O (A17) P3 P5 P3 C2 P302 C25
I/O – – – – P247 Y26
I/O P4 P6 P4 D2 P301 D24
I/O – – – – P246 W25
I/O P5 P7 P5 D3 P300 E23
I/O – – P46 T1 P245 W24
I/O (TDI) P6 P8 P6 E4 P299 C26
I/O – – P47 R3 P244 V23
I/O (TCK) P7 P9 P7 C1 P298 E24
I/O – P38 P48 T2 P243 AA26
I/O – – – – P297 F24
I/O – P39 P49 U1 P242 Y25
I/O – – – – P296 E25
I/O P30 P40 P50 T3 P241 Y24
VCC – – – VCC* – VCC*
I/O P31 P41 P51 U2 P240 AA25
GND – – – GND* – GND* GND – – – GND* – GND*
I/O P8 P10 P8 D1 P295 D26 VCC – – – VCC* – VCC*
I/O P9 P11 P9 E3 P294 G24 I/O – – – – P239 AB25
I/O – P12 P10 E2 P293 F25 I/O – – – – P238 AA24
I/O – P13 P11 E1 P292 F26 I/O P32 P42 P52 V1 P237 Y23
I/O – – P12 F3 P291 H23
I/O P33 P43 P53 T4 P236 AC26

R
PAD NAME HQ160 HQ208 HQ240 BG256 HQ304 BG352 PAD NAME HQ160 HQ208 HQ240 BG256 HQ304 BG352
I/O P34 P44 P54 U3 P235 AA23 I/O – – – – P181 AE11
I/O P35 P45 P55 V2 P234 AB24 I/O – – – – P180 AD11
I/O P36 P46 P56 W1 P233 AD25 I/O – – P99 V13 P179 AF9
I/O, GCK2 P37 P47 P57 V3 P232 AC24 I/O – – P100 Y14 P178 AD10
O (M1) P38 P48 P58 W2 P231 AB23 VCC – – P101 VCC* P177 VCC*
GND P39 P49 P59 GND* P230 GND* I/O P66 P86 P102 Y15 P175 AE9
I (M0) P40 P50 P60 Y1 P229 AD24 I/O P67 P87 P103 V14 P174 AD9
VCC P41 P55 P61 VCC* P228 VCC* I/O P68 P88 P104 W15 P173 AC10
I (M2) P42 P56 P62 W3 P227 AC23 I/O P69 P89 P105 Y16 P172 AF7
I/O, GCK3 P43 P57 P63 Y2 P226 AE24 GND P70 P90 P106 GND* P171 GND*
I/O (HDC) P44 P58 P64 W4 P225 AD23 I/O – – – – P170 AE8
I/O P45 P59 P65 V4 P224 AC22 I/O – – – – P169 AD8
I/O P46 P60 P66 U5 P223 AF24 I/O – – – V15 P168 AC9
I/O P47 P61 P67 Y3 P222 AD22 I/O – – – W16 P167 AF6
I/O (/LDC) P48 P62 P68 Y4 P221 AE23 I/O – – – Y17 P166 AE7
I/O – – – – P220 AE22 I/O – P92 P110 V16 P165 AD7
I/O – – – – P219 AF23 I/O P71 P93 P111 W17 P164 AE6
VCC – – – VCC* – VCC* I/O P72 P94 P112 Y18 P163 AE5
GND – – – GND* – GND* GND – – – GND* – GND*
I/O P49 P63 P69 V5 P218 AD20 VCC – – – VCC* – VCC*
I/O P50 P64 P70 W5 P217 AE21 I/O – – – – P162 AD6
I/O – P65 P71 Y5 P216 AF21 I/O – – – – P161 AC7
I/O – P66 P72 V6 P215 AC19 I/O P73 P95 P113 U16 P160 AF4
I/O – – P73 W6 P214 AD19 I/O P74 P96 P114 V17 P159 AF3
I/O – – P74 Y6 P213 AE20 I/O P75 P97 P115 W18 P158 AD5
I/O – – – – P212 AF20 I/O P76 P98 P116 Y19 P157 AE3
I/O – – – – P211 AC18 I/O P77 P99 P117 V18 P156 AD4
GND P51 P67 P75 GND* P210 GND* I/O, GCK4 P78 P100 P118 W19 P155 AC5
I/O P52 P68 P76 W7 P209 AD18 GND P79 P101 P119 GND* P154 GND*
I/O P53 P69 P77 Y7 P208 AE19 DONE P80 P103 P120 Y20 P153 AD3
I/O P54 P70 P78 V8 P207 AC17 VCC P81 P106 P121 VCC* P152 VCC*
I/O P55 P71 P79 W8 P206 AD17 /PROGRAM P82 P108 P122 V19 P151 AC4
VCC – – P80 VCC* P204 VCC* I/O (D7) P83 P109 P123 U19 P150 AD2
I/O – P72 P81 Y8 P203 AE18 I/O, GCK5 P84 P110 P124 U18 P149 AC3
I/O – P73 P82 U9 P202 AF18 I/O P85 P111 P125 T17 P148 AB4
I/O – – – – P201 AE17 I/O P86 P112 P126 V20 P147 AD1
I/O – – – P200 AE16 I/O – – P127 U20 P146 AA4
GND – – P83 GND* – GND* I/O – – P128 T18 P145 AA3
I/O – – – V9 P199 AF16 I/O – – – – P144 AB2
I/O – – – W9 P198 AC15 I/O – – – – P143 AC1
I/O – – P84 Y9 P197 AD15 VCC – – – VCC* – VCC*
I/O – – P85 W10 P196 AE15 GND – – – GND* – GND*
I/O P56 P74 P86 V10 P195 AF15 I/O (D6) P87 P113 P129 T19 P142 Y3
I/O P57 P75 P87 Y10 P194 AD14 I/O P88 P114 P130 T20 P141 AA2
I/O P58 P76 P88 Y11 P193 AE14 I/O P89 P115 P131 R18 P140 AA1
I/O (/INIT) P59 P77 P89 W11 P192 AF14 I/O P90 P116 P132 R19 P139 W4
VCC P60 P78 P90 VCC* P191 VCC* I/O – P117 P133 R20 P138 W3
GND P61 P79 P91 GND* P190 GND* I/O – P118 P134 P18 P137 Y2
I/O P62 P80 P92 V11 P189 AE13 I/O – – – – P136 Y1
I/O P63 P81 P93 U11 P188 AC13 I/O – – – – P135 V4
I/O P64 P82 P94 Y12 P187 AD13 GND P91 P119 P135 GND* P134 GND*
I/O P65 P83 P95 W12 P186 AF12 I/O – – P136 P20 P133 V3
I/O – P84 P96 V12 P185 AE12 I/O – – P137 N18 P132 W2
I/O – P85 P97 U12 P184 AD12 I/O, FCLK3 P92 P120 P138 N19 P131 U4
I/O – – – Y13 P183 AC12 I/O P93 P121 P139 N20 P130 U3
I/O – – – W13 P182 AF11 VCC P140 VCC* P129 VCC*
GND – – P98 GND* – GND* I/O (D5) P94 P122 P141 M17 P127 V2

R
I/O (/CS0) P95 P123 P142 M18 P126 V1 I/O P125 P163 P185 C17 P72 D5
I/O – – – – P125 U2 I/O P126 P164 P186 D16 P71 A3
I/O – – – – P124 T2 I/O, (CS1, A2) P127 P165 P187 A18 P70 D6
GND – – P143* GND* – GND* I/O (A3) P128 P166 P188 A17 P69 C6
I/O – – – – P123 T1 I/O – – – – P68 B5
I/O – – – M19 P122 R4 I/O – – – – P67 A4
I/O – P124 P144 M20 P121 R3 VCC – – – VCC* – VCC*
I/O – P125 P145 L19 P120 R2 GND – – – GND* – GND*
I/O P96 P126 P146 L18 P119 R1 I/O – – P189 C16 P66 C7
I/O P97 P127 P147 L20 P118 P3 I/O – – P190 B16 P65 B6
I/O (D4) P98 P128 P148 K20 P117 P2 I/O P129 P167 P191 A16 P64 A6
I/O P99 P129 P149 K19 P116 P1 I/O P130 P168 P192 C15 P63 D8
VCC P100 P130 P150 VCC* P115 VCC* I/O – P169 P193 B15 P62 B7
GND P101 P131 P151 GND* P114 GND* I/O – P170 P194 A15 P61 A7
I/O (D3) P102 P132 P152 K18 P113 N2 I/O – – P195 –– P60 D9
I/O (/RS) P103 P133 P153 K17 P112 N4 I/O – – – – P59 C9
I/O P104 P134 P154 J20 P111 N3 GND P131 P171 P196 GND* P58 GND*
I/O P105 P135 P155 J19 P110 M1 I/O P132 P172 P197 B14 P57 B8
I/O – P136 P156 J18 P109 M2 I/O P133 P173 P198 A14 P56 D10
I/O – P137 P157 J17 P108 M3 I/O – – P199 C13 P55 C10
I/O – – – H20 P107 M4 I/O – – P200 B13 P54 B9
I/O – – – P106 L1 VCC* – – P201 VCC* P52 VCC*
GND – – – GND* – GND* I/O – – – A13 P51 A9 6
I/O – – – – P105 L2 I/O – – – D12 P50 D11
I/O – – – – P104 L3 I/O – – – – P49 B11
I/O (D2) P106 P138 P159 H19 P103 J1 I/O – – – – P48 A11
I/O P107 P139 P160 H18 P102 K3 GND – – – GND* – GND*
VCC P161 VCC* P101 VCC* I/O (A4) P134 P174 P202 C12 P47 D12
I/O P108 P140 P162 G19 P99 J2 I/O (A5) P135 P175 P203 B12 P46 C12
I/O, FCLK4 P109 P141 P163 F20 P98 J3 I/O P176 P205 A12 P45 B12
I/O – – P164 G18 P97 K4 I/O P136 P177 P206 B11 P44 A12
I/O – – P165 F19 P96 G1 I/O (A21) P137 P178 P207 C11 P43 C13
GND P110 P142 P166 GND* P95 GND* I/O (A20) P138 P179 P208 A11 P42 B13
I/O – – – – P94 H2 I/O (A6) P139 P180 P209 A10 P41 A13
I/O – – – – P93 H3 I/O (A7) P140 P181 P210 B10 P40 B14
I/O – – P167 F18 P92 J4 GND P141 P182 P211 GND* P39 GND*
I/O – – P168 E19 P91 F1 VCC – – – D6 – A10
I/O – P143 P169 D20 P90 G2 VCC – – – D11 – A17
I/O – P144 P170 E18 P89 G3 VCC – – – D15 – AC14
I/O P111 P145 P171 D19 P88 F2 VCC – – – F4 – AC20
I/O P112 P146 P172 C20 P87 E2 VCC – – – F17 – AC8
GND – – – GND* – GND* VCC – – – K4 – AF10
VCC – – – VCC* – VCC* VCC – – – L17 – AF17
I/O (D1) P113 P147 P173 E17 P86 F3 VCC – – – R4 – D7
I/O (/RCK, RDY_/BUSY) P114 P148 P174 D18 P85 G4 VCC – – – R17 – D13
I/O – – – – P84 D2 VCC – – – U6 – D19
I/O – – – – P83 F4 VCC – – – U10 – G23
I/O P115 P149 P175 C19 P82 E3 VCC – – – U15 – H4
I/O P116 P150 P176 B20 P81 C2 VCC – – – C14 – K1
I/O (D0, DIN) P117 P151 P177 C18 P80 D3 VCC – – – F1 – K26
I/O, GCK6 (DOUT) P118 P152 P178 B19 P79 E4 VCC – – – R2 – N23
CCLK P119 P153 P179 A20 P78 C3 VCC – – – E20 – P4
VCC P120 P154 P180 VCC* P77 VCC* VCC – – – P19 – U1
O, TDO P121 P159 P181 A19 P76 D4 VCC – – – V7 – U26
GND P122 P160 P182 GND* P75 GND* VCC – – – – – W23
I/O (A0, /WS) P123 P161 P183 B18 P74 B3 VCC – – – – – Y4
I/O, GCK7 (A1) P124 P162 P184 B17 P73 C4 VCC – – – D7 – B2

R
VCC – – – D14 – B25 NC – P158 – – – L23
VCC – – – G17 – AE2 NC – P206 – – – T3
VCC – – – P17 – AE25 NC – P207 – – – T4
VCC – – – W20 – – NC – P208 – – – T24
VCC – – – U14 – – NC – – – – – U25
VCC – – – U7 – – NC – – – – – AB3
VCC – – – P4 – – NC – – – – – AC2
VCC – – – G4 – – NC – – – – – AC6
NC – – – – – AC11
GND – – – A1 – A1 NC – – – – – AC16
GND – – – D4 – A14 NC – – – – – AC21
GND – – – D8 – A19 NC – – – – – AC25
GND – – – D13 – A2 NC – – – – – AD16
GND – – – D17 – A22 NC – – – – – AD21
GND – – – H4 – A25 NC – – – – – AD26
GND – – – H17 – A26 NC – – – – – AE4
GND – – – N4 – A5 NC – – – – – AE10
GND – – – N17 – A8 NC – – – – – C8
GND – – – U4 – AB1 NC – – – – – L4
GND – – – U8 – AB26 NC – – – – – K2
GND – – – U13 – AE1 12/18/98
GND – – – U17 – AE26
GND – – – B7 – AF1
GND – – – N3 – AF13
GND – – – W14 – AF19
GND – – – G20 – AF2
GND – – – – – AF22
GND – – – – – AF25
GND – – – – – AF26
GND – – – – – AF5
GND – – – – – AF8
GND – – – – – B1
GND – – – – – B26
GND – – – – – E1
GND – – – – – E26
GND – – – – – H1
GND – – – – – H26
GND – – – – – N1
GND – – – – – P26
GND – – – – – W1
GND – – – – – W26
GND – – P204 – – –
GND – – P219 – – –
NC – P1 – – P11 A18
NC – P3 – – P24 A24
NC – P51 – – P53 B4
NC – P52 – – P100 B10
NC – P53 – – P128 B23
NC – P54 – – P176 C1
NC – P102 – – P205 C5
NC – P104 – – P254 C11
NC – P105 – – P281 D1
NC – P107 – – – D16
NC – P155 – – – D25
NC – P156 – – – F23
NC – P157 – – – J26

XC4036XLA Pinout Table PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432
VCC – – – – VCC* VCC*
PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432
GND – – – – GND* GND*
VCC P142 P183 P212 P38 VCC* VCC*
I/O P8 P10 P8 P295 D26 H28
I/O (A8) P143 P184 P213 P37 D14 D17
I/O P9 P11 P9 P294 G24 H29
I/O (A9) P144 P185 P214 P36 C14 A17
I/O – P12 P10 P293 F25 G30
I/O (A19) P145 P186 P215 P35 A15 C18
I/O – P13 P11 P292 F26 H30
I/O (A18) P146 P187 P216 P34 B15 D18
I/O – – P12 P291 H23 J28
I/O – P188 P217 P33 C15 B18
I/O – – P13 P290 H24 J29
I/O – P189 P218 P32 D15 A19
I/O – – – P289 G25 H31
I/O (A10) P147 P190 P220 P31 A16 B19
I/O – – – P288 G26 J30
I/O (A11) P148 P191 P221 P30 B16 C19 GND P10 P14 P14 P287 GND* GND*
VCC – – – – VCC* VCC* I/O, FCLK1 P11 P15 P15 P286 J23 K28
GND – – – – GND* GND* I/O P12 P16 P16 P285 J24 K29
I/O – – – P29 C16 D19 I/O (TMS) P13 P17 P17 P284 H25 K30
I/O – – – P28 B17 A20 I/O P14 P18 P18 P283 K23 K31
I/O – – – – D16 B20
VCC – – P19 P282 VCC* VCC*
I/O – – – – A18 C20
I/O – – P20 P280 K24 L29
I/O – – – P27 C17 C21
I/O – P21 P279 J25 L30
I/O – – – P26 B18 A22
I/O – – – – J26 M29
I/O – – – – L23 M31
I/O – – P223 P23 C18 B22
I/O – – – P278 L24 N31
I/O – – P224 P22 D17 C22
I/O – – – P277 K25 N28
I/O P149 P192 P225 P21 A20 B23
GND P22 – GND* GND* 6
I/O P150 P193 P226 P20 B19 A24
GND P151 P194 P227 P19 GND* GND*
I/O – – – P276 L25 P30
I/O – – – P18 C19 D22
I/O – – – P275 L26 P28
I/O – – – P17 D18 C23
I/O – P19 P23 P274 M23 P29
I/O – P195 P228 P16 A21 B24
I/O – P20 P24 P273 M24 R31
I/O – P196 P229 P15 B20 C24
I/O P15 P21 P25 P272 M25 R30
I/O P152 P197 P230 P14 C20 A26
I/O P16 P22 P26 P271 M26 R28
I/O P153 P198 P231 P13 B21 C25
I/O P17 P23 P27 P270 N24 R29
I/O (A12) P154 P199 P232 P12 B22 D24 I/O P18 P24 P28 P269 N25 T31
I/O (A13) P155 P200 P233 P10 C21 B26 GND P19 P25 P29 P268 GND* GND*
GND – – – – GND* GND* VCC P20 P26 P30 P267 VCC* VCC*
VCC – – – – VCC* VCC* I/O P21 P27 P31 P266 N26 T30
I/O – – – P9 D20 A27 I/O P22 P28 P32 P265 P25 T29
I/O – – – P8 A23 D25
I/O P23 P29 P33 P264 P23 U31
I/O – – – – A24 C26
I/O P24 P30 P34 P263 P24 U30
I/O – – – – B23 B27
I/O – P31 P35 P262 R26 U28
I/O – – P234 P7 D21 C27
I/O – P32 P36 P261 R25 U29
I/O – – P235 P6 C22 B28
I/O – – – P260 R24 V30
I/O P156 P201 P236 P5 B24 D27
I/O – – – P259 R23 V29
I/O P157 P202 P237 P4 C23 B29
I/O (A14) P158 P203 P238 P3 D22 C28
GND – – P37 – GND* GND*
I/O, GCK8 (A15) P159 P204 P239 P2 C24 D28
I/O – – – P258 T26 W30
I/O – – – P257 T25 W29
I/O – – – – T24 Y30
I/O, GCK1 (A16) P2 P4 P2 P303 D23 D29
I/O – – – – U25 Y29
I/O (A17) P3 P5 P3 P302 C25 C30
I/O – – P38 P256 T23 Y28
I/O P4 P6 P4 P301 D24 E28
I/O – – P39 P255 V26 AA30
I/O P5 P7 P5 P300 E23 E29
I/O (TDI) P6 P8 P6 P299 C26 D30 I/O P25 P33 P41 P252 U24 AA29
I/O (TCK) P7 P9 P7 P298 E24 D31 I/O P26 P34 P42 P251 V25 AB31
I/O – – – – D25 E30 I/O P27 P35 P43 P250 V24 AB30
I/O – – – – F23 E31 I/O, FCLK2 P28 P36 P44 P249 U23 AB29
I/O – – – P297 F24 G28 GND P29 P37 P45 P248 GND* GND*
I/O – – – P296 E25 G29
I/O – – – P247 Y26 AB28

R
PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432 PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432
I/O – – – P246 W25 AC30 VCC – – – – VCC* VCC*
I/O – – P46 P245 W24 AC29 I/O – – – P199 AF16 AK19
I/O – – P47 P244 V23 AC28 I/O – – – P198 AC15 AJ18
I/O – P38 P48 P243 AA26 AD29 I/O – – P84 P197 AD15 AL19
I/O – P39 P49 P242 Y25 AD28 I/O – – P85 P196 AE15 AK18
I/O P30 P40 P50 P241 Y24 AE30 I/O P56 P74 P86 P195 AF15 AH17
I/O P31 P41 P51 P240 AA25 AE29 I/O P57 P75 P87 P194 AD14 AJ17
GND – – – – GND* GND* I/O P58 P76 P88 P193 AE14 AJ16
VCC – – – – VCC* VCC* I/O (/INIT) P59 P77 P89 P192 AF14 AK16
I/O – – – P239 AB25 AF31 VCC P60 P78 P90 P191 VCC* VCC*
I/O – – – P238 AA24 AE28 GND P61 P79 P91 P190 GND* GND*
I/O P32 P42 P52 P237 Y23 AG31 I/O P62 P80 P92 P189 AE13 AL16
I/O P33 P43 P53 P236 AC26 AF28 I/O P63 P81 P93 P188 AC13 AH15
I/O – – – – AD26 AG30 I/O P64 P82 P94 P187 AD13 AK15
I/O – – – – AC25 AG29 I/O P65 P83 P95 P186 AF12 AJ14
I/O P34 P44 P54 P235 AA23 AH31 I/O – P84 P96 P185 AE12 AH14
I/O P35 P45 P55 P234 AB24 AG28 I/O – P85 P97 P184 AD12 AK14
I/O P36 P46 P56 P233 AD25 AH30 I/O – – – P183 AC12 AL13
I/O, GCK2 P37 P47 P57 P232 AC24 AJ30 I/O – – – P182 AF11 AK13
O (M1) P38 P48 P58 P231 AB23 AH29 VCC – – – – VCC* VCC*
GND P39 P49 P59 P230 GND* GND* GND – – P98 – GND* GND*
I (M0) P40 P50 P60 P229 AD24 AH28 I/O – – – P181 AE11 AJ13
VCC P41 P55 P61 P228 VCC* VCC* I/O – – – P180 AD11 AH13
I (M2) P42 P56 P62 P227 AC23 AJ28 I/O – – – – AE10 AL12
I/O, GCK3 P43 P57 P63 P226 AE24 AK29 I/O – – – – AC11 AK12
I/O (HDC) P44 P58 P64 P225 AD23 AH27 I/O – – P99 P179 AF9 AH12
I/O P45 P59 P65 P224 AC22 AK28 I/O – – P100 P178 AD10 AJ11
I/O P46 P60 P66 P223 AF24 AJ27 VCC – – P101 P177 VCC* VCC*
I/O P47 P61 P67 P222 AD22 AL28 I/O P66 P86 P102 P175 AE9 AL10
I/O (/LDC) P48 P62 P68 P221 AE23 AH26 I/O P67 P87 P103 P174 AD9 AK10
I/O – – – – AC21 AL27 I/O P68 P88 P104 P173 AC10 AJ10
I/O – – – – AD21 AH25 I/O P69 P89 P105 P172 AF7 AK9
I/O – – – P220 AE22 AK26 GND P70 P90 P106 P171 GND* GND*
I/O – – – P219 AF23 AL26 I/O – – – P170 AE8 AL8
VCC – – – – VCC* VCC* I/O – – – P169 AD8 AH10
GND – – – – GND* GND* I/O – – P107 P168 AC9 AJ9
I/O P49 P63 P69 P218 AD20 AH24 I/O – – P108 P167 AF6 AK8
I/O P50 P64 P70 P217 AE21 AJ25 I/O – P91 P109 P166 AE7 AK7
I/O – P65 P71 P216 AF21 AK25 I/O – P92 P110 P165 AD7 AL6
I/O – P66 P72 P215 AC19 AJ24 I/O P71 P93 P111 P164 AE6 AJ7
I/O – – P73 P214 AD19 AL24 I/O P72 P94 P112 P163 AE5 AH8
I/O – – P74 P213 AE20 AH22 GND – – – – GND* GND*
I/O – – – P212 AF20 AJ23 VCC – – – – VCC* VCC*
I/O – – – P211 AC18 AK23 I/O – – – P162 AD6 AK6
GND P51 P67 P75 P210 GND* GND* I/O – – – P161 AC7 AL5
I/O P52 P68 P76 P209 AD18 AJ22 I/O P73 P95 P113 P160 AF4 AH7
I/O P53 P69 P77 P208 AE19 AK22 I/O P74 P96 P114 P159 AF3 AJ6
I/O P54 P70 P78 P207 AC17 AL22 I/O – – – – AE4 AK5
I/O P55 P71 P79 P206 AD17 AJ21 I/O – – – – AC6 AL4
VCC – – P80 P204 VCC* VCC* I/O P75 P97 P115 P158 AD5 AK4
I/O – P72 P81 P203 AE18 AH20 I/O P76 P98 P116 P157 AE3 AH5
I/O – P73 P82 P202 AF18 AK21 I/O P77 P99 P117 P156 AD4 AK3
I/O – – – – AC16 AK20 I/O, GCK4 P78 P100 P118 P155 AC5 AJ4
I/O – – – – AD16 AJ19 GND P79 P101 P119 P154 GND* GND*
I/O – – – P201 AE17 AL20 DONE P80 P103 P120 P153 AD3 AH4
I/O – – – P200 AE16 AH18 VCC P81 P106 P121 P152 VCC* VCC*
GND – – P83 – GND* GND* /PROGRAM P82 P108 P122 P151 AC4 AH3

R
I/O (D7) P83 P109 P123 P150 AD2 AJ2 I/O – – – – L4 M2
I/O, GCK5 P84 P110 P124 P149 AC3 AG4 I/O (D2) P106 P138 P159 P103 J1 L2
I/O P85 P111 P125 P148 AB4 AG3 I/O P107 P139 P160 P102 K3 L3
I/O P86 P112 P126 P147 AD1 AH2 VCC – – P161 P101 VCC* VCC*
I/O – – – – AB3 AH1 I/O P108 P140 P162 P99 J2 K1
I/O – – – – AC2 AF4 I/O, FCLK4 P109 P141 P163 P98 J3 K2
I/O – – P127 P146 AA4 AF3 I/O – – P164 P97 K4 K3
I/O – – P128 P145 AA3 AG2 I/O – – P165 P96 G1 K4
I/O – – – P144 AB2 AE3 GND P110 P142 P166 P95 GND* GND*
I/O – – – P143 AC1 AF2 I/O – – – P94 H2 J2
VCC – – – – VCC* VCC* I/O – – – P93 H3 J3
GND – – – – GND* GND* I/O – – P167 P92 J4 J4
I/O (D6) P87 P113 P129 P142 Y3 AF1 I/O – – P168 P91 F1 H1
I/O P88 P114 P130 P141 AA2 AD4 I/O – P143 P169 P90 G2 H2
I/O P89 P115 P131 P140 AA1 AD3 I/O – P144 P170 P89 G3 H3
I/O P90 P116 P132 P139 W4 AE2 I/O P111 P145 P171 P88 F2 H4
I/O – P117 P133 P138 W3 AC3 I/O P112 P146 P172 P87 E2 G2
I/O – P118 P134 P137 Y2 AD1 GND – – – – GND* GND*
I/O – – – P136 Y1 AC2 VCC – – – – VCC* VCC*
I/O – – – P135 V4 AB4 I/O (D1) P113 P147 P173 P86 F3 G4
GND P91 P119 P135 P134 GND* GND* I/O (/RCK, RDY_/BUSY) P114 P148 P174 P85 G4 F2
I/O – – P136 P133 V3 AB3 I/O – – – – D1 F3
I/O – – P137 P132 W2 AB2 I/O – – – – C1 E1 6
I/O, FCLK3 P92 P120 P138 P131 U4 AB1 I/O – – – P84 D2 E3
I/O P93 P121 P139 P130 U3 AA3 I/O – – – P83 F4 D1
VCC – – P140 P129 VCC* VCC* I/O P115 P149 P175 P82 E3 E4
I/O (D5) P94 P122 P141 P127 V2 AA2 I/O P116 P150 P176 P81 C2 D2
I/O (/CS0) P95 P123 P142 P126 V1 Y2 I/O (D0, DIN)” P117 P151 P177 P80 D3 C2
I/O – – – – T4 Y4 I/O, GCK6 (DOUT) P118 P152 P178 P79 E4 D3
I/O – – – – T3 Y3 CCLK P119 P153 P179 P78 C3 D4
I/O – – – P125 U2 W4 VCC P120 P154 P180 P77 VCC* VCC*
I/O – – – P124 T2 W3 O, TDO P121 P159 P181 P76 D4 C4
GND – – P143* – GND* GND* GND P122 P160 P182 P75 GND* GND*
VCC – – – – VCC* VCC* I/O (A0, /WS) P123 P161 P183 P74 B3 B3
I/O – – – P123 T1 V4 I/O, GCK7 (A1) P124 P162 P184 P73 C4 D5
I/O – – – P122 R4 V3 I/O P125 P163 P185 P72 D5 B4
I/O – P124 P144 P121 R3 U1 I/O P126 P164 P186 P71 A3 C5
I/O – P125 P145 P120 R2 U2 I/O – – – – C5 B5
I/O P96 P126 P146 P119 R1 U4 I/O – – – – B4 C6
I/O P97 P127 P147 P118 P3 U3 I/O, (CS1, A2) P127 P165 P187 P70 D6 A5
I/O (D4) P98 P128 P148 P117 P2 T1 I/O (A3) P128 P166 P188 P69 C6 D7
I/O P99 P129 P149 P116 P1 T2 I/O – – – P68 B5 B6
VCC P100 P130 P150 P115 VCC* VCC* I/O – – – P67 A4 A6
GND P101 P131 P151 P114 GND* GND* VCC – – – – VCC* VCC*
I/O (D3) P102 P132 P152 P113 N2 T3 GND – – – – GND* GND*
I/O (/RS) P103 P133 P153 P112 N4 R1 I/O – – P189 P66 C7 D8
I/O P104 P134 P154 P111 N3 R2 I/O – – P190 P65 B6 C7
I/O P105 P135 P155 P110 M1 R4 I/O P129 P167 P191 P64 A6 B7
I/O – P136 P156 P109 M2 R3 I/O P130 P168 P192 P63 D8 D9
I/O – P137 P157 P108 M3 P2 I/O – P169 P193 P62 B7 D10
I/O – – – P107 M4 P3 I/O – P170 P194 P61 A7 C9
I/O – – – P106 L1 P4 I/O – – P195 P60 D9 B9
VCC – – – – VCC* VCC* I/O – – – P59 C9 C10
GND – – P158 – GND* GND* GND P131 P171 P196 P58 GND* GND*
I/O – – – P105 L2 N3 I/O P132 P172 P197 P57 B8 B10
I/O – – – P104 L3 N4 I/O P133 P173 P198 P56 D10 A10
I/O – – – – K2 M1 I/O – – P199 P55 C10 C11

R
I/O – – P200 P54 B9 D12 GND – – – – AF1 B31
VCC – – P201 P52 VCC* VCC* GND – – – – AF13 C1
I/O – – – P51 A9 B11 GND – – – – AF19 C31
I/O – – – P50 D11 C12 GND – – – – AF2 D16
I/O – – – – C11 C13 GND – – – – AF22 G1
I/O – – – – B10 A12 GND – – – – AF25 G31
I/O – – – P49 B11 D14 GND – – – – AF26 J1
I/O – – – P48 A11 B13 GND – – – – AF5 J31
GND – – – – GND* GND* GND – – – – AF8 P1
VCC – – – – VCC* VCC* GND – – – – B1 P31
I/O (A4) P134 P174 P202 P47 D12 C14 GND – – – – B26 T4
I/O (A5) P135 P175 P203 P46 C12 A13 GND – – – – E1 T28
I/O – P176 P205 P45 B12 B14 GND – – – – E26 V1
I/O P136 P177 P206 P44 A12 D15 GND – – – – H1 V31
I/O (A21) P137 P178 P207 P43 C13 C15 GND – – – – H26 AC1
I/O (A20) P138 P179 P208 P42 B13 B15 GND – – – – N1 AC31
I/O (A6) P139 P180 P209 P41 A13 B16 GND – – – – P26 AE1
I/O (A7) P140 P181 P210 P40 B14 A16 GND – – – – W1 AE31
GND P141 P182 P211 P39 GND* GND* GND – – – – W26 AH16
VCC – – – – A10 A1 GND – – – – – AJ1
VCC – – – – A17 A11 GND – – – – – AJ31
VCC – – – – AC14 A21 GND – – – – – AK1
VCC – – – – AC20 A31 GND – – – – – AK2
VCC – – – – AC8 D11 GND – – – – – AK30
VCC – – – – AF10 D21 GND – – – – – AK31
VCC – – – – AF17 L1 GND – – – – – AL2
VCC – – – – D7 L4 GND – – – – – AL3
VCC – – – – D13 L28 GND – – – – – AL7
VCC – – – – D19 L31 GND – – – – – AL9
VCC – – – – G23 AA1 GND – – – – – AL14
VCC – – – – H4 AA4 GND – – – – – AL18
VCC – – – – K1 AA28 GND – – – – – AL23
VCC – – – – K26 AA31 GND – – – – – AL25
VCC – – – – N23 AH11 GND – – – – – AL29
VCC – – – – P4 AH21 GND – – – – – AL30
VCC – – – – U1 AL1 GND – P204 – – –
VCC – – – – U26 AL11 GND – P219 – – –
VCC – – – – W23 AL21
VCC – – – – Y4 AL31 NC – P1 – P11 C8 D26
VCC – – – – B2 C3 NC – P3 – P24 – A28
VCC – – – – B25 C29 NC – P51 – P53 – B25
VCC – – – – AE2 AJ3 NC – P52 – P100 – D23
VCC – – – – AE25 AJ29 NC – P53 – P128 – D20
NC – P54 – P176 – B21
GND – – – – A1 A2 NC – P102 – P205 – B17
GND – – – – A14 A3 NC – P104 – P254 – C17
GND – – – – A19 A7 NC – P105 – P281 – C16
GND – – – – A2 A9 NC – P107 – – – A15
GND – – – – A22 A14 NC – P155 – – – B12
GND – – – – A25 A18 NC – P156 – – – D13
GND – – – – A26 A23 NC – P157 – – – A8
GND – – – – A5 A25 NC – P158 – – – B8
GND – – – – A8 A29 NC – P206 – – – D6
GND – – – – AB1 A30 NC – P207 – – – A4
GND – – – – AB26 B1 NC – P208 – – – E2
GND – – – – AE1 B2 NC – – – – – F4
GND – – – – AE26 B30 NC – – – – – F1

R
XC4036XLA Pinout Table (Continued)

NC – – – – – G3
NC – – – – – M4
NC – – – – – M3
NC – – – – – N2
NC – – – – – N1
NC – – – – – V2
NC – – – – – W2
NC – – – – – W1
NC – – – – – Y1
NC – – – – – AC4
NC – – – – – AD2
NC – – – – – AE4
NC – – – – – AG1
NC – – – – – AJ5
NC – – – – – AH6
NC – – – – – AH9
NC – – – – – AJ8
NC – – – – – AK11
NC – – – – – AJ12
NC – – – – – AJ15
NC – – – – – AL15
NC – – – – – AL17
NC – – – – – AK17 6
NC – – – – – AH19
NC – – – – – AJ20
NC – – – – – AK24
NC – – – – – AH23
NC – – – – – AJ26
NC – – – – – AK27
NC – – – – – AF29
NC – – – – – AF30
NC – – – – – AD30
NC – – – – – AD31
NC – – – – – Y31
NC – – – – – W28
NC – – – – – W31
NC – – – – – V28
NC – – – – – N30
NC – – – – – N29
NC – – – – – M28
NC – – – – – M30
NC – – – – – F31
NC – – – – – F30
NC – – – – – F29
NC – – – – – F28
NC – – – – – C8
12/18/98

R

XC4044XLA Pinout Table PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432
I/O – – – – D25 E30
I/O – – – – F23 E31
I/O – – – P297 F24 G28
I/O (A8) P143 P184 P213 P37 D14 D17
I/O – – – P296 E25 G29
I/O (A9) P144 P185 P214 P36 C14 A17
I/O – – – – – C17
I/O – – – – – B17
I/O – – – – – F30
I/O (A19) P145 P186 P215 P35 A15 C18
I/O – – – – – F31
I/O (A18) P146 P187 P216 P34 B15 D18
I/O P8 P10 P8 P295 D26 H28
I/O – P188 P217 P33 C15 B18
I/O P9 P11 P9 P294 G24 H29
I/O – P189 P218 P32 D15 A19 I/O – P12 P10 P293 F25 G30
I/O (A10) P147 P190 P220 P31 A16 B19 I/O – P13 P11 P292 F26 H30
I/O (A11) P148 P191 P221 P30 B16 C19 I/O – – P12 P291 H23 J28
VCC – – – – VCC* VCC* I/O – – P13 P290 H24 J29
GND – – – – GND* GND* I/O – – P289 G25 H31
I/O – – – P29 C16 D19
I/O – – P288 G26 J30
I/O – – – P28 B17 A20
I/O – – – – D16 B20
I/O, FCLK1 P11 P15 P15 P286 J23 K28
I/O – – – – A18 C20
I/O P12 P16 P16 P285 J24 K29
I/O – – – P27 C17 C21
I/O (TMS) P13 P17 P17 P284 H25 K30
I/O – – – P26 B18 A22
I/O P14 P18 P18 P283 K23 K31
I/O – – P223 P23 C18 B22
I/O – – P20 P280 K24 L29 6
I/O – – P224 P22 D17 C22
I/O – – P21 P279 J25 L30
I/O P149 P192 P225 P21 A20 B23
I/O – – – – J26 M29
I/O P150 P193 P226 P20 B19 A24
I/O – – – – L23 M31
I/O – – – P278 L24 N31
I/O – – – P18 C19 D22
I/O – – – P277 K25 N28
I/O – – – P17 D18 C23
GND – – P22 – GND* GND*
I/O – P195 P228 P16 A21 B24
I/O – P196 P229 P15 B20 C24
I/O – – – – – N29
I/O – – – – – D23 I/O – – – – – N30
I/O – – – – – B25 I/O – – – P276 L25 P30
I/O P152 P197 P230 P14 C20 A26 I/O – – – P275 L26 P28
I/O P153 P198 P231 P13 B21 C25 I/O – P19 P23 P274 M23 P29
I/O (A12) P154 P199 P232 P12 B22 D24 I/O – P20 P24 P273 M24 R31
I/O (A13) P155 P200 P233 P10 C21 B26
I/O P15 P21 P25 P272 M25 R30
I/O P16 P22 P26 P271 M26 R28
I/O P17 P23 P27 P270 N24 R29
I/O – – – P9 D20 A27
I/O P18 P24 P28 P269 N25 T31
I/O – – – P8 A23 D25
I/O – – – – A24 C26
I/O – – – – B23 B27
I/O P21 P27 P31 P266 N26 T30
I/O – – P234 P7 D21 C27
I/O P22 P28 P32 P265 P25 T29
I/O – – P235 P6 C22 B28
I/O P23 P29 P33 P264 P23 U31
I/O P156 P201 P236 P5 B24 D27
I/O P24 P30 P34 P263 P24 U30
I/O P157 P202 P237 P4 C23 B29
I/O – P31 P35 P262 R26 U28
I/O (A14) P158 P203 P238 P3 D22 C28
I/O – P32 P36 P261 R25 U29
I/O, GCK8 (A15) P159 P204 P239 P2 C24 D28
I/O – – – P260 R24 V30
I/O – – – P259 R23 V29
I/O – – – – – V28
I/O, GCK1 (A16) P2 P4 P2 P303 D23 D29 I/O – – – – – W31
I/O (A17) P3 P5 P3 P302 C25 C30 VCC – – – – VCC* VCC*
I/O P4 P6 P4 P301 D24 E28 GND – – P37 – GND* GND*
I/O P5 P7 P5 P300 E23 E29 I/O – – – P258 T26 W30
I/O (TDI) P6 P8 P6 P299 C26 D30 I/O – – – P257 T25 W29
I/O (TCK) P7 P9 P7 P298 E24 D31
I/O – – – – T24 Y30

R
I/O – – – – U25 Y29 I/O – – – P211 AC18 AK23
I/O – – P38 P256 T23 Y28 GND P51 P67 P75 P210 GND* GND*
I/O – – P39 P255 V26 AA30 I/O P52 P68 P76 P209 AD18 AJ22
VCC – – P40 P253 VCC* VCC* I/O P53 P69 P77 P208 AE19 AK22
I/O P25 P33 P41 P252 U24 AA29 I/O P54 P70 P78 P207 AC17 AL22
I/O P26 P34 P42 P251 V25 AB31 I/O P55 P71 P79 P206 AD17 AJ21
I/O P27 P35 P43 P250 V24 AB30 VCC – – P80 P204 VCC* VCC*
I/O, FCLK2 P28 P36 P44 P249 U23 AB29 I/O – P72 P81 P203 AE18 AH20
GND P29 P37 P45 P248 GND* GND* I/O – P73 P82 P202 AF18 AK21
I/O – – – P247 Y26 AB28 I/O – – – – AC16 AK20
I/O – – – P246 W25 AC30 I/O – – – – AD16 AJ19
I/O – – P46 P245 W24 AC29 I/O – – – P201 AE17 AL20
I/O – – P47 P244 V23 AC28 I/O – – – P200 AE16 AH18
I/O – – – – – AD31 GND – – P83 – GND* GND*
I/O – – – – – AD30 VCC – – – – VCC* VCC*
I/O – P38 P48 P243 AA26 AD29 I/O – – – P199 AF16 AK19
I/O – P39 P49 P242 Y25 AD28 I/O – – – P198 AC15 AJ18
I/O P30 P40 P50 P241 Y24 AE30 I/O – – P84 P197 AD15 AL19
I/O P31 P41 P51 P240 AA25 AE29 I/O – – P85 P196 AE15 AK18
GND – – – – GND* GND* I/O P56 P74 P86 P195 AF15 AH17
VCC – – – – VCC* VCC* I/O P57 P75 P87 P194 AD14 AJ17
I/O – – – P239 AB25 AF31 I/O – – – – – AK17
I/O – – – P238 AA24 AE28 I/O – – – – – AL17
I/O P32 P42 P52 P237 Y23 AG31 I/O P58 P76 P88 P193 AE14 AJ16
I/O P33 P43 P53 P236 AC26 AF28 I/O (/INIT) P59 P77 P89 P192 AF14 AK16
I/O – – – – AD26 AG30 VCC P60 P78 P90 P191 VCC* VCC*
I/O – – – – AC25 AG29 GND P61 P79 P91 P190 GND* GND*
I/O P34 P44 P54 P235 AA23 AH31 I/O P62 P80 P92 P189 AE13 AL16
I/O P35 P45 P55 P234 AB24 AG28 I/O P63 P81 P93 P188 AC13 AH15
I/O P36 P46 P56 P233 AD25 AH30 I/O – – – – – AL15
I/O, GCK2 P37 P47 P57 P232 AC24 AJ30 I/O – – – – – AJ15
O (M1) P38 P48 P58 P231 AB23 AH29 I/O P64 P82 P94 P187 AD13 AK15
GND P39 P49 P59 P230 GND* GND* I/O P65 P83 P95 P186 AF12 AJ14
I (M0) P40 P50 P60 P229 AD24 AH28 I/O – P84 P96 P185 AE12 AH14
VCC P41 P55 P61 P228 VCC* VCC* I/O – P85 P97 P184 AD12 AK14
I (M2) P42 P56 P62 P227 AC23 AJ28 I/O – – – P183 AC12 AL13
I/O, GCK3 P43 P57 P63 P226 AE24 AK29 I/O – – – P182 AF11 AK13
I/O (HDC) P44 P58 P64 P225 AD23 AH27 VCC – – – – VCC* VCC*
I/O P45 P59 P65 P224 AC22 AK28 GND – – P98 – GND* GND*
I/O P46 P60 P66 P223 AF24 AJ27 I/O – – P181 AE11 AJ13
I/O P47 P61 P67 P222 AD22 AL28 I/O – – – P180 AD11 AH13
I/O (/LDC) P48 P62 P68 P221 AE23 AH26 I/O – – – – AE10 AL12
I/O – – – – AC21 AL27 I/O – – – – AC11 AK12
I/O – – – – AD21 AH25 I/O – – P99 P179 AF9 AH12
I/O – – P220 AE22 AK26 I/O – – P100 P178 AD10 AJ11
I/O – – – P219 AF23 AL26 VCC – – P101 P177 VCC* VCC*
VCC – – – – VCC* VCC* I/O P66 P86 P102 P175 AE9 AL10
GND – – – – GND* GND* I/O P67 P87 P103 P174 AD9 AK10
I/O P49 P63 P69 P218 AD20 AH24 I/O P68 P88 P104 P173 AC10 AJ10
I/O P50 P64 P70 P217 AE21 AJ25 I/O P69 P89 P105 P172 AF7 AK9
I/O – P65 P71 P216 AF21 AK25 GND P70 P90 P106 P171 GND* GND*
I/O – P66 P72 P215 AC19 AJ24 I/O – – – P170 AE8 AL8
I/O – – – – – AH23 I/O – – – P169 AD8 AH10
I/O – – – – – AK24 I/O – – P107 P168 AC9 AJ9
I/O – – P73 P214 AD19 AL24 I/O – – P108 P167 AF6 AK8
I/O – – P74 P213 AE20 AH22 I/O – – – – – AJ8
I/O – – – P212 AF20 AJ23 I/O – – – – – AH9

R
I/O – P91 P109 P166 AE7 AK7 I/O – – – – – V2
I/O – P92 P110 P165 AD7 AL6 I/O – – – P123 T1 V4
I/O P71 P93 P111 P164 AE6 AJ7 I/O – – – P122 R4 V3
I/O P72 P94 P112 P163 AE5 AH8 I/O – P124 P144 P121 R3 U1
GND – – – – GND* GND* I/O – P125 P145 P120 R2 U2
VCC – – – – VCC* VCC* I/O P96 P126 P146 P119 R1 U4
I/O – – – P162 AD6 AK6 I/O P97 P127 P147 P118 P3 U3
I/O – – – P161 AC7 AL5 I/O (D4) P98 P128 P148 P117 P2 T1
I/O P73 P95 P113 P160 AF4 AH7 I/O P99 P129 P149 P116 P1 T2
I/O P74 P96 P114 P159 AF3 AJ6 VCC P100 P130 P150 P115 VCC* VCC*
I/O – – – – AE4 AK5 GND P101 P131 P151 P114 GND* GND*
I/O – – – – AC6 AL4 I/O (D3) P102 P132 P152 P113 N2 T3
I/O P75 P97 P115 P158 AD5 AK4 I/O (/RS) P103 P133 P153 P112 N4 R1
I/O P76 P98 P116 P157 AE3 AH5 I/O P104 P134 P154 P111 N3 R2
I/O P77 P99 P117 P156 AD4 AK3 I/O P105 P135 P155 P110 M1 R4
I/O, GCK4 P78 P100 P118 P155 AC5 AJ4 I/O – P136 P156 P109 M2 R3
GND P79 P101 P119 P154 GND* GND* I/O – P137 P157 P108 M3 P2
DONE P80 P103 P120 P153 AD3 AH4 I/O – – – P107 M4 P3
VCC P81 P106 P121 P152 VCC* VCC* I/O – – – P106 L1 P4
/PROGRAM P82 P108 P122 P151 AC4 AH3 I/O – – – – – N1
I/O (D7) P83 P109 P123 P150 AD2 AJ2 I/O – – – – – N2
I/O, GCK5 P84 P110 P124 P149 AC3 AG4 VCC – – – – VCC* VCC*
I/O P85 P111 P125 P148 AB4 AG3 GND – – P158 – GND* GND* 6
I/O P86 P112 P126 P147 AD1 AH2 I/O – – – P105 L2 N3
I/O – – – – AB3 AH1 I/O – – – P104 L3 N4
I/O – – – – AC2 AF4 I/O – – – – K2 M1
I/O – – P127 P146 AA4 AF3 I/O – – – – L4 M2
I/O – – P128 P145 AA3 AG2 I/O (D2) P106 P138 P159 P103 J1 L2
I/O – – – P144 AB2 AE3 I/O P107 P139 P160 P102 K3 L3
I/O – – – P143 AC1 AF2 VCC P161 P101 VCC* VCC*
VCC – – – – VCC* VCC* I/O P108 P140 P162 P99 J2 K1
GND – – – – GND* GND* I/O, FCLK4 P109 P141 P163 P98 J3 K2
I/O (D6) P87 P113 P129 P142 Y3 AF1 I/O – – P164 P97 K4 K3
I/O P88 P114 P130 P141 AA2 AD4 I/O – – P165 P96 G1 K4
I/O P89 P115 P131 P140 AA1 AD3 GND P110 P142 P166 P95 GND* GND*
I/O P90 P116 P132 P139 W4 AE2 I/O – – – P94 H2 J2
I/O – – – – – AD2 I/O – – – P93 H3 J3
I/O – – – – – AC4 I/O – – P167 P92 J4 J4
I/O – P117 P133 P138 W3 AC3 I/O – – P168 P91 F1 H1
I/O – P118 P134 P137 Y2 AD1 I/O – P143 P169 P90 G2 H2
I/O – – – P136 Y1 AC2 I/O – P144 P170 P89 G3 H3
I/O – – – P135 V4 AB4 I/O P111 P145 P171 P88 F2 H4
GND P91 P119 P135 P134 GND* GND* I/O P112 P146 P172 P87 E2 G2
I/O – – P136 P133 V3 AB3 I/O – – – – – G3
I/O – – P137 P132 W2 AB2 I/O – – – – – F1
I/O, FCLK3 P92 P120 P138 P131 U4 AB1 GND – – – – GND* GND*
I/O P93 P121 P139 P130 U3 AA3 VCC – – – – VCC* VCC*
VCC – – P140 P129 VCC* VCC* I/O (D1) P113 P147 P173 P86 F3 G4
I/O (D5) P94 P122 P141 P127 V2 AA2 I/O (/RCK, P114 P148 P174 P85 G4 F2
I/O (/CS0) P95 P123 P142 P126 V1 Y2 RDY_/BUSY)
I/O – – – – T4 Y4 I/O – – – – D1 F3
I/O – – – – T3 Y3 I/O – – – – C1 E1
I/O – – – P125 U2 W4 I/O – – – P84 D2 E3
I/O – – – P124 T2 W3 I/O – – – P83 F4 D1
GND – – P143* – GND* GND* I/O P115 P149 P175 P82 E3 E4
VCC – – – – VCC* VCC* I/O P116 P150 P176 P81 C2 D2
I/O – – – – – W2 I/O (D0, DIN) P117 P151 P177 P80 D3 C2

R
I/O, GCK6 (DOUT) P118 P152 P178 P79 E4 D3 VCC – – – – AF10 D21
CCLK P119 P153 P179 P78 C3 D4 VCC – – – – AF17 L1
VCC P120 P154 P180 P77 VCC* VCC* VCC – – – – D7 L4
O, TDO P121 P159 P181 P76 D4 C4 VCC – – – – D13 L28
GND P122 P160 P182 P75 GND* GND* VCC – – – – D19 L31
I/O (A0, /WS) P123 P161 P183 P74 B3 B3 VCC – – – – G23 AA1
I/O, GCK7 (A1) P124 P162 P184 P73 C4 D5 VCC – – – – H4 AA4
I/O P125 P163 P185 P72 D5 B4 VCC – – – – K1 AA28
I/O P126 P164 P186 P71 A3 C5 VCC – – – – K26 AA31
I/O – – – – C5 B5 VCC – – – – N23 AH11
I/O – – – – B4 C6 VCC – – – – P4 AH21
I/O, (CS1, A2) P127 P165 P187 P70 D6 A5 VCC – – – – U1 AL1
I/O (A3) P128 P166 P188 P69 C6 D7 VCC – – – – U26 AL11
I/O – – – P68 B5 B6 VCC – – – – W23 AL21
I/O – – – P67 A4 A6 VCC – – – – Y4 AL31
VCC – – – – VCC* VCC* VCC – – – – B2 C3
GND – – – – GND* GND* VCC – – – – B25 C29
I/O – – P189 P66 C7 D8 VCC – – – – AE2 AJ3
I/O – – P190 P65 B6 C7 VCC – – – – AE25 AJ29
I/O P129 P167 P191 P64 A6 B7
I/O P130 P168 P192 P63 D8 D9 GND – – – – A1 A2
I/O – – – – C8 B8 GND – – – – A14 A3
I/O – – – – – A8 GND – – – – A19 A7
I/O – P169 P193 P62 B7 D10 GND – – – – A2 A9
I/O – P170 P194 P61 A7 C9 GND – – – – A22 A14
I/O – – P195 P60 D9 B9 GND – – – – A25 A18
I/O – – – P59 C9 C10 GND – – – – A26 A23
GND P131 P171 P196 P58 GND* GND* GND – – – – A5 A25
I/O P132 P172 P197 P57 B8 B10 GND – – – – A8 A29
I/O P133 P173 P198 P56 D10 A10 GND – – – – AB1 A30
I/O – – P199 P55 C10 C11 GND – – – – AB26 B1
I/O – – P200 P54 B9 D12 GND – – – – AE1 B2
VCC – – P201 P52 VCC* VCC* GND – – – – AE26 B30
I/O – – – P51 A9 B11 GND – – – – AF1 B31
I/O – – – P50 D11 C12 GND – – – – AF13 C1
I/O – – – – C11 C13 GND – – – – AF19 C31
I/O – – – – B10 A12 GND – – – – AF2 D16
I/O – – – P49 B11 D14 GND – – – – AF22 G1
I/O – – – P48 A11 B13 GND – – – – AF25 G31
GND – – – – GND* GND* GND – – – – AF26 J1
VCC – – – – VCC* VCC* GND – – – – AF5 J31
I/O (A4) P134 P174 P202 P47 D12 C14 GND – – – – AF8 P1
I/O (A5) P135 P175 P203 P46 C12 A13 GND – – – – B1 P31
I/O P176 P205 P45 B12 B14 GND – – – – B26 T4
I/O P136 P177 P206 P44 A12 D15 GND – – – – E1 T28
I/O (A21) P137 P178 P207 P43 C13 C15 GND – – – – E26 V1
I/O (A20) P138 P179 P208 P42 B13 B15 GND – – – – H1 V31
I/O – – – – – A15 GND – – – – H26 AC1
I/O – – – – – C16 GND – – – – N1 AC31
I/O (A6) P139 P180 P209 P41 A13 B16 GND – – – – P26 AE1
I/O (A7) P140 P181 P210 P40 B14 A16 GND – – – – W1 AE31
GND P141 P182 P211 P39 GND* GND* GND – – – – W26 AH16
VCC – – – – A10 A1 GND – – – – – AJ1
VCC – – – – A17 A11 GND – – – – – AJ31
VCC – – – – AC14 A21 GND – – – – – AK1
VCC – – – – AC20 A31 GND – – – – – AK2
VCC – – – – AC8 D11 GND – – – – – AK30

R

GND – – – – – AK31
GND – – – – – AL2
GND – – – – – AL3
GND – – – – – AL7
GND – – – – – AL9
GND – – – – – AL14
GND – – – – – AL18
GND – – – – – AL23
GND – – – – – AL25
GND – – – – – AL29
GND – – – – – AL30
GND – – P219 – – –
GND – – P204 – – –
NC – P1 – P11 – D26
NC – P3 – P24 – A28
NC – P51 – P53 – D20
NC – P52 – P100 – B21
NC – P53 – P128 – B12
NC – P54 – P176 – D13
NC – P102 – P205 – D6
NC – P104 – P254 – A4
NC – P105 – P281 – E2 6
NC – P107 – – – F4
NC – P155 – – – M4
NC – P156 – – – M3
NC – P157 – – – W1
NC – P158 – – – Y1
NC – P206 – – – AE4
NC – P207 – – – AG1
NC – P208 – – – AJ5
NC – – – – AH6
NC – – – – – AK11
NC – – – – – AJ12
NC – – – – – AH19
NC – – – – – AJ20
NC – – – – – AJ26
NC – – – – – AK27
NC – – – – – AF29
NC – – – – – AF30
NC – – – – – W28
NC – – – – – Y31
NC – – – – – M28
NC – – – – – M30
NC – – – – – F29
NC – – – – – F28
NC – – – – – C8
12/18/98

R

XC4052XLA Pinout Table PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432 BG560
VCC P160 P205 P240 P1 VCC* VCC* VCC*
PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432 BG560
GND P1 P2 P1 P304 GND* GND* GND*
I/O, GCK1 (A16) P2 P4 P2 P303 D23 D29 B33
I/O (A8) P143 P184 P213 P37 D14 D17 A17
I/O (A17) P3 P5 P3 P302 C25 C30 F29
I/O (A9) P144 P185 P214 P36 C14 A17 B18
I/O P4 P6 P4 P301 D24 E28 E30
I/O - - - - - C17 C18
I/O P5 P7 P5 P300 E23 E29 D31
I/O - - - - - B17 E18
I/O (TDI) P6 P8 P6 P299 C26 D30 F30
GND - - - - - GND* GND*
I/O (TCK) P7 P9 P7 P298 E24 D31 C33
I/O (A19) P145 P186 P215 P35 A15 C18 C19
GND - - - - - GND* GND*
I/O (A18) P146 P187 P216 P34 B15 D18 D19
I/O - - - - - F28 G29
I/O - P188 P217 P33 C15 B18 E19 I/O - - - - - F29 E31
I/O - P189 P218 P32 D15 A19 B20 I/O - - - - D25 E30 D32
I/O (A10) P147 P190 P220 P31 A16 B19 C20 I/O - - - - F23 E31 G30
I/O (A11) P148 P191 P221 P30 B16 C19 D20 I/O - - - P297 F24 G28 F31
VCC - - - - VCC* VCC* VCC* I/O - - - P296 E25 G29 H29
GND - - - - GND* GND* GND*
VCC - - - - VCC* VCC* VCC*
I/O - - - P29 C16 D19 A21
I/O - - - P28 B17 A20 E20
I/O - - - - - F30 H30
I/O - - - - D16 B20 B21
I/O - - - - - F31 G31
I/O - - - - A18 C20 C21
I/O P8 P10 P8 P295 D26 H28 J29
I/O - - - - - B21 D21
I/O P9 P11 P9 P294 G24 H29 F33
I/O - - - - - D20 B22
I/O - P12 P10 P293 F25 G30 G32
GND - - - - - GND* GND*
I/O - P13 P11 P292 F26 H30 J30 6
I/O - - - P27 C17 C21 C23
GND - - - - - GND* GND*
I/O - - - P26 B18 A22 E22
I/O - - P12 P291 H23 J28 K30
VCC - - P222 P25 VCC* VCC* VCC*
I/O - - P13 P290 H24 J29 H33
I/O - - P223 P23 C18 B22 B24
I/O - - - P289 G25 H31 L29
I/O - - P224 P22 D17 C22 D23
I/O - - - P288 G26 J30 K31
I/O P149 P192 P225 P21 A20 B23 C24
I/O P150 P193 P226 P20 B19 A24 A25
I/O, FCLK1 P11 P15 P15 P286 J23 K28 L30
I/O P12 P16 P16 P285 J24 K29 K32
I/O - - - P18 C19 D22 E23 I/O (TMS) P13 P17 P17 P284 H25 K30 J33
I/O - - - P17 D18 C23 B25 I/O P14 P18 P18 P283 K23 K31 M29
I/O - P195 P228 P16 A21 B24 D24 VCC - - P19 P282 VCC* VCC* VCC*
I/O - P196 P229 P15 B20 C24 C25 I/O - - P20 P280 K24 L29 L32
GND - - - - - GND* GND* I/O - - P21 P279 J25 L30 M31
I/O - - - - - D23 E25
GND - - - - - GND* GND*
I/O - - - - - B25 C27
I/O - - - - - M30 N29
I/O P152 P197 P230 P14 C20 A26 D26
I/O - - - - - M28 L33
I/O P153 P198 P231 P13 B21 C25 B28
I/O - - - - J26 M29 M32
I/O (A12) P154 P199 P232 P12 B22 D24 B29
I/O - - - - L23 M31 P29
I/O (A13) P155 P200 P233 P10 C21 B26 E26
I/O - - - P278 L24 N31 P30
I/O - - - P277 K25 N28 N33
GND - - P22 - GND* GND* GND*
I/O - - - P9 D20 A27 C28
I/O - - - P8 A23 D25 D27
I/O - - - - - N29 P31
I/O - - - - A24 C26 B30
I/O - - - - - N30 P32
I/O - - - - B23 B27 C29
I/O - - - P276 L25 P30 R29
I/O - - - - - A28 E27
I/O - - - P275 L26 P28 R30
I/O - - - - - D26 A31
I/O - P19 P23 P274 M23 P29 R31
GND - - - - - GND* GND*
I/O - P20 P24 P273 M24 R31 R33
I/O - - P234 P7 D21 C27 D28 GND - - - - - GND* GND*
I/O - - P235 P6 C22 B28 C30 I/O P15 P21 P25 P272 M25 R30 T31
I/O P156 P201 P236 P5 B24 D27 D29 I/O P16 P22 P26 P271 M26 R28 T29
I/O P157 P202 P237 P4 C23 B29 E28 I/O P17 P23 P27 P270 N24 R29 U32
I/O (A14) P158 P203 P238 P3 D22 C28 D30 I/O P18 P24 P28 P269 N25 T31 U31
I/O, GCK8 (A15) P159 P204 P239 P2 C24 D28 E29

R
PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432 BG560 PAD NAME HQ160 HQ208 HQ240 HQ304 BG352 BG432 BG560
VCC P20 P26 P30 P267 VCC* VCC* VCC* I (M0) P40 P50 P60 P229 AD24 AH28 AJ29
I/O P21 P27 P31 P266 N26 T30 U29 VCC P41 P55 P61 P228 VCC* VCC* VCC*
I/O P22 P28 P32 P265 P25 T29 U30 I (M2) P42 P56 P62 P227 AC23 AJ28 AN32
I/O P23 P29 P33 P264 P23 U31 V31 I/O, GCK3 P43 P57 P63 P226 AE24 AK29 AJ28
I/O P24 P30 P34 P263 P24 U30 V29 I/O (HDC) P44 P58 P64 P225 AD23 AH27 AK29
GND - - - - - GND* GND* I/O P45 P59 P65 P224 AC22 AK28 AL30
I/O - P31 P35 P262 R26 U28 V30 I/O P46 P60 P66 P223 AF24 AJ27 AK28
I/O - P32 P36 P261 R25 U29 W33 I/O P47 P61 P67 P222 AD22 AL28 AM31
I/O - - - P260 R24 V30 W31 I/O (/LDC) P48 P62 P68 P221 AE23 AH26 AJ27
I/O - - - P259 R23 V29 W30 GND - - - - - GND* GND*
I/O - - - - - V28 W29 I/O - - - - - AK27 AN31
I/O - - - - - W31 Y32 I/O - - - - - AJ26 AL29
VCC - - - - VCC* VCC* VCC* I/O - - - - AC21 AL27 AK27
GND - - P37 - GND* GND* GND* I/O - - - - AD21 AH25 AL28
I/O - - - P258 T26 W30 Y31 I/O - - - P220 AE22 AK26 AJ26
I/O - - - P257 T25 W29 Y30 I/O - - - P219 AF23 AL26 AM30
I/O - - - - - W28 AA32 VCC - - - - VCC* VCC* VCC
I/O - - - - - Y31 AA31 GND - - - - GND* GND* GND*
I/O - - - - T24 Y30 AA30 I/O P49 P63 P69 P218 AD20 AH24 AM29
I/O - - - - U25 Y29 AB32 I/O P50 P64 P70 P217 AE21 AJ25 AK26
GND - - - - - GND* GND* I/O - P65 P71 P216 AF21 AK25 AL27
I/O - - P38 P256 T23 Y28 AA29 I/O - P66 P72 P215 AC19 AJ24 AJ25
I/O - - P39 P255 V26 AA30 AB31 I/O - - - - - AH23 AN29
VCC - - P40 P253 VCC* VCC* VCC* I/O - - - - - AK24 AN28
I/O P25 P33 P41 P252 U24 AA29 AC31 GND - - - - - GND* GND*
I/O P26 P34 P42 P251 V25 AB31 AB29 I/O - - P73 P214 AD19 AL24 AL25
I/O P27 P35 P43 P250 V24 AB30 AD32 I/O - - P74 P213 AE20 AH22 AJ23
I/O, FCLK2 P28 P36 P44 P249 U23 AB29 AC30 I/O - - - P212 AF20 AJ23 AN26
GND P29 P37 P45 P248 GND* GND* GND* I/O - - - P211 AC18 AK23 AL24
I/O - - - P247 Y26 AB28 AD31 GND P51 P67 P75 P210 GND* GND* GND*
I/O - - - P246 W25 AC30 AE33 I/O P52 P68 P76 P209 AD18 AJ22 AK23
I/O - - P46 P245 W24 AC29 AC29 I/O P53 P69 P77 P208 AE19 AK22 AN25
I/O - - P47 P244 V23 AC28 AE32 I/O P54 P70 P78 P207 AC17 AL22 AJ22
GND - - - - - GND* GND* I/O P55 P71 P79 P206 AD17 AJ21 AL23
I/O - - - - - AD31 AG33 VCC - - P80 P204 VCC* VCC* VCC*
I/O - - - - - AD30 AH33 I/O - P72 P81 P203 AE18 AH20 AM24
I/O - P38 P48 P243 AA26 AD29 AE29 I/O - P73 P82 P202 AF18 AK21 AK22
I/O - P39 P49 P242 Y25 AD28 AG31 GND - - - - - GND* GND*
I/O P30 P40 P50 P241 Y24 AE30 AF30 I/O - - - - - AJ20 AK21
I/O P31 P41 P51 P240 AA25 AE29 AH32 I/O - - - - - AH19 AM22
GND - - - - GND* GND* GND* I/O - - - - AC16 AK20 AJ20
VCC - - - - VCC* VCC* VCC* I/O - - - - AD16 AJ19 AL21
I/O - - - P239 AB25 AF31 AJ32 I/O - - - P201 AE17 AL20 AN21
I/O - - - P238 AA24 AE28 AF29 I/O - - - P200 AE16 AH18 AK20
I/O - - - - - AF30 AH31 GND - - P83 - GND* GND* GND*
I/O - - - - - AF29 AG30 VCC - - - - VCC* VCC* VCC*
I/O P32 P42 P52 P237 Y23 AG31 AK32 I/O - - - P199 AF16 AK19 AL20
I/O P33 P43 P53 P236 AC26 AF28 AJ31 I/O - - - P198 AC15 AJ18 AJ19
GND - - - - - GND* GND* I/O - - P84 P197 AD15 AL19 AM20
I/O - - - - AD26 AG30 AG29 I/O - - P85 P196 AE15 AK18 AK19
I/O - - - - AC25 AG29 AL33 I/O P56 P74 P86 P195 AF15 AH17 AL19
I/O P34 P44 P54 P235 AA23 AH31 AH30 I/O P57 P75 P87 P194 AD14 AJ17 AN19
I/O P35 P45 P55 P234 AB24 AG28 AK31 GND - - - - - GND* GND*
I/O P36 P46 P56 P233 AD25 AH30 AJ30 I/O - - - - - AK17 AL18
I/O, GCK2 P37 P47 P57 P232 AC24 AJ30 AH29 I/O - - - - - AL17 AM18
O (M1) P38 P48 P58 P231 AB23 AH29 AK30 I/O P58 P76 P88 P193 AE14 AJ16 AK17
GND P39 P49 P59 P230 GND* GND* GND* I/O (/INIT) P59 P77 P89 P192 AF14 AK16 AJ17

R
VCC P60 P78 P90 P191 VCC* VCC* VCC* DONE P80 P103 P120 P153 AD3 AH4 AJ5
GND P61 P79 P91 P190 GND* GND* GND* VCC P81 P106 P121 P152 VCC* VCC* VCC*
I/O P62 P80 P92 P189 AE13 AL16 AL17 /PROGRAM P82 P108 P122 P151 AC4 AH3 AM1
I/O P63 P81 P93 P188 AC13 AH15 AM17 I/O (D7) P83 P109 P123 P150 AD2 AJ2 AH5
I/O - - - - - AL15 AN17 I/O, GCK5 P84 P110 P124 P149 AC3 AG4 AJ4
I/O - - - - - AJ15 AK16 I/O P85 P111 P125 P148 AB4 AG3 AK3
GND - - - - - GND* GND* I/O P86 P112 P126 P147 AD1 AH2 AH4
I/O P64 P82 P94 P187 AD13 AK15 AM16 I/O - - - - AB3 AH1 AL1
I/O P65 P83 P95 P186 AF12 AJ14 AL15 I/O - - - - AC2 AF4 AG5
I/O - P84 P96 P185 AE12 AH14 AK15 GND - - - - - GND* GND*
I/O - P85 P97 P184 AD12 AK14 AJ15 I/O - - P127 P146 AA4 AF3 AJ3
I/O - - - P183 AC12 AL13 AN15 I/O - - P128 P145 AA3 AG2 AK2
I/O - - - P182 AF11 AK13 AM14 I/O - - - - - AG1 AG4
VCC - - - - VCC* VCC* VCC* I/O - - - - - AE4 AH3
GND - - P98 - GND* GND* GND* I/O - - - P144 AB2 AE3 AF5
I/O - - - P181 AE11 AJ13 AL14 I/O - - - P143 AC1 AF2 AJ2
I/O - - - P180 AD11 AH13 AK14 VCC - - - - VCC* VCC* VCC*
I/O - - - - AE10 AL12 AJ14 GND - - - - GND* GND* GND*
I/O - - - - AC11 AK12 AN13 I/O (D6) P87 P113 P129 P142 Y3 AF1 AJ1
I/O - - - - - AJ12 AM13 I/O P88 P114 P130 P141 AA2 AD4 AF4
I/O - - - - - AK11 AL13 I/O P89 P115 P131 P140 AA1 AD3 AG3
GND - - - - - GND* GND* I/O P90 P116 P132 P139 W4 AE2 AE5
I/O - - P99 P179 AF9 AH12 AK12 I/O - - - - - AD2 AH1 6
I/O - - P100 P178 AD10 AJ11 AN11 I/O - - - - - AC4 AF3
VCC - - P101 P177 VCC* VCC* VCC* GND - - - - - GND* GND*
I/O P66 P86 P102 P175 AE9 AL10 AJ12 I/O - P117 P133 P138 W3 AC3 AE3
I/O P67 P87 P103 P174 AD9 AK10 AL11 I/O - P118 P134 P137 Y2 AD1 AC5
I/O P68 P88 P104 P173 AC10 AJ10 AK11 I/O - - - P136 Y1 AC2 AE1
I/O P69 P89 P105 P172 AF7 AK9 AM10 I/O - - - P135 V4 AB4 AD3
GND P70 P90 P106 P171 GND* GND* GND* GND P91 P119 P135 P134 GND* GND* GND*
I/O - - - P170 AE8 AL8 AL10 I/O - - P136 P133 V3 AB3 AC4
I/O - - - P169 AD8 AH10 AJ11 I/O - - P137 P132 W2 AB2 AD2
I/O - - P107 P168 AC9 AJ9 AN9 I/O, FCLK3 P92 P120 P138 P131 U4 AB1 AB5
I/O - - P108 P167 AF6 AK8 AK10 I/O P93 P121 P139 P130 U3 AA3 AC3
GND - - - - - GND* GND* VCC - - P140 P129 VCC* VCC* VCC*
I/O - - - - - AJ8 AN7 I/O (D5) P94 P122 P141 P127 V2 AA2 AA5
I/O - - - - - AH9 AJ9 I/O (/CS0) P95 P123 P142 P126 V1 Y2 AB3
I/O - P91 P109 P166 AE7 AK7 AL7 GND - - - - - GND* GND*
I/O - P92 P110 P165 AD7 AL6 AK8 I/O - - - - T4 Y4 AB2
I/O P71 P93 P111 P164 AE6 AJ7 AN6 I/O - - - - T3 Y3 AA4
I/O P72 P94 P112 P163 AE5 AH8 AM6 I/O - - - - - Y1 AA3
GND - - - - GND* GND* GND* I/O - - - - - W1 Y5
VCC - - - - VCC* VCC* VCC* I/O - - - P125 U2 W4 Y3
I/O - - - P162 AD6 AK6 AJ8 I/O - - - P124 T2 W3 Y2
I/O - - - P161 AC7 AL5 AL6 GND - - P143* - GND* GND* GND*
I/O P73 P95 P113 P160 AF4 AH7 AK7 VCC - - - - VCC* VCC* VCC*
I/O P74 P96 P114 P159 AF3 AJ6 AM5 I/O - - - - - W2 W5
I/O - - - - AE4 AK5 AM4 I/O - - - - - V2 W4
I/O - - - - AC6 AL4 AJ7 I/O - - - P123 T1 V4 W3
GND - - - - - GND* GND* I/O - - - P122 R4 V3 W1
I/O - - - - - AH6 AL5 I/O - P124 P144 P121 R3 U1 V3
I/O - - - - - AJ5 AK6 I/O - P125 P145 P120 R2 U2 V5
I/O P75 P97 P115 P158 AD5 AK4 AN3 GND - - - - - GND* GND*
I/O P76 P98 P116 P157 AE3 AH5 AK5 I/O P96 P126 P146 P119 R1 U4 V4
I/O P77 P99 P117 P156 AD4 AK3 AJ6 I/O P97 P127 P147 P118 P3 U3 V2
I/O, GCK4 P78 P100 P118 P155 AC5 AJ4 AL4 I/O (D4) P98 P128 P148 P117 P2 T1 U5
GND P79 P101 P119 P154 GND* GND* GND* I/O P99 P129 P149 P116 P1 T2 U4

R
VCC P100 P130 P150 P115 VCC* VCC* VCC* I/O, GCK6 P118 P152 P178 P79 E4 D3 D4
GND P101 P131 P151 P114 GND* GND* GND* (DOUT)
I/O (D3) P102 P132 P152 P113 N2 T3 U3 CCLK P119 P153 P179 P78 C3 D4 C4
I/O (/RS) P103 P133 P153 P112 N4 R1 T2 VCC P120 P154 P180 P77 VCC* VCC* VCC*
I/O P104 P134 P154 P111 N3 R2 T4 O, TDO P121 P159 P181 P76 D4 C4 E6
I/O P105 P135 P155 P110 M1 R4 R1 GND P122 P160 P182 P75 GND* GND* GND*
GND - - - - - GND* GND* I/O (A0, /WS) P123 P161 P183 P74 B3 B3 D5
I/O - P136 P156 P109 M2 R3 R3 I/O, GCK7 (A1) P124 P162 P184 P73 C4 D5 A2
I/O - P137 P157 P108 M3 P2 R4 I/O P125 P163 P185 P72 D5 B4 D6
I/O - - - P107 M4 P3 R5 I/O P126 P164 P186 P71 A3 C5 A3
I/O - - - P106 L1 P4 P2 I/O - - - - - A4 E7
I/O - - - - - N1 P3 I/O - - - - - D6 C5
I/O - - - - - N2 P4 GND - - - - - GND* GND*
VCC - - - - VCC* VCC* VCC* I/O - - - - C5 B5 B4
GND - - P158 - GND* GND* GND* I/O - - - - B4 C6 D7
I/O - - - P105 L2 N3 N1 I/O, (CS1, A2) P127 P165 P187 P70 D6 A5 C6
I/O - - - P104 L3 N4 P5 I/O (A3) P128 P166 P188 P69 C6 D7 E8
I/O - - - - K2 M1 N2 I/O - - - P68 B5 B6 B5
I/O - - - - L4 M2 N3 I/O - - - P67 A4 A6 A5
I/O - - - - - M3 N5 VCC - - - - VCC* VCC* VCC*
I/O - - - - - M4 M3 GND - - - - GND* GND* GND*
GND - - - - - GND* GND* I/O - - P189 P66 C7 D8 D8
I/O (D2) P106 P138 P159 P103 J1 L2 M4 I/O - - P190 P65 B6 C7 C7
I/O P107 P139 P160 P102 K3 L3 L1 I/O P129 P167 P191 P64 A6 B7 E9
VCC - - P161 P101 VCC* VCC* VCC* I/O P130 P168 P192 P63 D8 D9 A6
I/O P108 P140 P162 P99 J2 K1 K2 I/O - - - - C8 B8 B7
I/O, FCLK4 P109 P141 P163 P98 J3 K2 L4 I/O - - - - - A8 D9
I/O - - P164 P97 K4 K3 J1 GND - - - - - GND* GND*
I/O - - P165 P96 G1 K4 K3 I/O - P169 P193 P62 B7 D10 E11
GND P110 P142 P166 P95 GND* GND* GND* I/O - P170 P194 P61 A7 C9 A9
I/O - - - P94 H2 J2 L5 I/O - - P195 P60 D9 B9 C10
I/O - - - P93 H3 J3 J2 I/O - - - P59 C9 C10 D11
I/O - - P167 P92 J4 J4 K4 GND P131 P171 P196 P58 GND* GND* GND*
I/O - - P168 P91 F1 H1 J3 I/O P132 P172 P197 P57 B8 B10 B10
GND - - - - - GND* GND* I/O P133 P173 P198 P56 D10 A10 E12
I/O - P143 P169 P90 G2 H2 G1 I/O - - P199 P55 C10 C11 C11
I/O - P144 P170 P89 G3 H3 F1 I/O - - P200 P54 B9 D12 B11
I/O P111 P145 P171 P88 F2 H4 J5 VCC - - P201 P52 VCC* VCC* VCC*
I/O P112 P146 P172 P87 E2 G2 G3 I/O - - - P51 A9 B11 D12
I/O - - - - - G3 H4 I/O - - - P50 D11 C12 A11
I/O - - - - - F1 F2 GND - - - - - GND* GND*
GND - - - - GND* GND* GND* I/O - - - - - D13 C13
VCC - - - - VCC* VCC* VCC* I/O - - - - - B12 E14
I/O (D1) P113 P147 P173 P86 F3 G4 F3 I/O - - - - C11 C13 A13
I/O (/RCK, P114 P148 P174 P85 G4 F2 G4 I/O - - - - B10 A12 D14
RDY_/BUSY) I/O - - - P49 B11 D14 C14
I/O - - - - D1 F3 D2 I/O - - - P48 A11 B13 B14
I/O - - - - C1 E1 E3 GND - - - - GND* GND* GND*
I/O - - - - - F4 G5 VCC - - - - VCC* VCC* VCC*
I/O - - - - - E2 C1 I/O (A4) P134 P174 P202 P47 D12 C14 E15
GND - - - - - GND* GND* I/O (A5) P135 P175 P203 P46 C12 A13 D15
I/O - - - P84 D2 E3 F4 I/O - P176 P205 P45 B12 B14 C15
I/O - - - P83 F4 D1 D3 I/O P136 P177 P206 P44 A12 D15 A15
I/O P115 P149 P175 P82 E3 E4 B3 I/O (A21) P137 P178 P207 P43 C13 C15 C16
I/O P116 P150 P176 P81 C2 D2 F5 I/O (A20) P138 P179 P208 P42 B13 B15 E16
I/O (D0, DIN) P117 P151 P177 P80 D3 C2 E4 GND - - - - - GND* GND*
I/O - - - - - A15 B17

R
I/O - - - - - C16 C17 GND - - - - A8 A29 B1
I/O (A6) P139 P180 P209 P41 A13 B16 E17 GND - - - - AB1 A30 B6
I/O (A7) P140 P181 P210 P40 B14 A16 D17 GND - - - - AB26 B1 B9
GND P141 P182 P211 P39 GND* GND* GND* GND - - - - AE1 B2 B15
VCC - - - - A10 A1 A4 GND - - - - AE26 B30 B23
VCC - - - - A17 A11 A10 GND - - - - AF1 B31 B27
VCC - - - - AC14 A21 A16 GND - - - - AF13 C1 B31
VCC - - - - AC20 A31 A22 GND - - - - AF19 C31 C2
VCC - - - - AC8 D11 A26 GND - - - - AF2 D16 E1
VCC - - - - AF10 D21 A30 GND - - - - AF22 G1 F32
VCC - - - - AF17 L1 B2 GND - - - - AF25 G31 G2
VCC - - - - D7 L4 B13 GND - - - - AF26 J1 G33
VCC - - - - D13 L28 B19 GND - - - - AF5 J31 J32
VCC - - - - D19 L31 B32 GND - - - - AF8 P1 K1
VCC - - - - G23 AA1 C3 GND - - - - B1 P31 L2
VCC - - - - H4 AA4 C32 GND - - - - B26 T4 M33
VCC - - - - K1 AA28 D1 GND - - - - E1 T28 P1
VCC - - - - K26 AA31 D33 GND - - - - E26 V1 P33
VCC - - - - N23 AH11 H1 GND - - - - H1 V31 R32
VCC - - - - P4 AH21 K33 GND - - - - H26 AC1 T1
VCC - - - - U1 AL1 M1 GND - - - - N1 AC31 V33
VCC - - - - U26 AL11 N32 GND - - - - P26 AE1 W2
VCC - - - - W23 AL21 R2 GND - - - - W1 AE31 Y1 6
VCC - - - - Y4 AL31 T33 GND - - - - W26 AH16 Y33
VCC - - - - B2 - V1 GND - - - - - AJ1 AB1
VCC - - - - B25 - W32 GND - - - - - AJ31 AC32
VCC - - - - AE2 - AA2 GND - - - - - AK1 AD33
VCC - - - - AE25 - AB33 GND - - - - - AK2 AE2
VCC - - - - - - AD1 GND - - - - - AK30 AG1
VCC - - - - - - AF33 GND - - - - - AK31 AG32
VCC - - - - - - AK1 GND - - - - - AL2 AH2
VCC - - - - - - AK33 GND - - - - - AL3 AJ33
VCC - - - - - - AL2 GND - - - - - AL7 AL32
VCC - - - - - - AL3 GND - - - - - AL9 AM3
VCC - - - - - - AM2 GND - - - - - AL14 AM11
VCC - - - - - - AM15 GND - - - - - AL18 AM19
VCC - - - - - - AM21 GND - - - - - AL23 AM25
VCC - - - - - - AM32 GND - - - - - AL25 AM28
VCC - - - - - - AN4 GND - - - - - AL29 AM33
VCC - - - - - - AN8 GND - - - - - AL30 AM7
VCC - - - - - - AN12 GND - - - - - - AN2
VCC - - - - - - AN18 GND - - - - - - AN5
VCC - - - - - - AN24 GND - - - - - - AN10
VCC - - - - - - AN30 GND - - - - - - AN14
VCC - - - - - C3 AL31 GND - - - - - - AN16
VCC - - - - - C29 E5 GND - - - - - - AN20
VCC - - - - - AJ3 C31 GND - - - - - - AN22
VCC - - - - - AJ29 AK4 GND - - - - - - AN27
GND - - P219 - - - -
GND - - - - A1 A2 A7 GND - - P204 - - - -
GND - - - - A14 A3 A12
GND - - - - A19 A7 A14 NC - P1 - P11 - C8 A28
GND - - - - A2 A9 A18 NC - P3 - P24 - - A27
GND - - - - A22 A14 A20 NC - P51 - P53 - - D25
GND - - - - A25 A18 A24 NC - P52 - P100 - - C26
GND - - - - A26 A23 A29 NC - P53 - P128 - - A23
GND - - - - A5 A25 A32 NC - P54 - P176 - - D22

R
NC - P102 - P205 - - C22 NC - - - - - - E32
NC - P104 - P254 - - E21 NC - - - - - - AC2
NC - P105 - P281 - - D13 NC - - - - - - A1
NC - P107 - - - - B12 NC - - - - - - A33
NC - P155 - - - - C12 NC - - - - - - AN1
NC - P156 - - - - E13 NC - - - - - - AN33
NC - P157 - - - - A8 NC - - - - - - A19
NC - P158 - - - - B8 NC - - - - - - D18
NC - P206 - - - E10 NC - - - - - - E24
NC - P207 - - - - C8 NC - - - - - - B26
NC - P208 - - - - H5 NC - - - - - - K5
NC - - - - - - E2 NC - - - - - - H2
NC - - - - - - J4 NC - - - - - - B16
NC - - - - - - H3 NC - - - - - - D16
NC - - - - - - M5 NC - - - - - - T5
NC - - - - - - L3 NC - - - - - - T3
NC - - - - - - M2 NC - - - - - - U1
NC - - - - - - N4 NC - - - - - - U2
NC - - - - - - Y4 NC - - - - - - AD4
NC - - - - - - AA1 NC - - - - - - AF1
NC - - - - - - AC1 NC - - - - - - AL9
NC - - - - - - AB4 NC - - - - - - AM9
NC - - - - - - AF2 NC - - - - - - AL16
NC - - - - - - AD5 NC - - - - - - AJ16
NC - - - - - - AG2 NC - - - - - - AK18
NC - - - - - - AE4 NC - - - - - - AJ18
NC - - - - - - AL8 NC - - - - - - AK24
NC - - - - - - AK9 NC - - - - - - AM26
NC - - - - - - AM8 NC - - - - - - AE31
NC - - - - - - AJ10 NC - - - - - - AD30
NC - - - - - - AL12 NC - - - - - - V32
NC - - - - - - AM12 NC - - - - - - U33
NC - - - - - - AJ13 NC - - - - - - T32
NC - - - - - - AK13 NC - - - - - - T30
NC - - - - - - AN23 NC - - - - - - J31
NC - - - - - - AL22 NC - - - - - - H32
NC - - - - - - AJ21 NC - - - - - - C9
NC - - - - - - AM23 NC - - - - - - D10
NC - - - - - - AM27 12/18/98
NC - - - - - - AJ24
NC - - - - - - AL26
NC - - - - - - AK25
NC - - - - - - AE30
NC - - - - - - AF31
NC - - - - - - AD29
NC - - - - - - AF32
NC - - - - - - AC33
NC - - - - - - AB30
NC - - - - - - Y29
NC - - - - - - AA33
NC - - - - - - N31
NC - - - - - - N30
NC - - - - - - M30
NC - - - - - - L31
NC - - - - - - K29
NC - - - - - - H31
NC - - - - - - E33

I/O P156 P201 P236 P5 B24 D27 D29
I/O P157 P202 P237 P4 C23 B29 E28
VCC P142 P183 P212 P38 VCC* VCC* VCC* I/O (A14) P158 P203 P238 P3 D22 C28 D30
I/O (A8) P143 P184 P213 P37 D14 D17 A17 I/O, GCK8 (A15) P159 P204 P239 P2 C24 D28 E29
I/O (A9) P144 P185 P214 P36 C14 A17 B18 VCC P160 P205 P240 P1 VCC* VCC* VCC*
I/O - - - - - C17 C18 GND P1 P2 P1 P304 GND* GND* GND*
I/O - - - - - B17 E18 I/O, GCK1 (A16) P2 P4 P2 P303 D23 D29 B33
I/O - - - - - - D18
I/O (A17) P3 P5 P3 P302 C25 C30 F29
I/O - - - - - - A19
I/O P4 P6 P4 P301 D24 E28 E30
GND - - - - - GND* GND*
I/O P5 P7 P5 P300 E23 E29 D31
I/O (A19) P145 P186 P215 P35 A15 C18 C19
I/O (TDI) P6 P8 P6 P299 C26 D30 F30
I/O (A18) P146 P187 P216 P34 B15 D18 D19
I/O (TCK) P7 P9 P7 P298 E24 D31 C33
I/O - P188 P217 P33 C15 B18 E19
GND - - - - - GND* GND*
I/O - P189 P218 P32 D15 A19 B20
I/O - - - - - F28 G29
I/O (A10) P147 P190 P220 P31 A16 B19 C20
I/O - - - - - F29 E31
I/O (A11) P148 P191 P221 P30 B16 C19 D20
I/O - - - - D25 E30 D32
I/O - - - - F23 E31 G30
I/O - - - P297 F24 G28 F31
I/O - - - P29 C16 D19 A21
I/O - - - P296 E25 G29 H29
I/O - - - P28 B17 A20 E20
I/O - - - - D16 B20 B21
I/O - - - - A18 C20 C21
I/O - - - - - F30 H30
I/O - - - - - B21 D21
I/O - - - - - D20 B22
I/O - - - - - F31 G31 6
I/O P8 P10 P8 P295 D26 H28 J29
GND - - - - - GND* GND* I/O P9 P11 P9 P294 G24 H29 F33
I/O - - - P27 C17 C21 C23 I/O - P12 P10 P293 F25 G30 G32
I/O - - - P26 B18 A22 E22 I/O - P13 P11 P292 F26 H30 J30
GND - - - - - GND* GND*
I/O - - P223 P23 C18 B22 B24
I/O - - - - - - H32
I/O - - P224 P22 D17 C22 D23
I/O - - - - - - J31
I/O P149 P192 P225 P21 A20 B23 C24
I/O - - P12 P291 H23 J28 K30
I/O P150 P193 P226 P20 B19 A24 A25
I/O - - P13 P290 H24 J29 H33
I/O - - - P289 G25 H31 L29
I/O - - - P18 C19 D22 E23
I/O - - - P288 G26 J30 K31
I/O - - - P17 D18 C23 B25
I/O - P195 P228 P16 A21 B24 D24
I/O, FCLK1 P11 P15 P15 P286 J23 K28 L30
I/O - P196 P229 P15 B20 C24 C25
I/O P12 P16 P16 P285 J24 K29 K32
I/O - - - - - - B26
I/O (TMS) P13 P17 P17 P284 H25 K30 J33
I/O - - - - - - E24
I/O P14 P18 P18 P283 K23 K31 M29
GND - - - - - GND* GND*
I/O - - - - - D23 E25
I/O - - P20 P280 K24 L29 L32
I/O - - - - - B25 C27
I/O - - P21 P279 J25 L30 M31
I/O P152 P197 P230 P14 C20 A26 D26
GND - - - - - GND* GND*
I/O P153 P198 P231 P13 B21 C25 B28 I/O - - - - - M30 N29
I/O (A12) P154 P199 P232 P12 B22 D24 B29 I/O - - - - - M28 L33
I/O (A13) P155 P200 P233 P10 C21 B26 E26 I/O - - - - J26 M29 M32
GND - - - - GND* GND* GND* I/O - - - - L23 M31 P29
VCC - - - - VCC* VCC* VCC* I/O - - - P278 L24 N31 P30
I/O - - - P9 D20 A27 C28
I/O - - - P277 K25 N28 N33
I/O - - - P8 A23 D25 D27
I/O - - - - A24 C26 B30
I/O - - - - B23 B27 C29
I/O - - - - - N29 P31
I/O - - - - - A28 E27
I/O - - - - - N30 P32
I/O - - - - - D26 A31
I/O - - - P276 L25 P30 R29
GND - - - - - GND* GND*
I/O - - - P275 L26 P28 R30
I/O - - P234 P7 D21 C27 D28
I/O - P19 P23 P274 M23 P29 R31
I/O - - P235 P6 C22 B28 C30
I/O - P20 P24 P273 M24 R31 R33

R
GND - - - - - GND* GND* I/O - - - - - AF29 AG30
I/O P15 P21 P25 P272 M25 R30 T31 I/O P32 P42 P52 P237 Y23 AG31 AK32
I/O P16 P22 P26 P271 M26 R28 T29 I/O P33 P43 P53 P236 AC26 AF28 AJ31
I/O - - - - - - T30 GND - - - - - GND* GND*
I/O - - - - - - T32 I/O - - - - AD26 AG30 AG29
I/O P17 P23 P27 P270 N24 R29 U32 I/O - - - - AC25 AG29 AL33
I/O P18 P24 P28 P269 N25 T31 U31 I/O P34 P44 P54 P235 AA23 AH31 AH30
GND P19 P25 P29 P268 GND* GND* GND* I/O P35 P45 P55 P234 AB24 AG28 AK31
VCC P20 P26 P30 P267 VCC* VCC* VCC* I/O P36 P46 P56 P233 AD25 AH30 AJ30
I/O P21 P27 P31 P266 N26 T30 U29 I/O, GCK2 P37 P47 P57 P232 AC24 AJ30 AH29
I/O P22 P28 P32 P265 P25 T29 U30 O (M1) P38 P48 P58 P231 AB23 AH29 AK30
I/O - - - - - - U33 GND P39 P49 P59 P230 GND* GND* GND*
I/O - - - - - - V32 I (M0) P40 P50 P60 P229 AD24 AH28 AJ29
I/O P23 P29 P33 P264 P23 U31 V31 VCC P41 P55 P61 P228 VCC* VCC* VCC*
I/O P24 P30 P34 P263 P24 U30 V29 I (M2) P42 P56 P62 P227 AC23 AJ28 AN32
GND - - - - - GND* GND* I/O, GCK3 P43 P57 P63 P226 AE24 AK29 AJ28
I/O - P31 P35 P262 R26 U28 V30 I/O (HDC) P44 P58 P64 P225 AD23 AH27 AK29
I/O - P32 P36 P261 R25 U29 W33 I/O P45 P59 P65 P224 AC22 AK28 AL30
I/O - - - P260 R24 V30 W31 I/O P46 P60 P66 P223 AF24 AJ27 AK28
I/O - - - P259 R23 V29 W30 I/O P47 P61 P67 P222 AD22 AL28 AM31
I/O - - - - - V28 W29 I/O (/LDC) P48 P62 P68 P221 AE23 AH26 AJ27
I/O - - - - - W31 Y32 GND - - - - - GND* GND*
VCC - - - - VCC* VCC* VCC* I/O - - - - - AK27 AN31
GND - - P37 - GND* GND* GND* I/O - - - - - AJ26 AL29
I/O - - - P258 T26 W30 Y31 I/O - - - - AC21 AL27 AK27
I/O - - - P257 T25 W29 Y30 I/O - - - - AD21 AH25 AL28
I/O - - - - - W28 AA32 I/O - - - P220 AE22 AK26 AJ26
I/O - - - - - Y31 AA31 I/O - - - P219 AF23 AL26 AM30
I/O - - - - T24 Y30 AA30 VCC - - - - VCC* VCC* VCC*
I/O - - - - U25 Y29 AB32 GND - - - - GND* GND* GND*
GND - - - - - GND* GND* I/O P49 P63 P69 P218 AD20 AH24 AM29
I/O - - P38 P256 T23 Y28 AA29 I/O P50 P64 P70 P217 AE21 AJ25 AK26
I/O - - P39 P255 V26 AA30 AB31 I/O - P65 P71 P216 AF21 AK25 AL27
VCC - - P40 P253 VCC* VCC* VCC* I/O - P66 P72 P215 AC19 AJ24 AJ25
I/O P25 P33 P41 P252 U24 AA29 AC31 I/O - - - - - AH23 AN29
I/O P26 P34 P42 P251 V25 AB31 AB29 I/O - - - - - AK24 AN28
I/O P27 P35 P43 P250 V24 AB30 AD32 GND - - - - - GND* GND*
I/O, FCLK2 P28 P36 P44 P249 U23 AB29 AC30 I/O - - - - - - AM26
GND P29 P37 P45 P248 GND* GND* GND* I/O - - - - - - AK24
I/O - - - P247 Y26 AB28 AD31 I/O - - P73 P214 AD19 AL24 AL25
I/O - - - P246 W25 AC30 AE33 I/O - - P74 P213 AE20 AH22 AJ23
I/O - - P46 P245 W24 AC29 AC29 I/O - - - P212 AF20 AJ23 AN26
I/O - - P47 P244 V23 AC28 AE32 I/O - - - P211 AC18 AK23 AL24
I/O - - - - - - AD30 GND P51 P67 P75 P210 GND* GND* GND*
I/O - - - - - - AE31 I/O P52 P68 P76 P209 AD18 AJ22 AK23
GND - - - - - GND* GND* I/O P53 P69 P77 P208 AE19 AK22 AN25
I/O - - - - - AD31 AG33 I/O P54 P70 P78 P207 AC17 AL22 AJ22
I/O - - - - - AD30 AH33 I/O P55 P71 P79 P206 AD17 AJ21 AL23
I/O - P38 P48 P243 AA26 AD29 AE29 VCC - - P80 P204 VCC* VCC* VCC*
I/O - P39 P49 P242 Y25 AD28 AG31 I/O - P72 P81 P203 AE18 AH20 AM24
I/O P30 P40 P50 P241 Y24 AE30 AF30 I/O - P73 P82 P202 AF18 AK21 AK22
I/O P31 P41 P51 P240 AA25 AE29 AH32 GND - - - - - GND* GND*
GND - - - - GND* GND* GND* I/O - - - - - AJ20 AK21
VCC - - - - VCC* VCC* VCC* I/O - - - - - AH19 AM22
I/O - - - P239 AB25 AF31 AJ32 I/O - - - - AC16 AK20 AJ20
I/O - - - P238 AA24 AE28 AF29 I/O - - - - AD16 AJ19 AL21
I/O - - - - - AF30 AH31 I/O - - - P201 AE17 AL20 AN21

R
I/O - - - P200 AE16 AH18 AK20 I/O - P91 P109 P166 AE7 AK7 AL7
GND - - P83 - GND* GND* GND* I/O - P92 P110 P165 AD7 AL6 AK8
VCC - - - - VCC* VCC* VCC* I/O P71 P93 P111 P164 AE6 AJ7 AN6
I/O - - - P199 AF16 AK19 AL20 I/O P72 P94 P112 P163 AE5 AH8 AM6
I/O - - - P198 AC15 AJ18 AJ19 GND - - - - GND* GND* GND*
I/O - - P84 P197 AD15 AL19 AM20 VCC - - - - VCC* VCC* VCC*
I/O - - P85 P196 AE15 AK18 AK19 I/O - - - P162 AD6 AK6 AJ8
I/O P56 P74 P86 P195 AF15 AH17 AL19 I/O - - - P161 AC7 AL5 AL6
I/O P57 P75 P87 P194 AD14 AJ17 AN19 I/O P73 P95 P113 P160 AF4 AH7 AK7
GND - - - - - GND* GND* I/O P74 P96 P114 P159 AF3 AJ6 AM5
I/O - - - - - - AJ18 I/O - - - - AE4 AK5 AM4
I/O - - - - - - AK18 I/O - - - - AC6 AL4 AJ7
I/O - - - - - AK17 AL18 GND - - - - - GND* GND*
I/O - - - - - AL17 AM18 I/O - - - - - AH6 AL5
I/O P58 P76 P88 P193 AE14 AJ16 AK17 I/O - - - - - AJ5 AK6
I/O (/INIT) P59 P77 P89 P192 AF14 AK16 AJ17 I/O P75 P97 P115 P158 AD5 AK4 AN3
VCC P60 P78 P90 P191 VCC* VCC* VCC* I/O P76 P98 P116 P157 AE3 AH5 AK5
GND P61 P79 P91 P190 GND* GND* GND* I/O P77 P99 P117 P156 AD4 AK3 AJ6
I/O P62 P80 P92 P189 AE13 AL16 AL17 I/O, GCK4 P78 P100 P118 P155 AC5 AJ4 AL4
I/O P63 P81 P93 P188 AC13 AH15 AM17 GND P79 P101 P119 P154 GND* GND* GND*
I/O - - - - - AL15 AN17 DONE P80 P103 P120 P153 AD3 AH4 AJ5
I/O - - - - - AJ15 AK16 VCC P81 P106 P121 P152 VCC* VCC* VCC*
I/O - - - - - - AJ16 /PROGRAM P82 P108 P122 P151 AC4 AH3 AM1 6
I/O - - - - - - AL16 I/O (D7) P83 P109 P123 P150 AD2 AJ2 AH5
GND - - - - - GND* GND* I/O, GCK5 P84 P110 P124 P149 AC3 AG4 AJ4
I/O P64 P82 P94 P187 AD13 AK15 AM16 I/O P85 P111 P125 P148 AB4 AG3 AK3
I/O P65 P83 P95 P186 AF12 AJ14 AL15 I/O P86 P112 P126 P147 AD1 AH2 AH4
I/O - P84 P96 P185 AE12 AH14 AK15 I/O - - - - AB3 AH1 AL1
I/O - P85 P97 P184 AD12 AK14 AJ15 I/O - - - - AC2 AF4 AG5
I/O - - - P183 AC12 AL13 AN15 GND - - - - - GND* GND*
I/O - - - P182 AF11 AK13 AM14 I/O - - P127 P146 AA4 AF3 AJ3
VCC - - - - VCC* VCC* VCC* I/O - - P128 P145 AA3 AG2 AK2
GND - - P98 - GND* GND* GND* I/O - - - - - AG1 AG4
I/O - - - P181 AE11 AJ13 AL14 I/O - - - - - AE4 AH3
I/O - - - P180 AD11 AH13 AK14 I/O - - - P144 AB2 AE3 AF5
I/O - - - - AE10 AL12 AJ14 I/O - - - P143 AC1 AF2 AJ2
I/O - - - - AC11 AK12 AN13 VCC - - - - VCC* VCC* VCC*
I/O - - - - - AJ12 AM13 GND - - - - GND* GND* GND*
I/O - - - - - AK11 AL13 I/O (D6) P87 P113 P129 P142 Y3 AF1 AJ1
GND - - - - - GND* GND* I/O P88 P114 P130 P141 AA2 AD4 AF4
I/O - - P99 P179 AF9 AH12 AK12 I/O P89 P115 P131 P140 AA1 AD3 AG3
I/O - - P100 P178 AD10 AJ11 AN11 I/O P90 P116 P132 P139 W4 AE2 AE5
VCC - - P101 P177 VCC* VCC* VCC* I/O - - - - - AD2 AH1
I/O P66 P86 P102 P175 AE9 AL10 AJ12 I/O - - - - - AC4 AF3
I/O P67 P87 P103 P174 AD9 AK10 AL11 GND - - - - - GND* GND*
I/O P68 P88 P104 P173 AC10 AJ10 AK11 I/O - - - - - - AF1
I/O P69 P89 P105 P172 AF7 AK9 AM10 I/O - - - - - - AD4
GND P70 P90 P106 P171 GND* GND* GND* I/O - P117 P133 P138 W3 AC3 AE3
I/O - - - P170 AE8 AL8 AL10 I/O - P118 P134 P137 Y2 AD1 AC5
I/O - - - P169 AD8 AH10 AJ11 I/O - - - P136 Y1 AC2 AE1
I/O - - P107 P168 AC9 AJ9 AN9 I/O - - - P135 V4 AB4 AD3
I/O - - P108 P167 AF6 AK8 AK10 GND P91 P119 P135 P134 GND* GND* GND*
I/O - - - - - - AM9 I/O - - P136 P133 V3 AB3 AC4
I/O - - - - - - AL9 I/O - - P137 P132 W2 AB2 AD2
GND - - - - - GND* GND* I/O, FCLK3 P92 P120 P138 P131 U4 AB1 AB5
I/O - - - - - AJ8 AN7 I/O P93 P121 P139 P130 U3 AA3 AC3
I/O - - - - - AH9 AJ9 VCC - - P140 P129 VCC* VCC* VCC*

R
I/O (D5) P94 P122 P141 P127 V2 AA2 AA5 I/O - - - P93 H3 J3 J2
I/O (/CS0) P95 P123 P142 P126 V1 Y2 AB3 I/O - - P167 P92 J4 J4 K4
GND - - P143* - - GND* GND* I/O - - P168 P91 F1 H1 J3
I/O - - - - T4 Y4 AB2 I/O - - - - - - H2
I/O - - - - T3 Y3 AA4 I/O - - - - - - K5
I/O - - - - - Y1 AA3 GND - - - - - GND* GND*
I/O - - - - - W1 Y5 I/O - P143 P169 P90 G2 H2 G1
I/O - - - P125 U2 W4 Y3 I/O - P144 P170 P89 G3 H3 F1
I/O - - - P124 T2 W3 Y2 I/O P111 P145 P171 P88 F2 H4 J5
GND - - - - GND* GND* GND* I/O P112 P146 P172 P87 E2 G2 G3
VCC - - - - VCC* VCC* VCC* I/O - - - - - G3 H4
I/O - - - - - W2 W5 I/O - - - - - F1 F2
I/O - - - - - V2 W4 GND - - - - GND* GND* GND*
I/O - - - P123 T1 V4 W3 VCC - - - - VCC* VCC* VCC*
I/O - - - P122 R4 V3 W1 I/O (D1) P113 P147 P173 P86 F3 G4 F3
I/O - P124 P144 P121 R3 U1 V3 I/O (/RCK, P114 P148 P174 P85 G4 F2 G4
I/O - P125 P145 P120 R2 U2 V5 RDY_/BUSY)
GND - - - - - GND* GND* I/O - - - - D1 F3 D2
I/O P96 P126 P146 P119 R1 U4 V4 I/O - - - - C1 E1 E3
I/O P97 P127 P147 P118 P3 U3 V2 I/O - - - - - F4 G5
I/O - - - - - - U2 I/O - - - - - E2 C1
I/O - - - - - - U1 GND - - - - - GND* GND*
I/O (D4) P98 P128 P148 P117 P2 T1 U5 I/O - - - P84 D2 E3 F4
I/O P99 P129 P149 P116 P1 T2 U4 I/O - - - P83 F4 D1 D3
VCC P100 P130 P150 P115 VCC* VCC* VCC* I/O P115 P149 P175 P82 E3 E4 B3
GND P101 P131 P151 P114 GND* GND* GND* I/O P116 P150 P176 P81 C2 D2 F5
I/O (D3) P102 P132 P152 P113 N2 T3 U3 I/O (D0, DIN) P117 P151 P177 P80 D3 C2 E4
I/O (/RS) P103 P133 P153 P112 N4 R1 T2 I/O, GCK6 P118 P152 P178 P79 E4 D3 D4
(DOUT)
I/O - - - - - - T3
CCLK P119 P153 P179 P78 C3 D4 C4
I/O - - - - - - T5
I/O P104 P134 P154 P111 N3 R2 T4
O, TDO P121 P159 P181 P76 D4 C4 E6
I/O P105 P135 P155 P110 M1 R4 R1
GND - - - - - GND* GND*
I/O (A0, /WS) P123 P161 P183 P74 B3 B3 D5
I/O - P136 P156 P109 M2 R3 R3
I/O, GCK7 (A1) P124 P162 P184 P73 C4 D5 A2
I/O - P137 P157 P108 M3 P2 R4
I/O P125 P163 P185 P72 D5 B4 D6
I/O - - - P107 M4 P3 R5
I/O P126 P164 P186 P71 A3 C5 A3
I/O - - - P106 L1 P4 P2
I/O - - - - - A4 E7
I/O - - - - - N1 P3
I/O - - - - - D6 C5
I/O - - - - - N2 P4
GND - - - - - GND* GND*
I/O - - - - C5 B5 B4
I/O - - - - B4 C6 D7
I/O - - - P105 L2 N3 N1
I/O, (CS1, A2) P127 P165 P187 P70 D6 A5 C6
I/O - - - P104 L3 N4 P5
I/O (A3) P128 P166 P188 P69 C6 D7 E8
I/O - - - - K2 M1 N2
I/O - - - P68 B5 B6 B5
I/O - - - - L4 M2 N3
I/O - - - P67 A4 A6 A5
I/O - - - - - M3 N5
I/O - - - - - M4 M3
GND - - - - - GND* GND*
I/O - - P189 P66 C7 D8 D8
I/O (D2) P106 P138 P159 P103 J1 L2 M4
I/O - - P190 P65 B6 C7 C7
I/O P107 P139 P160 P102 K3 L3 L1
I/O P129 P167 P191 P64 A6 B7 E9
I/O P130 P168 P192 P63 D8 D9 A6
I/O P108 P140 P162 P99 J2 K1 K2
I/O - - - - C8 B8 B7
I/O, FCLK4 P109 P141 P163 P98 J3 K2 L4
I/O - - - - - A8 D9
I/O - - P164 P97 K4 K3 J1
GND - - - - - GND* GND*
I/O - - P165 P96 G1 K4 K3
I/O - - - - - - D10
I/O - - - - - - C9
I/O - - - P94 H2 J2 L5

R
I/O - P169 P193 P62 B7 D10 E11 VCC - - - - AE2 - AA2
I/O - P170 P194 P61 A7 C9 A9 VCC - - - - AE25 - AB33
I/O - - P195 P60 D9 B9 C10 VCC - - - - - - AD1
I/O - - - P59 C9 C10 D11 VCC - - - - - - AF33
GND P131 P171 P196 P58 GND* GND* GND* VCC - - - - - - AK1
I/O P132 P172 P197 P57 B8 B10 B10 VCC - - - - - - AK33
I/O P133 P173 P198 P56 D10 A10 E12 VCC - - - - - - AL2
I/O - - P199 P55 C10 C11 C11 VCC - - - - - - AL3
I/O - - P200 P54 B9 D12 B11 VCC - - - - - - AM2
VCC - - P201 P52 VCC* VCC* VCC* VCC - - - - - - AM15
I/O - - - P51 A9 B11 D12 VCC - - - - - - AM21
I/O - - - P50 D11 C12 A11 VCC - - - - - - AM32
GND - - - - - GND* GND* VCC - - - - - - AN4
I/O - - - - - D13 C13 VCC - - - - - - AN8
I/O - - - - - B12 E14 VCC - - - - - - AN12
I/O - - - - C11 C13 A13 VCC - - - - - - AN18
I/O - - - - B10 A12 D14 VCC - - - - - - AN24
I/O - - - P49 B11 D14 C14 VCC - - - - - - AN30
I/O - - - P48 A11 B13 B14 VCC - - - - - C3 AL31
GND - - - - GND* GND* GND* VCC - - - - - C29 E5
VCC - - - - VCC* VCC* VCC* VCC - - - - - AJ3 C31
I/O (A4) P134 P174 P202 P47 D12 C14 E15 VCC - - - - - AJ29 AK4
I/O (A5) P135 P175 P203 P46 C12 A13 D15 6
I/O - P176 P205 P45 B12 B14 C15 GND - - - - A1 A2 A7
I/O P136 P177 P206 P44 A12 D15 A15 GND - - - - A14 A3 A12
I/O (A21) P137 P178 P207 P43 C13 C15 C16 GND - - - - A19 A7 A14
I/O (A20) P138 P179 P208 P42 B13 B15 E16 GND - - - - A2 A9 A18
GND - - - - - GND* GND* GND - - - - A22 A14 A20
I/O - - - - - - D16 GND - - - - A25 A18 A24
I/O - - - - - - B16 GND - - - - A26 A23 A29
I/O - - - - - A15 B17 GND - - - - A5 A25 A32
I/O - - - - - C16 C17 GND - - - - A8 A29 B1
I/O (A6) P139 P180 P209 P41 A13 B16 E17 GND - - - - AB1 A30 B6
I/O (A7) P140 P181 P210 P40 B14 A16 D17 GND - - - - AB26 B1 B9
GND P141 P182 P211 P39 GND* GND* GND* GND - - - - AE1 B2 B15
VCC - - - - A10 A1 A4 GND - - - - AE26 B30 B23
VCC - - - - A17 A11 A10 GND - - - - AF1 B31 B27
VCC - - - - AC14 A21 A16 GND - - - - AF13 C1 B31
VCC - - - - AC20 A31 A22 GND - - - - AF19 C31 C2
VCC - - - - AC8 D11 A26 GND - - - - AF2 D16 E1
VCC - - - - AF10 D21 A30 GND - - - - AF22 G1 F32
VCC - - - - AF17 L1 B2 GND - - - - AF25 G31 G2
VCC - - - - D7 L4 B13 GND - - - - AF26 J1 G33
VCC - - - - D13 L28 B19 GND - - - - AF5 J31 J32
VCC - - - - D19 L31 B32 GND - - - - AF8 P1 K1
VCC - - - - G23 AA1 C3 GND - - - - B1 P31 L2
VCC - - - - H4 AA4 C32 GND - - - - B26 T4 M33
VCC - - - - K1 AA28 D1 GND - - - - E1 T28 P1
VCC - - - - K26 AA31 D33 GND - - - - E26 V1 P33
VCC - - - - N23 AH11 H1 GND - - - - H1 V31 R32
VCC - - - - P4 AH21 K33 GND - - - - H26 AC1 T1
VCC - - - - U1 AL1 M1 GND - - - - N1 AC31 V33
VCC - - - - U26 AL11 N32 GND - - - - P26 AE1 W2
VCC - - - - W23 AL21 R2 GND - - - - W1 AE31 Y1
VCC - - - - Y4 AL31 T33 GND - - - - W26 AH16 Y33
VCC - - - - B2 - V1 GND - - - - - AJ1 AB1
VCC - - - - B25 - W32 GND - - - - - AJ31 AC32

R
GND - - - - - AK1 AD33 NC - - - - - - AL8
GND - - - - - AK2 AE2 NC - - - - - - AK9
GND - - - - - AK30 AG1 NC - - - - - - AM8
GND - - - - - AK31 AG32 NC - - - - - - AJ10
GND - - - - - AL2 AH2 NC - - - - - - AL12
GND - - - - - AL3 AJ33 NC - - - - - - AM12
GND - - - - - AL7 AL32 NC - - - - - - AJ13
GND - - - - - AL9 AM3 NC - - - - - - AK13
GND - - - - - AL14 AM11 NC - - - - - - AN23
GND - - - - - AL18 AM19 NC - - - - - - AL22
GND - - - - - AL23 AM25 NC - - - - - - AJ21
GND - - - - - AL25 AM28 NC - - - - - - AM23
GND - - - - - AL29 AM33 NC - - - - - - AM27
GND - - - - - AL30 AM7 NC - - - - - - AJ24
GND - - - - - - AN2 NC - - - - - - AL26
GND - - - - - - AN5 NC - - - - - - AK25
GND - - - - - - AN10 NC - - - - - - AE30
GND - - - - - - AN14 NC - - - - - - AF31
GND - - - - - - AN16 NC - - - - - - AD29
GND - - - - - - AN20 NC - - - - - - AF32
GND - - - - - - AN22 NC - - - - - - AC33
GND - - - - - - AN27 NC - - - - - - AB30
GND - - P204 - - - - NC - - - - - - Y29
GND - - P219 - - - - NC - - - - - - AA33
NC - - - - - - N31
NC - P1 - P11 - C8 A28 NC - - - - - - N30
NC - P3 - P24 - - A27 NC - - - - - - M30
NC - P51 P53 - - D25 NC - - - - - - L31
NC - P52 - P100 - - C26 NC - - - - - - K29
NC - P53 - P128 - - A23 NC - - - - - - H31
NC - P54 - P176 - - D22 NC - - - - - - E33
NC - P102 - P205 - - C22 NC - - - - - - E32
NC - P104 - P254 - - E21 NC - - - - - - A1
NC - P105 - P281 - - D13 NC - - - - - - A33
NC - P107 - - - - B12 NC - - - - - - AN1
NC - P155 - - - - C12 NC - - - - - - AN33
NC - P156 - - - - E13 NC - - - - - - AC2
NC - P157 - - - - A8 12/18/98
NC - P158 - - - - B8
NC - P206 - - - - E10
NC - P207 - - - - C8
NC - P208 - - - - H5
NC - - - - - - E2
NC - - - - - - J4
NC - - - - - - H3
NC - - - - - - M5
NC - - - - - - L3
NC - - - - - - M2
NC - - - - - - N4
NC - - - - - - Y4
NC - - - - - - AA1
NC - - - - - - AC1
NC - - - - - - AB4
NC - - - - - - AF2
NC - - - - - - AD5
NC - - - - - - AG2
NC - - - - - - AE4

I/O - - - P9 D20 A27 C28
I/O - - - P8 A23 D25 D27
I/O - - - - A24 C26 B30
I/O (A8) P143 P184 P213 P37 D14 D17 A17
I/O - - - - B23 B27 C29
I/O (A9) P144 P185 P214 P36 C14 A17 B18
I/O - - - - - A28 E27
I/O - - - - - C17 C18
I/O - - - - - D26 A31
I/O - - - - - B17 E18
I/O - - - - - - D18
I/O - - P234 P7 D21 C27 D28
I/O - - - - - - A19
I/O - - P235 P6 C22 B28 C30
I/O P156 P201 P236 P5 B24 D27 D29
I/O (A19) P145 P186 P215 P35 A15 C18 C19 I/O P157 P202 P237 P4 C23 B29 E28
I/O (A18) P146 P187 P216 P34 B15 D18 D19 I/O (A14) P158 P203 P238 P3 D22 C28 D30
I/O - P188 P217 P33 C15 B18 E19 I/O, GCK8 (A15) P159 P204 P239 P2 C24 D28 E29
I/O - P189 P218 P32 D15 A19 B20 VCC P160 P205 P240 P1 VCC* VCC* VCC*
I/O (A10) P147 P190 P220 P31 A16 B19 C20 GND P1 P2 P1 P304 GND* GND* GND*
I/O (A11) P148 P191 P221 P30 B16 C19 D20
I/O, GCK1 (A16) P2 P4 P2 P303 D23 D29 B33
I/O (A17) P3 P5 P3 P302 C25 C30 F29
I/O P4 P6 P4 P301 D24 E28 E30
I/O - - - P29 C16 D19 A21
I/O P5 P7 P5 P300 E23 E29 D31
I/O - - - P28 B17 A20 E20
I/O (TDI) P6 P8 P6 P299 C26 D30 F30
I/O - - - - D16 B20 B21
I/O (TCK) P7 P9 P7 P298 E24 D31 C33
I/O - - - - A18 C20 C21
I/O - - - - - B21 D21
I/O - - - - - F28 G29 6
I/O - - - - - D20 B22
I/O - - - - - F29 E31
I/O - - - - - - E21
I/O - - - - D25 E30 D32
I/O - - - - - - C22
I/O - - - - F23 E31 G30
I/O - - - P297 F24 G28 F31
I/O - - - - - - D22
I/O - - - P296 E25 G29 H29
I/O - - - - - - A23
I/O - - - P27 C17 C21 C23
I/O - - - P26 B18 A22 E22
I/O - - - - - - E32
VCC - - P222 P25 VCC* VCC* VCC* I/O - - - - - - E33
I/O - - P223 P23 C18 B22 B24 I/O - - - - - F30 H30
I/O - - P224 P22 D17 C22 D23 I/O - - - - - F31 G31
I/O P149 P192 P225 P21 A20 B23 C24 I/O P8 P10 P8 P295 D26 H28 J29
I/O P150 P193 P226 P20 B19 A24 A25 I/O P9 P11 P9 P294 G24 H29 F33
I/O - P12 P10 P293 F25 G30 G32
I/O - - - P18 C19 D22 E23
I/O - P13 P11 P292 F26 H30 J30
I/O - - - P17 D18 C23 B25
I/O - P195 P228 P16 A21 B24 D24
I/O - P196 P229 P15 B20 C24 C25
I/O - - - - - - H31
I/O - - - - - - B26
I/O - - - - - - K29
I/O - - - - - - E24
I/O - - - - - - H32
I/O - - - - - - C26
I/O - - - - - - J31
I/O - - - - - - D25
I/O - - P12 P291 H23 J28 K30
I/O - - P13 P290 H24 J29 H33
I/O - - - P289 G25 H31 L29
I/O - - - - - - A27
I/O - - - P288 G26 J30 K31
I/O - - - - - - A28
I/O - - - - - D23 E25
I/O, FCLK1 P11 P15 P15 P286 J23 K28 L30
I/O - - - - - B25 C27
I/O P12 P16 P16 P285 J24 K29 K32
I/O P152 P197 P230 P14 C20 A26 D26 I/O (TMS) P13 P17 P17 P284 H25 K30 J33
I/O P153 P198 P231 P13 B21 C25 B28 I/O P14 P18 P18 P283 K23 K31 M29
I/O (A12) P154 P199 P232 P12 B22 D24 B29 VCC - - P19 P282 VCC* VCC* VCC*
I/O (A13) P155 P200 P233 P10 C21 B26 E26 I/O - - - - - - L31
GND - - - - GND* GND* GND* I/O - - - - - - M30
I/O - - P20 P280 K24 L29 L32

R
I/O - - P21 P279 J25 L30 M31 I/O P26 P34 P42 P251 V25 AB31 AB29
GND - - - - GND* GND* GND* I/O P27 P35 P43 P250 V24 AB30 AD32
I/O - - - - - M30 N29 I/O, FCLK2 P28 P36 P44 P249 U23 AB29 AC30
I/O - - - - - M28 L33 GND P29 P37 P45 P248 GND* GND* GND*
I/O - - - - - - N30 I/O - - - P247 Y26 AB28 AD31
I/O - - - - - - N31 I/O - - - P246 W25 AC30 AE33
I/O - - - - J26 M29 M32 I/O - - P46 P245 W24 AC29 AC29
I/O - - - - L23 M31 P29 I/O - - P47 P244 V23 AC28 AE32
I/O - - - P278 L24 N31 P30 I/O - - - - - - AD30
I/O - - - P277 K25 N28 N33 I/O - - - - - - AE31
GND - - P22 - GND* GND* GND* I/O - - - - - - AF32
VCC - - - - VCC* VCC* VCC* I/O - - - - - - AD29
I/O - - - - - N29 P31 GND - - - - GND* GND* GND*
I/O - - - - - N30 P32 VCC - - - - VCC* VCC* VCC*
I/O - - - P276 L25 P30 R29 I/O - - - - - - AF31
I/O - - - P275 L26 P28 R30 I/O - - - - - - AE30
I/O - P19 P23 P274 M23 P29 R31 I/O - - - - - AD31 AG33
I/O - P20 P24 P273 M24 R31 R33 I/O - - - - - AD30 AH33
GND - - - - GND* GND* GND* I/O - P38 P48 P243 AA26 AD29 AE29
I/O P15 P21 P25 P272 M25 R30 T31 I/O - P39 P49 P242 Y25 AD28 AG31
I/O P16 P22 P26 P271 M26 R28 T29 I/O P30 P40 P50 P241 Y24 AE30 AF30
I/O - - - - - - T30 I/O P31 P41 P51 P240 AA25 AE29 AH32
I/O - - - - - - T32 GND - - - - GND* GND* GND*
I/O P17 P23 P27 P270 N24 R29 U32 VCC - - - - VCC* VCC* VCC*
I/O P18 P24 P28 P269 N25 T31 U31 I/O - - - P239 AB25 AF31 AJ32
GND P19 P25 P29 P268 GND* GND* GND* I/O - - - P238 AA24 AE28 AF29
VCC P20 P26 P30 P267 VCC* VCC* VCC* I/O - - - - - AF30 AH31
I/O P21 P27 P31 P266 N26 T30 U29 I/O - - - - - AF29 AG30
I/O P22 P28 P32 P265 P25 T29 U30 I/O P32 P42 P52 P237 Y23 AG31 AK32
I/O - - - - - - U33 I/O P33 P43 P53 P236 AC26 AF28 AJ31
I/O - - - - - - V32 GND - - - - GND* GND* GND*
I/O P23 P29 P33 P264 P23 U31 V31 I/O - - - - AD26 AG30 AG29
I/O P24 P30 P34 P263 P24 U30 V29 I/O - - - - AC25 AG29 AL33
GND - - - - GND* GND* GND* I/O P34 P44 P54 P235 AA23 AH31 AH30
I/O - P31 P35 P262 R26 U28 V30 I/O P35 P45 P55 P234 AB24 AG28 AK31
I/O - P32 P36 P261 R25 U29 W33 I/O P36 P46 P56 P233 AD25 AH30 AJ30
I/O - - - P260 R24 V30 W31 I/O, GCK2 P37 P47 P57 P232 AC24 AJ30 AH29
I/O - - - P259 R23 V29 W30 O (M1) P38 P48 P58 P231 AB23 AH29 AK30
I/O - - - - - V28 W29 GND P39 P49 P59 P230 GND* GND* GND*
I/O - - - - - W31 Y32 I (M0) P40 P50 P60 P229 AD24 AH28 AJ29
VCC - - - - VCC* VCC* VCC* VCC P41 P55 P61 P228 VCC* VCC* VCC*
GND - - P37 - GND* GND* GND* I (M2) P42 P56 P62 P227 AC23 AJ28 AN32
I/O - - - P258 T26 W30 Y31 I/O, GCK3 P43 P57 P63 P226 AE24 AK29 AJ28
I/O - - - P257 T25 W29 Y30 I/O (HDC) P44 P58 P64 P225 AD23 AH27 AK29
I/O - - - - - - AA33 I/O P45 P59 P65 P224 AC22 AK28 AL30
I/O - - - - - - Y29 I/O P46 P60 P66 P223 AF24 AJ27 AK28
I/O - - - - - W28 AA32 I/O P47 P61 P67 P222 AD22 AL28 AM31
I/O - - - - - Y31 AA31 I/O (/LDC) P48 P62 P68 P221 AE23 AH26 AJ27
I/O - - - - T24 Y30 AA30 GND - - - - GND* GND* GND*
I/O - - - - U25 Y29 AB32 I/O - - - - - AK27 AN31
GND - - - - GND* GND* GND* I/O - - - - - AJ26 AL29
I/O - - P38 P256 T23 Y28 AA29 I/O - - - - AC21 AL27 AK27
I/O - - P39 P255 V26 AA30 AB31 I/O - - - - AD21 AH25 AL28
I/O - - - - - - AB30 I/O - - - P220 AE22 AK26 AJ26
I/O - - - - - - AC33 I/O - - - P219 AF23 AL26 AM30
VCC - - P40 P253 VCC* VCC* VCC* VCC - - - - VCC* VCC* VCC*
I/O P25 P33 P41 P252 U24 AA29 AC31 GND - - - - GND* GND* GND*

R
I/O P49 P63 P69 P218 AD20 AH24 AM29 I/O - - - - - AJ15 AK16
I/O P50 P64 P70 P217 AE21 AJ25 AK26 I/O - - - - - - AJ16
I/O - P65 P71 P216 AF21 AK25 AL27 I/O - - - - - - AL16
I/O - P66 P72 P215 AC19 AJ24 AJ25 GND - - - - GND* GND* GND*
I/O - - - - - AH23 AN29 I/O P64 P82 P94 P187 AD13 AK15 AM16
I/O - - - - - AK24 AN28 I/O P65 P83 P95 P186 AF12 AJ14 AL15
I/O - - - - - - AK25 I/O - P84 P96 P185 AE12 AH14 AK15
I/O - - - - - - AL26 I/O - P85 P97 P184 AD12 AK14 AJ15
VCC - - - VCC* VCC* VCC* I/O - - - P183 AC12 AL13 AN15
GND - - - - GND* GND* GND* I/O - - - P182 AF11 AK13 AM14
I/O - - - - - - AJ24 VCC - - - - VCC* VCC* VCC*
I/O - - - - - - AM27 GND - - P98 - GND* GND* GND*
I/O - - - - - - AM26 I/O - - - P181 AE11 AJ13 AL14
I/O - - - - - - AK24 I/O - - - P180 AD11 AH13 AK14
I/O - - P73 P214 AD19 AL24 AL25 I/O - - - - AE10 AL12 AJ14
I/O - - P74 P213 AE20 AH22 AJ23 I/O - - - - AC11 AK12 AN13
I/O - - - P212 AF20 AJ23 AN26 I/O - - - - - AJ12 AM13
I/O - - - P211 AC18 AK23 AL24 I/O - - - - - AK11 AL13
GND P51 P67 P75 P210 GND* GND* GND* I/O - - - - - - AK13
I/O P52 P68 P76 P209 AD18 AJ22 AK23 I/O - - - - - - AJ13
I/O P53 P69 P77 P208 AE19 AK22 AN25 GND - - - - GND* GND* GND*
I/O P54 P70 P78 P207 AC17 AL22 AJ22 I/O - - - - - - AM12
I/O P55 P71 P79 P206 AD17 AJ21 AL23 I/O - - - - - - AL12 6
VCC - - P80 P204 VCC* VCC* VCC* I/O - - P99 P179 AF9 AH12 AK12
I/O - P72 P81 P203 AE18 AH20 AM24 I/O - - P100 P178 AD10 AJ11 AN11
I/O - P73 P82 P202 AF18 AK21 AK22 VCC - - P101 P177 VCC* VCC* VCC*
I/O - - - - - - AM23 I/O P66 P86 P102 P175 AE9 AL10 AJ12
I/O - - - - - - AJ21 I/O P67 P87 P103 P174 AD9 AK10 AL11
GND - - - - GND* GND* GND* I/O P68 P88 P104 P173 AC10 AJ10 AK11
I/O - - - - - - AL22 I/O P69 P89 P105 P172 AF7 AK9 AM10
I/O - - - - - - AN23 GND P70 P90 P106 P171 GND* GND* GND*
I/O - - - - - AJ20 AK21 I/O - - - P170 AE8 AL8 AL10
I/O - - - - - AH19 AM22 I/O - - - P169 AD8 AH10 AJ11
I/O - - - - AC16 AK20 AJ20 I/O - - P107 P168 AC9 AJ9 AN9
I/O - - - - AD16 AJ19 AL21 I/O - - P108 P167 AF6 AK8 AK10
I/O - - - P201 AE17 AL20 AN21 I/O - - - - - - AM9
I/O - - - P200 AE16 AH18 AK20 I/O - - - - - - AL9
GND - - P83 - GND* GND* GND* I/O - - - - - - AJ10
VCC - - - - VCC* VCC* VCC* I/O - - - - - - AM8
I/O - - - P199 AF16 AK19 AL20 GND - - - - GND* GND* GND*
I/O - - - P198 AC15 AJ18 AJ19 VCC - - - - VCC* VCC* VCC*
I/O - - P84 P197 AD15 AL19 AM20 I/O - - - - - - AK9
I/O - - P85 P196 AE15 AK18 AK19 I/O - - - - - - AL8
I/O P56 P74 P86 P195 AF15 AH17 AL19 I/O - - - - - AJ8 AN7
I/O P57 P75 P87 P194 AD14 AJ17 AN19 I/O - - - - - AH9 AJ9
GND - - - - GND* GND* GND* I/O - P91 P109 P166 AE7 AK7 AL7
I/O - - - - - - AJ18 I/O - P92 P110 P165 AD7 AL6 AK8
I/O - - - - - - AK18 I/O P71 P93 P111 P164 AE6 AJ7 AN6
I/O - - - - - AK17 AL18 I/O P72 P94 P112 P163 AE5 AH8 AM6
I/O - - - - - AL17 AM18 GND - - - - GND* GND* GND*
I/O P58 P76 P88 P193 AE14 AJ16 AK17 VCC - - - - VCC* VCC* VCC*
I/O (/INIT) P59 P77 P89 P192 AF14 AK16 AJ17 I/O - - - P162 AD6 AK6 AJ8
VCC P60 P78 P90 P191 VCC* VCC* VCC* I/O - - - P161 AC7 AL5 AL6
GND P61 P79 P91 P190 GND* GND* GND* I/O P73 P95 P113 P160 AF4 AH7 AK7
I/O P62 P80 P92 P189 AE13 AL16 AL17 I/O P74 P96 P114 P159 AF3 AJ6 AM5
I/O P63 P81 P93 P188 AC13 AH15 AM17 I/O - - - - AE4 AK5 AM4
I/O - - - - - AL15 AN17 I/O - - - - AC6 AL4 AJ7

R
GND - - - - GND* GND* GND* I/O - - - - - Y1 AA3
I/O - - - - - AH6 AL5 I/O - - - - - W1 Y5
I/O - - - - - AJ5 AK6 I/O - - - - - - AA1
I/O P75 P97 P115 P158 AD5 AK4 AN3 I/O - - - - - - Y4
I/O P76 P98 P116 P157 AE3 AH5 AK5 I/O - - - P125 U2 W4 Y3
I/O P77 P99 P117 P156 AD4 AK3 AJ6 I/O - - - P124 T2 W3 Y2
I/O, GCK4 P78 P100 P118 P155 AC5 AJ4 AL4 GND - - - - GND* GND* GND*
GND P79 P101 P119 P154 GND* GND* GND* VCC - - - - VCC* VCC* VCC*
DONE P80 P103 P120 P153 AD3 AH4 AJ5 I/O - - - - - W2 W5
VCC P81 P106 P121 P152 VCC* VCC* VCC* I/O - - - - - V2 W4
/PROGRAM P82 P108 P122 P151 AC4 AH3 AM1 I/O - - - P123 T1 V4 W3
I/O (D7) P83 P109 P123 P150 AD2 AJ2 AH5 I/O - - - P122 R4 V3 W1
I/O, GCK5 P84 P110 P124 P149 AC3 AG4 AJ4 I/O - P124 P144 P121 R3 U1 V3
I/O P85 P111 P125 P148 AB4 AG3 AK3 I/O - P125 P145 P120 R2 U2 V5
I/O P86 P112 P126 P147 AD1 AH2 AH4 GND - - - - GND* GND* GND*
I/O - - - - AB3 AH1 AL1 I/O P96 P126 P146 P119 R1 U4 V4
I/O - - - - AC2 AF4 AG5 I/O P97 P127 P147 P118 P3 U3 V2
GND - - - - GND* GND* GND* I/O - - - - - - U2
I/O - - P127 P146 AA4 AF3 AJ3 I/O - - - - - - U1
I/O - - P128 P145 AA3 AG2 AK2 I/O (D4) P98 P128 P148 P117 P2 T1 U5
I/O - - - - - AG1 AG4 I/O P99 P129 P149 P116 P1 T2 U4
I/O - - - - - AE4 AH3 VCC P100 P130 P150 P115 VCC* VCC* VCC*
I/O - - - P144 AB2 AE3 AF5 GND P101 P131 P151 P114 GND* GND* GND*
I/O - - - P143 AC1 AF2 AJ2 I/O (D3) P102 P132 P152 P113 N2 T3 U3
VCC - - - - VCC* VCC* VCC* I/O (/RS) P103 P133 P153 P112 N4 R1 T2
GND - - - - GND* GND* GND* I/O - - - - - - T3
I/O (D6) P87 P113 P129 P142 Y3 AF1 AJ1 I/O - - - - - - T5
I/O P88 P114 P130 P141 AA2 AD4 AF4 I/O P104 P134 P154 P111 N3 R2 T4
I/O P89 P115 P131 P140 AA1 AD3 AG3 I/O P105 P135 P155 P110 M1 R4 R1
I/O P90 P116 P132 P139 W4 AE2 AE5 GND - - - - GND* GND* GND*
I/O - - - - - AD2 AH1 I/O - P136 P156 P109 M2 R3 R3
I/O - - - - - AC4 AF3 I/O - P137 P157 P108 M3 P2 R4
I/O - - - - - - AE4 I/O - - - P107 M4 P3 R5
I/O - - - - - - AG2 I/O - - - P106 L1 P4 P2
VCC - - - - VCC* VCC* VCC* I/O - - - - - N1 P3
GND - - - - GND* GND* GND* I/O - - - - - N2 P4
I/O - - - - - - AD5 VCC - - - - VCC* VCC* VCC*
I/O - - - - - - AF2 GND - - P158 - GND* GND* GND*
I/O - - - - - - AF1 I/O - - - P105 L2 N3 N1
I/O - - - - - - AD4 I/O - - - P104 L3 N4 P5
I/O - P117 P133 P138 W3 AC3 AE3 I/O - - - - K2 M1 N2
I/O - P118 P134 P137 Y2 AD1 AC5 I/O - - - - L4 M2 N3
I/O - - - P136 Y1 AC2 AE1 I/O - - - - - - N4
I/O - - - P135 V4 AB4 AD3 I/O - - - - - - M2
GND P91 P119 P135 P134 GND* GND* GND* I/O - - - - - M3 N5
I/O - - P136 P133 V3 AB3 AC4 I/O - - - - - M4 M3
I/O - - P137 P132 W2 AB2 AD2 GND - - - - GND* GND* GND*
I/O, FCLK3 P92 P120 P138 P131 U4 AB1 AB5 I/O (D2) P106 P138 P159 P103 J1 L2 M4
I/O P93 P121 P139 P130 U3 AA3 AC3 I/O P107 P139 P160 P102 K3 L3 L1
VCC - - P140 P129 VCC* VCC* VCC* I/O - - - - - - L3
I/O - - - - - - AB4 I/O - - - - - - M5
I/O - - - - - - AC1 VCC - - P161 P101 VCC* VCC* VCC*
I/O (D5) P94 P122 P141 P127 V2 AA2 AA5 I/O P108 P140 P162 P99 J2 K1 K2
I/O (/CS0) P95 P123 P142 P126 V1 Y2 AB3 I/O, FCLK4 P109 P141 P163 P98 J3 K2 L4
GND - - P143 - GND* GND* GND* I/O - - P164 P97 K4 K3 J1
I/O - - - - T4 Y4 AB2 I/O - - P165 P96 G1 K4 K3
I/O - - - - T3 Y3 AA4 GND P110 P142 P166 P95 GND* GND* GND*

R
I/O - - - P94 H2 J2 L5 I/O P130 P168 P192 P63 D8 D9 A6
I/O - - - P93 H3 J3 J2 I/O - - - - C8 B8 B7
I/O - - P167 P92 J4 J4 K4 I/O - - - - - A8 D9
I/O - - P168 P91 F1 H1 J3 I/O - - - - - - C8
I/O - - - - - - H2 I/O - - - - - - E10
I/O - - - - - - K5 VCC - - - - - VCC* VCC*
I/O - - - - - - H3 GND - - - - - GND* GND*
I/O - - - - - - J4 I/O - - - - - - B8
GND - - - - GND* GND* GND* I/O - - - - - - A8
VCC - - - - VCC* VCC* VCC* I/O - - - - - - D10
I/O - P143 P169 P90 G2 H2 G1 I/O - - - - - - C9
I/O - P144 P170 P89 G3 H3 F1 I/O - P169 P193 P62 B7 D10 E11
I/O P111 P145 P171 P88 F2 H4 J5 I/O - P170 P194 P61 A7 C9 A9
I/O P112 P146 P172 P87 E2 G2 G3 I/O - - P195 P60 D9 B9 C10
I/O - - - - - G3 H4 I/O - - - P59 C9 C10 D11
I/O - - - - - F1 F2 GND P131 P171 P196 P58 GND* GND* GND*
I/O - - - - - - E2 I/O P132 P172 P197 P57 B8 B10 B10
I/O - - - - - - H5 I/O P133 P173 P198 P56 D10 A10 E12
GND - - - - GND* GND* GND* I/O - - P199 P55 C10 C11 C11
VCC - - - - VCC* VCC* VCC* I/O - - P200 P54 B9 D12 B11
I/O (D1) P113 P147 P173 P86 F3 G4 F3 VCC - - P201 P52 VCC* VCC* VCC*
I/O (/RCK, P114 P148 P174 P85 G4 F2 G4 I/O - - - P51 A9 B11 D12
RDY_/BUSY) I/O - - - P50 D11 C12 A11 6
I/O - - - - D1 F3 D2 I/O - - - - - - E13
I/O - - - - C1 E1 E3 I/O - - - - - - C12
I/O - - - - - F4 G5 GND - - - - GND* GND* GND*
I/O - - - - - E2 C1 I/O - - - - - - B12
GND - - - - GND* GND* GND* I/O - - - - - - D13
I/O - - - P84 D2 E3 F4 I/O - - - - - D13 C13
I/O - - - P83 F4 D1 D3 I/O - - - - - B12 E14
I/O P115 P149 P175 P82 E3 E4 B3 I/O - - - - C11 C13 A13
I/O P116 P150 P176 P81 C2 D2 F5 I/O - - - - B10 A12 D14
I/O (D0, DIN) P117 P151 P177 P80 D3 C2 E4 I/O - - - P49 B11 D14 C14
“I/O, GCK6 P118 P152 P178 P79 E4 D3 D4 I/O - - - P48 A11 B13 B14
(DOUT)”
CCLK P119 P153 P179 P78 C3 D4 C4
I/O (A4) P134 P174 P202 P47 D12 C14 E15
O, TDO P121 P159 P181 P76 D4 C4 E6
I/O (A5) P135 P175 P203 P46 C12 A13 D15
I/O - P176 P205 P45 B12 B14 C15
I/O (A0, /WS) P123 P161 P183 P74 B3 B3 D5
I/O P136 P177 P206 P44 A12 D15 A15
“I/O, GCK7 P124 P162 P184 P73 C4 D5 A2
(A1)” I/O (A21) P137 P178 P207 P43 C13 C15 C16
I/O P125 P163 P185 P72 D5 B4 D6 I/O (A20) P138 P179 P208 P42 B13 B15 E16
I/O P126 P164 P186 P71 A3 C5 A3 GND - - - - GND* GND* GND*
I/O - - - - - A4 E7 I/O - - - - - - D16
I/O - - - - - D6 C5 I/O - - - - - - B16
GND - - - - GND* GND* GND* I/O - - - - - A15 B17
I/O - - - - C5 B5 B4 I/O - - - - - C16 C17
I/O - - - - B4 C6 D7 I/O (A6) P139 P180 P209 P41 A13 B16 E17
I/O, (CS1, A2) P127 P165 P187 P70 D6 A5 C6 I/O (A7) P140 P181 P210 P40 B14 A16 D17
I/O (A3) P128 P166 P188 P69 C6 D7 E8 GND P141 P182 P211 P39 GND* GND* GND*
I/O - - - P68 B5 B6 B5 VCC - - - - A10 A1 A4
I/O - - - P67 A4 A6 A5 VCC - - - - A17 A11 A10
VCC - - - - VCC* VCC* VCC* VCC - - - - AC14 A21 A16
GND - - - - GND* GND* GND* VCC - - - - AC20 A31 A22
I/O - - P189 P66 C7 D8 D8 VCC - - - - AC8 D11 A26
I/O - - P190 P65 B6 C7 C7 VCC - - - - AF10 D21 A30
I/O P129 P167 P191 P64 A6 B7 E9 VCC - - - - AF17 L1 B2

R
VCC - - - - D7 L4 B13 GND - - - - AF26 J1 G33
VCC - - - - D13 L28 B19 GND - - - - AF5 J31 J32
VCC - - - - D19 L31 B32 GND - - - - AF8 P1 K1
VCC - - - - G23 AA1 C3 GND - - - - B1 P31 L2
VCC - - - - H4 AA4 C32 GND - - - - B26 T4 M33
VCC - - - - K1 AA28 D1 GND - - - - E1 T28 P1
VCC - - - - K26 AA31 D33 GND - - - - E26 V1 P33
VCC - - - - N23 AH11 H1 GND - - - - H1 V31 R32
VCC - - - - P4 AH21 K33 GND - - - - H26 AC1 T1
VCC - - - - U1 AL1 M1 GND - - - - N1 AC31 V33
VCC - - - - U26 AL11 N32 GND - - - - P26 AE1 W2
VCC - - - - W23 AL21 R2 GND - - - - W1 AE31 Y1
VCC - - - - Y4 AL31 T33 GND - - - - W26 AH16 Y33
VCC - - - - B2 - V1 GND - - - - - AJ1 AB1
VCC - - - - B25 - W32 GND - - - - - AJ31 AC32
VCC - - - - AE2 - AA2 GND - - - - - AK1 AD33
VCC - - - - AE25 - AB33 GND - - - - - AK2 AE2
VCC - - - - - - AD1 GND - - - - - AK30 AG1
VCC - - - - - - AF33 GND - - - - - AK31 AG32
VCC - - - - - - AK1 GND - - - - - AL2 AH2
VCC - - - - - - AK33 GND - - - - - AL3 AJ33
VCC - - - - - - AL2 GND - - - - - AL7 AL32
VCC - - - - - - AL3 GND - - - - - AL9 AM3
VCC - - - - - - AM2 GND - - - - - AL14 AM11
VCC - - - - - - AM15 GND - - - - - AL18 AM19
VCC - - - - - - AM21 GND - - - - - AL23 AM25
VCC - - - - - - AM32 GND - - - - - AL25 AM28
VCC - - - - - - AN4 GND - - - - - AL29 AM33
VCC - - - - - - AN8 GND - - - - - AL30 AM7
VCC - - - - - - AN12 GND - - - - - AN2
VCC - - - - - - AN18 GND - - - - - - AN5
VCC - - - - - - AN24 GND - - - - - - AN10
VCC - - - - - - AN30 GND - - - - - - AN14
VCC - - - - - C3 AL31 GND - - - - - - AN16
VCC - - - - - C29 E5 GND - - - - - - AN20
VCC - - - - - AJ3 C31 GND - - - - - - AN22
VCC - - - - - AJ29 AK4 GND - - - - - - AN27
GND - - P204 - - - -
GND - - - - A1 A2 A7 GND - - P219 - - - -
GND - - - - A14 A3 A12
GND - - - - A19 A7 A14 NC - P1 - P11 - C8 A1
GND - - - - A2 A9 A18 NC - P3 - P24 - - A33
GND - - - - A22 A14 A20 NC - P51 - P53 - - AN1
GND - - - - A25 A18 A24 NC - P52 - P100 - - AN33
GND - - - - A26 A23 A29 NC - P53 - P128 - - AC2
GND - - - - A5 A25 A32 NC - P54 - P176 - - -
GND - - - - A8 A29 B1 NC - P102 - P205 - - -
GND - - - - AB1 A30 B6 NC - P104 - P254 - - -
GND - - - - AB26 B1 B9 NC - P105 - P281 - - -
GND - - - - AE1 B2 B15 NC - P107 - - - - -
GND - - - - AE26 B30 B23 NC - P155 - - - - -
GND - - - - AF1 B31 B27 NC - P156 - - - - -
GND - - - - AF13 C1 B31 NC - P157 - - - - -
GND - - - - AF19 C31 C2 NC - P158 - - - - -
GND - - - - AF2 D16 E1 NC - P206 - - - - -
GND - - - - AF22 G1 F32 NC - P207 - - - - -
GND - - - - AF25 G31 G2 NC - P208 - - - - -

R

12/18/98
* Note: Pads labelled GND* or VCC* are internally bonded

to Ground or VCC planes within the associated package.
They have no direct connection to any specific package pin.

XC40110XV Pinout Table XC40110XV Pinout Table (Continued)
XC40110XV Pinout Table PAD EPIC

HQ240 BG352 BG432 BG560 PG559
NAME PAD#
PAD EPIC I/O 38 - C17 C21 C23 R3
HQ240 BG352 BG432 BG560 PG559
NAME PAD#
I/O 39 - - - A23 U1
VCCIO - P240 VCCIO* VCCIO* VCCIO* VCCIO*
I/O 40 - - - D22 T4
I/O GCK8 1 P239 C24 D28 E29 E3
(A15) GND - - - GND* GND* GND*
I/O (A14) 2 P238 D22 C28 D30 F4 VCCIO - - - VCCIO* VCCIO* -
I/O 3 P237 C23 B29 E28 J7 I/O 41 - - - C22 U5
I/O 4 P236 B24 D27 D29 D2 I/O 42 - - - E21 V2
I/O 5 P235 C22 B28 C30 K8 I/O 43 - - D20 B22 U7
I/O 6 P234 D21 C27 D28 H6 I/O 44 - - B21 D21 U3
I/O 7 - - - - - I/O 45 - A18 C20 C21 Y2
I/O 8 - - - - - I/O 46 - D16 B20 B21 V4
VCCIO - - - VCCIO* VCCIO* - I/O 47 - B17 A20 E20 V6
GND - - - GND* GND* GND* I/O 48 - C16 D19 A21 W5
I/O 9 - - - - - GND - - GND* GND* GND* GND*
I/O 10 - - - - - VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 11 - - D26 A31 G5 I/O (A11) 49 P221 B16 C19 D20 W7
I/O 12 - - A28 E27 F2 I/O (A10) 50 P220 A16 B19 C20 Y4
I/O 13 - B23 B27 C29 H4 I/O 51 P218 D15 A19 B20 AA1
I/O 14 - A24 C26 B30 G1 I/O 52 P217 C15 B18 E19 Y6
I/O 15 - A23 D25 D27 J5 I/O (A18) 53 P216 B15 D18 D19 AB2
I/O 16 - D20 A27 C28 L7 I/O (A19) 54 P215 A15 C18 C19 Y8
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O 55 - - - - -
GND - - GND* GND* GND* GND* I/O 56 - - - - -
I/O (A13) 17 P233 C21 B26 E26 J3 GND - - - GND* GND* GND*
I/O (A12) 18 P232 B22 D24 B29 K4 VCCIO - - - VCCIO* VCCIO* -
I/O 19 P231 B21 C25 B28 H2 I/O 57 - - - - -
I/O 20 P230 C20 A26 D26 L5 I/O 58 - - - - -
I/O 21 - - B25 C27 J1 I/O 59 - - - A19 AA3
I/O 22 - - D23 E25 M6 I/O 60 - - - D18 AA5
I/O 23 - - - A28 K2 I/O 61 - - B17 E18 AC1
I/O 24 - - - A27 N7 I/O 62 - - C17 C18 AA7
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O (A9) 63 P214 C14 A17 B18 AB4
GND - - - GND* GND* GND* I/O (A8) 64 P213 D14 D17 A17 AB6
I/O 25 - - - D25 P8 VCCIO - P212 VCCIO* VCCIO* VCCIO* VCCIO*
I/O 26 - - - C26 L3 GND - P211 GND* GND* GND* GND*
I/O 27 - - - E24 L1 I/O (A7) 65 P210 B14 A16 D17 AC3
I/O 28 - - - B26 M4 I/O (A6) 66 P209 A13 B16 E17 AC5
I/O 29 P229 B20 C24 C25 P2 VCCINT - - - VCCINT* VCCINT* VCCINT*
I/O 30 P228 A21 B24 D24 N5 I/O 67 - - - - AD2
I/O 31 - D18 C23 B25 R1 I/O 68 - - A15 B17 AC7
I/O 32 - C19 D22 E23 N3 I/O 69 - - - B16 AF2
VCCIO - - - VCCIO* VCCIO* - I/O 70 - - - D16 AD4
GND - P227 GND* GND* GND* GND* I/O 71 - - - - -
I/O 33 P226 B19 A24 A25 R7 I/O 72 - - - - -
I/O 34 - - - - P4 VCCIO - - - VCCIO* VCCIO* -
VCCINT - P225 A20 VCCINT* VCCINT* VCCINT* GND - - - GND* GND* GND*
I/O 35 P224 D17 C22 D23 T2 I/O 73 - - - - -
I/O 36 P223 C18 B22 B24 T8 I/O 74 - - - - -
VCCIO - P222 VCCIO* VCCIO* VCCIO* VCCIO* I/O (A20) 75 P208 B13 B15 E16 AD6
I/O 37 - B18 A22 E22 R5 I/O (A21) 76 P207 C13 C15 C16 AE5

R
XC40110XV Pinout Table (Continued) XC40110XV Pinout Table (Continued)

PAD EPIC PAD EPIC
HQ240 BG352 BG432 BG560 PG559 HQ240 BG352 BG432 BG560 PG559
NAME PAD# NAME PAD#
I/O 77 P206 A12 D15 A15 AF4 I/O (A3) 115 P188 C6 D7 E8 BA1
I/O 78 P205 B12 B14 C15 AG1 I/O, 116 P187 D6 A5 C6 AU5
I/O (A5) 79 P203 C12 A13 D15 AD8 (CS1,A2)
I/O (A4) 80 P202 D12 C14 E15 AG3 I/O 117 - B4 C6 D7 AY2
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O 118 - C5 B5 B4 AT6
I/O 81 - A11 B13 B14 AH4 I/O 120 - - - - -
I/O 82 - B11 D14 C14 AE7 GND - - - GND* GND* GND*
I/O 83 - B10 A12 D14 AH2 VCCIO - - - VCCIO* VCCIO* -
I/O 84 - C11 C13 A13 AF6 I/O 121 - - - - -
I/O 85 - - B12 E14 AJ1 I/O 122 - - - - -
I/O 86 - - D13 C13 AG5 I/O 123 - - D6 C5 AP8
I/O 87 - - - D13 AJ3 I/O 124 - - A4 E7 AR7
I/O 88 - - - B12 AK4 I/O 125 - - C5 A3 AY4
VCCIO - - - VCCIO* VCCIO* - I/O 126 P186 A3 - - BB2
GND - - - GND* GND* GND* VCCINT - P185 D5 VCCINT* VCCINT* VCCINT*
I/O 89 - - - C12 AG7 I/O GCK7 127 P184 C4 D5 A2 AV6
(A1)
I/O 90 - - - E13 AK2
I/O 128 P183 B3 B3 D5 AT8
I/O 91 - D11 C12 A11 AJ5 (A0, /WS)
I/O 92 - A9 B11 D12 AL3 GND - P182 GND* GND* GND* GND*
6
VCCIO - P201 VCCIO* VCCIO* VCCIO* VCCIO* O, TDO - P181 D4 C4 E6 BA3
I/O 93 P200 B9 D12 B11 AM4 VCCIO - P180 - VCCIO* VCCIO* VCCIO*
I/O 94 P199 C10 C11 C11 AN1 CCLK - P179 C3 D4 C4 BA5
VCCINT - P198 D10 VCCINT* VCCINT* VCCINT* I/O GCK6 129 P178 E4 D3 D4 BB4
I/O 95 - - - - AL5 (DOUT)
I/O 96 P197 B8 B10 B10 AH8 I/O (D0, 130 P177 D3 C2 E4 AY6
GND - P196 GND* GND* GND* GND* DIN)
VCCIO - - - VCCIO* VCCIO* - I/O 131 P176 C2 D2 F5 BC3
I/O 97 - C9 C10 D11 AJ7 I/O 132 P175 E3 E4 B3 AW7
I/O 98 P195 D9 B9 C10 AP2 I/O 133 - F4 D1 D3 BB6
I/O 99 P194 A7 C9 A9 AN3 I/O 134 - D2 E3 F4 AU9
I/O 100 P193 B7 D10 E11 AP4 I/O 135 - - - - -
I/O 101 - - - C9 AR1 I/O 136 - - - - -
I/O 102 - - - D10 AN5 VCCIO - - - VCCIO* VCCIO* -
I/O 103 - - - A8 AM6 GND - - - GND* GND* GND*
I/O 104 - - - B8 AK8 I/O 137 - - - - -
GND - - - GND* GND* GND* I/O 138 - - - - -
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O 139 - - E2 C1 AT10
I/O 105 - - - E10 AL7 I/O 140 - - F4 G5 AV8
I/O 106 - - - C8 AT2 I/O 141 - C1 E1 E3 AY8
I/O 107 - - A8 D9 AR3 I/O 142 - D1 F3 D2 BC7
I/O 108 - - B8 B7 AU1 I/O (/RCK, 143 P174 G4 F2 G4 AW9
RDY_/
I/O 109 P192 D8 D9 A6 AT4
BUSY
I/O 110 P191 A6 B7 E9 AV2 I/O (D1) 144 P173 F3 G4 F3 BA9
I/O 111 P190 B6 C7 C7 AR5
VCCIO - - - VCCIO* VCCIO* VCCIO*
I/O 112 P189 C7 D8 D8 AN7
GND - - GND* GND* GND* GND* I/O 145 - - - H5 AU11
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 146 - - - E2 AY10
I/O 113 - A4 A6 A5 AW3
I/O 147 - - F1 F2 BB8
I/O 114 - B5 B6 B5 AV4 I/O 148 - - G3 H4 AW11

R

PAD EPIC PAD EPIC
NAME PAD# NAME PAD#
I/O 149 - - G2 G3 BC9 I/O 185 P155 M1 R4 R1 AW21
I/O 150 - - H4 J5 AV12 I/O 186 - - - T4 AU21
GND - - - - - - I/O 187 - - - T5 BB22
I/O 151 P172 E2 H3 F1 AU13 I/O 188 - - - - AY22
I/O 152 P171 F2 H2 G1 AT14 I/O 189 - - - - -
VCCIO - - - VCCIO* VCCIO* VCCIO* I/O 190 - - - - -
GND - - - GND* GND* GND* VCCINT - P154 N3 VCCINT* VCCINT* VCCINT*
I/O 153 - - - J4 BA11 I/O (/RS) 191 P153 N4 R1 T2 AV22
I/O 154 - - - H3 AY12 I/O (D3) 192 P152 N2 T3 U3 BA23
I/O 155 P170 G3 - K5 BB10 GND - P151 - GND* GND* GND*
I/O 156 P169 G2 - H2 AW13 VCCIO - P150 - VCCIO* VCCIO* VCCIO*
I/O 157 P168 F1 H1 J3 BC11 I/O 193 P149 P1 T2 U4 AW23
I/O 158 P167 J4 J4 K4 AU15 I/O (D4) 194 P148 P2 T1 U5 AY24
I/O 159 - H3 J3 J2 BB14 I/O 195 - - - - -
I/O 160 - H2 J2 L5 AT16 I/O 196 - - - - -
VCCIO - - - VCCIO* VCCIO* - I/O 197 - - - U1 BC23
GND - P166 GND* GND* GND* GND* I/O 198 - - - U2 BA27
I/O 161 P165 G1 K4 K3 BA13 I/O 199 P147 P3 U3 V2 AU23
I/O 162 - - - - AY14 I/O 200 P146 R1 U4 V4 AV24
VCCINT - P164 K4 VCCINT* VCCINT* VCCINT* VCCIO - - - VCCIO* VCCIO* -
I/O, 163 P163 J3 K2 L4 BC15 GND - - - GND* GND* GND*
FCLK4 I/O 201 P145 R2 U2 V5 AY26
I/O 164 P162 J2 K1 K2 AW15 I/O 202 P144 R3 U1 V3 BB24
VCCIO - P161 VCCIO* VCCIO* VCCIO* VCCIO* I/O 203 - R4 V3 W1 AW25
I/O 165 - - - M5 BA15 I/O 204 - T1 V4 W3 BB26
I/O 166 - - - L3 AU17 I/O 205 - - V2 W4 AT24
I/O 167 P160 K3 L3 L1 BB16 I/O 206 - - W2 W5 BA29
I/O (D2) 168 P159 J1 L2 M4 AY16 I/O 207 - - - - -
GND - - - GND* GND* GND* I/O 208 - - - - -
VCCIO - - - VCCIO* VCCIO* - VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 169 - - M4 M3 AW17 GND - P143 GND* GND* GND* GND*
I/O 170 - - M3 N5 AV18 I/O 209 - T2 W3 Y2 AY28
I/O 171 - - - M2 BA17 I/O 210 - U2 W4 Y3 AU25
I/O 172 - - - N4 BC17 I/O 211 - - - Y4 AV26
I/O 173 - L4 M2 N3 AU19 I/O 212 - - - AA1 BC27
I/O 174 - K2 M1 N2 BB18 I/O 213 - - W1 Y5 AW27
I/O 175 - L3 N4 P5 AY18 I/O 214 - - Y1 AA3 BB28
I/O 176 - L2 N3 N1 AW19 I/O 215 - T3 Y3 AA4 BA31
GND - P158 GND* GND* GND* GND* I/O 216 - T4 Y4 AB2 AY30
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* VCCIO - - - VCCIO* VCCIO* VCCIO*
I/O 177 - - - - - GND - - - GND* GND* GND*
I/O 178 - - - - - I/O (/CS0) 217 P142 V1 Y2 AB3 AW29
I/O 179 - - N2 P4 AV20 I/O (D5) 218 P141 V2 AA2 AA5 BC29
I/O 180 - - N1 P3 AT20 I/O 219 - - - AC1 AU27
I/O 181 - L1 P4 P2 BB20 I/O 220 - - - AB4 BA33
I/O 182 - M4 P3 R5 AY20 VCCIO - P140 VCCIO* VCCIO* VCCIO* VCCIO*
I/O 183 P157 M3 P2 R4 BC21 I/O 221 P139 U3 AA3 AC3 AY32
I/O 184 P156 M2 R3 R3 BA21 I/O, 222 P138 U4 AB1 AB5 AW31
GND - - - GND* GND* GND* FCLK3
VCCIO - - - VCCIO* VCCIO* - VCCINT - P137 W2 VCCINT* VCCINT* VCCINT*

R

PAD EPIC PAD EPIC
NAME PAD# NAME PAD#
I/O 223 - - - - BB30 I/O 259 - - - - AV42
I/O 224 P136 V3 AB3 AC4 BA35 I/O 260 P115 AD5 AK4 AN3 AR37
GND - P135 GND* GND* GND* GND* I/O 261 - - AJ5 AK6 AP36
VCCIO - - - VCCIO* VCCIO* - I/O 262 - - AH6 AL5 AT38
I/O 225 - V4 AB4 AD3 AT28 I/O 263 - - - - -
I/O 226 - Y1 AC2 AE1 AU29 I/O 264 - - - - -
I/O 227 P134 Y2 AD1 AC5 BC33 VCCIO - - - VCCIO* VCCIO* -
I/O 228 P133 W3 AC3 AE3 AY34 GND - - - GND* GND* GND*
I/O 229 - - - AD4 BB34 I/O 265 - - - - -
I/O 230 - - - AF1 AW33 I/O 266 - - - - -
I/O 231 - - - AF2 AU31 I/O 267 - AC6 AL4 AJ7 AU39
I/O 232 - - - AD5 AV32 I/O 268 - AE4 AK5 AM4 AU43
GND - - - GND* GND* GND* I/O 269 P114 AF3 AJ6 AM5 BA43
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O 270 P113 AF4 AH7 AK7 AT42
I/O 233 - - - AG2 AT30 I/O 271 - AC7 AL5 AL6 AR39
I/O 234 - - - AE4 AU33 I/O 272 - AD6 AK6 AJ8 AN37
I/O 235 - - AC4 AF3 AW35 VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 236 - - AD2 AH1 BC35 GND - - GND* GND* GND* GND*
I/O 237 P132 W4 AE2 AE5 AY36 I/O 273 P112 AE5 AH8 AM6 AT40
I/O 238 P131 AA1 AD3 AG3 BB36 I/O 274 P111 AE6 AJ7 AN6 AP40 6
I/O 239 P130 AA2 AD4 AF4 AV36 I/O 275 P110 AD7 AL6 AK8 AR43
I/O (D6) 240 P129 Y3 AF1 AJ1 AU35 I/O 276 P109 AE7 AK7 AL7 AN39
GND - - - GND* GND* GND* I/O 277 - - AH9 AJ9 AP42
VCCIO - - - VCCIO* VCCIO* VCCIO* I/O 278 - - AJ8 AN7 AM38
I/O 241 - AC1 AF2 AJ2 AT34 I/O 279 - - - AL8 AN43
I/O 242 - AB2 AE3 AF5 AW37 I/O 280 - - - AK9 AL37
I/O 244 - - - - - GND - - - GND* GND* GND*
I/O 245 - - AE4 AH3 BC37 I/O 281 - - - AM8 AK36
I/O 246 - - AG1 AG4 AY38 I/O 282 - - - AJ10 AR41
I/O 247 P128 AA3 AG2 AK2 BB38 I/O 283 - - - AL9 AL41
I/O 248 P127 AA4 AF3 AJ3 BA41 I/O 284 - - - AM9 AN41
GND - - - GND* GND* GND* I/O 285 P108 AF6 AK8 AK10 AK42
VCCIO - - - VCCIO* VCCIO* - I/O 286 P107 AC9 AJ9 AN9 AM40
I/O 249 - - - - - I/O 287 - AD8 AH10 AJ11 AJ43
I/O 250 - - - - - I/O 288 - AE8 AL8 AL10 AJ37
I/O 251 - AC2 AF4 AG5 AY40 VCCIO - - - VCCIO* VCCIO* -
I/O 252 - AB3 AH1 AL1 BB40 GND - P106 GND* GND* GND* GND*
I/O 253 P126 AD1 AH2 AH4 AT36 I/O 289 P105 AF7 AK9 AM10 AL39
I/O 254 P125 AB4 AG3 AK3 BA39 I/O 290 - - - - AH36
I/O, GCK5 255 P124 AC3 AG4 AJ4 AV38 VCCINT - P104 AC10 VCCINT* VCCINT* VCCINT*
I/O (D7) 256 P123 AD2 AJ2 AH5 BC41 I/O 291 P103 AD9 AK10 AL11 AH42
/PRO- - P122 AC4 AH3 AM1 BB42 I/O 292 P102 AE9 AL10 AJ12 AJ39
GRAM VCCIO - P101 VCCIO* VCCIO* VCCIO* VCCIO*
VCCIO - P121 VCCIO* VCCIO* VCCIO* VCCIO* I/O 293 P100 AD10 AJ11 AN11 AK40
DONE - P120 AD3 AH4 AJ5 AY42 I/O 294 P99 AF9 AH12 AK12 AG41
GND - P119 GND* GND* GND* GND* I/O 295 - - - AL12 AJ41
I/O, GCK4 257 P118 AC5 AJ4 AL4 AW41 I/O 296 - - - AM12 AH40
I/O 258 P117 AD4 AK3 AJ6 AV40 GND - - - GND* GND* GND*
VCCINT - P116 AE3 VCCINT* VCCINT* VCCINT* VCCIO - - - VCCIO* VCCIO* -

R

PAD EPIC PAD EPIC
NAME PAD# NAME PAD#
I/O 297 - - - AJ13 AG37 GND - P83 GND* GND* GND* GND*
I/O 298 - - - AK13 AG43 I/O 337 - AE16 AH18 AK20 V38
I/O 299 - - AK11 AL13 AG39 I/O 338 - AE17 AL20 AN21 U39
I/O 300 - - AJ12 AM13 AF38 I/O 339 - AD16 AJ19 AL21 V42
I/O 301 - AC11 AK12 AN13 AF42 I/O 340 - AC16 AK20 AJ20 R41
I/O 302 - AE10 AL12 AJ14 AD42 I/O 341 - - AH19 AM22 U43
I/O 303 - AD11 AH13 AK14 AF40 I/O 342 - - AJ20 AK21 P40
I/O 304 - AE11 AJ13 AL14 AE37 I/O 343 - - - AN23 T42
GND - P98 GND* GND* GND* GND* I/O 344 - - - AL22 U37
I/O 305 - AF11 AK13 AM14 AE39 GND - - - GND* GND* GND*
I/O 306 - AC12 AL13 AN15 AD40 I/O 345 - - - AJ21 R39
I/O 307 P97 AD12 AK14 AJ15 AC43 I/O 346 - - - AM23 N41
I/O 308 P96 AE12 AH14 AK15 AD38 I/O 347 P82 AF18 AK21 AK22 R43
I/O 309 P95 AF12 AJ14 AL15 AC41 I/O 348 P81 AE18 AH20 AM24 M40
I/O 310 P94 AD13 AK15 AM16 AD36 VCCIO - P80 VCCIO* VCCIO* VCCIO* -
I/O 311 - - - - - I/O 349 P79 AD17 AJ21 AL23 N39
I/O 312 - - - - - I/O 350 P78 AC17 AL22 AJ22 T36
GND - - - GND* GND* GND* VCCINT - P77 AE19 VCCINT* VCCINT* VCCINT*
VCCIO - - - VCCIO* VCCIO* - I/O 351 - - - - P42
I/O 313 - - - - - I/O 352 P76 AD18 AJ22 AK23 R37
I/O 314 - - - - - GND - P75 GND* GND* GND* GND*
I/O 315 - - - AL16 AC39 VCCIO - - - VCCIO* VCCIO* -
I/O 316 - - - AJ16 AC37 I/O 353 - AC18 AK23 AL24 L41
I/O 317 - - AJ15 AK16 AB40 I/O 354 - AF20 AJ23 AN26 L43
I/O 318 - - AL15 AN17 AB42 I/O 355 P74 AE20 AH22 AJ23 K40
VCCINT - P93 AC13 VCCINT* VCCINT* VCCINT* I/O 356 P73 AD19 AL24 AL25 K42
I/O 319 - - - - AB38 I/O 357 - - - AK24 L39
I/O 320 P92 AE13 AL16 AL17 AA41 I/O 358 - - - AM26 J43
GND - P91 GND* GND* GND* GND* I/O 359 - - - AM27 M38
VCCIO - P90 VCCIO* VCCIO* VCCIO* VCCIO* I/O 360 - - - AJ24 P36
I/O (/INIT) 321 P89 AF14 AK16 AJ17 AA39 GND - - - GND* GND* GND*
I/O 322 P88 AE14 AJ16 AK17 AA37 VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 323 - - AL17 AM18 Y40 I/O 361 - - - AL26 N37
I/O 324 - - AK17 AL18 Y38 I/O 362 - - - AK25 H42
I/O 325 - - - AK18 AA43 I/O 363 - - AK24 AN28 J41
I/O 326 - - - AJ18 W39 I/O 364 - - AH23 AN29 G43
I/O 327 - - - - - I/O 365 P72 AC19 AJ24 AJ25 H40
I/O 328 - - - - - I/O 366 P71 AF21 AK25 AL27 F42
VCCIO - - - VCCIO* VCCIO* - I/O 367 P70 AE21 AJ25 AK26 J39
GND - - - GND* GND* GND* I/O 368 P69 AD20 AH24 AM29 L37
I/O 331 P87 AD14 AJ17 AN19 V40 I/O 369 - AF23 AL26 AM30 E41
I/O 332 P86 AF15 AH17 AL19 Y36 I/O 370 - AE22 AK26 AJ26 F40
I/O 333 P85 AE15 AK18 AK19 U41 I/O 371 - AD21 AH25 AL28 C43
I/O 334 P84 AD15 AL19 AM20 Y42 I/O 372 - AC21 AL27 AK27 G39
I/O 335 - AC15 AJ18 AJ19 T40 I/O 373 - - AJ26 AL29 D42
I/O 336 - AF16 AK19 AL20 W37 I/O 374 - - AK27 AN31 H38

R

PAD EPIC PAD EPIC
NAME PAD# NAME PAD#
I/O 376 - - - - - I/O 412 - - - AD30 H28
GND - - - GND* GND* GND* I/O 413 P47 V23 AC28 AE32 B34
VCCIO - - - VCCIO* VCCIO* - I/O 414 P46 W24 AC29 AC29 C31
I/O 377 - - - - - I/O 415 - W25 AC30 AE33 A33
I/O 378 - - - - - I/O 416 - Y26 AB28 AD31 D30
I/O (/LDC) 379 P68 AE23 AH26 AJ27 K36 VCCIO - - - VCCIO* VCCIO* -
I/O 380 P67 AD22 AL28 AM31 J37 GND - P45 GND* GND* GND* GND*
I/O 381 P66 AF24 AJ27 AK28 B42 I/O, 417 P44 U23 AB29 AC30 E29
I/O 382 P65 AC22 AK28 AL30 D40 FCLK2
I/O (HDC) 383 P64 AD23 AH27 AK29 C41 I/O 418 - - - - C29
I/O, GCK3 384 P63 AE24 AK29 AJ28 F38 VCCINT - P43 V24 VCCINT* VCCINT* VCCINT*
I (M2) - P62 AC23 AJ28 AN32 H36 I/O 419 P42 V25 AB31 AB29 B30
VCCIO - P61 VCCIO* VCCIO* VCCIO* VCCIO* I/O 420 P41 U24 AA29 AC31 G27
I (M0) - P60 AD24 AH28 AJ29 C39 VCCIO - P40 VCCIO* VCCIO* VCCIO* VCCIO*
GND - P59 GND* GND* GND* GND* I/O 421 - - - AC33 D28
O (M1) - P58 AB23 AH29 AK30 D38 I/O 422 - - - AB30 A29
I/O, GCK2 385 P57 AC24 AJ30 AH29 E37 I/O 423 P39 V26 AA30 AB31 E27
I/O 386 P56 AD25 AH30 AJ30 B40 I/O 424 P38 T23 Y28 AA29 F26
VCCINT - P55 AB24 VCCINT* VCCINT* VCCINT* GND - - - GND* GND* GND*
I/O 387 - - - - H34 VCCIO - - - VCCIO* VCCIO* - 6
I/O 388 P54 AA23 AH31 AH30 G35 I/O 425 - U25 Y29 AB32 C27
I/O 389 - AC25 AG29 AL33 F36 I/O 426 - T24 Y30 AA30 B28
I/O 390 - AD26 AG30 AG29 D36 I/O 427 - - Y31 AA31 G25
I/O 391 - - - - - I/O 428 - - W28 AA32 A27
I/O 392 - - - - - I/O 429 - - - Y29 D26
VCCIO - - - VCCIO* VCCIO* - I/O 430 - - - AA33 B26
GND - - - GND* GND* GND* I/O 431 - T25 W29 Y30 E25
I/O 393 P53 AC26 AF28 AJ31 E35 I/O 432 - T26 W30 Y31 F24
I/O 394 P52 Y23 AG31 AK32 A41 GND - P37 GND* GND* GND* GND*
I/O 395 - - AF29 AG30 G33 VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 396 - - AF30 AH31 B38 I/O 433 - - - - -
I/O 397 - - - - - I/O 434 - - - - -
I/O 398 - - - - - I/O 435 - - W31 Y32 H24
I/O 399 - AA24 AE28 AF29 D34 I/O 436 - - V28 W29 B24
I/O 400 - AB25 AF31 AJ32 E33 I/O 437 - R23 V29 W30 D24
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O 438 - R24 V30 W31 A23
GND - - GND* GND* GND* GND* I/O 439 P36 R25 U29 W33 C23
I/O 401 P51 AA25 AE29 AH32 F32 I/O 440 P35 R26 U28 V30 E23
I/O 402 P50 Y24 AE30 AF30 G31 GND - - - GND* GND* GND*
I/O 403 P49 Y25 AD28 AG31 A37 VCCIO - - - VCCIO* VCCIO* -
I/O 404 P48 AA26 AD29 AE29 H30 I/O 441 P34 P24 U30 V29 G23
I/O 405 - - AD30 AH33 B36 I/O 442 P33 P23 U31 V31 B22
I/O 406 - - AD31 AG33 C33 I/O 443 - - - V32 D22
I/O 407 - - - AE30 C35 I/O 444 - - - U33 A21
I/O 408 - - - AF31 D32 I/O 445 - - - - -
GND - - - GND* GND* GND* I/O 447 P32 P25 T29 U30 F22
I/O 409 - - - AD29 E31 I/O 448 P31 N26 T30 U29 C21
I/O 410 - - - AF32 G29 VCCIO - P30 VCCIO* VCCIO* VCCIO* VCCIO*
I/O 411 - - - AE31 A35 GND - P29 GND* GND* GND* GND*

R

PAD EPIC PAD EPIC
NAME PAD# NAME PAD#
I/O 449 P28 N25 T31 U31 E21 I/O 487 P11 F26 - K29 E11
I/O 450 - - - - D20 I/O 488 P10 F25 - H31 G13
VCCINT - P27 N24 VCCINT* VCCINT* VCCINT* GND - - - GND* GND* GND*
I/O 452 - - - - - I/O 489 P9 G24 H30 J30 F12
I/O 453 - - - T32 C17 I/O 490 P8 D26 G30 G32 H14
I/O 454 - - - T30 G21 I/O 491 - - H29 F33 G11
I/O 455 P26 M26 R28 T29 F20 I/O 492 - - H28 J29 A7
I/O 456 P25 M25 R30 T31 D18 I/O 493 - - F31 G31 E9
VCCIO - - - VCCIO* VCCIO* - I/O 494 - - F30 H30 B6
GND - - - GND* GND* GND* I/O 495 - - - E33 D8
I/O 457 P24 M24 R31 R33 E19 I/O 496 - - - E32 F8
I/O 458 P23 M23 P29 R31 B20 GND - - GND* GND* GND* GND*
I/O 459 - L26 P28 R30 H20 VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 460 - L25 P30 R29 B18 I/O 497 - E25 G29 H29 G9
I/O 461 - - N30 P32 C15 I/O 498 - F24 G28 F31 H10
I/O 462 - - N29 P31 D16 I/O 499 - F23 E31 G30 B4
I/O 463 - - - - - I/O 500 - D25 E30 D32 E7
I/O 464 - - - - - I/O 501 - - F29 E31 C5
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O 502 - - F28 G29 D6
I/O 465 - K25 N28 N33 G19 I/O 504 - - - - -
I/O 466 - L24 N31 P30 A17 GND - - - GND* GND* GND*
I/O 467 - L23 M31 P29 F18 VCCIO - - - VCCIO* VCCIO* -
I/O 468 - J26 M29 M32 E17 I/O 505 - - - - -
I/O 469 - - - N31 B16 I/O 506 - - - - -
I/O 470 - - - N30 C13 I/O (TCK) 507 P7 E24 D31 C33 D4
I/O 471 - - M28 L33 A15 I/O (TDI) 508 P6 C26 D30 F30 H8
I/O 472 - - M30 N29 D14 I/O 509 P5 E23 E29 D31 A3
VCCIO - - - VCCIO* VCCIO* - I/O 510 - - - - F6
GND - - - GND* GND* GND* VCCINT - P4 D24 VCCINT* VCCINT* VCCINT*
I/O 473 P21 J25 L30 M31 E15 I/O (A17) 511 P3 C25 C30 F29 C3
I/O 474 P20 K24 L29 L32 G17 I/O, GCK1 512 P2 D23 D29 B33 C1
I/O 475 - - - M30 B14 (A16)
I/O 476 - - - L31 C11 GND - P1 GND* GND* GND* GND*
VCCIO - P19 VCCIO* VCCIO* VCCIO* VCCIO* VCCIO - - A10 A1 A4 A13
I/O 477 P18 K23 K31 M29 D12 VCCIO - - A17 A11 A10 A31
I/O (TMS) 478 P17 H25 K30 J33 A11 VCCIO - - AC14 A21 A16 A43
VCCINT - P16 J24 VCCINT* VCCINT* VCCINT* VCCIO - - AC20 A31 A22 B2
I/O 479 - - - - E13 VCCIO - - AC8 D11 A26 C7
I/O, 480 P15 J23 K28 L30 C9 VCCIO - - AF10 D21 A30 C19
FCLK1 VCCIO - - AF17 L1 B2 C25
GND - P14 GND* GND* GND* GND* VCCIO - - D7 L4 B13 C37
VCCIO - - - VCCIO* VCCIO* - VCCIO - - D13 L28 B19 F14
I/O 481 - G26 J30 K31 H16 VCCIO - - D19 L31 B32 F30
I/O 482 - G25 H31 L29 B10 VCCIO - - G23 AA1 C3 G3
I/O 483 P13 H24 J29 H33 G15 VCCIO - - H4 AA4 C31 G7
I/O 484 P12 H23 J28 K30 A9 VCCIO - - K1 AA28 C32 G37
I/O 485 - - - J31 D10 VCCIO - - K26 AA31 D1 G41
I/O 486 - - - H32 B8 VCCIO - - N23 AH11 D33 N1

R

PAD EPIC PAD EPIC
NAME PAD# NAME PAD#
VCCIO - - P4 AH21 E5 N43 GND - - A2 A9 A18 A39
VCCIO - - U1 AL1 H1 P6 GND - - A22 A14 A20 B12
VCCIO - - U26 AL11 K33 P38 GND - - A25 A18 A24 B32
VCCIO - - W23 AL21 M1 W3 GND - - A26 A23 A29 E1
VCCIO - - Y4 AL31 N32 W41 GND - - A5 A25 A32 E5
VCCIO - - B2 C3 R2 AE3 GND - - A8 A29 B1 E39
VCCIO - - B25 C29 T33 AE41 GND - - AB1 A30 B6 E43
VCCIO - - AE2 AJ3 V1 AK6 GND - - AB26 B1 B9 F10
VCCIO - - AE25 AJ29 W32 AK38 GND - - AE1 B2 B15 F16
VCCIO - - - - AA2 AL1 GND - - AE26 B30 B23 F28
VCCIO - - - - AB33 AL43 GND - - AF1 B31 B27 F34
VCCIO - - - - AD1 AU3 GND - - AF13 C1 B31 H22
VCCIO - - - - AF33 AU7 GND - - AF19 C31 C2 K6
VCCIO - - - - AK1 AU37 GND - - AF2 D16 E1 K38
VCCIO - - - - AK4 AU41 GND - - AF22 G1 F32 M2
VCCIO - - - - AK33 AV14 GND - - AF25 G31 G2 M42
VCCIO - - - - AL2 AV30 GND - - AF26 J1 G33 T6
VCCIO - - - - AL3 BA7 GND - - AF5 J31 J32 T38
VCCIO - - - - AL31 BA19 GND - - AF8 P1 K1 W1
VCCIO - - - - AM2 BA25 GND - - B1 P31 L2 W43 6
VCCIO - - - - AM15 BA37 GND - - B26 T4 M33 AB8
VCCIO - - - - AM21 BC1 GND - - E1 T28 P1 AB36
VCCIO - - - - AM32 BC13 GND - - E26 V1 P33 AE1
VCCIO - - - - AN4 BC31 GND - - H1 V31 R32 AE43
VCCIO - - - - AN8 BC43 GND - - H26 AC1 T1 AH6
VCCIO - - - - AN12 - GND - - N1 AC31 V33 AH38
VCCIO - - - - AN18 - GND - - P26 AE1 W2 AM2
VCCIO - - - - AN24 - GND - - W1 AE31 Y1 AM42
VCCIO - - - - AN30 - GND - - W26 AH16 Y33 AP6
GND - - - AJ1 AB1 AP38
VCCINT - - - A10 E12 H12 GND - - - AJ31 AC32 AT22
VCCINT - - - AB2 AD2 H18 GND - - - AK1 AD33 AV10
VCCINT - - - AB30 AD32 H26 GND - - - AK2 AE2 AV16
VCCINT - - - AG28 AK31 H32 GND - - - AK30 AG1 AV28
VCCINT - - - AH15 AM17 M8 GND - - - AK31 AG32 AV34
VCCINT - - - AH5 AK5 M36 GND - - - AL2 AH2 AW1
VCCINT - - - AJ10 AK11 V8 GND - - - AL3 AJ33 AW5
VCCINT - - - AK22 AN25 V36 GND - - - AL7 AL32 AW39
VCCINT - - - B23 C24 AF8 GND - - - AL9 AM3 AW43
VCCINT - - - B4 D6 AF36 GND - - - AL14 AM11 BB12
VCCINT - - - C16 C17 AM8 GND - - - AL18 AM19 BB32
VCCINT - - - E28 E30 AM36 GND - - - AL23 AM25 BC5
VCCINT - - - K29 K32 AT12 GND - - - AL25 AM28 BC19
VCCINT - - - K3 J1 AT18 GND - - - AL29 AM33 BC25
VCCINT - - - R2 T3 AT26 GND - - - AL30 AM7 BC39
VCCINT - - - R29 U32 AT32 GND - - - - AN2 -
GND - - - - AN5 -
GND - - A1 A2 A7 A5 GND - - - - AN10 -
GND - - A14 A3 A12 A19 GND - - - - AN14 -
GND - - A19 A7 A14 A25 GND - - - - AN16 -

R
XC40110XV Pinout Table (Continued)

PAD EPIC
HQ240 BG352 BG432 BG560 PG559
NAME PAD#
GND - - - - AN20 -
GND - - - - AN22 -
GND - - - - AN27 -
NC - P204 C8 C8 A1 -
NC - P219 - - A33 -
NC - - - - AC2 -
NC - - - - AN1 -
NC - - - - AN33 -


XC40150XV Pinout Table EPIC
PAD NAME HQ240 BG352 BG432 BG560 PG559
PAD #
EPIC I/O 44 - - - D22 T4
PAD NAME HQ240 BG352 BG432 BG560 PG559
PAD #
VCCIO - - - VCCIO* VCCIO* -
I/O, GCK8 (A15) 1 P239 C24 D28 E29 E3
I/O 45 - - - - -
I/O (A14) 2 P238 D22 C28 D30 F4
I/O 46 - - - - -
I/O 3 P237 C23 B29 E28 J7
I/O 47 - - - C22 U5
I/O 4 P236 B24 D27 D29 D2
I/O 48 - - - E21 V2
I/O 5 P235 C22 B28 C30 K8
I/O 49 - - D20 B22 U7
I/O 6 P234 D21 C27 D28 H6
I/O 50 - - B21 D21 U3
I/O 7 - - - - -
I/O 51 - A18 C20 C21 Y2
I/O 8 - - - - -
I/O 52 - D16 B20 B21 V4
I/O 53 - B17 A20 E20 V6
I/O 54 - C16 D19 A21 W5
I/O 9 - - - - -
GND - - GND* GND* GND* GND*
I/O 10 - - - - -
I/O 11 - - - - -
I/O (A11) 55 P221 B16 C19 D20 W7
I/O 12 - - - - -
I/O (A10) 56 P220 A16 B19 C20 Y4
I/O 13 - - D26 A31 G5
I/O 57 P218 D15 A19 B20 AA1
I/O 14 - - A28 E27 F2
I/O 58 P217 C15 B18 E19 Y6
I/O 15 - B23 B27 C29 H4
I/O (A18) 59 P216 B15 D18 D19 AB2
I/O 16 - A24 C26 B30 G1
I/O (A19) 60 P215 A15 C18 C19 Y8
I/O 17 - A23 D25 D27 J5
I/O 18 - D20 A27 C28 L7
I/O 61 - - - - - 6
I/O 62 - - - - -
I/O 63 - - - - -
I/O 64 - - - - -
I/O (A13) 19 P233 C21 B26 E26 J3
I/O (A12) 20 P232 B22 D24 B29 K4
I/O 21 P231 B21 C25 B28 H2
I/O 65 - - - - -
I/O 22 P230 C20 A26 D26 L5
I/O 66 - - - - -
I/O 23 - - B25 C27 J1
I/O 67 - - - A19 AA3
I/O 24 - - D23 E25 M6
I/O 68 - - - D18 AA5
I/O 25 - - - A28 K2
I/O 69 - - B17 E18 AC1
I/O 26 - - - A27 N7
I/O 70 - - C17 C18 AA7
I/O 27 - - - - -
I/O (A9) 71 P214 C14 A17 B18 AB4
I/O 28 - - - - -
I/O (A8) 72 P213 D14 D17 A17 AB6
GND - P211 GND* GND* GND* GND*
I/O 29 - - - D25 P8
I/O (A7) 73 P210 B14 A16 D17 AC3
I/O 30 - - - C26 L3
I/O (A6) 74 P209 A13 B16 E17 AC5
I/O 31 - - - E24 L1
VCCINT - - - VCCINT* VCCINT* VCCINT*
I/O 32 - - - B26 M4
I/O 75 - - - - AD2
I/O 33 P229 B20 C24 C25 P2
I/O 76 - - A15 B17 AC7
I/O 34 P228 A21 B24 D24 N5
I/O 77 - - - B16 AF2
I/O 35 - D18 C23 B25 R1
I/O 78 - - - D16 AD4
I/O 36 - C19 D22 E23 N3
I/O 79 - - - - -
I/O 80 - - - - -
GND - P227 GND* GND* GND* GND*
I/O 37 P226 B19 A24 A25 R7
I/O 38 - - - - P4
I/O 81 - - - - -
VCCINT - P225 A20 VCCINT* VCCINT* VCCINT*
I/O 82 - - - - -
I/O 39 P224 D17 C22 D23 T2
I/O 83 - - - - -
I/O 40 P223 C18 B22 B24 T8
I/O 84 - - - - -
I/O (A20) 85 P208 B13 B15 E16 AD6
I/O 41 - B18 A22 E22 R5
I/O (A21) 86 P207 C13 C15 C16 AE5
I/O 42 - C17 C21 C23 R3
I/O 87 P206 A12 D15 A15 AF4
I/O 43 - - - A23 U1
I/O 88 P205 B12 B14 C15 AG1

R

EPIC EPIC
PAD NAME HQ240 BG352 BG432 BG560 PG559 PAD NAME HQ240 BG352 BG432 BG560 PG559
PAD # PAD #
I/O (A5) 89 P203 C12 A13 D15 AD8 I/O 132 - C5 B5 B4 AT6
I/O (A4) 90 P202 D12 C14 E15 AG3 I/O 133 - - - - -
I/O 91 - A11 B13 B14 AH4 I/O 136 - - - - -
I/O 92 - B11 D14 C14 AE7 GND - - - GND* GND* GND*
I/O 93 - B10 A12 D14 AH2 VCCIO - - - VCCIO* VCCIO* -
I/O 94 - C11 C13 A13 AF6 I/O 137 - - - - -
I/O 95 - - B12 E14 AJ1 I/O 138 - - - - -
I/O 96 - - D13 C13 AG5 I/O 139 - - D6 C5 AP8
I/O 97 - - - D13 AJ3 I/O 140 - - A4 E7 AR7
I/O 98 - - - B12 AK4 I/O 141 - - C5 A3 AY4
I/O 99 - - - - - I/O 142 P186 A3 - - BB2
I/O 100 - - - - - VCCINT - P185 D5 VCCINT* VCCINT* VCCINT*
VCCIO - - - VCCIO* VCCIO* - I/O, GCK7 (A1) 143 P184 C4 D5 A2 AV6
GND - - - GND* GND* GND* I/O (A0, /WS) 144 P183 B3 B3 D5 AT8
I/O 101 - - - C12 AG7 GND - P182 GND* GND* GND* GND*
I/O 102 - - - E13 AK2 O, TDO - P181 D4 C4 E6 BA3
I/O 103 0 D11 C12 A11 AJ5 VCCIO - P180 - VCCIO* VCCIO* VCCIO*
I/O 104 0 A9 B11 D12 AL3 CCLK - P179 C3 D4 C4 BA5
VCCIO - P201 VCCIO* VCCIO* VCCIO* VCCIO* I/O, GCK6 145 P178 E4 D3 D4 BB4
I/O 105 P200 B9 D12 B11 AM4 (DOUT)
I/O 106 P199 C10 C11 C11 AN1 I/O (D0, DIN) 146 P177 D3 C2 E4 AY6
VCCINT - P198 D10 VCCINT* VCCINT* VCCINT* I/O 147 P176 C2 D2 F5 BC3
I/O 107 - - - - AL5 I/O 148 P175 E3 E4 B3 AW7
I/O 108 P197 B8 B10 B10 AH8 I/O 149 - F4 D1 D3 BB6
GND - P196 GND* GND* GND* GND* I/O 150 - D2 E3 F4 AU9
VCCIO - - - VCCIO VCCIO* - I/O 151 - - - - -
I/O 109 - C9 C10 D11 AJ7 I/O 152 - - - - -
I/O 110 P195 D9 B9 C10 AP2 VCCIO - - - VCCIO* VCCIO* -
I/O 111 P194 A7 C9 A9 AN3 GND - - - GND* GND* GND*
I/O 112 P193 B7 D10 E11 AP4 I/O 153 - - - - -
I/O 113 - - - C9 AR1 I/O 154 - - - - -
I/O 114 - - - D10 AN5 I/O 155 - - E2 C1 AT10
I/O 115 - - - A8 AM6 I/O 156 - - F4 G5 AV8
I/O 116 - - - B8 AK8 I/O 157 - C1 E1 E3 AY8
GND - - - GND* GND* GND* I/O 158 - D1 F3 D2 BC7
I/O 117 - - - - - I/O 160 - - - - -
I/O 118 - - - - - I/O (/RCK, 161 P174 G4 F2 G4 AW9

RDY_/BUSY)
I/O 119 - - - E10 AL7
I/O (D1) 162 P173 F3 G4 F3 BA9
I/O 120 - - - C8 AT2
VCCIO - - - VCCIO* VCCIO* VCCIO*
I/O 121 - - A8 D9 AR3
I/O 122 - - B8 B7 AU1
I/O 163 - - - - -
I/O 123 P192 D8 D9 A6 AT4
I/O 164 - - - - -
I/O 124 P191 A6 B7 E9 AV2
I/O 165 - - F1 H5 AU11
I/O 125 P190 B6 C7 C7 AR5
I/O 166 - - G3 E2 AY10
I/O 126 P189 C7 D8 D8 AN7
I/O 167 - - G2 F2 BB8
I/O 168 - - H4 H4 AW11
I/O 169 P172 E2 H3 G3 BC9
I/O 127 - A4 A6 A5 AW3
I/O 170 P171 F2 H2 J5 AV12
I/O 128 - B5 B6 B5 AV4
I/O 171 P170 G3 - F1 AU13
I/O (A3) 129 P188 C6 D7 E8 BA1
I/O 172 P169 G2 - G1 AT14
I/O, 130 P187 D6 A5 C6 AU5
(CS1, A2) VCCIO - - - VCCIO* VCCIO* VCCIO*
I/O 131 - B4 C6 D7 AY2 GND - - - GND* GND* GND*

R

EPIC EPIC
PAD # PAD #
I/O 173 - - - J4 BA11 VCCIO - P150 - VCCIO* VCCIO* VCCIO*
I/O 174 - - - H3 AY12 I/O 217 P149 P1 T2 U4 AW23
I/O 175 - - - K5 BB10 I/O (D4) 218 P148 P2 T1 U5 AY24
I/O 176 - - - H2 AW13 I/O 219 - - - - -
I/O 177 P168 F1 H1 J3 BC11 I/O 220 - - - - -
I/O 178 P167 J4 J4 K4 AU15 I/O 221 - - - U1 BC23
I/O 179 - H3 J3 J2 BB14 I/O 222 - - - U2 BA27
I/O 180 - H2 J2 L5 AT16 I/O 223 P147 P3 U3 V2 AU23
VCCIO - - - VCCIO* VCCIO* - I/O 224 P146 R1 U4 V4 AV24
GND - P166 GND* GND* GND* GND* VCCIO - - - VCCIO* VCCIO* -
I/O 181 P165 G1 K4 K3 BA13 GND - - - GND* GND* GND*
I/O 182 - - - - AY14 I/O 225 P145 R2 U2 V5 AY26
VCCINT - P164 K4 VCCINT* VCCINT* VCCINT* I/O 226 P144 R3 U1 V3 BB24
I/O, FCLK4 183 P163 J3 K2 L4 BC15 I/O 227 - R4 V3 W1 AW25
I/O 184 P162 J2 K1 K2 AW15 I/O 228 - T1 V4 W3 BB26
VCCIO - P161 VCCIO* VCCIO* VCCIO* VCCIO* I/O 229 - - V2 W4 AT24
I/O 185 - - - M5 BA15 I/O 230 - - W2 W5 BA29
I/O 186 - - - L3 AU17 I/O 231 - - - - -
I/O 187 P160 K3 L3 L1 BB16 I/O 232 - - - - -
I/O (D2) 188 P159 J1 L2 M4 AY16 I/O 233 - - - - -
GND - - - GND* GND* GND* I/O 234 - - - - -
VCCIO - - - VCCIO* VCCIO* - VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* 6
I/O 189 - - M4 M3 AW17 GND - P143 GND* GND* GND* GND*
I/O 190 - - M3 N5 AV18 I/O 235 - T2 W3 Y2 AY28
I/O 191 - - - M2 BA17 I/O 236 - U2 W4 Y3 AU25
I/O 192 - - - N4 BC17 I/O 237 - - - Y4 AV26
I/O 193 - - - - - I/O 238 - - - AA1 BC27
I/O 194 - - - - - I/O 239 - - - - -
I/O 195 - L4 M2 N3 AU19 I/O 240 - - - - -
I/O 196 - K2 M1 N2 BB18 I/O 241 - - W1 Y5 AW27
I/O 197 - L3 N4 P5 AY18 I/O 242 - - Y1 AA3 BB28
I/O 198 - L2 N3 N1 AW19 I/O 243 - T3 Y3 AA4 BA31
GND - P158 GND* GND* GND* GND* I/O 244 - T4 Y4 AB2 AY30
I/O 199 - - - - - GND - - - GND* GND* GND*
I/O 200 - - - - - I/O (/CS0) 245 P142 V1 Y2 AB3 AW29
I/O 201 - - - - - I/O (D5) 246 P141 V2 AA2 AA5 BC29
I/O 202 - - - - - I/O 247 - - - AC1 AU27
I/O 203 - - N2 P4 AV20 I/O 248 - - - AB4 BA33
I/O 204 - - N1 P3 AT20 VCCIO - P140 VCCIO* VCCIO* VCCIO* VCCIO*
I/O 205 - L1 P4 P2 BB20 I/O 249 P139 U3 AA3 AC3 AY32
I/O 206 - M4 P3 R5 AY20 I/O, FCLK3 250 P138 U4 AB1 AB5 AW31
I/O 207 P157 M3 P2 R4 BC21 VCCINT - P137 W2 VCCINT* VCCINT* VCCINT*
I/O 208 P156 M2 R3 R3 BA21 I/O 251 - - - - BB30
GND - - - GND* GND* GND* I/O 252 P136 V3 AB3 AC4 BA35
VCCIO - - - VCCIO* VCCIO* - GND - P135 GND* GND* GND* GND*
I/O 209 P155 M1 R4 R1 AW21 VCCIO - - - VCCIO* VCCIO* -
I/O 210 - - - T4 AU21 I/O 253 - V4 AB4 AD3 AT28
I/O 211 - - - T5 BB22 I/O 254 - Y1 AC2 AE1 AU29
I/O 212 - - - - AY22 I/O 255 P134 Y2 AD1 AC5 BC33
I/O 213 - - - - - I/O 256 P133 W3 AC3 AE3 AY34
I/O 214 - - - - - I/O 257 - - - AD4 BB34
VCCINT - P154 N3 VCCINT* VCCINT* VCCINT* I/O 258 - - - AF1 AW33
I/O (/RS) 215 P153 N4 R1 T2 AV22 I/O 259 - - - AF2 AU31
I/O (D3) 216 P152 N2 T3 U3 BA23 I/O 260 - - - AD5 AV32
GND - P151 - GND* GND* GND* GND - - - GND* GND* GND*

R

EPIC EPIC
PAD # PAD #
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O 305 - AC7 AL5 AL6 AR39
I/O 261 - - - - - I/O 306 - AD6 AK6 AJ8 AN37
I/O 263 - - - AG2 AT30 GND - - GND* GND* GND* GND*
I/O 264 - - - AE4 AU33 I/O 307 P112 AE5 AH8 AM6 AT40
I/O 265 - - AC4 AF3 AW35 I/O 308 P111 AE6 AJ7 AN6 AP40
I/O 266 - - AD2 AH1 BC35 I/O 309 P110 AD7 AL6 AK8 AR43
I/O 267 P132 W4 AE2 AE5 AY36 I/O 310 P109 AE7 AK7 AL7 AN39
I/O 268 P131 AA1 AD3 AG3 BB36 I/O 311 - - AH9 AJ9 AP42
I/O 269 P130 AA2 AD4 AF4 AV36 I/O 312 - - AJ8 AN7 AM38
I/O (D6) 270 P129 Y3 AF1 AJ1 AU35 I/O 313 - - - AL8 AN43
GND - - - GND* GND* GND* I/O 314 - - - AK9 AL37
VCCIO - - - VCCIO* VCCIO* VCCIO* I/O 315 - - - - -
I/O 271 - AC1 AF2 AJ2 AT34 I/O 316 - - - - -
I/O 272 - AB2 AE3 AF5 AW37 VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 273 - - - - - GND - - - GND* GND* GND*
I/O 274 - - - - - I/O 317 - - - AM8 AK36
I/O 275 - - - - - I/O 318 - - - AJ10 AR41
I/O 276 - - - - - I/O 319 - - - AL9 AL41
I/O 277 - - AE4 AH3 BC37 I/O 320 - - - AM9 AN41
I/O 278 - - AG1 AG4 AY38 I/O 321 P108 AF6 AK8 AK10 AK42
I/O 279 P128 AA3 AG2 AK2 BB38 I/O 322 P107 AC9 AJ9 AN9 AM40
I/O 280 P127 AA4 AF3 AJ3 BA41 I/O 323 - AD8 AH10 AJ11 AJ43
GND - - - GND* GND* GND* I/O 324 - AE8 AL8 AL10 AJ37
VCCIO - - - VCCIO* VCCIO* - VCCIO - - - VCCIO* VCCIO* -
I/O 282 - - - - - I/O 325 P105 AF7 AK9 AM10 AL39
I/O 283 - AC2 AF4 AG5 AY40 I/O 326 - - - - AH36
I/O 284 - AB3 AH1 AL1 BB40 VCCINT - P104 AC10 VCCINT* VCCINT* VCCINT*
I/O 285 P126 AD1 AH2 AH4 AT36 I/O 327 P103 AD9 AK10 AL11 AH42
I/O 286 P125 AB4 AG3 AK3 BA39 I/O 328 P102 AE9 AL10 AJ12 AJ39
I/O, GCK5 287 P124 AC3 AG4 AJ4 AV38 VCCIO - P101 VCCIO* VCCIO* VCCIO* VCCIO*
I/O (D7) 288 P123 AD2 AJ2 AH5 BC41 I/O 329 P100 AD10 AJ11 AN11 AK40
/PROGRAM - P122 AC4 AH3 AM1 BB42 I/O 330 P99 AF9 AH12 AK12 AG41
VCCIO - P121 VCCIO* VCCIO* VCCIO* VCCIO* I/O 331 - - - AL12 AJ41
DONE - P120 AD3 AH4 AJ5 AY42 I/O 332 - - - AM12 AH40
GND - P119 GND* GND* GND* GND* GND - - - GND* GND* GND*
I/O, GCK4 289 P118 AC5 AJ4 AL4 AW41 VCCIO - - - VCCIO* VCCIO* -
I/O 290 P117 AD4 AK3 AJ6 AV40 I/O 333 - - - - -
VCCINT - P116 AE3 VCCINT* VCCINT* VCCINT* I/O 334 - - - - -
I/O 291 - - - - AV42 I/O 335 - - - AJ13 AG37
I/O 292 P115 AD5 AK4 AN3 AR37 I/O 336 - - - AK13 AG43
I/O 293 - - AJ5 AK6 AP36 I/O 337 - - AK11 AL13 AG39
I/O 294 - - AH6 AL5 AT38 I/O 338 - - AJ12 AM13 AF38
I/O 295 - - - - - I/O 339 - AC11 AK12 AN13 AF42
I/O 296 - - - - - I/O 340 - AE10 AL12 AJ14 AD42
VCCIO - - - VCCIO* VCCIO* - I/O 341 - AD11 AH13 AK14 AF40
GND - - - GND* GND* GND* I/O 342 - AE11 AJ13 AL14 AE37
I/O 299 - - - - - I/O 343 - AF11 AK13 AM14 AE39
I/O 300 - - - - - I/O 344 - AC12 AL13 AN15 AD40
I/O 301 - AC6 AL4 AJ7 AU39 I/O 345 P97 AD12 AK14 AJ15 AC43
I/O 302 - AE4 AK5 AM4 AU43 I/O 346 P96 AE12 AH14 AK15 AD38
I/O 303 P114 AF3 AJ6 AM5 BA43 I/O 347 P95 AF12 AJ14 AL15 AC41
I/O 304 P113 AF4 AH7 AK7 AT42 I/O 348 P94 AD13 AK15 AM16 AD36

R

EPIC EPIC
PAD # PAD #
I/O 349 - - - - - I/O 393 P79 AD17 AJ21 AL23 N39
I/O 350 - - - - - I/O 394 P78 AC17 AL22 AJ22 T36
I/O 351 - - - - - VCCINT - P77 AE19 VCCINT* VCCINT* VCCINT*
I/O 352 - - - - - I/O 395 - - - - P42
GND - - - GND* GND* GND* I/O 396 P76 AD18 AJ22 AK23 R37
VCCIO - - - VCCIO* VCCIO* - GND - P75 GND* GND* GND* GND*
I/O 353 - - - - - VCCIO - - - VCCIO* VCCIO* -
I/O 354 - - - - - I/O 397 - AC18 AK23 AL24 L41
I/O 355 - - - AL16 AC39 I/O 398 - AF20 AJ23 AN26 L43
I/O 356 - - - AJ16 AC37 I/O 399 P74 AE20 AH22 AJ23 K40
I/O 357 - - AJ15 AK16 AB40 I/O 400 P73 AD19 AL24 AL25 K42
I/O 358 - - AL15 AN17 AB42 I/O 401 - - - AK24 L39
VCCINT - P93 AC13 VCCINT* VCCINT* VCCINT* I/O 402 - - - AM26 J43
I/O 359 - - - - AB38 I/O 403 - - - AM27 M38
I/O 360 P92 AE13 AL16 AL17 AA41 I/O 404 - - - AJ24 P36
GND - P91 GND* GND* GND* GND* GND - - - GND* GND* GND*
VCCIO - P90 VCCIO* VCCIO* VCCIO* VCCIO* VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O (/INIT) 361 P89 AF14 AK16 AJ17 AA39 I/O 405 - - - - -
I/O 362 P88 AE14 AJ16 AK17 AA37 I/O 406 - - - - -
I/O 363 - - AL17 AM18 Y40 I/O 407 - - - AL26 N37
I/O 364 - - AK17 AL18 Y38 I/O 408 - - - AK25 H42
I/O 365 - - - AK18 AA43 I/O 409 - - AK24 AN28 J41 6
I/O 366 - - - AJ18 W39 I/O 410 - - AH23 AN29 G43
I/O 367 - - - - - I/O 411 P72 AC19 AJ24 AJ25 H40
I/O 368 - - - - - I/O 412 P71 AF21 AK25 AL27 F42
VCCIO - - - VCCIO* VCCIO* - I/O 413 P70 AE21 AJ25 AK26 J39
GND - - - GND* GND* GND* I/O 414 P69 AD20 AH24 AM29 L37
I/O 371 - - - - - I/O 415 - AF23 AL26 AM30 E41
I/O 372 - - - - - I/O 416 - AE22 AK26 AJ26 F40
I/O 373 P87 AD14 AJ17 AN19 V40 I/O 417 - AD21 AH25 AL28 C43
I/O 374 P86 AF15 AH17 AL19 Y36 I/O 418 - AC21 AL27 AK27 G39
I/O 375 P85 AE15 AK18 AK19 U41 I/O 419 - - AJ26 AL29 D42
I/O 376 P84 AD15 AL19 AM20 Y42 I/O 420 - - AK27 AN31 H38
I/O 377 - AC15 AJ18 AJ19 T40 I/O 421 - - - - -
I/O 378 - AF16 AK19 AL20 W37 I/O 422 - - - - -
I/O 379 - AE16 AH18 AK20 V38 GND - - - GND* GND* GND*
I/O 380 - AE17 AL20 AN21 U39 VCCIO - - - VCCIO* VCCIO* -
I/O 381 - AD16 AJ19 AL21 V42 I/O 425 - - - - -
I/O 382 - AC16 AK20 AJ20 R41 I/O 426 - - - - -
I/O 383 - - AH19 AM22 U43 I/O (/LDC) 427 P68 AE23 AH26 AJ27 K36
I/O 384 - - AJ20 AK21 P40 I/O 428 P67 AD22 AL28 AM31 J37
I/O 385 - - - AN23 T42 I/O 429 P66 AF24 AJ27 AK28 B42
I/O 386 - - - AL22 U37 I/O 430 P65 AC22 AK28 AL30 D40
I/O 387 - - - - - I/O (HDC) 431 P64 AD23 AH27 AK29 C41
I/O 388 - - - - - I/O, GCK3 432 P63 AE24 AK29 AJ28 F38
VCCIO - - - VCCIO* VCCIO* VCCIO* I (M2) - P62 AC23 AJ28 AN32 H36
GND - - - GND* GND* GND* VCCIO - P61 VCCIO* VCCIO* VCCIO* VCCIO*
I/O 389 - - - AJ21 R39 I (M0) - P60 AD24 AH28 AJ29 C39
I/O 390 - - - AM23 N41 GND - P59 GND* GND* GND* GND*
I/O 391 P82 AF18 AK21 AK22 R43 O (M1) - P58 AB23 AH29 AK30 D38
I/O 392 P81 AE18 AH20 AM24 M40 I/O, GCK2 433 P57 AC24 AJ30 AH29 E37
VCCIO - P80 VCCIO* VCCIO* VCCIO* - I/O 434 P56 AD25 AH30 AJ30 B40

R

EPIC EPIC
PAD # PAD #
VCCINT - P55 AB24 VCCINT* VCCINT* VCCINT* I/O 478 - T24 Y30 AA30 B28
I/O 435 - - - - H34 I/O 479 - - Y31 AA31 G25
I/O 436 P54 AA23 AH31 AH30 G35 I/O 480 - - W28 AA32 A27
I/O 437 - AC25 AG29 AL33 F36 I/O 481 - - - - -
I/O 438 - AD26 AG30 AG29 D36 I/O 482 - - - - -
I/O 439 - - - - - I/O 483 - - - Y29 D26
I/O 440 - - - - - I/O 484 - - - AA33 B26
VCCIO - - - VCCIO* VCCIO* - I/O 485 - T25 W29 Y30 E25
GND - - - GND* GND* GND* I/O 486 - T26 W30 Y31 F24
I/O 441 P53 AC26 AF28 AJ31 E35 GND - P37 GND* GND* GND* GND*
I/O 442 P52 Y23 AG31 AK32 A41 VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 443 - - AF29 AG30 G33 I/O 487 - - - - -
I/O 444 - - AF30 AH31 B38 I/O 488 - - - - -
I/O 445 - - - - - I/O 489 - - - - -
I/O 446 - - - - - I/O 490 - - - - -
I/O 447 - - - - - I/O 491 - - W31 Y32 H24
I/O 448 - - - - - I/O 492 - - V28 W29 B24
I/O 449 - AA24 AE28 AF29 D34 I/O 493 - R23 V29 W30 D24
I/O 450 - AB25 AF31 AJ32 E33 I/O 494 - R24 V30 W31 A23
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* I/O 495 P36 R25 U29 W33 C23
GND - - GND* GND* GND* GND* I/O 496 P35 R26 U28 V30 E23
I/O 451 P51 AA25 AE29 AH32 F32 GND - - - GND* GND* GND*
I/O 452 P50 Y24 AE30 AF30 G31 VCCIO - - - VCCIO* VCCIO* -
I/O 453 P49 Y25 AD28 AG31 A37 I/O 497 P34 P24 U30 V29 G23
I/O 454 P48 AA26 AD29 AE29 H30 I/O 498 P33 P23 U31 V31 B22
I/O 455 - - AD30 AH33 B36 I/O 499 - - - V32 D22
I/O 456 - - AD31 AG33 C33 I/O 500 - - - U33 A21
I/O 457 - - - AE30 C35 I/O 501 - - - - -
I/O 458 - - - AF31 D32 I/O 502 - - - - -
I/O 459 - - - - - I/O 503 P32 P25 T29 U30 F22
I/O 460 - - - - - I/O 504 P31 N26 T30 U29 C21
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* VCCIO - P30 VCCIO* VCCIO* VCCIO* VCCIO*
GND - - - GND* GND* GND* GND - P29 GND* GND* GND* GND*
I/O 461 - - - AD29 E31 I/O 505 P28 N25 T31 U31 E21
I/O 462 - - - AF32 G29 I/O 506 - - - - D20
I/O 463 - - - AE31 A35 VCCINT - P27 N24 VCCINT* VCCINT* VCCINT*
I/O 464 - - - AD30 H28 I/O 507 - - - - -
I/O 465 P47 V23 AC28 AE32 B34 I/O 508 - - - - -
I/O 466 P46 W24 AC29 AC29 C31 I/O 509 - - - T32 C17
I/O 467 - W25 AC30 AE33 A33 I/O 510 - - - T30 G21
I/O 468 - Y26 AB28 AD31 D30 I/O 511 P26 M26 R28 T29 F20
VCCIO - - - VCCIO* VCCIO* - I/O 512 P25 M25 R30 T31 D18
GND - P45 GND* GND* GND* GND* VCCIO - - - VCCIO* VCCIO* -
I/O, FCLK2 469 P44 U23 AB29 AC30 E29 GND - - - GND* GND* GND*
I/O 470 - - - - C29 I/O 513 P24 M24 R31 R33 E19
VCCINT - P43 V24 VCCINT* VCCINT* VCCINT* I/O 514 P23 M23 P29 R31 B20
I/O 471 P42 V25 AB31 AB29 B30 I/O 515 - L26 P28 R30 H20
I/O 472 P41 U24 AA29 AC31 G27 I/O 516 - L25 P30 R29 B18
VCCIO - P40 VCCIO* VCCIO* VCCIO* VCCIO* I/O 517 - - N30 P32 C15
I/O 473 - - - AC33 D28 I/O 518 - - N29 P31 D16
I/O 474 - - - AB30 A29 I/O 519 - - - - -
I/O 475 P39 V26 AA30 AB31 E27 I/O 520 - - - - -
I/O 476 P38 T23 Y28 AA29 F26 I/O 521 - - - - -
GND - - - GND* GND* GND* I/O 522 - - - - -
VCCIO - - - VCCIO* VCCIO* - VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO*
I/O 477 - U25 Y29 AB32 C27 GND - P22 GND* GND* GND* GND*

R

EPIC EPIC
PAD # PAD #
I/O 523 - K25 N28 N33 G19 I/O 568 - - - - -
I/O 524 - L24 N31 P30 A17 GND - - - GND* GND* GND*
VCCINT - - - - - - VCCIO - - - VCCIO* VCCIO* -
I/O 525 - L23 M31 P29 F18 I/O 569 - - - - -
I/O 526 - J26 M29 M32 E17 I/O 570 - - - - -
I/O 527 - - - - - I/O (TCK) 571 P7 E24 D31 C33 D4
I/O 528 - - - - - I/O (TDI) 572 P6 C26 D30 F30 H8
I/O 529 - - - N31 B16 I/O 573 P5 E23 E29 D31 A3
I/O 530 - - - N30 C13 I/O 574 - - - - F6
I/O 531 - - M28 L33 A15 VCCINT - P4 D24 VCCINT* VCCINT* VCCINT*
I/O 532 - - M30 N29 D14 I/O (A17) 575 P3 C25 C30 F29 C3
VCCIO - - - VCCIO* VCCIO* - I/O, GCK1 (A16) 576 P2 D23 D29 B33 C1
GND - - - GND* GND* GND* GND - P1 GND* GND* GND* GND*
I/O 533 P21 J25 L30 M31 E15
I/O 534 P20 K24 L29 L32 G17 VCCIO - - A10 A1 A4 A13
I/O 535 - - - M30 B14 VCCIO - - A17 A11 A10 A31
I/O 536 - - - L31 C11 VCCIO - - AC14 A21 A16 A43
VCCIO - P19 VCCIO* VCCIO* VCCIO* VCCIO* VCCIO - - AC20 A31 A22 B2
I/O 537 P18 K23 K31 M29 D12 VCCIO - - AC8 D11 A26 C7
I/O (TMS) 538 P17 H25 K30 J33 A11 VCCIO - - AF10 D21 A30 C19
VCCINT - P16 J24 VCCINT* VCCINT* VCCINT* VCCIO - - AF17 L1 B2 C25
I/O 539 - - - - E13 VCCIO - - D7 L4 B13 C37 6
I/O, FCLK1 540 P15 J23 K28 L30 C9 VCCIO - - D13 L28 B19 F14
GND - P14 GND* GND* GND* GND* VCCIO - - D19 L31 B32 F30
VCCIO - - - VCCIO* VCCIO* - VCCIO - - G23 AA1 C3 G3
I/O 541 - G26 J30 K31 H16 VCCIO - - H4 AA4 C31 G7
I/O 542 - G25 H31 L29 B10 VCCIO - - K1 AA28 C32 G37
I/O 543 P13 H24 J29 H33 G15 VCCIO - - K26 AA31 D1 G41
I/O 544 P12 H23 J28 K30 A9 VCCIO - - N23 AH11 D33 N1
I/O 545 - - - J31 D10 VCCIO - - P4 AH21 E5 N43
I/O 546 - - - H32 B8 VCCIO - - U1 AL1 H1 P6
I/O 547 - - - K29 E11 VCCIO - - U26 AL11 K33 P38
I/O 548 - - - H31 G13 VCCIO - - W23 AL21 M1 W3
GND - - - GND* GND* GND* VCCIO - - Y4 AL31 N32 W41
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* VCCIO - - B2 C3 R2 AE3
I/O 549 P11 F26 - J30 F12 VCCIO - - B25 C29 T33 AE41
I/O 550 P10 F25 - G32 H14 VCCIO - - AE2 AJ3 V1 AK6
I/O 551 P9 G24 H30 F33 G11 VCCIO - - AE25 AJ29 W32 AK38
I/O 552 P8 D26 G30 J29 A7 VCCIO - - - - AA2 AL1
I/O 553 - - H29 G31 E9 VCCIO - - - - AB33 AL43
I/O 554 - - H28 H30 B6 VCCIO - - - - AD1 AU3
I/O 555 - - F31 E33 D8 VCCIO - - - - AF33 AU7
I/O 556 - - F30 E32 F8 VCCIO - - - - AK1 AU37
I/O 557 - - - - - VCCIO - - - - AK4 AU41
I/O 558 - - - - - VCCIO - - - - AK33 AV14
GND - - GND* GND* GND* GND* VCCIO - - - - AL2 AV30
VCCIO - - VCCIO* VCCIO* VCCIO* VCCIO* VCCIO - - - - AL3 BA7
I/O 559 - E25 G29 H29 G9 VCCIO - - - - AL31 BA19
I/O 560 - F24 G28 F31 H10 VCCIO - - - - AM2 BA25
I/O 561 - - - - - VCCIO - - - - AM15 BA37
I/O 562 - - - - - VCCIO - - - - AM21 BC1
I/O 563 - F23 E31 G30 B4 VCCIO - - - - AM32 BC13
I/O 564 - D25 E30 D32 E7 VCCIO - - - - AN4 BC31
I/O 565 - - F29 E31 C5 VCCIO - - - - AN8 BC43
I/O 566 - - F28 G29 D6 VCCIO - - - - AN12 -
I/O 567 - - - - - VCCIO - - - - AN18 -

R

EPIC EPIC
PAD # PAD #
VCCIO - - - - AN24 - GND - - E1 T28 P1 AB36
VCCIO - - - - AN30 - GND - - E26 V1 P33 AE1
GND - - H1 V31 R32 AE43
VCCINT - - - A10 E12 H12 GND - - H26 AC1 T1 AH6
VCCINT - - - AB2 AD2 H18 GND - - N1 AC31 V33 AH38
VCCINT - - - AB30 AD32 H26 GND - - P26 AE1 W2 AM2
VCCINT - - - AG28 AK31 H32 GND - - W1 AE31 Y1 AM42
VCCINT - - - AH15 AM17 M8 GND - - W26 AH16 Y33 AP6
VCCINT - - - AH5 AK5 M36 GND - - - AJ1 AB1 AP38
VCCINT - - - AJ10 AK11 V8 GND - - - AJ31 AC32 AT22
VCCINT - - - AK22 AN25 V36 GND - - - AK1 AD33 AV10
VCCINT - - - B23 C24 AF8 GND - - - AK2 AE2 AV16
VCCINT - - - B4 D6 AF36 GND - - - AK30 AG1 AV28
VCCINT - - - C16 C17 AM8 GND - - - AK31 AG32 AV34
VCCINT - - - E28 E30 AM36 GND - - - AL2 AH2 AW1
VCCINT - - - K29 K32 AT12 GND - - - AL3 AJ33 AW5
VCCINT - - - K3 J1 AT18 GND - - - AL7 AL32 AW39
VCCINT - - - R2 T3 AT26 GND - - - AL9 AM3 AW43
VCCINT - - - R29 U32 AT32 GND - - - AL14 AM11 BB12
GND - - - AL18 AM19 BB32
GND - - A1 A2 A7 A5 GND - - - AL23 AM25 BC5
GND - - A22 A14 A20 B12 GND - - - - AN2 -
GND - - A25 A18 A24 B32 GND - - - - AN5 -
GND - - A26 A23 A29 E1 GND - - - - AN10 -
GND - - A5 A25 A32 E5 GND - - - - AN14 -
GND - - A8 A29 B1 E39 GND - - - - AN16 -
GND - - AB1 A30 B6 E43 GND - - - - AN20 -
GND - - AB26 B1 B9 F10 GND - - - - AN22 -
GND - - AE1 B2 B15 F16 GND - - - - AN27 -
GND - - AE26 B30 B23 F28
GND - - AF1 B31 B27 F34 NC - P204 C8 C8 A1 -
GND - - AF13 C1 B31 H22 NC - P219 - - A33 -
GND - - AF19 C31 C2 K6 NC - - - - AC2 -
GND - - AF2 D16 E1 K38 NC - - - - AN1 -
GND - - AF22 G1 F32 M2 NC - - - - AN33 -
GND - - AF25 G31 G2 M42 12/21/98
GND - - AF26 J1 G33 T6
*Note: Pads labelled GND*, VCCINT*, or VCCIO* are inter-
GND - - AF5 J31 J32 T38
nally bonded to the corresponding power plane within the
GND - - AF8 P1 K1 W1
GND - - B1 P31 L2 W43
associated package. They have no direct connection to any
GND - - B26 T4 M33 AB8
specific package pin.


XC40200XV Pinout Table
EPIC
Pad Name BG432 BG560 PG559
EPIC Pad #
Pad # I/O 46 B22 B24 T8
VCCIO - VCCIO* VCCIO* VCCIO* VCCIO - VCCIO* VCCIO* VCCIO*
I/O, GCK8 (A15 1 D28 E29 E3 I/O 47 A22 E22 R5
I/O (A)14) 2 C28 D30 F4 I/O 48 C21 C23 R3
I/O 3 B29 E28 J7 I/O 49 - A23 U1
I/O 4 D27 D29 D2 I/O 50 - D22 T4
I/O 5 B28 C30 K8 I/O 51 - - -
I/O 6 C27 D28 H6 I/O 52 - - -
I/O 7 - - - I/O 53 - - -
I/O 8 - - - I/O 54 - - -
I/O 9 - - - GND - GND* GND* GND*
I/O 10 - - - VCCIO - VCCIO* VCCIO* -
VCCIO - VCCIO* VCCIO* - I/O 55 - - -
GND - GND* GND* GND* I/O 56 - - -
I/O 11 - - - I/O 57 - C22 U5
I/O 12 - - - I/O 58 - E21 V2
I/O 13 - - - I/O 59 D20 B22 U7
I/O 14 - - - I/O 60 B21 D21 U3
I/O 15 D26 A31 G5 I/O 61 C20 C21 Y2
I/O 16 A28 E27 F2 I/O 62 B20 B21 V4
I/O 17 B27 C29 H4 I/O 63 A20 E20 V6
I/O 18 C26 B30 G1 I/O 64 D19 A21 W5 6
I/O 19 D25 D27 J5 GND - GND* GND* GND*
I/O 20 A27 C28 L7 VCCIO - VCCIO* VCCIO* VCCIO*
VCCIO - VCCIO* VCCIO* VCCIO* I/O (A11) 65 C19 D20 W7
GND - GND* GND* GND* I/O (A10) 66 B19 C20 Y4
I/O (A13) 21 B26 E26 J3 I/O 67 A19 B20 AA1
I/O (A12) 22 D24 B29 K4 I/O 68 B18 E19 Y6
I/O 23 C25 B28 H2 I/O (A18) 69 D18 D19 AB2
I/O 24 A26 D26 L5 I/O (A19) 70 C18 C19 Y8
I/O 25 B25 C27 J1 I/O 71 - - -
I/O 26 D23 E25 M6 I/O 72 - - -
I/O 27 - A28 K2 I/O 73 - - -
I/O 28 - A27 N7 I/O 74 - - -
VCCIO - VCCIO* VCCIO* VCCIO* I/O 75 - - -
I/O 31 - - - I/O 77 - - -
I/O 32 - - - I/O 78 - - -
I/O 33 - - - I/O 79 - A19 AA3
I/O 34 - - - I/O 80 - D18 AA5
I/O 35 - D25 P8 I/O 81 B17 E18 AC1
I/O 36 - C26 L3 I/O 82 C17 C18 AA7
I/O 37 - E24 L1 I/O (A9) 83 A17 B18 AB4
I/O 38 - B26 M4 I/O (A8) 84 D17 A17 AB6
I/O 39 C24 C25 P2 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 40 B24 D24 N5 GND - GND* GND* GND*
I/O 41 C23 B25 R1 I/O (A7) 85 A16 D17 AC3
I/O 42 D22 E23 N3 I/O (A6) 86 B16 E17 AC5
VCCIO - VCCIO* VCCIO* - VCCINT - C16 C17 VCCINT*
GND - GND* GND* GND* I/O 87 - - AD2
I/O 43 A24 A25 R7 I/O 88 A15 B17 AC7
I/O 44 - - P4 I/O 89 - B16 AF2
VCCINT - B23 C24 VCCINT* I/O 90 - D16 AD4
I/O 45 C22 D23 T2 I/O 91 - - -

R

EPIC EPIC
Pad Name BG432 BG560 PG559 Pad Name BG432 BG560 PG559
Pad # Pad #
I/O 92 - - - I/O 137 - - -
I/O 93 - - - I/O 138 - - -
VCCIO - VCCIO* VCCIO* - VCCIO - VCCIO* VCCIO* VCCIO*
I/O 95 - - - I/O 140 - - -
I/O 96 - - - I/O 141 - E10 AL7
I/O 97 - - - I/O 142 - C8 AT2
I/O 98 - - - I/O 143 A8 D9 AR3
I/O (A20) 99 B15 E16 AD6 I/O 144 B8 B7 AU1
I/O (A21) 100 C15 C16 AE5 I/O 145 D9 A6 AT4
I/O 101 D15 A15 AF4 I/O 146 B7 E9 AV2
I/O 102 B14 C15 AG1 I/O 147 C7 C7 AR5
I/O (A5) 103 A13 D15 AD8 I/O 148 D8 D8 AN7
I/O (A4) 104 C14 E15 AG3 GND - GND* GND* GND*
VCCIO - VCCIO* VCCIO* VCCIO* VCCIO - VCCIO* VCCIO* VCCIO*
GND - GND* GND* GND* I/O 149 A6 A5 AW3
I/O 105 B13 B14 AH4 I/O 150 B6 B5 AV4
I/O 106 D14 C14 AE7 I/O (A3) 151 D7 E8 BA1
I/O 107 A12 D14 AH2 I/O, (CS1, A2) 152 A5 C6 AU5
I/O 108 C13 A13 AF6 I/O 153 C6 D7 AY2
I/O 109 B12 E14 AJ1 I/O 154 B5 B4 AT6
I/O 110 D13 C13 AG5 I/O 155 - - -
I/O 111 - D13 AJ3 I/O 156 - - -
I/O 112 - B12 AK4 I/O 157 - - -
I/O 113 - - - I/O 158 - - -
VCCIO - VCCIO* VCCIO* - VCCIO - VCCIO* VCCIO* -
I/O 115 - - - I/O 160 - - -
I/O 116 - - - I/O 161 - - -
I/O 117 - - - I/O 162 - - -
I/O 118 - - - I/O 163 D6 C5 AP8
I/O 119 - C12 AG7 I/O 164 A4 E7 AR7
I/O 120 - E13 AK2 I/O 165 C5 A3 AY4
I/O 121 C12 A11 AJ5 I/O 166 - - BB2
I/O 122 B11 D12 AL3 I/O, GCK7 (A1) 167 D5 A2 AV6
VCCIO - VCCIO* VCCIO* VCCIO* I/O (A0, /WS) 168 B3 D5 AT8
I/O 123 D12 B11 AM4 GND - GND* GND* GND*
I/O 124 C11 C11 AN1 O, TDO - C4 E6 BA3
VCCINT - A10 E12 VCCINT* VCCIO - VCCIO* VCCIO* VCCIO*
I/O 125 - - AL5 CCLK - D4 C4 BA5
VCCINT - - - - I/O, GCK6 (DOUT) 169 D3 D4 BB4
I/O 126 B10 B10 AH8 I/O (D0, DIN) 170 C2 E4 AY6
GND - GND* GND* GND* I/O 171 D2 F5 BC3
VCCIO - VCCIO VCCIO* - I/O 172 E4 B3 AW7
I/O 127 C10 D11 AJ7 I/O 173 D1 D3 BB6
I/O 128 B9 C10 AP2 I/O 174 E3 F4 AU9
I/O 129 C9 A9 AN3 I/O 176 - - -
I/O 130 D10 E11 AP4 I/O 177 - - -
I/O 131 - C9 AR1 I/O 178 - - -
I/O 132 - D10 AN5 VCCIO - VCCIO* VCCIO* -
I/O 133 - A8 AM6 GND - GND* GND* GND*
I/O 134 - B8 AK8 I/O 179 - - -
I/O 135 - - - I/O 180 - - -
I/O 136 - - - I/O 181 E2 C1 AT10

R

EPIC EPIC
Pad # Pad #
I/O 182 F4 G5 AV8 I/O 228 - - -
I/O 183 E1 E3 AY8 I/O 229 M2 N3 AU19
I/O 184 F3 D2 BC7 I/O 230 M1 N2 BB18
I/O 185 - - - I/O 231 N4 P5 AY18
I/O 186 - - - I/O 232 N3 N1 AW19
I/O (/RCK,RDY_/BUSY) 187 F2 G4 AW9 GND - GND* GND* GND*
I/O (D1) 188 G4 F3 BA9 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 189 - - - I/O 235 - - -
I/O 190 - - - I/O 236 - - -
I/O 191 - H5 AU11 I/O 237 N2 P4 AV20
I/O 192 - E2 AY10 I/O 238 N1 P3 AT20
I/O 193 F1 F2 BB8 I/O 239 P4 P2 BB20
I/O 194 G3 H4 AW11 I/O 240 P3 R5 AY20
I/O 195 G2 G3 BC9 I/O 241 P2 R4 BC21
I/O 196 H4 J5 AV12 I/O 242 R3 R3 BA21
I/O 197 H3 F1 AU13 GND - GND* GND* GND*
I/O 198 H2 G1 AT14 VCCIO - VCCIO* VCCIO* -
VCCIO - VCCIO* VCCIO* VCCIO* I/O 243 R4 R1 AW21
GND - GND* GND* GND* I/O 244 - T4 AU21
I/O 199 - - - I/O 245 - T5 BB22 6
I/O 200 - - - I/O 246 - - AY22
I/O 201 - - - I/O 247 - - -
I/O 202 - - - I/O 248 - - -
I/O 203 - J4 BA11 I/O 249 - - -
I/O 204 - H3 AY12 I/O 250 - - -
I/O 205 - K5 BB10 VCCINT - R2 T3 VCCINT*
I/O 206 - H2 AW13 I/O (/RS) 251 R1 T2 AV22
I/O 207 H1 J3 BC11 I/O (D3) 252 T3 U3 BA23
I/O 208 J4 K4 AU15 GND - GND* GND* GND*
I/O 209 J3 J2 BB14 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 210 J2 L5 AT16 I/O 253 T2 U4 AW23
VCCIO - VCCIO* VCCIO* - I/O (D4) 254 T1 U5 AY24
I/O 211 K4 K3 BA13 I/O 256 - - -
I/O 212 - - AY14 I/O 257 - - -
VCCINT - K3 J1 VCCINT* I/O 258 - - -
I/O, FCLK4 213 K2 L4 BC15 I/O 259 - U1 BC23
I/O 214 K1 K2 AW15 I/O 260 - U2 BA27
VCCIO - VCCIO* VCCIO* VCCIO* I/O 261 U3 V2 AU23
I/O 215 - - - I/O 262 U4 V4 AV24
I/O 218 - - - I/O 263 U2 V5 AY26
I/O 219 - M5 BA15 I/O 264 U1 V3 BB24
I/O 220 - L3 AU17 I/O 265 V3 W1 AW25
I/O 221 L3 L1 BB16 I/O 266 V4 W3 BB26
I/O (D2) 222 L2 M4 AY16 I/O 267 V2 W4 AT24
GND - GND* GND* GND* I/O 268 W2 W5 BA29
I/O 223 M4 M3 AW17 I/O 270 - - -
I/O 224 M3 N5 AV18 I/O 271 - - -
I/O 225 - M2 BA17 I/O 272 - - -
I/O 226 - N4 BC17 VCCIO - VCCIO* VCCIO* VCCIO*

R

EPIC EPIC
Pad # Pad #
I/O 273 W3 Y2 AY28 I/O 319 - - -
I/O 274 W4 Y3 AU25 I/O 320 - - -
I/O 275 - Y4 AV26 I/O 321 - - -
I/O 276 - AA1 BC27 I/O 322 - - -
I/O 277 - - - I/O 323 AE4 AH3 BC37
I/O 278 - - - I/O 324 AG1 AG4 AY38
I/O 279 W1 Y5 AW27 I/O 325 AG2 AK2 BB38
I/O 280 Y1 AA3 BB28 I/O 326 AF3 AJ3 BA41
I/O 281 Y3 AA4 BA31 GND - GND* GND* GND*
I/O 282 Y4 AB2 AY30 VCCIO - VCCIO* VCCIO* -
I/O (/CS0) 283 Y2 AB3 AW29 I/O 329 - - -
I/O (D5) 284 AA2 AA5 BC29 I/O 330 - - -
I/O 285 - AC1 AU27 I/O 331 AF4 AG5 AY40
I/O 286 - AB4 BA33 I/O 332 AH1 AL1 BB40
I/O 287 - - - I/O 333 AH2 AH4 AT36
I/O 288 - - - I/O 334 AG3 AK3 BA39
I/O 289 - - - I/O, GCK5 335 AG4 AJ4 AV38
I/O 290 - - - I/O (D7) 336 AJ2 AH5 BC41
VCCIO - VCCIO* VCCIO* VCCIO* /PROGRAM - AH3 AM1 BB42
I/O 291 AA3 AC3 AY32 VCCIO - VCCIO* VCCIO* VCCIO*
I/O, FCLK3 292 AB1 AB5 AW31 DONE - AH4 AJ5 AY42
VCCINT - AB2 AD2 VCCINT* GND - GND* GND* GND*
I/O 293 - - BB30 I/O, GCK4 337 AJ4 AL4 AW41
I/O 294 AB3 AC4 BA35 I/O 338 AK3 AJ6 AV40
GND - GND* GND* GND* VCCINT - AH5 AK5 VCCINT*
VCCIO - VCCIO* VCCIO* - I/O 339 - - AV42
I/O 295 AB4 AD3 AT28 I/O 340 AK4 AN3 AR37
I/O 296 AC2 AE1 AU29 I/O 341 AJ5 AK6 AP36
I/O 297 AD1 AC5 BC33 I/O 342 AH6 AL5 AT38
I/O 298 AC3 AE3 AY34 I/O 343 - - -
I/O 299 - AD4 BB34 I/O 344 - - -
I/O 300 - AF1 AW33 I/O 345 - - -
I/O 301 - AF2 AU31 I/O 346 - - -
I/O 302 - AD5 AV32 VCCIO - VCCIO* VCCIO* -
I/O 304 - - - I/O 347 - - -
I/O 305 - - - I/O 348 - - -
I/O 306 - - - I/O 349 - - -
VCCIO - VCCIO* VCCIO* VCCIO* I/O 351 AL4 AJ7 AU39
I/O 307 - - - I/O 352 AK5 AM4 AU43
I/O 308 - - - I/O 353 AJ6 AM5 BA43
I/O 309 - AG2 AT30 I/O 354 AH7 AK7 AT42
I/O 310 - AE4 AU33 I/O 355 AL5 AL6 AR39
I/O 311 AC4 AF3 AW35 I/O 356 AK6 AJ8 AN37
I/O 312 AD2 AH1 BC35 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 313 AE2 AE5 AY36 GND - GND* GND* GND*
I/O 314 AD3 AG3 BB36 I/O 357 AH8 AM6 AT40
I/O 315 AD4 AF4 AV36 I/O 358 AJ7 AN6 AP40
I/O (D6) 316 AF1 AJ1 AU35 I/O 359 AL6 AK8 AR43
GND - GND* GND* GND* I/O 360 AK7 AL7 AN39
VCCIO - VCCIO* VCCIO* VCCIO* I/O 361 AH9 AJ9 AP42
I/O 317 AF2 AJ2 AT34 I/O 362 AJ8 AN7 AM38
I/O 318 AE3 AF5 AW37 I/O 363 - AL8 AN43

R

EPIC EPIC
Pad # Pad #
I/O 364 - AK9 AL37 I/O 410 - - -
I/O 367 - - - I/O 413 - - -
I/O 368 - - - I/O 414 - - -
I/O 369 - - - I/O 415 - AL16 AC39
I/O 370 - - - I/O 416 - AJ16 AC37
I/O 371 - AM8 AK36 I/O 417 AJ15 AK16 AB40
I/O 372 - AJ10 AR41 I/O 418 AL15 AN17 AB42
I/O 373 - AL9 AL41 VCCINT - AH15 AM17 VCCINT*
I/O 374 - AM9 AN41 I/O 419 - - AB38
I/O 375 AK8 AK10 AK42 I/O 420 AL16 AL17 AA41
I/O 376 AJ9 AN9 AM40 GND - GND* GND* GND*
I/O 377 AH10 AJ11 AJ43 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 378 AL8 AL10 AJ37 I/O (/INIT) 421 AK16 AJ17 AA39
VCCIO - VCCIO* VCCIO* - I/O 422 AJ16 AK17 AA37
GND - GND* GND* GND* I/O 423 AL17 AM18 Y40
I/O 379 AK9 AM10 AL39 I/O 424 AK17 AL18 Y38
I/O 380 - - AH36 I/O 425 - AK18 AA43
VCCINT - AJ10 AK11 VCCINT* I/O 426 - AJ18 W39 6
I/O 381 AK10 AL11 AH42 I/O 427 - - -
I/O 382 AL10 AJ12 AJ39 I/O 428 - - -
I/O 383 AJ11 AN11 AK40 I/O 430 - - -
I/O 384 AH12 AK12 AG41 VCCIO - VCCIO* VCCIO* -
I/O 385 - AL12 AJ41 GND - GND* GND* GND*
I/O 386 - AM12 AH40 I/O 431 - - -
I/O 387 - - - I/O 432 - - -
I/O 388 - - - I/O 433 - - -
I/O 389 - - - I/O 434 - - -
I/O 390 - - - I/O 435 AJ17 AN19 V40
GND - GND* GND* GND* I/O 436 AH17 AL19 Y36
VCCIO - VCCIO* VCCIO* - I/O 437 AK18 AK19 U41
I/O 391 - - - I/O 438 AL19 AM20 Y42
I/O 392 - - - I/O 439 AJ18 AJ19 T40
I/O 393 - AJ13 AG37 I/O 440 AK19 AL20 W37
I/O 394 - AK13 AG43 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 395 AK11 AL13 AG39 GND - GND* GND* GND*
I/O 396 AJ12 AM13 AF38 I/O 441 AH18 AK20 V38
I/O 397 AK12 AN13 AF42 I/O 442 AL20 AN21 U39
I/O 398 AL12 AJ14 AD42 I/O 443 AJ19 AL21 V42
I/O 399 AH13 AK14 AF40 I/O 444 AK20 AJ20 R41
I/O 400 AJ13 AL14 AE37 I/O 445 AH19 AM22 U43
GND - GND* GND* GND* I/O 446 AJ20 AK21 P40
VCCIO - VCCIO* VCCIO* VCCIO* VCCIO - - - -
I/O 401 AK13 AM14 AE39 GND - - - -
I/O 402 AL13 AN15 AD40 I/O 447 - AN23 T42
I/O 403 AK14 AJ15 AC43 I/O 448 - AL22 U37
I/O 404 AH14 AK15 AD38 I/O 449 - - -
I/O 405 AJ14 AL15 AC41 I/O 450 - - -
I/O 406 AK15 AM16 AD36 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 408 - - - I/O 451 - - -
I/O 409 - - - I/O 452 - - -

R

EPIC EPIC
Pad # Pad #
I/O 453 - - - I/O 498 - - -
I/O 454 - - - I/O (/LDC) 499 AH26 AJ27 K36
I/O 455 - AJ21 R39 I/O 500 AL28 AM31 J37
I/O 456 - AM23 N41 I/O 501 AJ27 AK28 B42
I/O 457 AK21 AK22 R43 I/O 502 AK28 AL30 D40
I/O 458 AH20 AM24 M40 I/O (HDC) 503 AH27 AK29 C41
VCCIO - VCCIO* VCCIO* - I/O, GCK3 504 AK29 AJ28 F38
I/O 459 AJ21 AL23 N39 I (M2) - AJ28 AN32 H36
I/O 460 AL22 AJ22 T36 VCCIO - VCCIO* VCCIO* VCCIO*
VCCINT - AK22 AN25 VCCINT* I (M0) - AH28 AJ29 C39
I/O 461 - - P42 GND - GND* GND* GND*
I/O 462 AJ22 AK23 R37 O (M1) - AH29 AK30 D38
GND - GND* GND* GND* I/O, GCK2 505 AJ30 AH29 E37
VCCIO - VCCIO* VCCIO* - I/O 506 AH30 AJ30 B40
I/O 463 AK23 AL24 L41 VCCINT - AG28 AK31 VCCINT*
I/O 464 AJ23 AN26 L43 I/O 507 - - H34
I/O 465 AH22 AJ23 K40 I/O 508 AH31 AH30 G35
I/O 466 AL24 AL25 K42 I/O 509 AG29 AL33 F36
I/O 467 - AK24 L39 I/O 510 AG30 AG29 D36
I/O 468 - AM26 J43 I/O 511 - - -
I/O 469 - AM27 M38 I/O 512 - - -
I/O 470 - AJ24 P36 I/O 513 - - -
I/O 471 - - - I/O 514 - - -
I/O 474 - - - I/O 515 AF28 AJ31 E35
GND - GND* GND* GND* I/O 516 AG31 AK32 A41
VCCIO - VCCIO* VCCIO* VCCIO* I/O 517 AF29 AG30 G33
I/O 475 - - - I/O 518 AF30 AH31 B38
I/O 476 - - - I/O 519 - - -
I/O 477 - AL26 N37 I/O 520 - - -
I/O 478 - AK25 H42 I/O 521 - - -
I/O 479 AK24 AN28 J41 I/O 522 - - -
I/O 480 AH23 AN29 G43 I/O 523 AE28 AF29 D34
I/O 481 AJ24 AJ25 H40 I/O 524 AF31 AJ32 E33
I/O 482 AK25 AL27 F42 VCCIO - VCCIO* VCCIO* VCCIO*
VCCINT - - - - GND - GND* GND* GND*
I/O 483 AJ25 AK26 J39 I/O 525 AE29 AH32 F32
I/O 484 AH24 AM29 L37 I/O 526 AE30 AF30 G31
GND - GND* GND* GND* I/O 527 AD28 AG31 A37
VCCIO - VCCIO* VCCIO* VCCIO* I/O 528 AD29 AE29 H30
I/O 485 AL26 AM30 E41 I/O 529 AD30 AH33 B36
I/O 486 AK26 AJ26 F40 I/O 530 AD31 AG33 C33
I/O 487 AH25 AL28 C43 I/O 531 - AE30 C35
I/O 488 AL27 AK27 G39 I/O 532 - AF31 D32
I/O 489 AJ26 AL29 D42 I/O 533 - - -
I/O 490 AK27 AN31 H38 I/O 534 - - -
I/O 491 - - - VCCIO - VCCIO* VCCIO* VCCIO*
I/O 493 - - - I/O 535 - - -
I/O 494 - - - I/O 536 - - -
I/O 495 - - - I/O 539 - AD29 E31
I/O 496 - - - I/O 540 - AF32 G29
I/O 497 - - - I/O 541 - AE31 A35

R

EPIC EPIC
Pad # Pad #
I/O 542 - AD30 H28 I/O 588 T30 U29 C21
I/O 543 AC28 AE32 B34 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 544 AC29 AC29 C31 GND - GND* GND* GND*
I/O 545 AC30 AE33 A33 I/O 589 T31 U31 E21
I/O 546 AB28 AD31 D30 I/O 590 - - D20
VCCIO - VCCIO* VCCIO* - VCCINT - R29 U32 VCCINT*
I/O, FCLK2 547 AB29 AC30 E29 I/O 592 - - -
I/O 548 - - C29 I/O 593 - - -
VCCINT - AB30 AD32 VCCINT* I/O 594 - - -
I/O 549 AB31 AB29 B30 I/O 595 - T32 C17
I/O 550 AA29 AC31 G27 I/O 596 - T30 G21
VCCIO - VCCIO* VCCIO* VCCIO* I/O 597 R28 T29 F20
I/O 551 - - - I/O 598 R30 T31 D18
I/O 554 - - - I/O 599 R31 R33 E19
I/O 555 - AC33 D28 I/O 600 P29 R31 B20
I/O 556 - AB30 A29 I/O 601 P28 R30 H20
I/O 557 AA30 AB31 E27 I/O 602 P30 R29 B18
I/O 558 Y28 AA29 F26 I/O 603 N30 P32 C15
GND - GND* GND* GND* I/O 604 N29 P31 D16 6
I/O 559 Y29 AB32 C27 I/O 606 - - -
I/O 560 Y30 AA30 B28 I/O 607 - - -
I/O 561 Y31 AA31 G25 I/O 608 - - -
I/O 562 W28 AA32 A27 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 564 - - - I/O 609 N28 N33 G19
I/O 565 - Y29 D26 I/O 610 N31 P30 A17
I/O 566 - AA33 B26 I/O 611 M31 P29 F18
I/O 567 W29 Y30 E25 I/O 612 M29 M32 E17
I/O 568 W30 Y31 F24 I/O 613 - - -
VCCIO - VCCIO* VCCIO* VCCIO* I/O 615 - N31 B16
I/O 569 - - - I/O 616 - N30 C13
I/O 570 - - - I/O 617 M28 L33 A15
I/O 571 - - - I/O 618 M30 N29 D14
I/O 573 W31 Y32 H24 GND - GND* GND* GND*
I/O 574 V28 W29 B24 I/O 619 L30 M31 E15
I/O 575 V29 W30 D24 I/O 620 L29 L32 G17
I/O 576 V30 W31 A23 I/O 621 - M30 B14
I/O 577 U29 W33 C23 I/O 622 - L31 C11
I/O 578 U28 V30 E23 I/O 623 - - -
I/O 579 U30 V29 G23 I/O 626 - - -
I/O 580 U31 V31 B22 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 581 - V32 D22 I/O 627 K31 M29 D12
I/O 582 - U33 A21 I/O (TMS) 628 K30 J33 A11
I/O 583 - - - VCCINT - K29 K32 VCCINT*
I/O 584 - - - I/O 629 - - E13
I/O 585 - - - I/O, FCLK1 630 K28 L30 C9
I/O 587 T29 U30 F22 VCCIO - VCCIO* VCCIO* -

R

EPIC EPIC
Pad # Pad #
I/O 631 J30 K31 H16 VCCIO - A31 A22 B2
I/O 632 H31 L29 B10 VCCIO - D11 A26 C7
I/O 633 J29 H33 G15 VCCIO - D21 A30 C19
I/O 634 J28 K30 A9 VCCIO - L1 B2 C25
I/O 635 - J31 D10 VCCIO - L4 B13 C37
I/O 636 - H32 B8 VCCIO - L28 B19 F14
I/O 637 - K29 E11 VCCIO - L31 B32 F30
I/O 638 - H31 G13 VCCIO - AA1 C3 G3
I/O 639 - - - VCCIO - AA4 C31 G7
I/O 640 - - - VCCIO - AA28 C32 G37
I/O 641 - - - VCCIO - AA31 D1 G41
I/O 642 - - - VCCIO - AH11 D33 N1
GND - GND* GND* GND* VCCIO - AH21 E5 N43
VCCIO - VCCIO* VCCIO* VCCIO* VCCIO - AL1 H1 P6
I/O 643 H30 J30 F12 VCCIO - AL11 K33 P38
I/O 644 G30 G32 H14 VCCIO - AL21 M1 W3
I/O 645 H29 F33 G11 VCCIO - AL31 N32 W41
I/O 646 H28 J29 A7 VCCIO - C3 R2 AE3
I/O 647 F31 G31 E9 VCCIO - C29 T33 AE41
I/O 648 F30 H30 B6 VCCIO - AJ3 V1 AK6
I/O 649 - E33 D8 VCCIO - AJ29 W32 AK38
I/O 650 - E32 F8 VCCIO - - AA2 AL1
I/O 651 - - - VCCIO - - AB33 AL43
I/O 652 - - - VCCIO - - AD1 AU3
GND - GND* GND* GND* VCCIO - - AF33 AU7
VCCIO - VCCIO* VCCIO* VCCIO* VCCIO - - AK1 AU37
I/O 653 G29 H29 G9 VCCIO - - AK4 AU41
I/O 654 G28 F31 H10 VCCIO - - AK33 AV14
I/O 655 - - - VCCIO - - AL2 AV30
I/O 656 - - - VCCIO - - AL3 BA7
I/O 657 E31 G30 B4 VCCIO - - AL31 BA19
I/O 658 E30 D32 E7 VCCIO - - AM2 BA25
I/O 659 F29 E31 C5 VCCIO - - AM15 BA37
I/O 660 F28 G29 D6 VCCIO - - AM21 BC1
I/O 661 - - - VCCIO - - AM32 BC13
I/O 662 - - - VCCIO - - AN4 BC31
GND - GND* GND* GND* VCCIO - - AN8 BC43
VCCIO - VCCIO* VCCIO* - VCCIO - - AN12 -
I/O 663 - - - VCCIO - - AN18 -
I/O 664 - - - VCCIO - - AN24 -
I/O 665 - - - VCCIO - - AN30 -
I/O 666 - - -
GND - - - - VCCINT - - - H12
VCCIO - - - - VCCINT - - - H18
I/O (TCK) 667 D31 C33 D4 VCCINT - - - H26
I/O (TDI) 668 D30 F30 H8 VCCINT - - - H32
I/O 669 E29 D31 A3 VCCINT - - - M8
I/O 670 - - F6 VCCINT - - - M36
VCCINT - E28 E30 VCCINT* VCCINT - - - V8
I/O (A17) 671 C30 F29 C3 VCCINT - - - V36
I/O, GCK1 (A16) 672 D29 B33 C1 VCCINT - - - AF8
GND - GND* GND* GND* VCCINT - - - AF36
VCCINT - - - AM8
VCCIO - A1 A4 A13 VCCINT - - - AM36
VCCIO - A11 A10 A31 VCCINT - - - AT12

R

EPIC EPIC
Pad # Pad #
VCCINT - - - AT26 GND - AJ1 AB1 AP38
VCCINT - - - AT32 GND - AJ31 AC32 AT22
GND - AK1 AD33 AV10
GND - A2 A7 A5 GND - AK2 AE2 AV16
GND - A3 A12 A19 GND - AK30 AG1 AV28
GND - A7 A14 A25 GND - AK31 AG32 AV34
GND - A9 A18 A39 GND - AL2 AH2 AW1
GND - A14 A20 B12 GND - AL3 AJ33 AW5
GND - A18 A24 B32 GND - AL7 AL32 AW39
GND - A23 A29 E1 GND - AL9 AM3 AW43
GND - A25 A32 E5 GND - AL14 AM11 BB12
GND - A29 B1 E39 GND - AL18 AM19 BB32
GND - A30 B6 E43 GND - AL23 AM25 BC5
GND - B1 B9 F10 GND - AL25 AM28 BC19
GND - B31 B27 F34 GND - - AN2 -
GND - C1 B31 H22 GND - - AN5 -
GND - C31 C2 K6 GND - - AN10 -
GND - D16 E1 K38 GND - - AN14 -
GND - G1 F32 M2 GND - - AN16 -
GND - G31 G2 M42 GND - - AN20 - 6
GND - J1 G33 T6 GND - - AN22 -
GND - J31 J32 T38 GND - - AN27 -
GND - P1 K1 W1
GND - P31 L2 W43 NC - C8 A1 -
GND - T4 M33 AB8 NC - - A33 -
GND - T28 P1 AB36 NC - - AC2 -
GND - V1 P33 AE1 NC - - AN1 -
GND - V31 R32 AE43 NC - - AN33 -
GND - AC1 T1 AH6 12/21/98
GND - AC31 V33 AH38
GND - AE1 W2 AM2
GND - AE31 Y1 AM42
GND - AH16 Y33 AP6

R

XC40250XV Pinout Table XC40250XV Pinout Table (Continued)
XC40250XV Pinout Table EPIC

Pad #
EPIC GND - GND* GND* GND*
Pad #
I/O 47 A24 A25 R7
VCCIO - VCCIO* VCCIO* VCCIO*
I/O 48 - - P4
I/O, GCK8 (A15) 1 D28 E29 E3
VCCINT - B23 C24 VCCINT*
I/O (A14) 2 C28 D30 F4
I/O 49 C22 D23 T2
I/O 3 B29 E28 J7
I/O 50 B22 B24 T8
I/O 4 D27 D29 D2
I/O 5 B28 C30 K8
I/O 51 A22 E22 R5
I/O 6 C27 D28 H6 I/O 52 C21 C23 R3
I/O 7 - - - I/O 53 - A23 U1
I/O 8 - - - I/O 54 - D22 T4
I/O 9 - - - I/O 55 - - -
I/O 10 - - - I/O 56 - - -
VCCIO - VCCIO* VCCIO* -
I/O 57 - - -
I/O 58 - - -
I/O 11 - - -
I/O 12 - - -
I/O 13 - - -
I/O 59 - - -
I/O 14 - - -
I/O 60 - - -
I/O 15 - - -
I/O 61 - - -
I/O 16 - - -
I/O 62 - - -
I/O 17 D26 A31 G5
I/O 63 - C22 U5
I/O 18 A28 E27 F2 6
I/O 64 - E21 V2
I/O 19 B27 C29 H4
I/O 65 D20 B22 U7
I/O 20 C26 B30 G1
I/O 66 B21 D21 U3
I/O 21 D25 D27 J5
I/O 67 C20 C21 Y2
I/O 22 A27 C28 L7
I/O 68 B20 B21 V4
I/O 69 A20 E20 V6
I/O 70 D19 A21 W5
I/O (A13) 23 B26 E26 J3 GND - GND* GND* GND*
I/O (A12) 24 D24 B29 K4 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 25 C25 B28 H2 I/O (A11) 71 C19 D20 W7
I/O 26 A26 D26 L5 I/O (A10) 72 B19 C20 Y4
I/O 27 B25 C27 J1 I/O 73 A19 B20 AA1
I/O 28 D23 E25 M6
I/O 74 B18 E19 Y6
I/O 29 - A28 K2
I/O (A18) 75 D18 D19 AB2
I/O 30 - A27 N7
I/O (A19) 76 C18 C19 Y8
I/O 31 - - -
I/O 77 - - -
I/O 32 - - -
I/O 78 - - -
I/O 33 - - -
I/O 79 - - -
I/O 34 - - -
I/O 80 - - -
I/O 81 - - -
I/O 82 - - -
I/O 35 - - -
I/O 36 - - -
I/O 37 - - -
I/O 83 - - -
I/O 38 - - -
I/O 84 - - -
I/O 39 - D25 P8
I/O 85 - - -
I/O 40 - C26 L3
I/O 86 - - -
I/O 41 - E24 L1 I/O 87 - A19 AA3
I/O 42 - B26 M4 I/O 88 - D18 AA5
I/O 43 C24 C25 P2 I/O 89 B17 E18 AC1
I/O 44 B24 D24 N5 I/O 90 C17 C18 AA7
I/O 45 C23 B25 R1 I/O (A9) 91 A17 B18 AB4
I/O 46 D22 E23 N3
I/O (A8) 92 D17 A17 AB6

R
EPIC EPIC
Pad # Pad #
I/O (A7) 93 A16 D17 AC3 VCCIO - VCCIO VCCIO* -
I/O (A6) 94 B16 E17 AC5 I/O 139 C10 D11 AJ7
VCCINT - C16 C17 VCCINT* I/O 140 B9 C10 AP2
I/O 95 - - AD2 I/O 141 C9 A9 AN3
I/O 96 A15 B17 AC7 I/O 142 D10 E11 AP4
I/O 97 - B16 AF2 I/O 143 - C9 AR1
I/O 98 - D16 AD4 I/O 144 - D10 AN5
I/O 99 - - - I/O 145 - A8 AM6
I/O 100 - - - I/O 146 - B8 AK8
I/O 101 - - - I/O 147 - - -
I/O 102 - - - I/O 148 - - -
I/O 105 - - - I/O 151 - - -
I/O 106 - - - I/O 152 - - -
I/O 107 - - - I/O 153 - - -
I/O 108 - - - I/O 154 - - -
I/O (A20) 109 B15 E16 AD6 I/O 155 - E10 AL7
I/O (A21) 110 C15 C16 AE5 I/O 156 - C8 AT2
I/O 111 D15 A15 AF4 I/O 157 A8 D9 AR3
I/O 112 B14 C15 AG1 I/O 158 B8 B7 AU1
I/O (A5) 113 A13 D15 AD8 I/O 159 D9 A6 AT4
I/O (A4) 114 C14 E15 AG3 I/O 160 B7 E9 AV2
VCCIO - VCCIO* VCCIO* VCCIO* I/O 161 C7 C7 AR5
GND - GND* GND* GND* I/O 162 D8 D8 AN7
I/O 115 B13 B14 AH4 GND - GND* GND* GND*
I/O 116 D14 C14 AE7 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 117 A12 D14 AH2 I/O 163 A6 A5 AW3
I/O 118 C13 A13 AF6 I/O 164 B6 B5 AV4
I/O 119 B12 E14 AJ1 I/O (A3) 165 D7 E8 BA1
I/O 120 D13 C13 AG5 I/O,, (CS1, A2) 166 A5 C6 AU5
I/O 121 - D13 AJ3 I/O 167 C6 D7 AY2
I/O 122 - B12 AK4 I/O 168 B5 B4 AT6
I/O 123 - - - I/O 169 - - -
I/O 124 - - - I/O 170 - - -
I/O 125 - - - I/O 171 - - -
I/O 126 - - - I/O 172 - - -
I/O 129 - - - I/O 175 - - -
I/O 130 - - - I/O 176 - - -
I/O 131 - C12 AG7 I/O 177 - - -
I/O 132 - E13 AK2 I/O 178 - - -
I/O 133 C12 A11 AJ5 I/O 179 D6 C5 AP8
I/O 134 B11 D12 AL3 I/O 180 A4 E7 AR7
VCCIO - VCCIO* VCCIO* VCCIO* I/O 181 C5 A3 AY4
I/O 135 D12 B11 AM4 I/O 182 - - BB2
I/O 136 C11 C11 AN1 VCCINT - B4 D6 VCCINT*
VCCINT - A10 E12 VCCINT* I/O, GCK7 (A1) 183 D5 A2 AV6
I/O 137 - - AL5 I/O (A0, /WS) 184 B3 D5 AT8
I/O 138 B10 B10 AH8 GND - GND* GND* GND*

R
EPIC EPIC
Pad # Pad #
O, TDO - C4 E6 BA3 GND - GND* GND* GND*
VCCIO - VCCIO* VCCIO* VCCIO* I/O 231 K4 K3 BA13
CCLK - D4 C4 BA5 I/O 232 - - AY14
I/O,CK6 (DOUT) 185 D3 D4 BB4 VCCINT - K3 J1 VCCINT*
I/O (D0, DIN) 186 C2 E4 AY6 I/O, FCLK4 233 K2 L4 BC15
I/O 187 D2 F5 BC3 I/O 234 K1 K2 AW15
I/O 188 E4 B3 AW7 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 189 D1 D3 BB6 I/O 235 - - -
I/O 190 E3 F4 AU9 I/O 236 - - -
I/O 191 - - - I/O 237 - - -
I/O 192 - - - I/O 238 - - -
I/O 193 - - - I/O 239 - M5 BA15
I/O 194 - - - I/O 240 - L3 AU17
VCCIO - VCCIO* VCCIO* - I/O 241 L3 L1 BB16
GND - GND* GND* GND* I/O (D2) 242 L2 M4 AY16
I/O 197 - - - I/O 243 M4 M3 AW17
I/O 198 - - - I/O 244 M3 N5 AV18
I/O 199 E2 C1 AT10 I/O 245 - M2 BA17
I/O 200 F4 G5 AV8 I/O 246 - N4 BC17
I/O 201 E1 E3 AY8 I/O 247 - - -
6
I/O 202 F3 D2 BC7 I/O 248 - - -
I/O 203 - - - I/O 249 - - -
I/O 204 - - - I/O 250 - - -
I/O (/RCK, RDY_/BUSY) 205 F2 G4 AW9 I/O 251 M2 N3 AU19
I/O (D1) 206 G4 F3 BA9 I/O 252 M1 N2 BB18
VCCIO - VCCIO* VCCIO* VCCIO* I/O 253 N4 P5 AY18
GND - GND* GND* GND* I/O 254 N3 N1 AW19
I/O 209 - H5 AU11 I/O 255 - - -
I/O 210 - E2 AY10 I/O 256 - - -
I/O 211 F1 F2 BB8 I/O 257 - - -
I/O 212 G3 H4 AW11 I/O 258 - - -
I/O 213 G2 G3 BC9 I/O 259 - - -
I/O 214 H4 J5 AV12 I/O 260 - - -
I/O 215 H3 F1 AU13 I/O 261 N2 P4 AV20
I/O 216 H2 G1 AT14 I/O 262 N1 P3 AT20
I/O 217 - - - I/O 263 P4 P2 BB20
I/O 218 - - - I/O 264 P3 R5 AY20
VCCIO - VCCIO* VCCIO* VCCIO* I/O 265 P2 R4 BC21
GND - GND* GND* GND* I/O 266 R3 R3 BA21
I/O 221 - - - I/O 267 R4 R1 AW21
I/O 222 - - - I/O 268 - T4 AU21
I/O 223 - J4 BA11 I/O 269 - T5 BB22
I/O 224 - H3 AY12 I/O 270 - - AY22
I/O 225 - K5 BB10 I/O 271 - - -
I/O 226 - H2 AW13 I/O 272 - - -
I/O 227 H1 J3 BC11 I/O 273 - - -
I/O 228 J4 K4 AU15 I/O 274 - - -
I/O 229 J3 J2 BB14 VCCINT - R2 T3 VCCINT*
I/O 230 J2 L5 AT16 I/O (/RS) 275 R1 T2 AV22
VCCIO - VCCIO* VCCIO* - I/O (D3) 276 T3 U3 BA23

R
EPIC EPIC
Pad # Pad #
VCCIO - VCCIO* VCCIO* VCCIO* VCCIO - VCCIO* VCCIO* -
I/O 277 T2 U4 AW23 I/O 323 AB4 AD3 AT28
I/O (D4) 278 T1 U5 AY24 I/O 324 AC2 AE1 AU29
I/O 279 - - - I/O 325 AD1 AC5 BC33
I/O 280 - - - I/O 326 AC3 AE3 AY34
I/O 281 - - - I/O 327 - AD4 BB34
I/O 282 - - - I/O 328 - AF1 AW33
I/O 283 - U1 BC23 I/O 329 - AF2 AU31
I/O 284 - U2 BA27 I/O 330 - AD5 AV32
I/O 285 U3 V2 AU23 I/O 331 - - -
I/O 286 U4 V4 AV24 I/O 332 - - -
I/O 287 U2 V5 AY26 GND - GND* GND* GND*
I/O 288 U1 V3 BB24 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 289 V3 W1 AW25 I/O 335 - - -
I/O 290 V4 W3 BB26 I/O 336 - - -
I/O 291 V2 W4 AT24 I/O 337 - - -
I/O 292 W2 W5 BA29 I/O 338 - - -
I/O 293 - - - I/O 339 - AG2 AT30
I/O 294 - - - I/O 340 - AE4 AU33
I/O 295 - - - I/O 341 AC4 AF3 AW35
I/O 296 - - - I/O 342 AD2 AH1 BC35
I/O 297 - - - I/O 343 AE2 AE5 AY36
I/O 298 - - - I/O 344 AD3 AG3 BB36
VCCIO - VCCIO* VCCIO* VCCIO* I/O 345 AD4 AF4 AV36
GND - GND* GND* GND* I/O (D6) 346 AF1 AJ1 AU35
I/O 299 W3 Y2 AY28 GND - GND* GND* GND*
I/O 300 W4 Y3 AU25 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 301 - Y4 AV26 I/O 347 AF2 AJ2 AT34
I/O 302 - AA1 BC27 I/O 348 AE3 AF5 AW37
I/O 303 - - - I/O 349 - - -
I/O 304 - - - I/O 350 - - -
I/O 305 - - - I/O 351 - - -
I/O 306 - - - I/O 352 - - -
I/O 307 W1 Y5 AW27 I/O 353 AE4 AH3 BC37
I/O 308 Y1 AA3 BB28 I/O 354 AG1 AG4 AY38
I/O 309 Y3 AA4 BA31 I/O 355 AG2 AK2 BB38
I/O 310 Y4 AB2 AY30 I/O 356 AF3 AJ3 BA41
I/O (/CS0) 311 Y2 AB3 AW29 GND - GND* GND* GND*
I/O (D5) 312 AA2 AA5 BC29 VCCIO - VCCIO* VCCIO* -
I/O 313 - AC1 AU27 I/O 359 - - -
I/O 314 - AB4 BA33 I/O 360 - - -
I/O 315 - - - I/O 361 - - -
I/O 316 - - - I/O 362 - - -
I/O 317 - - - I/O 363 AF4 AG5 AY40
I/O 318 - - - I/O 364 AH1 AL1 BB40
VCCIO - VCCIO* VCCIO* VCCIO* I/O 365 AH2 AH4 AT36
I/O 319 AA3 AC3 AY32 I/O 366 AG3 AK3 BA39
I/O, FCLK3 320 AB1 AB5 AW31 I/O, GCK5 367 AG4 AJ4 AV38
VCCINT - AB2 AD2 VCCINT* I/O (D7) 368 AJ2 AH5 BC41
I/O 321 - - BB30 /PROGRAM - AH3 AM1 BB42
I/O 322 AB3 AC4 BA35 VCCIO - VCCIO* VCCIO* VCCIO*

R
EPIC EPIC
Pad # Pad #
DONE - AH4 AJ5 AY42 GND - GND* GND* GND*
GND - GND* GND* GND* I/O 415 AK9 AM10 AL39
I/O, GCK4 369 AJ4 AL4 AW41 I/O 416 - - AH36
I/O 370 AK3 AJ6 AV40 VCCINT - AJ10 AK11 VCCINT*
VCCINT - AH5 AK5 VCCINT* I/O 417 AK10 AL11 AH42
I/O 371 - - AV42 I/O 418 AL10 AJ12 AJ39
I/O 372 AK4 AN3 AR37 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 373 AJ5 AK6 AP36 I/O 419 AJ11 AN11 AK40
I/O 374 AH6 AL5 AT38 I/O 420 AH12 AK12 AG41
I/O 375 - - - I/O 421 - AL12 AJ41
I/O 376 - - - I/O 422 - AM12 AH40
I/O 377 - - - I/O 423 - - -
I/O 378 - - - I/O 424 - - -
I/O 381 - - - I/O 427 - - -
I/O 382 - - - I/O 428 - - -
I/O 383 - - - I/O 429 - - -
I/O 384 - - - I/O 430 - - -
I/O 385 AL4 AJ7 AU39 I/O 431 - AJ13 AG37
6
I/O 386 AK5 AM4 AU43 I/O 432 - AK13 AG43
I/O 387 AJ6 AM5 BA43 I/O 433 AK11 AL13 AG39
I/O 388 AH7 AK7 AT42 I/O 434 AJ12 AM13 AF38
I/O 389 AL5 AL6 AR39 I/O 435 AK12 AN13 AF42
I/O 390 AK6 AJ8 AN37 I/O 436 AL12 AJ14 AD42
VCCIO - VCCIO* VCCIO* VCCIO* I/O 437 AH13 AK14 AF40
GND - GND* GND* GND* I/O 438 AJ13 AL14 AE37
I/O 391 AH8 AM6 AT40 GND - GND* GND* GND*
I/O 392 AJ7 AN6 AP40 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 393 AL6 AK8 AR43 I/O 439 AK13 AM14 AE39
I/O 394 AK7 AL7 AN39 I/O 440 AL13 AN15 AD40
I/O 395 AH9 AJ9 AP42 I/O 441 AK14 AJ15 AC43
I/O 396 AJ8 AN7 AM38 I/O 442 AH14 AK15 AD38
I/O 397 - AL8 AN43 I/O 443 AJ14 AL15 AC41
I/O 398 - AK9 AL37 I/O 444 AK15 AM16 AD36
I/O 399 - - - I/O 445 - - -
I/O 400 - - - I/O 446 - - -
I/O 401 - - - I/O 447 - - -
I/O 402 - - - I/O 448 - - -
I/O 405 - - - I/O 451 - - -
I/O 406 - - - I/O 452 - - -
I/O 407 - AM8 AK36 I/O 453 - - -
I/O 408 - AJ10 AR41 I/O 454 - - -
I/O 409 - AL9 AL41 I/O 455 - AL16 AC39
I/O 410 - AM9 AN41 I/O 456 - AJ16 AC37
I/O 411 AK8 AK10 AK42 I/O 457 AJ15 AK16 AB40
I/O 412 AJ9 AN9 AM40 I/O 458 AL15 AN17 AB42
I/O 413 AH10 AJ11 AJ43 VCCINT - AH15 AM17 VCCINT*
I/O 414 AL8 AL10 AJ37 I/O 459 - - AB38
VCCIO - VCCIO* VCCIO* - I/O 460 AL16 AL17 AA41

R
EPIC EPIC
Pad # Pad #
VCCIO - VCCIO* VCCIO* VCCIO* VCCIO - VCCIO* VCCIO* -
I/O (/INIT) 461 AK16 AJ17 AA39 I/O 507 AK23 AL24 L41
I/O 462 AJ16 AK17 AA37 I/O 508 AJ23 AN26 L43
I/O 463 AL17 AM18 Y40 I/O 509 AH22 AJ23 K40
I/O 464 AK17 AL18 Y38 I/O 510 AL24 AL25 K42
I/O 465 - AK18 AA43 I/O 511 - AK24 L39
I/O 466 - AJ18 W39 I/O 512 - AM26 J43
I/O 467 - - - I/O 513 - AM27 M38
I/O 468 - - - I/O 514 - AJ24 P36
I/O 469 - - - I/O 515 - - -
I/O 470 - - - I/O 516 - - -
I/O 473 - - - I/O 519 - - -
I/O 474 - - - I/O 520 - - -
I/O 475 - - - I/O 521 - - -
I/O 476 - - - I/O 522 - - -
I/O 477 AJ17 AN19 V40 I/O 523 - AL26 N37
I/O 478 AH17 AL19 Y36 I/O 524 - AK25 H42
I/O 479 AK18 AK19 U41 I/O 525 AK24 AN28 J41
I/O 480 AL19 AM20 Y42 I/O 526 AH23 AN29 G43
I/O 481 AJ18 AJ19 T40 I/O 527 AJ24 AJ25 H40
I/O 482 AK19 AL20 W37 I/O 528 AK25 AL27 F42
VCCIO - VCCIO* VCCIO* VCCIO* I/O 529 AJ25 AK26 J39
GND - GND* GND* GND* I/O 530 AH24 AM29 L37
I/O 483 AH18 AK20 V38 GND - GND* GND* GND*
I/O 484 AL20 AN21 U39 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 485 AJ19 AL21 V42 I/O 531 AL26 AM30 E41
I/O 486 AK20 AJ20 R41 I/O 532 AK26 AJ26 F40
I/O 487 AH19 AM22 U43 I/O 533 AH25 AL28 C43
I/O 488 AJ20 AK21 P40 I/O 534 AL27 AK27 G39
I/O 489 - AN23 T42 I/O 535 AJ26 AL29 D42
I/O 490 - AL22 U37 I/O 536 AK27 AN31 H38
I/O 491 - - - I/O 537 - - -
I/O 492 - - - I/O 538 - - -
I/O 493 - - - I/O 539 - - -
I/O 494 - - - I/O 540 - - -
I/O 497 - - - I/O 543 - - -
I/O 498 - - - I/O 544 - - -
I/O 499 - AJ21 R39 I/O 545 - - -
I/O 500 - AM23 N41 I/O 546 - - -
I/O 501 AK21 AK22 R43 I/O (/LDC) 547 AH26 AJ27 K36
I/O 502 AH20 AM24 M40 I/O 548 AL28 AM31 J37
VCCIO - VCCIO* VCCIO* - I/O 549 AJ27 AK28 B42
I/O 503 AJ21 AL23 N39 I/O 550 AK28 AL30 D40
I/O 504 AL22 AJ22 T36 I/O (HDC) 551 AH27 AK29 C41
VCCINT - AK22 AN25 VCCINT* I/O, GCK3 552 AK29 AJ28 F38
I/O 505 - - P42 I (M2) - AJ28 AN32 H36
I/O 506 AJ22 AK23 R37 VCCIO - VCCIO* VCCIO* VCCIO*

R
EPIC EPIC
Pad # Pad #
I (M0) - AH28 AJ29 C39 VCCIO - VCCIO* VCCIO* -
O (M1) - AH29 AK30 D38 I/O, FCLK2 599 AB29 AC30 E29
I/O, GCK2 553 AJ30 AH29 E37 I/O 600 - - C29
I/O 554 AH30 AJ30 B40 VCCINT - AB30 AD32 VCCINT*
VCCINT - AG28 AK31 VCCINT* I/O 601 AB31 AB29 B30
I/O 555 - - H34 I/O 602 AA29 AC31 G27
I/O 556 AH31 AH30 G35 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 557 AG29 AL33 F36 I/O 603 - - -
I/O 558 AG30 AG29 D36 I/O 604 - - -
I/O 559 - - - I/O 605 - - -
I/O 560 - - - I/O 606 - - -
I/O 561 - - - I/O 607 - AC33 D28
I/O 562 - - - I/O 608 - AB30 A29
VCCIO - VCCIO* VCCIO* - I/O 609 AA30 AB31 E27
GND - GND* GND* GND* I/O 610 Y28 AA29 F26
I/O 563 AF28 AJ31 E35 GND - GND* GND* GND*
I/O 564 AG31 AK32 A41 VCCIO - VCCIO* VCCIO* -
I/O 565 AF29 AG30 G33 I/O 611 Y29 AB32 C27
I/O 566 AF30 AH31 B38 I/O 612 Y30 AA30 B28
I/O 567 - - - I/O 613 Y31 AA31 G25
I/O 568 - - - I/O 614 W28 AA32 A27
6
I/O 569 - - - I/O 615 - - -
I/O 570 - - - I/O 616 - - -
I/O 571 - - - I/O 617 - - -
I/O 572 - - - I/O 618 - - -
I/O 573 AE28 AF29 D34 I/O 619 - Y29 D26
I/O 574 AF31 AJ32 E33 I/O 620 - AA33 B26
VCCIO - VCCIO* VCCIO* VCCIO* I/O 621 W29 Y30 E25
GND - GND* GND* GND* I/O 622 W30 Y31 F24
I/O 575 AE29 AH32 F32 GND - GND* GND* GND*
I/O 576 AE30 AF30 G31 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 577 AD28 AG31 A37 I/O 623 - - -
I/O 578 AD29 AE29 H30 I/O 624 - - -
I/O 579 AD30 AH33 B36 I/O 625 - - -
I/O 580 AD31 AG33 C33 I/O 626 - - -
I/O 581 - AE30 C35 I/O 627 - - -
I/O 582 - AF31 D32 I/O 628 - - -
I/O 583 - - - I/O 629 W31 Y32 H24
I/O 584 - - - I/O 630 V28 W29 B24
I/O 585 - - - I/O 631 V29 W30 D24
I/O 586 - - - I/O 632 V30 W31 A23
VCCIO - VCCIO* VCCIO* VCCIO* I/O 633 U29 W33 C23
GND - GND* GND* GND* I/O 634 U28 V30 E23
I/O 589 - - - I/O 635 U30 V29 G23
I/O 590 - - - I/O 636 U31 V31 B22
I/O 591 - AD29 E31 I/O 637 - V32 D22
I/O 592 - AF32 G29 I/O 638 - U33 A21
I/O 593 - AE31 A35 I/O 639 - - -
I/O 594 - AD30 H28 I/O 640 - - -
I/O 595 AC28 AE32 B34 I/O 641 - - -
I/O 596 AC29 AC29 C31 I/O 642 - - -
I/O 597 AC30 AE33 A33 I/O 643 T29 U30 F22
I/O 598 AB28 AD31 D30 I/O 644 T30 U29 C21

R
EPIC EPIC
Pad # Pad #
VCCIO - VCCIO* VCCIO* VCCIO* I/O, FCLK1 690 K28 L30 C9
I/O 645 T31 U31 E21 VCCIO - VCCIO* VCCIO* -
I/O 646 - - D20 I/O 691 J30 K31 H16
VCCINT - R29 U32 VCCINT* I/O 692 H31 L29 B10
I/O 647 - - - I/O 693 J29 H33 G15
I/O 648 - - - I/O 694 J28 K30 A9
I/O 649 - - - I/O 695 - J31 D10
I/O 650 - - - I/O 696 - H32 B8
I/O 651 - T32 C17 I/O 697 - K29 E11
I/O 652 - T30 G21 I/O 698 - H31 G13
I/O 653 R28 T29 F20 I/O 699 - - -
I/O 654 R30 T31 D18 I/O 700 - - -
I/O 655 R31 R33 E19 GND - GND* GND* GND*
I/O 656 P29 R31 B20 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 657 P28 R30 H20 I/O 703 - - -
I/O 658 P30 R29 B18 I/O 704 - - -
I/O 659 N30 P32 C15 I/O 705 H30 J30 F12
I/O 660 N29 P31 D16 I/O 706 G30 G32 H14
I/O 661 - - - I/O 707 H29 F33 G11
I/O 662 - - - I/O 708 H28 J29 A7
I/O 663 - - - I/O 709 F31 G31 E9
I/O 664 - - - I/O 710 F30 H30 B6
I/O 665 - - - I/O 711 - E33 D8
I/O 666 - - - I/O 712 - E32 F8
I/O 667 N28 N33 G19 GND - GND* GND* GND*
I/O 668 N31 P30 A17 VCCIO - VCCIO* VCCIO* VCCIO*
I/O 669 M31 P29 F18 I/O 715 G29 H29 G9
I/O 670 M29 M32 E17 I/O 716 G28 F31 H10
I/O 671 - - - I/O 717 - - -
I/O 672 - - - I/O 718 - - -
I/O 673 - - - I/O 719 E31 G30 B4
I/O 674 - - - I/O 720 E30 D32 E7
I/O 675 - N31 B16 I/O 721 F29 E31 C5
I/O 676 - N30 C13 I/O 722 F28 G29 D6
I/O 677 M28 L33 A15 I/O 723 - - -
I/O 678 M30 N29 D14 I/O 724 - - -
I/O 679 L30 M31 E15 GND - GND* GND* GND*
I/O 680 L29 L32 G17 VCCIO - VCCIO* VCCIO* -
I/O 681 - M30 B14 I/O 727 - - -
I/O 682 - L31 C11 I/O 728 - - -
I/O 683 - - - I/O 729 - - -
I/O 684 - - - I/O 730 - - -
I/O 685 - - - I/O (TCK) 731 D31 C33 D4
I/O 686 - - - I/O (TDI) 732 D30 F30 H8
VCCIO - VCCIO* VCCIO* VCCIO* I/O 733 E29 D31 A3
I/O 687 K31 M29 D12 I/O 734 - - F6
I/O (TMS) 688 K30 J33 A11 VCCINT - E28 E30 VCCINT*
VCCINT - K29 K32 VCCINT* I/O (A17) 735 C30 F29 C3
I/O 689 - - E13 I/O, GCK1 (A16) 736 D29 B33 C1

R
EPIC EPIC
Pad # Pad #
GND - GND* GND* GND* VCCINT - - - AF36
VCCINT - - - AM8
VCCIO - A1 A4 A13 VCCINT - - - AM36
VCCIO - A31 A22 B2 VCCINT - - - AT26
VCCIO - D11 A26 C7 VCCINT - - - AT32
VCCIO - D21 A30 C19
VCCIO - L1 B2 C25 GND - A2 A7 A5
VCCIO - L4 B13 C37 GND - A3 A12 A19
VCCIO - L28 B19 F14 GND - A7 A14 A25
VCCIO - L31 B32 F30 GND - A9 A18 A39
VCCIO - AA1 C3 G3 GND - A14 A20 B12
VCCIO - AA4 C31 G7 GND - A18 A24 B32
VCCIO - AA28 C32 G37 GND - A23 A29 E1
VCCIO - AA31 D1 G41 GND - A25 A32 E5
VCCIO - AH11 D33 N1 GND - A29 B1 E39
VCCIO - AH21 E5 N43 GND - A30 B6 E43
VCCIO - AL1 H1 P6 GND - B1 B9 F10
VCCIO - AL11 K33 P38 GND - B2 B15 F16
VCCIO - AL21 M1 W3 GND - B30 B23 F28
VCCIO - AL31 N32 W41 GND - B31 B27 F34
6
VCCIO - C3 R2 AE3 GND - C1 B31 H22
VCCIO - C29 T33 AE41 GND - C31 C2 K6
VCCIO - AJ3 V1 AK6 GND - D16 E1 K38
VCCIO - AJ29 W32 AK38 GND - G1 F32 M2
VCCIO - - AA2 AL1 GND - G31 G2 M42
VCCIO - - AB33 AL43 GND - J1 G33 T6
VCCIO - - AD1 AU3 GND - J31 J32 T38
VCCIO - - AF33 AU7 GND - P1 K1 W1
VCCIO - - AK1 AU37 GND - P31 L2 W43
VCCIO - - AK4 AU41 GND - T4 M33 AB8
VCCIO - - AK33 AV14 GND - T28 P1 AB36
VCCIO - - AL2 AV30 GND - V1 P33 AE1
VCCIO - - AL3 BA7 GND - V31 R32 AE43
VCCIO - - AL31 BA19 GND - AC1 T1 AH6
VCCIO - - AM2 BA25 GND - AC31 V33 AH38
VCCIO - - AM15 BA37 GND - AE1 W2 AM2
VCCIO - - AM21 BC1 GND - AE31 Y1 AM42
VCCIO - - AM32 BC13 GND - AH16 Y33 AP6
VCCIO - - AN4 BC31 GND - AJ1 AB1 AP38
VCCIO - - AN8 BC43 GND - AJ31 AC32 AT22
VCCIO - - AN12 - GND - AK1 AD33 AV10
VCCIO - - AN18 - GND - AK2 AE2 AV16
VCCIO - - AN24 - GND - AK30 AG1 AV28
VCCIO - - AN30 - GND - AK31 AG32 AV34
GND - AL2 AH2 AW1
VCCINT - - - H12 GND - AL3 AJ33 AW5
VCCINT - - - H18 GND - AL7 AL32 AW39
VCCINT - - - H26 GND - AL9 AM3 AW43
VCCINT - - - H32 GND - AL14 AM11 BB12
VCCINT - - - M8 GND - AL18 AM19 BB32
VCCINT - - - M36 GND - AL23 AM25 BC5
VCCINT - - - V8 GND - AL25 AM28 BC19
VCCINT - - - V36 GND - AL29 AM33 BC25
VCCINT - - - AF8 GND - AL30 AM7 BC39

R
EPIC
Pad #
GND - - AN2 -
GND - - AN5 -
GND - - AN10 -
GND - - AN14 -
GND - - AN16 -
GND - - AN20 -
GND - - AN22 -
GND - - AN27 -
NC - C8 A1 -
NC - - A33 -
NC - - AC2 -
NC - - AN1 -
NC - - AN33 -
12/21/98


R
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

R
Table of Contents
XC3000 Series Field Programmable Gate Arrays

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
XC3000 Series Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
Crystal Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18
Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-19
General XC3000 Series Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-34
Device Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-35
Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-36
XC3000A Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-41
XC3000L Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-47
XC3000 Series Pin Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-65

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-83
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-83
XC5200 Family Compared to XC4000/Spartan™ and XC3000 Series . . . . . . . . . . . . . . . . . . . . . . . . . 7-84
Architectural Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-85
VersaBlock Routing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-93
General Routing Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-94
Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-98
Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-104
XC5200 Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-127
Revisions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-155
0
R
XC3000 Series Table of Contents
0 7*

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
Additional XC3100A Features. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
Low-Voltage Versions Available . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
XC3000 Series Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
New XC3000 Series Compared to Original XC3000 Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
Configuration Memory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
I/O Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-7
Configurable Logic Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
Programmable Interconnect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-10
Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-19
Initialization Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-19
Configuration Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-21
Daisy Chain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-22
Special Configuration Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-23
Configuration Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-25 7
Master Parallel Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-27
Peripheral Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-29
Program Readback Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-33
General XC3000 Series Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-34
Device Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-35
Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-36
Power Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-36
Dynamic Power Consumption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-37
Power Consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-37
Permanently Dedicated Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-38
User I/O Pins That Can Have Special Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-38
Unrestricted User I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-39
Pin Functions During Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-40
XC3000A Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-41
XC3000A DC Characteristics Over Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-41
XC3000A Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-42
XC3000A Global Buffer Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 7-42
XC3000A CLB Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-43
XC3000A IOB Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-45
XC3000L Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-47
XC3000L DC Characteristics Over Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-47
XC3000L Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-48
XC3000L Global Buffer Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 7-48
XC3000L CLB Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-49
7-1
R
XC3000L IOB Switching Characteristics Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-51

XC3100A Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-53
XC3100A DC Characteristics Over Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-53
XC3100A Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-54
XC3100A Global Buffer Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 7-54
XC3100A CLB Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-55
XC3100A IOB Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-57
XC3100L Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-59
XC3100L DC Characteristics Over Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-59
XC3100L Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-60
XC3100L Global Buffer Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 7-60
XC3100L CLB Switching Characteristics Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-61
XC3100L IOB Switching Characteristics Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-63
XC3000 Series Pin Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-65
XC3000 Series 44-Pin PLCC Pinouts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-65
XC3000 Series 64-Pin Plastic VQFP Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-66
XC3000 Series 68-Pin PLCC, 84-Pin PLCC and PGA Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . 7-67
XC3064A/XC3090A/XC3195A 84-Pin PLCC Pinouts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-68
XC3000 Series 100-Pin QFP Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-69
XC3000 Series 132-Pin Ceramic and Plastic PGA Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-70
XC3000 Series 144-Pin Plastic TQFP Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-71
XC3000 Series 160-Pin PQFP Pinouts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-72
XC3000 Series 175-Pin Ceramic and Plastic PGA Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-73
XC3000 Series 176-Pin TQFP Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-74
XC3000 Series 208-Pin PQFP Pinouts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-75
XC3195A PQ208 Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-76
Number of Available I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-78
7-2
0
XC3000 Series
R
(XC3000A/L, XC3100A/L)
November 9, 1998 (Version 3.1) 0 7* Product Description
Features • Complete Development System

- Schematic capture, automatic place and route
• Complete line of four related Field Programmable Gate - Logic and timing simulation
Array product families - Interactive design editor for design optimization
- XC3000A, XC3000L, XC3100A, XC3100L - Timing calculator
• Ideal for a wide range of custom VLSI design tasks - Interfaces to popular design environments like
- Replaces TTL, MSI, and other PLD logic Viewlogic, Cadence, Mentor Graphics, and others
- Integrates complete sub-systems into a single
package Additional XC3100A Features
- Avoids the NRE, time delay, and risk of conventional
• Ultra-high-speed FPGA family with six members
masked gate arrays
- 50-85 MHz system clock rates
• High-performance CMOS static memory technology
- 190 to 370 MHz guaranteed flip-flop toggle rates
- Guaranteed toggle rates of 70 to 370 MHz, logic
- 1.55 to 4.1 ns logic delays
delays from 7 to 1.5 ns
• High-end additional family member in the 22 X 22 CLB
- System clock speeds over 85 MHz
array-size XC3195A device
- Low quiescent and active power consumption
• 8 mA output sink current and 8 mA source current
• Flexible FPGA architecture
• Maximum power-down and quiescent current is 5 mA
- Compatible arrays ranging from 1,000 to 7,500 gate
• 100% architecture and pin-out compatible with other
complexity
XC3000 families
- Extensive register, combinatorial, and I/O 7
• Software and bitstream compatible with the XC3000,
capabilities
XC3000A, and XC3000L families
- High fan-out signal distribution, low-skew clock nets
- Internal 3-state bus capabilities XC3100A combines the features of the XC3000A and
- TTL or CMOS input thresholds XC3100 families:
- On-chip crystal oscillator amplifier • Additional interconnect resources for TBUFs and CE
• Unlimited reprogrammability inputs
- Easy design iteration • Error checking of the configuration bitstream
- In-system logic changes • Soft startup holds all outputs slew-rate limited during
• Extensive packaging options initial power-up
- Over 20 different packages • More advanced CMOS process
- Plastic and ceramic surface-mount and pin-grid-
array packages Low-Voltage Versions Available
- Thin and Very Thin Quad Flat Pack (TQFP and
• Low-voltage devices function at 3.0 - 3.6 V
VQFP) options
• XC3000L - Low-voltage versions of XC3000A devices
• Ready for volume production
• XC3100L - Low-voltage versions of XC3100A devices
- Standard, off-the-shelf product availability
- 100% factory pre-tested devices
- Excellent reliability record
Max Logic Typical Gate User I/Os Horizontal Configuration
Device CLBs Array Flip-Flops
Gates Range Max Longlines Data Bits
XC3020A, 3020L, 3120A 1,500 1,000 - 1,500 64 8x8 64 256 16 14,779
XC3030A, 3030L, 3130A 2,000 1,500 - 2,000 100 10 x 10 80 360 20 22,176
XC3042A, 3042L, 3142A, 3142L 3,000 2,000 - 3,000 144 12 x 12 96 480 24 30,784
XC3064A, 3064L, 3164A 4,500 3,500 - 4,500 224 16 x 14 120 688 32 46,064
XC3090A, 3090L, 3190A, 3190L 6,000 5,000 - 6,000 320 16 x 20 144 928 40 64,160
XC3195A 7,500 6,500 - 7,500 484 22 x 22 176 1,320 44 94,984

R
Introduction Here is a simple overview of those XC3000 products cur-

rently emphasized:
XC3000-Series Field Programmable Gate Arrays (FPGAs)
provide a group of high-performance, high-density, digital • XC3000A Family — The XC3000A is an enhanced
integrated circuits. Their regular, extendable, flexible, version of the basic XC3000 family, featuring additional
user-programmable array architecture is composed of a interconnect resources and other user-friendly
configuration program store plus three types of config- enhancements.
urable elements: a perimeter of I/O Blocks (IOBs), a core • XC3000L Family — The XC3000L is identical in
array of Configurable Logic Bocks (CLBs) and resources architecture and features to the XC3000A family, but
for interconnection. The general structure of an FPGA is operates at a nominal supply voltage of 3.3 V. The
shown in Figure 2. The development system provides XC3000L is the right solution for battery-operated and
schematic capture and auto place-and-route for design low-power applications.
entry. Logic and timing simulation, and in-circuit emulation • XC3100A Family — The XC3100A is a
are available as design verification alternatives. The design performance-optimized relative of the XC3000A family.
editor is used for interactive design optimization, and to While both families are bitstream and footprint
compile the data pattern that represents the configuration compatible, the XC3100A family extends toggle rates to
program. 370 MHz and in-system performance to over 80 MHz.
The XC3100A family also offers one additional array
The FPGA user logic functions and interconnections are size, the XC3195A.
determined by the configuration program data stored in • XC3100L Family — The XC3100L is identical in
internal static memory cells. The program can be loaded in architectures and features to the XC3100A family, but
any of several modes to accommodate various system operates at a nominal supply voltage of 3.3V.
requirements. The program data resides externally in an
EEPROM, EPROM or ROM on the application circuit Figure 1 illustrates the relationships between the families.
board, or on a floppy disk or hard disk. On-chip initialization Compared to the original XC3000 family, XC3000A offers
logic provides for optional automatic loading of program additional functionality and increased speed. The XC3000L
data at power-up. The companion XC17XX Serial Configu- family offers the same additional functionality, but reduced
ration PROMs provide a very simple serial configuration speed due to its lower supply voltage of 3.3 V. The
program storage in a one-time programmable package. XC3100A family offers substantially higher speed and
higher density with the XC3195A.
The XC3000 Field Programmable Gate Array families pro-
vide a variety of logic capacities, package styles, tempera- New XC3000 Series Compared to Original
ture ranges and speed grades. XC3000 Family
XC3000 Series Overview For readers already familiar with the original XC3000 family
of FPGAs, the major new features in the XC3000A,
There are now four distinct family groupings within the
XC3000L, XC3100A, and XC3100L families are listed in
XC3000 Series of FPGA devices:
this section.
• XC3000A Family
All of these new families are upward-compatible extensions
• XC3000L Family
of the original XC3000 FPGA architecture. Any bitstream
• XC3100A Family
used to configure an XC3000 device will configure the cor-
• XC3100L Family
responding XC3000A, XC3000L, XC3100A, or XC3100L
All four families share a common architecture, develop- device exactly the same way.
ment software, design and programming methodology, and
The XC3100A and XC3100L FPGA architectures are
also common package pin-outs. An extensive Product
upward-compatible extensions of the XC3000A and
Description covers these common aspects.
XC3000L architectures. Any bitstream used to configure an
Detailed parametric information for the XC3000A, XC3000A or XC3000L device will configure the corre-
XC3000L, XC3100A, and XC3100L product families is then sponding XC3100A or XC3100L device exactly the same
provided. (The XC3000 and XC3100 families are not rec- way.
ommended for new designs.)

R
Improvements in the XC3000A and XC3000L

Families
nality
The XC3000A and XC3000L families offer the following Functio
enhancements over the popular XC3000 family:

0A
The XC3000A and XC3000L families have additional inter- XC310
0 00L
connect resources to drive the I-inputs of TBUFs driving XC310XC31
horizontal Longlines. The CLB Clock Enable input can be
0A
driven from a second vertical Longline. These two additions XC300
0L
result in more efficient and faster designs when horizontal XC300
Longlines are used for data bussing. Speed
During configuration, the XC3000A and XC3000L devices

check the bit-stream format for stop bits in the appropriate
positions. Any error terminates the configuration and pulls
INIT Low.
)
195A
When the configuration process is finished and the device (XC3
starts up in user mode, the first activation of the outputs is

automatically slew-rate limited. This feature, called Soft city
Capa
Startup, avoids the potential ground bounce when all Gate
out-puts are turned on simultaneously. After start-up, the

X7068
slew rate of the individual outputs is, as in the XC3000 fam-
ily, determined by the individual configuration option. Figure 1: XC3000 FPGA Families
Improvements in the XC3100A and XC3100L

Families
Based on a more advanced CMOS process, the XC3100A 7
and XC3100L families are architecturally-identical, perfor-
mance-optimized relatives of the XC3000A and XC3000L
families. While all families are footprint compatible, the
XC3100A family extends achievable system performance
beyond 85 MHz.

R
Detailed Functional Description

The perimeter of configurable Input/Output Blocks (IOBs) data may be either bit serial or byte parallel. The develop-
provides a programmable interface between the internal ment system generates the configuration program bit-
logic array and the device package pins. The array of Con- stream used to configure the device. The memory loading
figurable Logic Blocks (CLBs) performs user-specified logic process is independent of the user logic functions.
functions. The interconnect resources are programmed to
form networks, carrying logic signals among blocks, analo- Configuration Memory
gous to printed circuit board traces connecting MSI/SSI The static memory cell used for the configuration memory
packages. in the Field Programmable Gate Array has been designed
The block logic functions are implemented by programmed specifically for high reliability and noise immunity. Integrity
look-up tables. Functional options are implemented by pro- of the device configuration memory based on this design is
gram-controlled multiplexers. Interconnecting networks assured even under adverse conditions. As shown in
between blocks are implemented with metal segments Figure 3, the basic memory cell consists of two CMOS
joined by program-controlled pass transistors. inverters plus a pass transistor used for writing and reading
cell data. The cell is only written during configuration and
These FPGA functions are established by a configuration
only read during readback. During normal operation, the
program which is loaded into an internal, distributed array
cell provides continuous control and the pass transistor is
of configuration memory cells. The configuration program
off and does not affect cell stability. This is quite different
is loaded into the device at power-up and may be reloaded
from the operation of conventional memory devices, in
on command. The FPGA includes logic and control signals
which the cells are frequently read and rewritten.
to implement automatic or passive configuration. Program
PWR P9 P8 P7 P6 P5 P4 P3 P2 GND
DN
I/O Blocks
P11 3-State Buffers With Access Configurable Logic

to Horizontal Long Lines Blocks
TCL
KIN
AA AB AC AD
P12
Interconnect Area
P13
BA BB
Frame Pointer
U61
X3241
Figure 2: Field Programmable Gate Array Structure.

It consists of a perimeter of programmable I/O blocks, a core of configurable logic blocks and their interconnect resources.
These are all controlled by the distributed array of configuration program memory cells.

R
testing, no soft errors have been observed even in the

presence of very high doses of alpha radiation.
Q
Configuration The method of loading the configuration data is selectable.
Control Two methods use serial data, while three use byte-wide
Read or Q
Write data. The internal configuration logic utilizes framing infor-
mation, embedded in the program data by the development
Data
system, to direct memory-cell loading. The serial-data
X5382 framing and length-count preamble provide programming
Figure 3: Static Configuration Memory Cell. compatibility for mixes of various FPGA device devices in a
It is loaded with one bit of configuration program and con- synchronous, serial, daisy-chain fashion.
trols one program selection in the Field Programmable
Gate Array.
I/O Block
Each user-configurable IOB shown in Figure 4, provides an
interface between the external package pin of the device
The memory cell outputs Q and Q use ground and VCC lev- and the internal user logic. Each IOB includes both regis-
els and provide continuous, direct control. The additional tered and direct input paths. Each IOB provides a program-
capacitive load together with the absence of address mable 3-state output buffer, which may be driven by a
decoding and sense amplifiers provide high stability to the registered or direct output signal. Configuration options
cell. Due to the structure of the configuration memory cells, allow each IOB an inversion, a controlled slew rate and a
they are not affected by extreme power-supply excursions high impedance pull-up. Each input circuit also provides
or very high levels of alpha particle radiation. In reliability input clamping diodes to provide electrostatic protection,
and circuits to inhibit latch-up produced by input currents.
Vcc
PROGRAM-CONTROLLED MEMORY CELLS
OUT 3-STATE OUTPUT SLEW PASSIVE 7

INVERT INVERT SELECT RATE PULL UP
3- STATE T
(OUTPUT ENABLE)
O D Q OUTPUT
OUT
BUFFER
FLIP
FLOP
I/O PAD
R
I
DIRECT IN
Q
REGISTERED IN Q D
FLIP TTL or
FLOP CMOS
or INPUT
LATCH THRESHOLD
OK IK (GLOBAL RESET)
CK1
CK2
PROGRAM
CONTROLLED
MULTIPLEXER = PROGRAMMABLE INTERCONNECTION POINT or PIP X3029
Figure 4: Input/Output Block.

Each IOB includes input and output storage elements and I/O options selected by configuration memory cells. A choice
of two clocks is available on each die edge. The polarity of each clock line (not each flip-flop or latch) is programmable.
A clock line that triggers the flip-flop on the rising edge is an active Low Latch Enable (Latch transparent) signal and vice
versa. Passive pull-up can only be enabled on inputs, not on outputs. All user inputs are programmed for TTL or CMOS
thresholds.

R
The input-buffer portion of each IOB provides threshold output and 3-state signal nets so that the buffer output is
detection to translate external signals applied to the pack- enabled only for a Low.
age pin to internal logic levels. The global input-buffer
Configuration program bits for each IOB control features
threshold of the IOBs can be programmed to be compatible
such as optional output register, logic signal inversion, and
with either TTL or CMOS levels. The buffered input signal
3-state and slew-rate control of the output.
drives the data input of a storage element, which may be
configured as either a flip-flop or a latch. The clocking The program-controlled memory cells of Figure 4 control
polarity (rising/falling edge-triggered flip-flop, High/Low the following options.
transparent latch) is programmable for each of the two • Logic inversion of the output is controlled by one
clock lines on each of the four die edges. Note that a clock configuration program bit per IOB.
line driving a rising edge-triggered flip-flop makes any latch • Logic 3-state control of each IOB output buffer is
driven by the same line on the same edge Low-level trans- determined by the states of configuration program bits
parent and vice versa (falling edge, High transparent). All that turn the buffer on, or off, or select the output buffer
Xilinx primitives in the supported schematic-entry pack- 3-state control interconnection (IOB pin T). When this
ages, however, are positive edge-triggered flip-flops or IOB output control signal is High, a logic one, the buffer
High transparent latches. When one clock line must drive is disabled and the package pin is high impedance.
flip-flops as well as latches, it is necessary to compensate When this IOB output control signal is Low, a logic zero,
for the difference in clocking polarities with an additional the buffer is enabled and the package pin is active.
inverter either in the flip-flop clock input or the latch-enable Inversion of the buffer 3-state control-logic sense
input. I/O storage elements are reset during configuration (output enable) is controlled by an additional
or by the active-Low chip RESET input. Both direct input configuration program bit.
(from IOB pin I) and registered input (from IOB pin Q) sig- • Direct or registered output is selectable for each IOB.
nals are available for interconnect. The register uses a positive-edge, clocked flip-flop. The
For reliable operation, inputs should have transition times clock source may be supplied (IOB pin OK) by either of
of less than 100 ns and should not be left floating. Floating two metal lines available along each die edge. Each of
CMOS input-pin circuits might be at threshold and produce these lines is driven by an invertible buffer.
oscillations. This can produce additional power dissipation • Increased output transition speed can be selected to
and system noise. A typical hysteresis of about 300 mV improve critical timing. Slower transitions reduce
reduces sensitivity to input noise. Each user IOB includes a capacitive-load peak currents of non-critical outputs
programmable high-impedance pull-up resistor, which may and minimize system noise.
be selected by the program to provide a constant High for • An internal high-impedance pull-up resistor (active by
otherwise undriven package pins. Although the Field Pro- default) prevents unconnected inputs from floating.
grammable Gate Array provides circuitry to provide input Unlike the original XC3000 series, the XC3000A,
protection for electrostatic discharge, normal CMOS han- XC3000L, XC3100A, and XC3100L families include the
dling precautions should be observed. Soft Startup feature. When the configuration process is fin-
Flip-flop loop delays for the IOB and logic-block flip-flops ished and the device starts up in user mode, the first activa-
are short, providing good performance under asynchro- tion of the outputs is automatically slew-rate limited. This
nous clock and data conditions. Short loop delays minimize feature avoids potential ground bounce when all outputs
the probability of a metastable condition that can result are turned on simultaneously. After start-up, the slew rate
from assertion of the clock during data transitions. Because of the individual outputs is determined by the individual
of the short-loop-delay characteristic in the Field Program- configuration option.
mable Gate Array, the IOB flip-flops can be used to syn-
Summary of I/O Options
chronize external signals applied to the device. Once
• Inputs
synchronized in the IOB, the signals can be used internally
- Direct
without further consideration of their clock relative timing,
- Flip-flop/latch
except as it applies to the internal logic and routing-path
- CMOS/TTL threshold (chip inputs)
delays.
- Pull-up resistor/open circuit
IOB output buffers provide CMOS-compatible 4-mA • Outputs
source-or-sink drive for high fan-out CMOS or TTL- com- - Direct/registered
patible signal levels (8 mA in the XC3100A family). The net- - Inverted/not
work driving IOB pin O becomes the registered or direct - 3-state/on/off
data source for the output buffer. The 3-state control signal - Full speed/slew limited
(IOB) pin T can control output activity. An open-drain output - 3-state/output enable (inverse)
may be obtained by using the same signal for driving the

R
Configurable Logic Block resources adjacent to the blocks. Each CLB also has two
outputs (X and Y) which may drive interconnect networks.
The array of CLBs provides the functional elements from
which the user’s logic is constructed. The logic blocks are Data input for either flip-flop within a CLB is supplied from
arranged in a matrix within the perimeter of IOBs. For the function F or G outputs of the combinatorial logic, or the
example, the XC3020A has 64 such blocks arranged in 8 block input, DI. Both flip-flops in each CLB share the asyn-
rows and 8 columns. The development system is used to chronous RD which, when enabled and High, is dominant
compile the configuration data which is to be loaded into over clocked inputs. All flip-flops are reset by the
the internal configuration memory to define the operation active-Low chip input, RESET, or during the configuration
and interconnection of each block. User definition of CLBs process. The flip-flops share the enable clock (EC) which,
and their interconnecting networks may be done by auto- when Low, recirculates the flip-flops’ present states and
matic translation from a schematic-capture logic diagram or inhibits response to the data-in or combinatorial function
optionally by installing library or user macros. inputs on a CLB. The user may enable these control inputs
and select their sources. The user may also select the
Each CLB has a combinatorial logic section, two flip-flops,
clock net input (K), as well as its active sense within each
and an internal control section. See Figure 5. There are:
CLB. This programmable inversion eliminates the need to
five logic inputs (A, B, C, D and E); a common clock input route both phases of a clock signal throughout the device.
(K); an asynchronous direct RESET input (RD); and an
enable clock (EC). All may be driven from the interconnect
DI
DATA IN
0
MUX D Q
F 1
DIN
G
QX
RD
QX X
A
B
F F 7
LOGIC C COMBINATORIAL
FUNCTION CLB OUTPUTS
VARIABLES D
E
G G
QY Y
F QY
DIN
G 0
MUX D Q
1
EC
ENABLE CLOCK RD
1 (ENABLE)
K
CLOCK
DIRECT RD
RESET
0 (INHIBIT)
(GLOBAL RESET)
X3032
Figure 5: Configurable Logic Block.

Each CLB includes a combinatorial logic section, two flip-flops and a program memory controlled multiplexer selection of
function. It has the following:
- five logic variable inputs A, B, C, D, and E
- a direct data in DI
- an enable clock EC
- a clock (invertible) K
- an asynchronous direct RESET RD
- two outputs X and Y

R
Flexible routing allows use of common or individual CLB

clocking.
A
The combinatorial-logic portion of the CLB uses a 32 by 1 B
QX
look-up table to implement Boolean functions. Variables Any Function
QY F
selected from the five logic inputs and two internal block of Up to 4
Variables
flip-flops are used as table address inputs. The combinato- C
D
rial propagation delay through the network is independent
of the logic function generated and is spike free for single E
input variable changes. This technique can generate two

A
independent logic functions of up to four variables each as B
QX
shown in Figure 6a, or a single function of five variables as
Any Function
shown in Figure 6b, or some functions of seven variables QY of Up to 4 G
Variables
as shown in Figure 6c. Figure 7 shows a modulo-8 binary C
D
counter with parallel enable. It uses one CLB of each type.
The partial functions of six or seven variables are imple- E 5a
mented using the input variable (E) to dynamically select
between two functions of four different variables. For the A
B
two functions of four variables each, the independent QX F
results (F and G) may be used as data inputs to either Any Function
QY of 5 Variables
flip-flop or either logic block output. For the single function
C G
of five variables and merged functions of six or seven vari- D
ables, the F and G outputs are identical. Symmetry of the F E
5b
and G functions and the flip-flops allows the interchange of
CLB outputs to optimize routing efficiencies of the networks
A
interconnecting the CLBs and IOBs. B
QX
Programmable Interconnect QY
Any Function
of Up to 4
Variables
Programmable-interconnection resources in the Field Pro- C
grammable Gate Array provide routing paths to connect D

F
inputs and outputs of the IOBs and CLBs into logic net- M
U
works. Interconnections between blocks are composed of a X
A G
two-layer grid of metal segments. Specially designed pass B
QX
transistors, each controlled by a configuration bit, form pro-
Any Function
grammable interconnect points (PIPs) and switching matri- QY of Up to 4
Variables
ces used to implement the necessary connections between C
selected metal segments and block pins. Figure 8 is an D
example of a routed net. The development system provides E 5c
FGM
Mode
automatic routing of these interconnections. Interactive
X5442
routing is also available for design optimization. The inputs
of the CLBs or IOBs are multiplexers which can be pro- Figure 6: Combinational Logic Options
grammed to select an input network from the adjacent 6a. Combinatorial Logic Option FG generates two func-
interconnect segments. Since the switch connections to tions of four variables each. One variable, A, must be
block inputs are unidirectional, as are block outputs, common to both functions. The second and third variable
they are usable only for block input connection and not can be any choice of B, C, QX and QY. The fourth vari-
for routing. Figure 9 illustrates routing access to logic able can be any choice of D or E.
block input variables, control inputs and block outputs. 6b. Combinatorial Logic Option F generates any function
Three types of metal resources are provided to accommo- of five variables: A, D, E and two choices out of B, C, QX,
date various network interconnect requirements. QY.
6c. Combinatorial Logic Option FGM allows variable E to
• General Purpose Interconnect select between two functions of four variables: Both have
• Direct Connection common inputs A and D and any choice out of B, C, QX
• Longlines (multiplexed busses and wide AND gates) and QY for the remaining two variables. Option 3 can
then implement some functions of six or seven variables.

R
Count Enable Terminal

Parallel Enable Count
Clock
Dual Function of 4 Variables
D Q Q0
D0
FG
Mode
D Q Q1
D1
Function of 5 Variables
F
Mode
D Q Q2
D2 Figure 8: A Design Editor view of routing resources
used to form a typical interconnection network from
Function of 6 Variables
FGM
CLB GA.
Mode
X5383
Figure 7: Counter. and to the right. The other PIPs adjacent to the matrices
The modulo-8 binary counter with parallel enable and are accessed to or from Longlines. The development sys-
clock enable uses one combinatorial logic block of each tem automatically defines the buffer direction based on the 7
option. location of the interconnection network source. The delay
calculator of the development system automatically calcu-
lates and displays the block, interconnect and buffer delays
for any paths selected. Generation of the simulation netlist
General Purpose Interconnect
with a worst-case delay model is provided.
General purpose interconnect, as shown in Figure 10, con-
sists of a grid of five horizontal and five vertical metal seg- Direct Interconnect
ments located between the rows and columns of logic and Direct interconnect, shown in Figure 12, provides the most
IOBs. Each segment is the height or width of a logic block. efficient implementation of networks between adjacent
Switching matrices join the ends of these segments and CLBs or I/O Blocks. Signals routed from block to block
allow programmed interconnections between the metal grid using the direct interconnect exhibit minimum interconnect
segments of adjoining rows and columns. The switches of propagation and use no general interconnect resources.
an unprogrammed device are all non-conducting. The con- For each CLB, the X output may be connected directly to
nections through the switch matrix may be established by the B input of the CLB immediately to its right and to the C
the automatic routing or by selecting the desired pairs of input of the CLB to its left. The Y output can use direct inter-
matrix pins to be connected or disconnected. The legiti- connect to drive the D input of the block immediately above
mate switching matrix combinations for each pin are indi- and the A input of the block below. Direct interconnect
cated in Figure 11. should be used to maximize the speed of high-performance
Special buffers within the general interconnect areas pro- portions of logic. Where logic blocks are adjacent to IOBs,
vide periodic signal isolation and restoration for improved direct connect is provided alternately to the IOB inputs (I)
performance of lengthy nets. The interconnect buffers are and outputs (O) on all four edges of the die. The right edge
available to propagate signals in either direction on a given provides additional direct connects from CLB outputs to
general interconnect segment. These bidirectional (bidi) adjacent IOBs. Direct interconnections of IOBs with CLBs
buffers are found adjacent to the switching matrices, above are shown in Figure 13.

R
Figure 9: Design Editor Locations of interconnect access, CLB control inputs, logic inputs and outputs. The dot pattern
represents the available programmable interconnection points (PIPs).
Some of the interconnect PIPs are directional.

R
Figure 12: CLB X and Y Outputs.

Figure 10: FPGA General-Purpose Interconnect.
The X and Y outputs of each CLB have single contact,
Composed of a grid of metal segments that may be inter-
direct access to inputs of adjacent CLBs
connected through switch matrices to form networks for
CLB and IOB inputs and outputs.
7
Figure 11: Switch Matrix Interconnection Options for

Each Pin.
Switch matrices on the edges are different.

R
Global Buffer Direct Input Global Buffer Inerconnect
* Unbonded IOBs (6 Places) Alternate Buffer Direct Input
Figure 13: XC3020A Die-Edge IOBs. The XC3020A die-edge IOBs are provided with direct access to adjacent CLBs.

R
Longlines umn are connectable half-length lines. On the XC3020A

and XC3120A FPGAs, only the outer Longlines are con-
The Longlines bypass the switch matrices and are intended
nectable half-length lines.
primarily for signals that must travel a long distance, or
must have minimum skew among multiple destinations. Longlines can be driven by a logic block or IOB output on a
Longlines, shown in Figure 14, run vertically and horizon- column-by-column basis. This capability provides a com-
tally the height or width of the interconnect area. Each inter- mon low skew control or clock line within each column of
connection column has three vertical Longlines, and each logic blocks. Interconnections of these Longlines are
interconnection row has two horizontal Longlines. Two shown in Figure 15. Isolation buffers are provided at each
additional Longlines are located adjacent to the outer sets input to a Longline and are enabled automatically by the
of switching matrices. In devices larger than the XC3020A development system when a connection is made.
and XC3120A FPGAs, two vertical Longlines in each col-
Figure 14: Horizontal and Vertical Longlines. These Longlines provide high fan-out, low-skew signal distribution in
each row and column. The global buffer in the upper left die corner drives a common line throughout the FPGA.

R
Figure 15: Programmable Interconnection of Longlines. This is provided at the edges of the routing area.
Three-state buffers allow the use of horizontal Longlines to form on-chip wired AND and multiplexed buses. The left two
non-clock vertical Longlines per column (except XC3020A) and the outer perimeter Longlines may be programmed as
connectable half-length lines.
VCC VCC
Z = DA • DB • DC • ... • DN
(LOW)
DA DB DC DN X3036
Figure 16: 3-State Buffers Implement a Wired-AND Function. When all the buffer 3-state lines are High, (high
impedance), the pull-up resistor(s) provide the High output. The buffer inputs are driven by the control signals or a Low.
Z = DA • A + DB • B + DC • C + … + DN • N
DA DB DC DN
WEAK
A B C N
KEEPER CIRCUIT X1741A
Figure 17: 3-State Buffers Implement a Multiplexer. The selection is accomplished by the buffer 3-state signal.

R
A buffer in the upper left corner of the FPGA chip drives a of the 3-state buffer controls allows them to implement wide
global net which is available to all K inputs of logic blocks. multiplexing functions. Any 3-state buffer input can be
Using the global buffer for a clock signal provides a selected as drive for the horizontal long-line bus by apply-
skew-free, high fan-out, synchronized clock for use at any ing a Low logic level on its 3-state control line. See
or all of the IOBs and CLBs. Configuration bits for the K Figure 16. The user is required to avoid contention which
input to each logic block can select this global line or can result from multiple drivers with opposing logic levels.
another routing resource as the clock source for its Control of the 3-state input by the same signal that drives
flip-flops. This net may also be programmed to drive the die the buffer input, creates an open-drain wired-AND function.
edge clock lines for IOB use. An enhanced speed, CMOS A logic High on both buffer inputs creates a high imped-
threshold, direct access to this buffer is available at the sec- ance, which represents no contention. A logic Low enables
ond pad from the top of the left die edge. the buffer to drive the Longline Low. See Figure 17. Pull-up
resistors are available at each end of the Longline to pro-
A buffer in the lower right corner of the array drives a hori-
vide a High output when all connected buffers are non-con-
zontal Longline that can drive programmed connections to
ducting. This forms fast, wide gating functions. When data
a vertical Longline in each interconnection column. This
drives the inputs, and separate signals drive the 3-state
alternate buffer also has low skew and high fan-out. The
control lines, these buffers form multiplexers (3-state bus-
network formed by this alternate buffer’s Longlines can be
ses). In this case, care must be used to prevent contention
selected to drive the K inputs of the CLBs. CMOS thresh-
through multiple active buffers of conflicting levels on a
old, high speed access to this buffer is available from the
common line. Each horizontal Longline is also driven by a
third pad from the bottom of the right die edge.
weak keeper circuit that prevents undefined floating levels
Internal Busses by maintaining the previous logic level when the line is not
driven by an active buffer or a pull-up resistor. Figure 18
A pair of 3-state buffers, located adjacent to each CLB, per-
shows 3-state buffers, Longlines and pull-up resistors.
mits logic to drive the horizontal Longlines. Logic operation
3 VERTICAL LONG
BIDIRECTIONAL GLOBAL NET LINES PER COLUMN
INTERCONNECT 7
BUFFERS
I/O CLOCKS
GG GH
P48
HORIZONTAL LONG LINE

PULL-UP RESISTOR
HORIZONTAL LONG LINE
OSCILLATOR
AMPLIFIER OUTPUT
P47 DIRECTINPUT OF P47

TO AUXILIARY BUFFER
BCL
HG HH KIN CRYSTAL OSCILLATOR
BUFFER
3-STATE INPUT
OS
C
3-STATE CONTROL
P46
.l .lk
.q
.ck
.Q
D 3-STATE BUFFER
P
G
M
ALTERNATE BUFFER
P40 P41 P42 P43 RST
X1245
Figure 18: Design Editor.

An extra large view of possible interconnections in the lower right corner of the XC3020A.

R
Crystal Oscillator series resistor R2 may be included to add to the amplifier

output impedance when needed for phase-shift control,
Figure 18 also shows the location of an internal high speed crystal resistance matching, or to limit the amplifier input
inverting amplifier that may be used to implement an swing to control clipping at large amplitudes. Excess feed-
on-chip crystal oscillator. It is associated with the auxiliary back voltage may be corrected by the ratio of C2/C1. The
buffer in the lower right corner of the die. When the oscilla- amplifier is designed to be used from 1 MHz to about
tor is configured and connected as a signal source, two one-half the specified CLB toggle frequency. Use at fre-
special user IOBs are also configured to connect the oscil- quencies below 1 MHz may require individual characteriza-
lator amplifier with external crystal oscillator components tion with respect to a series resistance. Crystal oscillators
as shown in Figure 19. A divide by two option is available to above 20 MHz generally require a crystal which operates in
assure symmetry. The oscillator circuit becomes active a third overtone mode, where the fundamental frequency
early in the configuration process to allow the oscillator to must be suppressed by an inductor across C2, turning this
stabilize. Actual internal connection is delayed until com- parallel resonant circuit to double the fundamental crystal
pletion of configuration. In Figure 19 the feedback resistor frequency, i.e., 2/3 of the desired third harmonic frequency
R1, between the output and input, biases the amplifier at network. When the oscillator inverter is not used, these
threshold. The inversion of the amplifier, together with the IOBs and their package pins are available for general user
R-C networks and an AT-cut series resonant crystal, pro- I/O.
duce the 360-degree phase shift of the Pierce oscillator. A
D Q
Internal External
Alternate XTAL1
Clock Buffer
XTAL2
(IN)
R1
Suggested Component Values

R1 0.5 – 1 MΩ R2
R2 0 – 1 kΩ
(may be required for low frequency, phase Y1
shift and/or compensation level for crystal Q)
C1, C2 10 – 40 pF C1 C2
Y1 1 – 20 MHz AT-cut parallel resonant
44 PIN 68 PIN 84 PIN 100 PIN 132 PIN 160 PIN 164 PIN 175 PIN 176 PIN 208 PIN
PLCC PLCC PLCC PGA CQFP PQFP PGA PQFP CQFP PGA TQFP PQFP
XTAL 1 (OUT) 30 47 57 J11 67 82 P13 82 105 T14 91 110
XTAL 2 (IN) 26 43 53 L11 61 76 M13 76 99 P15 85 100
X7064
Figure 19: Crystal Oscillator Inverter. When activated, and by selecting an output network for its buffer, the crystal
oscillator inverter uses two unconfigured package pins and external components to implement an oscillator. An optional
divide-by-two mode is available to assure symmetry.

R
Configuration In Master configuration modes, the device becomes the

source of the Configuration Clock (CCLK). The beginning
Initialization Phase of configuration of devices using Peripheral or Slave
modes must be delayed long enough for their initialization
An internal power-on-reset circuit is triggered when power to be completed. An FPGA with mode lines selecting a
is applied. When VCC reaches the voltage at which portions
Master configuration mode extends its initialization state
of the FPGA device begin to operate (nominally 2.5 to 3 V),
using four times the delay (43 to 130 ms) to assure that all
the programmable I/O output buffers are 3-stated and a daisy-chained slave devices, which it may be driving, will
high-impedance pull-up resistor is provided for the user be ready even if the master is very fast, and the slave(s)
I/O pins. A time-out delay is initiated to allow the power very slow. Figure 20 shows the state sequences. At the end
supply voltage to stabilize. During this time the power-down of Initialization, the device enters the Clear state where it
mode is inhibited. The Initialization state time-out (about 11
clears the configuration memory. The active Low,
to 33 ms) is determined by a 14-bit counter driven by a
open-drain initialization signal INIT indicates when the Ini-
self-generated internal timer. This nominal 1-MHz timer is tialization and Clear states are complete. The FPGA tests
subject to variations with process, temperature and power for the absence of an external active Low RESET before it
supply. As shown in Table 1, five configuration mode makes a final sample of the mode lines and enters the Con-
choices are available as determined by the input levels of figuration state. An external wired-AND of one or more INIT
three mode pins; M0, M1 and M2.
pins can be used to control configuration by the assertion of
Table 1: Configuration Mode Choices the active-Low RESET of a master mode device or to sig-
nal a processor that the FPGAs are not yet initialized.
M0 M1 M2 CCLK Mode Data
0 0 0 output Master Bit Serial If a configuration has begun, a re-assertion of RESET for a
0 0 1 output Master Byte Wide Addr. = 0000 up minimum of three internal timer cycles will be recognized
0 1 0 — reserved — and the FPGA will initiate an abort, returning to the Clear
0 1 1 output Master Byte Wide Addr. = FFFF down state to clear the partially loaded configuration memory
1 0 0 — reserved — words. The FPGA will then resample RESET and the mode
1 0 1 output Peripheral Byte Wide lines before re-entering the Configuration state.
1 1 0 — reserved —
7
During configuration, the XC3000A, XC3000L, XC3100A,
1 1 1 input Slave Bit Serial and XC3100L devices check the bit-stream format for stop
bits in the appropriate positions. Any error terminates the
configuration and pulls INIT Low.
All User I/O Pins 3-Stated with High Impedance Pull-Up, HDC=High, LDC=Low
INIT Output = Low Power Down

No HDC, LDC
or Pull-Up
PWRDWN
Initialization Inactive
Power-On PWRDWN
Time Delay Active
Active RESET
Clear Test Configuration

Configuration RESET Start-Up Operational
No Mode Pins Program Mode Mode
Memory Active
Active RESET
Operates on
User Logic
Low on DONE/PROGRAM and RESET
Clear Is
~ 200 Cycles for the XC3020A—130 to 400 µs
Power-On Delay is ~ 250 Cycles for the XC3030A—165 to 500 µs
214 Cycles for Non-Master Mode—11 to 33 ms ~ 290 Cycles for the XC3042A—195 to 580 µs
216 Cycles for Master Mode—43 to 130 ms ~ 330 Cycles for the XC3064A—220 to 660 µs
~ 375 Cycles for the XC3090A—250 to 750 µs X3399
Figure 20: A State Diagram of the Configuration Process for Power-up and Reprogram.

R
A re-program is initiated.when a configured XC3000 series generated by the development system begins with a pre-
device senses a High-to-Low transition and subsequent >6 amble of 111111110010 followed by a 24-bit length count
µs Low level on the DONE/PROG package pin, or, if this representing the total number of configuration clocks
pin is externally held permanently Low, a High-to-Low tran- needed to complete loading of the configuration pro-
sition and subsequent >6 µs Low time on the RESET pack- gram(s). The data framing is shown in Figure 21. All
age pin. FPGAs connected in series read and shift preamble and
length count in on positive and out on negative configura-
The device returns to the Clear state where the configura-
tion clock edges. A device which has received the pream-
tion memory is cleared and mode lines re-sampled, as for
ble and length count then presents a High Data Out until it
an aborted configuration. The complete configuration pro-
has intercepted the appropriate number of data frames.
gram is cleared and loaded during each configuration pro-
When the configuration program memory of an FPGA is full
gram cycle.
and the length count does not yet compare, the device
Length count control allows a system of multiple Field Pro- shifts any additional data through, as it did for preamble
grammable Gate Arrays, of assorted sizes, to begin opera- and length count. When the FPGA configuration memory is
tion in a synchronized fashion. The configuration program full and the length count compares, the device will execute
11111111 —Dummy Bits*

0010 —Preamble Code Header
< 24-Bit Length Count > —Configuration Program Length
1111 —Dummy Bits (4 Bits Minimum)
0 <Data Frame # 001 > 111

0 <Data Frame # 002 > 111 For XC3120
0 <Data Frame # 003 > 111
. . . 197 Configuration Data Frames Program Data
. . .
. . . (Each Frame Consists of: Repeated for Each Logic
0 <Data Frame # 196 > 111 A Start Bit (0) Cell Array in a Daisy Chain
0 <Data Frame # 197 > 111 A 71-Bit Data Field
Three Stop Bits
1111 Postamble Code (4 Bits Minimum)

*The LCA Device Require Four Dummy Bits Min; Software Generates Eight Dummy Bits
X5300_01
XC3042A XC3090A
XC3020A XC3030A XC3042L XC3064A XC3090L
XC3020L XC3030L XC3142A XC3064L XC3190A
Device XC3120A XC3130A XC3142L XC3164A XC3190L XC3195A
Gates 1,000 to 1,500 1,500 to 2,000 2,000 to 3,000 3,500 to 4,500 5,000 to 6,000 6,500 to 7,500
CLBs 64 100 144 224 320 484
Row x Col (8 x 8) (10 x 10) (12 x 12) (16 x 14) (20 x 16) (22 x 22)
IOBs 64 80 96 120 144 176
Flip-flops 256 360 480 688 928 1,320
Horizontal Longlines 16 20 24 32 40 44
TBUFs/Horizontal LL 9 11 13 15 17 23
Bits per Frame 75 92 108 140 172 188
(including1 start and 3 stop bits)
Frames 197 241 285 329 373 505
Program Data = 14,779 22,176 30,784 46,064 64,160 94,944
Bits x Frames + 4 bits
(excludes header)
PROM size (bits) = 14,819 22,216 30,824 46,104 64,200 94,984
Program Data
+ 40-bit Header
Figure 21: Internal Configuration Data Structure for an FPGA. This shows the preamble, length count and data
frames generated by the Development System.
The Length Count produced by the program = [(40-bit preamble + sum of program data + 1 per daisy chain device)
rounded up to multiple of 8] – (2 ≤ K ≤ 4) where K is a function of DONE and RESET timing selected. An additional 8 is
added if roundup increment is less than K. K additional clocks are needed to complete start-up after length count is
reached.

R
a synchronous start-up sequence and become operational. Configuration Data

See Figure 22. Two CCLK cycles after the completion of
Configuration data to define the function and interconnec-
loading configuration data, the user I/O pins are enabled as
tion within a Field Programmable Gate Array is loaded from
configured. As selected, the internal user-logic RESET is
an external storage at power-up and after a re-program sig-
released either one clock cycle before or after the I/O pins
nal. Several methods of automatic and controlled loading of
become active. A similar timing selection is programmable
the required data are available. Logic levels applied to
for the DONE/PROG output signal. DONE/PROG may also
be programmed to be an open drain or include a pull-up mode selection pins at the start of configuration time deter-
resistor to accommodate wired ANDing. The High During mine the method to be used. See Table 1. The data may be
Configuration (HDC) and Low During Configuration (LDC) either bit-serial or byte-parallel, depending on the configu-
are two user I/O pins which are driven active while an ration mode. The different FPGAs have different sizes and
numbers of data frames. To maintain compatibility between
FPGA is in its Initialization, Clear or Configure states. They
various device types, the Xilinx product families use com-
and DONE/PROG provide signals for control of external
logic signals such as RESET, bus enable or PROM enable patible configuration formats. For the XC3020A, configura-
during configuration. For parallel Master configuration tion requires 14779 bits for each device, arranged in 197
modes, these signals provide PROM enable control and data frames. An additional 40 bits are used in the header.
allow the data pins to be shared with user logic signals. See Figure 22. The specific data format for each device is
produced by the development system and one or more of
User I/O inputs can be programmed to be either TTL or these files can then be combined and appended to a length
CMOS compatible thresholds. At power-up, all inputs have count preamble and be transformed into a PROM format
TTL thresholds and can change to CMOS thresholds at the file by the development system. A compatibility exception
completion of configuration if the user has selected CMOS precludes the use of an XC2000-series device as the mas-
thresholds. The threshold of PWRDWN and the direct clock ter for XC3000-series devices if their DONE or RESET are
inputs are fixed at a CMOS level. programmed to occur after their outputs become active.
If the crystal oscillator is used, it will begin operation before The Tie Option defines output levels of unused blocks of a
configuration is complete to allow time for stabilization design and connects these to unused routing resources.
before it is connected to the internal circuitry. This prevents indeterminate levels that might produce par-
7
asitic supply currents. If unused blocks are not sufficient to
complete the tie, the user can indicate nets which must not
Postamble
Last Frame
Data Frame
12 24 4
3 3 4
STOP
DIN Stop
Preamble Length Count Data
Start
Bit Length Count*
Start
Bit
The configuration data consists of a composite
* 40-bit preamble/length count, followed by one or Weak Pull-Up
I/O Active
more concatenated FPGA programs, separated by
4-bit postambles. An additional final postamble bit
is added for each slave device and the result rounded
up to a byte boundary. The length count is two less
than the number of resulting bits.
PROGRAM DONE
Timing of the assertion of DONE and
termination of the INTERNAL RESET
may each be programmed to occur
one cycle before or after the I/O outputs
become active. Internal Reset
Heavy lines indicate the default condition

X5988
Figure 22: Configuration and Start-up of One or More FPGAs.

R
be used to drive the remaining unused routing, as that Peripheral Mode

might affect timing of user nets. Tie can be omitted for quick
Peripheral mode provides a simplified interface through
breadboard iterations where a few additional milliamps of
which the device may be loaded byte-wide, as a processor
Icc are acceptable.
peripheral. Figure 27 shows the peripheral mode connec-
The configuration bitstream begins with eight High pream- tions. Processor write cycles are decoded from the com-
ble bits, a 4-bit preamble code and a 24-bit length count. mon assertion of the active low Write Strobe (WS), and two
When configuration is initiated, a counter in the FPGA is set active low and one active high Chip Selects (CS0, CS1,
to zero and begins to count the total number of configura- CS2). The FPGA generates a configuration clock from the
tion clock cycles applied to the device. As each configura- internal timing generator and serializes the parallel input
tion data frame is supplied to the device, it is internally data for internal framing or for succeeding slaves on Data
assembled into a data word, which is then loaded in parallel Out (DOUT). A output High on READY/BUSY pin indicates
into one word of the internal configuration memory array. the completion of loading for each byte when the input reg-
The configuration loading process is complete when the ister is ready for a new byte. As with Master modes, Periph-
current length count equals the loaded length count and the eral mode may also be used as a lead device for a
required configuration program data frames have been daisy-chain of slave devices.
written. Internal user flip-flops are held Reset during config-
uration. Slave Serial Mode
Two user-programmable pins are defined in the unconfig- Slave Serial mode provides a simple interface for loading
ured Field Programmable Gate Array. High During Config- the Field Programmable Gate Array configuration as
uration (HDC) and Low During Configuration (LDC) as well shown in Figure 29. Serial data is supplied in conjunction
as DONE/PROG may be used as external control signals with a synchronizing input clock. Most Slave mode applica-
during configuration. In Master mode configurations it is tions are in daisy-chain configurations in which the data
convenient to use LDC as an active-Low EPROM Chip input is driven from the previous FPGA’s data out, while the
Enable. After the last configuration data bit is loaded and clock is supplied by a lead device in Master or Peripheral
the length count compares, the user I/O pins become mode. Data may also be supplied by a processor or other
active. Options allow timing choices of one clock earlier or special circuits.
later for the timing of the end of the internal logic RESET
and the assertion of the DONE signal. The open-drain Daisy Chain
DONE/PROG output can be AND-tied with multiple devices The development system is used to create a composite
and used as an active-High READY, an active-Low PROM configuration for selected FPGAs including: a preamble, a
enable or a RESET to other portions of the system. The length count for the total bitstream, multiple concatenated
state diagram of Figure 20 illustrates the configuration pro- data programs and a postamble plus an additional fill bit
cess. per device in the serial chain. After loading and passing-on
the preamble and length count to a possible daisy-chain, a
Configuration Modes lead device will load its configuration data frames while pro-
viding a High DOUT to possible down-stream devices as
Master Mode
shown in Figure 25. Loading continues while the lead
In Master mode, the FPGA automatically loads configura- device has received its configuration program and the cur-
tion data from an external memory device. There are three rent length count has not reached the full value. The addi-
Master modes that use the internal timing source to supply tional data is passed through the lead device and appears
the configuration clock (CCLK) to time the incoming data. on the Data Out (DOUT) pin in serial form. The lead device
Master Serial mode uses serial configuration data supplied also generates the Configuration Clock (CCLK) to synchro-
to Data-in (DIN) from a synchronous serial source such as nize the serial output data and data in of down-stream
the Xilinx Serial Configuration PROM shown in Figure 23. FPGAs. Data is read in on DIN of slave devices by the pos-
Master Parallel Low and High modes automatically use itive edge of CCLK and shifted out the DOUT on the nega-
parallel data supplied to the D0–D7 pins in response to the tive edge of CCLK. A parallel Master mode device uses its
16-bit address generated by the FPGA. Figure 25 shows internal timing generator to produce an internal CCLK of 8
an example of the parallel Master mode connections times its EPROM address rate, while a Peripheral mode
required. The HEX starting address is 0000 and increments device produces a burst of 8 CCLKs for each chip select
for Master Low mode and it is FFFF and decrements for and write-strobe cycle. The internal timing generator con-
Master High mode. These two modes provide address tinues to operate for general timing and synchronization of
compatibility with microprocessors which begin execution inputs in all modes.
from opposite ends of memory.

R
Special Configuration Functions configuration, each Readback frame has only one Stop bit
(read back as a zero). The third leading dummy bit men-
The configuration data includes control over several spe- tioned above can be considered the Start bit of the first
cial functions in addition to the normal user logic functions frame. All data frames must be read back to complete the
and interconnect. process and return the Mode Select and CCLK pins to their
• Input thresholds normal functions.
• Readback disable Readback data includes the current state of each CLB
• DONE pull-up resistor
flip-flop, each input flip-flop or latch, and each device pad.
• DONE timing
These data are imbedded into unused configuration bit
• RESET timing positions during Readback. This state information is used
• Oscillator frequency divided by two by the development system In-Circuit Verifier to provide
Each of these functions is controlled by configuration data visibility into the internal operation of the logic while the
bits which are selected as part of the normal development system is operating. To readback a uniform time-sample of
system bitstream generation process. all storage elements, it may be necessary to inhibit the sys-
tem clock.
Input Thresholds
Prior to the completion of configuration all FPGA input
Reprogram
thresholds are TTL compatible. Upon completion of config- To initiate a re-programming cycle, the dual-function pin
uration, the input thresholds become either TTL or CMOS DONE/PROG must be given a High-to-Low transition. To
compatible as programmed. The use of the TTL threshold reduce sensitivity to noise, the input signal is filtered for two
option requires some additional supply current for thresh- cycles of the FPGA internal timing generator. When repro-
old shifting. The exception is the threshold of the gram begins, the user-programmable I/O output buffers are
PWRDWN input and direct clocks which always have a disabled and high-impedance pull-ups are provided for the
CMOS input. Prior to the completion of configuration the package pins. The device returns to the Clear state and
user I/O pins each have a high impedance pull-up. The clears the configuration memory before it indicates ‘initial-
configuration program can be used to enable the IOB ized’. Since this Clear operation uses chip-individual inter-
pull-up resistors in the Operational mode to act either as an nal timing, the master might complete the Clear operation
7
input load or to avoid a floating input on an otherwise and then start configuration before the slave has completed
unused pin. the Clear operation. To avoid this problem, the slave INIT
pins must be AND-wired and used to force a RESET on the
Readback master (see Figure 25). Reprogram control is often imple-
The contents of a Field Programmable Gate Array may be mented using an external open-collector driver which pulls
read back if it has been programmed with a bitstream in DONE/PROG Low. Once a stable request is recognized,
which the Readback option has been enabled. Readback the DONE/PROG pin is held Low until the new configura-
may be used for verification of configuration and as a tion has been completed. Even if the re-program request is
method of determining the state of internal logic nodes dur- externally held Low beyond the configuration period, the
ing debugging. There are three options in generating the FPGA will begin operation upon completion of configura-
configuration bitstream. tion.
• “Never” inhibits the Readback capability. DONE Pull-up
• “One-time,” inhibits Readback after one Readback has
DONE/PROG is an open-drain I/O pin that indicates the
been executed to verify the configuration.
FPGA is in the operational state. An optional internal
• “On-command” allows unrestricted use of Readback.
pull-up resistor can be enabled by the user of the develop-
Readback is accomplished without the use of any of the ment system. The DONE/PROG pins of multiple FPGAs in
user I/O pins; only M0, M1 and CCLK are used. The initia- a daisy-chain may be connected together to indicate all are
tion of Readback is produced by a Low to High transition of DONE or to direct them all to reprogram.
the M0/RTRIG (Read Trigger) pin. The CCLK input must
then be driven by external logic to read back the configura- DONE Timing
tion data. The first three Low-to-High CCLK transitions The timing of the DONE status signal can be controlled by
clock out dummy data. The subsequent Low-to-High CCLK a selection to occur either a CCLK cycle before, or after, the
transitions shift the data frame information out on the outputs going active. See Figure 22. This facilitates control
M1/RDATA (Read Data) pin. Note that the logic polarity is of external functions such as a PROM enable or holding a
always inverted, a zero in configuration becomes a one in system in a wait state.
Readback, and vice versa. Note also that each Readback
frame has one Start bit (read back as a one) but, unlike in

R
RESET Timing but with incorrect configuration and the possibility of inter-
nal contention.
As with DONE timing, the timing of the release of the inter-
nal reset can be controlled to occur either a CCLK cycle An XC3000A/XC3100A/XC3000L/XC3100L device starts
before, or after, the outputs going active. See Figure 22. any new frame only if the three preceding bits are all ones.
This reset keeps all user programmable flip-flops and If this check fails, it pulls INIT Low and stops the internal
latches in a zero state during configuration. configuration, although the Master CCLK keeps running.
The user must then start a new configuration by applying a
Crystal Oscillator Division >6 µs Low level on RESET.
A selection allows the user to incorporate a dedicated This simple check does not protect against random bit
divide-by-two flip-flop between the crystal oscillator and the errors, but it offers almost 100 percent protection against
alternate clock line. This guarantees a symmetrical clock erroneous configuration files, defective configuration data
signal. Although the frequency stability of a crystal oscilla- sources, synchronization errors between configuration
tor is very good, the symmetry of its waveform can be source and FPGA, or PC-board level defects, such as bro-
affected by bias or feedback drive. ken lines or solder-bridges.
Bitstream Error Checking Reset Spike Protection
Bitstream error checking protects against erroneous con- A separate modification slows down the RESET input
figuration. before configuration by using a two-stage shift register
Each Xilinx FPGA bitstream consists of a 40-bit preamble, driven from the internal clock. It tolerates submicrosecond
followed by a device-specific number of data frames. The High spikes on RESET before configuration. The XC3000
number of bits per frame is also device-specific; however, master can be connected like an XC4000 master, but with
each frame ends with three stop bits (111) followed by a its RESET input used instead of INIT. (On XC3000, INIT is
start bit for the next frame (0). output only).
All devices in all XC3000 families start reading in a new Soft Start-up
frame when they find the first 0 after the end of the previous
After configuration, the outputs of all FPGAs in a
frame. An original XC3000 device does not check for the
daisy-chain become active simultaneously, as a result of
correct stop bits, but XC3000A, XC3100A, XC3000L, and
the same CCLK edge. In the original XC3000/3100
XC3100L devices check that the last three bits of any frame
devices, each output becomes active in either fast or
are actually 111.
slew-rate limited mode, depending on the way it is config-
Under normal circumstances, all these FPGAs behave the ured. This can lead to large ground-bounce signals. In
same way; however, if the bitstream is corrupted, an XC3000A, XC3000L, XC3100A, and XC3100L devices, all
XC3000 device will always start a new frame as soon as it outputs become active first in slew-rate limited mode,
finds the first 0 after the end of the previous frame, even if reducing the ground bounce. After this soft start-up, each
the data is completely wrong or out-of-sync. Given suffi- individual output slew rate is again controlled by the
cient zeros in the data stream, the device will also go Done, respective configuration bit.

R
Configuration Timing
This section describes the configuration modes in detail.
Master Serial Mode

In Master Serial mode, the CCLK output of the lead FPGA DOUT changes on the falling CCLK edge, and the next
drives a Xilinx Serial PROM that feeds the DIN input. Each device in the daisy-chain accepts data on the subsequent
rising edge of the CCLK output increments the Serial rising CCLK edge.
PROM internal address counter. This puts the next data bit
The SPROM CE input can be driven from either LDC or
on the SPROM data output, connected to the DIN pin. The DONE. Using LDC avoids potential contention on the DIN
lead FPGA accepts this data on the subsequent rising pin, if this pin is configured as user-I/O, but LDC is then
CCLK edge. restricted to be a permanently High user output. Using
The lead FPGA then presents the preamble data (and all DONE also avoids contention on DIN, provided the early
data that overflows the lead device) on its DOUT pin. There DONE option is invoked.
is an internal delay of 1.5 CCLK periods, which means that
*
IF READBACK IS
ACTIVATED, A * +5 V
5-kΩ RESISTOR IS
REQUIRED IN
SERIES WITH M1
M0 M1 PWRDWN
DURING CONFIGURATION TO DIN OF OPTIONAL

THE 5 kΩ M2 PULL-DOWN DOUT DAISY-CHAINED LCAs WITH
RESISTOR OVERCOMES THE DIFFERENT CONFIGURATIONS
INTERNAL PULL-UP, M2
BUT IT ALLOWS M2 TO TO CCLK OF OPTIONAL
BE USER I/O. +5V DAISY-CHAINED LCAs WITH
HDC DIFFERENT CONFIGURATIONS
LDC
GENERAL-
PURPOSE INIT
USER I/O
PINS
•
•
7
• OTHER
• I/O PINS TO CCLK OF OPTIONAL
• SLAVE LCAs WITH IDENTICAL
CONFIGURATIONS
XC3000
FPGA TO DIN OF OPTIONAL
DEVICE SLAVE LCAs WITH IDENTICAL
CONFIGURATIONS
+5 V
RESET RESET
VCC VPP
DIN DATA DATA
CCLK CLK CLK CASCADED

SCP
SERIAL
D/P CE CEO CE MEMORY
INIT OE/RESET OE/RESET

XC17xx
(LOW RESETS THE XC17xx ADDRESS POINTER) X5989_01
Figure 23: Master Serial Mode Circuit Diagram

R
CCLK
(Output)
2 TCKDS
1 TDSCK

(Output)
X3223

Data In setup 1 TDSCK 60 ns
CCLK
Data In hold 2 CKDS 0 ns
Notes: 1. At power-up, VCC must rise from 2.0 V to VCC min in less than 25 ms. If this is not possible, configuration can be delayed by
holding RESET Low until VCC has reached 4.0 V (2.5 V for the XC3000L). A very long VCC rise time of >100 ms, or a
non-monotonically rising VCC may require >6-µs High level on RESET, followed by a >6-µs Low level on RESET and D/P
after VCC has reached 4.0 V (2.5 V for the XC3000L).
2. Configuration can be controlled by holding RESET Low with or until after the INIT of all daisy-chain slave-mode devices is
High.
3. Master-serial-mode timing is based on slave-mode testing.

R
Master Parallel Mode

In Master Parallel mode, the lead FPGA directly addresses nal delay of 1.5 CCLK periods, after the rising CCLK edge
an industry-standard byte-wide EPROM and accepts eight that accepts a byte of data, and also changes the EPROM
data bits right before incrementing (or decrementing) the address, until the falling CCLK edge that makes the LSB
address outputs. (D0) of this byte appear at DOUT. This means that DOUT
changes on the falling CCLK edge, and the next device in
The eight data bits are serialized in the lead FPGA, which
then presents the preamble data (and all data that over-
CCLK edge.
flows the lead device) on the DOUT pin. There is an inter-
* If Readback is +5 V * +5 V
+5 V * +5 V *
Activated, a
5-kΩ Resistor is M0 M1PWRDWN M0 M1PWRDWN M0 M1PWRDWN
Required in
Series With M1
5 kΩ CCLK CCLK 5 kΩ CCLK 5 kΩ
M2 FPGA
Slave #1
... FPGA
Slave #n
HDC M2 M2
RCLK A15 A15 HDC HDC
General- General- General-
Purpose A14 A14 LDC LDC Purpose
Purpose
User I/O A13 User I/O User I/O
A13
Pins EPROM Other Pins Other Pins
...
...
A12 A12
I/O Pins I/O Pins
.....
Other
I/O Pins A11 A11
FPGA A10 A10 INIT INIT

Master A9 D/P D/P
A9
D7 A8 A8 RESET Reset
D6 A7 A7 D7
D5 A6 A6 D6 7
D4 A5 A5 D5 Note: XC2000 Devices Do Not
Have INIT to Hold Off a Master
D3 A4 A4 D4 Device. Reset of a Master Device
D2 A3 D3 Should be Asserted by an External
A3 Timing Circuit to Allow for LCA CCLK
D1 A2 A2 D2 Variations in Clear State Time.
D0 A1 A1 D1
A0 A0 D0
D/P OE
RESET INIT CE +5 V
N.C.
8 5 kΩ Each
Open
Reprogram Collector
System Reset
X5990

R
A0-A15
1 TRAC
D0-D7
Byte
2 TDRC 3 TRCD
RCLK
(output)
7 CCLKs CCLK
CCLK
(output)
DOUT
(output) D6 D7
Byte n - 1 X5380

To address valid 1 TRAC 0 200 ns
To data setup 2 TDRC 60 ns
RCLK To data hold 3 TRCD 0 ns
RCLK High TRCH 600 ns
RCLK Low TRCL 4.0 µs
non-monotonically rising VCC may require a >6-µs High level on RESET, followed by a >6-µs Low level on RESET and D/P
2. Configuration can be controlled by holding RESET Low with or until after the INIT of all daisy-chain slave-mode devices is
High.
This timing diagram shows that the EPROM requirements are extremely relaxed:
EPROM access time can be longer than 4000 ns. EPROM data output has no hold time requirements.

R
Peripheral Mode
Peripheral mode uses the trailing edge of the logic AND when the byte-wide input buffer has transferred its informa-
condition of the CS0, CS1, CS2, and WS inputs to accept tion into the shift register, and the buffer is ready to receive
byte-wide data from a microprocessor bus. In the lead new data. The length of the BUSY signal depends on the
FPGA, this data is loaded into a double-buffered UART-like activity in the UART. If the shift register had been empty
parallel-to-serial converter and is serially shifted into the when the new byte was received, the BUSY signal lasts for
internal logic. The lead FPGA presents the preamble data only two CCLK periods. If the shift register was still full
(and all data that overflows the lead device) on the DOUT when the new byte was received, the BUSY signal can be
pin. as long as nine CCLK periods.
The Ready/Busy output from the lead device acts as a Note that after the last byte has been entered, only seven
handshake signal to the microprocessor. RDY/BUSY goes of its bits are shifted out. CCLK remains High with DOUT
Low when a byte has been received, and goes High again equal to bit 6 (the next-to-last bit) of the last byte entered.
+5 V
CONTROL ADDRESS DATA
*
SIGNALS BUS BUS * IF READBACK IS
ACTIVATED, A
8 5 kΩ 5-kΩ RESISTOR IS
M0 M1 PWR
REQUIRED IN SERIES
DWN
WITH M1
D0–7
D0–7 CCLK OPTIONAL
DAISY-CHAINED
FPGAs WITH DIFFERENT
DOUT CONFIGURATIONS
ADDRESS
DECODE CS0 M2
...
LOGIC HDC
+5 V FPGA LDC
GENERAL- 7
PURPOSE
CS1 USER I/O
PINS
CS2
OTHER
...
WS
I/O PINS
RDY/BUSY
INIT
REPROGRAM
OC D/P
RESET
X5991
Figure 27: Peripheral Mode Circuit Diagram

R
WRITE TO FPGA
WS, CS0, CS1
CS2 1
TCA
2 TCD
TDC 3
D0-D7 Valid
CCLK
4 TWTRB TBUSY
6
RDY/BUSY
DOUT D6 D7 D0 D1 D2
Previous Byte New Byte

X5992

Effective Write time required 1 TCA 100 ns
(Assertion of CS0, CS1, CS2, WS)
DIN Setup time required 2 TDC 60 ns
WRITE
DIN Hold time required 3 TCD 0 ns
RDY/BUSY delay after end of WS 4 TWTRB 60 ns
Earliest next WS after end of BUSY 5 TRBWT 0 ns

RDY
BUSY Low time generated 6 TBUSY 2.5 9 CCLK
periods
2. Configuration must be delayed until the INIT of all FPGAs is High.
3. Time from end of WS to CCLK cycle for the new byte of data depends on completion of previous byte processing and the
phase of the internal timing generator for CCLK.
5. TBUSY indicates that the double-buffered parallel-to-serial converter is not yet ready to receive new data. The shortest TBUSY
occurs when a byte is loaded into an empty parallel-to-serial converter. The longest TBUSY occurs when a new word is
loaded into the input register before the second-level buffer has started shifting out data.
Note: This timing diagram shows very relaxed requirements: Data need not be held beyond the rising edge of WS. BUSY
will go active within 60 ns after the end of WS. BUSY will stay active for several microseconds. WS may be asserted
immediately after the end of BUSY.
Figure 28: Peripheral Mode Programming Switching Characteristics

R
Slave Serial Mode

In Slave Serial mode, an external signal drives the CCLK flows the lead device) on its DOUT pin. There is an internal
input(s) of the FPGA(s). The serial configuration bitstream delay of 0.5 CCLK periods, which means that DOUT
must be available at the DIN input of the lead FPGA a short changes on the falling CCLK edge, and the next device in
set-up time before each rising CCLK edge. The lead device the daisy-chain accepts data on the subsequent rising
then presents the preamble data (and all data that over- CCLK edge.
* If Readback is
+5 V *
Activated, a
5-kΩ Resistor is
Required in
Series with M1
M0 M1 PWRDWN
Micro 5 kΩ
Computer
Optional
STRB CCLK M2 Daisy-Chained
LCAs with
D0 DIN DOUT Different
Configurations
D1 HDC
I/O D2 LDC General-

Port Purpose
+5 V User I/O
D3 Pins
FPGA
D4
Other
...
D5
I/O Pins
D6
D/P
7
D7 INIT
RESET RESET
X5993
Figure 29: Slave Serial Mode Circuit Diagram

R
DIN Bit n Bit n + 1
CCLK
4 TCCH 3 TCCO
DOUT
Bit n - 1 Bit n
(Output)
X5379

To DOUT 3 TCCO 100 ns

CCLK DIN hold 2 TCCD 0 ns
High time 4 TCCH 0.05 µs
Low time (Note 1) 5 TCCL 0.05 5.0 µs
Notes: 1. The max limit of CCLK Low time is caused by dynamic circuitry inside the FPGA.
2. Configuration must be delayed until the INIT of all FPGAs is High.
3. At power-up, VCC must rise from 2.0 V to VCC min in less than 25 ms. If this is not possible, configuration can be delayed by

R
Program Readback Switching Characteristics
DONE/PROG
(OUTPUT)
1 TRTH
RTRIG (M0)
2 TRTCC
4 TCCL
4 TCCL
CCLK(1)
5
3 TCCRD
HI-Z VALID VALID

M1 Input/
READBACK OUTPUT READBACK OUTPUT
RDATA Output
X6116
Description Symbol Min Max Units 7

RTRIG RTRIG High 1 TRTH 250 ns
RTRIG setup 2 TRTCC 200 ns
RDATA delay 3 TCCRD 100 ns
CCLK
High time 4 TCCHR 0.5 µs
Low time 5 TCCLR 0.5 5 µs
Notes: 1. During Readback, CCLK frequency may not exceed 1 MHz.
2. RETRIG (M0 positive transition) shall not be done until after one clock following active I/O pins.
3. Readback should not be initiated until configuration is complete.
4. TCCLR is 5 µs min to 15 µs max for XC3000L.

R
General XC3000 Series Switching Characteristics
4 TMRW
RESET
2 TMR
3 TRM
M0/M1/M2
5 TPGW
DONE/PROG
6 TPGI
INIT User State Clear State Configuration State

(Output)
PWRDWN
Note 3
VCC (Valid)
VCCPD
X5387

M0, M1, M2 setup time required 2 TMR 1 µs
RESET (2) M0, M1, M2 hold time required 3 TRM 4.5 µs
RESET Width (Low) req. for Abort 4 TMRW 6 µs
Width (Low) required for Re-config. 5 TPGW 6 µs
DONE/PROG
INIT response after D/P is pulled Low 6 TPGI 7 µs
PWRDWN (3) Power Down VCC VCCPD 2.3 V
holding RESET Low until Vcc has reached 4.0 V (2.5 V for XC3000L). A very long Vcc rise time of >100 ms, or a
after Vcc has reached 4.0 V (2.5 V for XC3000L).
2. RESET timing relative to valid mode lines (M0, M1, M2) is relevant when RESET is used to delay configuration. The
specified hold time is caused by a shift-register filter slowing down the response to RESET during configuration.
3. PWRDWN transitions must occur while VCC >4.0 V(2.5 V for XC3000L).

R
Device Performance logic block is a function of supply voltage and temperature.

See Figure 32.
The XC3000 families of FPGAs can achieve very high per-
formance. This is the result of Interconnect performance depends on the routing
resources used to implement the signal path. Direct inter-
• A sub-micron manufacturing process, developed and connects to the neighboring CLB provide an extremely fast
continuously being enhanced for the production of path. Local interconnects go through switch matrices
state-of-the-art CMOS SRAMs. (magic boxes) and suffer an RC delay, equal to the resis-
• Careful optimization of transistor geometries, circuit tance of the pass transistor multiplied by the capacitance of
design, and lay-out, based on years of experience with the driven metal line. Longlines carry the signal across the
the XC3000 family. length or breadth of the chip with only one access delay.
• A look-up table based, coarse-grained architecture that Generous on-chip signal buffering makes performance rel-
can collapse multiple-layer combinatorial logic into a atively insensitive to signal fan-out; increasing fan-out from
single function generator. One CLB can implement up 1 to 8 changes the CLB delay by only 10%. Clocks can be
to four layers of conventional logic in as little as 1.5 ns. distributed with two low-skew clock distribution networks.
Actual system performance is determined by the timing of The tools in the Development System used to place and
critical paths, including the delay through the combinatorial route a design in an XC3000 FPGA automatically calculate
and sequential logic elements within CLBs and IOBs, plus the actual maximum worst-case delays along each signal
the delay in the interconnect routing. The AC-timing speci- path. This timing information can be back-annotated to the
fications state the worst-case timing parameters for the var- design’s netlist for use in timing simulation or examined
ious logic resources available in the XC3000-families with, a static timing analyzer.
architecture. Figure 31 shows a variety of elements
involved in determining system performance. Actual system performance is applications dependent. The
maximum clock rate that can be used in a system is deter-
Logic block performance is expressed as the propagation mined by the critical path delays within that system. These
time from the interconnect point at the input to the block to delays are combinations of incremental logic and routing
the output of the block in the interconnect area. Since com- delays, and vary from design to design. In a synchronous
binatorial logic is implemented with a memory lookup table system, the maximum clock rate depends on the number of 7
within a CLB, the combinatorial delay through the CLB, combinatorial logic layers between re-synchronizing
called TILO, is always the same, regardless of the function flip-flops. Figure 33 shows the achievable clock rate as a
being implemented. For the combinatorial logic function function of the number of CLB layers.
driving the data input of the storage element, the critical
timing is data set-up relative to the clock edge provided to
the flip-flop element. The delay from the clock source to the
output of the logic block is critical in the timing signals pro-
duced by storage elements. Loading of a logic-block output
is limited only by the resulting propagation delay of the
larger interconnect network. Speed performance of the
Clock to Output Combinatorial Setup

TCKO TILO TICK TOP
CLB CLB CLB IOB
Logic Logic
PAD
(K) (K)
CLOCK
TCKO
IOB
PAD
T PID TOKPO
X3178
Figure 31: Primary Block Speed Factors. Actual timing is a function of various block factors combined with routing.
factors. Overall performance can be evaluated with the timing calculator or by an optional simulation.

R
SPECIFIED WORST-CASE VALUES
1.00
)
V) 4.5 V
4.75 RY (
L( LITA
CIA X MI
MER MA
COM
MAX
0.80
NORMALIZED DELAY
0.60 TYPICAL COMMERCIAL

(+ 5.0 V, 25°C)
TYPICAL MILITARY
0.40 V)
ERCIAL (4
.75 V) ARY (4.5
MIN COMM MIN MILIT
M ERCIAL (5.25 V)
MIN COM
Y (5.5 V)
MIN MILITAR
0.20
– 55 – 40 – 20 0 25 40 70 80 100 125
TEMPERATURE (°C)
Figure 32: Relative Delay as a Function of Temperature, Supply Voltage and Processing Variations X6094
Power
300
Power Distribution
250 Power for the FPGA is distributed through a grid to achieve
System Clock (MHz)

200
Inside the FPGA, a dedicated VCC and ground ring sur-
150 rounding the logic array provides power to the I/O drivers.
100 An independent matrix of VCC and groundlines supplies the
XC3100A-3
interior logic of the device. This power distribution grid pro-
50
XC3000A--6
vides a stable supply and ground for all internal logic, pro-
0 viding the external package power pins are all connected
CLB Levels: 4 CLBs 3 CLBs 2 CLBs 1 CLB Toggle
Gate Levels: (4-16) (3-12) (2-8) (1-4) Rate and appropriately decoupled. Typically a 0.1-µF capacitor
X7065
connected near the VCC and ground pins will provide ade-
quate decoupling.
Figure 33: Clock Rate as a Function of Logic
Complexity (Number of Combinational Levels between Output buffers capable of driving the specified 4- or 8-mA
Flip-Flops) loads under worst-case conditions may be capable of driv-
ing as much as 25 to 30 times that current in a best case.
tance and reducing simultaneous output transitions in the
same direction. It may also be beneficial to locate heavily
loaded output buffers near the ground pads. The I/O Block
output buffers have a slew-limited mode which should be
used where output rise and fall times are not speed critical.
Slew-limited outputs maintain their dc drive capability, but
generate less external reflections and internal noise.

R
Dynamic Power Consumption

XC3042A XC3042L XC3142A
One CLB driving three local interconnects 0.25 0.17 0.25 mW per MHz
One global clock buffer and clock line 2.25 1.40 1.70 mW per MHz
One device output with a 50 pF load 1.25 1.25 1.25 mW per MHz
Power Consumption
The Field Programmable Gate Array exhibits the low power has built in powerdown logic which, when activated, will
consumption characteristic of CMOS ICs. For any design, disable normal operation of the device and retain only the
the configuration option of TTL chip input threshold configuration data. All internal operation is suspended and
requires power for the threshold reference. The power output buffers are placed in their high-impedance state with
required by the static memory cells that hold the configura- no pull-ups. Different from the XC3000 family which can be
tion data is very low and may be maintained in a powered down to a current consumption of a few micro-
power-down mode. amps, the XC3100A draws 5 mA, even in power-down.
This makes power-down operation less meaningful. In con-
Typically, most of power dissipation is produced by external
trast, ICCPD for the XC3000L is only 10 µA.
capacitive loads on the output buffers. This load and fre-
quency dependent power is 25 µW/pF/MHz per output. To force the FPGA into the Powerdown state, the user must
Another component of I/O power is the external dc loading pull the PWRDWN pin Low and continue to supply a reten-
on all output pins. tion voltage to the VCC pins. When normal power is
restored, VCC is elevated to its normal operating voltage
Internal power dissipation is a function of the number and
and PWRDWN is returned to a High. The FPGA resumes
size of the nodes, and the frequency at which they change.
operation with the same internal sequence that occurs at
In an FPGA, the fraction of nodes changing on a given
the conclusion of configuration. Internal-I/O and logic-block
clock is typically low (10-20%). For example, in a long
storage elements will be reset, the outputs will become
binary counter, the total activity of all counter flip-flops is
equivalent to that of only two CLB outputs toggling at the
enabled and the DONE/PROG pin will be released. 7
clock frequency. Typical global clock-buffer power is When VCC is shut down or disconnected, some power
between 2.0 mW/MHz for the XC3020A and 3.5 mW/MHz might unintentionally be supplied from an incoming signal
for the XC3090A. The internal capacitive load is more a driving an I/O pin. The conventional electrostatic input pro-
function of interconnect than fan-out. With a typical load of tection is implemented with diodes to the supply and
three general interconnect segments, each CLB output ground. A positive voltage applied to an input (or output)
requires about 0.25 mW per MHz of its output frequency. will cause the positive protection diode to conduct and drive
the VCC connection. This condition can produce invalid
Because the control storage of the FPGA is CMOS static
power conditions and should be avoided. A large series
memory, its cells require a very low standby current for data
resistor might be used to limit the current or a bipolar buffer
retention. In some systems, this low data retention current
may be used to isolate the input signal.
characteristic can be used as a method of preserving con-
figurations in the event of a primary power loss. The FPGA

R
Pin Descriptions Once configuration is done, a High-to-Low transition of this

pin will cause an initialization of the FPGA and start a
Permanently Dedicated Pins reconfiguration.
VCC M0/RTRIG
Two to eight (depending on package type) connections to As Mode 0, this input is sampled on power-on to determine
the positive V supply voltage. All must be connected. the power-on delay (214 cycles if M0 is High, 216 cycles if M0
is Low). Before the start of configuration, this input is again
GND sampled together with M1, M2 to determine the configura-
Two to eight (depending on package type) connections to tion mode to be used.
ground. All must be connected. A Low-to-High input transition, after configuration is com-
plete, acts as a Read Trigger and initiates a Readback of
PWRDWN
configuration and storage-element data clocked by CCLK.
A Low on this CMOS-compatible input stops all internal By selecting the appropriate Readback option when gener-
activity, but retains configuration. All flip-flops and latches ating the bitstream, this operation may be limited to a single
are reset, all outputs are 3-stated, and all inputs are inter- Readback, or be inhibited altogether.
preted as High, independent of their actual level. When
PWDWN returns High, the FPGA becomes operational M1/RDATA
with DONE Low for two cycles of the internal 1-MHz clock. As Mode 1, this input and M0, M2 are sampled before the
Before and during configuration, PWRDWN must be High. start of configuration to establish the configuration mode to
If not used, PWRDWN must be tied to VCC. be used. If Readback is never used, M1 can be tied directly
to ground or VCC. If Readback is ever used, M1 must use a
RESET
5-kΩ resistor to ground or VCC, to accommodate the
This is an active Low input which has three functions. RDATA output.
Prior to the start of configuration, a Low input will delay the As an active-Low Read Data, after configuration is com-
start of the configuration process. An internal circuit senses plete, this pin is the output of the Readback data.
the application of power and begins a minimal time-out
cycle. When the time-out and RESET are complete, the User I/O Pins That Can Have Special
levels of the M lines are sampled and configuration begins. Functions
If RESET is asserted during a configuration, the FPGA is M2
re-initialized and restarts the configuration at the termina-
tion of RESET. During configuration, this input has a weak pull-up resistor.
Together with M0 and M1, it is sampled before the start of
If RESET is asserted after configuration is complete, it pro-
configuration to establish the configuration mode to be
vides a global asynchronous RESET of all IOB and CLB
used. After configuration, this pin is a user-programmable
storage elements of the FPGA.
I/O pin.
CCLK
HDC
During configuration, Configuration Clock is an output of an
During configuration, this output is held at a High level to
FPGA in Master mode or Peripheral mode, but an input in
indicate that configuration is not yet complete. After config-
Slave mode. During Readback, CCLK is a clock input for
uration, this pin is a user-programmable I/O pin.
shifting configuration data out of the FPGA.
CCLK drives dynamic circuitry inside the FPGA. The Low LDC
time may, therefore, not exceed a few microseconds. When During Configuration, this output is held at a Low level to
used as an input, CCLK must be “parked High”. An internal indicate that the configuration is not yet complete. After
pull-up resistor maintains High when the pin is not being configuration, this pin is a user-programmable I/O pin. LDC
driven. is particularly useful in Master mode as a Low enable for an
EPROM, but it must then be programmed as a High after
DONE/PROG (D/P) configuration.
DONE is an open-drain output, configurable with or without
an internal pull-up resistor of 2 to 8 k Ω. At the completion of INIT
configuration, the FPGA circuitry becomes active in a syn- This is an active Low open-drain output with a weak pull-up
chronous order; DONE is programmed to go active High and is held Low during the power stabilization and internal
one cycle either before or after the outputs go active. clearing of the configuration memory. It can be used to indi-
cate status to a configuring microprocessor or, as a wired

R
AND of several slave mode devices, a hold-off signal for a D0-D7

master mode device. After configuration this pin becomes a
This set of eight pins represents the parallel configuration
user-programmable I/O pin.
byte for the parallel Master and Peripheral modes. After
BCLKIN configuration is complete, they are user-programmed I/O
pins.
This is a direct CMOS level input to the alternate clock
buffer (Auxiliary Buffer) in the lower right corner. A0-A15
XTL1 During Master Parallel mode, these 16 pins present an

address output for a configuration EPROM. After configura-
This user I/O pin can be used to operate as the output of an tion, they are user-programmable I/O pins.
amplifier driving an external crystal and bias circuitry.
DIN
XTL2
During Slave or Master Serial configuration, this pin is used
This user I/O pin can be used as the input of an amplifier as a serial-data input. In the Master or Peripheral configu-
connected to an external crystal and bias circuitry. The I/O ration, this is the Data 0 input. After configuration is com-
Block is left unconfigured. The oscillator configuration is plete, this pin becomes a user-programmed I/O pin.
activated by routing a net from the oscillator buffer symbol
output and by the MakeBits program. DOUT
CS0, CS1, CS2, WS During configuration this pin is used to output serial-config-
uration data to the DIN pin of a daisy-chained slave. After
These four inputs represent a set of signals, three active configuration is complete, this pin becomes a user-pro-
Low and one active High, that are used to control configu- grammed I/O pin.
ration-data entry in the Peripheral mode. Simultaneous
assertion of all four inputs generates a Write to the internal TCLKIN
data buffer. The removal of any assertion clocks in the This is a direct CMOS-level input to the global clock buffer.
D0-D7 data. In Master-Parallel mode, WS and CS2 are the This pin can also be configured as a user programmable
A0 and A1 outputs. After configuration, these pins are 7
I/O pin. However, since TCLKIN is the preferred input to the
user-programmable I/O pins. global clock net, and the global clock net should be used as
RDY/BUSY the primary clock source, this pin is usually the clock input
to the chip.
During Peripheral Parallel mode configuration this pin indi-
cates when the chip is ready for another byte of data to be Unrestricted User I/O Pins
written to it. After configuration is complete, this pin
becomes a user-programmed I/O pin. I/O
An I/O pin may be programmed by the user to be an Input
RCLK
or an Output pin following configuration. All unrestricted I/O
During Master Parallel mode configuration, each change pins, plus the special pins mentioned on the following page,
on the A0-15 outputs is preceded by a rising edge on have a weak pull-up resistor that becomes active as soon
RCLK, a redundant output signal. After configuration is as the device powers up, and stays active until the end of
complete, this pin becomes a user-programmed I/O pin. configuration.
Note: Before and during configuration, all outputs that are not used for the configuration process are 3-stated with a weak
pull-up resistor.

R
Pin Functions During Configuration

Configuration Mode <M2:M1:M0> *** ** ****
SLAVE MASTER- MASTER- MASTER- 100
SERIAL SERIAL PERIPH HIGH LOW 44 64 68 84 84 100 VQFP 132 144 160 175 176 208 User
<1:1:1> <0:0:0> <1:0:1> <1:1:0> <1:0:0> PLCC VQFP PLCC PLCC PGA PQFP TQFP PGA TQFP PQFP PGA TQFP PQFP Function
POWR POWER POWER POWER POWER POWER

DWN DWN DWN DWN DWN DWN
(I) (I) (I) (I) (I) 7 17 10 12 B2 29 26 A1 1 159 B2 1 3 (1)
M1 (HIGH) (I) M1 (LOW) (I) M1 (LOW) (I) M1 (HIGH) (I) M1 (LOW) (I) 16 31 25 31 J2 52 49 B13 36 40 B14 45 48 RDATA
M0 (HIGH) (I) M0 (LOW) (I) M0 (HIGH) (I) M0 (LOW) (I) M0 (LOW) (I) 17 32 26 32 L1 54 51 A14 38 42 B15 47 50 RTRIG (I)
M2 (HIGH) (I) M2 (LOW) (I) M2 (HIGH) (I) M2 (HIGH) (I) M2 (HIGH) (I) 18 33 27 33 K2 56 53 C13 40 44 C15 49 56 I/O
HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) 19 34 28 34 K3 57 54 B14 41 45 E14 50 57 I/O
LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) 20 36 30 36 L3 59 56 D14 45 49 D16 54 61 I/O
INIT* INIT* INIT* INIT* INIT* 22 40 34 42 K6 65 62 G14 53 59 H15 65 77 I/O
GND GND GND GND GND 23 41 35 43 J6 66 63 H12 55 61 J14 67 79 GND
26 47 43 53 L11 76 73 M13 69 76 P15 85 100 XTL2 OR I/O
RESET (I) RESET (I) RESET (I) RESET (I) RESET (I) 27 48 44 54 K10 78 75 P14 71 78 R15 87 102 RESET (I)
DONE DONE DONE DONE DONE 28 49 45 55 J10 80 77 N13 73 80 R14 89 107 PROGRAM (I)
DATA 7 (I) DATA 7 (I) DATA 7 (I) 50 46 56 K11 81 78 M12 74 81 N13 90 109 I/O
30 51 47 57 J11 82 79 P13 75 82 T14 91 110 XTL1 OR I/O
DATA 6 (I) DATA 6 (I) DATA 6 (I) 52 48 58 H10 83 80 N11 78 86 P12 96 115 I/O
DATA 5 (I) DATA 5 (I) DATA 5 (I) 53 49 60 F10 87 84 M9 84 92 T11 102 122 I/O
CS0 (I) 54 50 61 G10 88 85 N9 85 93 R10 103 123 I/O
DATA 4 (I) DATA 4 (I) DATA 4 (I) 55 51 62 G11 89 86 N8 88 96 R9 108 128 I/O
DATA 3 (I) DATA 3 (I) DATA 3 (I) 57 53 65 F11 92 89 N7 92 102 P8 112 132 I/O
CS1 (I) 58 54 66 E11 93 90 P6 93 103 R8 113 133 I/O
DATA 2 (I) DATA 2 (I) DATA 2 (I) 59 55 67 E10 94 91 M6 96 106 R7 118 138 I/O
DATA 1 (I) DATA 1 (I) DATA 1 (I) 60 56 70 D10 98 95 M5 102 114 R5 124 145 I/O
RDY/BUSY RCLK RCLK 61 57 71 C11 99 96 N4 103 115 P5 125 146 I/O
DIN (I) DIN (I) DATA 0 (I) DATA 0 (I) DATA 0 (I) 38 62 58 72 B11 100 97 N2 106 119 R3 130 151 I/O
DOUT DOUT DOUT DOUT DOUT 39 63 59 73 C10 1 98 M3 107 120 N4 131 152 I/O
CCLK (I) CCLK (O) CCLK (O) CCLK (O) CCLK (O) 40 64 60 74 A11 2 99 P1 108 121 R2 132 153 CCLK (I)
WS (I) A0 A0 1 61 75 B10 5 2 M2 111 124 P2 135 161 I/O
CS2 (I) A1 A1 2 62 76 B9 6 3 N1 112 125 M3 136 162 I/O
A2 A2 3 63 77 A10 8 5 L2 115 128 P1 140 165 I/O
A3 A3 4 64 78 A9 9 6 L1 116 129 N1 141 166 I/O
A15 A15 65 81 B6 12 9 K1 119 132 M1 146 172 5
A4 A4 5 66 82 B7 13 10 J2 120 133 L2 147 173 I/O
A14 A14 6 67 83 A7 14 11 H1 123 136 K2 150 178 I/O
A5 A5 7 68 84 C7 15 12 H2 124 137 K1 151 179 I/O
A13 A13 9 2 2 A6 17 14 G2 128 141 H2 156 184 I/O
A6 A6 10 3 3 A5 18 15 G1 129 142 H1 157 185 I/O
A12 A12 11 4 4 B5 19 16 F2 133 147 F2 164 192 I/O
A7 A7 12 5 5 C5 20 17 E1 134 148 E1 165 193 I/O
A11 A11 13 6 8 A3 23 20 D1 137 151 D1 169 199 I/O
A8 A8 14 7 9 A2 24 21 D2 138 152 C1 170 200 I/O
A10 A10 15 8 10 B3 25 22 B1 141 155 E3 173 203 I/O
A9 A9 16 9 11 A1 26 26 C2 142 156 C2 174 204 I/O
All Others
X X X X XC3x20A etc.
X X X X X X X XC3x30A etc.
X X X X X X XC3x42A etc.
X** X X XC3x64A etc.
X** X X X X X XC3x90A etc.
Notes: X** X X X XC3195A
Generic I/O pins are not shown.
For a detailed description of the configuration modes, see page 25 through page 34.
For pinout details, see page 65 through page 76.
Represents a weak pull-up before and during configuration.
* INIT is an open drain output during configuration.
(I) Represents an input.
** Pin assignment for the XC3064A/XC3090A and XC3195A differ from those shown.
*** Peripheral mode and master parallel mode are not supported in the PC44 package.
**** Pin assignments for the XC3195A PQ208 differ from those shown.
Pin assignments of PGA Footprint PLCC sockets and PGA packages are not identical.
The information on this page is provided as a convenient summary. For detailed pin descriptions, see the preceding two pages.
Note: Before and during configuration, all outputs that are not used for the configuration process are 3-stated with a weak pull-up resistor.

R
XC3000A Switching Characteristics

Xilinx maintains test specifications for each product as controlled documents. To insure the use of the most recently released
device performance parameters, please request a copy of the current test-specification revision.
XC3000A Operating Conditions

VCC Supply voltage relative to GND Commercial 0°C to +85°C junction 4.75 5.25 V
Supply voltage relative to GND Industrial -40°C to +100°C junction 4.5 5.5 V
VIHT High-level input voltage — TTL configuration 2.0 VCC V
VILT Low-level input voltage — TTL configuration 0 0.8 V
VIHC High-level input voltage — CMOS configuration 70% 100% VCC
VILC Low-level input voltage — CMOS configuration 0 20% VCC
XC3000A DC Characteristics Over Operating Conditions

VOH High-level output voltage (@ IOH = –4.0 mA, VCC min) 3.86 V
Commercial
VOL Low-level output voltage (@ IOL = 4.0 mA, VCC min) 0.40 V 7
Industrial
VOL Low-level output voltage (@ IOL = 4.0 mA, VCC min) 0.40 V
VCCPD Power-down supply voltage (PWRDWN must be Low) 2.30 V
ICCPD Power-down supply current
(VCC(MAX) @ TMAX) 3020A 100 µA
3030A 160 µA
3042A 240 µA
3064A 340 µA
3090A 500 µA
Quiescent FPGA supply current in addition to ICCPD
ICCO Chip thresholds programmed as CMOS levels 500 µA
Chip thresholds programmed as TTL levels 10 µA
IIL Input Leakage Current –10 +10 µA
Input capacitance, all packages except PGA175
(sample tested)
All Pins except XTL1 and XTL2 10 pF
XTL1 and XTL2 15 pF
CIN
Input capacitance, PGA 175
(sample tested)
XTL1 and XTL2 20 pF
IRIN Pad pull-up (when selected) @ VIN = 0 V3 0.02 0.17 mA
IRLL Horizontal Longline pull-up (when selected) @ logic Low 3.4 mA
Notes: 1. With no output current loads, no active input or Longline pull-up resistors, all package pins at VCC or GND, and the FPGA
device configured with a tie option.
2. Total continuous output sink current may not exceed 100 mA per ground pin. Total continuous output source may not exceed
100 mA per VCC pin. The number of ground pins varies from the XC3020A to the XC3090A.
3. Not tested. Allow an undriven pin to float High. For any other purposes use an external pull-up.

R
XC3000A Absolute Maximum Ratings

VCC Supply voltage relative to GND –0.5 to +7.0 V
VIN Input voltage with respect to GND –0.5 to VCC +0.5 V
VTS Voltage applied to 3-state output –0.5 to VCC +0.5 V
TSTG Storage temperature (ambient) –65 to +150 °C
TSOL Maximum soldering temperature (10 s @ 1/16 in.) +260 °C
Junction temperature plastic +125 °C
TJ
Junction temperature ceramic +150 °C
Note: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are
Recommended Operating Conditions is not implied. Exposure to Absolute Maximum Ratings conditions for extended
periods of time may affect device reliability.
XC3000A Global Buffer Switching Characteristics Guidelines

Speed Grade -7 -6
Description Symbol Max Max Units
Global and Alternate Clock Distribution1
Either: Normal IOB input pad through clock buffer
to any CLB or IOB clock input TPID 7.5 7.0 ns
Or: Fast (CMOS only) input pad through clock
buffer to any CLB or IOB clock input TPIDC 6.0 5.7 ns
TBUF driving a Horizontal Longline (L.L.)1
I to L.L. while T is Low (buffer active) TIO 4.5 4.0 ns
T↓ to L.L. active and valid with single pull-up resistor TON 9.0 8.0 ns
T↓ to L.L. active and valid with pair of pull-up resistors TON 11.0 10.0 ns
T↑ to L.L. High with single pull-up resistor TPUS 16.0 14.0 ns
T↑ to L.L. High with pair of pull-up resistors TPUF 10.0 8.0 ns
BIDI
Bidirectional buffer delay TBIDI 1.7 1.5 ns
Note: 1. Timing is based on the XC3042A, for other devices see timing calculator.

R
XC3000A CLB Switching Characteristics Guidelines

functionally tested. Since many internal timing parameters cannot be measured directly, they are derived from benchmark
timing patterns. The following guidelines reflect worst-case values over the recommended operating conditions. For more
detailed, more precise, and more up-to-date timing information, use the values provided by the timing calculator and used
in the simulator.
Speed Grade -7 -6
Description Symbol Min Max Min Max Units
Combinatorial Delay
Logic Variables A, B, C, D, E, to outputs X or Y
FG Mode 1 TILO 5.1 4.1 ns
F and FGM Mode 5.6 4.6 ns
Sequential delay
Clock k to outputs X or Y 8 TCKO 4.5 4.0 ns
Clock k to outputs X or Y when Q is returned
through function generators F or G to drive X or Y
FG Mode TQLO 9.5 8.0 ns
Set-up time before clock K
Logic Variables A, B, C, D, E
FG Mode 2 TICK 4.5 3.5 ns
Data In DI 4 TDICK 4.0 3.0 ns
Enable Clock EC 6 TECCK 4.5 4.0 ns
Hold Time after clock K 7
Logic Variables A, B, C, D, E 3 TCKI 0 0 ns
Data In DI2 5 TCKDI 1.0 1.0 ns
Enable Clock EC 7 TCKEC 2.0 2.0 ns
Clock
Clock High time 11 TCH 4.0 3.5 ns
Clock Low time 12 TCL 4.0 3.5 ns
Max. flip-flop toggle rate FCLK 113.0 135.0 MHz
Reset Direct (RD)
RD width 13 TRPW 6.0 5.0 ns
delay from RD to outputs X or Y 9 TRIO 6.0 5.0 ns
Global Reset (RESET Pad)1
RESET width (Low) TMRW 16.0 14.0 ns
delay from RESET pad to outputs X or Y TMRQ 19.0 17.0 ns
Notes: 1. Timing is based on the XC3042A, for other devices see timing calculator.
2. The CLB K to Q output delay (TCKO, #8) of any CLB, plus the shortest possible interconnect delay, is always longer than the
Data In hold time requirement (TCKDI, #5) of any CLB on the same die.

R
XC3000A CLB Switching Characteristics Guidelines (continued)
CLB Output (X, Y)

(Combinatorial)
1 T ILO
CLB Input (A,B,C,D,E)
2 T ICK 3 T CKI
CLB Clock
12 TCL 11 T CH
4 TDICK 5 TCKDI
CLB Input
(Direct In)
6 T ECCK 7 TCKEC
CLB Input
(Enable Clock)
8 TCKO
CLB Output
(Flip-Flop)
CLB Input
(Reset Direct)
13 TRPW
9 T RIO T
CLB Output
(Flip-Flop)
X5424

R
XC3000A IOB Switching Characteristics Guidelines

in the simulator.
Speed Grade -7 -6
Propagation Delays (Input)
Pad to Direct In (I) 3 TPID 4.0 3.0 ns
Pad to Registered In (Q) with latch transparent TPTG 15.0 14.0 ns
Clock (IK) to Registered In (Q) 4 TIKRI 3.0 2.5 ns
Set-up Time (Input)
Pad to Clock (IK) set-up time 1 TPICK 14.0 12.0 ns
Propagation Delays (Output)
Clock (OK) to Pad (fast) 7 TOKPO 8.0 7.0 ns
same (slew rate limited) 7 TOKPO 18.0 15.0 ns
Output (O) to Pad (fast) 10 TOPF 6.0 5.0 ns
same (slew-rate limited) 10 TOPS 16.0 13.0 ns
3-state to Pad begin hi-Z (fast) 9 TTSHZ 10.0 9.0 ns
same (slew-rate limited) 9 TTSHZ 20.0 12.0 ns
3-state to Pad active and valid (fast) 8 TTSON 11.0 10.0 ns
same (slew -rate limited) 8 TTSON 21.0 18.0 ns
Set-up and Hold Times (Output)
Output (O) to clock (OK) set-up time 5 TOOK 8.0 7.0 ns 7
Output (O) to clock (OK) hold time 6 TOKO 0 0 ns
Clock
Clock High time 11 TIOH 4.0 3.5 ns
Clock Low time 12 TIOL 4.0 3.5 ns
Max. flip-flop toggle rate FCLK 113.0 135.0 MHz
Global Reset Delays (based on XC3042A)
RESET Pad to Registered In (Q) 13 TRRI 24.0 23.0 ns
RESET Pad to output pad (fast) 15 TRPO 33.0 29.0 ns
(slew-rate limited) 15 TRPO 43.0 37.0 ns
Notes: 1. Timing is measured at pin threshold, with 50 pF external capacitive loads (incl. test fixture). Typical slew rate limited output
rise/fall times are approximately four times longer.
2. Voltage levels of unused (bonded and unbonded) pads must be valid logic levels. Each can be configured with the internal
pull-up resistor or alternatively configured as a driven output or driven from an external source.
3. Input pad set-up time is specified with respect to the internal clock (ik). In order to calculate system set-up time, subtract
clock delay (pad to ik) from the input pad set-up time value. Input pad holdtime with respect to the internal clock (ik) is
negative. This means that pad level changes immediately before the internal clock edge (ik) will not be recognized.
4. TPID, TPTG, and TPICK are 3 ns higher for XTL2 when the pin is configured as a user input.

R
XC3000A IOB Switching Characteristics Guidelines (continued)
I/O Block (I)
3 T PID
I/O Pad Input
1 T PICK
I/O Clock (IK/OK)
12 TIOL 11 TIOH
I/O Block (RI)
4 TIKRI 13 TRRI
RESET
5 TOOK 6 TOKO 15 TRPO
I/O Block (O)
10 TOP
I/O Pad Output

(Direct)
7 TOKPO
I/O Pad Output
(Registered)
I/O Pad TS
8 TTSON 9 T TSHZ
I/O Pad Output
X5425
Vcc
OUT 3-STATE OUTPUT SLEW PASSIVE

3- STATE T
(OUTPUT ENABLE)
O D Q OUTPUT
OUT
BUFFER
FLIP
FLOP
I/O PAD
R
I
DIRECT IN
Q
REGISTERED IN Q D
FLIP TTL or
FLOP CMOS
or INPUT
LATCH THRESHOLD
CK1
CK2
PROGRAM
CONTROLLED

R
XC3000L Switching Characteristics

XC3000L Operating Conditions

VIH High-level input voltage — TTL configuration 2.0 VCC+0.3 V
VIL Low-level input voltage — TTL configuration -0.3 0.8 V
Notes: 1. At junction temperatures above those listed as Operating Conditions, all delay parameters increase by 0.3% per °C.
2. Although the present (1996) devices operate over the full supply voltage range from 3.0 to 5.25 V, Xilinx reserves the right to
restrict operation to the 3.0 to 3.6 V range later, when smaller device geometries might preclude operation at 5V. Operating
conditions are guaranteed in the 3.0 – 3.6 V VCC range.
XC3000L DC Characteristics Over Operating Conditions

VOH High-level output voltage (@ IOH = –4.0 mA, VCC min) VCC -0.2 V
7
ICCPD Power-down supply current (VCC(MAX) @ TMAX) 10 µA
Quiescent FPGA supply current in addition to ICCPD1
ICCO
Chip thresholds programmed as CMOS levels 20 µA
(sample tested)
XTL1 and XTL2 15 pF
CIN
(sample tested)
XTL1 and XTL2 20 pF
IRLL Horizontal Longline pull-up (when selected) @ logic Low 2.50 mA
Notes: 1. With no output current loads, no active input or Longline pull-up resistors, all package pins at VCC or GND, and the FPGA
device configured with a tie option. ICCO is in addition to ICCPD.
2. Total continuous output sink current may not exceed 100 mA per ground pin. Total continuous output source may not exceed
100 mA per VCC pin. The number of ground pins varies from the XC3020L to the XC3090L.
3. Not tested. Allows an undriven pin to float High. For any other purpose, use an external pull-up.

R
XC3000L Absolute Maximum Ratings

TJ
XC3000L Global Buffer Switching Characteristics Guidelines

Speed Grade -8
Description Symbol Max Units
to any CLB or IOB clock input TPID 9.0 ns
buffer to any CLB or IOB clock input TPIDC 7.0 ns
I to L.L. while T is Low (buffer active) TIO 5.0 ns
T↓ to L.L. active and valid with single pull-up resistor TON 12.0 ns
T↑ to L.L. High with single pull-up resistor TPUS 24.0 ns
BIDI
Bidirectional buffer delay TBIDI 2.0 ns
Notes: 1. Timing is based on the XC3042A, for other devices see timing calculator.
2. The use of two pull-up resistors per Longline, available on other XC3000 devices, is not a valid option for XC3000L devices.

R
XC3000L CLB Switching Characteristics Guidelines

in the simulator.
Speed Grade -8
Combinatorial Delay
Logic Variables A, B, C, D, E, to outputs X or Y
FG Mode 1 TILO 6.7 ns
F and FGM Mode 7.5 ns
Sequential delay
Clock k to outputs X or Y 8 TCKO 7.5 ns
through function generators F or G to drive X or Y
FG Mode TQLO 14.0 ns
Logic Variables A, B, C, D, E
FG Mode 2 TICK 5.0 ns
Data In DI 4 TDICK 5.0 ns
Enable Clock EC 6 TECCK 6.0 ns
Hold Time after clock K 7
Logic Variables A, B, C, D, E 3 TCKI 0 ns
Data In DI2 5 TCKDI 2.0 ns
Enable Clock EC 7 TCKEC 2.0 ns
Clock
Clock High time 11 TCH 5.0 ns
Clock Low time 12 TCL 5.0 ns
Max. flip-flop toggle rate FCLK 80.0 MHz
Reset Direct (RD)
RD width 13 TRPW 7.0 ns
delay from RD to outputs X or Y 9 TRIO 7.0 ns
RESET width (Low) TMRW 16.0 ns
delay from RESET pad to outputs X or Y TMRQ 23.0 ns
Notes: 1. Timing is based on the XC3042L, for other devices see timing calculator.
2. The CLB K to Q output delay (TCKO, #8) of any CLB, plus the shortest possible interconnect delay, is always longer than the

R
XC3000L CLB Switching Characteristics Guidelines (continued)
CLB Output (X, Y)

(Combinatorial)
1 T ILO
2 T ICK 3 T CKI
CLB Clock
12 TCL 11 T CH
4 TDICK 5 TCKDI
CLB Input
(Direct In)
6 T ECCK 7 TCKEC
CLB Input
(Enable Clock)
8 TCKO
CLB Output
(Flip-Flop)
CLB Input
(Reset Direct)
13 TRPW
9 T RIO T
CLB Output
(Flip-Flop)
X5424

R
XC3000L IOB Switching Characteristics Guidelines

in the simulator.
Speed Grade -8
Pad to Direct In (I) 3 TPID 5.0 ns
Pad to Registered In (Q) with latch transparent TPTG 24.0 ns
Clock (IK) to Registered In (Q) 4 TIKRI 6.0 ns
Set-up Time (Input)
Pad to Clock (IK) set-up time 1 TPICK 22.0 ns
Clock (OK) to Pad (fast) 7 TOKPO 12.0 ns
same (slew rate limited) 7 TOKPO 28.0 ns
Output (O) to Pad (fast) 10 TOPF 9.0 ns
same (slew-rate limited) 10 TOPS 25.0 ns
3-state to Pad begin hi-Z (fast) 9 TTSHZ 12.0 ns
same (slew-rate limited) 9 TTSHZ 28.0 ns
3-state to Pad active and valid (fast) 8 TTSON 16.0 ns
same (slew -rate limited) 8 TTSON 32.0 ns
Output (O) to clock (OK) set-up time 5 TOOK 12.0 ns 7
Output (O) to clock (OK) hold time 6 TOKO 0 ns
Clock
Clock High time 11 TIOH 5.0 ns
Clock Low time 12 TIOL 5.0 ns
Max. flip-flop toggle rate FCLK 80.0 MHz
Global Reset Delays (based on XC3042L)
RESET Pad to Registered In (Q) 13 TRRI 25.0 ns
RESET Pad to output pad (fast) 15 TRPO 35.0 ns
(slew-rate limited) 15 TRPO 51.0 ns

R
XC3000L IOB Switching Characteristics Guidelines (continued)
I/O Block (I)
3 T PID
I/O Pad Input
1 T PICK
I/O Clock (IK/OK)
12 TIOL 11 TIOH
I/O Block (RI)
4 TIKRI 13 TRRI
RESET
I/O Block (O)
10 TOP
I/O Pad Output

(Direct)
7 TOKPO
I/O Pad Output
(Registered)
I/O Pad TS
8 TTSON 9 T TSHZ
I/O Pad Output
X5425
Vcc

3- STATE T
(OUTPUT ENABLE)
O D Q OUTPUT
OUT
BUFFER
FLIP
FLOP
I/O PAD
R
I
DIRECT IN
Q
REGISTERED IN Q D
FLIP TTL or
FLOP CMOS
or INPUT
LATCH THRESHOLD
CK1
CK2
PROGRAM
CONTROLLED

R
XC3100A Switching Characteristics

XC3100A Operating Conditions

Supply voltage relative to GND Commercial 0°C to +85°C junction 4.25 5.25 V
VCC
Supply voltage relative to GND Industrial -40°C to +100°C junction 4.5 5.5 V
XC3100A DC Characteristics Over Operating Conditions

Commercial
VOL Low-level output voltage (@ IOL = 8.0 mA, VCC min) 0.40 V 7
Industrial
Quiescent LCA supply current in addition to ICCPD1
ICCO Chip thresholds programmed as CMOS levels 8 mA
Chip thresholds programmed as TTL levels 14 mA
(sample tested)
XTL1 and XTL2 15 pF
CIN
(sample tested)
XTL1 and XTL2 20 pF
Notes: 1. With no output current loads, no active input or Longline pull-up resistors, all package pins at VCC or GND, and the LCA
device configured with a tie option.
2. Total continuous output sink current may not exceed 100 mA per ground pin. The number of ground pins varies from two for
the XC3120A in the PC84 package, to eight for the XC3195A in the PQ208 package.
3. Not tested. Allows an undriven pin to float High. For any other purpose, use an external pull-up.

R
XC3100A Absolute Maximum Ratings

TJ
XC3100A Global Buffer Switching Characteristics Guidelines

Speed Grade -4 -3 -2 -1 -09
Description Symbol Max Max Max Max Max Units
to any CLB or IOB clock input TPID 6.5 5.6 4.7 4.3 3.9 ns
buffer to any CLB or IOB clock input TPIDC 5.1 4.3 3.7 3.5 3.1 ns
I to L.L. while T is Low (buffer active) (XC3100) TIO 3.7 3.1 ns
(XC3100A) TIO 3.6 3.1 3.1 2.9 2.1 ns
T↓ to L.L. active and valid with single pull-up resistor TON 5.0 4.2 4.2 4.0 3.1 ns
T↓ to L.L. active and valid with pair of pull-up resistors TON 6.5 5.7 5.7 5.5 4.6 ns
T↑ to L.L. High with single pull-up resistor TPUS 13.5 11.4 11.4 10.4 8.9 ns
T↑ to L.L. High with pair of pull-up resistors TPUF 10.5 8.8 8.1 7.1 5.9 ns
BIDI
Bidirectional buffer delay TBIDI 1.2 1.0 0.9 0.85 0.75 ns
Prelim
Note: 1. Timing is based on the XC3142A, for other devices see timing calculator.
The use of two pull-up resistors per longline, available on other XC3000 devices, is not a valid design option for XC3100A
devices.

R
XC3100A CLB Switching Characteristics Guidelines

in the simulator.
Speed Grade -4 -3 -2 -1 -09
Description Symbol Min Max Min Max Min Max Min Max Min Max Units
Combinatorial Delay
Logic Variables A, B, C, D, E, 1 TILO 3.3 2.7 2.2 1.75 1.5 ns
to outputs X or Y
Sequential delay
Clock k to outputs X or Y 8 TCKO 2.5 2.1 1.7 1.4 1.25 ns
through function generators F or G to drive
X or Y TQLO 5.2 4.3 3.5 3.1 2.7 ns
Logic Variables A, B, C, D, E 2 TICK 2.5 2.1 1.8 1.7 1.5 ns
Data In DI 4 TDICK 1.6 1.4 1.3 1.2 1.0 ns
Enable Clock EC 6 TECCK 3.2 2.7 2.5 2.3 2.05 ns
Reset Direct inactive RD 1.0 1.0 1.0 1.0 1.0 ns
Hold Time after clock K
Logic Variables A, B, C, D, E 3 TCKI 0 0 0 0 0 ns
Data In DI 5 TCKDI 1.0 0.9 0.9 0.8 0.7 ns
Enable Clock EC 7 TCKEC 0.8 0.7 0.7 0.6 0.55 ns
Clock
Clock High time 11 TCH 2.0 1.6 1.3 1.3 1.3 ns
Clock Low time 12 TCL 2.0 1.6 1.3 1.3 1.3 ns
Max. flip-flop toggle rate FCLK 227 270 323 323 370 MHz 7
Reset Direct (RD)
RD width 13 TRPW 3.2 2.7 2.3 2.3 2.05 ns
delay from RD to outputs X or Y 9 TRIO 3.7 3.1 2.7 2.4 2.15 ns
RESET width (Low) (XC3142A) TMRW 14.0 12.0 12.0 12.0 12.0 ns
delay from RESET pad to outputs X or Y TMRQ 14.0 12.0 12.0 12.0 12.0 ns
Prelim
Notes: 1. The CLB K to Q output delay (TCKO, #8) of any CLB, plus the shortest possible interconnect delay, is always longer than the
2. TILO, TQLO and TICK are specified for 4-input functions. For 5-input functions or base FGM functions, each of these
specifications for the XC3100A family increases by 0.50 ns (-5), 0.42 ns (-4) and 0.35 ns (-3), 0.35 ns (-2), 0.30 ns (-1), and
0.30 ns (-09).

R
XC3100A CLB Switching Characteristics Guidelines (continued)
CLB Output (X, Y)

(Combinatorial)
1 T ILO
2 T ICK 3 T CKI
CLB Clock
12 TCL 11 T CH
4 TDICK 5 TCKDI
CLB Input
(Direct In)
6 T ECCK 7 TCKEC
CLB Input
(Enable Clock)
8 TCKO
CLB Output
(Flip-Flop)
CLB Input
(Reset Direct)
13 TRPW
9 T RIO T
CLB Output
(Flip-Flop)
X5424

R
XC3100A IOB Switching Characteristics Guidelines

in the simulator.
Speed Grade -4 -3 -2 -1 -09
Description Symbol Min Max Min Max Min Max Min Max Min Max Units
Pad to Direct In (I) 3 TPID 2.5 2.2 2.0 1.7 1.55 ns
Pad to Registered In (Q)
with latch transparent(XC3100A)Clock (IK) TPTG 12.0 11.0 11.0 10.0 9.2 ns
to Registered In (Q) 4 TIKRI 2.5 2.2 1.9 1.7 1.55 ns
Set-up Time (Input)
Pad to Clock (IK) set-up time
XC3120A, XC3130A 1 TPICK 10.6 9.4 8.9 8.0 7.2 ns
XC3142A 10.7 9.5 9.0 8.1 7.3 ns
XC3164A 11.0 9.7 9.2 8.3 7.5 ns
XC3190A 11.2 9.9 9.4 8.5 7.7 ns
XC3195A 11.6 10.3 9.8 8.9 8.1 ns
Clock (OK) to Pad (fast) 7 TOKPO 5.0 4.4 3.7 3.4 3.3 ns
same (slew rate limited) 7 TOKPO 12.0 10.0 9.7 8.4 6.9 ns
Output (O) to Pad (fast) 10 TOPF 3.7 3.3 3.0 3.0 2.9 ns
same (slew-rate limited) ns
(XC3100A) 10 TOPS 11.0 9.0 8.7 8.0 6.5 ns
3-state to Pad
begin hi-Z (fast) 9 TTSHZ 6.2 5.5 5.0 4.5 4.05 ns
same (slew-rate limited) 9 TTSHZ 6.2 5.5 5.0 4.5 4.05 ns 7
3-state to Pad
active and valid (fast) (XC3100A) 8 TTSON 10.0 9.0 8.5 6.5 5.0 ns
same (slew -rate limited) 8 TTSON 17.0 15.0 14.2 11.5 8.6 ns
Output (O) to clock (OK) set-up time
(XC3100A) 5 TOOK 4.5 3.6 3.2 2.9 ns
Output (O) to clock (OK) hold time 6 TOKO 0 0 0 ns
Clock
Clock High time 11 TIOH 2.0 1.6 1.3 1.3 1.3 ns
Clock Low time 12 TIOL 2.0 1.6 1.3 1.3 1.3 ns
Max. flip-flop toggle rate FCLK 227 270 323 323 370 MHz
Global Reset Delays
RESET Pad to Registered In (Q)
(XC3142A) 13 TRRI 15.0 13.0 13.0 13.0 14.4 ns
(XC3190A) 25.5 21.0 21.0 21.0 21.0 ns
RESET Pad to output pad (fast) 15 TRPO 20.0 17.0 17.0 17.0 17.0 ns
(slew-rate limited) 15 TRPO 27.0 23.0 23.0 22.0 21.0 ns
Preliminary
Notes: 1. Timing is measured at pin threshold, with 50 pF external capacitive loads (incl. test fixture). For larger capacitive loads, see
XAPP024. Typical slew rate limited output rise/fall times are approximately four times longer.

R
XC3100A IOB Switching Characteristics Guidelines (continued)
I/O Block (I)
3 T PID
I/O Pad Input
1 T PICK
I/O Clock (IK/OK)
12 TIOL 11 TIOH
I/O Block (RI)
4 TIKRI 13 TRRI
RESET
I/O Block (O)
10 TOP
I/O Pad Output

(Direct)
7 TOKPO
I/O Pad Output
(Registered)
I/O Pad TS
8 TTSON 9 T TSHZ
I/O Pad Output
X5425
Vcc

3- STATE T
(OUTPUT ENABLE)
O D Q OUTPUT
OUT
BUFFER
FLIP
FLOP
I/O PAD
R
I
DIRECT IN
Q
REGISTERED IN Q D
FLIP TTL or
FLOP CMOS
or INPUT
LATCH THRESHOLD
CK1
CK2
PROGRAM
CONTROLLED

R
XC3100L Switching Characteristics

XC3100L Operating Conditions

VIH High-level input voltage 2.0 VCC + 0.3 V
VIL Low-level input voltage -0.3 0.8 V
Notes: 1. At junction temperatures above those listed as Operating Conditions, all delay parameters increase by 0.3% per °C.
2. Although the present (1996) devices operate over the full supply voltage range from 3.0 V to 5.25 V, Xilinx reserves the right
to restrict operation to the 3.0 and 3.6 V range later, when smaller device geometries might preclude operation @ 5 V.
Operating conditions are guaranteed in the 3.0 – 3.6 V VCC range.
XC3100L DC Characteristics Over Operating Conditions

High-level output voltage (@ IOH = -4.0 mA, VCC min) 2.4 V
VOH
High-level output voltage (@ IOH = -100.0 µA, VCC min) VCC -0.2 V
Low-level output voltage (@ IOH = 4.0 mA, VCC min) 0.40 V
VOL
Low-level output voltage (@ IOH = +100.0 µA, VCC min) 0.2 V 7
ICCO Quiescent FPGA supply current 1.5 mA
Chip thresholds programmed as CMOS levels1
IIL Input Leakage Current -10 +10 µA
Input capacitance
(sample tested)
CIN
All pins except XTL1 and XTL2 10 pF
XTL1 and XTL2 15 pF
IRIN Pad pull-up (when selected) @ VIN = 0 V 3 0.02 0.17 mA
IRLL Horizontal long line pull-up (when selected) @ logic Low 0.20 2.80 mA
Notes: 1. With no output current loads, no active input or long line pull-up resistors, all package pins at VCC or GND, and the FPGA
configured with a tie option.
2. Total continuous output sink current may not exceed 100 mA per ground pin. Total continuous output source current may not
exceed 100 mA per VCC pin. The number of ground pins varies from the XC3142L to the XC3190L.
3. Not tested. Allows undriven pins to float High. For any other purpose, use an external pull-up.

R
XC3100L Absolute Maximum Ratings

TJ
XC3100L Global Buffer Switching Characteristics Guidelines

Speed Grade -3 -2
Description Symbol Max Max Units
Either:Normal IOB input pad through clock buffer
to any CLB or IOB clock input TPID 5.6 4.7 ns
buffer to any CLB or IOB clock input TPIDC 4.3 3.7 ns
I to L.L. while T is Low (buffer active) TIO 3.1 3.1 ns
T↓ to L.L. active and valid with single pull-up resistor TON 4.2 4.2 ns
T↑ to L.L. High with single pull-up resistor TPUS 11.4 11.4 ns
BIDI
Bidirectional buffer delay TBIDI 1.0 0.9 ns
Advance
Notes: 1. Timing is based on the XC3142L, for other devices see timing calculator.
2. The use of two pull-up resistors per longline, available on other XC3000 devices, is not a valid option for XC3100L devices.

R
XC3100L CLB Switching Characteristics Guidelines

in the simulator.
Speed Grade -3 -2
Combinatorial Delay
Logic Variables A, B, C, D, E, to outputs X or Y 1 TILO 2.7 2.2 ns
Sequential delay
Clock k to outputs X or Y 8 TCKO 2.1 1.7 ns
through function generators F or G to drive X or Y TQLO 4.3 3.5 ns
Logic Variables A, B, C, D, E 2 TICK 2.1 1.8 ns
Data In DI 4 TDICK 1.4 1.3 ns
Enable Clock EC 6 TECCK 2.7 2.5 ns
Reset Direct Inactive RD 1.0 1.0 ns
Hold Time after clock K
Logic Variables A, B, C, D, E 3 TCKI 0 0 ns
Data In DI 5 TCKDI 0.9 0.9 ns
Enable Clock EC 7 TCKEC 0.7 0.7 ns
Clock
Clock High time 11 TCH 1.6 1.3 ns 7
Clock Low time 12 TCL 1.6 1.3 ns
Max. flip-flop toggle rate FCLK 270 325 MHz
Reset Direct (RD)
RD width 13 TRPW 2.7 2.3 ns
delay from RD to outputs X or Y 9 TRIO 3.1 2.7 ns
Global Reset (RESET Pad)
RESET width (Low) ns
(XC3142L) TMRW 12.0 12.0 ns
delay from RESET pad to outputs X or Y TMRQ 12.0 12.0
Advance
Notes: 1. The CLB K to Q delay (TCKO, #8) of any CLB, plus the shortest possible interconnect delay, is always longer than the Data
In hold time requirement (TCKDI, #5) of any CLB on the same die.
2. TILO, TQLO and TICK are specified for 4-input functions. For 5-input functions or base FGM functions, each of these
specifications for the XC3100L family increase by 0.35 ns (-3) and 0.29 ns (-2).

R
XC3100L CLB Switching Characteristics Guidelines (continued)
CLB Output (X, Y)

(Combinatorial)
1 T ILO
2 T ICK 3 T CKI
CLB Clock
12 TCL 11 T CH
4 TDICK 5 TCKDI
CLB Input
(Direct In)
6 T ECCK 7 TCKEC
CLB Input
(Enable Clock)
8 TCKO
CLB Output
(Flip-Flop)
CLB Input
(Reset Direct)
13 TRPW
9 T RIO T
CLB Output
(Flip-Flop)
X5424

R
XC3100L IOB Switching Characteristics Guidelines

in the simulator.
Speed Grade -3 -2
Pad to Direct In (I) 3 TPID 2.2 2.0 ns
Pad to Registered In (Q) with latch (XC3100L) TPTG 11.0 11.0 ns
transparent
Clock (IK) to Registered In (Q) 4 TIKRI 2.2 1.9 ns
Set-up Time (Input)
Pad to Clock (IK) set-up time 1 TPICK
XC3142L 9.5 9.0 ns
XC3190L 9.9 9.4 ns
Clock (OK) to Pad (fast) 7 TOKPOTOK 4.4 4.0 ns
same (slew rate limited) 7 PO 10.0 9.7 ns
Output (O) to Pad (fast) 10 TOPF 3.3 3.0 ns
same (slew-rate limited)(XC3100L) 10 TOPF 9.0 8.7 ns
3-state to Pad begin hi-Z (fast) 9 TTSHZ 5.5 5.0 ns
same (slew-rate limited) 9 TTSHZ 5.5 5.0 ns
3-state to Pad active and valid (fast)(XC3100L) 8 TTSON 9.0 8.5 ns 7
same (slew -rate limited) 8 TTSON 15.0 14.2 ns
Output (O) to clock (OK) set-up time (XC3100L) 5 TOOK 4.0 3.6 ns
Output (O) to clock (OK) hold time 6 TOKO 0 0 ns
Clock
Clock High time 11 TIOH 1.6 1.3 ns
Clock Low time 12 TIOL 1.6 1.3 ns
Export Control Maximum flip-flop toggle rate FTOG 270 325 MHz
Global Reset Delays
RESET Pad to Registered In (Q)
(XC3142L) 13 TRRI 16.0 16.0 ns
(XC3190L) 21.0 21.0 ns
RESET Pad to output pad (fast) 15 TRPO 17.0 17.0 ns
(slew-rate limited) 15 TRPO 23.0 23.0 ns
Advance
3. Input pad set-up time is specified with respect to the internal clock (IK). In order to calculate system set-up time, subtract
clock delay (pad to ik) from the input pad set-up time value. Input pad holdtime with respect to the internal clock (IK) is
negative. This means that pad level changes immediately before the internal clock edge (IK) will not be recognized.

R
XC3100L IOB Switching Characteristics Guidelines (continued)
I/O Block (I)
3 T PID
I/O Pad Input
1 T PICK
I/O Clock (IK/OK)
12 TIOL 11 TIOH
I/O Block (RI)
4 TIKRI 13 TRRI
RESET
I/O Block (O)
10 TOP
I/O Pad Output

(Direct)
7 TOKPO
I/O Pad Output
(Registered)
I/O Pad TS
8 TTSON 9 T TSHZ
I/O Pad Output
X5425
Vcc

3- STATE T
(OUTPUT ENABLE)
O D Q OUTPUT
OUT
BUFFER
FLIP
FLOP
I/O PAD
R
I
DIRECT IN
Q
REGISTERED IN Q D
FLIP TTL or
FLOP CMOS
or INPUT
LATCH THRESHOLD
CK1
CK2
PROGRAM
CONTROLLED

R
XC3000 Series Pin Assignments

Xilinx offers the six different array sizes in the XC3000 families in a variety of surface-mount and through-hole package
types, with pin counts from 44 to 208.
Each chip is offered in several package types to accommodate the available PC board space and manufacturing technology.
Most package types are also offered with different chips to accommodate design changes without the need for PC board
changes.
Note that there is no perfect match between the number of bonding pads on the chip and the number of pins on a package.
In some cases, the chip has more pads than there are pins on the package, as indicated by the information (“unused” pads)
below the line in the following table. The IOBs of the unconnected pads can still be used as storage elements if the specified
propagation delays and set-up times are acceptable.
In other cases, the chip has fewer pads than there are pins on the package; therefore, some package pins are not connected
(n.c.), as shown above the line in the following table.
XC3000 Series 44-Pin PLCC Pinouts

XC3000A, XC3000L, and XC3100A families have identical pinouts
Pin No. XC3030A Pin No. XC3030A

1 GND 23 GND
2 I/O 24 I/O 7
3 I/O 25 I/O
4 I/O 26 XTL2(IN)-I/O
5 I/O 27 RESET
6 I/O 28 DONE-PGM
7 PWRDWN 29 I/O
8 TCLKIN-I/O 30 XTL1(OUT)-BCLK-I/O
9 I/O 31 I/O
10 I/O 32 I/O
11 I/O 33 I/O
12 VCC 34 VCC
13 I/O 35 I/O
14 I/O 36 I/O
15 I/O 37 I/O
16 M1-RDATA 38 DIN-I/O
17 M0-RTRIG 39 DOUT-I/O
18 M2-I/O 40 CCLK
19 HDC-I/O 41 I/O
20 LDC-I/O 42 I/O
21 I/O 43 I/O
22 INIT-I/O 44 I/O
Peripheral mode and Master Parallel mode are not supported in the PC44 package

R
XC3000 Series 64-Pin Plastic VQFP Pinouts

XC3000A, XC3000L, and XC3100A families have identical pinouts
Pin No. XC3030A Pin No. XC3030A

1 A0-WS-I/O 33 M2-I/O
2 A1-CS2-I/O 34 HDC-I/O
3 A2-I/O 35 I/O
4 A3-I/O 36 LDC-I/O
5 A4-I/O 37 I/O
6 A14-I/O 38 I/O
7 A5-I/O 39 I/O
8 GND 40 INIT-I/O
9 A13-I/O 41 GND
10 A6-I/O 42 I/O
11 A12-I/O 43 I/O
12 A7-I/O 44 I/O
13 A11-I/O 45 I/O
14 A8-I/O 46 I/O
15 A10-I/O 47 XTAL2(IN)-I/O
16 A9-I/O 48 RESET
17 PWRDN 49 DONE-PG
18 TCLKIN-I/O 50 D7-I/O
19 I/O 51 XTAL1(OUT)-BCLKIN-I/O
20 I/O 52 D6-I/O
21 I/O 53 D5-I/O
22 I/O 54 CS0-I/O
23 I/O 55 D4-I/O
24 VCC 56 VCC
25 I/O 57 D3-I/O
26 I/O 58 CS1-I/O
27 I/O 59 D2-I/O
28 I/O 60 D1-I/O
29 I/O 61 RDY/BUSY-RCLK-I/O
30 I/O 62 D0-DIN-I/O
31 M1-RDATA 63 DOUT-I/O
32 M0-RTRIG 64 CCLK

R
XC3000 Series 68-Pin PLCC, 84-Pin PLCC and PGA Pinouts

XC3000A, XC3000L, XC3100A, and XC3100L families have identical pinouts
68 PLCC XC3020A, XC3030A, 68 PLCC XC3020A, XC3030A,
XC3030A XC3020A XC3042A 84 PLCC XC3030A XC3020A XC3042A 84 PLCC
10 10 PWRDN 12 44 44 RESET 54
11 11 TCLKIN-I/O 13 45 45 DONE-PG 55
12 — I/O* 14 46 46 D7-I/O 56
13 12 I/O 15 47 47 XTL1(OUT)-BCLKIN-I/O 57
14 13 I/O 16 48 48 D6-I/O 58
— — I/O 17 — — I/O 59
15 14 I/O 18 49 49 D5-I/O 60
16 15 I/O 19 50 50 CS0-I/O 61
— 16 I/O 20 51 51 D4-I/O 62
17 17 I/O 21 — — I/O 63
18 18 VCC 22 52 52 VCC 64
19 19 I/O 23 53 53 D3-I/O 65
— — I/O 24 54 54 CS1-I/O 66
20 20 I/O 25 55 55 D2-I/O 67
— 21 I/O 26 — — I/O 68
21 22 I/O 27 — — I/O* 69
22 — I/O 28 56 56 D1-I/O 70
23 23 I/O 29 57 57 RDY/BUSY-RCLK-I/O 71
24 24 I/O 30 58 58 D0-DIN-I/O 72
25 25 M1-RDATA 31 59 59 DOUT-I/O 73
26 26 M0-RTRIG 32 60 60 CCLK 74
27 27 M2-I/O 33 61 61 A0-WS-I/O 75
28 28 HDC-I/O 34 62 62 A1-CS2-I/O 76 7
29 29 I/O 35 63 63 A2-I/O 77
30 30 LDC-I/O 36 64 64 A3-I/O 78
— 31 I/O 37 — — I/O* 79
— I/O* 38 — — I/O* 80
31 32 I/O 39 65 65 A15-I/O 81
32 33 I/O 40 66 66 A4-I/O 82
33 — I/O* 41 67 67 A14-I/O 83
34 34 INIT-I/O 42 68 68 A5-I/O 84
35 35 GND 43 1 1 GND 1
36 36 I/O 44 2 2 A13-I/O 2
37 37 I/O 45 3 3 A6-I/O 3
38 38 I/O 46 4 4 A12-I/O 4
39 39 I/O 47 5 5 A7-I/O 5
— 40 I/O 48 — — I/O* 6
— 41 I/O 49 — — I/O* 7
40 I/O* 50 6 6 A11-I/O 8
41 I/O* 51 7 7 A8-I/O 9
42 42 I/O 52 8 8 A10-I/O 10
43 43 XTL2(IN)-I/O 53 9 9 A9-I/O 11
Unprogrammed IOBs have a default pull-up. This prevents an undefined pad level for unbonded or unused IOBs.
Programmed outputs are default slew-rate limited.
This table describes the pinouts of three different chips in three different packages. The pin-description column lists 84 of the
118 pads on the XC3042A (and 84 of the 98 pads on the XC3030A) that are connected to the 84 package pins. Ten pads,
indicated by an asterisk, do not exist on the XC3020A, which has 74 pads; therefore the corresponding pins on the 84-pin
packages have no connections to an XC3020A. Six pads on the XC3020A and 16 pads on the XC3030A, indicated by a
dash (—) in the 68 PLCC column, have no connection to the 68 PLCC, but are connected to the 84-pin packages.

R
XC3064A/XC3090A/XC3195A 84-Pin PLCC Pinouts

PLCC Pin Number XC3064A, XC3090A, XC3195A PLCC Pin Number XC3064A, XC3090A, XC3195A
12 PWRDN 54 RESET
13 TCLKIN-I/O 55 DONE-PG
14 I/O 56 D7-I/O
15 I/O 57 XTL1(OUT)-BCLKIN-I/O
16 I/O 58 D6-I/O
17 I/O 59 I/O
18 I/O 60 D5-I/O
19 I/O 61 CS0-I/O
20 I/O 62 D4-I/O
21 GND* 63 I/O
22 VCC 64 VCC
23 I/O 65 GND*
24 I/O 66 D3-I/O*
25 I/O 67 CS1-I/O*
26 I/O 68 D2-I/O*
27 I/O 69 I/O
28 I/O 70 D1-I/O
29 I/O 71 RDY/BUSY-RCLK-I/O
30 I/O 72 D0-DIN-I/O
31 M1-RDATA 73 DOUT-I/O
32 M0-RTRIG 74 CCLK
33 M2-I/O 75 A0-WS-I/O
34 HDC-I/O 76 A1-CS2-I/O
35 I/O 77 A2-I/O
36 LDC-I/O 78 A3-I/O
37 I/O 79 I/O
38 I/O 80 I/O
39 I/O 81 A15-I/O
40 I/O 82 A4-I/O
41 INIT/I/O* 83 A14-I/O
42 VCC* 84 A5-I/O
43 GND 1 GND
44 I/O 2 VCC*
45 I/O 3 A13-I/O*
46 I/O 4 A6-I/O*
47 I/O 5 A12-I/O*
48 I/O 6 A7-I/O*
49 I/O 7 I/O
50 I/O 8 A11-I/O
51 I/O 9 A8-I/O
52 I/O 10 A10-I/O
53 XTL2(IN)-I/O 11 A9-I/O
* In the PC84 package, XC3064A, XC3090A and XC3195A have additional VCC and GND pins and thus a different pin
definition than XC3020A/XC3030A/XC3042A.

R
XC3000 Series 100-Pin QFP Pinouts

Pin No. XC3020A Pin No. XC3020A Pin No. XC3020A
TQFP XC3030A TQFP XC3030A TQFP XC3030A
PQFP VQFP XC3042A PQFP VQFP XC3042A PQFP VQFP XC3042A
16 13 GND 50 47 I/O* 84 81 I/O*
17 14 A13-I/O 51 48 I/O* 85 82 I/O*
18 15 A6-I/O 52 49 M1-RD 86 83 I/O
19 16 A12-I/O 53 50 GND* 87 84 D5-I/O
20 17 A7-I/O 54 51 MO-RT 88 85 CS0-I/O
21 18 I/O* 55 52 VCC* 89 86 D4-I/O
22 19 I/O* 56 53 M2-I/O 90 87 I/O
23 20 A11-I/O 57 54 HDC-I/O 91 88 VCC
24 21 A8-I/O 58 55 I/O 92 89 D3-I/O
25 22 A10-I/O 59 56 LDC-I/O 93 90 CS1-I/O
26 23 A9-I/O 60 57 I/O* 94 91 D2-I/O
27 24 VCC* 61 58 I/O* 95 92 I/O
28 25 GND* 62 59 I/O 96 93 I/O*
29 26 PWRDN 63 60 I/O 97 94 I/O*
30 27 TCLKIN-I/O 64 61 I/O 98 95 D1-I/O
31 28 I/O** 65 62 INIT-I/O 99 96 RDY/BUSY-RCLK-I/O
32 29 I/O* 66 63 GND 100 97 DO-DIN-I/O
33 30 I/O* 67 64 I/O 1 98 DOUT-I/O
34 31 I/O 68 65 I/O 2 99 CCLK
35 32 I/O 69 66 I/O 3 100 VCC* 7
36 33 I/O 70 67 I/O 4 1 GND*
37 34 I/O 71 68 I/O 5 2 AO-WS-I/O
38 35 I/O 72 69 I/O 6 3 A1-CS2-I/O
39 36 I/O 73 70 I/O 7 4 I/O**
40 37 I/O 74 71 I/O* 8 5 A2-I/O
41 38 VCC 75 72 I/O* 9 6 A3-I/O
42 39 I/O 76 73 XTL2-I/O 10 7 I/O*
43 40 I/O 77 74 GND* 11 8 I/O*
44 41 I/O 78 75 RESET 12 9 A15-I/O
45 42 I/O 79 76 VCC* 13 10 A4-I/O
46 43 I/O 80 77 DONE-PG 14 11 A14-I/O
47 44 I/O 81 78 D7-I/O 15 12 A5-I/O
48 45 I/O 82 79 BCLKIN-XTL1-I/O
49 46 I/O 83 80 D6-I/O
* This table describes the pinouts of three different chips in three different packages. The pin-description column lists 100 of
the 118 pads on the XC3042A that are connected to the 100 package pins. Two pads, indicated by double asterisks, do not
exist on the XC3030A, which has 98 pads; therefore the corresponding pins have no connections. Twenty-six pads,
indicated by single or double asterisks, do not exist on the XC3020A, which has 74 pads; therefore, the corresponding pins
have no connections. (See table on page 65.)

R
XC3000 Series 132-Pin Ceramic and Plastic PGA Pinouts

PGA PGA PGA PGA

Pin XC3042A Pin XC3042A Pin XC3042A Pin XC3042A
Number XC3064A Number XC3064A Number XC3064A Number XC3064A
C4 GND B13 M1-RD P14 RESET M3 DOUT-I/O
A1 PWRDN C11 GND M11 VCC P1 CCLK
C3 I/O-TCLKIN A14 M0-RT N13 DONE-PG M4 VCC
B2 I/O D12 VCC M12 D7-I/O L3 GND
B3 I/O C13 M2-I/O P13 XTL1-I/O-BCLKIN M2 A0-WS-I/O
A2 I/O* B14 HDC-I/O N12 I/O N1 A1-CS2-I/O
B4 I/O C14 I/O P12 I/O M1 I/O
C5 I/O E12 I/O N11 D6-I/O K3 I/O
A3 I/O* D13 I/O M10 I/O L2 A2-I/O
A4 I/O D14 LDC-I/O P11 I/O* L1 A3-I/O
B5 I/O E13 I/O* N10 I/O K2 I/O
C6 I/O F12 I/O P10 I/O J3 I/O
A5 I/O E14 I/O M9 D5-I/O K1 A15-I/O
B6 I/O F13 I/O N9 CS0-I/O J2 A4-I/O
A6 I/O F14 I/O P9 I/O* J1 I/O*
B7 I/O G13 I/O P8 I/O* H1 A14-I/O
C7 GND G14 INIT-I/O N8 D4-I/O H2 A5-I/O
C8 VCC G12 VCC P7 I/O H3 GND
A7 I/O H12 GND M8 VCC G3 VCC
B8 I/O H14 I/O M7 GND G2 A13-I/O
A8 I/O H13 I/O N7 D3-I/O G1 A6-I/O
A9 I/O J14 I/O P6 CS1-I/O F1 I/O*
B9 I/O J13 I/O N6 I/O* F2 A12-I/O
C9 I/O K14 I/O P5 I/O* E1 A7-I/O
A10 I/O J12 I/O M6 D2-I/O F3 I/O
B10 I/O K13 I/O N5 I/O E2 I/O
A11 I/O* L14 I/O* P4 I/O D1 A11-I/O
C10 I/O L13 I/O P3 I/O D2 A8-I/O
B11 I/O K12 I/O M5 D1-I/O E3 I/O
A12 I/O* M14 I/O N4 RDY/BUSY-RCLK-I/O C1 I/O
B12 I/O N14 I/O P2 I/O B1 A10-I/O
A13 I/O* M13 XTL2(IN)-I/O N3 I/O C2 A9-I/O
C12 I/O L12 GND N2 D0-DIN-I/O D3 VCC
* Indicates unconnected package pins (14) for the XC3042A.

R
XC3000 Series 144-Pin Plastic TQFP Pinouts

XC3042A XC3042A XC3042A
Pin Pin Pin
XC3064A XC3064A XC3064A
Number Number Number
XC3090A XC3090A XC3090A
1 PWRDN 49 I/O 97 I/O
2 I/O-TCLKIN 50 I/O* 98 I/O
3 I/O* 51 I/O 99 I/O*
4 I/O 52 I/O 100 I/O
5 I/O 53 INIT-I/O 101 I/O*
6 I/O* 54 VCC 102 D1-I/O
7 I/O 55 GND 103 RDY/BUSY-RCLK-I/O
8 I/O 56 I/O 104 I/O
9 I/O* 57 I/O 105 I/O
10 I/O 58 I/O 106 D0-DIN-I/O
11 I/O 59 I/O 107 DOUT-I/O
12 I/O 60 I/O 108 CCLK
13 I/O 61 I/O 109 VCC
14 I/O 62 I/O 110 GND
15 I/O* 63 I/O* 111 A0-WS-I/O
16 I/O 64 I/O* 112 A1-CS2-I/O
17 I/O 65 I/O 113 I/O
18 GND 66 I/O 114 I/O
19 VCC 67 I/O 115 A2-I/O
20 I/O 68 I/O 116 A3-I/O
21 I/O 69 XTL2(IN)-I/O 117 I/O
22 I/O 70 GND 118 I/O
23 I/O 71 RESET 119 A15-I/O 7
24 I/O 72 VCC 120 A4-I/O
25 I/O 73 DONE-PG 121 I/O*
26 I/O 74 D7-I/O 122 I/O*
27 I/O 75 XTL1(OUT)-BCLKIN-I/O 123 A14-I/O
28 I/O* 76 I/O 124 A5-I/O
29 I/O 77 I/O 125 I/O (XC3090 only)
30 I/O 78 D6-I/O 126 GND
31 I/O* 79 I/O 127 VCC
32 I/O* 80 I/O* 128 A13-I/O
33 I/O 81 I/O 129 A6-I/O
34 I/O* 82 I/O 130 I/O*
35 I/O 83 I/O* 131 I/O (XC3090 only)
36 M1-RD 84 D5-I/O 132 I/O*
37 GND 85 CS0-I/O 133 A12-I/O
38 M0-RT 86 I/O* 134 A7-I/O
39 VCC 87 I/O* 135 I/O
40 M2-I/O 88 D4-I/O 136 I/O
41 HDC-I/O 89 I/O 137 A11-I/O
42 I/O 90 VCC 138 A8-I/O
43 I/O 91 GND 139 I/O
44 I/O 92 D3-I/O 140 I/O
45 LDC-I/O 93 CS1-I/O 141 A10-I/O
46 I/O* 94 I/O* 142 A9-I/O
47 I/O 95 I/O* 143 VCC
48 I/O 96 D2-I/O 144 GND

R
XC3000 Series 160-Pin PQFP Pinouts

PQFP Pin XC3064A, XC3090A, PQFP Pin XC3064A, XC3090A, PQFP Pin XC3064A, XC3090A, PQFP Pin XC3064A, XC3090A,
Number XC3195A Number XC3195A Number XC3195A Number XC3195A
1 I/O* 41 GND 81 D7-I/O 121 CCLK
2 I/O* 42 M0–RTRIG 82 XTL1-I/O-BCLKIN 122 VCC
3 I/O* 43 VCC 83 I/O* 123 GND
4 I/O 44 M2-I/O 84 I/O 124 A0-WS-I/O
5 I/O 45 HDC-I/O 85 I/O 125 A1-CS2-I/O
6 I/O 46 I/O 86 D6-I/O 126 I/O
7 I/O 47 I/O 87 I/O 127 I/O
8 I/O 48 I/O 88 I/O 128 A2-I/O
9 I/O 49 LDC-I/O 89 I/O 129 A3-I/O
10 I/O 50 I/O* 90 I/O 130 I/O
11 I/O 51 I/O* 91 I/O 131 I/O
12 I/O 52 I/O 92 D5-I/O 132 A15-I/O
13 I/O 53 I/O 93 CS0-I/O 133 A4-I/O
14 I/O 54 I/O 94 I/O* 134 I/O
15 I/O 55 I/O 95 I/O* 135 I/O
16 I/O 56 I/O 96 I/O 136 A14-I/O
17 I/O 57 I/O 97 I/O 137 A5-I/O
18 I/O 58 I/O 98 D4-I/O 138 I/O*
19 GND 59 INIT-I/O 99 I/O 139 GND
20 VCC 60 VCC 100 VCC 140 VCC
21 I/O* 61 GND 101 GND 141 A13-I/O
22 I/O 62 I/O 102 D3-I/O 142 A6-I/O
23 I/O 63 I/O 103 CS1-I/O 143 I/O*
24 I/O 64 I/O 104 I/O 144 I/O*
25 I/O 65 I/O 105 I/O 145 I/O
26 I/O 66 I/O 106 I/O* 146 I/O
27 I/O 67 I/O 107 I/O* 147 A12-I/O
28 I/O 68 I/O 108 D2-I/O 148 A7-I/O
29 I/O 69 I/O 109 I/O 149 I/O
30 I/O 70 I/O 110 I/O 150 I/O
31 I/O 71 I/O 111 I/O 151 A11-I/O
32 I/O 72 I/O 112 I/O 152 A8-I/O
33 I/O 73 I/O 113 I/O 153 I/O
34 I/O 74 I/O 114 D1-I/O 154 I/O
35 I/O 75 I/O* 115 RDY/BUSY-RCLK-I/O 155 A10-I/O
36 I/O 76 XTL2-I/O 116 I/O 156 A9-I/O
37 I/O 77 GND 117 I/O 157 VCC
38 I/O* 78 RESET 118 I/O* 158 GND
39 I/O* 79 VCC 119 D0-DIN-I/O 159 PWRDWN
40 M1-RDATA 80 DONE/PG 120 DOUT-I/O 160 TCLKIN-I/O
Programmed IOBs are default slew-rate limited.

R
XC3000 Series 175-Pin Ceramic and Plastic PGA Pinouts

PGA Pin PGA Pin PGA Pin PGA Pin

XC3090A, XC3195A XC3090A, XC3195A XC3090A, XC3195A XC3090A, XC3195A
Number Number Number Number
B2 PWRDN D13 I/O R14 DONE-PG N4 DOUT-I/O
D4 TCLKIN-I/O B14 M1-RDATA N13 D7-I/O R2 CCLK
B3 I/O C14 GND T14 XTL1(OUT)-BCLKIN-I/O P3 VCC
C4 I/O B15 M0-RTRIG P13 I/O N3 GND
B4 I/O D14 VCC R13 I/O P2 A0-WS-I/O
A4 I/O C15 M2-I/O T13 I/O M3 A1-CS2-I/O
D5 I/O E14 HDC-I/O N12 I/O R1 I/O
C5 I/O B16 I/O P12 D6-I/O N2 I/O
B5 I/O D15 I/O R12 I/O P1 A2-I/O
A5 I/O C16 I/O T12 I/O N1 A3-I/O
C6 I/O D16 LDC-I/O P11 I/O L3 I/O
D6 I/O F14 I/O N11 I/O M2 I/O
B6 I/O E15 I/O R11 I/O M1 A15-I/O
A6 I/O E16 I/O T11 D5-I/O L2 A4-I/O
B7 I/O F15 I/O R10 CS0-I/O L1 I/O
C7 I/O F16 I/O P10 I/O K3 I/O
D7 I/O G14 I/O N10 I/O K2 A14-I/O
A7 I/O G15 I/O T10 I/O K1 A5-I/O
A8 I/O G16 I/O T9 I/O J1 I/O
B8 I/O H16 I/O R9 D4-I/O J2 I/O
C8 I/O H15 INIT-I/O P9 I/O J3 GND
D8 GND H14 VCC N9 VCC H3 VCC
D9 VCC J14 GND N8 GND H2 A13-I/O 7
C9 I/O J15 I/O P8 D3-I/O H1 A6-I/O
B9 I/O J16 I/O R8 CS1-I/O G1 I/O
A9 I/O K16 I/O T8 I/O G2 I/O
A10 I/O K15 I/O T7 I/O G3 I/O
D10 I/O K14 I/O N7 I/O F1 I/O
C10 I/O L16 I/O P7 I/O F2 A12-I/O
B10 I/O L15 I/O R7 D2-I/O E1 A7-I/O
A11 I/O M16 I/O T6 I/O E2 I/O
B11 I/O M15 I/O R6 I/O F3 I/O
D11 I/O L14 I/O N6 I/O D1 A11-I/O
C11 I/O N16 I/O P6 I/O C1 A8-I/O
A12 I/O P16 I/O T5 I/O D2 I/O
B12 I/O N15 I/O R5 D1-I/O B1 I/O
C12 I/O R16 I/O P5 RDY/BUSY-RCLK-I/O E3 A10-I/O
D12 I/O M14 I/O N5 I/O C2 A9-I/O
A13 I/O P15 XTL2(IN)-I/O T4 I/O D3 VCC
B13 I/O N14 GND R4 I/O C3 GND
C13 I/O R15 RESET P4 I/O
A14 I/O P14 VCC R3 D0-DIN-I/O
Pins A2, A3, A15, A16, T1, T2, T3, T15 and T16 are not connected. Pin A1 does not exist.

R
XC3000 Series 176-Pin TQFP Pinouts

Pin Pin Pin Pin

XC3090A XC3090A XC3090A XC3090A
Number Number Number Number
1 PWRDWN 45 M1-RDATA 89 DONE-PG 133 VCC
2 TCLKIN-I/O 46 GND 90 D7-I/O 134 GND
3 I/O 47 M0-RTRIG 91 XTAL1(OUT)-BCLKIN-I/O 135 A0-WS-I/O
4 I/O 48 VCC 92 I/O 136 A1-CS2-I/O
5 I/O 49 M2-I/O 93 I/O 137 –
6 I/O 50 HDC-I/O 94 I/O 138 I/O
7 I/O 51 I/O 95 I/O 139 I/O
8 I/O 52 I/O 96 D6-I/O 140 A2-I/O
9 I/O 53 I/O 97 I/O 141 A3-I/O
10 I/O 54 LDC-I/O 98 I/O 142 –
11 I/O 55 – 99 I/O 143 –
12 I/O 56 I/O 100 I/O 144 I/O
13 I/O 57 I/O 101 I/O 145 I/O
14 I/O 58 I/O 102 D5-I/O 146 A15-I/O
15 I/O 59 I/O 103 CS0-I/O 147 A4-I/O
16 I/O 60 I/O 104 I/O 148 I/O
17 I/O 61 I/O 105 I/O 149 I/O
18 I/O 62 I/O 106 I/O 150 A14-I/O
19 I/O 63 I/O 107 I/O 151 A5-I/O
20 I/O 64 I/O 108 D4-I/O 152 I/O
21 I/O 65 INIT-I/O 109 I/O 153 I/O
22 GND 66 VCC 110 VCC 154 GND
23 VCC 67 GND 111 GND 155 VCC
24 I/O 68 I/O 112 D3-I/O 156 A13-I/O
25 I/O 69 I/O 113 CS1-I/O 157 A6-I/O
26 I/O 70 I/O 114 I/O 158 I/O
27 I/O 71 I/O 115 I/O 159 I/O
28 I/O 72 I/O 116 I/O 160 –
29 I/O 73 I/O 117 I/O 161 –
30 I/O 74 I/O 118 D2-I/O 162 I/O
31 I/O 75 I/O 119 I/O 163 I/O
32 I/O 76 I/O 120 I/O 164 A12-I/O
33 I/O 77 I/O 121 I/O 165 A7-I/O
34 I/O 78 I/O 122 I/O 166 I/O
35 I/O 79 I/O 123 I/O 167 I/O
36 I/O 80 I/O 124 D1-I/O 168 –
37 I/O 81 I/O 125 RDY/BUSY-RCLK-I/O 169 A11-I/O
38 I/O 82 – 126 I/O 170 A8-I/O
39 I/O 83 – 127 I/O 171 I/O
40 I/O 84 I/O 128 I/O 172 I/O
41 I/O 85 XTAL2(IN)-I/O 129 I/O 173 A10-I/O
42 I/O 86 GND 130 D0-DIN-I/O 174 A9-I/O
43 I/O 87 RESET 131 DOUT-I/O 175 VCC
44 – 88 VCC 132 CCLK 176 GND

R
XC3000 Series 208-Pin PQFP Pinouts

XC3000A, and XC3000L families have identical pinouts
Pin Number XC3090A Pin Number XC3090A Pin Number XC3090A Pin Number XC3090A
1 – 53 – 105 – 157 –
2 GND 54 – 106 VCC 158 –
3 PWRDWN 55 VCC 107 D/P 159 –
4 TCLKIN-I/O 56 M2-I/O 108 – 160 GND
5 I/O 57 HDC-I/O 109 D7-I/O 161 WS-A0-I/O
6 I/O 58 I/O 110 XTL1-BCLKIN-I/O 162 CS2-A1-I/O
7 I/O 59 I/O 111 I/O 163 I/O
8 I/O 60 I/O 112 I/O 164 I/O
9 I/O 61 LDC-I/O 113 I/O 165 A2-I/O
10 I/O 62 I/O 114 I/O 166 A3-I/O
11 I/O 63 I/O 115 D6-I/O 167 I/O
12 I/O 64 – 116 I/O 168 I/O
13 I/O 65 – 117 I/O 169 –
14 I/O 66 – 118 I/O 170 –
15 – 67 – 119 – 171 –
16 I/O 68 I/O 120 I/O 172 A15-I/O
17 I/O 69 I/O 121 I/O 173 A4-I/O
18 I/O 70 I/O 122 D5-I/O 174 I/O
19 I/O 71 I/O 123 CS0-I/O 175 I/O
20 I/O 72 – 124 I/O 176 –
21 I/O 73 – 125 I/O 177 –
22 I/O 74 I/O 126 I/O 178 A14-I/O
23 I/O 75 I/O 127 I/O 179 A5-I/O
24 I/O 76 I/O 128 D4-I/O 180 I/O
25 GND 77 INIT-I/O 129 I/O 181 I/O 7
26 VCC 78 VCC 130 VCC 182 GND
27 I/O 79 GND 131 GND 183 VCC
28 I/O 80 I/O 132 D3-I/O 184 A13-I/O
29 I/O 81 I/O 133 CS1-I/O 185 A6-I/O
30 I/O 82 I/O 134 I/O 186 I/O
31 I/O 83 – 135 I/O 187 I/O
32 I/O 84 – 136 I/O 188 –
33 I/O 85 I/O 137 I/O 189 –
34 I/O 86 I/O 138 D2-I/O 190 I/O
35 I/O 87 I/O 139 I/O 191 I/O
36 I/O 88 I/O 140 I/O 192 A12-I/O
37 – 89 I/O 141 I/O 193 A7-I/O
38 I/O 90 – 142 – 194 –
39 I/O 91 – 143 I/O 195 –
40 I/O 92 – 144 I/O 196 –
41 I/O 93 I/O 145 D1-I/O 197 I/O
42 I/O 94 I/O 146 RDY/BUSY-RCLK-I/O 198 I/O
43 I/O 95 I/O 147 I/O 199 A11-I/O
44 I/O 96 I/O 148 I/O 200 A8-I/O
45 I/O 97 I/O 149 I/O 201 I/O
46 I/O 98 I/O 150 I/O 202 I/O
47 I/O 99 I/O 151 DIN-D0-I/O 203 A10-I/O
48 M1-RDATA 100 XTL2-I/O 152 DOUT-I/O 204 A9-I/O
49 GND 101 GND 153 CCLK 205 VCC
50 M0-RTRIG 102 RESET 154 VCC 206 –
51 – 103 – 155 – 207 –
52 – 104 – 156 – 208 –
* In PQ208, XC3090A and XC3195A have different pinouts.

R
XC3195A PQ208 Pinouts
Pin Description PQ208 Pin Description PQ208 Pin Description PQ208 Pin Description PQ208
A9-I/O 206 D0-DIN-I/O 154 I/O 102 I/O 48
A10-I/O 205 I/O 153 I/O 101 I/O 47
I/O 204 I/O 152 I/O 100 I/O 46
I/O 203 I/O 151 I/O 99 I/O 45
I/O 202 I/O 150 I/O 98 I/O 44
I/O 201 RDY/BUSY-RCLK-I/O 149 I/O 97 I/O 43
A8-I/O 200 D1-I/O 148 I/O 96 I/O 42
A11-I/O 199 I/O 147 I/O 95 I/O 41
I/O 198 I/O 146 I/O 94 I/O 40
I/O 197 I/O 145 I/O 93 I/O 39
I/O 196 I/O 144 I/O 92 I/O 38
I/O 194 I/O 141 I/O 89 I/O 37
A7-I/O 193 I/O 140 I/O 88 I/O 36
A12-I/O 192 I/O 139 I/O 87 I/O 35
I/O 191 D2-I/O 138 I/O 86 I/O 34
I/O 190 I/O 137 I/O 85 I/O 33
I/O 189 I/O 136 I/O 84 I/O 32
I/O 188 I/O 135 I/O 83 I/O 31
I/O 187 I/O 134 I/O 82 I/O 30
I/O 186 CS1-I/O 133 I/O 81 I/O 29
A6-I/O 185 D3-I/O 132 I/O 80 I/O 28
A13-I/O 184 GND 131 GND 79 VCC 27
VCC 183 VCC 130 VCC 78 GND 26
GND 182 I/O 129 INIT 77 I/O 25
I/O 181 D4-I/O 128 I/O 76 I/O 24
I/O 180 I/O 127 I/O 75 I/O 23
A5-I/O 179 I/O 126 I/O 74 I/O 22
A14-I/O 178 I/O 125 I/O 73 I/O 21
I/O 177 I/O 124 I/O 72 I/O 20
I/O 176 CS0-I/O 123 I/O 71 I/O 19
I/O 175 D5-I/O 122 I/O 70 I/O 18
I/O 174 I/O 121 I/O 69 I/O 17
A4-I/O 173 I/O 120 I/O 68 I/O 14
A15-I/O 172 I/O 119 I/O 67 I/O 13
I/O 171 I/O 118 I/O 66 I/O 12
I/O 169 I/O 117 I/O 63 I/O 11
I/O 168 I/O 116 I/O 62 I/O 10
I/O 167 I/O 115 I/O 61 I/O 9
A3-I/O 166 D6-I/O 114 I/O 60 I/O 8
A2-I/O 165 I/O 113 LDC-I/O 59 I/O 7
I/O 164 I/O 112 I/O 58 I/O 6
I/O 163 I/O 111 I/O 57 I/O 5
I/O 162 I/O 110 I/O 56 I/O 4
I/O 161 XTLX1(OUT)BCLKN-I/O 109 HDC-I/O 55 I/O 3
A1-CS2-I/O 160 D7-I/O 108 M2-I/O 54 TCLKIN-I/O 2
A0-WS-I/O 159 D/P 107 VCC 53 PWRDN 1
GND 158 VCC 106 M0-RTIG 52 GND 208
VCC 157 RESET 105 GND 51 VCC 207
CCLK 156 GND 104 M1/RDATA 50

DOUT-I/O 155 XTL2(IN)-I/O 103 I/O 49
Unprogrammed IOBs have a default pull-up. This prevents an undefined pad level for unbonded or unused IOBs. Programmed outputs are
default slew-rate limited.
In the PQ208 package, pins 15, 16, 64, 65, 90, 91, 142, 143, 170 and 195 are not connected.
* In PQ208, XC3090A and XC3195A have different pinouts.

R
Pins 44 64 68 84 100 132 144 160 175 176 208
Plast. Plast. Plast. Plast. Cer. Plast. Plast. Plast. Plast. Cer. Plast. Plast. Plast. Cer. Plast. Plast.
Type PLCC VQFP PLCC PLCC PGA PQFP TQFP VQFP PGA PGA TQFP PQFP PGA PGA TQFP PQFP
Code PC44 VQ64 PC68 PC84 PG84 PQ100 TQ100 VQ100 PP132 PG132 TQ144 PQ160 PP175 PG175 TQ176 PQ208
-7 CI CI CI
XC3020A
-6 C C C
XC3030A
-6 C C C C C C
XC3042A
-6 C C C C C C
-7 CI CI CI CI CI
XC3064A
-6 C C C C C
XC3090A
-6 C C C C C C C
XC3020L -8 CI
XC3030L -8 CI CI CI
XC3042L -8 CI CI CI
XC3064L -8 CI CI
XC3090L -8 CI CI CI
-4 CI CI CI
-3 CI CI CI
XC3120A
-2 CI CI CI
-1 C C C
-09 C C C
XC3130A
-1 C C C C C C
7
-09 C C C C C C
-4 CI CI C CI
-3 CI CI CI CI
XC3142A -2 CI CI CI CI
-1 C C C C
-09 C C C C
-4 CI CI CI
-3 CI CI CI
XC3164A
-2 CI CI CI
-1 C C C
-09 C C C
XC3190A
-1 C C C C C C C
-09 C C C C C C C
-4 CI CI CI CI CI
-3 CI CI CI CI CI
XC3195A -2 CI CI CI CI CI
-1 C C C C C
-09 C C C C C

R
Pins 44 64 68 84 100 132 144 160 175 176 208
Plast. Plast. Plast. Plast. Cer. Plast. Plast. Plast. Plast. Cer. Plast. Plast. Plast. Cer. Plast. Plast.
Type PLCC VQFP PLCC PLCC PGA PQFP TQFP VQFP PGA PGA TQFP PQFP PGA PGA TQFP PQFP
Code PC44 VQ64 PC68 PC84 PG84 PQ100 TQ100 VQ100 PP132 PG132 TQ144 PQ160 PP175 PG175 TQ176 PQ208
C C C
XC3142L
C C C
C C C
XC3190L
C C C
Notes: C = Commercial, TJ= 0° to +85°C I = Industrial, TJ = -40° to +100°C
Number of Available I/O Pins

Number of Package Pins
Max I/O 44 64 68 84 100 132 144 160 175 176 208
XC3020A/XC3120A 64 58 64 64
XC3030A/XC3130A 80 34 54 58 74 80
XC3042A/3142A 96 74 82 96 96
XC2064A/XC3164A 120 70 110 120 120
XC3090A/XC3190A 144 70 122 138 144 144 144
XC3195A 176 70 138 144 176
Example: XC3030A-3 PC44C

Speed Grade Number of Pins
Package Type
Revision History
Date Revision
11/98 Revised version number to 3.1, removed XC3100A-5 obsolete packages.

0
R
0 7*

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-83
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-83
XC5200 Family Compared to XC4000/Spartan™ and XC3000 Series . . . . . . . . . . . . . . . . . . . . . . . . . 7-84
Configurable Logic Block (CLB) Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-84
Input/Output Block (IOB) Resources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-84
Routing Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-84
Configuration and Readback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-84
Architectural Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-85
VersaBlock: Abundant Local Routing Plus Versatile Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-85
VersaRing I/O Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-86
General Routing Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-86
Performance Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-87
Taking Advantage of Reconfiguration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-87
5-Input Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-87
Carry Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-88
Cascade Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-89
CLB Flip-Flops and Latches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-89 7
Data Inputs and Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-89
Clock Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-89
Clock Enable. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-89
Clear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-89
Global Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-90
Using FPGA Flip-Flops and Latches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-90
Three-State Buffers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-90
Input/Output Blocks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-91
IOB Input Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-91
Optional Delay Guarantees Zero Hold Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-91
IOB Output Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-92
Other IOB Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-92
Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-93
VersaBlock Routing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-93
Local Interconnect Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-93
Direct Connects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-94
General Routing Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-94
Single- and Double-Length Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-96
Longlines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-96
Global Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-96
VersaRing Input/Output Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-98
Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-98
Data Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-100
Instruction Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-100
Bit Sequence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-100
Including Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-100
Avoiding Inadvertent Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-101
7-79
R
Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-104
Special Purpose Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-104
Master Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-104
Peripheral Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-105
Slave Serial Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-105
Express Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-106
Cyclic Redundancy Check (CRC) for Configuration and Readback . . . . . . . . . . . . . . . . . . . . . . 7-107
Power-On Time-Out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-108
Initialization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-108
Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-110
Start-Up. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-110
DONE Goes High to Signal End of Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-111
Release of User I/O After DONE Goes High . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-111
Release of Global Reset After DONE Goes High . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-112
Configuration Complete After DONE Goes High . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-112
Readback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-112
Readback Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-113
Read Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-113
Read Abort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-113
Clock Select . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-113
Violating the Maximum High and Low Time Specification for the Readback Clock . . . . . . . . . . 7-113
Readback with the XChecker Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-113
Synchronous Peripheral Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-118
Asynchronous Peripheral Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-120
Write to FPGA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-120
Status Read . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-120
Express Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-122
Configuration Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-125
Master Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-125
Slave and Peripheral Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-125
XC5200 Program Readback Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . 7-126
XC5200 Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-127
XC5200 Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-127
XC5200 DC Characteristics Over Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-127
XC5200 Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-127
XC5200 Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-128
XC5200 Longline Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-128
XC5200 CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-129
XC5200 Guaranteed Input and Output Parameters (Pin-to-Pin) . . . . . . . . . . . . . . . . . . . . . . . . 7-130
XC5200 IOB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-131
XC5200 Boundary Scan (JTAG) Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . 7-132
Pin Locations for XC5202 Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-133
Additional No Connect (N.C.) Connections on TQ144 Package. . . . . . . . . . . . . . . . . . . . . . . . . 7-135
Additional No Connect (N.C.) Connections for PQ160 Package . . . . . . . . . . . . . . . . . . . . . . . . 7-138
7-80
R

Additional No Connect (N.C.) Connections for PQ208 and TQ176 Packages . . . . . . . . . . . . . . 7-142
Additional No Connect (N.C.) Connections for PQ208 and PQ240 Packages . . . . . . . . . . . . . . 7-147
Additional No Connect (N.C.) Connections for HQ208 and HQ240 Packages . . . . . . . . . . . . . . 7-153
Revisions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-155
7-81
R
7-82
0
R
XC5200 Series
Features - Footprint compatibility in common packages within

the XC5200 Series and with the XC4000 Series
• Low-cost, register/latch rich, SRAM based - Over 150 device/package combinations, including
reprogrammable architecture advanced BGA, TQ, and VQ packaging available
- 0.5µm three-layer metal CMOS process technology • Fully Supported by Xilinx Development System
- 256 to 1936 logic cells (3,000 to 23,000 “gates”) - Automatic place and route software
- Price competitive with Gate Arrays - Wide selection of PC and Workstation platforms
• System Level Features - Over 100 3rd-party Alliance interfaces
- System performance beyond 50 MHz - Supported by shrink-wrap Foundation software
- 6 levels of interconnect hierarchy
- VersaRing™ I/O Interface for pin-locking Description
- Dedicated carry logic for high-speed arithmetic
functions The XC5200 Field-Programmable Gate Array Family is
- Cascade chain for wide input functions engineered to deliver low cost. Building on experiences
- Built-in IEEE 1149.1 JTAG boundary scan test gained with three previous successful SRAM FPGA fami-
circuitry on all I/O pins lies, the XC5200 family brings a robust feature set to pro-
- Internal 3-state bussing capability grammable logic design. The VersaBlock™ logic module,
- Four dedicated low-skew clock or signal distribution the VersaRing I/O interface, and a rich hierarchy of inter-
nets connect resources combine to enhance design flexibility
• Versatile I/O and Packaging and reduce time-to-market. Complete support for the
XC5200 family is delivered through the familiar Xilinx soft-
7
- Innovative VersaRing™ I/O interface provides a high
logic cell to I/O ratio, with up to 244 I/O signals ware environment. The XC5200 family is fully supported on
- Programmable output slew-rate control maximizes popular workstation and PC platforms. Popular design
performance and reduces noise entry methods are fully supported, including ABEL, sche-
- Zero Flip-Flop hold time for input registers simplifies matic capture, VHDL, and Verilog HDL synthesis. Design-
system timing ers utilizing logic synthesis can use their existing tools to
- Independent Output Enables for external bussing design with the XC5200 devices.
.
Table 1: XC5200 Field-Programmable Gate Array Family Members
Device XC5202 XC5204 XC5206 XC5210 XC5215
Logic Cells 256 480 784 1,296 1,936
Max Logic Gates 3,000 6,000 10,000 16,000 23,000
Typical Gate Range 2,000 - 3,000 4,000 - 6,000 6,000 - 10,000 10,000 - 16,000 15,000 - 23,000
VersaBlock Array 8x8 10 x 12 14 x 14 18 x 18 22 x 22
CLBs 64 120 196 324 484
Flip-Flops 256 480 784 1,296 1,936
I/Os 84 124 148 196 244
TBUFs per Longline 10 14 16 20 24

R
XC5200 Family Compared to Table 2: Xilinx Field-Programmable Gate Array

Families
XC4000/Spartan™ and XC3000
Series Parameter XC5200 Spartan XC4000 XC3000
For readers already familiar with the XC4000/Spartan and CLB function
4 3 3 2
XC3000 FPGA Families, this section describes significant generators
differences between them and the XC5200 family. Unless CLB inputs 20 9 9 5
otherwise indicated, comparisons refer to both
XC4000/Spartan and XC3000 devices. CLB outputs 12 4 4 2
Configurable Logic Block (CLB) Resources Global buffers 4 8 8 2

Each XC5200 CLB contains four independent 4-input func- User RAM no yes yes no
tion generators and four registers, which are configured as
four independent Logic Cells™ (LCs). The registers in each Edge decoders no no yes no
XC5200 LC are optionally configurable as edge-triggered
D-type flip-flops or as transparent level-sensitive latches. Cascade chain yes no no no
The XC5200 CLB includes dedicated carry logic that pro- Fast carry logic yes yes yes no
vides fast arithmetic carry capability. The dedicated carry
logic may also be used to cascade function generators for Internal 3-state yes yes yes yes
implementing wide arithmetic functions.
Boundary scan yes yes yes no
XC4000 family: XC5200 devices have no wide edge
decoders. Wide decoders are implemented using cascade Slew-rate control yes yes yes yes
logic. Although sacrificing speed for some designs, lack of
wide edge decoders reduces the die area and hence cost Routing Resources
of the XC5200.
The XC5200 family provides a flexible coupling of logic and
XC4000/Spartan family: XC5200 dedicated carry logic local routing resources called the VersaBlock. The XC5200
differs from that of the XC4000/Spartan family in that the
VersaBlock element includes the CLB, a Local Interconnect
sum is generated in an additional function generator in the
Matrix (LIM), and direct connects to neighboring Versa-
adjacent column. This design reduces XC5200 die size and
Blocks.
hence cost for many applications. Note, however, that a
loadable up/down counter requires the same number of The XC5200 provides four global buffers for clocking or
function generators in both families. XC3000 has no dedi- high-fanout control signals. Each buffer may be sourced by
cated carry. means of its dedicated pad or from any internal source.
XC4000/Spartan family: XC5200 lookup tables are opti- Each XC5200 TBUF can drive up to two horizontal and two
mized for cost and hence cannot implement RAM. vertical Longlines. There are no internal pull-ups for
XC5200 Longlines.
Input/Output Block (IOB) Resources
Configuration and Readback
The XC5200 family maintains footprint compatibility with
the XC4000 family, but not with the XC3000 family. The XC5200 supports a new configuration mode called
Express mode.
To minimize cost and maximize the number of I/O per Logic
Cell, the XC5200 I/O does not include flip-flops or latches. XC4000/Spartan family: The XC5200 family provides a
global reset but not a global set.
For high performance paths, the XC5200 family provides
direct connections from each IOB to the registers in the XC5200 devices use a different configuration process than
adjacent CLB in order to emulate IOB registers. that of the XC3000 family, but use the same process as the
XC4000 and Spartan families.
Each XC5200 I/O Pin provides a programmable delay ele-
ment to control input set-up time. This element can be used XC3000 family: Although their configuration processes dif-
to avoid potential hold-time problems. Each XC5200 I/O fer, XC5200 devices may be used in daisy chains with
Pin is capable of 8-mA source and sink currents. XC3000 devices.
IEEE 1149.1-type boundary scan is supported in each XC3000 family: The XC5200 PROGRAM pin is a sin-
XC5200 I/O. gle-function input pin that overrides all other inputs. The
PROGRAM pin does not exist in XC3000.

R
XC3000 family: XC5200 devices support an additional pro-

Input/Output Blocks (IOBs)
gramming mode: Peripheral Synchronous.
XC3000 family: The XC5200 family does not support
Power-down, but offers a Global 3-state input that does not
reset any flip-flops. VersaRing
XC3000 family: The XC5200 family does not provide an GRM GRM GRM
on-chip crystal oscillator amplifier, but it does provide an Versa- Versa- Versa-
internal oscillator from which a variety of frequencies up to Block Block Block
12 MHz are available.
VersaRing
VersaRing
GRM GRM GRM
Architectural Overview Versa- Versa- Versa-
Block Block Block
Figure 1 presents a simplified, conceptual overview of the
XC5200 architecture. Similar to conventional FPGAs, the
XC5200 family consists of programmable IOBs, program- GRM GRM GRM
mable logic blocks, and programmable interconnect. Unlike Versa- Versa- Versa-
Block Block Block
other FPGAs, however, the logic and local routing
resources of the XC5200 family are combined in flexible VersaRing
VersaBlocks (Figure 2). General-purpose routing connects
to the VersaBlock through the General Routing Matrix X4955
(GRM).
Figure 1: XC5200 Architectural Overview
VersaBlock: Abundant Local Routing Plus
Versatile Logic
GRM
4 4
The basic logic element in each VersaBlock structure is the
24
Logic Cell, shown in Figure 3. Each LC contains a 4-input 24
7
function generator (F), a storage device (FD), and control TS
CLB
logic. There are five independent inputs and three outputs LC3
to each LC. The independence of the inputs and outputs LC2
4
4 4
allows the software to maximize the resource utilization 4 4
LC1
within each LC. Each Logic Cell also contains a direct
LC0
feedthrough path that does not sacrifice the use of either
the function generator or the register; this feature is a first LIM
for FPGAs. The storage device is configurable as either a D
4 4
flip-flop or a latch. The control logic consists of carry logic
for fast implementation of arithmetic functions, which can Direct Connects X5707
also be configured as a cascade chain allowing decode of

very wide input functions. Figure 2: VersaBlock
CO
DO
DI
D Q
F4
FD
F3
F2 F
F1
X
CI CE CK CLR
X4956
Figure 3: XC5200 Logic Cell (Four LCs per CLB)

R
The XC5200 CLB consists of four LCs, as shown in The LIM provides 100% connectivity of the inputs and out-
Figure 4. Each CLB has 20 independent inputs and 12 puts of each LC in a given CLB. The benefit of the LIM is
independent outputs. The top and bottom pairs of LCs can that no general routing resources are required to connect
be configured to implement 5-input functions. The chal- feedback paths within a CLB. The LIM connects to the
lenge of FPGA implementation software has always been GRM via 24 bidirectional nodes.
to maximize the usage of logic resources. The XC5200 The direct connects allow immediate connections to neigh-
family addresses this issue by surrounding each CLB with boring CLBs, once again without using any of the general
two types of local interconnect — the Local Interconnect
interconnect. These two layers of local routing resource
Matrix (LIM) and direct connects. These two interconnect
improve the granularity of the architecture, effectively mak-
resources, combined with the CLB, form the VersaBlock, ing the XC5200 family a “sea of logic cells.” Each
represented in Figure 2. Versa-Block has four 3-state buffers that share a common
enable line and directly drive horizontal and vertical Lon-
CO glines, creating robust on-chip bussing capability. The
LC3
DO VersaBlock allows fast, local implementation of logic func-
DI tions, effectively implementing user designs in a hierarchi-
D Q cal fashion. These resources also minimize local routing
F4 congestion and improve the efficiency of the general inter-
FD
F3 connect, which is used for connecting larger groups of
F2 F logic. It is this combination of both fine-grain and
F1
coarse-grain architecture attributes that maximize logic uti-
X lization in the XC5200 family. This symmetrical structure
takes full advantage of the third metal layer, freeing the
placement software to pack user logic optimally with mini-
LC2 mal routing restrictions.
DO
DI
D Q
VersaRing I/O Interface
F4 The interface between the IOBs and core logic has been
FD
F3 redesigned in the XC5200 family. The IOBs are completely
F2 F decoupled from the core logic. The XC5200 IOBs contain
F1 dedicated boundary-scan logic for added board-level test-
X ability, but do not include input or output registers. This
approach allows a maximum number of IOBs to be placed
around the device, improving the I/O-to-gate ratio and
LC1
DO decreasing the cost per I/O. A “freeway” of interconnect
DI cells surrounding the device forms the VersaRing, which
D Q provides connections from the IOBs to the internal logic.
F4 These incremental routing resources provide abundant
FD
F3 connections from each IOB to the nearest VersaBlock, in
F2 F addition to Longline connections surrounding the device.
The VersaRing eliminates the historic trade-off between
F1
X high logic utilization and pin placement flexibility. These
incremental edge resources give users increased flexibility
in preassigning (i.e., locking) I/O pins before completing
LC0 their logic designs. This ability accelerates time-to-market,
DO
DI since PCBs and other system components can be manu-
D Q
factured concurrent with the logic design.
F4
FD General Routing Matrix
F3
F2 F The GRM is functionally similar to the switch matrices
F1
found in other architectures, but it is novel in its tight cou-
X pling to the logic resources contained in the VersaBlocks.
CI CE CK CLR Advanced simulation tools were used during the develop-
ment of the XC5200 architecture to determine the optimal
X4957
level of routing resources required. The XC5200 family
Figure 4: Configurable Logic Block contains six levels of interconnect hierarchy — a series of

R
single-length lines, double-length lines, and Longlines all Detailed Functional Description
routed through the GRM. The direct connects, LIM, and
logic-cell feedthrough are contained within each Configurable Logic Blocks (CLBs)
Versa-Block. Throughout the XC5200 interconnect, an effi-
Figure 4 shows the logic in the XC5200 CLB, which con-
cient multiplexing scheme, in combination with three layer
sists of four Logic Cells (LC[3:0]). Each Logic Cell consists
metal (TLM), was used to improve the overall efficiency of
of an independent 4-input Lookup Table (LUT), and a
silicon usage.
D-Type flip-flop or latch with common clock, clock enable,
Performance Overview and clear, but individually selectable clock polarity. Addi-
tional logic features provided in the CLB are:
The XC5200 family has been benchmarked with many
designs running synchronous clock rates beyond 66 MHz. • An independent 5-input LUT by combining two 4-input
The performance of any design depends on the circuit to be LUTs.
implemented, and the delay through the combinatorial and • High-speed carry propagate logic.
sequential logic elements, plus the delay in the intercon- • High-speed pattern decoding.
nect routing. A rough estimate of timing can be made by • High-speed direct connection to flip-flop D-inputs.
assuming 3-6 ns per logic level, which includes direct-con- • Individual selection of either a transparent,
nect routing delays, depending on speed grade. More level-sensitive latch or a D flip-flop.
accurate estimations can be made using the information in • Four 3-state buffers with a shared Output Enable.
the Switching Characteristic Guideline section.
5-Input Functions
Taking Advantage of Reconfiguration Figure 5 illustrates how the outputs from the LUTs from
FPGA devices can be reconfigured to change logic function LC0 and LC1 can be combined with a 2:1 multiplexer
while resident in the system. This capability gives the sys- (F5_MUX) to provide a 5-input function. The outputs from
tem designer a new degree of freedom not available with the LUTs of LC2 and LC3 can be similarly combined.
any other type of logic.
Hardware can be changed as easily as software. Design 7
CO
updates or modifications are easy, and can be made to
DO
products already in the field. An FPGA can even be recon- DI
D Q
figured dynamically to perform different functions at differ- FD
ent times. I1
F4
F3
I2
I3 F2 F
Reconfigurable logic can be used to implement system I4 F1 X
self-diagnostics, create systems capable of being reconfig- LC1

F5_MUX
ured for different environments or operations, or implement DO
DI out
I5
multi-purpose hardware for a given application. As an
D Q
added benefit, using reconfigurable FPGA devices simpli- Qout
FD
fies hardware design and debugging and shortens product F4
F3
time-to-market. F2 F X
F1
CI CE CK CLR LC0
5-Input Function X5710
Figure 5: Two LUTs in Parallel Combined to Create a

5-input Function

R
carry out
CO carry3
A3 DO CO DO
DI DI
or
B3 D Q D Q
F4 CY_MUX FD FD
F4
F3 F3
A3 and B3
F2 XOR F2 XOR
to any two
F1 X half sum3 F1 X sum3
LC3 LC3
A2
DI DO carry2 DO
or DI
B2 D Q D Q
CY_MUX FD FD
F4 F4
F3 F3
A2 and B2
to any two F2 XOR F2 XOR
half sum2 sum2
F1 X F1 X
LC2 LC2
A1 DI DO carry1 DO
or DI
B1 D Q D Q
F4
FD FD
CY_MUX F4
F3 F3
A1 and B1
to any two F2 XOR F2 XOR
half sum1 sum1
F1 X F1 X
LC1 LC1
A0 carry0
DI DO DO
or DI
B0 D Q D Q
CY_MUX FD FD
F4 F4
F3 F3
A0 and B0
to any two F2 XOR F2
half sum0 XOR
F1 X sum0
F1 X
CI CE CK CLR LC0 CI CE CK CLR LC0
carry in
0
CY_MUX
F=0
Initialization of
carry chain (One Logic Cell) X5709
Figure 6: XC5200 CY_MUX Used for Adder Carry Propagate
Carry Function requires two LCs per bit. Notice that the carry chain
requires an initialization stage, which the XC5200 family
The XC5200 family supports a carry-logic feature that accomplishes using the carry initialize (CY_INIT) macro
enhances the performance of arithmetic functions such as and one additional LC. The carry chain can propagate ver-
counters, adders, etc. A carry multiplexer (CY_MUX) sym- tically up a column of CLBs.
bol is used to indicate the XC5200 carry logic. This symbol
represents the dedicated 2:1 multiplexer in each LC that The XC5200 library contains a set of Relationally-Placed
performs the one-bit high-speed carry propagate per logic Macros (RPMs) and arithmetic functions designed to take
cell (four bits per CLB). advantage of the dedicated carry logic. Using and modify-
ing these macros makes it much easier to implement cus-
While the carry propagate is performed inside the LC, an
adjacent LC must be used to complete the arithmetic func-
tion. Figure 6 represents an example of an adder function.
The carry propagate is performed on the CLB shown,
which also generates the half-sum for the four-bit adder. An
adjacent CLB is responsible for XORing the half-sum with
the corresponding carry-out. Thus an adder or counter

R
tomized RPMs, freeing the designer from the need to results or other incoming data in flip-flops, and connect
become an expert on architectures. their outputs to the interconnect network as well. The CLB
storage elements can also be configured as latches.
cascade out Table 3: CLB Storage Element Functionality
(active rising edge is shown)
CO
DI DO
out
Mode CK CE CLR D Q
D Q Power-Up or
X X X X 0
A15 F4
CY_MUX FD GR
A14 F3
X X 1 X 0
A13 F2 AND
A12 F1 X Flip-Flop __/ 1* 0* D D
LC3 0 X 0* X Q
DO
DI 1 1* 0* X Q
D Q Latch
0 1* 0* D D
CY_MUX FD
A11 F4 Both X 0 0* X Q
A10 F3
A9 F2 AND Legend:
A8 F1 X X Don’t care
LC2 __/ Rising edge
DI DO 1* Input is High or unconnected (default value)
D Q
FD
A7 F4 CY_MUX
A6 F3
Data Inputs and Outputs
A5 F2 AND
A4 F1 X
The source of a storage element data input is programma-
LC1
ble. It is driven by the function F, or by the Direct In (DI)
DO
block input. The flip-flops or latches drive the Q CLB out-
DI
Q puts. 7
D
CY_MUX FD Four fast feed-through paths from DI to DO are available,

A3 F4
A2 F3 as shown in Figure 4. This bypass is sometimes used by
A1 F2 AND
A0 F1 X
the automated router to repower internal signals. In addi-
LC0
tion to the storage element (Q) and direct (DO) outputs,
CI CE CK CLR
cascade in there is a combinatorial output (X) that is always sourced
by the Lookup Table.
CY_MUX
F=0 Initialization of The four edge-triggered D-type flip-flops or level-sensitive
carry chain (One Logic Cell) X5708
latches have common clock (CK) and clock enable (CE)
inputs. Any of the clock inputs can also be permanently
Figure 7: XC5200 CY_MUX Used for Decoder Cascade enabled. Storage element functionality is described in
Logic Table 3.
Clock Input
Cascade Function
The flip-flops can be triggered on either the rising or falling
Each CY_MUX can be connected to the CY_MUX in the
clock edge. The clock pin is shared by all four storage ele-
adjacent LC to provide cascadable decode logic. Figure 7
ments with individual polarity control. Any inverter placed
illustrates how the 4-input function generators can be con-
on the clock input is automatically absorbed into the CLB.
figured to take advantage of these four cascaded
CY_MUXes. Note that AND and OR cascading are specific Clock Enable
cases of a general decode. In AND cascading all bits are
The clock enable signal (CE) is active High. The CE pin is
decoded equal to logic one, while in OR cascading all bits
shared by the four storage elements. If left unconnected
are decoded equal to logic zero. The flexibility of the LUT
for any, the clock enable for that storage element defaults
achieves this result. The XC5200 library contains gate
to the active state. CE is not invertible within the CLB.
macros designed to take advantage of this function.
Clear
CLB Flip-Flops and Latches
An asynchronous storage element input (CLR) can be used
The CLB can pass the combinatorial output(s) to the inter- to reset all four flip-flops or latches in the CLB. This input
connect network, but can also store the combinatorial

R
can also be independently disabled for any flip-flop. CLR is Three-State Buffers
active High. It is not invertible within the CLB.
The XC5200 family has four dedicated Three-State Buffers
(TBUFs, or BUFTs in the schematic library) per CLB (see
STARTUP
Figure 9). The four buffers are individually configurable
PAD GR Q2 through four configuration bits to operate as simple
GTS Q3 non-inverting buffers or in 3-state mode. When in 3-state
IBUF
Q1Q4 mode the CLB output enable (TS) control signal drives the
CLK DONEIN enable to all four buffers. Each TBUF can drive up to two
horizontal and/or two vertical Longlines. These 3-state buff-
X9009
ers can be used to implement multiplexed or bidirectional
Figure 8: Schematic Symbols for Global Reset buses on the horizontal or vertical longlines, saving logic
resources.
Global Reset
The 3-state buffer enable is an active-High 3-state (i.e. an
A separate Global Reset line clears each storage element active-Low enable), as shown in Table 4.
during power-up, reconfiguration, or when a dedicated
Table 4: Three-State Buffer Functionality
Reset net is driven active. This global net (GR) does not
compete with other routing resources; it uses a dedicated IN T OUT
distribution network. X 1 Z
GR can be driven from any user-programmable pin as a IN 0 IN
global reset input. To use this global net, place an input pad
and input buffer in the schematic or HDL code, driving the Another 3-state buffer with similar access is located near
GR pin of the STARTUP symbol. (See Figure 9.) A specific each I/O block along the right and left edges of the array.
pin location can be assigned to this input using a LOC The longlines driven by the 3-state buffers have a weak
attribute or property, just as with any other user-program- keeper at each end. This circuit prevents undefined float-
mable pad. An inverter can optionally be inserted after the ing levels. However, it is overridden by any driver. To
input buffer to invert the sense of the Global Reset signal. ensure the longline goes high when no buffers are on, add
Alternatively, GR can be driven from any internal node. an additional BUFT to drive the output High during all of the
previously undefined states.
Using FPGA Flip-Flops and Latches
Figure 10 shows how to use the 3-state buffers to imple-
The abundance of flip-flops in the XC5200 Series invites ment a multiplexer. The selection is accomplished by the
pipelined designs. This is a powerful way of increasing per- buffer 3-state signal.
formance by breaking the function into smaller subfunc-
tions and executing them in parallel, passing on the results
through pipeline flip-flops. This method should be seriously TS
considered wherever throughput is more important than
latency.
To include a CLB flip-flop, place the appropriate library
symbol. For example, FDCE is a D-type flip-flop with clock
enable and asynchronous clear. The corresponding latch CLB
CLB
symbol is called LDCE. LC3
In XC5200-Series devices, the flip-flops can be used as LC2

registers or shift registers without blocking the function LC1
generators from performing a different, perhaps unrelated
task. This ability increases the functional capacity of the LC0
devices.
The CLB setup time is specified between the function gen-
Horizontal
erator inputs and the clock input CK. Therefore, the speci- Longlines
fied CLB flip-flop setup time includes the delay through the
function generator.
X9030
Figure 9: XC5200 3-State Buffers

R
Z = DA • A + DB • B + DC • C + DN • N
~100 kΩ
DA DB DC DN
BUFT BUFT BUFT BUFT
A B C N
X6466
"Weak Keeper"
Input/Output Blocks
User-configurable input/output blocks (IOBs) provide the Table 5: Supported Sources for XC5200-Series Device
interface between external package pins and the internal Inputs
logic. Each IOB controls one package pin and can be con-
figured for input, output, or bidirectional signals. XC5200 Input Mode
5 V, 5 V,
The I/O block, shown in Figure 11, consists of an input Source
TTL CMOS
buffer and an output buffer. The output driver is an 8-mA
Any device, Vcc = 3.3 V,
full-rail CMOS buffer with 3-state control. Two slew-rate √
CMOS outputs Unreliable
control modes are supported to minimize bus transients.
Any device, Vcc = 5 V, Data
Both the output buffer and the 3-state control are invertible. √
The input buffer has globally selected CMOS or TTL input TTL outputs
thresholds. The input buffer is invertible and also provides a Any device, Vcc = 5 V,
√ √
programmable delay line to assure reliable chip-to-chip CMOS outputs
set-up and hold times. Minimum ESD protection is 3 KV
Optional Delay Guarantees Zero Hold Time
using the Human Body Model.
XC5200 devices do not have storage elements in the IOBs.
However, XC5200 IOBs can be efficiently routed to CLB 7
Vcc
flip-flops or latches to store the I/O signals.
Input Delay
Buffer The data input to the register can optionally be delayed by
Pullup
I several nanoseconds. With the delay enabled, the setup
time of the input flip-flop is increased so that normal clock
PAD Output
Buffer routing does not result in a positive hold-time requirement.
Pulldown O A positive hold time requirement can lead to unreliable,
temperature- or processing-dependent operation.
T
The input flip-flop setup time is defined between the data
Slew Rate
X9001
Control measured at the device I/O pin and the clock input at the
CLB (not at the clock pin). Any routing delay from the
device clock pin to the clock input of the CLB must, there-
Figure 11: XC5200 I/O Block fore, be subtracted from this setup time to arrive at the real
setup time requirement relative to the device pins. A short
IOB Input Signals specified setup time might, therefore, result in a negative
The XC5200 inputs can be globally configured for either setup time at the device pins, i.e., a positive hold-time
TTL (1.2V) or CMOS thresholds, using an option in the bit- requirement.
stream generation software. There is a slight hysteresis of When a delay is inserted on the data line, more clock delay
about 300mV. can be tolerated without causing a positive hold-time
The inputs of XC5200-Series 5-Volt devices can be driven requirement. Sufficient delay eliminates the possibility of a
by the outputs of any 3.3-Volt device, if the 5-Volt inputs are data hold-time requirement at the external pin. The maxi-
in TTL mode. mum delay is therefore inserted as the software default.
Supported sources for XC5200-Series device inputs are The XC5200 IOB has a one-tap delay element: either the
shown in Table 5. delay is inserted (default), or it is not. The delay guarantees
a zero hold time with respect to clocks routed through any
of the XC5200 global clock buffers. (See “Global Lines” on
page 96 for a description of the global clock buffers in the
XC5200.) For a shorter input register setup time, with

R
non-zero hold, attach a NODELAY attribute or property to For XC5200 devices, maximum total capacitive load for
the flip-flop or input buffer. simultaneous fast mode switching in the same direction is
200 pF for all package pins between each Power/Ground
IOB Output Signals pin pair. For some XC5200 devices, additional internal
Output signals can be optionally inverted within the IOB, Power/Ground pin pairs are connected to special Power
and pass directly to the pad. As with the inputs, a CLB and Ground planes within the packages, to reduce ground
flip-flop or latch can be used to store the output signal. bounce.
An active-High 3-state signal can be used to place the out- For slew-rate limited outputs this total is two times larger for
put buffer in a high-impedance state, implementing 3-state each device type: 400 pF for XC5200 devices. This maxi-
outputs or bidirectional I/O. Under configuration control, mum capacitive load should not be exceeded, as it can
the output (OUT) and output 3-state (T) signals can be result in ground bounce of greater than 1.5 V amplitude and
inverted. The polarity of these signals is independently more than 5 ns duration. This level of ground bounce may
configured for each IOB. cause undesired transient behavior on an output, or in the
internal logic. This restriction is common to all high-speed
The XC5200 devices provide a guaranteed output sink cur-
digital ICs, and is not particular to Xilinx or the XC5200
rent of 8 mA.
Series.
Supported destinations for XC5200-Series device outputs
XC5200-Series devices have a feature called “Soft
are shown in Table 6.(For a detailed discussion of how to
Start-up,” designed to reduce ground bounce when all out-
interface between 5 V and 3.3 V devices, see the 3V Prod-
puts are turned on simultaneously at the end of configura-
ucts section of The Programmable Logic Data Book.)
tion. When the configuration process is finished and the
An output can be configured as open-drain (open-collector) device starts up, the first activation of the outputs is auto-
by placing an OBUFT symbol in a schematic or HDL code, matically slew-rate limited. Immediately following the initial
then tying the 3-state pin (T) to the output signal, and the activation of the I/O, the slew rate of the individual outputs
input pin (I) to Ground. (See Figure 12.) is determined by the individual configuration option for
Table 6: Supported Destinations for XC5200-Series each IOB.
Outputs
Global Three-State
XC5200 Output Mode A separate Global 3-State line (not shown in Figure 11)
5 V, forces all FPGA outputs to the high-impedance state,
Destination
CMOS unless boundary scan is enabled and is executing an
XC5200 device, VCC=3.3 V, EXTEST instruction. This global net (GTS) does not com-
√ pete with other routing resources; it uses a dedicated distri-
CMOS-threshold inputs
Any typical device, VCC = 3.3 V, bution network.
some1
CMOS-threshold inputs GTS can be driven from any user-programmable pin as a
Any device, VCC = 5 V, global 3-state input. To use this global net, place an input
√
TTL-threshold inputs pad and input buffer in the schematic or HDL code, driving
Any device, VCC = 5 V, the GTS pin of the STARTUP symbol. A specific pin loca-
√ tion can be assigned to this input using a LOC attribute or
CMOS-threshold inputs
property, just as with any other user-programmable pad. An
1. Only if destination device has 5-V tolerant inputs
inverter can optionally be inserted after the input buffer to
invert the sense of the Global 3-State signal. Using GTS is
similar to Global Reset. See Figure 8 on page 90 for
details. Alternatively, GTS can be driven from any internal
node.
OPAD
OBUFT Other IOB Options
X6702
There are a number of other programmable options in the
Figure 12: Open-Drain Output
XC5200-Series IOB.
Pull-up and Pull-down Resistors

Output Slew Rate
Programmable IOB pull-up and pull-down resistors are
The slew rate of each output buffer is, by default, reduced, useful for tying unused pins to Vcc or Ground to minimize
to minimize power bus transients when switching non-criti- power consumption and reduce noise sensitivity. The con-
cal signals. For critical signals, attach a FAST attribute or figurable pull-up resistor is a p-channel transistor that pulls
property to the output buffer or flip-flop.

R
to Vcc. The configurable pull-down resistor is an n-channel

transistor that pulls to Ground.
The value of these resistors is 20 kΩ − 100 kΩ. This high OSC1
value makes them unsuitable as wired-AND pull-up resis- OSCS
OSC2
tors.
The pull-up resistors for most user-programmable IOBs are
active during the configuration process. See Table 13 on
page 124 for a list of pins with pull-ups active before and OSC1
during configuration. CK_DIV
OSC2
After configuration, voltage levels of unused pads, bonded
or unbonded, must be valid logic levels, to reduce noise
5200_14
sensitivity and avoid excess current. Therefore, by default,
unused pads are configured with the internal pull-up resis- Figure 13: XC5200 Oscillator Macros
tor active. Alternatively, they can be individually configured
with the pull-down resistor, or as a driven output, or to be
driven by an external source. To activate the internal VersaBlock Routing
pull-up, attach the PULLUP library component to the net The General Routing Matrix (GRM) connects to the
attached to the pad. To activate the internal pull-down, Versa-Block via 24 bidirectional ports (M0-M23). Excluding
attach the PULLDOWN library component to the net direct connections, global nets, and 3-statable Longlines,
attached to the pad. all VersaBlock inputs and outputs connect to the GRM via
these 24 ports. Four 3-statable unidirectional signals
JTAG Support (TQ0-TQ3) drive out of the VersaBlock directly onto the
Embedded logic attached to the IOBs contains test struc- horizontal and vertical Longlines. Two horizontal global
tures compatible with IEEE Standard 1149.1 for boundary nets and two vertical global nets connect directly to every
scan testing, simplifying board-level testing. More informa- CLB clock pin; they can connect to other CLB inputs via the
tion is provided in “Boundary Scan” on page 98. GRM. Each CLB also has four unidirectional direct con- 7
nects to each of its four neighboring CLBs. These direct
Oscillator connects can also feed directly back to the CLB (see
XC5200 devices include an internal oscillator. This oscilla- Figure 14).
tor is used to clock the power-on time-out, clear configura- In addition, each CLB has 16 direct inputs, four direct con-
tion memory, and source CCLK in Master configuration nections from each of the neighboring CLBs. These direct
modes. The oscillator runs at a nominal 12 MHz frequency connections provide high-speed local routing that
that varies with process, Vcc, and temperature. The output bypasses the GRM.
CCLK frequency is selectable as 1 MHz (default), 6 MHz,
or 12 MHz. Local Interconnect Matrix
The XC5200 oscillator divides the internal 12-MHz clock or The Local Interconnect Matrix (LIM) is built from input and
a user clock. The user then has the choice of dividing by 4, output multiplexers. The 13 CLB outputs (12 LC outputs
16, 64, or 256 for the “OSC1” output and dividing by 2, 8, plus a Vcc/GND signal) connect to the eight VersaBlock
32, 128, 1024, 4096, 16384, or 65536 for the “OSC2” out- outputs via the output multiplexers, which consist of eight
put. The division is specified via a “DIVIDEn_BY=x” fully populated 13-to-1 multiplexers. Of the eight
attribute on the symbol, where n=1 for OSC1, or n=2 for VersaBlock outputs, four signals drive each neighboring
OSC2. These frequencies can vary by as much as -50% or CLB directly, and provide a direct feedback path to the input
+ 50%. multiplexers. The four remaining multiplexer outputs can
drive the GRM through four TBUFs (TQ0-TQ3). All eight
The OSC5 macro is used where an internal oscillator is
multiplexer outputs can connect to the GRM through the
required. The CK_DIV macro is applicable when a user
bidirectional M0-M23 signals. All eight signals also connect
clock input is specified (see Figure 13).
to the input multiplexers and are potential inputs to that
CLB.

R
To GRM
M0-M23
24 8
TS
Global Nets 4
COUT 4 To
Longlines
4 and GRM
North 4 TQ0-TQ3
CLB
South 4
5 3
East 4 LC3
West 4 Input Output
5 3
Multiplexers LC2 VCC /GND Multiplexers
4
Direct to
8 East
5 3
LC1 4
5 3
LC0
Direct North
CLK
4 CE
Feedback
4 CLR CIN
Direct West 4
Direct South X5724
Figure 14: VersaBlock Details

CLB inputs have several possible sources: the 24 signals The direct connects also provide a high-speed path from
from the GRM, 16 direct connections from neighboring the edge CLBs to the VersaRing input/output buffers, and
VersaBlocks, four signals from global, low-skew buffers, thus reduce pin-to-pin set-up time, clock-to-out, and combi-
and the four signals from the CLB output multiplexers. national propagation delay. Direct connects from the input
Unlike the output multiplexers, the input multiplexers are buffers to the CLB DI pin (direct flip-flop input) are only
not fully populated; i.e., only a subset of the available sig- available on the left and right edges of the device. CLB
nals can be connected to a given CLB input. The flexibility look-up table inputs and combinatorial/registered outputs
of LUT input swapping and LUT mapping compensates for have direct connects to input/output buffers on all four
this limitation. For example, if a 2-input NAND gate is sides.
required, it can be mapped into any of the four LUTs, and
The direct connects are ideal for developing customized
use any two of the four inputs to the LUT.
RPM cells. Using direct connects improves the macro per-
formance, and leaves the other routing channels intact for
Direct Connects
improved routing. Direct connects can also route through a
The unidirectional direct-connect segments are connected CLB using one of the four cell-feedthrough paths.
to the logic input/output pins through the CLB input and out-
put multiplexer arrays, and thus bypass the general routing General Routing Matrix
matrix altogether. These lines increase the routing channel
The General Routing Matrix, shown in Figure 15, provides
utilization, while simultaneously reducing the delay
flexible bidirectional connections to the Local Interconnect
incurred in speed-critical connections.
Matrix through a hierarchy of different-length metal seg-
ments in both the horizontal and vertical directions. A pro-

R
GRM GRM GRM
Versa- Versa- Versa-

Block Block Block
1
GRM GRM GRM

Block Block Block
2
GRM GRM GRM

Block Block Block 3
7
Six Levels of Routing Hierarchy GRM 4 4
1 Single-length Lines 24
24
TS
2 Double-length Lines CLB
LC3
4
3 Direct Connects
LC2
4 4
4 4
4 Longlines and Global Lines LC1
6 LC0
5 LIM Local Interconnect Matrix
LIM 5
Logic Cell Feedthrough
6 Path (Contained within each
Logic Cell) 4 4
Direct Connects X4963
Figure 15: XC5200 Interconnect Structure

grammable interconnect point (PIP) establishes an electri- provide ten routing channels to each of the four
cal connection between two wire segments. The PIP, con- neighboring CLBs in four directions.
sisting of a pass transistor switch controlled by a memory • Sixteen bidirectional double-length segments per CLB
element, provides bidirectional (in some cases, unidirec- provide four routing channels to each of four other
tional) connection between two adjoining wires. A collec- (non-neighboring) CLBs in four directions.
tion of PIPs inside the General Routing Matrix and in the • Eight horizontal and eight vertical bidirectional Longline
Local Interconnect Matrix provides connectivity between segments span the width and height of the chip,
various types of metal segments. A hierarchy of PIPs and respectively.
associated routing segments combine to provide a power-
Two low-skew horizontal and vertical unidirectional glo-
ful interconnect hierarchy: bal-line segments span each row and column of the chip,
• Forty bidirectional single-length segments per CLB respectively.

R
Single- and Double-Length Lines simple as adding an additional TBUF to drive the bus High
when the previously undefined states are activated.
The single- and double-length bidirectional line segments
make up the bulk of the routing channels. The dou- Global Lines
ble-length lines hop across every other CLB to reduce the
propagation delays in speed-critical nets. Regenerating the Global buffers in Xilinx FPGAs are special buffers that drive
signal strength is recommended after traversing three or a dedicated routing network called Global Lines, as shown
four such segments. Xilinx place-and-route software auto- in Figure 16. This network is intended for high-fanout
matically connects buffers in the path of the signal as nec- clocks or other control signals, to maximize speed and min-
essary. Single- and double-length lines cannot drive onto imize skewing while distributing the signal to many loads.
Longlines and global lines; Longlines and global lines can, The XC5200 family has a total of four global buffers (BUFG
however, drive onto single- and double-length lines. As a symbol in the library), each with its own dedicated routing
general rule, Longline and global-line connections to the channel. Two are distributed vertically and two horizontally
general routing matrix are unidirectional, with the signal throughout the FPGA.
direction from these lines toward the routing matrix.
The global lines provide direct input only to the CLB clock
Longlines pins. The global lines also connect to the General Routing
Matrix to provide access from these lines to the function
Longlines are used for high-fan-out signals, 3-state busses, generators and other control signals.
low-skew nets, and faraway destinations. Row and column
splitter PIPs in the middle of the array effectively double the Four clock input pads at the corners of the chip, as shown
total number of Longlines by electrically dividing them into in Figure 16, provide a high-speed, low-skew clock network
two separated half-lines. Longlines are driven by the to each of the four global-line buffers. In addition to the ded-
3-state buffers in each CLB, and are driven by similar buff- icated pad, the global lines can be sourced by internal
ers at the periphery of the array from the VersaRing I/O logic. PIPs from several routing channels within the Ver-
Interface. saRing can also be configured to drive the global-line buff-
ers.
Bus-oriented designs are easily implemented by using Lon-
glines in conjunction with the 3-state buffers in the CLB and Details of all the programmable interconnect for a CLB is
in the VersaRing. Additionally, weak keeper cells at the shown in Figure 17.
periphery retain the last valid logic level on the Longlines
when all buffers are in 3-state mode.
Longlines connect to the single-length or double-length GCK4
GCK1
lines, or to the logic inside the CLB, through the General
Routing Matrix. The only manner in which a Longline can
be driven is through the four 3-state buffers; therefore, a
Longline-to-Longline or single-line-to-Longline connection
through PIPs in the General Routing Matrix is not possible.
Again, as a general rule, long- and global-line connections
to the General Routing Matrix are unidirectional, with the
signal direction from these lines toward the routing matrix.
The XC5200 family has no pull-ups on the ends of the Lon-

glines sourced by TBUFs, unlike the XC4000 Series. Con-
sequently, wired functions (i.e., WAND and WORAND) and
wide multiplexing functions requiring pull-ups for undefined
states (i.e., bus applications) must be implemented in a dif- GCK3
GCK2
ferent way. In the case of the wired functions, the same
X5704
functionality can be achieved by taking advantage of the
carry/cascade logic described above, implementing a wide
Figure 16: Global Lines
logic function in place of the wired function. In the case of
3-state bus applications, the user must insure that all states .
of the multiplexing function are defined. This process is as

R
x9010
LONG
CLB
SINGLE
CARRY
DOUBLE
7
GLOBAL
DIRECT
DIRECT
DIRECT
LONG
GLOBAL
DOUBLE
SINGLE
Figure 17: Detail of Programmable Interconnect Associated with XC5200 Series CLB

R
VersaRing Input/Output Interface XC5200 devices support all the mandatory boundary-scan
instructions specified in the IEEE standard 1149.1. A Test
The VersaRing, shown in Figure 18, is positioned between Access Port (TAP) and registers are provided that imple-
the core logic and the pad ring; it has all the routing ment the EXTEST, SAMPLE/PRELOAD, and BYPASS
resources of a VersaBlock without the CLB logic. The Ver- instructions. The TAP can also support two USERCODE
saRing decouples the core logic from the I/O pads. Each
instructions. When the boundary scan configuration option
VersaRing Cell provides up to four pad-cell connections on
is selected, three normal user I/O pins become dedicated
one side, and connects directly to the CLB ports on the inputs for these functions. Another user output pin
other side. becomes the dedicated boundary scan output.
Boundary-scan operation is independent of individual IOB
VersaRing configuration and package type. All IOBs are treated as
independently controlled bidirectional pins, including any
2 unbonded IOBs. Retaining the bidirectional test capability
8
8 8 after configuration provides flexibility for interconnect test-
2 2 ing.
Also, internal signals can be captured during EXTEST by
2 Pad
connecting them to unbonded IOBs, or to the unused out-
Pad
puts in IOBs used as unidirectional input pins. This tech-
GRM 10
Interconnect nique partially compensates for the lack of INTEST
Pad support.
4
VersaBlock 4 The user can serially load commands and data into these
Pad
devices to control the driving of their outputs and to exam-
ine their inputs. This method is an improvement over
8
bed-of-nails testing. It avoids the need to over-drive device
8
outputs, and it reduces the user interface to four pins. An
2
optional fifth pin, a reset for the control logic, is described in
the standard but is not implemented in Xilinx devices.
2 Pad
The dedicated on-chip logic implementing the IEEE 1149.1
GRM 10 Pad functions includes a 16-state machine, an instruction regis-
Interconnect
ter and a number of data registers. The functional details
Pad
4 can be found in the IEEE 1149.1 specification and are also
VersaBlock 4 Pad discussed in the Xilinx application note XAPP 017: “Bound-
2 ary Scan in XC4000 and XC5200 Series devices”
Figure 19 on page 99 is a diagram of the XC5200-Series
8 8
2 boundary scan logic. It includes three bits of Data Register
X5705
per IOB, the IEEE 1149.1 Test Access Port controller, and
the Instruction Register with decodes.
Figure 18: VersaRing I/O Interface The public boundary-scan instructions are always available
prior to configuration. After configuration, the public instruc-
Boundary Scan tions and any USERCODE instructions are only available if
The “bed of nails” has been the traditional method of testing specified in the design. While SAMPLE and BYPASS are
electronic assemblies. This approach has become less available during configuration, it is recommended that
appropriate, due to closer pin spacing and more sophisti- boundary-scan operations not be performed during this
cated assembly methods like surface-mount technology transitory period.
and multi-layer boards. The IEEE boundary scan standard In addition to the test instructions outlined above, the
1149.1 was developed to facilitate board-level testing of boundary-scan circuitry can be used to configure the FPGA
electronic assemblies. Design and test engineers can device, and to read back the configuration data.
imbed a standard test logic structure in their device to
All of the XC4000 boundary-scan modes are supported in
achieve high fault coverage for I/O and internal logic. This
the XC5200 family. Three additional outputs for the User-
structure is easily implemented with a four-pin interface on
Register are provided (Reset, Update, and Shift), repre-
any boundary scan-compatible IC. IEEE 1149.1-compatible
senting the decoding of the corresponding state of the
devices may be serial daisy-chained together, connected in
boundary-scan internal state machine.
parallel, or a combination of the two.

R
DATA IN
1 sd
D Q D Q
0
LE
1
IOB.O 0
IOB.T 0
1 sd
D Q D Q 1
0
IOB IOB sd
1
D Q D Q
0
IOB IOB
LE
IOB IOB
1
IOB.I
0
IOB IOB
1 sd
IOB IOB D Q D Q
0
LE
IOB IOB
1
0
IOB IOB IOB.O
BYPASS
REGISTER
IOB.T 0
M TDO
U 1 sd
TDI
X 0
LE 7
M
TDI
U INSTRUCTION REGISTER
TDO X
BYPASS 1 sd
REGISTER D Q D Q
IOB IOB 0
LE
IOB IOB
1
IOB IOB IOB.I
0
IOB IOB 1 sd
D Q D Q
0
IOB IOB
LE
IOB IOB 0
1
IOB IOB IOB.O
IOB IOB IOB IOB IOB SHIFT/ CLOCK DATA

CAPTURE REGISTER
X1523_01
Figure 19: XC5200-Series Boundary Scan Logic
XC5200-Series devices can also be configured through the

boundary scan logic. See XAPP 017 for more information.

R
Data Registers Bit Sequence

The primary data register is the boundary scan register. The bit sequence within each IOB is: 3-State, Out, In. The
For each IOB pin in the FPGA, bonded or not, it includes data-register cells for the TAP pins TMS, TCK, and TDI
three bits for In, Out and 3-State Control. Non-IOB pins have an OR-gate that permanently disables the output
have appropriate partial bit population for In or Out only. buffer if boundary-scan operation is selected. Conse-
PROGRAM, CCLK and DONE are not included in the quently, it is impossible for the outputs in IOBs used by TAP
boundary scan register. Each EXTEST CAPTURE-DR inputs to conflict with TAP operation. TAP data is taken
state captures all In, Out, and 3-State pins. directly from the pin, and cannot be overwritten by injected
The data register also includes the following non-pin bits: boundary-scan data.
TDO.T, and TDO.O, which are always bits 0 and 1 of the The primary global clock inputs (PGCK1-PGCK4) are
data register, respectively, and BSCANT.UPD, which is taken directly from the pins, and cannot be overwritten with
always the last bit of the data register. These three bound- boundary-scan data. However, if necessary, it is possible to
ary scan bits are special-purpose Xilinx test signals. drive the clock input from boundary scan. The external
clock source is 3-stated, and the clock net is driven with
The other standard data register is the single flip-flop
BYPASS register. It synchronizes data being passed boundary scan data through the output driver in the
through the FPGA to the next downstream boundary scan clock-pad IOB. If the clock-pad IOBs are used for non-clock
device. signals, the data may be overwritten normally.
The FPGA provides two additional data registers that can Pull-up and pull-down resistors remain active during
be specified using the BSCAN macro. The FPGA provides boundary scan. Before and during configuration, all pins
two user pins (BSCAN.SEL1 and BSCAN.SEL2) which are are pulled up. After configuration, the choice of internal
the decodes of two user instructions, USER1 and USER2. pull-up or pull-down resistor must be taken into account
when designing test vectors to detect open-circuit PC
For these instructions, two corresponding pins
traces.
(BSCAN.TDO1 and BSCAN.TDO2) allow user scan data to
be shifted out on TDO. The data register clock From a cavity-up view of the chip (as shown in XDE or
(BSCAN.DRCK) is available for control of test logic which Epic), starting in the upper right chip corner, the boundary
the user may wish to implement with CLBs. The NAND of scan data-register bits are ordered as shown in Table 8.
TCK and RUN-TEST-IDLE is also provided (BSCAN.IDLE). The device-specific pinout tables for the XC5200 Series
include the boundary scan locations for each IOB pin.
Instruction Set
Table 8: Boundary Scan Bit Sequence
The XC5200-Series boundary scan instruction set also
includes instructions to configure the device and read back Bit Position I/O Pad Location
the configuration data. The instruction set is coded as Bit 0 (TDO) Top-edge I/O pads (right to left)
shown in Table 7. Bit 1 ...
Table 7: Boundary Scan Instructions ... Left-edge I/O pads (top to bottom)
... Bottom-edge I/O pads (left to right)
Instruction Test I/O Data
TDO Source ... Right-edge I/O pads (bottom to top)
I2 I1 I0 Selected Source
Bit N (TDI) BSCANT.UPD
0 0 0 EXTEST DR DR
0 0 1 SAMPLE/ DR Pin/Logic BSDL (Boundary Scan Description Language) files for
PRELOAD XC5200-Series devices are available on the Xilinx web site
0 1 0 USER 1 BSCAN. User Logic in the File Download area.
TDO1
0 1 1 USER 2 BSCAN. User Logic
Including Boundary Scan
TDO2 If boundary scan is only to be used during configuration, no
1 0 0 READBACK Readback Pin/Logic special elements need be included in the schematic or HDL
Data code. In this case, the special boundary scan pins TDI,
1 0 1 CONFIGURE DOUT Disabled TMS, TCK and TDO can be used for user functions after
configuration.
1 1 0 Reserved — —
1 1 1 BYPASS Bypass — To indicate that boundary scan remain enabled after config-
Register uration, include the BSCAN library symbol and connect pad
symbols to the TDI, TMS, TCK and TDO pins, as shown in
Figure 20.

R
Ground pins of the package will provide adequate decou-

Optional
pling.
To User
Logic
Output buffers capable of driving/sinking the specified 8 mA
IBUF
BSCAN
loads under specified worst-case conditions may be capa-
RESET ble of driving/sinking up to 10 times as much current under
UPDATE
best case conditions.
SHIFT
TDI TDO
TMS DRCK tance and reducing simultaneous output transitions in the
TCK IDLE same direction. It may also be beneficial to locate heavily
To User
From TDO1 SEL1 Logic loaded output buffers near the Ground pads. The I/O Block
User Logic TDO2 SEL2 output buffers have a slew-rate limited mode (default)
X9000
which should be used where output rise and fall times are
Figure 20: Boundary Scan Schematic Example not speed-critical.
Even if the boundary scan symbol is used in a schematic,

GND
the input pins TMS, TCK, and TDI can still be used as
inputs to be routed to internal logic. Care must be taken not Ground and
to force the chip into an undesired boundary scan state by Vcc Ring for
I/O Drivers
inadvertently applying boundary scan input patterns to
these pins. The simplest way to prevent this is to keep
TMS High, and then apply whatever signal is desired to TDI
and TCK.
Vcc Vcc
Avoiding Inadvertent Boundary Scan
If TMS or TCK is used as user I/O, care must be taken to Logic
Power Grid
ensure that at least one of these pins is held constant dur- 7
ing configuration. In some applications, a situation may
occur where TMS or TCK is driven during configuration.
This may cause the device to go into boundary scan mode
GND X5422
and disrupt the configuration process.
To prevent activation of boundary scan during configura- Figure 21: XC5200-Series Power Distribution
tion, do either of the following:
• TMS: Tie High to put the Test Access Port controller Pin Descriptions
in a benign RESET state There are three types of pins in the XC5200-Series
• TCK: Tie High or Low—do not toggle this clock input. devices:
For more information regarding boundary scan, refer to the • Permanently dedicated pins
Xilinx Application Note XAPP 017, “Boundary Scan in • User I/O pins that can have special functions
XC4000 and XC5200 Devices.“ • Unrestricted user-programmable I/O pins.
Power Distribution Before and during configuration, all outputs not used for the
configuration process are 3-stated and pulled high with a
Power for the FPGA is distributed through a grid to achieve 20 kΩ - 100 kΩ pull-up resistor.
After configuration, if an IOB is unused it is configured as
Inside the FPGA, a dedicated Vcc and Ground ring sur-
an input with a 20 kΩ - 100 kΩ pull-up resistor.
rounding the logic array provides power to the I/O drivers,
as shown in Figure 21. An independent matrix of Vcc and Device pins for XC5200-Series devices are described in
Ground lines supplies the interior logic of the device. Table 9. Pin functions during configuration for each of the
seven configuration modes are summarized in “Pin Func-
This power distribution grid provides a stable supply and
tions During Configuration” on page 124, in the “Configura-
ground for all internal logic, providing the external package
tion Timing” section.
power pins are all connected and appropriately decoupled.
Typically, a 0.1 µF capacitor connected near the Vcc and

R
I/O I/O
During After
Five or more (depending on package) connections to the nominal +5 V supply voltage.
VCC I I All must be connected, and each must be decoupled with a 0.01 - 0.1 µF capacitor to
Ground.
Four or more (depending on package type) connections to Ground. All must be con-
GND I I
nected.
During configuration, Configuration Clock (CCLK) is an output in Master modes or Asyn-
chronous Peripheral mode, but is an input in Slave mode, Synchronous Peripheral
mode, and Express mode. After configuration, CCLK has a weak pull-up resistor and
CCLK I or O I can be selected as the Readback Clock. There is no CCLK High time restriction on
XC5200-Series devices, except during Readback. See “Violating the Maximum High
and Low Time Specification for the Readback Clock” on page 113 for an explanation of
this exception.
indicates the completion of the configuration process. As an input, a Low level on
DONE can be configured to delay the global logic initialization and the enabling of out-
DONE I/O O puts.
The exact timing, the clock source for the Low-to-High transition, and the optional pull-
up resistor are selected as options in the program that creates the configuration bit-
stream. The resistor is included by default.
PROGRAM is an active Low input that forces the FPGA to clear its configuration mem-
PROGRAM I I
executes a complete clear cycle, before it goes into a WAIT state and releases INIT.
The PROGRAM pin has an optional weak pull-up after configuration.
During Peripheral mode configuration, this pin indicates when it is appropriate to write
another byte of data into the FPGA. The same status is also available on D7 in Asyn-
RDY/BUSY O I/O chronous Peripheral mode, if a read operation is performed when the device is selected.
After configuration, RDY/BUSY is a user-programmable I/O pin.
RDY/BUSY is pulled High with a high-impedance pull-up prior to INIT going High.
During Master Parallel configuration, each change on the A0-A17 outputs is preceded
by a rising edge on RCLK, a redundant output signal. RCLK is useful for clocked
RCLK O I/O
PROMs. It is rarely used during configuration. After configuration, RCLK is a user-pro-
grammable I/O pin.
As Mode inputs, these pins are sampled before the start of configuration to determine
the configuration mode to be used. After configuration, M0, M1, and M2 become user-
programmable I/O.
M0, M1, M2 I I/O
During configuration, these pins have weak pull-up resistors. For the most popular con-
figuration mode, Slave Serial, the mode pins can thus be left unconnected. A pull-down
resistor value of 3.3 kΩ is recommended for other modes.
this pin is a 3-state output, after configuration is completed.
TDO O O This pin can be user output only when called out by special schematic definitions. To
use this pin, place the library component TDO instead of the usual pad symbol. An out-
put buffer must still be used.

R
I/O I/O
During After
I/O
I or I
(JTAG)
In this case, they must be called out by special schematic definitions. To use these pins,
High During Configuration (HDC) is driven High until the I/O go active. It is available as
INIT I/O I/O
occurred. After the I/O go active, INIT is a user-programmable I/O pin. 7
Four Global inputs each drive a dedicated internal global net with short delay and min-
imal skew. These internal global nets can also be driven from internal logic. If not used
GCK1 - Weak to drive a global net, any of these pins is a user-programmable I/O pin.
I or I/O
GCK4 Pull-up The GCK1-GCK4 pins provide the shortest path to the four Global Buffers. Any input
pad symbol connected directly to the input of a BUFG symbol is automatically placed on
one of these pins.
These four inputs are used in Asynchronous Peripheral mode. The chip is selected
when CS0 is Low and CS1 is High. While the chip is selected, a Low on Write Strobe
(WS) loads the data present on the D0 - D7 inputs into the internal data buffer. A Low
CS0, CS1, on Read Strobe (RS) changes D7 into a status output — High if Ready, Low if Busy —
I I/O
WS, RS and drives D0 - D6 High.
In Express mode, CS1 is used as a serial-enable signal for daisy-chaining.
WS and RS should be mutually exclusive, but if both are Low simultaneously, the Write
Strobe overrides. After configuration, these are user-programmable I/O pins.
During Master Parallel configuration, these 18 output pins address the configuration
A0 - A17 O I/O
EPROM. After configuration, they are user-programmable I/O pins.
During Master Parallel, Peripheral, and Express configuration, these eight input pins re-
D0 - D7 I I/O
ceive configuration data. After configuration, they are user-programmable I/O pins.
DIN I I/O input receiving data on the rising edge of CCLK. During Parallel configuration, DIN is
the D0 input. After configuration, DIN is a user-programmable I/O pin.
During configuration in any mode but Express mode, DOUT is the serial configuration
data output that can drive the DIN of daisy-chained slave FPGAs. DOUT data changes
on the falling edge of CCLK.
DOUT O I/O In Express mode, DOUT is the status output that can drive the CS1 of daisy-chained
FPGAs, to enable and disable downstream devices.

R
I/O I/O
During After
Weak
Pull-up
tor (20 kΩ - 100 kΩ) that defines the logic level as High.
Configuration M1, and M0 inputs. There are three self-loading Master

modes, two Peripheral modes, and a Serial Slave mode,
Configuration is the process of loading design-specific pro-
gramming data into one or more FPGAs to define the func- Table 10: Configuration Modes
tional operation of the internal blocks and their Mode M2 M1 M0 CCLK Data
Master Serial 0 0 0 output Bit-Serial
mand registers of a programmable peripheral chip.
XC5200-Series devices use several hundred bits of config- Slave Serial 1 1 1 input Bit-Serial
uration data per CLB and its associated interconnects. Master 1 0 0 output Byte-Wide,
Each configuration bit defines the state of a static memory Parallel Up increment
cell that controls either a function look-up table bit, a multi- from 00000
plexer input, or an interconnect pass transistor. The devel- Master 1 1 0 output Byte-Wide,
opment system translates the design into a netlist file. It Parallel Down decrement
automatically partitions, places and routes the logic and from 3FFFF
generates the configuration data in PROM format. Peripheral 0 1 1 input Byte-Wide
Synchronous*
Special Purpose Pins Peripheral 1 0 1 output Byte-Wide
Three configuration mode pins (M2, M1, M0) are sampled Asynchronous
prior to configuration to determine the configuration mode. Express 0 1 0 input Byte-Wide
After configuration, these pins can be used as auxiliary I/O Reserved 0 0 1 — —
connections. The development system does not use these
resources unless they are explicitly specified in the design Note :*Peripheral Synchronous can be considered byte-wide
entry. This is done by placing a special pad symbol called Slave Parallel
MD2, MD1, or MD0 instead of the input or output pad sym- which is used primarily for daisy-chained devices. The sev-
bol. enth mode, called Express mode, is an additional slave
mode that allows high-speed parallel configuration. The
In XC5200-Series devices, the mode pins have weak
coding for mode selection is shown in Table 10.
pull-up resistors during configuration. With all three mode
pins High, Slave Serial mode is selected, which is the most Note that the smallest package, VQ64, only supports the
popular configuration mode. Therefore, for the most com- Master Serial, Slave Serial, and Express modes.A detailed
mon configuration mode, the mode pins can be left uncon- description of each configuration mode, with timing infor-
nected. (Note, however, that the internal pull-up resistor mation, is included later in this data sheet. During configu-
value can be as high as 100 kΩ.) After configuration, these ration, some of the I/O pins are used temporarily for the
pins can individually have weak pull-up or pull-down resis- configuration process. All pins used during configuration
tors, as specified in the design. A pull-down resistor value are shown in Table 13 on page 124.
of 3.3kΩ is recommended.
Master Modes
These pins are located in the lower left chip corner and are
near the readback nets. This location allows convenient The three Master modes use an internal oscillator to gener-
routing if compatibility with the XC2000 and XC3000 family ate a Configuration Clock (CCLK) for driving potential slave
conventions of M0/RT, M1/RD is desired. devices. They also generate address and timing for exter-
nal PROM(s) containing the configuration data.
Configuration Modes Master Parallel (Up or Down) modes generate the CCLK
XC5200 devices have seven configuration modes. These signal and PROM addresses and receive byte parallel
modes are selected by a 3-bit input code applied to the M2, data. The data is internally serialized into the FPGA
data-frame format. The up and down selection generates
starting addresses at either zero or 3FFFF, for compatibility
with different microprocessor addressing conventions. The

R
Master Serial mode generates CCLK and receives the con- Multi-Family Daisy Chain
figuration data in serial form from a Xilinx serial-configura-
All Xilinx FPGAs of the XC2000, XC3000, XC4000, and
tion PROM.
XC5200 Series use a compatible bitstream format and can,
CCLK speed is selectable as 1 MHz (default), 6 MHz, or 12 therefore, be connected in a daisy chain in an arbitrary
MHz. Configuration always starts at the default slow fre- sequence. There is, however, one limitation. If the chain
quency, then can switch to the higher frequency during the contains XC5200-Series devices, the master normally can-
first frame. Frequency tolerance is -50% to +50%. not be an XC2000 or XC3000 device.
Peripheral Modes The reason for this rule is shown in Figure 25 on page 109.
Since all devices in the chain store the same length count
The two Peripheral modes accept byte-wide data from a value and generate or receive one common sequence of
bus. A RDY/BUSY status is available as a handshake sig- CCLK pulses, they all recognize length-count match on the
nal. In Asynchronous Peripheral mode, the internal oscilla- same CCLK edge, as indicated on the left edge of
tor generates a CCLK burst signal that serializes the Figure 25. The master device then generates additional
byte-wide data. CCLK can also drive slave devices. In the CCLK pulses until it reaches its finish point F. The different
synchronous mode, an externally supplied clock input to families generate or require different numbers of additional
CCLK serializes the data. CCLK pulses until they reach F. Not reaching F means that
Slave Serial Mode the device does not really finish its configuration, although
DONE may have gone High, the outputs became active,
In Slave Serial mode, the FPGA receives serial configura- and the internal reset was released. For the
tion data on the rising edge of CCLK and, after loading its XC5200-Series device, not reaching F means that read-
configuration, passes additional data out, resynchronized back cannot be initiated and most boundary scan instruc-
on the next falling edge of CCLK. tions cannot be used.
Multiple slave devices with identical configurations can be The user has some control over the relative timing of these
wired with parallel DIN inputs. In this way, multiple devices events and can, therefore, make sure that they occur at the
can be configured simultaneously. proper time and the finish point F is reached. Timing is con-
trolled using options in the bitstream generation software. 7
Serial Daisy Chain
XC5200 devices always have the same number of CCLKs
Multiple devices with different configurations can be con-
in the power up delay, independent of the configuration
nected together in a “daisy chain,” and a single combined
mode, unlike the XC3000/XC4000 Series devices. To guar-
bitstream used to configure the chain of slave devices.
antee all devices in a daisy chain have finished the
To configure a daisy chain of devices, wire the CCLK pins power-up delay, tie the INIT pins together, as shown in
of all devices in parallel, as shown in Figure 28 on page Figure 27.
114. Connect the DOUT of each device to the DIN of the
next. The lead or master FPGA and following slaves each XC3000 Master with an XC5200-Series Slave
passes resynchronized configuration data coming from a Some designers want to use an XC3000 lead device in
single source. The header data, including the length count, peripheral mode and have the I/O pins of the
is passed through and is captured by each FPGA when it XC5200-Series devices all available for user I/O. Figure 22
recognizes the 0010 preamble. Following the length-count provides a solution for that case.
data, each FPGA outputs a High on DOUT until it has
This solution requires one CLB, one IOB and pin, and an
received its required number of data frames.
internal oscillator with a frequency of up to 5 MHz as a
After an FPGA has received its configuration data, it clock source. The XC3000 master device must be config-
passes on any additional frame start bits and configuration ured with late Internal Reset, which is the default option.
One CLB and one IOB in the lead XC3000-family device
clocks applied after memory initialization equals the value
are used to generate the additional CCLK pulse required by
of the 24-bit length count, the FPGAs begin the start-up
the XC5200-Series devices. When the lead device
sequence and become operational together. FPGA I/O are
removes the internal RESET signal, the 2-bit shift register
normally released two CCLK cycles after the last configura-
responds to its clock input and generates an active Low
tion bit is received. Figure 25 on page 109 shows the
output signal for the duration of the subsequent clock
start-up timing for an XC5200-Series device.
period. An external connection between this output and
The daisy-chained bitstream is not simply a concatenation CCLK thus creates the extra CCLK pulse.
of the individual bitstreams. The PROM file formatter must
be used to combine the bitstreams for a daisy-chained con-
figuration.

R
Pseudo Daisy Chain

Multiple devices with different configurations can be con-
nected together in a pseudo daisy chain, provided that all of
the devices are in Express mode. A single combined bit-
stream is used to configure the chain of Express mode
devices, but the input data bus must drive D0-D7 of each
device. Tie High the CS1 pin of the first device to be config-
OE/T
Output
Connected
ured, or leave it floating in the XC5200 since it has an inter-
Reset to CCLK nal pull-up. Connect the DOUT pin of each FPGA to the
0 0
1 0 Active Low Output
CS1 pin of the next device in the chain. The D0-D7 inputs
1 1
0 1
Active High Output are wired to each device in parallel. The DONE pins are
0 1 wired together, with one or more internal DONE pull-ups
. etc .
. . X5223 activated. Alternatively, a 4.7 kΩ external resistor can be
Figure 22: CCLK Generation for XC3000 Master used, if desired. (See Figure 37 on page 122.) CCLK pins
Driving an XC5200-Series Slave are tied together.
The requirement that all DONE pins in a daisy chain be
Express Mode wired together applies only to Express mode, and only if all
Express mode is similar to Slave Serial mode, except the devices in the chain are to become active simultaneously.
data is presented in parallel format, and is clocked into the All devices in Express mode are synchronized to the DONE
target device a byte at a time rather than a bit at a time. The pin. User I/O for each device become active after the
data is loaded in parallel into eight different columns: it is DONE pin for that device goes High. (The exact timing is
not internally serialized. Eight bits of configuration data are determined by options to the bitstream generation soft-
loaded with every CCLK cycle, therefore this configuration ware.) Since the DONE pin is open-drain and does not
mode runs at eight times the data rate of the other six drive a High value, tying the DONE pins of all devices
modes. In this mode the XC5200 family is capable of sup- together prevents all devices in the chain from going High
porting a CCLK frequency of 10 MHz, which is equivalent to until the last device in the chain has completed its configu-
an 80 MHz serial rate, because eight bits of configuration ration cycle.
data are being loaded per CCLK cycle. An XC5210 in the The status pin DOUT is pulled LOW two internal-oscillator
Express mode, for instance, can be configured in about 2 cycles (nominally 1 MHz) after INIT is recognized as High,
ms. The Express mode does not support CRC error check- and remains Low until the device’s configuration memory is
ing, but does support constant-field error checking. A full. Then DOUT is pulled High to signal the next device in
length count is not used in Express mode. the chain to accept the configuration data on the D7-D0
In the Express configuration mode, an external signal bus. All devices receive and recognize the six bytes of pre-
drives the CCLK input(s). The first byte of parallel configu- amble and length count, irrespective of the level on CS1;
ration data must be available at the D inputs of the FPGA but subsequent frame data is accepted only when CS1 is
devices a short set-up time before the second rising CCLK High and the device’s configuration memory is not already
edge. Subsequent data bytes are clocked in on each con- full.
secutive rising CCLK edge. See Figure 38 on page 123.
Setting CCLK Frequency
Bitstream generation currently generates a bitstream suffi-
For Master modes, CCLK can be generated in one of three
cient to program in all configuration modes except Express.
frequencies. In the default slow mode, the frequency is
Extra CCLK cycles are necessary to complete the configu-
nominally 1 MHz. In fast CCLK mode, the frequency is
ration, since in this mode data is read at a rate of eight bits
nominally 12 MHz. In medium CCLK mode, the frequency
per CCLK cycle instead of one bit per cycle. Normally the
is nominally 6 MHz. The frequency range is -50% to +50%.
entire start-up sequence requires a number of bits that is
The frequency is selected by an option when running the
equal to the number of CCLK cycles needed. An additional
bitstream generation software. If an XC5200-Series Master
five CCLKs (equivalent to 40 extra bits) will guarantee com-
is driving an XC3000- or XC2000-family slave, slow CCLK
pletion of configuration, regardless of the start-up options
mode must be used. Slow mode is the default.
chosen.
Multiple slave devices with identical configurations can be
wired with parallel D0-D7 inputs. In this way, multiple
devices can be configured simultaneously.

R
Table 11: XC5200 Bitstream Format Detection of an error results in the suspension of data load-
ing and the pulling down of the INIT pin. In Master modes,
Data Type Value Occurrences CCLK and address signals continue to operate externally.
Fill Byte 11111111 Once per bit- The user must detect INIT and initialize a new configuration
Preamble 11110010 stream
by pulsing the PROGRAM pin Low or cycling Vcc.
Length Counter COUNT(23:0)
Fill Byte 11111111
Start Byte 11111110 Once per data Table 12: Internal Configuration Data Structure
Data Frame * DATA(N-1:0) frame
Cyclic Redundancy Check or CRC(3:0) or PROM Xilinx
VersaBlock
Constant Field Check 0110 Device Size Serial PROM
Array
Fill Nibble 1111 (bits) Needed
Extend Write Cycle FFFFFF XC5202 8x8 42,416 XC1765D
Postamble 11111110 Once per de- XC5204 10 x 12 70,704 XC17128D
Fill Bytes (30) FFFF…FF vice
XC5206 14 x 14 106,288 XC17128D
Start-Up Byte FF Once per bit-
stream XC5210 18 x 18 165,488 XC17256D
*Bits per Frame (N) depends on device size, as described for XC5215 22 x 22 237,744 XC17256D
table 11. Bits per Frame = (34 x number of Rows) + 28 for the top + 28 for
the bottom + 4 splitter bits + 8 start bits + 4 error check bits + 4 fill
Data Stream Format bits * + 24 extended write bits
= (34 x number of Rows) + 100
The data stream (“bitstream”) format is identical for all con- * In the XC5202 (8 x 8), there are 8 fill bits per frame, not 4
figuration modes, with the exception of Express mode. In Number of Frames = (12 x number of Columns) + 7 for the left
Express mode, the device becomes active when DONE edge + 8 for the right edge + 1 splitter bit
= (12 x number of Columns) + 16
goes High, therefore no length count is required. Addition-
Program Data = (Bits per Frame x Number of Frames) + 48
ally, CRC error checking is not supported in Express mode.
header bits + 8 postamble bits + 240 fill bits + 8 start-up bits
The data stream formats are shown in Table 11. Express = (Bits per Frame x Number of Frames) + 304
7
mode data is shown with D0 at the left and D7 at the right. PROM Size = Program Data
For all other modes, bit-serial data is read from left to right,
and byte-parallel data is effectively assembled from this Cyclic Redundancy Check (CRC) for
serial bitstream, with the first bit in each byte assigned to Configuration and Readback
D0.
The Cyclic Redundancy Check is a method of error detec-
The configuration data stream begins with a string of eight tion in data transmission applications. Generally, the trans-
ones, a preamble code, followed by a 24-bit length count mitting system performs a calculation on the serial
and a separator field of ones (or 24 fill bits, in Express bitstream. The result of this calculation is tagged onto the
mode). This header is followed by the actual configuration data stream as additional check bits. The receiving system
data in frames. The length and number of frames depends performs an identical calculation on the bitstream and com-
on the device type (see Table 12). Each frame begins with pares the result with the received checksum.
a start field and ends with an error check. In all modes Each data frame of the configuration bitstream has four
except Express mode, a postamble code is required to sig-
error bits at the end, as shown in Table 11. If a frame data
nal the end of data for a single device. In all cases, addi-
error is detected during the loading of the FPGA, the con-
tional start-up bytes of data are required to provide four
figuration process with a potentially corrupted bitstream is
clocks for the startup sequence at the end of configuration. terminated. The FPGA pulls the INIT pin Low and goes into
Long daisy chains require additional startup bytes to shift a Wait state.
the last data through the chain. All startup bytes are
don’t-cares; these bytes are not included in bitstreams cre- During Readback, 11 bits of the 16-bit checksum are added
ated by the Xilinx software. to the end of the Readback data stream. The checksum is
computed using the CRC-16 CCITT polynomial, as shown
In Express mode, only non-CRC error checking is sup- in Figure 23. The checksum consists of the 11 most signif-
ported. In all other modes, a selection of CRC or non-CRC
icant bits of the 16-bit code. A change in the checksum indi-
error checking is allowed by the bitstream generation soft-
cates a change in the Readback bitstream. A comparison
ware. The non-CRC error checking tests for a designated to a previous checksum is meaningful only if the readback
end-of-frame field for each frame. For CRC error checking, data is independent of the current device state. CLB out-
the software calculates a running CRC and inserts a unique puts should not be included (Read Capture option not
four-bit partial check at the end of each frame. The 11-bit used). Statistically, one error out of 2048 might go undetec-
CRC check of the last frame of an FPGA includes the last
ted.
seven data bits.

R
Initialization
X2 X15
X16
This phase clears the configuration memory and estab-
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 lishes the configuration mode.
SERIAL DATA IN
The configuration memory is cleared at the rate of one
Polynomial: X16 + X15 + X2 + 1
frame per internal clock cycle (nominally 1 MHz). An
open-drain bidirectional signal, INIT, is released when the
configuration memory is completely cleared. The device
1 1 1 1 1 0 15 14 13 12 11 10 9 8 7 6 5
then tests for the absence of an external active-low level on
START BIT
LAST DATA FRAME CRC – CHECKSUM

INIT. The mode lines are sampled two internal clock cycles
later (nominally 2 µs).
X1789
The master device waits an additional 32 µs to 256 µs
(nominally 64-128 µs) to provide adequate time for all of the
slave devices to recognize the release of INIT as well. Then
Configuration Sequence the master device enters the Configuration phase.
There are four major steps in the XC5200-Series power-up

configuration sequence. Boundary Scan VCC No
Instructions 3V
• Power-On Time-Out Available:

Yes
• Initialization
Generate
• Configuration One Time-Out Pulse PROGRAM
of 4 ms = Low
• Start-Up Yes
EXTEST*
The full process is illustrated in Figure 24. SAMPLE/PRELOAD*
Completely Clear
BYPASS Configuration ~1.3 µs per Frame
CONFIGURE* Memory
Power-On Time-Out (*only when PROGRAM = High)
An internal power-on reset circuit is triggered when power INIT

High? if
No
is applied. When VCC reaches the voltage at which portions Master

Yes
of the FPGA begin to operate (i.e., performs a

Sample
write-and-read test of a sample pair of configuration mem- Mode Lines
ory bits), the programmable I/O buffers are 3-stated with Master CCLK
Goes Active after
active high-impedance pull-up resistors. A time-out delay 50 to 250 µs
Load One
— nominally 4 ms — is initiated to allow the power-supply Configuration
Data Frame
voltage to stabilize. For correct operation the power supply
must reach VCC(min) by the end of the time-out, and must Frame Yes Pull INIT Low
not dip below it thereafter. Error and Stop
No
There is no distinction between master and slave modes SAMPLE/PRELOAD Config-
uration No
with regard to the time-out delay. Instead, the INIT line is BYPASS
memory
Full
used to ensure that all daisy-chained devices have com- Yes
pleted initialization. Since XC2000 devices do not have this Pass

Configuration
signal, extra care must be taken to guarantee proper oper- Data to DOUT
ation when daisy-chaining them with XC5200 devices. For

proper operation with XC3000 devices, the RESET signal, CCLK
Count Equals No
Length
which is used in XC3000 to delay configuration, should be Count
Yes
connected to INIT.
Start-Up
If the time-out delay is insufficient, configuration should be Sequence
F
delayed by holding the INIT pin Low until the power supply
I/O Active
Operational
EXTEST
has reached operating levels. SAMPLE PRELOAD
BYPASS
If Boundary Scan
This delay is applied only on power-up. It is not applied USER 1
is Selected
USER 2
when reconfiguring an FPGA by pulsing the PROGRAM CONFIGURE X9017
READBACK
pin Low. During all three phases — Power-on, Initialization,
and Configuration — DONE is held Low; HDC, LDC, and Figure 24: Configuration Sequence
INIT are active; DOUT is driven; and all I/O buffers are dis-
abled.

R
Length Count Match

CCLK Period
CCLK
F
DONE
I/O
XC2000
Global Reset
F = Finished, no more
F
DONE Daisy-chain lead device
XC3000 must have latest F
I/O
default timing
Global Reset
F
DONE
C1 C2 C3 C4
XC4000E/EX I/O
XC5200/ C2 C3 C4
CCLK_NOSYNC
GSR Active
C2 C3 C4
DONE IN
F
DONE 7
C1, C2 or C3
XC4000E/EX I/O
XC5200/
Di Di+1
CCLK_SYNC
GSR Active
Di Di+1
F
DONE
C1 U2 U3 U4
I/O
XC4000E/EX
XC5200/ U2 U3 U4
UCLK_NOSYNC
GSR Active
U2 U3 U4
DONE IN
F
DONE
C1 U2
I/O
XC4000E/EX
XC5200/ Di Di+1 Di+2
UCLK_SYNC
GSR Active
Di Di+1 Di+2
Synchronization
X6700

R
Configuration Start-Up
The length counter begins counting immediately upon entry Start-up is the transition from the configuration process to
into the configuration state. In slave-mode operation it is the intended user operation. This transition involves a
important to wait at least two cycles of the internal 1-MHz change from one clock source to another, and a change
clock oscillator after INIT is recognized before toggling from interfacing parallel or serial configuration data where
CCLK and feeding the serial bitstream. Configuration will most outputs are 3-stated, to normal operation with I/O pins
not begin until the internal configuration logic reset is active in the user-system. Start-up must make sure that
released, which happens two cycles after INIT goes High. the user-logic ‘wakes up’ gracefully, that the outputs
A master device’s configuration is delayed from 32 to 256 become active without causing contention with the configu-
µs to ensure proper operation with any slave devices driven ration signals, and that the internal flip-flops are released
by the master device. from the global Reset at the right time.
The 0010 preamble code, included for all modes except Figure 25 describes start-up timing for the three Xilinx fam-
Express mode, indicates that the following 24 bits repre- ilies in detail. Express mode configuration always uses
sent the length count. The length count is the total number either CCLK_SYNC or UCLK_SYNC timing, the other con-
of configuration clocks needed to load the complete config- figuration modes can use any of the four timing sequences.
uration data. (Four additional configuration clocks are To access the internal start-up signals, place the STARTUP
required to complete the configuration process, as dis- library symbol.
cussed below.) After the preamble and the length count
have been passed through to all devices in the daisy chain, Start-up Timing
DOUT is held High to prevent frame start bits from reaching
Different FPGA families have different start-up sequences.
any daisy-chained devices. In Express mode, the length
count bits are ignored, and DOUT is held Low, to disable The XC2000 family goes through a fixed sequence. DONE
the next device in the pseudo daisy chain. goes High and the internal global Reset is de-activated one
CCLK period after the I/O become active.
A specific configuration bit, early in the first frame of a mas-
ter device, controls the configuration-clock rate and can The XC3000A family offers some flexibility. DONE can be
increase it by a factor of eight. Therefore, if a fast configu- programmed to go High one CCLK period before or after
ration clock is selected by the bitstream, the slower clock the I/O become active. Independent of DONE, the internal
rate is used until this configuration bit is detected. global Reset is de-activated one CCLK period before or
after the I/O become active.
Each frame has a start field followed by the frame-configu-
ration data bits and a frame error field. If a frame data error The XC4000/XC5200 Series offers additional flexibility.
is detected, the FPGA halts loading, and signals the error The three events — DONE going High, the internal Reset
by pulling the open-drain INIT pin Low. After all configura- being de-activated, and the user I/O going active — can all
tion frames have been loaded into an FPGA, DOUT again occur in any arbitrary sequence. Each of them can occur
follows the input data so that the remaining data is passed one CCLK period before or after, or simultaneous with, any
on to the next device. In Express mode, when the first of the others. This relative timing is selected by means of
device is fully programmed, DOUT goes High to enable the software options in the bitstream generation software.
next device in the chain. The default option, and the most practical one, is for DONE
to go High first, disconnecting the configuration data source
Delaying Configuration After Power-Up
and avoiding any contention when the I/Os become active
To delay master mode configuration after power-up, pull one clock later. Reset is then released another clock period
the bidirectional INIT pin Low, using an open-collector later to make sure that user-operation starts from stable
(open-drain) driver. (See Figure 12.) internal conditions. This is the most common sequence,
Using an open-collector or open-drain driver to hold INIT shown with heavy lines in Figure 25, but the designer can
Low before the beginning of master mode configuration modify it to meet particular requirements.
causes the FPGA to wait after completing the configuration Normally, the start-up sequence is controlled by the internal
memory clear operation. When INIT is no longer held Low device oscillator output (CCLK), which is asynchronous to
externally, the device determines its configuration mode by the system clock.
capturing its mode pins, and is ready to start the configura-
XC4000/XC5200 Series offers another start-up clocking
tion process. A master device waits up to an additional 250
option, UCLK_NOSYNC. The three events described
µs to make sure that any slaves in the optional daisy chain
above need not be triggered by CCLK. They can, as a con-
have seen that INIT is High.
figuration option, be triggered by a user clock. This means
that the device can wake up in synchronism with the user
system.

R
When the UCLK_SYNC option is enabled, the user can ship between CCLK and the user clock. This arbitration
externally hold the open-drain DONE output Low, and thus causes an unavoidable one-cycle uncertainty in the timing
stall all further progress in the start-up sequence until of the rest of the start-up sequence.
DONE is released and has gone High. This option can be
used to force synchronization of several FPGAs to a com- DONE Goes High to Signal End of Configuration
mon user clock, or to guarantee that all devices are suc- In all configuration modes except Express mode,
cessfully configured before any I/Os go active. XC5200-Series devices read the expected length count
If either of these two options is selected, and no user clock from the bitstream and store it in an internal register. The
is specified in the design or attached to the device, the chip length count varies according to the number of devices and
could reach a point where the configuration of the device is the composition of the daisy chain. Each device also
complete and the Done pin is asserted, but the outputs do counts the number of CCLKs during configuration.
not become active. The solution is either to recreate the Two conditions have to be met in order for the DONE pin to
bitstream specifying the start-up clock as CCLK, or to sup- go high:
ply the appropriate user clock.
• the chip's internal memory must be full, and
Start-up Sequence • the configuration length count must be met, exactly.
The Start-up sequence begins when the configuration This is important because the counter that determines
memory is full, and the total number of configuration clocks when the length count is met begins with the very first
received since INIT went High equals the loaded value of CCLK, not the first one after the preamble.
the length count. Therefore, if a stray bit is inserted before the preamble, or
The next rising clock edge sets a flip-flop Q0, shown in the data source is not ready at the time of the first CCLK,
Figure 26. Q0 is the leading bit of a 5-bit shift register. The the internal counter that holds the number of CCLKs will be
outputs of this register can be programmed to control three one ahead of the actual number of data bits read. At the
events. end of configuration, the configuration memory will be full,
but the number of bits in the internal counter will not match
• The release of the open-drain DONE output
the expected length count. 7
• The change of configuration-related pins to the user
function, activating all IOBs. As a consequence, a Master mode device will continue to
• The termination of the global Set/Reset initialization of send out CCLKs until the internal counter turns over to
all CLB and IOB storage elements. zero, and then reaches the correct length count a second
time. This will take several seconds [224 ∗ CCLK period]
The DONE pin can also be wire-ANDed with DONE pins of — which is sometimes interpreted as the device not config-
other FPGAs or with other external signals, and can then uring at all.
be used as input to bit Q3 of the start-up register. This is
called “Start-up Timing Synchronous to Done In” and is If it is not possible to have the data ready at the time of the
selected by either CCLK_SYNC or UCLK_SYNC. first CCLK, the problem can be avoided by increasing the
number in the length count by the appropriate value.
When DONE is not used as an input, the operation is called
“Start-up Timing Not Synchronous to DONE In,” and is In Express mode, there is no length count. The DONE pin
selected by either CCLK_NOSYNC or UCLK_NOSYNC. for each device goes High when the device has received its
quota of configuration data. Wiring the DONE pins of sev-
As a configuration option, the start-up control register
eral devices together delays start-up of all devices until all
beyond Q0 can be clocked either by subsequent CCLK are fully configured.
pulses or from an on-chip user net called STARTUP.CLK.
These signals can be accessed by placing the STARTUP Note that DONE is an open-drain output and does not go
library symbol. High unless an internal pull-up is activated or an external
pull-up is attached. The internal pull-up is activated as the
Start-up from CCLK default by the bitstream generation software.
If CCLK is used to drive the start-up, Q0 through Q3 pro- Release of User I/O After DONE Goes High
vide the timing. Heavy lines in Figure 25 show the default
timing, which is compatible with XC2000 and XC3000 By default, the user I/O are released one CCLK cycle after
devices using early DONE and late Reset. The thin lines the DONE pin goes High. If CCLK is not clocked after
indicate all other possible timing options. DONE goes High, the outputs remain in their initial state —
3-stated, with a 20 kΩ - 100 kΩ pull-up. The delay from
Start-up from a User Clock (STARTUP.CLK) DONE High to active user I/O is controlled by an option to
When, instead of CCLK, a user-supplied start-up clock is the bitstream generation software.
selected, Q1 is used to bridge the unknown phase relation-

R
Q3 Q1/Q4
STARTUP DONE
Q2
IN
* IOBs OPERATIONAL PER CONFIGURATION
* GLOBAL RESET OF
ALL CLB FLIP-FLOPS/LATCHES
1
0
GR ENABLE
GR INVERT
STARTUP.GR CONTROLLED BY STARTUP SYMBOL
IN THE USER SCHEMATIC (SEE
STARTUP.GTS LIBRARIES GUIDE)
GTS INVERT
GTS ENABLE
0
GLOBAL 3-STATE OF ALL IOBs
1
Q S
* DONE
" FINISHED "

1 1 ENABLES BOUNDARY
0 0 SCAN, READBACK AND
CONTROLS THE OSCILLATOR
Q0 Q1 Q2 Q3 Q4
FULL 1
S Q D Q D Q D Q D Q
LENGTH COUNT 0
K K K * K K
CLEAR MEMORY
CCLK 0
STARTUP.CLK 1
USER NET
M
CONFIGURATION BIT OPTIONS SELECTED BY USER
* * X9002
Figure 26: Start-up Logic
Release of Global Reset After DONE Goes High For detailed information, refer to the Xilinx application note
XAPP017, “Boundary Scan in XC4000 and XC5200
By default, Global Reset (GR) is released two CCLK cycles
Devices.”
after the DONE pin goes High. If CCLK is not clocked twice
after DONE goes High, all flip-flops are held in their initial Readback
reset state. The delay from DONE High to GR inactive is
controlled by an option to the bitstream generation soft- The user can read back the content of configuration mem-
ware. ory and the level of certain internal nodes without interfer-
ing with the normal operation of the device.
Configuration Complete After DONE Goes High
Readback not only reports the downloaded configuration
Three full CCLK cycles are required after the DONE pin bits, but can also include the present state of the device,
goes High, as shown in Figure 25 on page 109. If CCLK is represented by the content of all flip-flops and latches in
not clocked three times after DONE goes High, readback CLBs.
cannot be initiated and most boundary scan instructions
Note that in XC5200-Series devices, configuration data is
cannot be used.
not inverted with respect to configuration as it is in XC2000
Configuration Through the Boundary Scan and XC3000 families.
Pins Readback of Express mode bitstreams results in data that
does not resemble the original bitstream, because the bit-
XC5200-Series devices can be configured through the
stream format differs from other modes.
boundary scan pins.

R
XC5200-Series Readback does not use any dedicated Read Abort

pins, but uses four internal nets (RDBK.TRIG,
When the Read Abort option is selected, a High-to-Low
RDBK.DATA, RDBK.RIP and RDBK.CLK) that can be
transition on RDBK.TRIG terminates the readback opera-
routed to any IOB. To access the internal Readback sig-
tion and prepares the logic to accept another trigger.
nals, place the READBACK library symbol and attach the
appropriate pad symbols, as shown in Figure 27. After an aborted readback, additional clocks (up to one
readback clock per configuration frame) may be required to
After Readback has been initiated by a Low-to-High transi-
re-initialize the control logic. The status of readback is indi-
tion on RDBK.TRIG, the RDBK.RIP (Read In Progress)
cated by the output control net RDBK.RIP. RDBK.RIP is
output goes High on the next rising edge of RDBK.CLK.
High whenever a readback is in progress.
Subsequent rising edges of this clock shift out Readback
data on the RDBK.DATA net. Clock Select
Readback data does not include the preamble, but starts CCLK is the default clock. However, the user can insert
with five dummy bits (all High) followed by the Start bit another clock on RDBK.CLK. Readback control and data
(Low) of the first frame. The first two data bits of the first are clocked on rising edges of RDBK.CLK. If readback
frame are always High. must be inhibited for security reasons, the readback control
Each frame ends with four error check bits. They are read nets are simply not connected.
read back as High. An additional Start bit (Low) and an Violating the Maximum High and Low Time
11-bit Cyclic Redundancy Check (CRC) signature follow, Specification for the Readback Clock
before RDBK.RIP returns Low. The readback clock has a maximum High and Low time
IF UNCONNECTED, specification. In some cases, this specification cannot be
DEFAULT IS CCLK
met. For example, if a processor is controlling readback,
an interrupt may force it to stop in the middle of a readback.
This necessitates stopping the clock, and thus violating the
CLK DATA READ_DATA
MD1 specification.
READ_TRIGGER TRIG
READBACK
RIP
OBUF 7
MD0 The specification is mandatory only on clocking data at the
IBUF X1786
end of a frame prior to the next start bit. The transfer mech-
Figure 27: Readback Schematic Example anism will load the data to a shift register during the last six
clock cycles of the frame, prior to the start bit of the follow-
Readback Options ing frame. This loading process is dynamic, and is the
Readback options are: Read Capture, Read Abort, and source of the maximum High and Low time requirements.
Clock Select. They are set with the bitstream generation Therefore, the specification only applies to the six clock
software. cycles prior to and including any start bit, including the
Read Capture clocks before the first start bit in the readback data stream.
When the Read Capture option is selected, the readback the register is not dynamic. Thus, it can be shifted out just
data stream includes sampled values of CLB and IOB sig- like a regular shift register.
nals. The rising edge of RDBK.TRIG latches the inverted
values of the CLB outputs and the IOB output and input sig- The user must precisely calculate the location of the read-
nals. Note that while the bits describing configuration back data relative to the frame. The system must keep
(interconnect and function generators) are not inverted, the track of the position within a data frame, and disable inter-
CLB and IOB output signals are inverted. rupts before frame boundaries. Frame lengths and data for-
mats are listed in Table 11 and Table 12.
When the Read Capture option is not selected, the values
of the capture bits reflect the configuration data originally Readback with the XChecker Cable
written to those memory locations.
The readback signals are located in the lower-left corner of Logic Probe uses the readback feature for bitstream verifi-
the device. cation. It can also display selected internal signals on the
PC or workstation screen, functioning as a low-cost in-cir-
cuit emulator.

R
Configuration Timing There is an internal delay of 0.5 CCLK periods, which

The seven configuration modes are discussed in detail in the next FPGA in the daisy chain accepts data on the sub-
this section. Timing specifications are included. sequent rising CCLK edge.
Slave Serial Mode Figure 28 shows a full master/slave system. An
XC5200-Series device in Slave Serial mode should be con-
In Slave Serial mode, an external signal drives the CCLK nected as shown in the third device from the left.
input of the FPGA. The serial configuration bitstream must
be available at the DIN input of the lead FPGA a short Slave Serial mode is selected by a <111> on the mode pins
setup time before each rising CCLK edge. (M2, M1, M0). Slave Serial is the default mode if the mode
pins are left unconnected, as they have weak pull-up resis-
tors during configuration.
NOTE: NOTE:
M2, M1, M0 can be shorted M2, M1, M0 can be shorted
to Ground if not used as I/O to VCC if not used as I/O
VCC
N/C
4.7 KΩ 4.7 KΩ 4.7 KΩ 4.7 KΩ 4.7 KΩ
4.7 KΩ
M0 M1 M0 M1 M0 M1 PWRDN
M2 N/C M2 M2
VCC CCLK CCLK

XC5200
Spartan,
MASTER XC1700E +5 V XC3100A
4.7 KΩ XC4000E/EX,
SERIAL SLAVE
XC5200
CCLK CLK VPP
SLAVE
DIN DATA
PROGRAM X9003_01
DIN Bit n Bit n + 1
CCLK
4 TCCH 3 TCCO
DOUT
Bit n - 1 Bit n
(Output)
X5379

CCLK

R
Master Serial Mode The value increases from a nominal 1 MHz, to a nominal 12
MHz. Be sure that the serial PROM and slaves are fast
In Master Serial mode, the CCLK output of the lead FPGA enough to support this data rate. The Medium ConfigRate
drives a Xilinx Serial PROM that feeds the FPGA DIN input. option changes the frequency to a nominal 6 MHz.
Each rising edge of the CCLK output increments the Serial XC2000, XC3000/A, and XC3100A devices do not support
PROM internal address counter. The next data bit is put on
the Fast or Medium ConfigRate options.
the SPROM data output, connected to the FPGA DIN pin.
The lead FPGA accepts this data on the subsequent rising The SPROM CE input can be driven from either LDC or
CCLK edge. DONE. Using LDC avoids potential contention on the DIN
pin, if this pin is configured as user-I/O, but LDC is then
The lead FPGA then presents the preamble data—and all restricted to be a permanently High user output after con-
data that overflows the lead device—on its DOUT pin. figuration. Using DONE can also avoid contention on DIN,
There is an internal pipeline delay of 1.5 CCLK periods, provided the DONE before I/O enable option is invoked.
which means that DOUT changes on the falling CCLK
edge, and the next FPGA in the daisy chain accepts data Figure 28 on page 114 shows a full master/slave system.
on the subsequent rising CCLK edge. The leftmost device is in Master Serial mode.
In the bitstream generation software, the user can specify Master Serial mode is selected by a <000> on the mode
Fast ConfigRate, which, starting several bits into the first pins (M2, M1, M0).
frame, increases the CCLK frequency by a factor of twelve.
CCLK
(Output)
2 TCKDS
1 TDSCK

7

(Output)
X3223

CCLK

In the two Master Parallel modes, the lead FPGA directly The PROM address pins can be incremented or decre-
addresses an industry-standard byte-wide EPROM, and mented, depending on the mode pin settings. This option
accepts eight data bits just before incrementing or decre- allows the FPGA to share the PROM with a wide variety of
menting the address outputs. microprocessors and microcontrollers. Some processors
must boot from the bottom of memory (all zeros) while oth-
The eight data bits are serialized in the lead FPGA, which
ers must boot from the top. The FPGA is flexible and can
then presents the preamble data—and all data that over-
load its configuration bitstream from either end of the mem-
flows the lead device—on its DOUT pin. There is an inter-
ory.
nal delay of 1.5 CCLK periods, after the rising CCLK edge
that accepts a byte of data (and also changes the EPROM Master Parallel Up mode is selected by a <100> on the
address) until the falling CCLK edge that makes the LSB mode pins (M2, M1, M0). The EPROM addresses start at
(D0) of this byte appear at DOUT. This means that DOUT 00000 and increment.
changes on the falling CCLK edge, and the next FPGA in
Master Parallel Down mode is selected by a <110> on the
mode pins. The EPROM addresses start at 3FFFF and
CCLK edge.
decrement.

R
HIGH TO DIN OF OPTIONAL

or DAISY-CHAINED FPGAS
3.3 K N/C
LOW
N/C
M0 M1 M2 TO CCLK OF OPTIONAL
DAISY-CHAINED FPGAS
CCLK
DOUT
NOTE:M0 can be shorted
to Ground if not used A17 ... M0 M1 M2
as I/O.
XC5200 A16 ... DIN DOUT
VCC Master
Parallel A15 ... EPROM
(8K x 8) CCLK
A14 ...
4.7K (OR LARGER)
USER CONTROL OF HIGHER XC5200/
INIT A13 ... XC4000E/EX/
ORDER PROM ADDRESS BITS
CAN BE USED TO SELECT BETWEEN Spartan
A12 A12 SLAVE
ALTERNATIVE CONFIGURATIONS
A11 A11
PROGRAM
A10 A10
PROGRAM A9 A9
DONE INIT
D7 A8 A8
D6 A7 A7 D7
D5 A6 A6 D6
D4 A5 A5 D5
D3 A4 A4 D4
D2 A3 A3 D3
D1 A2 A2 D2
D0 A1 A1 D1
A0 A0 D0
DONE OE
CE
DATA BUS 8
PROGRAM
X9004_01

R
.
A0-A17
1 TRAC
D0-D7
Byte
2 TDRC 3 TRCD
RCLK
(output)
7 CCLKs CCLK
CCLK
(output)
DOUT
(output) D6 D7
Byte n - 1 X6078

Delay to Address valid 1 TRAC 0 200 ns
CCLK Data setup time 2 TDRC 60 ns
Data hold time 3 TRCD 0 ns 7
Note: 1. At power-up, VCC must rise from 2.0 V to VCC min in less then 25 ms, otherwise delay configuration by pulling PROGRAM
Low until VCC is Valid.
2. The first Data byte is loaded and CCLK starts at the end of the first RCLK active cycle (rising edge).
This timing diagram shows that the EPROM requirements are extremely relaxed. EPROM access time can be longer than
500 ns. EPROM data output has no hold-time requirements.

R
Synchronous Peripheral Mode for test purposes. Note that RDY/BUSY is pulled High with
a high-impedance pullup prior to INIT going High.
Synchronous Peripheral mode can also be considered
Slave Parallel mode. An external signal drives the CCLK The lead FPGA serializes the data and presents the pre-
input(s) of the FPGA(s). The first byte of parallel configura- amble data (and all data that overflows the lead device) on
tion data must be available at the Data inputs of the lead its DOUT pin. There is an internal delay of 1.5 CCLK peri-
FPGA a short setup time before the rising CCLK edge. ods, which means that DOUT changes on the falling CCLK
Subsequent data bytes are clocked in on every eighth con- edge, and the next FPGA in the daisy chain accepts data
secutive rising CCLK edge. on the subsequent rising CCLK edge.
The same CCLK edge that accepts data, also causes the In order to complete the serial shift operation, 10 additional
RDY/BUSY output to go High for one CCLK period. The pin CCLK rising edges are required after the last data byte has
name is a misnomer. In Synchronous Peripheral mode it is been loaded, plus one more CCLK cycle for each
really an ACKNOWLEDGE signal. Synchronous operation daisy-chained device.
does not require this response, but it is a meaningful signal Synchronous Peripheral mode is selected by a <011> on
the mode pins (M2, M1, M0).
NOTE:
M2 can be shorted to Ground
if not used as I/O
N/C 3.3 kΩ N/C
M0 M1 M2 M0 M1 M2
CLOCK CCLK CCLK
OPTIONAL
8 DAISY-CHAINED
DATA BUS D0-7 FPGAs
DOUT DIN DOUT
VCC XC5200
SYNCHRO- XC5200E/EX
NOUS SLAVE
4.7 kΩ PERIPHERAL
CONTROL RDY/BUSY
SIGNALS INIT DONE INIT DONE
3.3 kΩ
X9005
Figure 33: Synchronous Peripheral Mode Circuit Diagram

R
TCCL
CCLK
1 TIC
3 TCD
INIT 2 TDC
BYTE BYTE
D0 - D7 0 1
BYTE 0 OUT BYTE 1 OUT
DOUT 0 1 2 3 4 5 6 7 0 1
RDY/BUSY
X6096

INIT (High) setup time 1 TIC 5 µs
D0 - D7 setup time 2 TDC 60 ns
D0 - D7 hold time 3 TCD 0 ns
CCLK
CCLK Frequency FCC 8 MHz 7
Notes: 1. Peripheral Synchronous mode can be considered Slave Parallel mode. An external CCLK provides timing, clocking in the
first data byte on the second rising edge of CCLK after INIT goes high. Subsequent data bytes are clocked in on every
eighth consecutive rising edge of CCLK.
2. The RDY/BUSY line goes High for one CCLK period after data has been clocked in, although synchronous operation does
not require such a response.
3. The pin name RDY/BUSY is a misnomer. In synchronous peripheral mode this is really an ACKNOWLEDGE signal.
4.Note that data starts to shift out serially on the DOUT pin 0.5 CCLK periods after it was loaded in parallel. Therefore,
additional CCLK pulses are clearly required after the last byte has been loaded.
Figure 34: Synchronous Peripheral Mode Programming Switching Characteristics

R
Asynchronous Peripheral Mode The READY/BUSY handshake can be ignored if the delay
from any one Write to the end of the next Write is guaran-
Write to FPGA teed to be longer than 10 CCLK periods.
Asynchronous Peripheral mode uses the trailing edge of Status Read
the logic AND condition of WS and CS0 being Low and RS
and CS1 being High to accept byte-wide data from a micro- The logic AND condition of the CS0, CS1 and RS inputs
processor bus. In the lead FPGA, this data is loaded into a puts the device status on the Data bus.
double-buffered UART-like parallel-to-serial converter and • D7 High indicates Ready
is serially shifted into the internal logic. • D7 Low indicates Busy
The lead FPGA presents the preamble data (and all data • D0 through D6 go unconditionally High
that overflows the lead device) on its DOUT pin. The It is mandatory that the whole start-up sequence be started
RDY/BUSY output from the lead FPGA acts as a hand- and completed by one byte-wide input. Otherwise, the pins
shake signal to the microprocessor. RDY/BUSY goes Low used as Write Strobe or Chip Enable might become active
when a byte has been received, and goes High again when outputs and interfere with the final byte transfer. If this
the byte-wide input buffer has transferred its information transfer does not occur, the start-up sequence is not com-
into the shift register, and the buffer is ready to receive new pleted all the way to the finish (point F in Figure 25 on page
data. A new write may be started immediately, as soon as 109).
the RDY/BUSY output has gone Low, acknowledging
In this case, at worst, the internal reset is not released. At
receipt of the previous data. Write may not be terminated
best, Readback and Boundary Scan are inhibited. The
until RDY/BUSY is High again for one CCLK period. Note
that RDY/BUSY is pulled High with a high-impedance length-count value, as generated by the software, ensures
pull-up prior to INIT going High. that these problems never occur.
The length of the BUSY signal depends on the activity in Although RDY/BUSY is brought out as a separate signal,
microprocessors can more easily read this information on
the UART. If the shift register was empty when the new
one of the data lines. For this purpose, D7 represents the
byte was received, the BUSY signal lasts for only two
CCLK periods. If the shift register was still full when the RDY/BUSY status when RS is Low, WS is High, and the
two chip select lines are both active.
new byte was received, the BUSY signal can be as long as
nine CCLK periods. Asynchronous Peripheral mode is selected by a <101> on
Note that after the last byte has been entered, only seven the mode pins (M2, M1, M0).
of its bits are shifted out. CCLK remains High with DOUT
equal to bit 6 (the next-to-last bit) of the last byte entered.
N/C
N/C N/C
3.3 kΩ
M0 M1 M2 M0 M1 M2
DATA 8
D0–7 CCLK CCLK
BUS
OPTIONAL
DAISY-CHAINED
FPGAs
DOUT DIN DOUT
VCC ADDRESS CS0
ADDRESS DECODE XC5200
...
BUS LOGIC
ASYNCHRO- XC5200/
NOUS XC4000E/EX
4.7 kΩ
PERIPHERAL SLAVE
4.7 kΩ CS1
RS
WS
CONTROL RDY/BUSY
SIGNALS
INIT INIT
DONE DONE
REPROGRAM
PROGRAM PROGRAM
3.3 kΩ
X9006
Figure 35: Asynchronous Peripheral Mode Circuit Diagram

R
Write to LCA Read Status
WS/CS0 RS, CS0
RS, CS1 WS, CS1

1 TCA
3 TCD 7 4
2 TDC
READY D7
D0-D7 BUSY
CCLK
TWTRB 4
6 TBUSY
RDY/BUSY
DOUT Previous Byte D6 D7 D0 D1 D2
X6097

Effective Write time 1 TCA 100 ns
(CSO, WS=Low; RS, CS1=High
Write
DIN setup time 2 TDC 60 ns
DIN hold time 3 TCD 0 ns
7
RDY/BUSY delay after end of 4 TWTRB 60 ns
Write or Read
RDY/BUSY active after beginning 7 60 ns
RDY
of Read
RDY/BUSY Low output (Note 4) 6 TBUSY 2 9 CCLK
periods
Notes: 1. Configuration must be delayed until INIT pins of all daisy-chained FPGAs are high.
2. The time from the end of WS to CCLK cycle for the new byte of data depends on the completion of previous byte processing
and the phase of internal timing generator for CCLK.
4. TBUSY indicates that the double-buffered parallel-to-serial converter is not yet ready to receive new data. The shortest TBUSY
occurs when a byte is loaded into an empty parallel-to-serial converter. The longest TBUSY occurs when a new word is
loaded into the input register before the second-level buffer has started shifting out data.
This timing diagram shows very relaxed requirements. Data need not be held beyond the rising edge of WS. RDY/BUSY will
go active within 60 ns after the end of WS. A new write may be asserted immediately after RDY/BUSY goes Low, but write
may not be terminated until RDY/BUSY has been High for one CCLK period.
Figure 36: Asynchronous Peripheral Mode Programming Switching Characteristics

R
Express Mode ration memory is not already full. The status pin DOUT is
pulled Low two internal-oscillator cycles after INIT is recog-
Express mode is similar to Slave Serial mode, except that
nized as High, and remains Low until the device’s configu-
data is processed one byte per CCLK cycle instead of one
ration memory is full. DOUT is then pulled High to signal
bit per CCLK cycle. An external source is used to drive
the next device in the chain to accept the configuration data
CCLK, while byte-wide data is loaded directly into the con-
on the D0-D7 bus.
figuration data shift registers. A CCLK frequency of 10
MHz is equivalent to an 80 MHz serial rate, because eight The DONE pins of all devices in the chain should be tied
bits of configuration data are loaded per CCLK cycle. together, with one or more active internal pull-ups. If a
Express mode does not support CRC error checking, but large number of devices are included in the chain, deacti-
does support constant-field error checking. vate some of the internal pull-ups, since the Low-driving
DONE pin of the last device in the chain must sink the cur-
In Express mode, an external signal drives the CCLK input rent from all pull-ups in the chain. The DONE pull-up is
of the FPGA device. The first byte of parallel configuration activated by default. It can be deactivated using an option
data must be available at the D inputs of the FPGA a short
in the bitstream generation software.
setup time before the second rising CCLK edge. Subse-
quent data bytes are clocked in on each consecutive rising XC5200 devices in Express mode are always synchronized
CCLK edge. to DONE. The device becomes active after DONE goes
High. DONE is an open-drain output. With the DONE pins
If the first device is configured in Express mode, additional
tied together, therefore, the external DONE signal stays low
devices may be daisy-chained only if every device in the until all devices are configured, then all devices in the daisy
chain is also configured in Express mode. CCLK pins are chain become active simultaneously. If the DONE pin of a
tied together and D0-D7 pins are tied together for all device is left unconnected, the device becomes active as
devices along the chain. A status signal is passed from soon as that device has been configured.
DOUT to CS1 of successive devices along the chain. The
lead device in the chain has its CS1 input tied High (or float- Express mode is selected by a <010> on the mode pins
ing, since there is an internal pullup). Frame data is (M2, M1, M0).
accepted only when CS1 is High and the device’s configu-
VCC
NOTE:
M2, M1, M0 can be shorted
to Ground if not used as I/O
3.3 kΩ
8
To Additional
M0 M1 M2 M0 M1 M2 Optional
Daisy-Chained
Devices
CS1 DOUT CS1 DOUT
8 8
Optional
VCC XC5200 Daisy-Chained
XC5200
4.7KΩ
CCLK CCLK
To Additional
Optional
Daisy-Chained
CCLK
Devices
X6611_01

R
CCLK
1 TIC
INIT
TCD 3
2 T
DC
D0-D7 BYTE BYTE BYTE BYTE

0 1 2 3
Serial Data Out

(DOUT)
FPGA Filled
Internal INIT
RDY/BUSY
CS1 X5087

INIT (High) Setup time required 1 TIC 5 µs
DIN Setup time required 2 TDC 30 ns
DIN hold time required 3 TCD 0 ns 7
CCLK
CCLK frequency FCC 10 MHz
Note: If not driven by the preceding DOUT, CS1 must remain high until the device is fully configured.
Figure 38: Express Mode Programming Switching Characteristics

R
Table 13. Pin Functions During Configuration

CONFIGURATION MODE: <M2:M1:M0>
USER
SLAVE MASTER-SER SYN.PERIPH ASYN.PERIPH MASTER-HIGH MASTER-LOW EXPRESS OPERATION
<1:1:1> <0:0:0> <0:1:1> <1:0:1> <1:1:0> <1:0:0> <0:1:0>
A16 A16 GCK1-I/O
A17 A17 I/O
TDI TDI TDI TDI TDI TDI TDI TDI-I/O
TCK TCK TCK TCK TCK TCK TCK TCK-I/O
TMS TMS TMS TMS TMS TMS TMS TMS-I/O
I/O
M1 (HIGH) (I) M1 (LOW) (I) M1 (HIGH) (I) M1 (LOW) (I) M1 (HIGH) (I) M1 (LOW) (I) M1 (HIGH) (I) I/O
M0 (HIGH) (I) M0 (LOW) (I) M0 (HIGH) (I) M0 (HIGH) (I) M0 (LOW) (I) M0 (LOW) (I) M0 (LOW) (I) I/O
M2 (HIGH) (I) M2 (LOW) (I) M2 (LOW) (I) M2 (HIGH) (I) M2 (HIGH) (I) M2 (HIGH) (I) M2 (LOW) (I) I/O
GCK2-I/O
HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) HDC (HIGH) I/O
LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) LDC (LOW) I/O
INIT-ERROR INIT-ERROR INIT-ERROR INIT-ERROR INIT-ERROR INIT-ERROR INIT-ERROR I/O
I/O
DONE DONE DONE DONE DONE DONE DONE DONE
PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM (I) PROGRAM
DATA 7 (I) DATA 7 (I) DATA 7 (I) DATA 7 (I) DATA 7 (I) I/O
GCK3-I/O
CSO (I) I/O
RS (I) I/O
RDY/BUSY RDY/BUSY RCLK RCLK I/O
DIN (I) DIN (I) DATA 0 (I) DATA 0 (I) DATA 0 (I) DATA 0 (I) DATA 0 (I) I/O
DOUT DOUT DOUT DOUT DOUT DOUT DOUT I/O
CCLK (I) CCLK (O) CCLK (I) CCLK (O) CCLK (O) CCLK (O) CCLK (I) CCLK (I)
TDO TDO TDO TDO TDO TDO TDO TDO-I/O
WS (I) A0 A0 I/O
A1 A1 GCK4-I/O
CS1 (I) A2 A2 CS1 (I) I/O
A3 A3 I/O
A4 A4 I/O
A5 A5 I/O
A6 A6 I/O
A7 A7 I/O
A8 A8 I/O
A9 A9 I/O
A10 A10 I/O
A11 A11 I/O
A12 A12 I/O
A13 A13 I/O
A14 A14 I/O
A15 A15 I/O
ALL OTHERS
Notes: 1. A shaded table cell represents a 20-kΩ to 100-kΩ pull-up resistor before and during configuration.
2. (I) represents an input (O) represents an output.

R
Vcc T POR
RE-PROGRAM
>300 ns
PROGRAM
T PI
INIT
T ICCK TCCLK
<300 ns
M0, M1, M2
VALID DONE RESPONSE
(Required)
X1532
<300 ns
I/O
Master Modes
Power-On-Reset TPOR 2 15 ms
Program Latency TPI 6 70 µs per CLB column
CCLK (output) Delay TICCK 40 375 µs 7
period (slow) TCCLK 640 3000 ns
period (fast) TCCLK 100 375 ns
Slave and Peripheral Modes

Power-On-Reset TPOR 2 15 ms
Program Latency TPI 6 70 µs per CLB column
period (required) TCCLK 100 ns
Note: At power-up, VCC must rise from 2.0 to VCC min in less than 15 ms, otherwise delay configuration using PROGRAM until
VCC is valid.

R
XC5200 Program Readback Switching Characteristic Guidelines

Finished
Internal Net
3
T RTL
rdbk.TRIG
T RCRT
T RTRC 2
1
rdclk.I
4 T RCL T RCH 5
rdbk.RIP
6
T RCRR
T RCRD
7
X1790


Note 2: rdbk.TRIG is High prior to Finished, Finished will trigger the first Readback

R
XC5200 Switching Characteristics

Definition of Terms
follows:
families. Use as estimates, not for production.
Unmarked: Specifications not identified as either Advance or Preliminary are to be considered Final.1
XC5200 Operating Conditions

VCC Supply voltage relative to GND Commercial: 0°C to 85°C junction 4.75 5.25 V
Supply voltage relative to GND Industrial: -40°C to 100°C junction 4.5 5.5 V
XC5200 DC Characteristics Over Operating Conditions

Symbol Description Min Max Units 7
VOH High-level output voltage @ IOH = -8.0 mA, VCC min 3.86 V
VOL Low-level output voltage @ IOL = 8.0 mA, VCC max 0.4 V
IIL Leakage current -10 +10 µA
IRIN Pad pull-up (when selected) @ VIN = 0V (sample tested) 0.02 0.30 mA
Note: 1. With no output current loads, all package pins at Vcc or GND, either TTL or CMOS inputs, and the FPGA configured with a
tie option.
XC5200 Absolute Maximum Ratings

VIN Input voltage with respect to GND -0.5 to VCC +0.5 V
VTS Voltage applied to 3-state output -0.5 to VCC +0.5 V
TJ Junction temperature in plastic packages +125 °C
Junction temperature in ceramic packages +150 °C
Note: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress
ratings only, and functional operation of the device at these or any other conditions beyond those listed under Recommended
1. Notwithstanding the definition of the above terms, all specifications are subject to change without notice.

R
XC5200 Global Buffer Switching Characteristic Guidelines

in the simulator.
Max Max Max Max
(ns) (ns) (ns) (ns)
Global Signal Distribution TBUFG XC5202 9.1 8.5 8.0 6.9
From pad through global buffer, to any clock (CK) XC5204 9.3 8.7 8.2 7.6
XC5206 9.4 8.8 8.3 7.7
XC5210 9.4 8.8 8.5 7.7
XC5215 10.5 9.9 9.8 9.6
XC5200 Longline Switching Characteristic Guidelines

in the simulator.
Max Max Max Max
(ns) (ns) (ns) (ns)
TBUF driving a Longline TIO XC5202 6.0 3.8 3.0 2.0
TS XC5204 6.4 4.1 3.2 2.3
I O
XC5206 6.6 4.2 3.3 2.7
TBUF
XC5210 6.6 4.2 3.3 2.9
I to Longline, while TS is Low; i.e., buffer is constantly ac- XC5215 7.3 4.6 3.8 3.2
tive
TS going Low to Longline going from floating High or Low TON XC5202 7.8 5.6 4.7 4.0
to active Low or High XC5204 8.3 5.9 4.9 4.3
XC5206 8.4 6.0 5.0 4.4
XC5210 8.4 6.0 5.0 4.4
XC5215 8.9 6.3 5.3 4.5
TS going High to TBUF going inactive, not driving TOFF XC52xx 3.0 2.8 2.6 2.4
Longline
Note: 1. Die-size-dependent parameters are based upon XC5215 characterization. Production specifications will vary with array
size.

R
XC5200 CLB Switching Characteristic Guidelines

in the simulator.
Description Symbol
(ns) (ns) (ns) (ns) (ns) (ns) (ns) (ns)
F inputs to X output TILO 5.6 4.6 3.8 3.0
F inputs via transparent latch to Q TITO 8.0 6.6 5.4 4.3
DI inputs to DO output (Logic-Cell TIDO 4.3 3.5 2.8 2.4
Feedthrough)
F inputs via F5_MUX to DO output TIMO 7.2 5.8 5.0 4.3
Carry Delays
Incremental delay per bit TCY 0.7 0.6 0.5 0.5
Carry-in overhead from DI TCYDI 1.8 1.6 1.5 1.4
Carry-in overhead from F TCYL 3.7 3.2 2.9 2.4
Carry-out overhead to DO TCYO 4.0 3.2 2.5 2.1
Sequential Delays
Clock (CK) to out (Q) (Flip-Flop) TCKO 5.8 4.9 4.0 4.0
Gate (Latch enable) going active to out (Q) TGO 9.2 7.4 5.9 5.5
Set-up Time Before Clock (CK) 7
F inputs TICK 2.3 1.8 1.4 1.3
F inputs via F5_MUX TMICK 3.8 3.0 2.5 2.4
DI input TDICK 0.8 0.5 0.4 0.4
CE input TEICK 1.6 1.2 0.9 0.9
Hold Times After Clock (CK)
F inputs TCKI 0 0 0 0
F inputs via F5_MUX TCKMI 0 0 0 0
DI input TCKDI 0 0 0 0
CE input TCKEI 0 0 0 0
Clock Widths
Clock High Time TCH 6.0 6.0 6.0 6.0
Clock Low Time TCL 6.0 6.0 6.0 6.0
Toggle Frequency (MHz) (Note 3) FTOG 83 83 83 83
Reset Delays
Width (High) TCLRW 6.0 6.0 6.0 6.0
Delay from CLR to Q (Flip-Flop) TCLR 7.7 6.3 5.1 4.0
Delay from CLR to Q (Latch) TCLRL 6.5 5.2 4.2 3.0
Global Reset Delays
Width (High) TGCLRW 6.0 6.0 6.0 6.0
Delay from internal GR to Q TGCLR 14.7 12.1 9.1 8.0
Note: 1. The CLB K to Q output delay (TCKO) of any CLB, plus the shortest possible interconnect delay, is always longer than the
Data In hold-time requirement (TCKDI) of any CLB on the same die.
2. Timing is based upon the XC5215 device. For other devices, see Timing Calculator.
3. Maximum flip-flop toggle rate for export control purposes.

R
XC5200 Guaranteed Input and Output Parameters (Pin-to-Pin)

All values listed below are tested directly, and guaranteed over the operating conditions. The same parameters can also be
derived indirectly from the Global Buffer specifications. The delay calculator uses this indirect method, and may
overestimate because of worst-case assumptions. When there is a discrepancy between these two methods, the values
listed below should be used, and the derived values should be considered conservative overestimates.
Max Max Max Max
(ns) (ns) (ns) (ns)
Global Clock to Output Pad (fast) TICKOF XC5202 16.9 15.1 10.9 9.8
CLB Direct IOB
XC5204 17.1 15.3 11.3 9.9
BUFG Q
Connect (Max) XC5206 17.2 15.4 11.9 10.8
..
FAST .. XC5210 17.2 15.4 12.8 11.2
XC5215 19.0 17.0 12.8 11.7
Global Clock to Output Pad (slew-limited) TICKO XC5202 21.4 18.7 12.6 11.5
CLB Direct IOB
XC5204 21.6 18.9 13.3 11.9
Connect (Max) XC5206 21.7 19.0 13.6 12.5
BUFG Q ..
.. XC5210 21.7 19.0 15.0 12.9
Global Clock-to-Output Delay XC5215 24.3 21.2 15.0 13.1
Input Set-up Time (no delay) to CLB Flip-Flop TPSUF XC5202 2.5 2.0 1.9 1.9
IOB(NODELAY) Direct CLB XC5204 2.3 1.9 1.9 1.9
Connect
Input F,DI (Min)
Set-up XC5206 2.2 1.9 1.9 1.9
& Hold
Time XC5210 2.2 1.9 1.9 1.8
BUFG XC5215 2.0 1.8 1.7 1.7
Input Hold Time (no delay) to CLB Flip-Flop TPHF XC5202 3.8 3.8 3.5 3.5
IOB(NODELAY) Direct CLB XC5204 3.9 3.9 3.8 3.6
Connect
Input F,DI (Min)
Set-up XC5206 4.4 4.4 4.4 4.3
& Hold
Time XC5210 5.1 5.1 4.9 4.8
BUFG XC5215 5.8 5.8 5.7 5.6
Input Set-up Time (with delay) to CLB Flip-Flop DI Input TPSU XC5202 7.3 6.6 6.6 6.6
IOB Connect
Direct CLB XC5204 7.3 6.6 6.6 6.6
Input DI
Set-up XC5206 7.2 6.5 6.4 6.3
& Hold
Time XC5210 7.2 6.5 6.0 6.0
BUFG XC5215 6.8 5.7 5.7 5.7
Input Set-up Time (with delay) to CLB Flip-Flop F Input TPSUL XC5202 8.8 7.7 7.5 7.5
IOB Connect
Direct CLB XC5204 8.6 7.5 7.5 7.5
Input F (Min)
Set-up XC5206 8.5 7.4 7.4 7.4
& Hold
Time XC5210 8.5 7.4 7.4 7.3
BUFG XC5215 8.5 7.4 7.4 7.2
Input Hold Time (with delay) to CLB Flip-Flop TPH XC52xx 0 0 0 0
IOB Connect
Direct CLB
Input F,DI
Set-up (Min)
& Hold
Time
BUFG
Note: 1. These measurements assume that the CLB flip-flop uses a direct interconnect to or from the IOB. The INREG/ OUTREG
properties, or XACT-Performance, can be used to assure that direct connects are used. tPSU applies only to the CLB input
DI that bypasses the look-up table, which only offers direct connects to IOBs on the left and right edges of the die. tPSUL
applies to the CLB inputs F that feed the look-up table, which offers direct connect to IOBs on all four edges, as do the CLB
Q outputs.
2. When testing outputs (fast or slew-limited), half of the outputs on one side of the device are switching.

R
XC5200 IOB Switching Characteristic Guidelines

in the simulator.
Max Max Max Max
Description Symbol
(ns) (ns) (ns) (ns)
Input
Propagation Delays from CMOS or TTL Levels
Pad to I (no delay) TPI 5.7 5.0 4.8 3.3
Pad to I (with delay) TPID 11.4 10.2 10.2 9.5
Output
Propagation Delays to CMOS or TTL Levels
Output (O) to Pad (fast) TOPF 4.6 4.5 4.5 3.5
Output (O) to Pad (slew-limited) TOPS 9.5 8.4 8.0 5.0
From clock (CK) to output pad (fast), using direct connect between Q TOKPOF 10.1 9.3 8.3 7.5
and output (O)
From clock (CK) to output pad (slew-limited), using direct connect be- TOKPOS 14.9 13.1 11.8 10.0
tween Q and output (O)
3-state to Pad active (fast) TTSONF 5.6 5.2 4.9 4.6
3-state to Pad active (slew-limited) TTSONS 10.4 9.0 8.3 6.0
Internal GTS to Pad active TGTS 17.7 15.9 14.7 13.5 7
Note: 1. Timing is measured at pin threshold, with 50-pF external capacitance loads. Slew-limited output rise/fall times are
approximately two times longer than fast output rise/fall times.
2. Unused and unbonded IOBs are configured by default as inputs with internal pull-up resistors.
3. Timing is based upon the XC5215 device. For other devices, see Timing Calculator.

R
XC5200 Boundary Scan (JTAG) Switching Characteristic Guidelines

of nanoseconds and apply to all XC5200 devices unless otherwise noted.
Setup and Hold
Input (TDI) to clock (TCK) TTDITCK 30.0 30.0 30.0 30.0
setup time
Input (TDI) to clock (TCK) TTCKTDI 0 0 0 0
hold time
Input (TMS) to clock (TCK) TTMSTCK 15.0 15.0 15.0 15.0
setup time
Input (TMS) to clock (TCK) TTCKTMS 0 0 0 0
hold time
Propagation Delay
Clock (TCK) to Pad (TDO) TTCKPO 30.0 30.0 30.0 30.0
Clock
Clock (TCK) High TTCKH 30.0 30.0 30.0 30.0
Clock (TCK) Low TTCKL 30.0 30.0 30.0 30.0
FMAX (MHz) FMAX 10.0 10.0 10.0 10.0
Note 1: Input pad setup and hold times are specified with respect to the internal clock.

R

Device-specific tables include all packages for each XC5200-Series device. They follow the pad locations around the die,
and include boundary scan register locations.
Pin Locations for XC5202 Devices

The following table may contain pinout information for unsupported device/package combinations. Please see the
availability charts elsewhere in the XC5200 Series data sheet for availability information.
Pin Description VQ64* PC84 PQ100 VQ100 TQ144 PG156 Boundary Scan Order
VCC - 2 92 89 128 H3 -
1. I/O (A8) 57 3 93 90 129 H1 51
2. I/O (A9) 58 4 94 91 130 G1 54
3. I/O - - 95 92 131 G2 57
4. I/O - - 96 93 132 G3 63
5. I/O (A10) - 5 97 94 133 F1 66
6. I/O (A11) 59 6 98 95 134 F2 69
GND - - - - 137 F3 -
7. I/O (A12) 60 7 99 96 138 E3 78
8. I/O (A13) 61 8 100 97 139 C1 81
9. I/O (A14) 62 9 1 98 142 B1 90
10. I/O (A15) 63 10 2 99 143 B2 93
VCC 64 11 3 100 144 C3 -
GND - 12 4 1 1 C4 -
11. GCK1 (A16, I/O) 1 13 5 2 2 B3 102
12. I/O (A17) 2 14 6 3 3 A1 105 7
13. I/O (TDI) 3 15 7 4 6 B4 111
14. I/O (TCK) 4 16 8 5 7 A3 114
GND - - - - 8 C6 -
15. I/O (TMS) 5 17 9 6 11 A5 117
16. I/O 6 18 10 7 12 C7 123
17. I/O - - - - 13 B7 126
18. I/O - - 11 8 14 A6 129
19. I/O - 19 12 9 15 A7 135
20. I/O 7 20 13 10 16 A8 138
GND 8 21 14 11 17 C8 -
VCC 9 22 15 12 18 B8 -
21. I/O - 23 16 13 19 C9 141
22. I/O 10 24 17 14 20 B9 147
23. I/O - 18 15 21 A9 150
24. I/O - - - 22 B10 153
25. I/O - 25 19 16 23 C10 159
26. I/O 11 26 20 17 24 A10 162
GND - - - 27 C11 -
27. I/O 12 27 21 18 28 B12 165
28. I/O - 22 19 29 A13 171
29. I/O 13 28 23 20 32 B13 174
30. I/O 14 29 24 21 33 B14 177
31. M1 (I/O) 15 30 25 22 34 A15 186
GND - 31 26 23 35 C13 -
32. M0 (I/O) 16 32 27 24 36 A16 189
VCC - 33 28 25 37 C14 -
33. M2 (I/O) 17 34 29 26 38 B15 192
34. GCK2 (I/O) 18 35 30 27 39 B16 195

R
35. I/O (HDC) 19 36 31 28 40 D14 204
36. I/O - - 32 29 43 E14 207
37. I/O (LDC) 20 37 33 30 44 C16 210
GND - - - - 45 F14 -
38. I/O - 38 34 31 48 F16 216
39. I/O 21 39 35 32 49 G14 219
40. I/O - - 36 33 50 G15 222
41. I/O - - 37 34 51 G16 228
42. I/O 22 40 38 35 52 H16 231
43. I/O (ERR, INIT) 23 41 39 36 53 H15 234
VCC 24 42 40 37 54 H14 -
GND 25 43 41 38 55 J14 -
44. I/O 26 44 42 39 56 J15 240
45. I/O 27 45 43 40 57 J16 243
46. I/O - - 44 41 58 K16 246
47. I/O - - 45 42 59 K15 252
48. I/O 28 46 46 43 60 K14 255
49. I/O 29 47 47 44 61 L16 258
GND - - - - 64 L14 -
50. I/O - 48 48 45 65 P16 264
51. I/O 30 49 49 46 66 M14 267
52. I/O - 50 50 47 69 N14 276
53. I/O 31 51 51 48 70 R16 279
GND - 52 52 49 71 P14 -
DONE 32 53 53 50 72 R15 -
VCC 33 54 54 51 73 P13 -
PROG 34 55 55 52 74 R14 -
54. I/O (D7) 35 56 56 53 75 T16 288
55. GCK3 (I/O) 36 57 57 54 76 T15 291
56. I/O (D6) 37 58 58 55 79 T14 300
57. I/O - - 59 56 80 T13 303
GND - - - - 81 P11 -
58. I/O (D5) 38 59 60 57 84 T10 306
59. I/O (CS0) - 60 61 58 85 P10 312
60. I/O - - 62 59 86 R10 315
61. I/O - - 63 60 87 T9 318
62. I/O (D4) 39 61 64 61 88 R9 324
63. I/O - 62 65 62 89 P9 327
VCC 40 63 66 63 90 R8 -
GND 41 64 67 64 91 P8 -
64. I/O (D3) 42 65 68 65 92 T8 336
65. I/O (RS) 43 66 69 66 93 T7 339
66. I/O - - 70 67 94 T6 342
67. I/O - - - - 95 R7 348
68. I/O (D2) 44 67 71 68 96 P7 351
69. I/O - 68 72 69 97 T5 360
GND - - - - 100 P6 -
70. I/O (D1) 45 69 73 70 101 T3 363
71. I/O (RCLK-BUSY/ - 70 74 71 102 P5 366
RDY)
72. I/O (D0, DIN) 46 71 75 72 105 P4 372
73. I/O (DOUT) 47 72 76 73 106 T2 375

R
CCLK 48 73 77 74 107 R2 -
VCC - 74 78 75 108 P3 -
74. I/O (TDO) 49 75 79 76 109 T1 0
GND - 76 80 77 110 N3 -
75. I/O (A0, WS) 50 77 81 78 111 R1 9
76. GCK4 (A1, I/O) 51 78 82 79 112 P2 15
77. I/O (A2, CS1) 52 79 83 80 115 P1 18
78. I/O (A3) - 80 84 81 116 N1 21
GND - - - - 118 L3 -
79. I/O (A4) - 81 85 82 121 K3 27
80. I/O (A5) 53 82 86 83 122 K2 30
81. I/O - - 87 84 123 K1 33
82. I/O - - 88 85 124 J1 39
83. I/O (A6) 54 83 89 86 125 J2 42
84. I/O (A7) 55 84 90 87 126 J3 45
GND 56 1 91 88 127 H2 -
* VQ64 package supports Master Serial, Slave Serial, and Express configuration modes only.
Additional No Connect (N.C.) Connections on TQ144 Package

TQ144
135 9 41 67 98 117
136 10 42 68 99 119
140 25 46 77 103 120
7
141 26 47 78 104
4 30 62 82 113
5 31 63 83 114
Notes: Boundary Scan Bit 0 = TDO.T

Boundary Scan Bit 1 = TDO.O
Boundary Scan Bit 1056 = BSCAN.UPD

Pin Description PC84 PQ100 VQ100 TQ144 PG156 PQ160 Boundary Scan Order
VCC 2 92 89 128 H3 142 -
1. I/O (A8) 3 93 90 129 H1 143 78
2. I/O (A9) 4 94 91 130 G1 144 81
3. I/O - 95 92 131 G2 145 87
4. I/O - 96 93 132 G3 146 90
5. I/O (A10) 5 97 94 133 F1 147 93
6. I/O (A11) 6 98 95 134 F2 148 99
7. I/O - - - 135 E1 149 102
8. I/O - - - 136 E2 150 105
GND - - - 137 F3 151 -
9. I/O - - - - D1 152 111
10. I/O - - - - D2 153 114
11. I/O (A12) 7 99 96 138 E3 154 117
12. I/O (A13) 8 100 97 139 C1 155 123
13. I/O - - - 140 C2 156 126

R
14. I/O - - - 141 D3 157 129
15. I/O (A14) 9 1 98 142 B1 158 138
16. I/O (A15) 10 2 99 143 B2 159 141
VCC 11 3 100 144 C3 160 -
GND 12 4 1 1 C4 1 -
17. GCK1 (A16, I/O) 13 5 2 2 B3 2 150
18. I/O (A17) 14 6 3 3 A1 3 153
19. I/O - - - 4 A2 4 159
20. I/O - - - 5 C5 5 162
21. I/O (TDI) 15 7 4 6 B4 6 165
22. I/O (TCK) 16 8 5 7 A3 7 171
GND - - - 8 C6 10 -
23. I/O - - - 9 B5 11 174
24. I/O - - - 10 B6 12 177
25. I/O (TMS) 17 9 6 11 A5 13 180
26. I/O 18 10 7 12 C7 14 183
27. I/O - - - 13 B7 15 186
28. I/O - 11 8 14 A6 16 189
29. I/O 19 12 9 15 A7 17 195
30. I/O 20 13 10 16 A8 18 198
GND 21 14 11 17 C8 19 -
VCC 22 15 12 18 B8 20 -
31. I/O 23 16 13 19 C9 21 201
32. I/O 24 17 14 20 B9 22 207
33. I/O - 18 15 21 A9 23 210
34. I/O - - - 22 B10 24 213
35. I/O 25 19 16 23 C10 25 219
36. I/O 26 20 17 24 A10 26 222
37. I/O - - - 25 A11 27 225
38. I/O - - - 26 B11 28 231
GND - - - 27 C11 29 -
39. I/O 27 21 18 28 B12 32 234
40. I/O - 22 19 29 A13 33 237
41. I/O - - - 30 A14 34 240
42. I/O - - - 31 C12 35 243
43. I/O 28 23 20 32 B13 36 246
44. I/O 29 24 21 33 B14 37 249
45. M1 (I/O) 30 25 22 34 A15 38 258
GND 31 26 23 35 C13 39 -
46. M0 (I/O) 32 27 24 36 A16 40 261
VCC 33 28 25 37 C14 41 -
47. M2 (I/O) 34 29 26 38 B15 42 264
48. GCK2 (I/O) 35 30 27 39 B16 43 267
49. I/O (HDC) 36 31 28 40 D14 44 276
50. I/O - - - 41 C15 45 279
51. I/O - - - 42 D15 46 282
52. I/O - 32 29 43 E14 47 288
53. I/O (LDC) 37 33 30 44 C16 48 291
54. I/O - - - - E15 49 294
55. I/O - - - - D16 50 300
GND - - - 45 F14 51 -
56. I/O - - - 46 F15 52 303

R
57. I/O - - - 47 E16 53 306
58. I/O 38 34 31 48 F16 54 312
59. I/O 39 35 32 49 G14 55 315
60. I/O - 36 33 50 G15 56 318
61. I/O - 37 34 51 G16 57 324
62. I/O 40 38 35 52 H16 58 327
63. I/O (ERR, INIT) 41 39 36 53 H15 59 330
VCC 42 40 37 54 H14 60 -
GND 43 41 38 55 J14 61 -
64. I/O 44 42 39 56 J15 62 336
65. I/O 45 43 40 57 J16 63 339
66. I/O - 44 41 58 K16 64 348
67. I/O - 45 42 59 K15 65 351
68. I/O 46 46 43 60 K14 66 354
69. I/O 47 47 44 61 L16 67 360
70. I/O - - - 62 M16 68 363
71. I/O - - - 63 L15 69 366
GND - - - 64 L14 70 -
72. I/O - - - - N16 71 372
73. I/O - - - - M15 72 375
74. I/O 48 48 45 65 P16 73 378
75. I/O 49 49 46 66 M14 74 384
76. I/O - - - 67 N15 75 387
77. I/O - - - 68 P15 76 390
7
78. I/O 50 50 47 69 N14 77 396
79. I/O 51 51 48 70 R16 78 399
GND 52 52 49 71 P14 79 -
DONE 53 53 50 72 R15 80 -
VCC 54 54 51 73 P13 81 -
PROG 55 55 52 74 R14 82 -
80. I/O (D7) 56 56 53 75 T16 83 408
81. GCK3 (I/O) 57 57 54 76 T15 84 411
82. I/O - - - 77 R13 85 420
83. I/O - - - 78 P12 86 423
84. I/O (D6) 58 58 55 79 T14 87 426
85. I/O - 59 56 80 T13 88 432
GND - - - 81 P11 91 -
86. I/O - - - 82 R11 92 435
87. I/O - - - 83 T11 93 438
88. I/O (D5) 59 60 57 84 T10 94 444
89. I/O (CS0) 60 61 58 85 P10 95 447
90. I/O - 62 59 86 R10 96 450
91. I/O - 63 60 87 T9 97 456
92. I/O (D4) 61 64 61 88 R9 98 459
93. I/O 62 65 62 89 P9 99 462
VCC 63 66 63 90 R8 100 -
GND 64 67 64 91 P8 101 -
94. I/O (D3) 65 68 65 92 T8 102 468
95. I/O (RS) 66 69 66 93 T7 103 471
96. I/O - 70 67 94 T6 104 474
97. I/O - - - 95 R7 105 480
98. I/O (D2) 67 71 68 96 P7 106 483

R
99. I/O 68 72 69 97 T5 107 486
100. I/O - - - 98 R6 108 492
101. I/O - - - 99 T4 109 495
GND - - - 100 P6 110 -
102. I/O (D1) 69 73 70 101 T3 113 498
103. I/O (RCLK-BUSY/ 70 74 71 102 P5 114 504
RDY)
104. I/O - - - 103 R4 115 507
105. I/O - - - 104 R3 116 510
106. I/O (D0, DIN) 71 75 72 105 P4 117 516
107. I/O (DOUT) 72 76 73 106 T2 118 519
CCLK 73 77 74 107 R2 119 -
VCC 74 78 75 108 P3 120 -
108. I/O (TDO) 75 79 76 109 T1 121 0
GND 76 80 77 110 N3 122 -
109. I/O (A0, WS) 77 81 78 111 R1 123 9
110. GCK4 (A1, I/O) 78 82 79 112 P2 124 15
111. I/O - - - 113 N2 125 18
112. I/O - - - 114 M3 126 21
113. I/O (A2, CS1) 79 83 80 115 P1 127 27
114. I/O (A3) 80 84 81 116 N1 128 30
115. I/O - - - 117 M2 129 33
116. I/O - - - - M1 130 39
GND - - - 118 L3 131 -
117. I/O - - - 119 L2 132 42
118. I/O - - - 120 L1 133 45
119. I/O (A4) 81 85 82 121 K3 134 51
120. I/O (A5) 82 86 83 122 K2 135 54
121. I/O - 87 84 123 K1 137 57
122. I/O - 88 85 124 J1 138 63
123. I/O (A6) 83 89 86 125 J2 139 66
124. I/O (A7) 84 90 87 126 J3 140 69
GND 1 91 88 127 H2 141 -
Additional No Connect (N.C.) Connections for PQ160 Package

PQ160
8 30 89 111 136
9 31 90 112


R

Pin Description PC84 PQ100 VQ100 TQ144 PQ160 TQ176 PG191 PQ208 Boundary Scan Order
VCC 2 92 89 128 142 155 J4 183 -
1. I/O (A8) 3 93 90 129 143 156 J3 184 87
2. I/O (A9) 4 94 91 130 144 157 J2 185 90
3. I/O - 95 92 131 145 158 J1 186 93
4. I/O - 96 93 132 146 159 H1 187 99
5. I/O - - - - - 160 H2 188 102
6. I/O - - - - - 161 H3 189 105
7. I/O (A10) 5 97 94 133 147 162 G1 190 111
8. I/O (A11) 6 98 95 134 148 163 G2 191 114
9. I/O - - - 135 149 164 F1 192 117
10. I/O - - - 136 150 165 E1 193 123
GND - - - 137 151 166 G3 194 -
11. I/O - - - - 152 168 C1 197 126
12. I/O - - - - 153 169 E2 198 129
13. I/O (A12) 7 99 96 138 154 170 F3 199 138
14. I/O (A13) 8 100 97 139 155 171 D2 200 141
15. I/O - - - 140 156 172 B1 201 150
16. I/O - - - 141 157 173 E3 202 153
17. I/O (A14) 9 1 98 142 158 174 C2 203 162
18. I/O (A15) 10 2 99 143 159 175 B2 204 165
VCC 11 3 100 144 160 176 D3 205 -
7
GND 12 4 1 1 1 1 D4 2 -
19. GCK1 (A16, I/O) 13 5 2 2 2 2 C3 4 174
20. I/O (A17) 14 6 3 3 3 3 C4 5 177
21. I/O - - - 4 4 4 B3 6 183
22. I/O - - - 5 5 5 C5 7 186
23. I/O (TDI) 15 7 4 6 6 6 A2 8 189
24. I/O (TCK) 16 8 5 7 7 7 B4 9 195
25. I/O - - - - 8 8 C6 10 198
26. I/O - - - - 9 9 A3 11 201
GND - - - 8 10 10 C7 14 -
27. I/O - - - 9 11 11 A4 15 207
28. I/O - - - 10 12 12 A5 16 210
29. I/O (TMS) 17 9 6 11 13 13 B7 17 213
30. I/O 18 10 7 12 14 14 A6 18 219
31. I/O - - - - - 15 C8 19 222
32. I/O - - - - - 16 A7 20 225
33. I/O - - - 13 15 17 B8 21 234
34. I/O - 11 8 14 16 18 A8 22 237
35. I/O 19 12 9 15 17 19 B9 23 246
36. I/O 20 13 10 16 18 20 C9 24 249
GND 21 14 11 17 19 21 D9 25 -
VCC 22 15 12 18 20 22 D10 26 -
37. I/O 23 16 13 19 21 23 C10 27 255
38. I/O 24 17 14 20 22 24 B10 28 258
39. I/O - 18 15 21 23 25 A9 29 261
40. I/O - - - 22 24 26 A10 30 267
41. I/O - - - - - 27 A11 31 270

R
42. I/O - - - - - 28 C11 32 273
43. I/O 25 19 16 23 25 29 B11 33 279
44. I/O 26 20 17 24 26 30 A12 34 282
45. I/O - - - 25 27 31 B12 35 285
46. I/O - - - 26 28 32 A13 36 291
GND - - - 27 29 33 C12 37 -
47. I/O - - - - 30 34 A15 40 294
48. I/O - - - - 31 35 C13 41 297
49. I/O 27 21 18 28 32 36 B14 42 303
50. I/O - 22 19 29 33 37 A16 43 306
51. I/O - - - 30 34 38 B15 44 309
52. I/O - - - 31 35 39 C14 45 315
53. I/O 28 23 20 32 36 40 A17 46 318
54. I/O 29 24 21 33 37 41 B16 47 321
55. M1 (I/O) 30 25 22 34 38 42 C15 48 330
GND 31 26 23 35 39 43 D15 49 -
56. M0 (I/O) 32 27 24 36 40 44 A18 50 333
VCC 33 28 25 37 41 45 D16 55 -
57. M2 (I/O) 34 29 26 38 42 46 C16 56 336
58. GCK2 (I/O) 35 30 27 39 43 47 B17 57 339
59. I/O (HDC) 36 31 28 40 44 48 E16 58 348
60. I/O - - - 41 45 49 C17 59 351
61. I/O - - - 42 46 50 D17 60 354
62. I/O - 32 29 43 47 51 B18 61 360
63. I/O (LDC) 37 33 30 44 48 52 E17 62 363
64. I/O - - - - 49 53 F16 63 372
65. I/O - - - - 50 54 C18 64 375
GND - - - 45 51 55 G16 67 -
66. I/O - - - 46 52 56 E18 68 378
67. I/O - - - 47 53 57 F18 69 384
68. I/O 38 34 31 48 54 58 G17 70 387
69. I/O 39 35 32 49 55 59 G18 71 390
70. I/O - - - - - 60 H16 72 396
71. I/O - - - - - 61 H17 73 399
72. I/O - 36 33 50 56 62 H18 74 402
73. I/O - 37 34 51 57 63 J18 75 408
74. I/O 40 38 35 52 58 64 J17 76 411
75. I/O (ERR, INIT) 41 39 36 53 59 65 J16 77 414
VCC 42 40 37 54 60 66 J15 78 -
GND 43 41 38 55 61 67 K15 79 -
76. I/O 44 42 39 56 62 68 K16 80 420
77. I/O 45 43 40 57 63 69 K17 81 423
78. I/O - 44 41 58 64 70 K18 82 426
79. I/O - 45 42 59 65 71 L18 83 432
80. I/O - - - - - 72 L17 84 435
81. I/O - - - - - 73 L16 85 438
82. I/O 46 46 43 60 66 74 M18 86 444
83. I/O 47 47 44 61 67 75 M17 87 447
84. I/O - - - 62 68 76 N18 88 450
85. I/O - - - 63 69 77 P18 89 456
GND - - - 64 70 78 M16 90 -
86. I/O - - - - 71 79 T18 93 459

R
87. I/O - - - - 72 80 P17 94 468
88. I/O 48 48 45 65 73 81 N16 95 471
89. I/O 49 49 46 66 74 82 T17 96 480
90. I/O - - - 67 75 83 R17 97 483
91. I/O - - - 68 76 84 P16 98 486
92. I/O 50 50 47 69 77 85 U18 99 492
93. I/O 51 51 48 70 78 86 T16 100 495
GND 52 52 49 71 79 87 R16 101 -
DONE 53 53 50 72 80 88 U17 103 -
VCC 54 54 51 73 81 89 R15 106 -
PROG 55 55 52 74 82 90 V18 108 -
94. I/O (D7) 56 56 53 75 83 91 T15 109 504
95. GCK3 (I/O) 57 57 54 76 84 92 U16 110 507
96. I/O - - - 77 85 93 T14 111 516
97. I/O - - - 78 86 94 U15 112 519
98. I/O (D6) 58 58 55 79 87 95 V17 113 522
99. I/O - 59 56 80 88 96 V16 114 528
100. I/O - - - - 89 97 T13 115 531
101. I/O - - - - 90 98 U14 116 534
GND - - - 81 91 99 T12 119 -
102. I/O - - - 82 92 100 U13 120 540
103. I/O - - - 83 93 101 V13 121 543
104. I/O (D5) 59 60 57 84 94 102 U12 122 552
105. I/O (CS0) 60 61 58 85 95 103 V12 123 555
7
106. I/O - - - - - 104 T11 124 558
107. I/O - - - - - 105 U11 125 564
108. I/O - 62 59 86 96 106 V11 126 567
109. I/O - 63 60 87 97 107 V10 127 570
110. I/O (D4) 61 64 61 88 98 108 U10 128 576
111. I/O 62 65 62 89 99 109 T10 129 579
VCC 63 66 63 90 100 110 R10 130 -
GND 64 67 64 91 101 111 R9 131 -
112. I/O (D3) 65 68 65 92 102 112 T9 132 588
113. I/O (RS) 66 69 66 93 103 113 U9 133 591
114. I/O - 70 67 94 104 114 V9 134 600
115. I/O - - - 95 105 115 V8 135 603
116. I/O - - - - - 116 U8 136 612
117. I/O - - - - - 117 T8 137 615
118. I/O (D2) 67 71 68 96 106 118 V7 138 618
119. I/O 68 72 69 97 107 119 U7 139 624
120. I/O - - - 98 108 120 V6 140 627
121. I/O - - - 99 109 121 U6 141 630
GND - - - 100 110 122 T7 142 -
122. I/O - - - - 111 123 U5 145 636
123. I/O - - - - 112 124 T6 146 639
124. I/O (D1) 69 73 70 101 113 125 V3 147 642
125. I/O (RCLK- 70 74 71 102 114 126 V2 148 648
BUSY/RDY)
126. I/O - - - 103 115 127 U4 149 651
127. I/O - - - 104 116 128 T5 150 654
128. I/O (D0, DIN) 71 75 72 105 117 129 U3 151 660
129. I/O (DOUT) 72 76 73 106 118 130 T4 152 663

R
CCLK 73 77 74 107 119 131 V1 153 -
VCC 74 78 75 108 120 132 R4 154 -
130. I/O (TDO) 75 79 76 109 121 133 U2 159 -
GND 76 80 77 110 122 134 R3 160 -
131. I/O (A0, WS) 77 81 78 111 123 135 T3 161 9
132. GCK4 (A1, I/O) 78 82 79 112 124 136 U1 162 15
133. I/O - - - 113 125 137 P3 163 18
134. I/O - - - 114 126 138 R2 164 21
135. I/O (A2, CS1) 79 83 80 115 127 139 T2 165 27
136. I/O (A3) 80 84 81 116 128 140 N3 166 30
137. I/O - - - 117 129 141 P2 167 33
138. I/O - - - - 130 142 T1 168 42
GND - - - 118 131 143 M3 171 -
139. I/O - - - 119 132 144 P1 172 45
140. I/O - - - 120 133 145 N1 173 51
141. I/O (A4) 81 85 82 121 134 146 M2 174 54
142. I/O (A5) 82 86 83 122 135 147 M1 175 57
143. I/O - - - - - 148 L3 176 63
144. I/O - - - - 136 149 L2 177 66
145. I/O - 87 84 123 137 150 L1 178 69
146. I/O - 88 85 124 138 151 K1 179 75
147. I/O (A6) 83 89 86 125 139 152 K2 180 78
148. I/O (A7) 84 90 87 126 140 153 K3 181 81
GND 1 91 88 127 141 154 K4 182 -
Additional No Connect (N.C.) Connections for PQ208 and TQ176 Packages

PQ208 TQ176
195 1 39 65 104 143 158 167
196 3 51 66 105 144 169
206 12 52 91 107 155 170
207 13 53 92 117 156
208 38 54 102 118 157


Boundary Scan
Pin Description PC84 TQ144 PQ160 TQ176 PQ208 PG223 BG225 PQ240
Order
VCC 2 128 142 155 183 J4 VCC* 212 -
1. I/O (A8) 3 129 143 156 184 J3 E8 213 111
2. I/O (A9) 4 130 144 157 185 J2 B7 214 114
3. I/O - 131 145 158 186 J1 A7 215 117
4. I/O - 132 146 159 187 H1 C7 216 123
5. I/O - - - 160 188 H2 D7 217 126
6. I/O - - - 161 189 H3 E7 218 129

R
Boundary Scan
Order
7. I/O (A10) 5 133 147 162 190 G1 A6 220 135
8. I/O (A11) 6 134 148 163 191 G2 B6 221 138
VCC - - - - - - VCC* 222 -
9. I/O - - - - - H4 C6 223 141
10. I/O - - - - - G4 F7 224 150
11. I/O - 135 149 164 192 F1 A5 225 153
12. I/O - 136 150 165 193 E1 B5 226 162
GND - 137 151 166 194 G3 GND* 227 -
13. I/O - - - - 195 F2 D6 228 165
14. I/O - - - 167 196 D1 C5 229 171
15. I/O - - 152 168 197 C1 A4 230 174
16. I/O - - 153 169 198 E2 E6 231 177
17. I/O (A12) 7 138 154 170 199 F3 B4 232 183
18. I/O (A13) 8 139 155 171 200 D2 D5 233 186
19. I/O - - - - - F4 A3 234 189
20. I/O - - - - - E4 C4 235 195
21. I/O - 140 156 172 201 B1 B3 236 198
22. I/O - 141 157 173 202 E3 F6 237 201
23. I/O (A14) 9 142 158 174 203 C2 A2 238 210
24. I/O (A15) 10 143 159 175 204 B2 C3 239 213
VCC 11 144 160 176 205 D3 VCC* 240 -
GND 12 1 1 1 2 D4 GND* 1 -
25. GCK1 (A16, I/O) 13 2 2 2 4 C3 D4 2 222
26. I/O (A17) 14 3 3 3 5 C4 B1 3 225
7
27. I/O - 4 4 4 6 B3 C2 4 231
28. I/O - 5 5 5 7 C5 E5 5 234
29. I/O (TDI) 15 6 6 6 8 A2 D3 6 237
30. I/O (TCK) 16 7 7 7 9 B4 C1 7 243
31. I/O - - 8 8 10 C6 D2 8 246
32. I/O - - 9 9 11 A3 G6 9 249
33. I/O - - - - 12 B5 E4 10 255
34. I/O - - - - 13 B6 D1 11 258
35. I/O - - - - - D5 E3 12 261
36. I/O - - - - - D6 E2 13 267
GND - 8 10 10 14 C7 GND* 14 -
37. I/O - 9 11 11 15 A4 F5 15 270
38. I/O - 10 12 12 16 A5 E1 16 273
39. I/O (TMS) 17 11 13 13 17 B7 F4 17 279
40. I/O 18 12 14 14 18 A6 F3 18 282
VCC - - - - - - VCC* 19 -
41. I/O - - - - - D7 F2 20 285
42. I/O - - - - - D8 F1 21 291
43. I/O - - - 15 19 C8 G4 23 294
44. I/O - - - 16 20 A7 G3 24 297
45. I/O - 13 15 17 21 B8 G2 25 306
46. I/O - 14 16 18 22 A8 G1 26 309
47. I/O 19 15 17 19 23 B9 G5 27 318
48. I/O 20 16 18 20 24 C9 H3 28 321
GND 21 17 19 21 25 D9 GND* 29 -
VCC 22 18 20 22 26 D10 VCC* 30 -
49. I/O 23 19 21 23 27 C10 H4 31 327

R
Boundary Scan
Order
50. I/O 24 20 22 24 28 B10 H5 32 330
51. I/O - 21 23 25 29 A9 J2 33 333
52. I/O - 22 24 26 30 A10 J1 34 339
53. I/O - - - 27 31 A11 J3 35 342
54. I/O - - - 28 32 C11 J4 36 345
55. I/O - - - - - D11 J5 38 351
56. I/O - - - - - D12 K1 39 354
VCC - - - - - - VCC* 40 -
57. I/O 25 23 25 29 33 B11 K2 41 357
58. I/O 26 24 26 30 34 A12 K3 42 363
59. I/O - 25 27 31 35 B12 J6 43 366
60. I/O - 26 28 32 36 A13 L1 44 369
GND - 27 29 33 37 C12 GND* 45 -
61. I/O - - - - - D13 L2 46 375
62. I/O - - - - - D14 K4 47 378
63. I/O - - - - 38 B13 L3 48 381
64. I/O - - - - 39 A14 M1 49 387
65. I/O - - 30 34 40 A15 K5 50 390
66. I/O - - 31 35 41 C13 M2 51 393
67. I/O 27 28 32 36 42 B14 L4 52 399
68. I/O - 29 33 37 43 A16 N1 53 402
69. I/O - 30 34 38 44 B15 M3 54 405
70. I/O - 31 35 39 45 C14 N2 55 411
71. I/O 28 32 36 40 46 A17 K6 56 414
72. I/O 29 33 37 41 47 B16 P1 57 417
73. M1 (I/O) 30 34 38 42 48 C15 N3 58 426
GND 31 35 39 43 49 D15 GND* 59 -
74. M0 (I/O) 32 36 40 44 50 A18 P2 60 429
VCC 33 37 41 45 55 D16 VCC* 61 -
75. M2 (I/O) 34 38 42 46 56 C16 M4 62 432
76. GCK2 (I/O) 35 39 43 47 57 B17 R2 63 435
77. I/O (HDC) 36 40 44 48 58 E16 P3 64 444
78. I/O - 41 45 49 59 C17 L5 65 447
79. I/O - 42 46 50 60 D17 N4 66 450
80. I/O - 43 47 51 61 B18 R3 67 456
81. I/O (LDC) 37 44 48 52 62 E17 P4 68 459
82. I/O - - 49 53 63 F16 K7 69 462
83. I/O - - 50 54 64 C18 M5 70 468
84. I/O - - - - 65 D18 R4 71 471
85. I/O - - - - 66 F17 N5 72 474
86. I/O - - - - - E15 P5 73 480
87. I/O - - - - - F15 L6 74 483
GND - 45 51 55 67 G16 GND* 75 -
88. I/O - 46 52 56 68 E18 R5 76 486
89. I/O - 47 53 57 69 F18 M6 77 492
90. I/O 38 48 54 58 70 G17 N6 78 495
91. I/O 39 49 55 59 71 G18 P6 79 504
VCC - - - - - - VCC* 80 -
92. I/O - - - 60 72 H16 R6 81 507
93. I/O - - - 61 73 H17 M7 82 510
94. I/O - - - - - G15 N7 84 516

R
Boundary Scan
Order
95. I/O - - - - - H15 P7 85 519
96. I/O - 50 56 62 74 H18 R7 86 522
97. I/O - 51 57 63 75 J18 L7 87 528
98. I/O 40 52 58 64 76 J17 N8 88 531
99. I/O (ERR, INIT) 41 53 59 65 77 J16 P8 89 534
VCC 42 54 60 66 78 J15 VCC* 90 -
GND 43 55 61 67 79 K15 GND* 91 -
100. I/O 44 56 62 68 80 K16 L8 92 540
101. I/O 45 57 63 69 81 K17 P9 93 543
102. I/O - 58 64 70 82 K18 R9 94 546
103. I/O - 59 65 71 83 L18 N9 95 552
104. I/O - - - 72 84 L17 M9 96 555
105. I/O - - - 73 85 L16 L9 97 558
106. I/O - - - - - L15 R10 99 564
107. I/O - - - - - M15 P10 100 567
VCC - - - - - - VCC* 101 -
108. I/O 46 60 66 74 86 M18 N10 102 570
109. I/O 47 61 67 75 87 M17 K9 103 576
110. I/O - 62 68 76 88 N18 R11 104 579
111. I/O - 63 69 77 89 P18 P11 105 588
GND - 64 70 78 90 M16 GND* 106 -
112. I/O - - - - - N15 M10 107 591
113. I/O - - - - - P15 N11 108 600
114. I/O - - - - 91 N17 R12 109 603
7
115. I/O - - - - 92 R18 L10 110 606
116. I/O - - 71 79 93 T18 P12 111 612
117. I/O - - 72 80 94 P17 M11 112 615
118. I/O 48 65 73 81 95 N16 R13 113 618
119. I/O 49 66 74 82 96 T17 N12 114 624
120. I/O - 67 75 83 97 R17 P13 115 627
121. I/O - 68 76 84 98 P16 K10 116 630
122. I/O 50 69 77 85 99 U18 R14 117 636
123. I/O 51 70 78 86 100 T16 N13 118 639
GND 52 71 79 87 101 R16 GND* 119 -
DONE 53 72 80 88 103 U17 P14 120 -
VCC 54 73 81 89 106 R15 VCC* 121 -
PROG 55 74 82 90 108 V18 M12 122 -
124. I/O (D7) 56 75 83 91 109 T15 P15 123 648
125. GCK3 (I/O) 57 76 84 92 110 U16 N14 124 651
126. I/O - 77 85 93 111 T14 L11 125 660
127. I/O - 78 86 94 112 U15 M13 126 663
128. I/O - - - - - R14 N15 127 666
129. I/O - - - - - R13 M14 128 672
130. I/O (D6) 58 79 87 95 113 V17 J10 129 675
131. I/O - 80 88 96 114 V16 L12 130 678
132. I/O - - 89 97 115 T13 M15 131 684
133. I/O - - 90 98 116 U14 L13 132 687
134. I/O - - - - 117 V15 L14 133 690
135. I/O - - - - 118 V14 K11 134 696
GND - 81 91 99 119 T12 GND* 135 -
136. I/O - - - - - R12 L15 136 699

R
Boundary Scan
Order
137. I/O - - - - - R11 K12 137 708
138. I/O - 82 92 100 120 U13 K13 138 711
139. I/O - 83 93 101 121 V13 K14 139 714
VCC - - - - - - VCC* 140 -
140. I/O (D5) 59 84 94 102 122 U12 K15 141 720
141. I/O (CS0) 60 85 95 103 123 V12 J12 142 723
142. I/O - - - 104 124 T11 J13 144 726
143. I/O - - - 105 125 U11 J14 145 732
144. I/O - 86 96 106 126 V11 J15 146 735
145. I/O - 87 97 107 127 V10 J11 147 738
146. I/O (D4) 61 88 98 108 128 U10 H13 148 744
147. I/O 62 89 99 109 129 T10 H14 149 747
VCC 63 90 100 110 130 R10 VCC* 150 -
GND 64 91 101 111 131 R9 GND* 151 -
148. I/O (D3) 65 92 102 112 132 T9 H12 152 756
149. I/O (RS) 66 93 103 113 133 U9 H11 153 759
150. I/O - 94 104 114 134 V9 G14 154 768
151. I/O - 95 105 115 135 V8 G15 155 771
152. I/O - - - 116 136 U8 G13 156 780
153. I/O - - - 117 137 T8 G12 157 783
154. I/O (D2) 67 96 106 118 138 V7 G11 159 786
155. I/O 68 97 107 119 139 U7 F15 160 792
VCC - - - - - - VCC* 161 -
156. I/O - 98 108 120 140 V6 F14 162 795
157. I/O - 99 109 121 141 U6 F13 163 798
158. I/O - - - - - R8 G10 164 804
159. I/O - - - - - R7 E15 165 807
GND - 100 110 122 142 T7 GND* 166 -
160. I/O - - - - - R6 E14 167 810
161. I/O - - - - - R5 F12 168 816
162. I/O - - - - 143 V5 E13 169 819
163. I/O - - - - 144 V4 D15 170 822
164. I/O - - 111 123 145 U5 F11 171 828
165. I/O - - 112 124 146 T6 D14 172 831
166. I/O (D1) 69 101 113 125 147 V3 E12 173 834
167. I/O (RCLK-BUSY/RDY) 70 102 114 126 148 V2 C15 174 840
168. I/O - 103 115 127 149 U4 D13 175 843
169. I/O - 104 116 128 150 T5 C14 176 846
170. I/O (D0, DIN) 71 105 117 129 151 U3 F10 177 855
171. I/O (DOUT) 72 106 118 130 152 T4 B15 178 858
CCLK 73 107 119 131 153 V1 C13 179 -
VCC 74 108 120 132 154 R4 VCC* 180 -
172. I/O (TDO) 75 109 121 133 159 U2 A15 181 -
GND 76 110 122 134 160 R3 GND* 182 -
173. I/O (A0, WS) 77 111 123 135 161 T3 A14 183 9
174. GCK4 (A1, I/O) 78 112 124 136 162 U1 B13 184 15
175. I/O - 113 125 137 163 P3 E11 185 18
176. I/O - 114 126 138 164 R2 C12 186 21
177. I/O (CS1, A2) 79 115 127 139 165 T2 A13 187 27
178. I/O (A3) 80 116 128 140 166 N3 B12 188 30
179. I/O - - - - - P4 F9 189 33

R
Boundary Scan
Order
180. I/O - - - - - N4 D11 190 39
181. I/O - 117 129 141 167 P2 A12 191 42
182. I/O - - 130 142 168 T1 C11 192 45
183. I/O - - - - 169 R1 B11 193 51
184. I/O - - - - 170 N2 E10 194 54
- - - - - - - GND* -
GND - 118 131 143 171 M3 - 196 -
185. I/O - 119 132 144 172 P1 A11 197 57
186. I/O - 120 133 145 173 N1 D10 198 66
187. I/O - - - - - M4 C10 199 69
188. I/O - - - - - L4 B10 200 75
VCC - - - - - - VCC* 201 -
189. I/O (A4) 81 121 134 146 174 M2 A10 202 78
190. I/O (A5) 82 122 135 147 175 M1 D9 203 81
191. I/O - - - 148 176 L3 C9 205 87
192. I/O - - 136 149 177 L2 B9 206 90
193. I/O - 123 137 150 178 L1 A9 207 93
194. I/O - 124 138 151 179 K1 E9 208 99
195. I/O (A6) 83 125 139 152 180 K2 C8 209 102
196. I/O (A7) 84 126 140 153 181 K3 B8 210 105
GND 1 127 141 154 182 K4 GND* 211 -
7
Additional No Connect (N.C.) Connections for PQ208 and PQ240 Packages
PQ208 PQ240
1 53 105 157 208 22 143 219
3 54 107 158 37 158
51 102 155 206 83 195
52 104 156 207 98 204
Notes: * Pins labeled VCC* are internally bonded to a VCC plane within the BG225 package. The external pins are: B2, D8, H15, R8,
B14, R1, H1, and R15.
Pins labeled GND* are internally bonded to a ground plane within the BG225 package. The external pins are: A1, D12, G7,
G9, H6, H8, H10, J8, K8, A8, F8, G8, H2, H7, H9, J7, J9, M8.
Boundary Scan Bit 0 = TDO.T

Pin Description PQ160 HQ208 HQ240 PG299 BG225 BG352 Boundary Scan Order
VCC 142 183 212 K1 VCC* VCC* -
1. I/O (A8) 143 184 213 K2 E8 D14 138
2. I/O (A9) 144 185 214 K3 B7 C14 141
3. I/O 145 186 215 K5 A7 A15 147
4. I/O 146 187 216 K4 C7 B15 150
5. I/O - 188 217 J1 D7 C15 153
6. I/O - 189 218 J2 E7 D15 159
7. I/O (A10) 147 190 220 H1 A6 A16 162

R
8. I/O (A11) 148 191 221 J3 B6 B16 165
9. I/O - - - H2 - C17 171
10. I/O - - - G1 - B18 174
VCC - - 222 E1 VCC* VCC* -
11. I/O - - 223 H3 C6 C18 177
12. I/O - - 224 G2 F7 D17 183
13. I/O 149 192 225 H4 A5 A20 186
14. I/O 150 193 226 F2 B5 B19 189
GND 151 194 227 F1 GND* GND* -
15. I/O - - - H5 - C19 195
16. I/O - - - G3 - D18 198
17. I/O - 195 228 D1 D6 A21 201
18. I/O - 196 229 G4 C5 B20 207
19. I/O 152 197 230 E2 A4 C20 210
20. I/O 153 198 231 F3 E6 B21 213
21. I/O (A12) 154 199 232 G5 B4 B22 219
22. I/O (A13) 155 200 233 C1 D5 C21 222
23. I/O - - - F4 - D20 225
24. I/O - - - E3 - A23 234
25. I/O - - 234 D2 A3 D21 237
26. I/O - - 235 C2 C4 C22 243
27. I/O 156 201 236 F5 B3 B24 246
28. I/O 157 202 237 E4 F6 C23 249
29. I/O (A14) 158 203 238 D3 A2 D22 258
30. I/O (A15) 159 204 239 C3 C3 C24 261
VCC 160 205 240 A2 VCC* VCC* -
GND 1 2 1 B1 GND* GND* -
31. GCK1 (A16, I/O) 2 4 2 D4 D4 D23 270
32. I/O (A17) 3 5 3 B2 B1 C25 273
33. I/O 4 6 4 B3 C2 D24 279
34. I/O 5 7 5 E6 E5 E23 282
35. I/O (TDI) 6 8 6 D5 D3 C26 285
36. I/O (TCK) 7 9 7 C4 C1 E24 294
37. I/O - - - A3 - F24 297
38. I/O - - - D6 - E25 303
39. I/O 8 10 8 E7 D2 D26 306
40. I/O 9 11 9 B4 G6 G24 309
41. I/O - 12 10 C5 E4 F25 315
42. I/O - 13 11 A4 D1 F26 318
43. I/O - - 12 D7 E3 H23 321
44. I/O - - 13 C6 E2 H24 327
45. I/O - - - E8 - G25 330
46. I/O - - - B5 - G26 333
GND 10 14 14 A5 GND* GND* -
47. I/O 11 15 15 B6 F5 J23 339
48. I/O 12 16 16 D8 E1 J24 342
49. I/O (TMS) 13 17 17 C7 F4 H25 345
50. I/O 14 18 18 B7 F3 K23 351
VCC - - 19 A6 VCC* VCC* -
51. I/O - - 20 C8 F2 L24 354
52. I/O - - 21 E9 F1 K25 357
53. I/O - - - B8 - L25 363

R
54. I/O - - - A8 - L26 366
55. I/O - 19 23 C9 G4 M23 369
56. I/O - 20 24 B9 G3 M24 375
57. I/O 15 21 25 E10 G2 M25 378
58. I/O 16 22 26 A9 G1 M26 381
59. I/O 17 23 27 D10 G5 N24 390
60. I/O 18 24 28 C10 H3 N25 393
GND 19 25 29 A10 GND* GND* -
VCC 20 26 30 A11 VCC* VCC* -
61. I/O 21 27 31 B10 H4 N26 399
62. I/O 22 28 32 B11 H5 P25 402
63. I/O 23 29 33 C11 J2 P23 405
64. I/O 24 30 34 E11 J1 P24 411
65. I/O - 31 35 D11 J3 R26 414
66. I/O - 32 36 A12 J4 R25 417
67. I/O - - - B12 - R24 423
68. I/O - - - A13 - R23 426
69. I/O - - 38 E12 J5 T26 429
70. I/O - - 39 B13 K1 T25 435
VCC - - 40 A16 VCC* VCC* -
71. I/O 25 33 41 A14 K2 U24 438
72. I/O 26 34 42 C13 K3 V25 441
73. I/O 27 35 43 B14 J6 V24 447
74. I/O 28 36 44 D13 L1 U23 450
7
GND 29 37 45 A15 GND* GND* -
75. I/O - - - B15 - Y26 453
76. I/O - - - E13 - W25 459
77. I/O - - 46 C14 L2 W24 462
78. I/O - - 47 A17 K4 V23 465
79. I/O - 38 48 D14 L3 AA26 471
80. I/O - 39 49 B16 M1 Y25 474
81. I/O 30 40 50 C15 K5 Y24 477
82. I/O 31 41 51 E14 M2 AA25 483
83. I/O - - - A18 - AB25 486
84. I/O - - - D15 - AA24 489
85. I/O 32 42 52 C16 L4 Y23 495
86. I/O 33 43 53 B17 N1 AC26 498
87. I/O 34 44 54 B18 M3 AA23 501
88. I/O 35 45 55 E15 N2 AB24 507
89. I/O 36 46 56 D16 K6 AD25 510
90. I/O 37 47 57 C17 P1 AC24 513
91. M1 (I/O) 38 48 58 A20 N3 AB23 522
GND 39 49 59 A19 GND* GND* -
92. M0 (I/O) 40 50 60 C18 P2 AD24 525
VCC 41 55 61 B20 VCC* VCC* -
93. M2 (I/O) 42 56 62 D17 M4 AC23 528
94. GCK2 (I/O) 43 57 63 B19 R2 AE24 531
95. I/O (HDC) 44 58 64 C19 P3 AD23 540
96. I/O 45 59 65 F16 L5 AC22 543
97. I/O 46 60 66 E17 N4 AF24 546
98. I/O 47 61 67 D18 R3 AD22 552
99. I/O (LDC) 48 62 68 C20 P4 AE23 555

R
100. I/O - - - F17 - AE22 558
101. I/O - - - G16 - AF23 564
102. I/O 49 63 69 D19 K7 AD20 567
103. I/O 50 64 70 E18 M5 AE21 570
104. I/O - 65 71 D20 R4 AF21 576
105. I/O - 66 72 G17 N5 AC19 579
106. I/O - - 73 F18 P5 AD19 582
107. I/O - - 74 H16 L6 AE20 588
108. I/O - - - E19 - AF20 591
109. I/O - - - F19 - AC18 594
GND 51 67 75 E20 GND* GND* -
110. I/O 52 68 76 H17 R5 AD18 600
111. I/O 53 69 77 G18 M6 AE19 603
112. I/O 54 70 78 G19 N6 AC17 606
113. I/O 55 71 79 H18 P6 AD17 612
VCC - - 80 F20 VCC* VCC* -
114. I/O - 72 81 J16 R6 AE17 615
115. I/O - 73 82 G20 M7 AE16 618
116. I/O - - - H20 - AF16 624
117. I/O - - - J18 - AC15 627
118. I/O - - 84 J19 N7 AD15 630
119. I/O - - 85 K16 P7 AE15 636
120. I/O 56 74 86 J20 R7 AF15 639
121. I/O 57 75 87 K17 L7 AD14 642
122. I/O 58 76 88 K18 N8 AE14 648
123. I/O (ERR, INIT) 59 77 89 K19 P8 AF14 651
VCC 60 78 90 L20 VCC* VCC* -
GND 61 79 91 K20 GND* GND* -
124. I/O 62 80 92 L19 L8 AE13 660
125. I/O 63 81 93 L18 P9 AC13 663
126. I/O 64 82 94 L16 R9 AD13 672
127. I/O 65 83 95 L17 N9 AF12 675
128. I/O - 84 96 M20 M9 AE12 678
129. I/O - 85 97 M19 L9 AD12 684
130. I/O - - - N20 - AC12 687
131. I/O - - - M18 - AF11 690
132. I/O - - 99 N19 R10 AE11 696
133. I/O - - 100 P20 P10 AD11 699
VCC - - 101 T20 VCC* VCC* -
134. I/O 66 86 102 N18 N10 AE9 702
135. I/O 67 87 103 P19 K9 AD9 708
136. I/O 68 88 104 N17 R11 AC10 711
137. I/O 69 89 105 R19 P11 AF7 714
GND 70 90 106 R20 GND* GND* -
138. I/O - - - N16 - AE8 720
139. I/O - - - P18 - AD8 723
140. I/O - - 107 U20 M10 AC9 726
141. I/O - - 108 P17 N11 AF6 732
142. I/O - 91 109 T19 R12 AE7 735
143. I/O - 92 110 R18 L10 AD7 738
144. I/O 71 93 111 P16 P12 AE6 744
145. I/O 72 94 112 V20 M11 AE5 747

R
146. I/O - - - R17 - AD6 750
147. I/O - - - T18 - AC7 756
148. I/O 73 95 113 U19 R13 AF4 759
149. I/O 74 96 114 V19 N12 AF3 768
150. I/O 75 97 115 R16 P13 AD5 771
151. I/O 76 98 116 T17 K10 AE3 774
152. I/O 77 99 117 U18 R14 AD4 780
153. I/O 78 100 118 X20 N13 AC5 783
GND 79 101 119 W20 GND* GND* -
DONE 80 103 120 V18 P14 AD3 -
VCC 81 106 121 X19 VCC* VCC* -
PROG 82 108 122 U17 M12 AC4 -
154. I/O (D7) 83 109 123 W19 P15 AD2 792
155. GCK3 (I/O) 84 110 124 W18 N14 AC3 795
156. I/O 85 111 125 T15 L11 AB4 804
157. I/O 86 112 126 U16 M13 AD1 807
158. I/O - - 127 V17 N15 AA4 810
159. I/O - - 128 X18 M14 AA3 816
160. I/O - - - U15 - AB2 819
161. I/O - - - T14 - AC1 828
162. I/O (D6) 87 113 129 W17 J10 Y3 831
163. I/O 88 114 130 V16 L12 AA2 834
164. I/O 89 115 131 X17 M15 AA1 840
165. I/O 90 116 132 U14 L13 W4 843
7
166. I/O - 117 133 V15 L14 W3 846
167. I/O - 118 134 T13 K11 Y2 852
168. I/O - - - W16 - Y1 855
169. I/O - - - W15 - V4 858
GND 91 119 135 X16 GND* GND* -
170. I/O - - 136 U13 L15 V3 864
171. I/O - - 137 V14 K12 W2 867
172. I/O 92 120 138 W14 K13 U4 870
173. I/O 93 121 139 V13 K14 U3 876
VCC - - 140 X15 VCC* VCC* -
174. I/O (D5) 94 122 141 T12 K15 V2 879
175. I/O (CS0) 95 123 142 X14 J12 V1 882
176. I/O - - - X13 - T1 888
177. I/O - - - V12 - R4 891
178. I/O - 124 144 W12 J13 R3 894
179. I/O - 125 145 T11 J14 R2 900
180. I/O 96 126 146 X12 J15 R1 903
181. I/O 97 127 147 U11 J11 P3 906
182. I/O (D4) 98 128 148 V11 H13 P2 912
183. I/O 99 129 149 W11 H14 P1 915
VCC 100 130 150 X10 VCC* VCC* -
GND 101 131 151 X11 GND* GND* -
184. I/O (D3) 102 132 152 W10 H12 N2 924
185. I/O (RS) 103 133 153 V10 H11 N4 927
186. I/O 104 134 154 T10 G14 N3 936
187. I/O 105 135 155 U10 G15 M1 939
188. I/O - 136 156 X9 G13 M2 942
189. I/O - 137 157 W9 G12 M3 948

R
190. I/O - - - X8 - M4 951
191. I/O - - - V9 - L1 954
192. I/O (D2) 106 138 159 W8 G11 J1 960
193. I/O 107 139 160 X7 F15 K3 963
VCC - - 161 X5 VCC* VCC*
194. I/O 108 140 162 V8 F14 J2 966
195. I/O 109 141 163 W7 F13 J3 972
196. I/O - - 164 U8 G10 K4 975
197. I/O - - 165 W6 E15 G1 978
GND 110 142 166 X6 GND* GND*
198. I/O - - - T8 - H2 984
199. I/O - - - V7 - H3 987
200. I/O - - 167 X4 E14 J4 990
201. I/O - - 168 U7 F12 F1 996
202. I/O - 143 169 W5 E13 G2 999
203. I/O - 144 170 V6 D15 G3 1002
204. I/O 111 145 171 T7 F11 F2 1008
205. I/O 112 146 172 X3 D14 E2 1011
206. I/O (D1) 113 147 173 U6 E12 F3 1014
207. I/O (RCLK-BUSY/RDY) 114 148 174 V5 C15 G4 1020
208. I/O - - - W4 - D2 1023
209. I/O - - - W3 - F4 1032
210. I/O 115 149 175 T6 D13 E3 1035
211. I/O 116 150 176 U5 C14 C2 1038
212. I/O (D0, DIN) 117 151 177 V4 F10 D3 1044
213. I/O (DOUT) 118 152 178 X1 B15 E4 1047
CCLK 119 153 179 V3 C13 C3 -
VCC 120 154 180 W1 VCC* VCC* -
214. I/O (TDO) 121 159 181 U4 A15 D4 0
GND 122 160 182 X2 GND* GND* -
215. I/O (A0, WS) 123 161 183 W2 A14 B3 9
216. GCK4 (A1, I/O) 124 162 184 V2 B13 C4 15
217. I/O 125 163 185 R5 E11 D5 18
218. I/O 126 164 186 T4 C12 A3 21
219. I/O (A2, CS1) 127 165 187 U3 A13 D6 27
220. I/O (A3) 128 166 188 V1 B12 C6 30
221. I/O - - - R4 - B5 33
222. I/O - - - P5 - A4 39
223. I/O - - 189 U2 F9 C7 42
224. I/O - - 190 T3 D11 B6 45
225. I/O 129 167 191 U1 A12 A6 51
226. I/O 130 168 192 P4 C11 D8 54
227. I/O - 169 193 R3 B11 B7 57
228. I/O - 170 194 N5 E10 A7 63
229. I/O - - 195 T2 - D9 66
230. I/O - - - R2 - C9 69
GND 131 171 196 T1 GND* GND* -
231. I/O 132 172 197 N4 A11 B8 75
232. I/O 133 173 198 P3 D10 D10 78
233. I/O - - 199 P2 C10 C10 81
234. I/O - - 200 N3 B10 B9 87
VCC - - 201 R1 VCC* VCC* -

R
235. I/O - - - M5 - B11 90
236. I/O - - - P1 - A11 93
237. I/O (A4) 134 174 202 N1 A10 D12 99
238. I/O (A5) 135 175 203 M3 D9 C12 102
239. I/O - 176 205 M2 C9 B12 105
240. I/O 136 177 206 L5 B9 A12 111
241. I/O 137 178 207 M1 A9 C13 114
242. I/O 138 179 208 L4 E9 B13 117
243. I/O (A6) 139 180 209 L3 C8 A13 126
244. I/O (A7) 140 181 210 L2 B8 B14 129
GND 141 182 211 L1 GND* GND* -
Additional No Connect (N.C.) Connections for HQ208 and HQ240 Packages

HQ208 HQ240
206 102 219
207 104 22
208 105 37
1 107 83
3 155 98
51 156 143
52 157 158
53 158 204
54 - - 7
Notes: * Pins labeled VCC* are internally bonded to a VCC plane within the BG225 and BG352 packages. The external pins for the
BG225 are: B2, D8, H15, R8, B14, R1, H1, and R15. The external pins for the BG352 are: A10, A17, B2, B25, D13, D19, D7,
G23, H4, K1, K26, N23, P4, U1, U26, W23, Y4, AC14, AC20, AC8, AE2, AE25, AF10, and AF17.
Pins labeled GND* are internally bonded to a ground plane within the BG225 and BG352 packages. The external pins for
the BG225 are: A1, D12, G7, G9, H6, H8, H10, J8, K8, A8, F8, G8, H2, H7, H9, J7, J9, M8. The external pins for the BG352
are: A1, A2, A5, A8, A14, A19, A22, A25, A26, B1, B26, E1, E26, H1, H26, N1, P26, W1, W26, AB1, AB26, AE1, AE26, AF1,
AF13, AF19, AF2, AF22, AF25, AF26, AF5, AF8.
Boundary Scan Bit 0 = TDO.T


R
PINS 64 84 100 100 144 156 160 176 191 208 208 223 225 240 240 299 352
High-Perf.
High-Perf.
Ceram.
Ceram.
Ceram.
Ceram.
VQFP
PQFP
VQFP
PQFP
PQFP
PQFP
TQFP
TQFP
PLCC
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
Plast.
PGA
PGA
PGA
BGA
PGA
BGA
QFP
QFP
TYPE
HQ208
HQ240
PQ100
VQ100
PG156
PQ160
PG191
PQ208
PG223
BG225
PQ240
PG299
BG352
TQ144
TQ176
VQ64*
PC84
CODE
XC5202
-4 C C C C C C
-3 C C C C C C
XC5204 -4 C C C C C C
-3 C C C C C C
-6 CI CI CI CI CI CI CI CI
XC5206 -4 C C C C C C C C
-3 C C C C C C C C
XC5210 -4 C C C C C C C C
-3 C C C C C C C C
-5 C C C C C C
XC5215 -4 C C C C C C
-3 C C C C C C
7/8/98
* VQ64 package supports Master Serial, Slave Serial, and Express configuration modes only.

Max
Package Type
Device I/O VQ64 PC84 PQ100 VQ100 TQ144 PG156 PQ160 TQ176 PG191 HQ208 PQ208 PG223 BG225 HQ240 PQ240 PG299 BG352
XC5202 84 52 65 81 81 84 84
XC5204 124 65 81 81 117 124 124
XC5206 148 65 81 81 117 133 148 148 148
XC5210 196 65 117 133 149 164 196 196 196
XC5215 244 133 164 196 197 244 244
7/8/98
Example: XC5210-6PQ208C
Speed Grade Number of Pins
Package Type

R
Revisions
Version Description
12/97 Rev 5.0 added -3, -4 specification
7/98 Rev 5.1 added Spartan family to comparison, removed HQ304
11/98 Rev 5.2 added changes to grahics and ?

R
SPROM Products and
Programming Support
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

0
R
SPROM Products and Programming
Support Table of Contents
0 8*
XC1700L Series High Density Serial Configuration PROMs

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
Pin Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
Controlling Serial PROMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5
FPGA Master Serial Mode Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5
Standby Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-7
Programming. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-7
XC1701 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
DC Characteristics Over Operating Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
XC1704L, XC1702L, XC1701L, XQ1701L, & XC17512L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9
AC Characteristics Over Operating Condition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-10
Marking Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-12
XC1700E Family of Serial Configuration PROMs 8

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-15
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-15
Pin Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-16
Standby Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-20
Programming. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-20
XC1736E, XC1765E, XC17128E and XC17256E. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-21
XC1765EL, XC17128EL and XC17256EL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-22
Spartan and SpartanXL Families of Serial Configuration PROMs

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-27
Spartan SPROM Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-27
Pin Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-28
Standby Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-31
8-1
R
SPROM Products and Programming Support Contents
Programming the Spartan Family Serial PROMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-31

XC17S05, XC17S10, XC17S20, XC17S30, XC17S40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-32
XC17S05XL, XC17S10XL, XC17S20XL, XC17S30XL, XC17S40XL . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-33
HW-130 Programmer
8-2
0
XC1700L Series High Density Serial

R
Configuration PROMs
Including XQ1701L QPROTM Series
January 27, 1999 (Version 2.2) 0 8* Product Specification
Features Description
• Serial Configuration one-time programmable (OTP) The XC1704L, XC1702L, XC1701L, and the XC17512L are
read-only memory designed to store configuration Xilinxs 3.3V series of high density serial configuration
bitstreams of Xilinx FPGA devices PROMs (SPROMs). Included within this family are the
• Simple interface to the FPGA; requires only one user XC1701 (5V) and the XQ1701L (3.3V) SPROMs to provide
I/O pin an easy-to-use, cost-effective method for storing large Xil-
• Cascadable for storing longer or multiple bitstreams inx FPGA configuration bitstreams.
• Programmable reset polarity (active High or active Low)
When the FPGA is in Master Serial mode, it generates a
for compatibility with different FPGA solutions
configuration clock that drives the SPROM. A short access
• Supports XC4000EX/XL/XLA/XV fast configuration
time after the rising clock edge, data appears on the
mode (15.0 MHz)
SPROM DATA output pin that is connected to the FPGA
• Low-power CMOS Floating Gate process
DIN pin. The FPGA generates the appropriate number of
• XC1704L, XC1702L, XC1701L, XQ1701L and the
clock pulses to complete the configuration. Once config-
XC17512L are 3.3 V devices
ured, it disables the SPROM. When the FPGA is in Slave
• XC1701 is a 5 V device only
Serial mode, the SPROM and the FPGA must both be
• Available in compact plastic packages: 8-pin PDIP,
clocked by an incoming signal.
20-pin SOIC, 20-pin PLCC, 44-pin PLCC or 44-pin
VQFP. Multiple devices can be concatenated by using the CEO
• QPROTM parts available in 44-pin ceramic LCC and output to drive the CE input of the following device. The
20-pin SOIC. clock inputs and the DATA outputs of all SPROMs in this
• Programming support by leading programmer chain are interconnected. All devices are compatible and
manufacturers. can be cascaded with other members of the family. 8
• Design support using the Xilinx Alliance and For device programming, either the Xilinx Alliance or Foun-
Foundation series software packages. dation series development system compiles the FPGA
design file into a standard Hex format, which is then trans-
ferred to most commercial PROM programmers.
VCC VPP GND
CE CEO
RESET/
OE or
OE/
RESET
Address Counter
CLK TC
EPROM OE
Output
Cell DATA
Matrix
X3185
Figure 1: Simplified Block Diagram (does not show programming circuit)

R
Pin Description Serial PROM Pinouts

44-Pin
DATA Pin Name
8-Pin 20-Pin 20-Pin 44-Pin
PLCC
PDIP SOIC PLCC VQFP
Data output, 3-stated when either CE or OE are inactive. CLCC
During programming, the DATA pin is I/O. Note that OE can DATA 1 1 2 40 2
be programmed to be either active High or active Low. CLK 2 3 4 43 5
RESET/OE 3 8 6 13 19
CLK (OE/RESET)
CE 4 10 8 15 21
Each rising edge on the CLK input increments the internal
GND 5 11 10 18 & 41 24 & 3
address counter, if both CE and OE are active.
CEO 6 13 14 21 27
RESET/OE VPP 7 18 17 35 41
VCC 8 20 20 38 44
When High, this input holds the address counter reset and
3-states the DATA output. The polarity of this input pin is Capacity
programmable as either RESET/OE or OE/RESET. To
avoid confusion, this document describes the pin as Devices Configuration Bits
RESET/OE, although the opposite polarity is possible on all XC1704L 4,194,304
devices. When RESET is active, the address counter is XC1702L 2,097,152
held at zero, and the DATA output is 3-stated. The polarity XC1701L 1,048,576
of this input is programmable. The default is active High XC1701 1,048,576
RESET, but the preferred option is active Low RESET, XC17512L 524,288
because it can be driven by the FPGA’s INIT pin.
Xilinx FPGAs and Compatible SPROMs.
The polarity of this pin is controlled in the programmer inter-
face. This input pin is easily inverted using the Xilinx Device Configuration Bits SPROM
HW-130 Programmer. Third-party programmers have dif- XC4010XL 283,424 XC17512L
ferent methods to invert this pin. XC4013XL/XLA 393,632 XC17512L
XC4020E 329,312 XC1701
CE XC4020XL/XLA 521,880 XC17512L
When High, this pin disables the internal address counter, XC4025E 422,176 XC1701
3-states the DATA output, and forces the device into low-ICC XC4028XL/XLA 668,184 XC1701L
standby mode. XC4028EX 668,184 XC1701
XC4036EX 832,528 XC1701
CEO XC4036XL/XLA 832,528 XC1701L
Chip Enable output, to be connected to the CE input of the XC4044XL/XLA 1,014,928 XC1701L
next SPROM in the daisy chain. This output is Low when XC4052XL/XLA 1,215,368 XC1702L
the CE and OE inputs are both active AND the internal XC4062XL/XLA 1,433,864 XC1702L
address counter has been incremented beyond its Terminal XC4085XL/XLA 1,924,992 XC1702L
Count (TC) value. In other words: when the PROM has XC40110XV 2,686,136 XC1704L
been read, CEO will follow CE as long as OE is active. XC40150XV 3,373,448 XC1704L
When OE goes inactive, CEO stays High until the PROM is XC40200XV 4,551,056 XC1704L +
reset. Note that OE can be programmed to be either active XC17512L
High or active Low. XC40250XV 5,433,888 XC1704L +
XC1702L
VPP XCV50 559,232 XC1701L
XCV100 781,248 XC1701L
Programming voltage. No overshoot above the specified
XCV150 1,041,128 XC1701L
max voltage is permitted on this pin. For normal read oper-
XCV200 1,335,872 XC1702L
ation, this pin must be connected to VCC. Failure to do so
may lead to unpredictable, temperature-dependent opera- XCV300 1,751,840 XC1702L
tion and severe problems in circuit debugging. Do not leave XCV400 2,546,080 XC1704L
VPP floating! XCV600 3,608,000 XC1704L
XCV800 4,715,648 XC1704L +
VCC and GND XC1701L
XCV1000 6,127,776 XC1704L +
Positive supply and ground pins. XC1702L

R
Controlling Serial PROMs If the user-programmable, dual-function DIN pin on the

FPGA is used only for configuration, it must still be held at
Connecting the FPGA device with the SPROM. a defined level during normal operation. The Xilinx FPGA
• The DATA output(s) of the of the SPROM(s) drives the families take care of this automatically with an on-chip
DIN input of the lead FPGA device. default pull-up resistor.
• The Master FPGA CCLK output drives the CLK input(s)
of the SPROM(s). Programming the FPGA With Counters
• The CEO output of a SPROM drives the CE input of the Unchanged Upon Completion
next SPROM in a daisy chain (if any). When multiple FPGA-configurations for a single FPGA are
• The RESET/OE input of all SPROMs is best driven by stored in a SPROM, the OE pin should be tied Low. Upon
the INIT output of the lead FPGA device. This
connection assures that the SPROM address counter is power-up, the internal address counters are reset and con-
reset before the start of any (re)configuration, even figuration begins with the first program stored in memory.
when a reconfiguration is initiated by a VCC glitch. Other Since the OE pin is held Low, the address counters are left
methods – such as driving RESET/OE from LDC or unchanged after configuration is complete. Therefore, to
system reset – assume the SPROM internal reprogram the FPGA with another program, the DONE line
power-on-reset is always in step with the FPGA’s is pulled Low and configuration begins at the last value of
internal power-on-reset. This may not be a safe the address counters.
assumption.
• The SPROM CE input can be driven from either the This method fails if a user applies RESET during the FPGA
LDC or DONE pins. Using LDC avoids potential configuration process. The FPGA aborts the configuration
contention on the DIN pin. and then restarts a new configuration, as intended, but the
• The CE input of the lead (or only) SPROM is driven by Serial PROM does not reset its address counter, since it
the DONE output of the lead FPGA device, provided never saw a High level on its OE input. The new configura-
that DONE is not permanently grounded. Otherwise, tion, therefore, reads the remaining data in the PROM and
LDC can be used to drive CE, but must then be interprets it as preamble, length count etc. Since the FPGA
unconditionally High during user operation. CE can also is the master, it issues the necessary number of CCLK
be permanently tied Low, but this keeps the DATA pulses, up to 16 million (224) and DONE goes High. How-
output active and causes an unnecessary supply ever, the FPGA configuration will be completely wrong, with
current of 10 mA maximum. potential contentions inside the FPGA and on its output
pins. This method must, therefore, never be used when 8
FPGA Master Serial Mode Summary there is any chance of external reset during configuration.
The I/O and logic functions of the Configurable Logic Block Cascading Serial Configuration PROMs
(CLB) and their associated interconnections are estab-
For multiple FPGAs configured as a daisy-chain, or for
lished by a configuration program. The program is loaded
future FPGAs requiring larger configuration memories, cas-
either automatically upon power up, or on command,
caded SPROMs provide additional memory. After the last
depending on the state of the three FPGA mode pins. In
bit from the first SPROM is read, the next clock signal to the
Master Serial mode, the FPGA automatically loads the con-
SPROM asserts its CEO output Low and disables its DATA
figuration program from an external memory. The Xilinx
line. The second SPROM recognizes the Low level on its
SPROMs have been designed for compatibility with the
CE input and enables its DATA output. See Figure 2.
Master Serial mode.
After configuration is complete, the address counters of all
Upon power-up or reconfiguration, an FPGA enters the
cascaded SPROMs are reset if the FPGA RESET pin goes
Master Serial mode whenever all three of the FPGA
Low, assuming the SPROM reset polarity option has been
mode-select pins are Low (M0=0, M1=0, M2=0). Data is
inverted.
read from the SPROM sequentially on a single data line.
Synchronization is provided by the rising edge of the tem- To reprogram the FPGA with another program, the DONE
porary signal CCLK, which is generated during configura- line goes Low and configuration begins where the address
tion. counters had stopped. In this case, avoid contention
between DATA and the configured I/O use of DIN.
Master Serial Mode provides a simple configuration inter-
face. Only a serial data line and two control lines are
required to configure an FPGA. Data from the SPROM is
read sequentially, accessed via the internal address and bit
counters which are incremented on every valid rising edge
of CCLK.

R
Vcc
DOUT OPTIONAL
Daisy-chained
FPGAs with
Different
Configurations
FPGA OPTIONAL
Slave FPGAs
with Identical
MODES Configurations
Vcc
RESET RESET
VCC VPP
DIN DATA DATA
CCLK CLK CLK Cascaded
SPROM Serial
DONE CE CEO CE Memory
(Low Resets the Address Pointer)
CCLK
(OUTPUT)
DIN
DOUT
(OUTPUT)
X8256_01
Figure 2: Master Serial Mode. The one-time-programmable SPROM supports automatic loading of configuration
programs. Multiple devices can be cascaded to support additional FPGAs. An early DONE inhibits the PROM data output
one CCLK cycle before the FPGA I/Os become active.

R
Standby Mode Programming

The PROM enters a low-power standby mode whenever The devices can be programmed on programmers supplied
CE is asserted High. The output remains in a high imped- by Xilinx or qualified third-party vendors. The user must
ance state regardless of the state of the OE input. ensure that the appropriate programming algorithm and the
latest version of the programmer software are used. The
wrong choice can permanently damage the device.
Table 1: Truth Table for XC1700 Control Inputs
Control Inputs Outputs

Internal Address
RESET CE DATA CEO Icc
Inactive Low if address < TC: increment active High active
if address > TC: don’t change 3-state Low reduced
Active Low Held reset 3-state High active
Inactive High Not changing 3-state High standby
Active High Held reset 3-state High standby
Notes: 1. The XC1700 RESET input has programmable polarity
2. TC = Terminal Count = highest address value. TC+1 = address 0.
IMPORTANT: Always tie the VPP pin to VCC in your application. Never leave VPP floating.

R
XC1701
VPP Supply voltage relative to GND -0.5 to +12.5 V
VIN Input voltage relative to GND -0.5 to VCC +0.5 V
Operating Conditions
VCC Commercial Supply voltage relative to GND (TA = 0°C to +70°C) 4.75 5.25 V
Industrial Supply voltage relative to GND (TA = -40°C to +85°C) 4.50 5.50 V
Note: During normal read operation VPP must be connect to VCC
DC Characteristics Over Operating Condition

VIH High-level input voltage 2.0 VCC V
VOH High-level output voltage (IOH = -4 mA) Commercial 3.86 V
VOL Low-level output voltage (IOL = +4 mA) 0.32 V
VOH High-level output voltage (IOH = -4 mA) Industrial 3.76 V
ICCA Supply current, active mode (at maximum frequency) 10.0 mA
ICCS Supply current, standby mode 100.0 µA
IL Input or output leakage current -10.0 10.0 µA
CIN Input Capacitance (VIN = GND, f = 1.0 MHZ) 10.0 pF
COUT Output Capacitance (VIN = GND, f = 1.0 MHZ) 10.0 pF

R
XC1704L, XC1702L, XC1701L, XQ1701L, & XC17512L

reliability.
Supply voltage relative to GND 3.0 3.6 V

Ceramic Package (TC = -55°C to +125°C)
Military 8
Supply voltage relative to GND 3.0 3.6 V
Plastic Package (TJ = -55°C to +125°C)
Note: During normal read operation VPP must be connected to VCC

VOH High-level output voltage (IOH = -3 mA) 2.4 V
ICCS Supply current, standby mode Commercial/Industrial 50.0 µA
Military 100.0 µA
CIN Input Capacitance (VIN = GND, f = 1.0 MHZ) 10.0 pF
COUT Output Capacitance (VIN = GND, f = 1.0 MHZ) 10.0 pF

R
AC Characteristics Over Operating Condition
CE
9 10
9 TSCE TSCE THCE
RESET/OE
11 THOE
TLC 8 THC 6 TCYC
7
CLK
3 4
TOE 1 TCAC TOH 5 TDF
2 TCE
DATA
4 TOH
X2634
XC1704L,
XC1702L,
XC1701 XC1701L,
Symbol Description XQ1701L & Units
XC17512L
Min Max Min Max
1 TOE OE to Data Delay 25 30 ns
2 TCE CE to Data Delay 45 45 ns
3 TCAC CLK to Data Delay 45 45 ns
4 TOH Data Hold From CE, OE, or CLK 0 0 ns
5 TDF CE or OE to Data Float Delay2 50 50 ns
6 TCYC Clock Periods 67 67 ns
7 TLC CLK Low Time3 20 25 ns
8 THC CLK High Time3 20 25 ns
9 TSCE CE Setup Time to CLK (to guarantee proper counting) 20 25 ns
10 THCE CE Hold Time to CLK (to guarantee proper counting) 0 0 ns
11 THOE OE Hold Time (guarantees counters are reset) 20 25 ns
Notes: 1. AC test load = 50 pF
2. Float delays are measured with 5 pF AC loads. Transition is measured at +/- 200mV from steady state active levels.
3. Guaranteed by design, not tested.
4. All AC parameters are measured with VIL = 0.0 V and VIH = 3.0 V.

R
AC Characteristics Over Operating Condition When Cascading
RESET/OE
CE
CLK
12 TCDF
DATA Last Bit First Bit
13 TOCK 15 TOOE
CEO
14 TOCE 14 TOCE
X3183

12 TCDF CLK to Data Float Delay2, 3 50 ns
13 TOCK CLK to CEO Delay3 30 ns
14 TOCE CE to CEO Delay3 35 ns
15 TOOE RESET/OE to CEO Delay3 30 ns
2. Float delays are measured with 5 pF AC loads. Transition is measured at +/- 200mV from steady state active levels. 8

R
XC 1701L PC20 C
Device Type
XC Commercial
XQ QPROTM Operating Range/Processing
C = Commercial (TA = 0° to +70°C)
Device Number Package Type I = Industrial (TA = –40° to +85°C)
1704L N = Military Plastic (TJ = –55° to +125°C)
PD8 = 8-Pin Plastic DIP
1702L M = Military (TC = –55° to +125°C)
SO20 = 20-Pin Plastic Small-Outline Package
1701L B = Military (TC = –55° to +125°C)
PC20 = 20-Pin Plastic Leaded Chip Carrier
1701 QML certified to MIL-PRF-38535
VQ44 = 44-Pin Plastic Quad Flat Package
17512L PC44 = 44-Pin Plastic Chip Carrier
CC44 = 44-Pin Ceramic Chip Carrier
Valid Ordering Combinations

XC1704LVQ44C XC1702LVQ44C XC1701LPD8C XC1701PD8C XC17512LPD8C
XC1704LPC44C XC1702LPC44C XC1701LSO20C XC1701SO20C XC17512LSO20C
XC1701LPC20C XC1701PC20C XC17512LPC20C
XC1704LVQ44I XC1702LVQ44I XC1701LPD8I XC1701PD8I XC17512LPD8I

XC1704LPC44I XC1702LPC44I XC1701LSO20I XC1701SO20I XC17512LSO20I
XC1701PC20I XC1701PC20I XC17512LPC20I
XQ1701LCC44M
XQ1701LCC44B
XQ1701LS020N
Marking Information
Due to the small size of the commercial serial PROM packages, the complete ordering part number cannot be marked on
the package. The XC prefix is deleted and the package code is simplified. The XQ CC44 packages are marked as ordered.
Device marking on the commercial and military plastic packages is as follows:
1701L J C
Device Number Operating Range/Processing

XC1704L C = Commercial (TA = 0° to +70°C)
XC1702L Package Type I = Industrial (TA = –40° to +85°C)
XC1701L P = 8-Pin Plastic DIP N = Military Plastic (TJ = –55° to +125°C)
XQ1701L S = 20-Pin Plastic Small-Outline Package M = Military (TC = –55° to +125°C)
XC1701 J = 20-Pin Plastic Leaded Chip Carrier B = Military (TC = –55° to +125°C)
XC17512L VQ44 = 44-Pin Plastic Quad Flat Package QML certified to MIL-PRF-38535
PC44 = 44-Pin Plastic Chip Carrier
CC44 = 44-Pin Ceramic Chip Carrier

R
Revision Control
Date Revision
7/14/98 Major revisions to include the XC1704L, XC1702L, and the XQ1701L devices, packages and op-
erating conditions. Also revised the timing specifications on page 10.
9/8/98 Revised the marking information on page 12 for the VQ44. Updated “DC Characteristics Over Op-
erating Condition” on page 8. and page 9. Added references to the XC4000XLA and XC4000XV
families in “Xilinx FPGAs and Compatible SPROMs.” on page 4. and Figure 2 on page 6.
12/18/98 Added Virtex FPGAs to “Xilinx FPGAs and Compatible SPROMs.” on page 4. Added the PC44
package for the XC1702L & XC1704L products.
1/27/99 Changed Military ICCS on page 9.

R

0
R
XC1700E Family of
Serial Configuration PROMs
December 7, 1998 (Version 1.4) 0 8* Product Specification
Features Description
• Serial Configuration one-time programmable (OTP) The XC1700 family of serial configuration PROMs
read-only memory designed to store configuration (SPROMs) provides an easy-to-use, cost-effective method
bitstreams of Xilinx FPGA devices for storing Xilinx FPGA configuration bitstreams.
• Simple interface to the FPGA requires only one user
When the FPGA is in Master Serial mode, it generates a
I/O pin
configuration clock that drives the SPROM. A short access
• Cascadable for storing longer or multiple bitstreams
time after the rising clock edge, data appears on the
SPROM DATA output pin that is connected to the FPGA
for compatibility with different FPGA solutions
DIN pin. The FPGA generates the appropriate number of
• The XC17128E/EL and XC17256E/EL devices support
clock pulses to complete the configuration. Once config-
the XC4000EX/XL/XLA/XV fast configuration mode
ured, it disables the SPROM. When the FPGA is in Slave
(15.0 MHz)
Serial mode, the SPROM and the FPGA must both be
• Low-power CMOS floating gate process
clocked by an incoming signal.
• Available in 5 V and 3.3 V versions
• Available in compact plastic 8-pin DIP, 8-pin SOIC, 8-pin Multiple devices can be concatenated by using the CEO
VOIC, or 20-pin PLCC packages. output to drive the CE input of the following device. The
• Programming support by leading programmer clock inputs and the DATA outputs of all SPROMs in this
manufacturers. chain are interconnected. All devices are compatible and
• Design support using the Xilinx Alliance and can be cascaded with other members of the family.
Foundation series software packages. For device programming, either the Xilinx Alliance or the
Foundation series development systems compiles the
FPGA design file into a standard HEX format which is then 8
transferred to most commercial PROM programmers.
VCC VPP GND
CE CEO
RESET/
OE or
OE/
RESET
Address Counter
CLK TC
EPROM OE
Output
Cell DATA
Matrix
X3185

R
XC1700E Family of Serial Configuration PROMs
Pin Description Serial PROM Pinouts

Pin Name 8-Pin 20-Pin
DATA
DATA 1 2
Data output, 3-stated when either CE or OE are inactive.
During programming, the DATA pin is I/O. Note that OE can CLK 2 4
be programmed to be either active High or active Low. RESET/OE (OE/RESET) 3 6
CLK CE 4 8
Each rising edge on the CLK input increments the internal GND 5 10
address counter, if both CE and OE are active. CEO 6 14
RESET/OE VPP 7 17
When High, this input holds the address counter reset and VCC 8 20
3-states the DATA output. The polarity of this input pin is
programmable as either RESET/OE or OE/RESET. To Capacity
avoid confusion, this document describes the pin as
Device Configuration Bits
RESET/OE, although the opposite polarity is possible on all
devices. When RESET is active, the address counter is XC1736E 36,288
held at zero, and the DATA output is 3-stated. The polarity XC1765E or EL 65,536
of this input is programmable. The default is active High XC17128E or EL 131,072
RESET, but the preferred option is active Low RESET, XC17256E or EL 262,144
because it can be driven by the FPGA’s INIT pin. XC17512L 524,288
XC1701, XC1701L or XQ1701L 1,048,576
The polarity of this pin is controlled in the programmer inter- XC1702L 2,097,152
face. This input pin is easily inverted using the Xilinx XC1704L 4,194,304
HW-130 programmer software. Third-party programmers Note: The XC17512L and larger SPROMs are specified in a sep-
have different methods to invert this pin. arate datasheet.
CE
When High, this pin disables the internal address counter,
3-states the DATA output, and forces the device into low-ICC
standby mode.
CEO
Chip Enable output, to be connected to the CE input of the
next SPROM in the daisy chain. This output is Low when
the CE and OE inputs are both active AND the internal
address counter has been incremented beyond its Terminal
Count (TC) value. In other words: when the PROM has
been read, CEO will follow CE as long as OE is active.
When OE goes inactive, CEO stays High until the PROM is
reset. Note that OE can be programmed to be either active
High or active Low.
VPP
Programming voltage. No overshoot above the specified
max voltage is permitted on this pin. For normal read oper-
ation, this pin must be connected to VCC. Failure to do so
may lead to unpredictable, temperature-dependent opera-
tion and severe problems in circuit debugging. Do not leave
VPP floating!
VCC and GND

VCC is positive supply pin and GND is ground pin.

R
Number of Configuration Bits, Including Header Controlling Serial PROMs

for Xilinx FPGAs and Compatible SPROMs Connecting the FPGA device with the SPROM.
Device Configuration Bits SPROM
• The DATA output(s) of the SPROM(s) drives the DIN
XC4003E 53,984 XC17128E1
input of the lead FPGA device.
XC4005E 95,008 XC17128E • The Master FPGA CCLK output drives the CLK input(s)
XC4006E 119,840 XC17128E of the SPROM(s).
XC4008E 147,552 XC17256E • The CEO output of a SPROM drives the CE input of the
XC4010E 178,144 XC17256E next SPROM in a daisy chain (if any).
XC4013E 247,968 XC17256E • The RESET/OE input of all SPROMs is best driven by
XC4020E 329,312 XC1701 the INIT output of the lead FPGA device. This
XC4025E 422,176 XC1701 connection assures that the SPROM address counter is
XC4002XL 61,100 XC17128EL1 reset before the start of any (re)configuration, even
when a reconfiguration is initiated by a VCC glitch. Other
XC4005XL 151,960 XC17256EL
methods – such as driving RESET/OE from LDC or
XC4010XL 283,424 XC17512L
system reset – assume the SPROM internal
XC4013XL/XLA 393,632 XC17512L power-on-reset is always in step with the FPGA’s
XC4020XL/XLA 521,880 XC17512L internal power-on-reset. This may not be a safe
XC4028XL/XLA 668,184 XC1701L assumption.
XC4028EX 668,184 XC1701 • The SPROM CE input can be driven from either the
XC4036EX/XL/XLA 832,528 XC1701L LDC or DONE pins. Using LDC avoids potential
XC4036EX 832,528 XC1701 contention on the DIN pin.
XC4044XL/XLA 1,014,928 XC1701L • The CE input of the lead (or only) SPROM is driven by
XC4052XL/XLA 1,215,368 XC1702L
the DONE output of the lead FPGA device, provided
that DONE is not permanently grounded. Otherwise,
XC4062XL/XLA 1,433,864 XC1702L
LDC can be used to drive CE, but must then be
XC4085XL/XLA 1,924,992 XC1702L
unconditionally High during user operation. CE can also
XC40110XV 2,686,136 XC1704L
be permanently tied Low, but this keeps the DATA
XC40150XV 3,373,448 XC1704L
output active and causes an unnecessary supply
XC40200XV 4,551,056 XC1704L +
current of 10 mA maximum.
XC17512L 8
XC40250XV 5,433,888 XC1704L+
XC1702L FPGA Master Serial Mode Summary
XC5202 42,416 XC1765E The I/O and logic functions of the Configurable Logic Block
XC5204 70,704 XC17128E (CLB) and their associated interconnections are estab-
XC5206 106,288 XC17128E lished by a configuration program. The program is loaded
XC5210 165,488 XC17256E either automatically upon power up, or on command,
XC5215 237,744 XC17256E depending on the state of the three FPGA mode pins. In
XCV50 559,232 XC1701L Master Serial mode, the FPGA automatically loads the con-
XCV100 781,248 XC1701L figuration program from an external memory. The Xilinx
XCV150 1,041,128 XC1701L SPROMs have been designed for compatibility with the
XCV200 1,335,872 XC1702L Master Serial mode.
XCV300 1,751,840 XC1702L Upon power-up or reconfiguration, an FPGA enters the
XCV400 2,546,080 XC1704L Master Serial mode whenever all three of the FPGA
XCV600 3,608,000 XC1704L mode-select pins are Low (M0=0, M1=0, M2=0). Data is
XCV800 4,715,648 XC1704L + read from the SPROM sequentially on a single data line.
XC1701L Synchronization is provided by the rising edge of the tem-
XCV1000 6,127,776 XC1704L + porary signal CCLK, which is generated during configura-
XC1702L tion.
Note: 1.The suggested SPROM is determined by compatibility with Master Serial Mode provides a simple configuration inter-
the higher configuration frequency of the Xilinx FPGA face. Only a serial data line and two control lines are
CCLK. Designers using the default slow configuration
frequency (CCLK) can use the XC1765E or XC1765EL for
required to configure an FPGA. Data from the SPROM is
the noted FPGA devices. read sequentially, accessed via the internal address and bit
2. The XC1701, XC1701L, XC1702L, XC1704L & XC17512L counters which are incremented on every valid rising edge
are specified in a separate XC1700L High Density
datasheet.
of CCLK.

R
If the user-programmable, dual-function DIN pin on the the master, it issues the necessary number of CCLK
FPGA is used only for configuration, it must still be held at pulses, up to 16 million (224) and DONE goes High. How-
a defined level during normal operation. Xilinx FPGAs take ever, the FPGA configuration will be completely wrong, with
care of this automatically with an on-chip default pull-up potential contentions inside the FPGA and on its output
resistor. pins. This method must, therefore, never be used when
there is any chance of external reset during configuration.
Programming the FPGA With Counters
Unchanged Upon Completion Cascading Serial Configuration PROMs
When multiple FPGA-configurations for a single FPGA are For multiple FPGAs configured as a daisy-chain, or for
stored in a SPROM, the OE pin should be tied Low. Upon future FPGAs requiring larger configuration memories, cas-
power-up, the internal address counters are reset and con- caded SPROMs provide additional memory. After the last
figuration begins with the first program stored in memory. bit from the first SPROM is read, the next clock signal to the
Since the OE pin is held Low, the address counters are left SPROM asserts its CEO output Low and disables its DATA
unchanged after configuration is complete. Therefore, to line. The second SPROM recognizes the Low level on its
reprogram the FPGA with another program, the DONE line CE input and enables its DATA output. See Figure 2.
is pulled Low and configuration begins at the last value of After configuration is complete, the address counters of all
the address counters. cascaded SPROMs are reset if the FPGA RESET pin goes
This method fails if a user applies RESET during the FPGA Low, assuming the SPROM reset polarity option has been
configuration process. The FPGA aborts the configuration inverted.
and then restarts a new configuration, as intended, but the To reprogram the FPGA with another program, the DONE
SPROM does not reset its address counter, since it never line goes Low and configuration begins where the address
saw a High level on its OE input. The new configuration, counters had stopped. In this case, avoid contention
therefore, reads the remaining data in the PROM and inter- between DATA and the configured I/O use of DIN.
prets it as preamble, length count etc. Since the FPGA is

R
Vcc
DOUT OPTIONAL
Daisy-chained
FPGAs with
Different
Configurations
FPGA OPTIONAL
Slave FPGAs
with Identical
MODES Configurations
Vcc
RESET RESET
VCC VPP
DIN DATA DATA
CCLK CLK CLK Cascaded
SPROM Serial
DONE CE CEO CE Memory
CCLK
(OUTPUT)
8
DIN
DOUT
(OUTPUT)
X8256_01
Figure 2: Master Serial Mode. The one-time-programmable SPROM supports automatic loading of configuration
programs. Multiple devices can be cascaded to support additional FPGAs. An early DONE inhibits the SPROM data output
one CCLK cycle before the FPGA I/Os become active.

R
Standby Mode Programming

The SPROM enters a low-power standby mode whenever The devices can be programmed on programmers supplied
CE is asserted High. The output remains in a high imped- by Xilinx or qualified third-party vendors. The user must
ance state regardless of the state of the OE input. ensure that the appropriate programming algorithm and the
Table 1: Truth Table for XC1700 Control Inputs

Internal Address
RESET CE DATA CEO Icc
Inactive Low if address < TC: increment active High active
if address > TC: don’t change 3-state Low reduced
Active Low Held reset 3-state High active
Inactive High Not changing 3-state High standby
Active High Held reset 3-state High standby
Notes: 1. The XC1700 RESET input has programmable polarity
Important: Always tie the VPP pin to VCC in your application. Never leave VPP floating.

R
XC1736E, XC1765E, XC17128E and XC17256E

DC Characteristics Over Operating Condition 8

Commercial

Industrial
CIN Input Capacitance (VIN = GND, f = 1.0MHz) 10.0 pF
COUT Output Capacitance (VIN = GND, f = 1.0MHz) 10.0 pF

R
XC1765EL, XC17128EL and XC17256EL

reliability.
Industrial Supply voltage relative to GND (TA = -40°C to +85°C ) 3.0 3.6 V


R
CE
9 10
9 TSCE TSCE THCE
RESET/OE
11 THOE
TLC 8 THC 6 TCYC
7
CLK
3 4
2 TCE
DATA
4 TOH
X2634
XC1736E XC17128E XC17128EL

XC1765EL
Symbol Description XC1765E XC17256E XC17256EL Units
1 TOE OE to Data Delay 45 45 25 30 ns
2 TCE CE to Data Delay 60 60 45 45 ns
3 TCAC CLK to Data Delay 80 200 45 45 ns
4 TOH Data Hold From CE, OE, or CLK3 0 0 0 0 ns 8
5 TDF CE or OE to Data Float Delay2 & 3 50 50 50 50 ns
6 TCYC Clock Periods 100 400 67 67 ns
3
7 TLC CLK Low Time 50 100 20 25 ns
8 THC CLK High Time3 50 100 20 25 ns
CE Setup Time to CLK (to guarantee
9 TSCE 25 40 20 25 ns
proper counting)
CE Hold Time to CLK (to guarantee
10 THCE 0 0 0 0 ns
proper counting)
OE Hold Time (guarantees counters
11 THOE 100 100 20 25 ns
are reset)

R
AC Characteristics Over Operating Condition When Cascading
RESET/OE
CE
CLK
12 TCDF
DATA Last Bit First Bit
13 TOCK 15 TOOE
CEO
14 TOCE 14 TOCE
X3183

2, 3
12 TCDF CLK to Data Float Delay 50 ns
3
13 TOCK CLK to CEO Delay 30 ns
14 TOCE CE to CEO Delay3 35 ns
15 TOOE RESET/OE to CEO Delay3 30 ns


R
XC17256E VO8 C

XC1736E C = Commercial (TA = 0°C to +70°C)
XC1765E Package Type I = Industrial (TA = –40°C to +85°C)
XC1765EL
XC17128E
XC17128EL
VO8 = 8-Pin Plastic Small-Outline Thin Package
XC17256E
PC20 = 20-Pin Plastic Leaded Chip Carrier
XC17256EL

XC17128EPD8C XC17256EPD8C XC1736EPD8C XC1765EPD8C XC1701PD8C XC1702LVQ44C
XC17128EVO8C XC17256EVO8C XC1736ESO8C XC1765ESO8C XC1701PC20C XC1702LPC44C
XC17128EPC20C XC17256EPC20C XC1736EVO8C XC1765EVO8C XC1701SO20C XC1704LVQ44C
XC17128EPD8I XC17256EPD8I XC1736EPC20C XC1765EPC20C XC1701PD8I XC1704LPC44C
XC17128EVO8I XC17256EVO8I XC1736EPD8I XC1765EPD8I XC1701PC20I XC1702LVQ44I
XC17128EPC20I XC17256EPC20I XC1736ESO8I XC1765ESO8I XC1701SO20I XC1702LPC44I
XC1736EVO8I XC1765EVO8I XC1704LVQ44I
XC1736EPC20I XC1765EPC20I XC1704LPC44I
XC17128ELPD8C XC17256ELPD8C XC1765ELPD8C XC1701LPD8C XC17512LPD8C

XC17128ELVO8C XC17256ELVO8C XC1765ELSO8C XC1701LPC20C XC17512LPC20C
XC17128ELPC20C XC17256ELPC20C XC1765ELVO8C XC1701LSO20C XC17512LSO20C
XC17128ELPD8I XC17256ELPD8I XC1765ELPC20C XC1701LPD8I XC17512LPD8I
XC17128ELVO8I XC17256ELVO8I XC1765ELPD8I XC1701LPC20I XC17512LPC20I 8
XC17128ELPC20I XC17256ELPC20I XC1765ELSO8I XC1701LSO20I XC17512LSO20I
XC1765ELVO8I XQ1701LCC44M
XC1765ELPC20I XQ1701LCC44B
XQ1701LS020N
Note: The XC1701, XC1701L, XQ1701L, XC1702L, XC1704L and XC17512L products are specified in the XC1700L High Density
datasheet.
Marking Information
Due to the small size of the serial PROM package, the complete ordering part number cannot be marked on the package.
The XC prefix is deleted and the package code is simplified. Device marking is as follows.
17256E V C

XC1736E C = Commercial (TA = 0°C to +70°C)
XC1765E Package Type I = Industrial (TA = –40°C to +85°C)
XC1765X P = 8-Pin Plastic DIP
XC17128E S = 8-Pin Plastic Small-Outline Package
XC17128X V = 8-Pin Plastic Small-Outline Thin Package
XC17256E J = 20-Pin Plastic Leaded Chip Carrier
XC17256X
Note: When marking the device number on the EL parts, an X is used in place of an EL.

R
Revision Control
Date Revision
7/14/98 Revised ICCS on page 21 and page 22; revised VCC specifications for TA on page 21 and page 22; revised
VCC on page 21; revised Note 2 on page 22 and page 23; added TA to operating range specifications on
page 25.
9/8/98 Revised the references to FPGAs to include the XC4000XLA and XC4000XV families.
9/30/98 Updated the Valid Ordering Combinations on page 25 to include high density products.
12/7/98 Updated the references to compatible FPGAs to include the Virtex family.

0
R
Spartan and SpartanXL Families of
Serial Configuration PROMs
September 8, 1998 (Version 1.2) 0 8* Product Specification
Introduction Spartan SPROM Features

The Spartan™ family of Serial Configuration PROMs • Serial Configuration one-time programmable (OTP)
(SPROM) provides and easy-to-use, cost-effective method read-only memory designed to store configuration
for storing Spartan device configuration bitstreams. bitstreams of Spartan FPGA devices
• Simple interface to the Spartan device requires only
When the Spartan device is in Master Serial mode, it gen-
one user I/O pin
erates a configuration clock that drives the Spartan
SPROM. A short access time after the rising clock edge,
• Low-power CMOS floating gate process
data appears on the SPROM DATA output pin that is con-
• Available in 5 V and 3.3 V versions
nected to the Spartan device DIN pin. The Spartan device
• Available in compact plastic 8-pin DIP, 8-pin VOIC, or
generates the appropriate number of clock pulses to com-
20-pin SOIC (XC17S40 only) packages.
plete the configuration. Once configured, it disables the
• Programming support by leading programmer
SPROM. When a Spartan device is in Slave Serial mode,
manufacturers.
the SPROM and the Spartan device must both be clocked
• Design support using the Xilinx Alliance and
by an incoming signal.
Foundation series software packages.
For device programming, either the Xilinx Alliance or the
Foundation series development systems compiles the
Spartan device design file into a standard HEX format
which is then transferred to most commercial PROM pro-
grammers.
Spartan FPGA Configuration Bits Compatible Spartan SPROM

XCS05 53,984 XC17S05
XCS05XL 54,544 XC17S05XL
XCS10 95,008 XC17S10
XCS20 178,144 XC17S20
XCS30 247,968 XC17S30
XCS40 329,312 XC17S40

R
Spartan and SpartanXL Families of Serial Configuration
Pin Description
Table 1: Spartan PROM Pinouts
8-Pin 20-Pin
Pin Name Pin Description
PDIP & VOIC SOIC
DATA 1 1 Data output, 3-stated when either CE or OE are inactive. During pro-
gramming, the DATA pin is I/O. Note that OE can be programmed to
be either active High or active Low.
CLK 2 3 Each rising edge on the CLK input increments the internal address
counter, if both CE and OE are active.
When High, this input holds the address counter reset and 3-states
the DATA output. The polarity of this input pin is programmable as ei-
ther RESET/OE or OE/RESET. To avoid confusion, this document
describes the pin as RESET/OE, although the opposite polarity is
possible on all devices. When RESET is active, the address counter
is held at zero, and the DATA output is 3-stated. The polarity of this
RESET/OE
3 8 input is programmable. The default is active High RESET, but the
(OE/RESET)
preferred option is active Low RESET, because it can be driven by
the FPGA’s INIT pin.
The polarity of this pin is controlled in the programmer interface. This
input pin is easily inverted using the Xilinx HW-130 programmer soft-
ware. Third-party programmers have different methods to invert this
pin.
CE 4 10 When High, this pin disables the internal address counter, 3-states
the DATA output, and forces the device into low-ICC standby mode.
GND 5 11 GND is the ground connection.
VCC 7, 8 18, 20 The VCC pins are to be connected to the positive voltage supply.

R
Controlling Serial PROMs Programming the FPGA With Counters

Connecting the Spartan device with the SPROM: Unchanged Upon Completion
• The DATA output of the SPROM drives the DIN input of When multiple-configurations for a single Spartan device
the lead Spartan device. are stored in a Serial Configuration PROM, the OE pin
• The Master Spartan device CCLK output drives the should be tied Low. Upon power-up, the internal address
CLK input of the SPROM. counters are reset and configuration begins with the first
• The RESET/OE input of the SPROM is driven by the program stored in memory. Since the OE pin is held Low,
INIT output of the Spartan device. This connection the address counters are left unchanged after configuration
assures that the SPROM address counter is reset is complete. Therefore, to reprogram the FPGA with
before the start of any (re)configuration, even when a another program, the DONE line is pulled Low and config-
reconfiguration is initiated by a VCC glitch. Other uration begins at the last value of the address counters.
methods – such as driving RESET/OE from LDC or This method fails if a user applies RESET during the Spar-
system reset – assume that the SPROM internal tan device configuration process. The Spartan device
power-on-reset is always in step with the FPGA’s aborts the configuration and then restarts a new configura-
internal power-on-reset, which may not be a safe tion, as intended, but the SPROM does not reset its
assumption. address counter, since it never saw a High level on its OE
• The CE input of the SPROM is driven by the DONE input. The new configuration, therefore, reads the remain-
output of the Spartan device, provided that DONE is not ing data in the PROM and interprets it as preamble, length
permanently grounded. Otherwise, LDC can be used to count etc. Since the Spartan device is the Master, it issues
drive CE, but must then be unconditionally High during the necessary number of CCLK pulses, up to 16 million
user operation. CE can also be permanently tied Low, (224) and DONE goes High. However, the Spartan device
but this keeps the DATA output active and causes an configuration will be completely wrong, with potential con-
unnecessary supply current of 10 mA maximum. tentions inside the Spartan device and on its output pins.
This method must, therefore, never be used when there is
FPGA Master Serial Mode Summary any chance of external reset during configuration.
The I/O and logic functions of the Configurable Logic Block
(CLB) and their associated interconnections are estab-
lished by a configuration program. The program is loaded 8
either automatically upon power up, or on command,
depending on the state of the Spartan device MODE pin. In
Master Serial mode, the Spartan device automatically
loads the configuration program from an external memory.
The Spartan SPROM has been designed for compatibility
with the Master Serial mode.
Upon power-up or reconfiguration, the Spartan device
enters the Master Serial mode when the MODE pin is Low.
Data is read from the Serial Configuration PROM sequen-
tially on a single data line. Synchronization is provided by
the rising edge of the temporary signal CCLK, which is gen-
erated during configuration.
Master Serial mode provides a simple configuration inter-
face. Only a serial data line and two control lines are
required to configure the Spartan device. Data from the
Serial Configuration PROM is read sequentially, accessed
via the internal address and bit counters which are incre-
mented on every valid rising edge of CCLK.
If the user-programmable, dual-function DIN pin on the
Spartan device is used only for configuration, it must still be
held at a defined level during normal operation. The Spar-
tan family takes care of this automatically with an on-chip
default pull-up resistor.

R
Vcc
SPARTAN
FPGA
MODE
Vcc
VCC
DIN DATA
CCLK CLK SPARTAN
DONE CE SPROM
RESET RESET INIT OE/RESET
CCLK
(OUTPUT)
DIN
DOUT
(OUTPUT)
S8472_02
Figure 1: Master Serial Mode. The one-time-programmable Spartan SPROM supports automatic loading of
configuration programs. An early DONE inhibits the PROM data output one CCLK cycle before the Spartan FPGA I/Os
become active.

R
Standby Mode Programming the Spartan Family

The SPROM enters a low-power standby mode whenever Serial PROMs
CE is asserted High. The output remains in a high imped- The devices can be programmed on programmers supplied
ance state regardless of the state of the OE input. by Xilinx or qualified third-party vendors. The user must
ensure that the appropriate programming algorithm and the
VCC GND
RESET/
OE CE
or
OE/
RESET
Address Counter
CLK TC
EPROM OE
Output
Cell DATA
Matrix
X8473_01
8

Important: Always tie the two VCC pins together in your application.
Table 2: Truth Table for XC17S00 Control Inputs

Internal Address
RESET CE DATA Icc
Inactive Low if address < TC: increment active active
if address > TC: don’t change 3-state reduced
Active Low Held reset 3-state active
Inactive High Not changing 3-state standby
Active High Held reset 3-state standby
Notes: 1. The XC17S00 RESET input has programmable polarity


R
XC17S05, XC17S10, XC17S20, XC17S30, XC17S40

Note: During normal read operation both VCC pins must be connected together.

VOH High-level output voltage (IOH = -4 mA) Commercial 3.86 V
VOH High-level output voltage (IOH = -4 mA) Industrial 3.76 V
ICCS Supply current, standby mode XC17S05, XC17S10, 50.0 µA

XC17S20, XC17S30
XC17S40 100.0 µA

R
XC17S05XL, XC17S10XL, XC17S20XL, XC17S30XL, XC17S40XL

reliability.
Note: During normal read operation both VCC pins must be connected together.
DC Characteristics Over Operating Condition 8


R

CE
9 10
9 TSCE TSCE THCE
RESET/OE
11 THOE
TLC 8 THC 6 TCYC
7
CLK
3 4
2 TCE
DATA
4 TOH
X2634

1 TOE OE to Data Delay 45 ns
2 TCE CE to Data Delay 60 ns
3 TCAC CLK to Data Delay 80 ns
4 TOH Data Hold From CE, OE, or CLK3 0 ns
5 TDF CE or OE to Data Float Delay2 & 3 50 ns
6 TCYC Clock Periods (TCCLK on FPGA) 100 ns
3
7 TLC CLK Low Time 50 ns
8 THC CLK High Time3 50 ns
9 TSCE CE Setup Time to CLK (to guarantee proper 25 ns
counting)
10 THCE CE Hold Time to CLK (to guarantee proper 0 ns
counting)
11 THOE OE Hold Time (guarantees counters are reset) 25 ns

R
XC17S20XL VO8 C

XC17S05 C = Commercial (TA = 0°C to +70°C)
XC17S05XL Package Type I = Industrial (TA = –40°C to +85°C)
XC17S10
XC17S10XL
VO8 = 8-Pin Plastic Small-Outline Thin Package
XC17S20
XC17S20XL
XC17S30
XC17S30XL
XC17S40
XC17S40XL

XC17S05PD8C XC17S10PD8C XC17S20PD8C XC17S30PD8C XC17S40PDC
XC17S05VO8C XC17S10VO8C XC17S20VO8C XC17S30VO8C XC17S40SO20C
XC17S05PD8I XC17S10PD8I XC17S20PD8I XC17S30PD8I XC17S40PD8I
XC17S05VO8I XC17S10VO8I XC17S20VO8I XC17S30VO8I XC17S40SO20I
XC17S05XLPD8C XC17S10XLPD8C XC17S20XLPD8C XC17S30XLPD8C XC17S40XLPD8C
XC17S05XLVO8C XC17S10XLVO8C XC17S20XLVO8C XC17S30XLVO8C XC17S40XLSO20C
XC17S05XLPD8I XC17S10XLPD8I XC17S20XLPD8I XC17S30XLPD8I XC17S40XLPD8I
XC17S05XLVO8I XC17S10XLVO8I XC17S20XLVO8I XC17S30XLVO8I XC17S40XLSO20I
8
Marking Information
Due to the small size of the serial PROM package, the complete ordering part number cannot be marked on the package.
The XC prefix is deleted and the package code is simplified. Device marking is as follows.
17S20L V C

XC17S05 C = Commercial (TA = 0°C to +70°C)
XC17S05L Package Type I = Industrial (TA = –40°C to +85°C)
XC17S10 P = 8-Pin Plastic DIP
XC17S10L V = 8-Pin Plastic Small-Outline Thin Package
XC17S20 S = 20-Pin Plastic Small-Outline Package
XC17S20L
XC17S30
XC17S30L
XC17S40
XC17S40L
Note: When marking the device number on the XL parts, an L is used in place of an XL.
Revision Control
Date Revision
7/14/98 Cosmetic edits for pages 27, 28, 29 & 30.
9/8/98 Clarified the SPARTAN FPGA & PROM interface by removing references to CEO pin. Removed the ESD
notation in Absolute Maximum table since it is now included in Xilinx’s Reliability Monitor Report.

R

0
R
HW-130 Programmer
Device and Package Support Programming Socket Adapters

• XC1700 Serial PROMs • Supports all package styles: PLCC, PQFP, TQFP,
• XC9500 CPLDs VQFP, HQFP, BGA, SOIC, VOIC, CSP and DIP.
• Supports all Xilinx package types • Purchase as needed for specific package support.
Programmer Accessories Electrical Requirements and Physical

• Universal power supply Specifications
• Power cord options for US/Asia, UK, European and • Operating voltage: 100-250 VAC, 50-60 Hz
Japanese standards. • Power consumption: 0.5 Amp
• Serial download cable and adapters • Dimensions: 6” x 8” x 2” programmer, 3” x 5” x 1.75”
• Users manual power supply
• Programmer interface software • Weight: 1.2 lb. programmer, 1.3 lb. power supply
• Vacuum handling tool • Safety standards: approved by UL, CSA, TUV, CE
Interface Software and System New Programming Algorithm Support

Requirements The new programmer algorithms are available via the Xilinx
The programmer software operates on a variety of different WEB site:
platforms. Table 1 indicates the minimum system require-
• To access programmer software from the Xilinx WEB
ments for each. In all cases, a CD-ROM drive and an
site, go to www.xilinx.com. The URL for the index of
RS-232 serial port are required. A mouse is recommended.
HW-130 software can be found at:
Programmer Functional Specifications www.xilinx.com/techdocs/htm_index/sw_programmer.htm 8
• Device programming, erasing and verification
• CPLD security control
• Serial PROM reset polarity control
• Checksum calculation and comparison
• Blank check and signature ID tests
• Master device upload
• File transfer and comparison
• Self check and auto calibration
Table 1: Interface Software and System Requirements

Requirements Windows 95 Windows 98 Windows NT Sun OS Solaris HP9000/700 IBM RS6000
Memory Needed 8MB 16MB 16MB — — — —
Hard Disk Space 2MB 2MB 2MB 6MB 6MB 6MB 6MB
System Software 4.00 4.10 3.1 or greater SunOS SunOS 5.3 or greater, HP-UX A09.05 AIX 3.2.5 or
or greater or greater 4.1.3 (Solaris 2.3 or greater) or greater greater
or greater

R
HW-130 Programmer
Adapter Selection Table

Product Family Package Types Adapter P/N
XC9500/XL PLCC44 HW-133-PC44
XC9500 VQFP 44 HW-133-VQ44
XC9500/XL CSP 48 HW-133-CS48
XC9500XL VQFP 64 HW-133-VQ64
XC9500 PLCC 84 HW-133-PC84
XC9500 PQFP 100 HW-133-PQ100
XC9500/XL TQFP 100 HW-133-TQ100
XC9500XL CSP 144 HW-133-CS144
XC9500XL TQFP 144 HW-133-TQ144
XC95001 PQFP 160 HW-133-PQ160
XC9500/XL PQFP/HQFP 208 HW-133-HQ208
XC9500/XL BGA 352 HW-133-BG352
XC1700 & XC17S00 DIP 8 HW-137-DIP8

XC1700 & XC17S00 PLCC20/SO8/VO8 HW-137-PC20/SO8
XC1700 & XC17S00 SO20 HW-137-SO20
XC1700 CLCC 44/VQFP 44 HW-137-LCC44/VQ44
Calibration Adapter - HW-130-CAL

1) Xilinx has manufactured two versions of the HW-133-PQ160 adapter. The current and correct adapter for programming
XC9500 devices has “CPLD” written on the front label, at the top left side, under the Xilinx logo.
Revision Control
Date Revision
6/13/98 Added Windows 98 system requirements, added HW-137-LCC44/VQ44 to Adapter selection table.
11/27/98 Removed DOS and Windows 3.1 support. Revised the Physical Specifications to include the power
supply dimensions and weight. Removed the obsolete XC7200/7300 product families. Added pack-
age adapters for the XC9500/XL series in CSP 48, VQFP 64, CSP and TQFP 144 packages.

1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

Table of Contents
Xilinx HardWire™ Product Family Overview

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
Xilinx HardWire Technology Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
Advantages of the Xilinx HardWire Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
Re-verifying the Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
Fault Coverage and Test Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
Packaging and Silicon Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
Support for the Entire Product Life Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
HardWire Design/ Production Interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-3
Design Submittal Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-5
Summary of the Conversion Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-5
The HardWire Product Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-5
Xilinx HardWire Product Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-5
HardWire Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-6
1
R
Xilinx HardWire™ Product Family
Overview
February 2, 1998 (Version 2.2) 1 9*
Introduction Advantages of the Xilinx HardWire

HardWire software and silicon provide a simple, turnkey Methodology
path, used for reducing the cost of FPGA designs. When a Converting from an FPGA to a HardWire device has many
system incorporating Xilinx FPGA’s moves to high volume advantages over standard gate array redesign. The most
production, HardWire products should be the first consider- important advantage is HardWire devices are developed
ation for cost reduction. HardWire products are developed using a fully turnkey process. Customer engineering
specifically to be pin compatible replacements of Xilinx resources are required to review conversion reports or dis-
FPGAs. The HardWire conversion flow coupled with the cuss special features implemented in the FPGA design. No
HardWire test methodology provides the lowest risk path customer engineering resources are required to convert
for customers to achieve dramatic cost reductions. Using the programmable logic design into a fully tested, com-
Xilinx FPGA’s and HardWire technology provides the cus- pletely verified HardWire device. This ease of conversion is
tomer with a single source for systems, software and sili- available only from Xilinx. HardWire devices are developed
con. This combination provides the fastest method for using the actual physical database previously created and
prototype development and production of systems based verified in the process of developing the FPGA design. The
on leading edge programmable logic technology. Each HardWire conversion methodology preserves the place-
HardWire product family is developed to match the perfor- ment and routing attributes of CLB's used for logic in the
mance and features of specific Xilinx FPGAs. The latest original physical database file. If the design is mapped to a
additions to the family of HardWire products are XH3 and third party library at the schematic level for conversion to
XH3L. They are designed to provide significant cost reduc- another technology, the design must be verified and proto-
tion for Xilinx FPGAs including E, EX and XL technology. typed. Third party implementations will change the place-
ment and routing, thereby changing the design's
Xilinx HardWire Technology performance characteristics. This means the new device
Overview must be re-verified and re-tested in the system to be certain
HardWire Products are a family of state-of-the-art that the performance and functionality still meet the appli- 9
sea-of-gates .5µ and .35µ multi-mask ASIC devices. The cations requirements. A comparison of the activities
HardWire product families have been developed to match required to convert a HardWire versus a generic gate array
the performance and features of Xilinx FPGAs.
The HardWire flow is the easiest method for cost reducing Re-verifying the Design
an FPGA based system. The Xilinx “Design Once” method-
In conventional gate array conversion (re-design), the
ology offers Xilinx customer the advantages of developing
design must be re-verified after the schematic is translated
prototypes, building pre-production and initial production
or recaptured. The process of re-verifying a design is rigor-
volumes using Xilinx FPGAs. Once the design is stable and
ous and time consuming. Functional simulation vectors
cost reduction is critical; a customer can convert the FPGA
need to be created, and the device must be exhaustively
to a HardWire device developed especially to match the
simulated before and after place and route. A suitable test
features and performance of that FPGA.
methodology must be considered and implemented. All this
The HardWire development process is turnkey, and uses is usually done by the customer, at the customers' expense
few customer engineering resources. It allows production and risk.
quality HardWire prototypes to be produced in half the time
In contrast, no additional customer engineering effort is
of traditional gate arrays. HardWire devices provide a cost -
required when converting to a HardWire device. Customer
effective alternative to traditional gate arrays.
engineering is involved to discuss features implemented in
the FPGA design and review conversion reports. The Hard-
Wire design is self-verifying because the actual FPGA
database files are used for the conversion. This makes the
HardWire conversion process a fully turnkey conversion.

R
Working Xilinx FPGA Design
Generic Gate Array Xilinx Hardwire

• Convert netlist to G/A format • Design Check/Evaluation
• Logic changes for design compatibility
• Logic changes for pin compatibility
• Logic changes for configuration emulation
• Logic changes for Boundary Scan • Design Conversion
• Design Check
• Functional Simulation
• Place and Route • Review/Approve Conversion
Reports
• Back-Annotation
• Timing simulation and new models
• Test Vector generation • Create Custom Masks
• Create custom masks
Figure 2: Steps Involved in Converting a PLD Design to a Gate Array as Compared to a HardWire Device
Fault Coverage and Test Vectors Packaging and Silicon

All designs need to be testable. In a traditional gate array, Considerations
the designer is required to build in testability and generate HardWire devices are functionally compatible with Xilinx
test vectors to verify chip performance by exercising as FPGAs. They are also, whenever possible, physically and
much of the device circuitry as possible. Most designers mechanically the same. HardWire devices are manufac-
strive for greater than 90% fault coverage. However, they tured in the same fabrication facilities used by Xilinx for the
often settle for significantly less because the iterative pro- production of FPGA’s. This allows a significant reduction in
cess is time consuming and increases exponentially as the time and cost associated with qualifying HardWire
fault coverage is increased. A third party conversion from a devices.
Xilinx FPGA to a generic gate array or other similar technol-
ogy will require test vector generation. Typically, the original Converting from a Xilinx programmable logic device to any
designers create test vectors, since they are most familiar third party device means a change in silicon, packaging,
with the FPGA’s design. This method misuses valuable assembly and test. Each of these changes adds an ele-
design resources and reverses the value of the decision to ment of risk into the qualification process.
use programmable logic for their ease of design and
time-to-market advantage. Another method is to contract Support for the Entire Product Life
with the conversion or gate array vendor to create the test Cycle
vectors. This method is both expensive and time consum-
Figure 3 shows the typical life cycle of a high-volume prod-
ing. In some cases, conversion or gate array vendors will
uct. It illustrates the optimal way of using the programmable
accept a design without test vectors, but the customer and HardWire devices. During development, prototyping
accepts the liability of determining whether the resulting and initial production cycles, the programmable device is
device is production worthy. In today’s competitive market, the best choice. As the system moves into higher volume
most projects can not afford the risk of possible re-spins if production and no additional modifications are being made
the design doesn’t work.
to the design, a HardWire device can be used in place of
Converting from a Xilinx FPGA to a HardWire device the original programmable logic device.
requires no test vector generation by the customer. Hard-
Wire devices use a combination of industry standard and
Xilinx patented test generation methods to achieve the
most complete fault coverage possible. All HardWire
devices are tested using a full scan test methodology.

R
Programmable Logic Volume

HardWire FpgASIC Volume
Unplanned Upside
Production
Ramp-Up
V End-of-Life
O
L
U
M
E HardWire
Device
99012001
Figure 3: Typical High Volume Product Life Cycle
Since the HardWire device and the programmable logic HardWire Design/ Production
device are functionally and physically compatible, produc-
tion can be switched back to the programmable device if
Interface
the situation warrants. For example, if the demand for the Figure 4 illustrates how the design, development and pro-
customer’s product increases dramatically, production can duction activities for HardWire devices are sequenced in
be increased immediately by fulfilling the additional comparison to gate arrays. Using the Xilinx “Design Once”
demand with programmable devices. The change can be methodology, limited customer activity is needed to develop
made immediately since there is virtually no lead-time for the HardWire. All HardWire devices are developed using
an off-the-shelf programmable device. Production can also the FPGA actual design database. Using the FPGA design
be switched to the programmable device as the product database in conjunction with other documentation supplied
ends its life cycle and volume decreases. This eliminates by the customer reduces the amount of time required to
the need for end-of-life buys and the risk of obsolescence. develop the HardWire device. It allows Xilinx to perform a
simple design check procedure prior to generating the 9
Furthermore, designs implemented with multiple program-
HardWire device. The design check along with the custom-
mable devices can be cost reduced incrementally, convert-
ers conversion report review reduces the risk of errors.
ing one or more of the programmable devices to a
After the design check is complete the HardWire proto-
HardWire device with the balance remaining as FPGAs. As
types can be manufactured. The customer then performs
each FPGA is converted to a HardWire device, the user
in-system verification of the prototypes. Once this verifica-
benefits by having a lower price for that device. This also
tion is complete the HardWire device can be released to
allows the user to maintain the ease-of-use of off-the-shelf
production. Since the functionality of the FPGA and Hard-
programmable logic in the other sockets. When all of the
Wire are compatible, few customer engineering resources
devices are converted, the storage element (PROM) can
are needed to move from the programmable to the Hard-
be removed, giving even further cost reductions. This flexi-
Wire or vice versa. By comparison, using a traditional gate
bility is unique to Xilinx, and allows customers to achieve
array to reproduce functions implemented in the FPGA
cost reduction quickly with minimal effort.
would require extensive simulation and test development.

R
Customer Xilinx Xilinx

Design Development Manufacturing
Functions Interface Functions
Design Concept
Production
FPGA
Design Considerations
and Verification
Foundation
or
Alliance
Programmable FPGA
• Schematic Capture
• Design Entry
• Simulation
Programmable
• Logic Simulation FPGA Flow
• Design Libraries
• Automatic Design
Xilinx Netlist Format Implementation
Programmable FPGA
• Design
Implementation
Array and Bitstream

Production
EPROM/PROM
Programming
Initial Production with Inventory of

Programmable FPGA Programmable
Logic Devices
High Volume
Production Achieved
HardWire Purchase Order

Design
• DRC Analysis/Report HardWire
Initial Design Development Flow
Review Submittal
.MBO
.BIT Xilinx Design
.RBT Conversion Automatic
.NCD Test Generation
Final HardWire Design File
Design Verification HardWire Design Verification
NRE Invoiced Custom
Masks Made
Prototypes Delivered HardWire

Prototypes
System Verification
HardWire Protoptye Built
Approval Signoff
HardWire Volume Units HardWire

Production Production Builds
99012000
Figure 4: Programmable/HardWire Design/Production Interface

R
Design Submittal Process the report has been reviewed and the customer is satisfied,
conversion begins. At the completion of the conversion the
HardWire development time will vary with the addition of final timing report along with the Design Verification Form
features such as RAM, Configuration Emulation and JTAG. (DVF) is sent to the customer. Once the DVF is completed
Once the complete design submittal kit is received the the HardWire files are sent to the mask shop for prototyp-
HardWire conversion will begin. A complete design submit- ing. If any custom markings are required they must be sub-
tal kit contains the following: mitted to Xilinx with the Design Verification Form (DVF).
1. Files: A complete list of required files can be found in the Prototypes are produced, tested and shipped to the cus-
HardWire Design Submittal workbook. tomer for in-system testing. The customer signs the proto-
type approval form and returns it to Xilinx. Production can
2. Hard copy of a board level schematic showing how the
begin.
FPGA interfaces with other components on the board (if
possible).
The HardWire Product Families
3. A detailed explanation of any special requirements for
Each HardWire product family is developed to support the
the conversion.
features, density and performance of a specific generation
4. A design submittal form and NRE PO. of Xilinx FPGA’s. The initial family of HardWire devices
All forms can be found in the HardWire Design Submittal (XC23xx, XC33xx and XC43xx) supports XC2xxx, XC3xxx
workbook and on the Xilinx web page under HardWire and XC4xxx (no E features) FPGAs. The second genera-
products. tion of HardWire devices (XC5400, XC4400 and
XC4400XL) support XC52xx, XC4xxxE, XC4xxxEX and
Summary of the Conversion XC4xxxXL FPGAs. These technologies are used to sup-
port production requirements only. New designs will be tar-
Process geted to more current HardWire technology. For new
The HardWire conversion process is the simplest way to designs developed using Xilinx XC4xxxXL family of
cost reduce systems designed using FPGAs. The cus- FPGAs, the XH3L HardWire family provides the most effec-
tomer is involved in tracking and approving milestones. Xil- tive technology, cost and performance. For customers
inx handles the day-to-day activities of converting the using fast, dense Xilinx XC4xxxE, EX FPGA’s the XH3
design to a HardWire device. Once Xilinx receives a com- product family provides the most efficient and cost effective
plete design submittal kit the conversion process begins. solution available. All HardWire devices support commer-
Xilinx first reviews the design to determine any items that cial and industrial temperature ranges. See Table 2 for
could impact the performance of the HardWire device. A product family details.
conversion evaluation report is sent to the customer. After 9
Table 2: HardWire Product Families
FPGA Device Speed Features

HardWire Family Voltage
Family Grade Supported
XC40xxXL All XL XH3xxL 3.3V
XC40xxEX All E, EX XH3xx 5V
XC40xxXL All XL XC44xxXL 3.3V
XC40xx All E, EX XC44xx 5V
XC52xx All All XC54xx 5V
XC4xxx All No E features XC43xx 5V
XC3xxx All All XC33xx 5V, 3.3V
XC2xxx All All XC23xx 5V, 3.3V
Xilinx HardWire Product Descriptions

XH3 and XH3L Product Descriptions RAM. The control logic for Configuration Emulation, Power
on Reset (POR), Oscillators and full JTAG are built into the
The XH3 and XH3L HardWire product families were devel-
XH3 and XH3L base arrays. RAM blocks are incorporated
oped to match the performance, density and features of Xil-
with maximum efficiency. Both XH3 and XH3L use the
inx XC4xxxE, XC4xxxEX and XC4xxxXL family of FPGA’s. dense sea-of-gates CMOS CBA technology, from Xilinx
These HardWire FpgASIC product families support the fea- partner SiARC, a division of Synopsys. The .5m (XH3) and
tures of Xilinx second generation FPGAs. This includes .35m (XH3L) process geometry used are small enough that
speed grades, Configuration Emulation (CE), JTAG and die sizes are driven by pad count and not gate count,

R
except for FPGAs designs with high RAM content. The Xil- - Multiple Masks, state-of-the-art .5µ process
inx patented DesignLock turnkey conversion methodology technology.
is used to develop all XH3 and XH3L devices. XH3 and - Pad counts and gate counts available for the densest
XH3L provide the most cost-effective method for converting FPGA devices.
XC4xxxE, XC4xxxEX and XC4xxxXL FPGA’s to low cost - On chip scan path test latches.
HardWire FpgASICs. - Fully pin-for-pin compatible with FPGA.
• Benefits
XH3L Summary - All Xilinx FPGA features supported, including CE,
• Features JTAG and Select RAM.
- Designed for conversion of XC4xxxXL 3.3v FPGAs. - Patented, turnkey conversion flow.
- Only used CLB's are mapped. - Pads, package and FPGA design content required
- Multiple Masks, state-of-the-art .35µ process determine device used.
technology. - No customer developed test vectors needed.
- On chip scan path test latches. - Drop in replacement for Xilinx FPGAs.
- Fully pin for pin compatible with FPGA
- Smallest possible die size. HardWire Summary
• Benefits Xilinx Hardwire products are a family of devices ranging
- All Xilinx FPGA features supported, including CE, from .5µ to .35µ state-of-the-art sea-of-gates multi-mask
JTAG and Select RAM. ASIC devices. The HardWire flow is the simplest method of
- Smallest possible die size used to achieve the lowest cost reduction for FPGA based systems. They are devel-
possible cost. oped using the FPGA's design files. This guarantees the
- Technology feature size matched to performance HardWire device will be functionally compatible with the
requirements. FPGA. No customer generated test vectors are required
- No customer developed test vectors needed. with HardWire. Each HardWire device is tested using a
- Drop in replacement for Xilinx FPGAs. combination of industry standard and Xilinx patented test
methods in a full scan methodology. HardWire prototypes
XH3 Summary can be developed in half the time of traditional gate array
• Features prototypes. HardWire process technologies, conversion
- Designed for conversion of high density XC4xxxE methods and testing procedures provide the most cost -
and EX 5V FPGAs. effective alternative to traditional gate arrays.
- Xilinx FPGA features built in to the base array.

High-Reliability and
QML Military Products
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

High-Reliability and QML Military
Products
Table of Contents
QPRO™ QML Certified and Radiation Hardened Products

Unmatched Hi-Rel Product Offering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Radiation Hardened Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Committed to the Hi-Rel Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Xilinx Hi-Rel and Radiation Hardened Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
QPROTM XQ4000E/EX QML High Reliability Field Programmable Gate Arrays

XQ4000E/EX High-Reliability Features. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-9
More Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-9
XQ4000E/EX Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-10
XQ4028EX Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-25
XQ4028EX Longline and Wide Decoder Timing Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-27
XQ4028EX Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-34
XQ4028EX Pin-to-Pin Input Parameter Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-35
CB191/196 Package for XQ4010E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-39
QPRO™ XQ4000XL Series QML High-Reliability FPGAs

XQ4000X Series Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-51
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-51
XQ4000XL Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-52
XQ4000XL Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-59
XQ4000XL Pin-to-Pin Input Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-61
Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-66
QPRO™ XQR4000XL Radiation Hardened Field Programmable Gate Arrays

XQR4000XL Series Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-71
Radiation Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-72
XQR4000XL Switching Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-72
XQR4000XL Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-79
XQR4000XL Pin-to-Pin Input Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-81
Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-86
0
R
QPRO™ QML Certified and
Radiation Hardened Products
Xilinx is the leading supplier of high reliability programma- full MIL-PRF-38535 QML/SMD versions are available as
ble logic devices to the aerospace and defense markets. standard, off-the-shelf products, in through-hole and sur-
These devices are used in a wide variety of applications face mount hermetic packages and plastic packages.
such as electronic warfare, missile guidance and targeting,
RADAR, SONAR, communications, signal processing, Radiation Hardened Products
aerospace, and avionics. The Xilinx QPRO family of Xilinx also offers a family of radiation hardened products for
ceramic and plastic QML products (Qualified Manufactur- space and satellite applications. WIth densities up to
ers Listing), certified to MIL-PRF-38535, provide you with 130,000 systems gates, these devices have guaranteed
advanced programmable logic solutions for next genera-
total ionizing dose of 60K Rad(Si), are latch-up immune
tion designs.
and have low soft upset rates. They are guaranteed to
The Xilinx QPRO family addresses the issues that are crit- meet full electrical specifications over -55°C to +125°C.
ical to the aerospace and defense market:
• QML/Best Commercial Practice- Commercial
Committed to the Hi-Rel Market
manufacturing strengths result in more effective Xilinx understands that you need to be able to count on
process flows. your Hi-Rel supplier. Xilinx is committed to our customers
• Performance-based Solution - Cost effective plastic and the Hi-Rel market for the long-term, and we are contin-
hermetic packages. ually expanding our Hi-Rel support and product portfolio.
• Reliability of Supply - Controlled mask sets and The unique capabilities of the Xilinx FPGA solution provide
processes insure that you receive quality devices, every increased design flexibility, field-upgradability and system
time, without variation, which remain in production over feature integration, while eliminating the NREs, lead-time
the long term. and inventory problems of custom logic and gate arrays.
• Off-the-Shelf ASIC Solutions - Devices are readily Now more than ever, Xilinx is your Hi-Rel logic solution.
available.
Xilinx Hi-Rel and Radiation
Unmatched Hi-Rel Product Offering Hardened Products
Xilinx offers a wide variety of devices, delivering the fastest Table 1 summarizes Xilinx high density, high performance
and biggest Hi-Rel devices available. Products up to 10
and radiation hardened product offerings. The following
130,000 system gates are available today, with even higher pages contain a complete listing of current Xilinx
densities to come. Xilinx offers multiple product families to QML/SMD (Standard Microcircuit Drawings) devices and
allow you to select the right device to meet your design “B” grade equivalents. Architectural descriptions for these
requirements. FPGA products can be found in Chapter 4. For additional
This broad range of devices is available in a wide variety of information, contact the nearest Xilinx Sales Office or Sales
speed and package options. Both military temperature and Representative.

R
Table 1: High Density High Performance and Radiation Hardened Products
Family Devices Features

XQ4000/E/EX XQ4003A Highest Density/Most Features Family
XQ4005/E • 3,000-28,000+ gates
XQ4010/E • Up to 256 user-definable I/Os
XQ4013/E • Extensive system features include on-chip user RAM, built-in 1149.1
XQ4025E test support and fast carry logic
XQ4028EX
XQ4000XL XQ4013XL Most Advanced Family
XQ4036XL • Up to 130,000 system gates
XQ4062XL • 3.3 V, 5V-compatible I/O
XQR4000XL XQR4013XL Radiation Hardened Family
XQR4036XL • Up to 130,000 System Gates
XQR4062XL • 60K Rads total dose, latchup immune
XC1700 Products (Serial PROMS)
XC1700 Products
SMD Number Speed Package Mark Loc
(Serial PROMS)
5962-9471701MPA XC1765DDD8B DD8 TOP
5962-9561701MPA XC17256DDD8B DD8 TOP
5962-9951401QXA XQ1701LB SO20 TOP
5962-9951401QYA XQ1701LB CC44 TOP
XC3000 Products1
SMD Number Equivalent “B” Grade P/N Speed Package Mark Loc
5962-8994801MXC XC3020-50PG84B2 -50 PG84 TOP
5962-8994802MXC XC3020-70PG84B2 -70 PG84 TOP
5962-8994803MXC XC3020-100PG84B -100 PG84 TOP
5962-8994801MNC XC3020-50CB100B2 -50 CB100 BASE
5962-8994802MNC XC3020-70CB100B2 -70 CB100 BASE
5962-8994803MNC XC3020-100CB100B -100 CB100 BASE
5962-8994801MMC XC3020-50CB100B2 -50 CB100 LID
5962-8994802MMC XC3020-70CB100B2 -70 CB100 LID
5962-8994803MMC XC3020-100CB100B -100 CB100 LID
5962-8971301MXC XC3042-50PG84B2 -50 PG84 TOP
5962-8971302MXC XC3042-70PG84B2 -70 PG84 TOP
5962-8971303MXC XC3042-100PG84B -100 PG84 TOP
5962-8971301MZC XC3042-50PG132B2 -50 PG132 TOP
5962-8971302MZC XC3042-70PG132B2 -70 PG132 TOP
5962-8971303MZC XC3042-100PG132B -100 PG132 TOP
5962-8971301M9C XC3042-50CB100B2 -50 CB100 BASE
5962-8971302M9C XC3042-70CB100B2 -70 CB100 BASE
5962-8971303M9C XC3042-100CB100B -100 CB100 BASE
5962-8971301MMC XC3042-50CB100B2 -50 CB100 LID
5962-8971302MMC XC3042-70CB100B2 -70 CB100 LID
5962-8971303MMC XC3042-100CB100B -100 CB100 LID

R
XC3000 Products1
5962-8982301MXC XC3090-50PG175B2 -50 PG175 TOP
5962-8982302MXC XC3090-70PG175B2 -70 PG175 TOP
5962-8982303MXC XC3090-100PG175B -100 PG175 TOP
5962-8982301MZC XC3090-50CB164B2 -50 CB164 BASE
5962-8982302MZC XC3090-70CB164B2 -70 CB164 BASE
5962-8982303MZC XC3090-100CB164B -100 CB164 BASE
5962-8982301MTC XC3090-50CB164B2 -50 CB164 LID
5962-8982302MTC XC3090-70CB164B2 -70 CB164 LID
5962-8982303MTC XC3090-100CB164B -100 CB164 LID
Note 1: All devices listed also available as military temperature only.
Note 2: Do not use for new designs (under obsolescence).
XC3100 Products
5962-9561001MXC XC3142A-5PG84B1 -5 PG84 TOP
5962-9561002MXC XC3142A-4PG84B1 -4 PG84 TOP
5962-9561001MUC XC3142A-5PG132B1 -5 PG132 TOP
5962-9561002MUC XC3142A-4PG132B1 -4 PG132 TOP
5962-9561001MYC XC3142A-5CB100B1 -5 CB100 BASE
5962-9561001MZC XC3142A-5CB100B1 -5 CB100 LID
5962-9561002MZC XC3142A-4CB100B1 -4 CB100 LID
5962-9561101MXC XC3190A-5PG175B1 -5 PG175 TOP
5962-9561102MXC XC3190A-4PG175B1 -4 PG175 TOP
5962-9561101MZC XC3190A-5CB164B1 -5 CB164 LID
5962-9561102MZC XC3190A-4CB164B1 -4 CB164 LID
10
5962-9561201MXC XC3195A-5PG175B1 -5 PG175 TOP
5962-9561202MXC XC3195A-4PG175B1 -4 PG175 TOP
5962-9561201MZC XC3195A-5CB164B1 -5 CB164 LID
5962-9561202MZC XC3195A-4CB164B1 -4 CB164 LID

R
XC4000 Products1
5962-9471201MXC XC4003A-10PG120B2 -10 PG120 TOP
5962-9471202MXC XC4003A-6PG120B2 -6 PG120 TOP
5962-9471201MZC XC4003A-10CB100B2 -10 CB100 LID
5962-9471202MZC XC4003A-6CB100B2 -6 CB100 LID
5962-9225201MXC XC4005-10PG156B2 -10 PG156 TOP
5962-9225202MXC XC4005-6PG156B2 -6 PG156 TOP
5962-9225203MXC XC4005-5PG156B -5 PG156 TOP
5962-9225201MYC XC4005-10CB164B2 -10 CB164 LID
5962-9225202MYC XC4005-6CB164B2 -6 CB164 LID
5962-9225203MYC XC4005-5CB164B -5 CB164 LID
5962-9225201MZC XC4005-10CB164B2 -10 CB164 BASE
5962-9225202MZC XC4005-6CB164B2 -6 CB164 BASE
5962-9225203MZC XC4005-5CB164B -5 CB164 BASE
5962-9230501MXC XC4010-10PG191B2 -10 PG191 TOP
5962-9230502MXC XC4010-6PG191B2 -6 PG191 TOP
5962-9230503MXC XC4010-5PG191B -5 PG191 TOP
5962-9230501MYC XC4010-10CB196B2 -10 CB196 BASE
5962-9230502MYC XC4010-6CB196B2 -6 CB196 BASE
5962-9230503MYC XC4010-5CB196B -5 CB196 BASE
5962-9230501MZC XC4010-10CB196B2 -10 CB196 LID
5962-9230502MZC XC4010-6CB196B2 -6 CB196 LID
5962-9230503MZC XC4010-5CB196B -5 CB196 LID
5962-9473001MXC XC4013-10PG223B2 -10 PG223 TOP
5962-9473002MXC XC4013-6PG223B -6 PG223 TOP
5962-9473001MYC XC4013-10CB228B2 -10 CB228 BASE
5962-9473002MYC XC4013-6CB228B -6 CB228 BASE
5962-9473001MZC XC4013-10CB228B2 -10 CB228 LID
5962-9473002MZC XC4013-6CB228B -6 CB228 LID

R
XC4000E Products1
5962-9752201QXC XC4005E-4PG156B -4 PG156 TOP
5962-9752201QYC XC4005E-4CB164B -4 CB164 BASE
5962-9752201QZC XC4005E-4CB164B -4 CB164 LID
5962-9752301QXC XC4010E-4PG191B -4 PG191 TOP
5962-9752301QZC XC4010E-4CB164B -4 CB196 LID
5962-9752401QXC XC4013E-4PG223B -4 PG223 TOP
5962-9752401QZC XC4013E-4CB228B -4 CB228 LID
5962-9752501QXC XC4025E-4PG299B -4 PG299 TOP
5962-9752501QZC XC4025E-4CB228B -4 CB228 LID
XQ4000EX Products1
5962-985901QTB XQ4028EX-4HG240N2 -4 HG240 TOP
5962-985901QUA XQ4028EX-4BG352N2 -4 BG352 TOP
5962-985901QXC XQ4028EX-4PG299B -4 PG299 TOP
5962-985901QYC XQ4028EX-4CB228B -4 CB228 BASE
5962-985901QZC XQ4028EX-44CB228B -4 CB228 LID
Note 2: Plastic package.
XQ4000XL Products1
10
5962-9851301QTB XQ4013XL-3PQ240N2 -3 PG240 TOP
5962-9851301QVA XQ4013XL-3BG256B2 -3 BG256 TOP
5962-9851301QXC XQ4013XL-3PG223B -3 PG223 TOP
5962-9851301QYC XQ4013XL-3CB228B -3 CB228 BASE
5962-9851301QZC XQ4013XL-3CB228B -3 CB288 LID
5962-9851001QTB XQ4036XL-3HQ240N2 -3 HQ240 TOP
5962-9851001QVA XQ4036XL-3BG352N2 -3 BG352 TOP
5962-9851001QYC XQ4036XL-3CB228B -3 PG228 BASE
5962-9851101QTB XQ4062XL-3HQ240N2 -3 HQ240 TOP
5962-9851101QVA XQ4062XL-3BG432N2 -3 BG432 TOP
5962-9851101QYC XQ4062XL-3CB228B -3 CB228 BASE
Note 2: Plastic Package

R
Revision History
Version Description
1/98 Version 1.1 -- High-Reliability and QML Military Products, correct erroneous information page 2
“XC3000 Products”, delete last page, table - “Mil-PRF-3853 QML, Xilinx M Grade and Plastic
Commercial Flows”
11/98 Version 1.2 - Added new products, corrected XC3000, XC4000 products.

1
QPRO™ XQ4000E/EX QML High-

Reliability Field Programmable
Gate Arrays Table of Contents
1 10*

Unmatched Hi-Rel Product Offering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Radiation Hardened Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Committed to the Hi-Rel Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Xilinx Hi-Rel and Radiation Hardened Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
QPRO™ XQ4000E/EX QML High-Reliability Field Programmable Gate

Arrays Table of Contents

XQ4000E/EX High-Reliability Features. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-11
More Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-11
XQ4000E/EX Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-12
XQ4000E/EX Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-12
XQ4000E/EX Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-12
XQ4000E/EX DC Characteristics Over Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . 10-13
XQ4000E Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-13
XQ4000E Horizontal Longline Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . 10-14
XQ4000E Wide Decoder Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 10-15
XQ4000E CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-16
XQ4000E CLB RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . . . . . . . . . 10-19
XQ4000E CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . 10-19
XQ4000E CLB Level-Sensitive RAM Switching Characteristic Guidelines. . . . . . . . . . . . . . . . . 10-20
XQ4000E CLB Level-Sensitive RAM Timing Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-21
XQ4000E Guaranteed Input and Output Parameters (Pin-to-Pin, TTL I/O) . . . . . . . . . . . . . . . . 10-22
XQ4000E IOB Input Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-23 10
XQ4000E IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-25
XQ4028EX Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-27
XQ4028EX Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-27
XQ4028EX Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-27
XQ4028EX DC Characteristics Over Recommended Operating Conditions . . . . . . . . . . . . . . . 10-28
XQ4028EX Longline and Wide Decoder Timing Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-29
XQ4028EX Wide Decoder Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . 10-29
XQ4028EX CLB Switching Characteristic Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-30
XQ4000EX CLB RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . . . . . . . . 10-33
XQ4000EX CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . 10-33
XQ4028EX CLB RAM Asynchronous (Level-Sensitive) Write and Read Operation Guidelines. 10-34
XQ4028EXX CLB RAM Asynchronous (Level-Sensitive) Timing Characteristics . . . . . . . . . . . 10-35
XQ4028EX Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-36
XQ4028EX Output MUX, Clock to Out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-36
XQ4028EX Output Level and Slew Rate Adjustments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-36
XQ4028EX Pin-to-Pin Input Parameter Guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-37
XQ4028EX Global Early Clock, Set-Up and Hold for IFF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-37
XQ4028EX Global Early Clock, Set-Up and Hold for FCL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-37
XQ4028EX Input Threshold Adjustments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-37
XQ4028EX IOB Input Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-38
10-7
XQ4028EX IOB Input Switching Characteristic Guidelines (Continued). . . . . . . . . . . . . . . . . . . 10-39

XQ4028EX IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-40
CB191/196 Package for XQ4010E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-41
Additional XQ4010E Package Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-43
Example Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-43
Additional XC4005E Package Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-45

Additional XC4010E Package Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-47
Example for SMD Part: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-51
Example for Military Temperature Only Part: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-51

XQ4000X Series Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-53
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-53
XQ4000XL Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-54
XQ4000XL DC Characteristics Over Recommended Operating Conditions . . . . . . . . . . . . . . . 10-55
XQ4000XL Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 10-56
XQ4000XL CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-57
XQ4000XL CLB RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . . . . . . . . 10-60
XQ4000XL CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . 10-60
XQ4000XL Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-61
XQ4000XL Pin-to-Pin Input Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-63
XQ4000XL Global Low Skew Clock, Set-Up and Hold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-63
XQ4000XL BUFGE #s 3, 4, 7, & 8 Global Early Clock, Set-up and Hold for IFF and FCL . . . . . 10-64
XQ4000XL BUFGE #s 1, 2, 5, & 6 Global Early Clock, Set-up and Hold for IFF and FCL . . . . . 10-65
XQ4000XL IOB Input Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-66
XQ4000XL IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-67
Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-68
CB228 Package for XQ4013XL/4036XL/4062XL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-68

XQR4000XL Series Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-73
Radiation Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-74
XQR4000XL Switching Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-74
XQR4000XL DC Characteristics Over Recommended Operating Conditions . . . . . . . . . . . . . . 10-75
XQR4000XL Global Buffer Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . 10-76
XQR4000XL CLB Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-77
XQR4000XL CLB RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . . . . . . . . . 10-80
10-8
XQR4000XL CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing . . . . . . . . . . . . 10-80
XQR4000XL Pin-to-Pin Output Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-81
XQR4000XL Pin-to-Pin Input Parameter Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-83
XQR4000XL Global Low Skew Clock, Set-Up and Hold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-83
XQR4000XL BUFGE #s 3, 4, 7, & 8 Global Early Clock, Set-up and Hold for IFF and FCL . . . 10-84
XQR4000XL BUFGE #s 1, 2, 5, & 6 Global Early Clock, Set-up and Hold for IFF and FCL . . . 10-85
XQR4000XL IOB Input Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-86
XQR4000XL IOB Output Switching Characteristic Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . 10-87
Pinouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-88
CB228 Package for XQR4013XL/4036XL/4062XL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-88
10
10-9
10-10
0
R
QPRO XQ4000E/EX QML High
TM
Reliability Field Programmable Gate Arrays
May 19, 1998 (Version 2.1) 0 10* Product Specification
XQ4000E/EX High-Reliability • Configured by Loading Binary File

- Unlimited reprogrammability
Features • Readback Capability
• Certified to MIL-PRF-38535, appendix A QML - Program verification
(Qualified Manufacturers Listing) - Internal node observability
• System featured Field-Programmable Gate Arrays • Backward Compatible with XC4000 Devices
- Select-RAMTM memory: on-chip ultra-fast RAM with • Development System runs on most common computer
- synchronous write option platforms
- dual-port RAM option - Interfaces to popular design environments
- Abundant flip-flops - Fully automatic mapping, placement and routing
- Flexible function generators - Interactive design editor for design optimization
- Dedicated high-speed carry logic • Available Speed Grades:
- Wide edge decoders on each edge - XQ4000E -3 for plastic packages only
- Hierarchy of interconnect lines - -4 for ceramic packages only
- Internal 3-state bus capability - XQ4028EX -4 for all packages
- 8 global low-skew clock or signal distribution
networks More Information
• System Performance beyond 60 MHz For more information refer to Xilinx XC4000E and
• Flexible Array Architecture XC4000X series Field Programmable Gate Arrays product
• Low Power Segmented Routing Architecture specification. This datasheet contains pinout tables for
• Systems-Oriented Features XQ4010E only. Refer to Xilinx 1998 Databook for pinout
- IEEE 1149.1-compatible boundary scan logic support
tables for other devices. (Pinouts for XQ4000E/EX are
- Individually programmable output slew rate
identical to XC4000E/EX.)
- Programmable input pull-up or pull-down resistors
- 12-mA sink current per XQ4000E/EX output
Table 1: XQ4000E/EX Field Programmable Gate Arrays

10
Max. Typical Max.
Logic Max. RAM Gate Range Number Decode
Gates Bits (Logic and CLB Total of Inputs Max.
Device (No RAM) (No Logic) RAM)* Matrix CLBs Flip-Flops per side User I/O Packages
XQ4005E 5,000 6,272 3,000 - 9,000 14 x 14 196 616 42 112 PG156,
CB164
XQ4010E 10,000 12,800 7,000 - 20,000 20 x 20 400 1,120 60 160 PG191,
CB196,
HQ208
XQ4013E 13,000 18,432 10,000 - 24 x 24 576 1,536 72 192 PG223,
30,000 CB228,
HQ240
XQ4025E 25,000 32,768 15,000 - 32 x 32 1,024 2,560 96 256 PG299,
45,000 CB228
XQ4028EX 28,000 32,768 18,000 - 32 x 32 1024 2560 96 256 PG299,
50,000 CB228,
HQ240,
BG352
Note: Max values of Typical Gate Range include 20-30% of CLBs used as RAM.
May 19, 1998 (Version 2.1) 10-11

QPROTM XQ4000E/EX QML High Reliability Field Program-
R
XQ4000E/EX Switching Characteristics

XQ4000E/EX Absolute Maximum Ratings
TJ Junction temperature Ceramic packages +150 °C
Note 1: Maximum DC overshoot or undershoot above Vcc or below GND must be limited to either 0.5 V or 10 mA, whichever is
easier to achieve. During transitions, the device pins may undershoot to -2.0 V or overshoot to Vcc + 2.0 V, provided this
over- or undershoot lasts less than 20 ns.
Note 2: Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are
XQ4000E/EX Recommended Operating Conditions

VCC Supply voltage relative to GND, TC = -55°C to +125°C Ceramic packages 4.5 5.5 V
VCC Supply voltage relative to GND, TJ = -55°C to +125°C Plastic packages 4.5 5.5 V
Note 1: At case temperatures above those listed as Recommended Operating Conditions, all delay parameters increase by 0.35%
per °C.
Note 2: Input and output Measurement thresholds are: 1.5V for TTL and 2.5V for CMOS.
Note 3: All specifications are subject to change without notice.
10-12 May 19, 1998 (Version 2.1)

R
XQ4000E/EX DC Characteristics Over Operating Conditions

VOL Low-level output voltage @ IOL = 12.0mA, VCC min (Note 1) TTL outputs 0.4 V
IRIN* Pad pull-up (when selected) @ VIN = 0V (sample tested) -0.02 -0.25 mA
IRLL* Horizontal Longline pull-up (when selected) @ logic Low 0.2 2.5 mA
Note 2: With no output current loads, no active input or Longline pull-up resistors, all package pins at Vcc or GND, and the FPGA
configured with the development system Tie option.
* Characterized Only.
XQ4000E Global Buffer Switching Characteristic Guidelines

When fewer vertical clock lines are connected, the clock distribution is faster; when multiple clock lines per column are
driven from the same global clock, the delay is longer. For more specific, more precise, and worst-case guaranteed data,
the static timing analyzer report. All timing parameters assume worst-case operating conditions (supply voltage and junction
temperature)
Speed Grade -3* -4**

Description Symbol Device Max Max Units
From pad through Primary buffer, to any TPG XQ4005E 7.0 ns
clock K XQ4010E 6.3 11.0 ns
XQ4013E 6.8 11.5 ns
XQ4025E 12.5 ns
10
From pad through Secondary buffer, to any TSG XQ4005E 7.5 ns
clock K XQ4010E 6.8 11.5 ns
XQ4013E 7.3 12.0 ns
XQ4025E 13.0 ns
* For plastic package options only.

** For ceramic package options only.
May 19, 1998 (Version 2.1) 10-13

R
XQ4000E Horizontal Longline Switching Characteristic Guidelines

worst-case operating conditions (supply voltage and junction temperature). Values apply to all XQ4000E/EX devices unless
otherwise noted.
Speed Grade -3 -4
TBUF driving a Horizontal Longline (LL):
I going High or Low to LL going High or Low, while T is Low. TIO1 XQ4005E 5.0 ns
Buffer is constantly active. (Note1) XQ4010E 6.4 8.0 ns
XQ4013E 7.2 9.0 ns
XQ4025E 11.0 ns
I going Low to LL going from resistive pull-up High to active Low. TIO2 XQ4005E 6.0 ns
TBUF configured as open-drain. (Note1) XQ4010E 6.9 10.5 ns
XQ4013E 7.7 11.0 ns
XQ4025E 12.0 ns
T going Low to LL going from resistive pull-up or floating High to TON XQ4005E 7.0 ns
active Low. XQ4010E 7.3 8.5 ns
XQ4013E 7.5 8.7 ns
TBUF configured as open-drain or active buffer with I = Low. XQ4025E 11.0 ns
(Note1)
T going High to TBUF going inactive, not driving LL TOFF XQ4005E 1.8 ns
XQ4010E 3.0 ns
XQ4013E 3.5 ns
XQ4025E 4.0 ns
T going High to LL going from Low to High, pulled up by a single TPUS XQ4005E 23.0 ns
resistor. (Note 1) XQ4010E 22 29.0 ns
XQ4013E 26 32.0 ns
XQ4025E 42.0 ns
T going High to LL going from Low to High, pulled up by two resis- TPUF XQ4005E 10.0 ns
tors. (Note1) XQ4010E 11 13.5 ns
XQ4013E 13 15.0 ns
XQ4025E 18.0 ns
Note 1: These values include a minimum load. Use the static timing analyzer to determine the delay for each destination.
10-14 May 19, 1998 (Version 2.1)

R
XQ4000E Wide Decoder Switching Characteristic Guidelines

otherwise noted.
Speed Grade -3 -4
Full length, both pull-ups, TWAF XQ4005E 9.5 ns
inputs from IOB I-pins XQ4010E 9.0 15.0 ns
XQ4013E 11.0 16.0 ns
XQ4025E 18.0 ns
Full length, both pull-ups, TWAFL XQ4005E 12.5 ns
inputs from internal logic XQ4010E 11.0 18.0 ns
XQ4013E 13.0 19.0 ns
XQ4025E 21.0 ns
Half length, one pull-up, TWAO XQ4005E 10.5 ns
inputs from IOB I-pins XQ4010E 10.0 16.0 ns
XQ4013E 12.0 17.0 ns
XQ4025E 19.0 ns
Half length, one pull-up, TWAOL XQ4005E 12.5 ns
inputs from internal logic XQ4010E 12.0 18.0 ns
XQ4013E 14.0 19.0 ns
XQ4025E 21.0 ns
Note 2: Fewer than the specified number of pullup resistors can be used, if desired. Using fewer pullups reduces power consumption
but increases delays. Use the static timing analyzer to determine delays if fewer pullups are used.
10
May 19, 1998 (Version 2.1) 10-15

R
XQ4000E CLB Switching Characteristic Guidelines

otherwise noted.
Speed Grade -3 -4
Units
C inputs via H to X/Y outputs THH1O 4.7 4.9 ns
Operand inputs (F1, F2, G1, G4) to COUT TOPCY 2.6 4.4 ns
Add/Subtract input (F3) to COUT TASCY 4.4 6.8 ns
Initialization inputs (F1, F3) to COUT TINCY 1.7 2.9 ns
CIN through function generators to X/Y outputs TSUM 4.4 5.0 ns
CIN to COUT, bypass function generators TBYP 0.7 1.0 ns
Sequential Delays
Clock K to outputs Q TCKO 5.0 ns
C inputs via DIN TDICK 2.4 3.0 ns
C inputs via EC TECCK 3.0 4.0 ns
C inputs via S/R, going Low (inactive) TRCK 4.0 4.2 ns
10-16 May 19, 1998 (Version 2.1)

R
XQ4000E CLB Switching Characteristic Guidelines (continued)

that are taken at device introduction, prior to any process improvements. For more detailed, more precise, and more
up-to-date information, use the values provided by the static timing analyzer and used in the simulator.
of nanoseconds and apply to all XQ4000E/EX devices unless otherwise noted.
Speed Grade -3 -4
Units
F/G inputs TCKI 0 0 ns
F/G inputs via H TCKIH 0 0 ns
C inputs via H1 through H TCKHH1 0 0 ns
C inputs via DIN TCKDI 0 0 ns
C inputs via EC TCKEC 0 0 ns
C inputs via SR, going Low (inactive) TCKR 0 0 ns
Clock
Set/Reset Direct
Master Set/Reset
Width (High or Low) TMRW 4005E 13.0 ns
4010E 11.5 55.0 ns
4013E 11.5 70.0 ns
4025E 112.0 ns
Delay from Global Set/Reset net to Q TMRQ 4005E 23.0 ns
4010E 18.7 60.0 ns
4013E 18.7 77.0 ns
4025E 134.0 ns
10
May 19, 1998 (Version 2.1) 10-17

R
XQ4000E CLB Edge-Triggered (Synchronous) RAM Switching Characteristic Guidelines

temperature). Values apply to all XQ4000E/EX devices unless otherwise noted.
Speed Grade -3 -4
Write Operation
32x1 TWCTS 14.4 15.0 ns
Clock K pulse width (active edge) 16x2 TWPS 7.2 7.5 1 ms ns
32x1 TWPTS 7.2 7.5 1 ms ns
Address setup time before clock K 16x2 TASS 2.4 2.8 ns
32x1 TASTS 2.4 2.8 ns
Address hold time after clock K 16x2 TAHS 0 0 ns
32x1 TAHTS 0 0 ns
32x1 TDSTS 1.9 2.5 ns
DIN hold time after clock K 16x2 TDHS 0 0 ns
32x1 TDHTS 0 0 ns
32x1 TWSTS 2.0 2.2 ns
WE hold time after clock K 16x2 TWHS 0 0 ns
32x1 TWHTS 0 0 ns
32x1 TWOTS 10.3 11.6 ns
Speed Grade -3 -4
Dual-Port RAM Units
Write Operation
Clock K pulse width (active edge) 16x1 TWPDS 7.2 7.5 1 ms ns
Address hold time after clock K 16x1 TAHDS 0 0 ns
DIN hold time after clock K 16x1 TDHDS 0 0 ns
WE hold time after clock K 16x1 TWHDS 0 0.3 ns
10-18 May 19, 1998 (Version 2.1)

R
XQ4000E CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS
WCLK (K)
TWSS TWHS
WE
TDSS TDHS
DATA IN
TASS TAHS
ADDRESS
TILO TILO
TWOS
DATA OUT OLD NEW
X6461
XQ4000E CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing
TWPDS
WCLK (K)
TWSDS TWHDS
WE
10
TDSDS TDHDS
DATA IN
TASDS TAHDS
ADDRESS
TILO TILO
TWODS
DATA OUT OLD NEW
X6474
May 19, 1998 (Version 2.1) 10-19

R
XQ4000E CLB Level-Sensitive RAM Switching Characteristic Guidelines

XQ4000E/EX devices unless otherwise noted.
Speed Grade -3 -4
Units
Description Size Symbol Min Max Max
Write Operation
Address write cycle time 16x2 TWC 8.0 8.0 ns

32x1 TWCT 8.0 8.0 ns
Write Enable pulse width (High) 16x2 TWP 4.0 4.0 ns

32x1 TWPT 4.0 4.0 ns
Address setup time before WE 16x2 TAS 2.0 2.0 ns

32x1 TAST 2.0 2.0 ns
Address hold time after end of WE 16x2 TAH 2.0 2.5 ns

32x1 TAHT 2.0 2.0 ns
DIN setup time before end of WE 16x2 TDS 2.2 4.0 ns

32x1 TDST 2.2 5.0 ns
DIN hold time after end of WE 16x2 TDH 2.0 2.0 ns

32x1 TDHT 2.0 2.0 ns
Read Operation

32x1 TRCT 5.5 6.5 ns
Data valid after address change 16x2 TILO 3.1 3.9 ns

(no Write Enable) 32x1 TIHO 5.5 5.9 ns

32x1 TIHCK 4.6 6.1 ns
Read During Write
Data valid after WE goes active (DIN stable before 16x2 TWO 6.0 10.0 ns
WE) 32x1 TWOT 7.3 12.0 ns
Data valid after DIN 16x2 TDO 6.6 9.0 ns

(DIN changes during WE) 32x1 TDOT 7.6 11.0 ns
WE setup time before clock K 16x2 TWCK 6.0 8.0 ns

32x1 TWCKT 6.8 9.6 ns
Data setup time before clock K 16x2 TDCK 5.2 7.0 ns

32x1 TDCKT 6.2 8.0 ns
10-20 May 19, 1998 (Version 2.1)

R
XQ4000E CLB Level-Sensitive RAM Timing Characteristics

T WC
ADDRESS
WRITE
TAS T WP T AH
WRITE ENABLE
T DS T DH
DATA IN REQUIRED
READ WITHOUT WRITE

T ILO

T ICK T CH
CLOCK
T CKO
VALID VALID
XQ, YQ OUTPUTS
(OLD) (NEW)
READ DURING WRITE

T WP
WRITE ENABLE
T DH
DATA IN
(stable during WE)
T WO
10
DATA IN
T WO T DO
VALID VALID VALID

X, Y OUTPUTS

T WP
WRITE ENABLE
T WCK
T DCK
DATA IN
CLOCK
T CKO
XQ, YQ OUTPUTS
X2640
May 19, 1998 (Version 2.1) 10-21

R
XQ4000E Guaranteed Input and Output Parameters (Pin-to-Pin, TTL I/O)

the static timing analyzer report. Values apply to all XQ4000E/EX devices unless otherwise noted.
Speed Grade -3 -4
Units
Global Clock to Output TICKOF XQ4005E 14.0 ns
(fast) using OFF XQ4010E 10.9 16.0 ns
XQ4013E 11.0 16.5 ns
XQ4025E 17.0 ns
TPG OFF
.
.
(Max)
.
.
.
X3202
Global Clock to Output TICKO XQ4005E 18.0 ns

(slew-limited) using OFF XQ4010E 14.9 20.0 ns
XQ4013E 15.0 20.5 ns
XQ4025E 21.0 ns
TPG OFF
.
. (Max)
.
.
.
X3202
Input Setup Time, using IFF TPSUF XQ4005E 2.0 ns

(no delay) XQ4010E 0.2 1.9 ns
XQ4013E 0 1.6 ns
D XQ4025E 1.5 ns
Input (Min)
Set - Up IFF
& TPG
Hold
Time
X3201
Input Hold Time, using IFF TPHF XQ4005E 4.6 ns

(no delay) XQ4010E 5.5 6.0 ns
XQ4013E 6.5 7.0 ns
D XQ4025E 8.0 ns
Input (Min)
Set - Up IFF
& TPG
Hold
Time
X3201
Input Setup Time, using IFF TPSU XQ4005E 8.5 ns

(with delay) XQ4010E 7.0 8.5 ns
XQ4013E 7.0 8.5 ns
D XQ4025E 9.5 ns
Input (Min) ns
Set - Up IFF
& TPG ns
Hold
Time ns
X3201
ns
Input Hold Time, using IFF TPH XQ4005E 0 ns

(with delay) XQ4010E 0 0 ns
XQ4013E 0 0 ns
D XQ4025E 0 ns
Input (Min)
Set - Up IFF
& TPG
Hold
Time
X3201
OFF = Output Flip-Flop IFF = Input Flip-Flop or Latch
10-22 May 19, 1998 (Version 2.1)

R
XQ4000E IOB Input Switching Characteristic Guidelines

the static timing analyzer report. Values apply to all XQ4000E/EX devices unless otherwise noted.
Speed Grade -3 -4
Units
Propagation Delays (TTL Inputs)
Pad to I1, I2 TPID All devices 2.5 3.0 ns
Pad to I1, I2 via transparent latch, no delay TPLI All devices 6.0 ns
with delay TPDLI XQ4005E 12.0 ns
XQ4010E 10.8 12.2 ns
XQ4013E 11.2 12.6 ns
XQ4025E 15.0 ns
Propagation Delays
Hold Times (Note 1)
Pad to Clock (IK), no delay TIKPI All devices 0 0 ns
with delay TIKPID All devices 0 0 ns
10
May 19, 1998 (Version 2.1) 10-23

R
XQ4000E IOB Input Switching Characteristic Guidelines (continued)

otherwise noted.
Speed Grade -3 -4
Units
Setup Times (TTL Inputs)
with delay TPICKD XQ4005E 10.9 ns
XQ4010E 9.8 11.3 ns
XQ4013E 10.2 11.8 ns
XQ4025E 14.0 ns
ns
ns
ns
ns
(TTL or CMOS)
Clock Enable (EC) to Clock (IK), no delay
with delay TECIK All devices 2.5 3.5 ns
TECIKD XQ4005E 10.4 ns
XQ4010E 9.7 10.7 ns
XQ4013E 10.1 11.1 ns
XQ4025E 14.0 ns
Delay from GSR net through Q to I1, I2 TRRI XQ4005E 12.0 ns
GSR width XQ4010E 7.8 21.0 ns
GSR inactive to first active Clock (IK) edge XQ4013E 7.8 23.0 ns
XQ4025E 29.0 ns
TMRW XQ4005E 13.0 ns
XQ4010E 11.5 55.0 ns
XQ4013E 11.5 70.0 ns
XQ4025E 112.0 ns
TRPO XQ4005E 15.0 ns
XQ4010E 11.8 20.3 ns
XQ4013E 11.8 22.0 ns
XQ4025E 28.0 ns
Note 3: Timing is based on the XQ4005E. For other devices see the static timing analyzer.
10-24 May 19, 1998 (Version 2.1)

R
XQ4000E IOB Output Switching Characteristic Guidelines

otherwise noted.
Speed Grade -3 -4
Units
Propagation Delays (TTL Output Levels)
Clock (OK) to Pad, fast TOKPOF 6.5 7.5 ns
slew-rate limited TOKPOS 9.5 11.5 ns
Output (O) to Pad, fast TOPF 5.5 8.0 ns
slew-rate limited TOPS 8.6 12.0 ns
3-state to Pad hi-Z TTSHZ 4.2 10.0 ns
and valid, fast TTSONF 8.1 10.0 ns
slew-rate limited TTSONS 11.1 13.7 ns
bounce, see the “Additional XQ4000 Data” section of the Programmable Logic Data Book.
10
May 19, 1998 (Version 2.1) 10-25

R
XQ4000E IOB Output Switching Characteristic Guidelines (continued)

otherwise noted. Values apply to all XQ4000E/EX devices unless otherwise noted.
Speed Grade -3 -4
Units
Setup and Hold
Output (O) to clock (OK) setup time TOOK 4.6 5.0 ns
Output (O) to clock (OK) hold time TOKO 0 0 ns
Clock
Clock High TCH 4.0 4.5 ns
Clock Low TCL 4.0 4.5 ns
bounce, see the “Additional XQ4000 Data” section of the Programmable Logic Data Book.
Note 3: Timing is based on the XQ4005E. For other devices see the static timing analyzer.
10-26 May 19, 1998 (Version 2.1)

R
XQ4028EX Switching Characteristics

Definition of Terms
follows:
Advance: Initial estimates based on simulation and/or extrapolation from other speed grades, devices, or device families.
Preliminary:Based on preliminary characterization. Further changes are not expected.
XQ4028EX Absolute Maximum Ratings

VCCt Longest Supply Voltage Rise Time from 1 V to 4 V 50 ms
TJ
XQ4028EX Recommended Operating Conditions

10
Supply voltage relative to GND, TC = -55 °C to +125°C Ceramic Packages 4.5 5.5 V
VCC
Supply voltage relative to GND, TJ = -55°C to +125°C Plastic Packages 4.5 5.5 V
High-level input voltage TTL inputs 2.0 VCC V
VIH
Low-level input voltage TTL inputs 0 0.8 V
VIL
Note 2: Input and output measurement thresholds for TTL are 1.5 V. Input and output measurement thresholds for CMOS are 2.5 V.
Note 3: All timing parameters are specified for Commercial temperature range only.
May 19, 1998 (Version 2.1) 10-27

R
XQ4028EX DC Characteristics Over Recommended Operating Conditions

High-level output voltage @ IOH = -4.0 mA, VCC min TTL outputs 2.4 V
VOH
High-level output voltage @ IOH = -1.0 mA CMOS outputs VCC-0.5 V
Low-level output voltage @ IOL = 12.0 mA, VCC min TTL outputs 0.4 V
VOL (Note 1) CMOS outputs 0.4 V
Input capacitance (sample tested) Plastic packages 10 pF
CIN
Ceramic packages 16 pF
Note 2: With no output current loads, no active input or Longline pull-up resistors, all package pins at Vcc or GND.
XQ4028EX Global Buffer Switching Characteristic Guidelines

When fewer vertical clock lines are connected, the clock distribution is faster; when multiple clock lines per column are
driven from the same global clock, the delay is longer. For more specific, more precise, and worst-case guaranteed data,
the static timing analyzer report. All timing parameters assume worst-case operating conditions (supply voltage and junction
temperature).
Speed Grade -4
Units
Description Symbol Max
From pad through Global Low Skew buffer, to any clock K TGLS 9.2 ns
From pad through Global Early buffer, to any clock K in same quadrant TGE 5.7 ns
10-28 May 19, 1998 (Version 2.1)

R
XQ4028EX Longline and Wide Decoder Timing Guidelines

worst-case operating conditions (supply voltage and junction temperature). Values apply to all XQ4000EX devices unless
otherwise noted. Fewer than the specified number of pullup resistors can be used, if desired. Using fewer pullups reduces
power consumption but increases delays. Use the static timing analyzer to determine delays if fewer pullups are used.
XQ4028EX Horizontal Longline Switching Characteristic Guidelines
Speed Grade -4
Units
TBUF driving a Horizontal Longline
I going High or Low to Horizontal Longline going High or Low, while T is Low. Buffer TIO1 13.7 ns
is constantly active.
T going Low to Horizontal Longline going from resistive pull-up or floating High to TON 14.7 ns
active Low. TBUF configured as open-drain or active buffer with I = Low.
T going High to Horizontal Longline going from Low to High, pulled up by two resis- TPU2 ns
tors. (Note 1) ns
TBUF driving Half a Horizontal Longline
I going High or Low to half of a Horizontal Longline going High or Low, while T is THIO1 6.3 ns
Low. Buffer is constantly active.
T going Low to half of a Horizontal Longline going from resistive pull-up or floating THON 7.2 ns
High to active Low. TBUF configured as open-drain or active buffer with I = Low.
T going High to half of a Horizontal Longline going from Low to High, pulled up by THPU4 ns
four resistors. (Note 1)
Note 1: These values include a minimum load of one output, spaced as far as possible from the activated pullup(s). Use the static timing ana-
lyzer to determine the delay for each destination.
10
XQ4028EX Wide Decoder Switching Characteristic Guidelines
Speed Grade -4
Units
Full length, two pull-ups, inputs from IOB I-pins TWAF2 ns
ns
Full length, two pull-ups, inputs from internal logic TWAF2L ns
ns
Half length, two pull-ups, inputs from IOB I-pins TWAO2 ns
ns
Half length, two pull-ups, inputs from internal logic TWAO2L ns
ns
May 19, 1998 (Version 2.1) 10-29

R
XQ4028EX CLB Switching Characteristic Guidelines

XQ4000EX devices unless otherwise noted.
Speed Grade -4
Units
Description Symbol Min Max
F/G inputs to X/Y outputs TILO 2.2 ns
F/G inputs via H’ to X/Y outputs TIHO 3.8 ns
F/G inputs via transparent latch to Q outputs TITO 3.2 ns
C inputs via SR/H0 via H’ to X/Y outputs THH0O 3.6 ns
C inputs via H1 via H’ to X/Y outputs THH1O 3.0 ns
C inputs via DIN/H2 via H’ to X/Y outputs THH2O 3.6 ns
C inputs via EC, DIN/H2 to YQ, XQ output (bypass) TCBYP 2.0 ns
Operand inputs (F1, F2, G1, G4) to COUT TOPCY 2.5 ns
Add/Subtract input (F3) to COUT TASCY 4.1 ns
Initialization inputs (F1, F3) to COUT TINCY 1.9 ns
CIN through function generators to X/Y outputs TSUM 3.0 ns
CIN to COUT, bypass function generators TBYP 0.60 ns
Carry Net Delay, COUT to CIN TNET 0.18 ns
Sequential Delays
Clock K to Flip-Flop outputs Q TCKO 2.2 ns
Clock K to Latch outputs Q TCKLO 2.2 ns
F/G inputs TICK 1.3 ns
F/G inputs via H’ TIHCK 3.0 ns
C inputs via H0 through H’ THH0CK 2.8 ns
C inputs via DIN TDICK 1.2 ns
C inputs via EC TECCK 1.2 ns
C inputs via S/R, going Low (inactive) TRCK 0.8 ns
CIN input via F’/G’ TCCK 2.2 ns
CIN input via F’/G’ and H’ TCHCK 3.9 ns
F/G inputs TCKI 0 ns
F/G inputs via H’ TCKIH 0 ns
C inputs via SR/H0 through H’ TCKHH0 0 ns
C inputs via H1 through H’ TCKHH1 0 ns
C inputs via DIN/H2 through H’ TCKHH2 0 ns
C inputs via DIN/H2 TCKDI 0 ns
C inputs via EC TCKEC 0 ns
C inputs via SR, going Low (inactive) TCKR 0 ns
Clock
Clock High time TCH 3.5 ns
Clock Low time TCL 3.5 ns
Set/Reset Direct
Width (High) TRPW 3.5 ns
Delay from C inputs via S/R, going High to Q TRIO 4.5 ns
10-30 May 19, 1998 (Version 2.1)

R
Speed Grade -4
Units
Global Set/Reset
Minimum GSR Pulse Width TMRW 13.0 ns
Delay from GSR input to any Q (XQ4028EX) TMRQ 22.8 ns
Delay from GSR input to any Q (XQ4036EX) TMRQ 24.0 ns
Toggle Frequency ) (for export control purposes) FTOG 143 MHz
XQ4028EX CLB RAM Synchronous (Edge-Triggered) Write Operation Guidelines

temperature). Values apply to all XQ4000EX devices unless otherwise noted.
Speed Grade -4
Size Symbol Min Max
Write Operation
Address write cycle time (clock K period) 16x2 TWCS 11.0 ns
32x1 TWCTS 11.0 ns
Clock K pulse width (active edge) 16x2 TWPS 5.5 ns
32x1 TWPTS 5.5 ns
Address setup time before clock K 16x2 TASS 2.7 ns
32x1 TASTS 2.6 ns
Address hold time after clock K 16x2 TAHS 0 ns
32x1 TAHTS 0 ns
DIN setup time before clock K 16x2 TDSS 2.4 ns
32x1 TDSTS 2.9 ns
DIN hold time after clock K 16x2 TDHS 0 ns 10
32x1 TDHTS 0 ns
WE setup time before clock K 16x2 TWSS 2.3 ns
32x1 TWSTS 2.1 ns
WE hold time after clock K 16x2 TWHS 0 ns
32x1 TWHTS 0 ns
Data valid after clock K 16x2 TWOS 8.2 ns
32x1 TWOTS 10.1 ns
May 19, 1998 (Version 2.1) 10-31

R
Speed Grade -4
Dual-Port RAM Units
Size Symbol Min Max
Write Operation
Address write cycle time (clock K period) 16x1 TWCDS 11.0 ns

Clock K pulse width (active edge) 16x1 TWPDS 5.5 ns
Address setup time before clock K 16x1 TASDS 3.1 ns
Address hold time after clock K 16x1 TAHDS 0 ns
DIN setup time before clock K 16x1 TDSDS 2.9 ns
DIN hold time after clock K 16x1 TDHDS 0 ns
WE setup time before clock K 16x1 TWSDS 2.1 ns
WE hold time after clock K 16x1 TWHDS 0 ns
Data valid after clock K 16x1 TWODS 9.4 ns
10-32 May 19, 1998 (Version 2.1)

R
XQ4000EX CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS
WCLK (K)
TWSS TWHS
WE
TDSS TDHS
DATA IN
TASS TAHS
ADDRESS
TILO TILO
TWOS
DATA OUT OLD NEW
X6461
XQ4000EX CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing
TWPDS
WCLK (K)
TWSDS TWHDS
WE
10
TDSDS TDHDS
DATA IN
TASDS TAHDS
ADDRESS
TILO TILO
TWODS
DATA OUT OLD NEW
X6474
May 19, 1998 (Version 2.1) 10-33

R
XQ4028EX CLB RAM Asynchronous (Level-Sensitive) Write and Read Operation Guidelines
XQ4000EX devices unless otherwise noted.
Speed Grade -4
Units
Description Size Symbol Min Max
Write Operation
Address write cycle time 16x2 TWC 10.6 ns

32x1 TWCT 10.6 ns
Write Enable pulse width (High) 16x2 TWP 5.3 ns

32x1 TWPT 5.3 ns
Address setup time before WE 16x2 TAS 2.8 ns

32x1 TAST 2.9 ns
Address hold time after end of WE 16x2 TAH 1.7 ns

32x1 TAHT 1.7 ns
DIN setup time before end of WE 16x2 TDS 1.1 ns

32x1 TDST 1.1 ns
DIN hold time after end of WE 16x2 TDH 6.6 ns

32x1 TDHT 6.6 ns
Read Operation
Address read cycle time 16x2 TRC 4.5 ns

32x1 TRCT 6.5 ns
Data valid after address change (no 16x2 TILO 2.2 ns

Write Enable) 32x1 TIHO 3.8 ns
Address setup time before clock K 16x2 TICK 1.5 ns

32x1 TIHCK 3.2 ns
Read During Write
Data valid after WE goes active 16x2 TWO 6.5 ns

(DIN stable before WE) 32x1 TWOT 7.4 ns
Data valid after DIN (DIN changes 16x2 TDO 7.7 ns

during WE) 32x1 TDOT 8.2 ns
WE setup time before clock K 16x2 TWCK 7.1 ns

32x1 TWCKT 9.2 ns
Data setup time before clock K 16x2 TDCK 5.9 ns

32x1 TDCKT 8.4 ns
10-34 May 19, 1998 (Version 2.1)

R
XQ4028EXX CLB RAM Asynchronous (Level-Sensitive) Timing Characteristics

T WC
ADDRESS
WRITE
TAS T WP T AH
WRITE ENABLE
T DS T DH
DATA IN REQUIRED
READ WITHOUT WRITE

T ILO

T ICK T CH
CLOCK
T CKO
VALID VALID
XQ, YQ OUTPUTS
(OLD) (NEW)
READ DURING WRITE

T WP
WRITE ENABLE
T DH
DATA IN
(stable during WE)
T WO
10
DATA IN
T WO T DO
VALID VALID VALID

X, Y OUTPUTS

T WP
WRITE ENABLE
T WCK
T DCK
DATA IN
CLOCK
T CKO
XQ, YQ OUTPUTS
X2640
May 19, 1998 (Version 2.1) 10-35

R
XQ4028EX Pin-to-Pin Output Parameter Guidelines

the static timing analyzer report. Values apply to all XQ4000EX devices unless otherwise noted.
XQ4028EX Output Flip-Flop, Clock to Out
Speed Grade -4
Units
Global Low Skew Clock to TTL Output (fast) using OFF TICKOF 16.6 ns
Global Early Clock to TTL Output (fast) using OFF TICKEOF 13.1 ns
XQ4028EX Output MUX, Clock to Out

Speed Grade -4
Units
Global Low Skew Clock to TTL Output (fast) using OMUX TPFPF 15.9 ns
Global Early Clock to TTL Output (fast) using OMUX TPEFPF 12.4 ns
OMUX = Output MUX
Note 1: Listed above are representative values where one global clock input drives one vertical clock line in each accessible column, and
where all accessible IOB and CLB flip-flops are clocked by the global clock net.
Note 2: Output timing is measured at TTL threshold with 50 pF external capacitive load.
Note 3: Set-up time is measured with the fastest route and the lightest load. Hold time is measured using the farthest distance and a refer-
ence load of one clock pin per two IOBs. Use the static timing analyzer to determine the setup and hold times under given design
conditions.
XQ4028EX Output Level and Slew Rate Adjustments

The following table must be used to adjust output parameters and output switching characteristics.
Speed Grade -4
For TTL output FAST add TTTLOF 0 ns
For TTL output SLOW add TTTLO 2.9 ns
For CMOS FAST output add TCMOSOF 1.0 ns
For CMOS SLOW output add TCMOSO 3.6 ns
10-36 May 19, 1998 (Version 2.1)

R
XQ4028EX Pin-to-Pin Input Parameter Guidelines

the static timing analyzer report. Values apply to all XQ4000EX devices unless otherwise noted
XQ4028EX Global Low Skew Clock, Set-Up and Hold

Speed Grade -4
Units
Description Symbol Min
Input Setup Time, using Global Low Skew clock and IFF (full delay) TPSD 8.0 ns
Input Hold Time, using Global Low Skew clock and IFF (full delay) TPHD 0 ns
XQ4028EX Global Early Clock, Set-Up and Hold for IFF

Speed Grade -4
Units
Input Setup Time, using Global Early clock and IFF (partial delay) TPSEP 6.5 ns
Input Hold Time, using Global Early clock and IFF (partial delay) TPHEP 0 ns
XQ4028EX Global Early Clock, Set-Up and Hold for FCL

Speed Grade -4
Units
Input Setup Time, using Global Early clock and FCL (partial delay) TPFSEP 3.4 ns
Input Hold Time, using Global Early clock and FCL (partial delay) TPFHEP 0 ns
FCL = Fast Capture Latch
Note 1: For CMOS input levels, see the “XQ4028EX Input Threshold Adjustments” on page 37. 10
Set-up time is measured with the fastest route and the lightest load. Use the static timing analyzer to determine the setup time
Note 2: Under given design conditions. Hold time is measured using the farthest distance and a reference load of one clock pin per two IOBs.
Use the static timing analyzer to determine the setup and hold times under given design conditions.
XQ4028EX Input Threshold Adjustments

The following table must be used to adjust input parameters and input switching characteristics.
Speed Grade -4
For TTL input add TTTLI 0 ns
For CMOS input add TCMOSI 0.3 ns
May 19, 1998 (Version 2.1) 10-37

R
XQ4028EX IOB Input Switching Characteristic Guidelines

otherwise noted.
Speed Grade -4 Units

Clocks
Delay from FCL enable (OK) active edge to IFF clock (IK) active edge TOKIK 3.2 ns
Propagation Delays Max
Pad to I1, I2 TPID 2.2 ns
Pad to I1, I2 via transparent input latch, no delay TPLI 3.8 ns
Pad to I1, I2 via transparent input latch, partial delay TPPLI 13.3 ns
Pad to I1, I2 via transparent input latch, full delay TPDLI 18.2 ns
Pad to I1, I2 via transparent FCL and input latch, no delay TPFLI 5.3 ns
Pad to I1, I2 via transparent FCL and input latch, partial delay TPPFLI 13.6 ns
Propagation Delays
Clock (IK) to I1, I2 (flip-flop) TIKRI 3.0 ns
Clock (IK) to I1, I2 (latch enable, active Low) TIKLI 3.2 ns
FCL Enable (OK) active edge to I1, I2 TOKLI 6.2 ns
Global Set/Reset
Delay from GSR input to any Q TRRI 22.8 ns
FCL = Fast Capture Latch, IFF = Input Flip-Flop or Latch
Note 1: For CMOS input levels, see the “XQ4028EX Input Threshold Adjustments” on page 37.
Note 2: For set-up and hold times with respect to the clock input pin, see the Global Low Skew Clock and Global Early Clock Set-up and Hold
tables on page 37.
.
10-38 May 19, 1998 (Version 2.1)

R
XQ4028EX IOB Input Switching Characteristic Guidelines (Continued)

otherwise noted.
Speed Grade -4
Units
Setup Times
Pad to Clock (IK), no delay TPICK 2.5 ns
Pad to Clock (IK), partial delay TPICKP 10.8 ns
Pad to Clock (IK), full delay TPICKD 15.7 ns
Pad to Clock (IK), via transparent Fast Capture Latch, no delay TPICKF 3.9 ns
Pad to Clock (IK), via transparent Fast Capture Latch, partial delay TPICKFP 12.3 ns
Pad to Fast Capture Latch Enable (OK), no delay TPOCK 0.8 ns
Pad to Fast Capture Latch Enable (OK), partial delay TPOCKP 9.1 ns
Setup Times (TTL or CMOS Inputs)
Clock Enable (EC) to Clock (IK) TECIK 0.3 ns
Hold Times
Pad to Clock (IK),
no delay TIKPI 0 ns
partial delay TIKPIP 0 ns
full delay TIKPID 0 ns
Pad to Clock (IK) via transparent Fast
Capture Latch,
no delay TIKFPI 0 ns
partial delay TIKFPIP 0 ns
full delay TIKFPID 0 ns
Clock Enable (EC) to Clock (IK),
no delay TIKEC 0 ns 10
partial delay TIKECP 0 ns
full delay TIKECD 0 ns
Pad to Fast Capture Latch Enable (OK),
no delay TOKPI 0 ns
partial delay TOKPIP 0 ns
Note 1: For CMOS input levels, see the “XQ4028EX Input Threshold Adjustments” on page 37.
tables on page 37.
May 19, 1998 (Version 2.1) 10-39

R
XQ4028EX IOB Output Switching Characteristic Guidelines

otherwise noted. Values apply to all XQ4000EX devices unless otherwise noted.
Speed Grade -4
Units
Description Symbol Min Max Max
Propagation Delays
Clock (OK) to Pad TOKPOF 7.4 6.0 ns
Output (O) to Pad TOPF 6.2 5.0 ns
3-state to Pad hi-Z (slew-rate independent) TTSHZ 4.9 4.1 ns
3-state to Pad active and valid TTSONF 6.2 5.0 ns
Output MUX Select (OK) to Pad TOKFPF 6.7 5.4 ns
Fast Path Output MUX Input (EC) to Pad TCEFPF 6.2 5.0 ns
Slowest Path Output MUX Input (O) to Pad TOFPF 7.3 5.9 ns
Output (O) to clock (OK) setup time TOOK 0.6 ns
Output (O) to clock (OK) hold time TOKO 0 ns
Clock Enable (EC) to clock (OK) setup TECOK 0 ns
Clock Enable (EC) to clock (OK) hold TOKEC 0 ns
Clock
Clock High TCH 3.5 ns
Clock Low TCL 3.5 ns
Global Set/Reset
Minimum GSR pulse width TMRW 13.0 ns
Delay from GSR input to any Pad (XQ4028EX) TRPO 30.2 ns
Delay from GSR input to any Pad (XQ4036EX) TRPO 31.4 ns
Note 1: Output timing is measured at TTL threshold, with 35pF external capacitive loads.
Note 2: For CMOS output levels, see the “XQ4028EX Output Level and Slew Rate Adjustments” on page 36.
10-40 May 19, 1998 (Version 2.1)

R
CB191/196 Package for XQ4010E Bound

Pin Description PG191 CB196 Scan
Bound I/O B14 P41 224
Pin Description PG191 CB196 Scan
I/O A16 P42 227
GND D4 P1 -
I/O B15 P43 230
PGCK1_(A16*I/0) C3 P2 122
I/O C14 P44 233
I/O_(A17) C4 P3 125
I/O A17 P45 236
I/0 B3 P4 128
SCGK2_(I/O) B16 P46 239
- - P5* -
M1 C15 P47 242
I/O C5 P6 131
GND D15 P48
I/O_(TDI) A2 P7 134
M0 A18 P49 245**
I/O_(TCK) B4 P8 137
VCC D16 P50 -
I/O C6 P9 140
M2 C16 P51 246**
I/O A3 P10 143
PGCK2_(I/O) B17 P52 247
I/O B5 P11 146
I/O_(HDC) E16 P53 250
I/O B6 P12 149
- - P54* -
GND C7 P13 -
I/O C17 P55 253
I/O A4 P14 152
I/0 D17 P56 256
I/O A5 P15 155
I/O B18 P57 259
I/O_(TMS) B7 P16 158
I/O_(LDC) E17 P58 262
I/O A6 P17 161
I/O F16 P59 265
I/O C8 P18 164
I/O C18 P60 268
I/O A7 P19 167
I/O D18 P61 271
I/O B8 P20 170
I/O F17 P62 274
I/O A8 P21 173
GND G16 P63 -
I/O B9 P22 176
I/O E18 P64 277
I/O C9 P23 179
I/O F18 P65 280
GND D9 P24 -
I/O G17 P66 283
VCC D10 P25 -
I/O G18 P67 286
I/O C10 P26 182
I/O H16 P68 286
I/O B10 P27 185 10
I/O H17 P69 291
I/O A9 P28 -
I/O H18 P70 295
I/O A10 P29 191
I/O J18 P71 298
I/O A11 P30 194
I/O J17 P72 301
I/O C11 P31 197
I/O_(/ERR_/INIT) J16 P73 304
I/O B11 P32 200
VCC J15 P74 -
I/O A12 P33 203
GND K15 P75 -
I/O B12 P34 206
I/O K16 P76 307
I/O A13 P35 209
I/O K17 P77 310
GND C12 P36 -
I/O K18 P78 313
I/O B13 P37 212
I/O L18 P79 316
I/O A14 P38 215
I/O L17 P80 319
I/O A15 P39 218
I/O L16 P81 322
I/O C13 P40 221
* Indicates unconnected package pins.
* Indicates unconnected package pins. ** Contributes only one bit (.I) to the boundary scan regis-
** Contributes only one bit (.I) to the boundary scan register.
ter. Boundary Scan BIt 0 = TD0.T
Boundary Scan BIt 0 = TD0.T Boundary Scan Bit 1 = TD0.0
Boundary Scan Bit 1 = TD0.0 Boundary Scan Bit 487 = BSCAN.UPD
May 19, 1998 (Version 2.1) 10-41

R
Bound Bound
Pin Description PG191 CB196 Scan Pin Description PG191 CB196 Scan
I/O M18 P82 325 VCC R10 P123 -
I/O M17 P83 328 GND R9 P124 -
I/O N18 P84 331 I/O_(D3) T9 P125 427
I/O P18 P85 334 I/O_(/RS) U9 P126 430
GND M16 P86 - I/O V9 P127 433
I/O N17 P87 337 I/O V8 P128 436
I/O R18 P88 340 I/O U8 P129 439
I/O T18 P89 343 I/O T8 P130 442
I/O P17 P90 349 I/O_(D2) V7 P131 445
I/O N16 P91 349 I/O U7 P132 448
I/O T17 P92 352 I/O V6 P133 451
I/O R17 P93 355 I/O U6 P134 454
I/O P16 P94 358 GND T7 P135 -
I/O U18 P95 361 I/O V5 P136 457
SGCK3_(I/O) T16 P96 364 I/O V4 P137 460
GND R16 P97 - I/O U5 P138 463
DONE U17 P98 - I/O T6 T139 446
VCC R15 P99 - I/O_(D1) V3 P140 469
/PROG V18 P100 - I/O_(RCLK-/BUSY/RDY) V2 P141 472
I/O_(D7) T15 P101 367 I/O U4 P142 475
PGCK3_(I/O) U16 P102 370 I/O T5 P143 478
- - P103* - I/O_(D0*_DIN) U3 P144 481
I/O T14 P104 376 SGCK4_(DOUT*_I/O) T4 P145 484
I/O U15 P105 376 CCLK V1 P146 -
I/O_(D6) V17 P106 379 VCC R4 P147 -
I/O V16 P107 382 TDO U2 P148 -
I/O T13 P108 385 GND R3 P149 -
I/O U14 P109 388 I/O_(A0*_WS) T3 P150 2
I/O V15 P110 391 PGCK4_(I/O*_A1) U1 P151 5
I/O V14 P111 394 - - P152* -
GND T12 P112 - I/O P3 P153 8
I/O U13 P113 397 I/O R2 P154 11
I/O V13 P114 400 I/O_(CS1*_A2) T2 P155 14
I/O_(D5) U12 P115 403 I/O_(A3) N3 P156 17
I/O_(/CSO) V12 P116 406 I/O P2 P157 20
I/O T11 P117 409 I/O T1 P158 23
I/O U11 P118 412 I/O R1 P159 26
I/O V11 P119 415 I/O N2 P160 29
I/O V1 P120 418 GND M3 P161 -
I/O_(D4) U10 P121 421 I/O P1 P162 32
I/O T10 P122 424 I/O N1 P163 35
* Indicates unconnected package pins. * Indicates unconnected package pins.
** Contributes only one bit (.I) to the boundary scan regis- ** Contributes only one bit (.I) to the boundary scan regis-
ter. ter.
Boundary Scan BIt 0 = TD0.T Boundary Scan BIt 0 = TD0.T
Boundary Scan Bit 1 = TD0.0 Boundary Scan Bit 1 = TD0.0
Boundary Scan Bit 487 = BSCAN.UPD Boundary Scan Bit 487 = BSCAN.UPD
10-42 May 19, 1998 (Version 2.1)

R
Bound Bound
Pin Description PG191 CB196 Scan Pin Description PG191 CB196 Scan
I/O_(A4) M2 P164 38 GND G3 P184 -
I/O_(A5) M1 P165 41 I/O F2 P185 92
I/O L3 P166 44 I/O D1 P186 96
I/O L2 P167 47 I/O C1 P187 98
I/O L1 P168 50 I/O E2 P188 101
I/O K1 P169 53 I/O_(A12) F3 P189 104
I/O_(A6) K2 P170 56 I/O_(A13 D2 P190 107
I/O_(A7) K3 P171 59 - - P192* -
GND K4 P172 - I/O E3 P193 113
VCC J4 P173 - I/O_(A14) C2 P194 116
I/O_(A8) J3 P174 62 SGCK1(A15*I/O) B2 P195 119
I/O_(A9) J2 P175 65 VCC D3 P196 -
I/O J1 P176 68 * Indicates unconnected package pins.
I/O H1 P177 71 ** Contributes only one bit (.I) to the boundary scan regis-
I/O H2 P178 74 ter.
Boundary Scan BIt 0 = TD0.T
I/O H3 P179 77
Boundary Scan Bit 1 = TD0.0
I/O_(A10) G1 P180 80
I/O_(A11) G2 P181 83
I/O F1 P182 86 Additional XQ4010E Package Pins
I/O E1 P183 89 CB196
N.C. Pins
* Indicates unconnected package pins.
P5 P54 P103 P152
** Contributes only one bit (.I) to the boundary scan regis-
P192 - - -
ter.
Boundary Scan BIt 0 = TD0.T Note: Information current as of 8/14/97.
Boundary Scan Bit 1 = TD0.0
Example Ordering Information

10
XQ 4010E -4 PG 191 M
MIL-PRF-38535 Temperature Range

(QML) Processing M = Ceramic (TC = -55o C to +125o C)
N = Plastic (TJ = -55o C to +125o C)
Device Type
XQ4005E Number of Pins
XQ4010E
XQ4013E
XQ4025E
XQ4028EX
Package Type
Speed Grade CB = Top Brazed Ceramic Quad Flat Pack
-3 PG = Ceramic Pin Grid Array
-4 HQ = Plastic Quad Flat Pack
BG = Plastic Ball Grid Array
May 19, 1998 (Version 2.1) 10-43

R
10-44 May 19, 1998 (Version 2.1)

QPROTM XQ4000E/EX QML High Reliability Field Programmable Gate Arrays May 19, 1998 (Version 2.1) 10

XC4005E PG CB Bndry XC4005E PG CB Bndry XC4005E PG CB Bndry
Pad Name 156† 164 Scan Pad Name 156† 164 Scan Pad Name 156† 164 Scan
VCC H3 P145 - I/O (HDC) D14 P45 178 I/O (D2) P7 P109 313
I/O (A8) H1 P146 44 I/O C15 P46 181 I/O T5 P110 316
I/O (A9) G1 P147 47 I/O D15 P48 184 I/O R6 P111 319
I/O G2 P148 50 I/O E14 P49 187 I/O T4 P112 322
I/O G3 P149 53 I/O (LDC) C16 P50 190 GND P6 P113 -
I/O (A10) F1 P150 56 GND F14 P53 - I/O (D1) T3 P115 325
I/O (A11) F2 P151 59 I/O F15 P54 193 I/O (RCLK, P5 P116 328
I/O E16 P55 196 RDY/BUSY)
I/O E1 P152 62
I/O F16 P56 199 I/O R4 P117 331
I/O E2 P153 65
I/O G14 P57 202 I/O R3 P119 334
GND F3 P154 -
I/O G15 P58 205 I/O (D0, DIN) P4 P120 337
I/O (A12) E3 P157 68
I/O G16 P59 208 I/O, SGCK4 T2 P121 340
I/O (A13) C1 P158 71
I/O H16 P60 211 (DOUT)
I/O C2 P160 74
I/O (INIT) H15 P61 214 CCLK R2 P122 -
I/O D3 P161 77
VCC H14 P62 - VCC P3 P123 -
I/O (A14) B1 P162 80
GND J14 P63 - O, TDO T1 P124 0
I/O, SGCK1 (A15) B2 P163 83
I/O J15 P64 217 GND N3 P125 -
VCC C3 P164 -
I/O J16 P65 220 I/O (A0, WS) R1 P126 2
GND C4 P1 -
I/O K16 P66 223 I/O, PGCK4 (A1) P2 P127 5
I/O, PGCK1 (A16) B3 P2 86
I/O K15 P67 226 I/O N2 P128 8
I/O (A17) A1 P3 89
I/O K14 P68 229 I/O M3 P130 11
I/O A2 P4 92
I/O L16 P69 232 I/O (CS1, A2) P1 P131 14
I/O C5 P5 95
I/O M16 P70 235 I/O (A3) N1 P132 17
I/O, TDI B4 P7 98
I/O L15 P71 238 GND L3 P135 -
I/O, TCK A3 P8 101
GND L14 P72 - I/O L2 P136 20
GND C6 P10 -
I/O P16 P75 241 I/O L1 P137 23
I/O B5 P11 104
I/O M14 P76 244 I/O (A4) K3 P138 26
I/O B6 P12 107
I/O N15 P77 247 I/O (A5) K2 P139 29
I/O, TMS A5 P13 110
I/O P15 P78 250 I/O K1 P140 32
I/O C7 P14 113
I/O N14 P79 253 I/O J1 P141 35
I/O B7 P15 116
I/O, SGCK3 R16 P80 256 I/O (A6) J2 P142 38
I/O A6 P16 119
GND P14 P81 - I/O (A7) J3 P143 41
I/O A7 P17 122
DONE R15 P82 - GND H2 P144 -
I/O A8 P18 125
GND C8 P19 - VCC P13 P83 - 8/13/97 10
PROGRAM R14 P84 -
VCC B8 P20 - Additional XC4005E Package
I/O C9 P21 128 I/O (D7) T16 P85 259
I/O B9 P22 131 I/O, PGCK3 T15 P86 262 Pins
I/O R13 P87 265 PG156
I/O A9 P23 134
I/O P12 P89 268 N.C. Pins
I/O B10 P24 137
I/O (D6) T14 P90 271 A4 A12 D1 D2
I/O C10 P26 140
I/O T13 P91 274 D16 E15 M1 M2
I/O A10 P27 143
GND P11 P94 - M15 N16 R5 R12
I/O A11 P28 146
I/O R11 P95 277 T12 - - -
I/O B11 P29 149
I/O T11 P96 280 8/14/97
GND C11 P30 -
I/O B12 P32 152 I/O (D5) T10 P97 283
I/O A13 P33 155 I/O (CS0) P10 P98 286
I/O A14 P34 158 I/O R10 P99 289 CB164
I/O C12 P35 161 I/O T9 P100 292 N.C. Pins
I/O B13 P37 164 I/O (D4) R9 P101 295 P6 P9 P25 P31
I/O, SGCK2 B14 P38 167 I/O P9 P102 298 P36 P47 P51 P52
O (M1) A15 P39 170 VCC R8 P103 - P73 P74 P88 P92
GND C13 P40 - GND P8 P104 - P93 P114 P118 P129
I (M0) A16 P41 173 I/O (D3) T8 P105 301 P133 P134 P155 P156
VCC C14 P42 - I/O (RS) T7 P106 304 P159 - - -
I (M2) B15 P43 174 I/O T6 P107 307 8/14/97
I/O, PGCK2 B16 P44 175 I/O R7 P108 310
May 19, 1998 (Version 2.1) 10-45


VCC J4 P183 - I/O B14 P41 224 I/O U15 P105 376
I/O (A8) J3 P184 62 I/O A16 P42 227 I/O (D6) V17 P106 379
I/O (A9) J2 P185 65 I/O B15 P43 230 I/O V16 P107 382
I/O (19) J1 P186 68 I/O C14 P44 233 I/O T13 P108 385
I/O (18) H1 P187 71 I/O A17 P45 236 I/O U14 P109 388
I/O H2 P188 74 I/O, SGCK2 B16 P46 239 I/O V15 P110 391
I/O H3 P189 77 O (M1) C15 P47 242 I/O V14 P111 394
I/O (A10) G1 P190 80 GND D15 P48 - GND T12 P112 -
I/O (A11) G2 P191 83 I (M0) A18 P49 245 I/O U13 P113 397
I/O F1 P192 86 VCC D16 P50 - I/O V13 P114 400
I/O E1 P193 89 I (M2) C16 P51 246 I/O (D5) U12 P115 403
GND G3 P194 - I/O, PGCK2 B17 P52 247 I/O (CS0) V12 P116 406
I/O F2 P195 92 I/O (HDC) E16 P53 250 I/O T11 P117 409
I/O D1 P196 95 I/O C17 P55 253 I/O U11 P118 412
I/O C1 P197 98 I/O D17 P56 256 I/O V11 P119 415
I/O E2 P198 101 I/O B18 P57 259 I/O V10 P120 418
I/O (A12) F3 P199 104 I/O (LDC) E17 P58 262 I/O (D4) U10 P121 421
I/O (A13) D2 P200 107 I/O F16 P59 265 I/O T10 P122 424
I/O B1 P201 110 I/O C18 P60 268 VCC R10 P123 -
I/O E3 P202 113 I/O D18 P61 271 GND R9 P124 -
I/O (A14) C2 P203 116 I/O F17 P62 274 I/O (D3) T9 P125 427
I/O, SGCK1 (A15) B2 P204 119 GND G16 P63 - I/O (RS) U9 P126 430
VCC D3 P205 - I/O E18 P64 277 I/O V9 P127 433
GND D4 P1 - I/O F18 P65 280 I/O V8 P128 436
I/O, PGCK1 (A16) C3 P2 122 I/O G17 P66 283 I/O U8 P129 439
I/O (A17) C4 P3 125 I/O G18 P67 286 I/O T8 P130 442
I/O B3 P4 128 I/O H16 P68 289 I/O (D2) V7 P131 445
I/O C5 P6 131 I/O H17 P69 292 I/O U7 P132 448
I/O, TDI A2 P7 134 I/O H18 P70 295 I/O V6 P133 451
I/O, TCK B4 P8 137 I/O J18 P71 298 I/O U6 P134 454
I/O C6 P9 140 I/O J17 P72 301 GND T7 P135 -
I/O A3 P10 143 I/O (INIT) J16 P73 304 I/O V5 P136 457
I/O B5 P11 146 VCC J15 P74 - I/O V4 P137 460
I/O B6 P12 149 GND K15 P75 - I/O U5 P138 463
GND C7 P13 - I/O K16 P76 307 I/O T6 P139 466
I/O A4 P14 152 I/O K17 P77 310 I/O (D1) V3 P140 469
I/O A5 P15 155 I/O K18 P78 313 I/O (RCLK, V2 P141 472
I/O L18 P79 316 RDY/BUSY)
I/O, TMS B7 P16 158
I/O L17 P80 319 I/O U4 P142 475
I/O A6 P17 161
I/O L16 P81 322 I/O T5 P143 478
I/O C8 P18 164
I/O M18 P82 325 I/O (D0, DIN) U3 P144 481
I/O A7 P19 167
I/O M17 P83 328 I/O, SGCK4 T4 P145 484
I/O B8 P20 170
I/O N18 P84 331 (DOUT)
I/O A8 P21 173
I/O P18 P85 334 CCLK V1 P146 -
I/O B9 P22 176
GND M16 P86 - VCC R4 P147 -
I/O C9 P23 179
I/O N17 P87 337 O, TDO U2 P148 0
GND D9 P24 -
I/O R18 P88 340 GND R3 P149 -
VCC D10 P25 -
I/O T18 P89 343 I/O (A0, WS) T3 P150 2
I/O C10 P26 182
I/O P17 P90 346 I/O, PGCK4 (A1) U1 P151 5
I/O B10 P27 185
I/O N16 P91 349 I/O P3 P153 8
I/O A9 P28 188
I/O T17 P92 352 I/O R2 P154 11
I/O A10 P29 191
I/O R17 P93 355 I/O (CS1, A2) T2 P155 14
I/O A11 P30 194
I/O P16 P94 358 I/O (A3) N3 P156 17
I/O C11 P31 197
I/O U18 P95 361 I/O P2 P157 20
I/O B11 P32 200
I/O, SGCK3 T16 P96 364 I/O T1 P158 23
I/O A12 P33 203
GND R16 P97 - I/O R1 P159 26
I/O B12 P34 206
DONE U17 P98 - I/O N2 P160 29
I/O A13 P35 209
VCC R15 P99 - GND M3 P161 -
GND C12 P36 -
PROGRAM V18 P100 - I/O P1 P162 32
I/O B13 P37 212
I/O (D7) T15 P101 367 I/O N1 P163 35
I/O A14 P38 215
I/O, PGCK3 U16 P102 370 I/O (A4) M2 P164 38
I/O A15 P39 218
I/O T14 P104 373 I/O (A5) M1 P165 41
I/O C13 P40 221
I/O L3 P166 44
10-46 May 19, 1998 (Version 2.1)

XC4010E PG CB Bndry Additional XC4010E Package
Pad Name 191† 196 Scan Pins
I/O L2 P167 47
CB196
I/O L1 P168 50
N.C. Pins
I/O K1 P169 53
P5 P54 P103 P152
I/O (A6) K2 P170 56
P192 - - -
I/O (A7) K3 P171 59
8/14/97
GND K4 P172 -
8/14/97

VCC J4 P201 - I/O A8 P24 209 I/O G17 P75 337
I/O (A8) J3 P202 74 I/O B9 P25 212 I/O G18 P76 340
I/O (A9) J2 P203 77 I/O C9 P26 215 I/O H16 P77 343
I/O J1 P204 80 GND D9 P27 - I/O H17 P78 346
I/O H1 P205 83 VCC D10 P28 - I/O G15 P79 349
I/O H2 P206 86 I/O C10 P29 218 I/O H15 P80 352
I/O H3 P207 89 I/O B10 P30 221 I/O H18 P81 355
I/O (A10) G1 P208 92 I/O A9 P31 224 I/O J18 P82 358
I/O (A11) G2 P209 95 I/O A10 P32 227 I/O J17 P83 361
VCC - P210 - I/O A11 P33 230 I/O (INIT) J16 P84 364
I/O H4 P211 98 I/O C11 P34 233 VCC J15 P85 -
I/O G4 P212 101 I/O D11 P35 236 GND K15 P86 -
I/O F1 P213 104 I/O D12 P36 239 I/O K16 P87 367
I/O E1 P214 107 VCC - P37 - I/O K17 P88 370
GND G3 P215 - I/O B11 P38 242 I/O K18 P89 373
I/O F2 P216 110 I/O A12 P39 245 I/O L18 P90 376
I/O D1 P217 113 I/O B12 P40 248 I/O L17 P91 379
I/O C1 P218 116 I/O A13 P41 251 I/O L16 P92 382
I/O E2 P219 119 GND C12 P42 - I/O L15 P93 385
I/O (A12) F3 P220 122 I/O D13 P43 254 I/O M15 P94 388
I/O (A13) D2 P221 125 I/O D14 P44 257 VCC - P95 -
I/O F4 P222 128 I/O B13 P45 260 I/O M18 P96 391
I/O E4 P223 131 I/O A14 P46 263 I/O M17 P97 394
I/O B1 P224 134 I/O A15 P47 266 I/O N18 P98 397
I/O E3 P225 137 I/O C13 P48 269 I/O P18 P99 400
I/O (A14) C2 P226 140 I/O B14 P49 272 GND M16 P100 - 10
I/O, SGCK1 (A15) B2 P227 143 I/O A16 P50 275 I/O N15 P101 403
VCC D3 P228 - I/O B15 P51 278 I/O P15 P102 406
GND D4 P1 - I/O C14 P52 281 I/O N17 P103 409
I/O, PGCK1(A16) C3 P2 146 I/O A17 P53 284 I/O R18 P104 412
I/O (A17) C4 P3 149 I/O, SGCK2 B16 P54 287 I/O T18 P105 415
I/O B3 P4 152 O (M1) C15 P55 290 I/O P17 P106 418
I/O C5 P5 155 GND D15 P56 - I/O N16 P107 421
I/O, TDI A2 P6 158 I (M0) A18 P57 293 I/O T17 P108 424
I/O, TCK B4 P7 161 VCC D16 P58 - I/O R17 P109 427
I/O C6 P8 164 I (M2) C16 P59 294 I/O P16 P110 430
I/O A3 P9 167 I/O, PGCK2 B17 P60 295 I/O U18 P111 433
I/O B5 P10 170 I/O (HDC) E16 P61 298 I/O, SGCK3 T16 P112 436
I/O B6 P11 173 I/O C17 P62 301 GND R16 P113 -
I/O D5 P12 176 I/O D17 P63 304 DONE U17 P114 -
I/O D6 P13 179 I/O B18 P64 307 VCC R15 P115 -
GND C7 P14 - I/O (LDC) E17 P65 310 PROGRAM V18 P116 -
I/O A4 P15 182 I/O F16 P66 313 I/O (D7) T15 P117 439
I/O A5 P16 185 I/O C18 P67 316 I/O, PGCK3 U16 P118 442
I/O, TMS B7 P17 188 I/O D18 P68 319 I/O T14 P119 445
I/O A6 P18 191 I/O F17 P69 322 I/O U15 P120 448
I/O D7 P19 194 I/O E15 P70 325 I/O R14 P121 451
I/O D8 P20 197 I/O F15 P71 328 I/O R13 P122 454
I/O C8 P21 200 GND G16 P72 - I/O (D6) V17 P123 457
I/O A7 P22 203 I/O E18 P73 331 I/O V16 P124 460
I/O B8 P23 206 I/O F18 P74 334 I/O T13 P125 463
May 19, 1998 (Version 2.1) 10-47


I/O U14 P126 466 I/O V6 P153 535 I/O (CS1, A2) T2 P178 14
I/O V15 P127 469 I/O U6 P154 538 I/O (A3) N3 P179 17
I/O V14 P128 472 I/O R8 P155 541 I/O P4 P180 20
GND T12 P129 - I/O R7 P156 544 I/O N4 P181 23
I/O R12 P130 475 GND T7 P157 - I/O P2 P182 26
I/O R11 P131 478 I/O R6 P158 547 I/O T1 P183 29
I/O U13 P132 481 I/O R5 P159 550 I/O R1 P184 32
I/O V13 P133 484 I/O V5 P160 553 I/O N2 P185 35
I/O (D5) U12 P134 487 I/O V4 P161 556 GND M3 P186 -
I/O (CS0) V12 P135 490 I/O U5 P162 559 I/O P1 P187 38
I/O T11 P136 493 I/O T6 P163 562 I/O N1 P188 41
I/O U11 P137 496 I/O (D1) V3 P164 565 I/O M4 P189 44
I/O V11 P138 499 I/O (RCLK, V2 P165 568 I/O L4 P190 47
I/O V10 P139 502 RDY/BUSY) VCC - P191 -
I/O (D4) U10 P140 505 I/O U4 P166 571 I/O (A4) M2 P192 50
I/O T10 P141 508 I/O T5 P167 574 I/O (A5) M1 P193 53
VCC R10 P142 - I/O (D0, DIN) U3 P168 577 I/O L3 P194 56
GND R9 P143 - I/O, SGCK4 T4 P169 580 I/O L2 P195 59
I/O (D3) T9 P144 511 (DOUT) I/O L1 P196 62
I/O (RS) U9 P145 514 CCLK V1 P170 - I/O K1 P197 65
I/O V9 P146 517 VCC R4 P171 - I/O (A6) K2 P198 68
I/O V8 P147 520 O, TDO U2 P172 0 I/O (A7) K3 P199 71
I/O U8 P148 523 GND R3 P173 - GND K4 P200 -
I/O T8 P149 526 I/O (A0, WS) T3 P174 2 8/14/97
I/O (D2) V7 P150 529 I/O, PGCK4 (A1) U1 P175 5
I/O U7 P151 532 I/O P3 P176 8
VCC - P152 - I/O R2 P177 11

XC4025E CB PG Bndry XC4025E CB PG Bndry XC4025E CB PG Bndry
Pad Name 228 299 Scan Pad Name 228 299 Scan Pad Name 228 299 Scan
VCC P201 K1 - I/O P223 C2 179 I/O - E9 257
I/O (A8) P202 K2 98 I/O P224 F5 182 I/O - A7 260
I/O (A9) P203 K3 101 I/O P225 E4 185 I/O - D9 263
I/O P204 K5 104 I/O (A14) P226 D3 188 I/O P19 B8 266
I/O P205 K4 107 I/O, SGCK1 (A15) P227 C3 191 I/O P20 A8 269
I/O P206 J1 110 VCC P228 A2 - I/O P21 C9 272
I/O P207 J2 113 GND P1 B1 - I/O P22 B9 275
I/O (A10) P208 H1 116 I/O, PGCK1 (A16) P2 D4 194 I/O P23 E10 278
I/O (A11) P209 J3 119 I/O (A17) P3 B2 197 I/O P24 A9 281
I/O - J4 122 I/O P4 B3 200 I/O P25 D10 284
I/O - J5 125 I/O P5 E6 203 I/O P26 C10 287
I/O - H2 128 I/O, TDI P6 D5 206 GND P27 A10 -
I/O - G1 131 I/O, TCK P7 C4 209 VCC P28 A11 -
VCC P210 E1 - I/O - A3 212 I/O P29 B10 290
I/O P211 H3 134 I/O - D6 215 I/O P30 B11 293
I/O P212 G2 137 I/O P8 E7 218 I/O P31 C11 296
I/O P213 H4 140 I/O P9 B4 221 I/O P32 E11 299
I/O P214 F2 143 I/O P10 C5 224 I/O P33 D11 302
GND P215 F1 - I/O P11 A4 227 I/O P34 A12 305
I/O - H5 146 I/O P12 D7 230 I/O - B12 308
I/O - G3 149 I/O P13 C6 233 I/O - A13 311
I/O P216 D1 152 I/O - E8 236 I/O - C12 314
I/O P217 G4 155 I/O - B5 239 I/O - D12 317
I/O P218 E2 158 GND P14 A5 - I/O P35 E12 320
I/O P219 F3 161 I/O P15 B6 242 I/O P36 B13 323
I/O (A12) P220 G5 164 I/O P16 D8 245 VCC P37 A16 -
I/O (A13) P221 C1 167 I/O, TMS P17 C7 248 I/O P38 A14 326
I/O - F4 170 I/O P18 B7 251 I/O P39 C13 329
I/O - E3 173 VCC - A6 - I/O P40 B14 332
I/O P222 D2 176 I/O - C8 254 I/O P41 D13 335
10-48 May 19, 1998 (Version 2.1)

XC4025E CB PG Bndry XC4025E CB PG Bndry XC4025E CB PG Bndry
Pad Name 228 299 Scan Pad Name 228 299 Scan Pad Name 228 299 Scan
GND P42 A15 - I/O - N20 505 I/O (D4) P140 V11 673
I/O - B15 338 I/O - M18 508 I/O P141 W11 676
I/O - E13 341 I/O - M17 511 VCC P142 X10 -
I/O P43 C14 344 I/O - M16 514 GND P143 X11 -
I/O P44 A17 347 I/O P93 N19 517 I/O (D3) P144 W10 679
I/O P45 D14 350 I/O P94 P20 520 I/O (RS) P145 V10 682
I/O P46 B16 353 VCC P95 T20 - I/O P146 T10 685
I/O P47 C15 356 I/O P96 N18 523 I/O P147 U10 688
I/O P48 E14 359 I/O P97 P19 526 I/O P148 X9 691
I/O - A18 362 I/O P98 N17 529 I/O P149 W9 694
I/O - D15 365 I/O P99 R19 532 I/O - X8 697
I/O P49 C16 368 GND P100 R20 - I/O - V9 700
I/O P50 B17 371 I/O - N16 535 I/O - U9 703
I/O P51 B18 374 I/O - P18 538 I/O - T9 706
I/O P52 E15 377 I/O P101 U20 541 I/O (D2) P150 W8 709
I/O P53 D16 380 I/O P102 P17 544 I/O P151 X7 712
I/O, SGCK2 P54 C17 383 I/O P103 T19 547 VCC P152 X5 -
O (M1) P55 A20 386 I/O P104 R18 550 I/O P153 V8 715
GND P56 A19 - I/O P105 P16 553 I/O P154 W7 718
I (M0) P57 C18 389 I/O P106 V20 556 I/O P155 U8 721
VCC P58 B20 - I/O - R17 559 I/O P156 W6 724
I (M2) P59 D17 390 I/O - T18 562 GND P157 X6 -
I/O, PGCK2 P60 B19 391 I/O P107 U19 565 I/O - T8 727
I/O (HDC) P61 C19 394 I/O P108 V19 568 I/O - V7 730
I/O P62 F16 397 I/O P109 R16 571 I/O P158 X4 733
I/O P63 E17 400 I/O P110 T17 574 I/O P159 U7 736
I/O P64 D18 403 I/O P111 U18 577 I/O P160 W5 739
I/O (LDC) P65 C20 406 I/O, SGCK3 P112 X20 580 I/O P161 V6 742
I/O - F17 409 GND P113 W20 - I/O P162 T7 745
I/O - G16 412 DONE P114 V18 - I/O P163 X3 748
I/O P66 D19 415 VCC P115 X19 - I/O (D1) P164 U6 751
I/O P67 E18 418 PROGRAM P116 U17 - I/O (RCLK, P165 V5 754
I/O P68 D20 421 I/O (D7) P117 W19 583 RDY/BUSY)
I/O P69 G17 424 I/O, PGCK3 P118 W18 586 I/O - W4 757
I/O P70 F18 427 I/O P119 T15 589 I/O - W3 760
I/O P71 H16 430 I/O P120 U16 592 I/O P166 T6 763
I/O - E19 433 I/O P121 V17 595 I/O P167 U5 766
I/O - F19 436 I/O P122 X18 598 I/O (D0, DIN) P168 V4 769
GND P72 E20 - I/O - U15 601 I/O, SGCK4 P169 X1 772 10
I/O P73 H17 439 I/O - T14 604 (DOUT)
I/O P74 G18 442 I/O (D6) P123 W17 607 CCLK P170 V3 -
I/O P75 G19 445 I/O P124 V16 610 VCC P171 VCC* -
I/O P76 H18 448 I/O P125 X17 613 O, TDO P172 U4 0
VCC - VCC* - I/O P126 U14 616 GND P173 GND* -
I/O P77 J16 451 I/O P127 V15 619 I/O (A0, WS) P174 W2 2
I/O P78 G20 454 I/O P128 T13 622 I/O, PGCK4 (A1) P175 V2 5
I/O - J17 457 I/O - W16 625 I/O P176 R5 8
I/O - H19 460 I/O - W15 628 I/O P177 T4 11
I/O - H20 463 GND P129 X16 - I/O (CS1, A2) P178 U3 14
I/O - J18 466 I/O P130 U13 631 I/O (A3) P179 V1 17
I/O P79 J19 469 I/O P131 V14 634 I/O P180 R4 20
I/O P80 K16 472 I/O P132 W14 637 I/O P181 P5 23
I/O P81 J20 475 I/O P133 V13 640 I/O P182 U2 26
I/O P82 K17 478 VCC - X15 - I/O P183 T3 29
I/O P83 K18 481 I/O (D5) P134 T12 643 I/O P184 U1 32
I/O (INIT) P84 K19 484 I/O (CS0) P135 X14 646 I/O P185 P4 35
VCC P85 L20 - I/O - U12 649 I/O - R3 38
GND P86 K20 - I/O - W13 652 I/O - N5 41
I/O P87 L19 487 I/O - X13 655 I/O - T2 44
I/O P88 L18 490 I/O - V12 658 I/O - R2 47
I/O P89 L16 493 I/O P136 W12 661 GND P186 T1 -
I/O P90 L17 496 I/O P137 T11 664 I/O P187 N4 50
I/O P91 M20 499 I/O P138 X12 667 I/O P188 P3 53
I/O P92 M19 502 I/O P139 U11 670 I/O P189 P2 56
May 19, 1998 (Version 2.1) 10-49

XC4025E CB PG Bndry
Pad Name 228 299 Scan
I/O P190 N3 59
VCC P191 R1 -
I/O - M5 62
I/O - P1 65
I/O - M4 68
I/O - N2 71
I/O (A4) P192 N1 74
I/O (A5) P193 M3 77
I/O P194 M2 80
I/O P195 L5 83
I/O P196 M1 86
I/O P197 L4 89
I/O (A6) P198 L3 92
I/O (A7) P199 L2 95
GND P200 L1 -
8/14/97
10-50 May 19, 1998 (Version 2.1)

Example for SMD Part: 5962-97523 01 Q X C
Generic Standard Lead Finish

Microcircuit Drawing C = Gold
(SMD) Prefix Package Type
Device Type X = Pin Grid
XC4005E = 97522 Y = Quad Flatpack
XC4010E = 97523 (Base Mark)
XC4013E = 97524 Z = Quad Flatpack
XC4025E = 97525 (Lid Mark)
QML Certified
Speed Grade
01 = -4
Example for XC4010E -4 PG 191 M

Military Temperature Only 10
XC4005E M = Military (TC = -55o C to +125o C)
XC4010E
XC4013E Number of Pins
XC4025E
Speed Grade
Package Type
PG = Ceramic Pin Grid Array
May 19, 1998 (Version 2.1) 10-51

10-52 May 19, 1998 (Version 2.1)

0
R
QPRO™ XQ4000XL Series QML
High-Reliability FPGAs
January 4, 1999 (Version 1.1) 0 10*
XQ4000X Series Features • Additional Routing Over XQ4000E

- almost twice the routing capacity for high-density
• Certified to MIL-PRF-38535 Appendix A QML designs
(Qualified Manufacturer Listing) • Buffered Interconnect for Maximum Speed
• Ceramic and plastic packages • New Latch Capability in Configurable Logic
• Also available under the following standard Blocks
microcircuit drawings (SMD) • Improved VersaRing™ I/O Interconnect for Better
- XQ4013XL 5962-98513 Fixed Pinout Flexibility
- XQ4036XL 5962-98510 - Virtually unlimited number of clock signals
- XQ4062XL 5962-98511
• Optional Multiplexer or 2-input Function
• For more information contact the Defense Supply
Center Columbus (DSCC)
Generator on Device Outputs
http://www.dscc.dla.mis/v/va/smd/smdsrch.html • 5V tolerant I/Os
• Available in -3 speed • 0.35m SRAM process
• System featured Field-Programmable Gate
Arrays Introduction
- Select-RAM™ memory: on-chip ultra-fast XQ4000X Series high-performance, high-capacity Field
RAM with Programmable Gate Arrays (FPGAs) provide the benefits
- synchronous write option of custom CMOS VLSI, while avoiding the initial cost, long
- dual-port RAM option development cycle, and inherent risk of a conventional
- Abundant flip-flops masked gate array.
- Flexible function generators The result of thirteen years of FPGA design experience and
- Dedicated high-speed carry logic feedback from thousands of customers, these FPGAs com-
- Wide edge decoders on each edge bine architectural versatility, on-chip Select-RAM memory
- Hierarchy of interconnect lines with edge-triggered and dual-port modes, increased
- Internal 3-state bus capability speed, abundant routing resources, and new, sophisticated
- 8 global low-skew clock or signal distribution soft-ware to achieve fully automated implementation of 10
networks complex, high-density, high-performance designs.
• System Performance beyond 50 MHz
Refer to the complete Commercial XC4000X Series Field
• Flexible Array Architecture Programmable Gate Arrays Data Sheet for more informa-
• Low Power Segmented Routing Architecture tion on device architecture and timing, and the latest Xilinx
• Systems-Oriented Features databook for package pinouts other than the CB228
- IEEE 1149.1-compatible boundary scan logic (included in this data sheet). (Pinouts for XQ4000XL device
support are identical to XC4000XL.)
- Individually programmable output slew rate
1
- Programmable input pull-up or pull-down resistors
- 12-mA Sink Current Per XQ4000XL Output
• Configured by Loading Binary File
- Unlimited reprogrammability
• Readback Capability
- Program verification
- Internal node observability
• Development System runs on most common
computer platforms
- - Interfaces to popular design environments
- - Fully automatic mapping, placement and routing
- - Interactive design editor for design optimization
• Highest Capacity — Over 130,000 Usable Gates

R
Table 1: XQ4000X Series High Reliability Field Progammable Gate Arrays

Max
Max. Typical Gate
Logic Logic CLB Total Number of Max.
Device RAM Bits Range Packages
Cells Gates Matrix CLBs Flip-Flops User I/O
(No Logic) (Logic and RAM)*
(No RAM)
XQ4013XL 2432 13,000 18,432 10,000-30,000 24x24 576 1,536 192 PG223, CB228
PQ240, BG256
XQ4036XL 3078 36,000 41,472 22,000-65,000 36x36 1,296 3,168 288 PG411, CB228,
HQ240, BG352
XQ4062XL 5472 62,000 73,728 40,000-130,000 48x48 2,304 5,376 384 PG475, CB228,
HQ240, BG432
* Maximum values of typical gate range includes 20% to 30% of CLBs used as RAM.
XQ4000XL Switching Characteristics

Definition of Terms
follows:

Ceramic Package +150 °C
TJ Junction temperature
Plastic Package +125 °C

R

Supply voltage relative to GND, TJ = -55°C to Plastic 3.0 3.6 V
+125°C
VCC
Supply voltage relative to GND, TC = -55°C to Ceramic 3.0 3.6 V
+125°C
Note 1: At junction temperatures above those listed as Operating Conditions, all delay parameters increase by 0.35% per ×C.
XQ4000XL DC Characteristics Over Recommended Operating Conditions

VOH
BGA, PQ, HQ, packages 10 pF
CIN Input capacitance (sample tested)
PGA packages 16 pF
10

R
XQ4000XL Global Buffer Switching Characteristic Guidelines

temperature).
Speed Grade -3
Units
Description Symbol Device Max
From pad through Global Low Skew buffer, to any clock K TGLS XQ4013XL 3.6 ns
XQ4036XL 4.8 ns
XQ4062XL 6.3 ns
From pad through Global Early buffer, to any IOB clockK. Values are for TGE XQ4013XL 2.4 ns
BUFGE #s 1, 2, 5 and 6. Add 1 - 2 ns for BUFGE #s 3, 4, 7 and 8 and for XQ4036XL 3.1 ns
all CLB clock Ks driven from any of the 8 BUFGEs, or consult TRCE. XQ4062XL 4.9 ns

R
XQ4000XL CLB Switching Characteristic Guidelines

XQ4000XL devices and expressed in nanoseconds unless otherwise noted.
Speed Grade -3
Units
C inputs via SR/H0 via H to X/Y outputs THH0O 2.5 ns
C inputs via H1 via H to X/Y outputs THH1O 2.4 ns
C inputs via DIN/H2 via H to X/Y outputs THH2O 2.5 ns
Sequential Delays
F/G inputs via H TIHCK 2.2 ns
C inputs via H0 through H THH0CK 2.0 ns
CIN input via F/G TCCK 2.3 ns
CIN input via F/G and H TCHCK 3.4 ns
10
F/G inputs via H TCKIH 0 ns
C inputs via SR/H0 through H TCKHH0 0 ns
C inputs via H1 through H TCKHH1 0 ns
C inputs via DIN/H2 through H TCKHH2 0 ns
Clock
Set/Reset Direct
Global Set/Reset
Delay from GSR input to any Q TMRQ See page 66 for TRRI
values per device.
Toggle Frequency (MHz) (for export control) FTOG 166 MHz

R
XQ4000XL RAM Synchronous (Edge-Triggered) Write Operation Guidelines

XQ4000XL devices and are expressed in nanoseconds unless otherwise noted.
Speed Grade -3
Size Symbol Min Max
Write Operation
32x1 TWCTS 9.0 ns
32x1 TWPTS 4.5 ns
32x1 TASTS 2.2 ns
32x1 TAHTS 0 ns
32x1 TDSTS 2.5 ns
DIN hold time after clock K 16x2 TDHS 0 ns
32x1 TDHTS 0 ns
32x1 TWSTS 1.8 ns
32x1 TWHTS 0 ns
32x1 TWOTS 8.1 ns
Read Operation
32x1 TRCT 6.5 ns
Data Valid after address change (no Write Enable) 16x2 TILO 1.6 ns
32x1 TIHO 2.7 ns
32x1 TIHCK 2.3 ns

R
Speed Grade -3
Dual Port RAM Units
Size Symbol Min Max
Write Operation
Note 1: Timing for16 x1 RAM option is identical to16 x 2 RAM.
10

R
XQ4000XL CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS
WCLK (K)
TWSS TWHS
WE
TDSS TDHS
DATA IN
TASS TAHS
ADDRESS
TILO TILO
TWOS
DATA OUT OLD NEW
X6461
XQ4000XL CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing
TWPDS
WCLK (K)
TWSDS TWHDS
WE
TDSDS TDHDS
DATA IN
TASDS TAHDS
ADDRESS
TILO TILO
TWODS
DATA OUT OLD NEW
X6474

R
XQ4000XL Pin-to-Pin Output Parameter Guidelines

XQ4000XL Output Flip-Flop, Clock to Out
Speed Grade -3
Units
Global Low Skew Clock to Output using OFF TICKOF XQ4013XL 8.6 ns
XQ4036XL 9.8 ns
XQ4062XL 11.3 ns
Global Early Clock to Output using OFF TICKEOF XQ4013XL 7.4 ns
Values are for BUFGE #s 3, 4, 7, and 8. Add XQ4036XL 8.1 ns
1.4 ns for BUFGE #s 1, 2, 5, and 6. XQ4062XL 9.9 ns
For output SLOW option add TSLOW All Devices 3.0 ns
Note 1: Listed above are representative values where one global clock input drives one vertical clock
line in each accessible column, and where all accessible IOB and CLB flip-flops are clocked by
the global clock net.
Note 2: Output timing is measured at ~50% VCC threshold with 50 pF external capacitive load. For dif-
ferent loads, see graph below.
XQ4000XL Output Mux, Clock to Out

Speed Grade -3
Units
Global Low Skew Clock to Output using OFF TICKOF XQ4013XL 8.8 ns
XQ4036XL 10.0 ns
XQ4062XL 11.4 ns 10
Global Early Clock to Output using OFF. Val- TICKEOF XQ4013XL 7.6 ns
ues are for BUFGE #s 3, 4, 7, and 8. Add 1.4 XQ4036XL 8.2 ns
ns for BUFGE #s 1, 2, 5, and 6. XQ4062XL 10.0 ns

R
Capacitive Load Factor

Delta Delay (ns)

delay. 0
-1
voltage and temperature and is independent of the output
slew rate control.
-2
0 20 40 60 80 100 120 140
Capacitance (pF)
X8257

R
XQ4000XL Pin-to-Pin Input Parameter Guidelines

XQ4000XL Global Low Skew Clock, Set-Up and Hold

Speed Grade -3
Units
Description Symbol Device Min
Input Setup and Hold Times Using
Global Low Skew Clock and IFF
No Delay TPSN/TPHN XQ4013XL 1.2 / 3.2 ns
XQ4036XL 1.2 / 5.5 ns
XQ4062XL 1.2 / 7.0 ns
Partial Delay TPSP/TPHP XQ4013XL 6.1 / 0.0 ns
XQ4036XL 6.4 / 1.0 ns
XQ4062XL 6.7 / 1.2 ns
Full Delay TPSD/TPHD XQ4013XL 6.4 / 0.0 ns
XQ4036XL 6. 6 / 0.0 ns
XQ4062XL 6.8 / 0.0 ns
Note 1: Setup time is measured with the fastest route and the lightest load. Hold time is measured
using the furthest distance and a reference load of one clock pin per two IOBs. Use the static
timing analyzer (TRCE) to determine the setup and hold times under given design conditions.
10

R
XQ4000XL BUFGE #s 3, 4, 7, & 8 Global Early Clock, Set-up and Hold for IFF and FCL
Speed Grade -3
No Delay XQ4013XL 1.2 / 4.7
Global Early Clock and IFF TPSEN/TPHEN XQ4036XL 1.2 / 6.7
Global Early Clock and FCL TPFSEN/TPFHEN XQ4062XL 1.2 / 8.4
Partial Delay XQ4013XL 5.4 / 0.0
Global Early Clock and IFF TPSEP/TPHEP XQ4036XL 6.4 / 0.8
Global Early Clock and FCL TPFSEP/TPFHEP XQ4062XL 8.4 / 1.5
Full Delay XQ4013XL 12.0 / 0.0
Global Early Clock and IFF TPSED/TPHED XQ4036XL 13.8 / 0.0
XQ4062XL 13.1 / 0.0
Note 1: Setup time is measured with the fastest route and the lightest load. Hold time is measured using
the furthest distance and a reference load of one clock pin per two IOBs. Use the static timing
analyzer(TRCE) to determine the setup and hold times under given design conditions.

R
XQ4000XL BUFGE #s 1, 2, 5, & 6 Global Early Clock, Set-up and Hold for IFF and FCL
Speed Grade -3
No Delay XQ4013XL 1.2 / 4.7
Global Early Clock and IFF TPSEN/TPHEN XQ4036XL 1.2 / 6.7
Global Early Clock and FCL TPFSEN/TPFHEN XQ4062XL 1.2 / 8.4
Partial Delay XQ4013XL 6.4 / 0.0
Global Early Clock and IFF TPSEP/TPHEP XQ4036XL 7.0 / 0.0
Global Early Clock and FCL TPFSEP/TPFHEP XQ4062XL 9.0 / 0.8
Full Delay XQ4013XL 10.0 / 0.0
Global Early Clock and IFF TPSED/TPHED XQ4036XL 12.2 / 0.0
XQ4062XL 13.1 / 0.0
using the furthest distance and a reference load of one clock pin per two IOBs. Use the
static timing analyzer(TRCE) to determine the setup and hold times under given design
conditions.
10

R
XQ4000XL IOB Input Switching Characteristic Guidelines

Speed Grade -3
Units
Clocks
Clock Enable (EC) to Clock (IK) TECIK All devices 0.3 ns
Delay from FCL enable (OK) active edge to IFF clock (IK) TOKIK All devices 1.7 ns
active edge
Setup Times
Pad to Clock (IK), no delay TPICK All devices 1.7 ns
Pad to Clock (IK), via transparent Fast Capture Latch, no TPICKF All devices 2.3 ns
delay
Pad to Fast Capture Latch Enable (OK), no delay TPOCK All devices 0.7 ns
Hold Times
All Hold Times All devices 0 ns
Global Set/Reset
Minimum GSR Pulse Width TMRW All devices 19.8 ns
Delay from GSR input to any Q TRRI XQ4013XL 15.9 ns
XQ4036XL 22.5 ns
XQ4062XL 29.1 ns
Pad to I1, I2 TPID All devices 1.6 ns
Pad to I1, I2 via transparent input latch, no delay TPLI All devices 2.6 ns
Pad to I1, I2 via transparent FCL and input latch, no delay TPFLI All devices 3.1 ns
Clock (IK) to I1, I2 (flip-flop) TIKRI All devices 1.8 ns
Clock (IK) to I1, I2 (latch enable, active Low) TIKLI All devices 1.9 ns
FCL Enable (OK) active edge to I1, I2 TOKLI All devices 3.6 ns

R
XQ4000XL IOB Output Switching Characteristic Guidelines

-3
Units
Clocks
Propagation Delays
Clock (OK) to Pad TOKPOF 5.0 ns
Output (O) to Pad TOPF 4.1 ns
3-state to Pad hi-Z (slew-rate independent) TTSHZ 4.4 ns
3-state to Pad active and valid TTSONF 4.1 ns
Output (O) to Pad via Fast Output MUX TOFPF 5.5 ns
Select (OK) to Pad via Fast MUX TOKFPF 5.1 ns
Output (O) to clock (OK) hold time TOKO 0.0 ns
Clock Enable (EC) to clock (OK) setup time TECOK 0.0 ns
Clock Enable (EC) to clock (OK) hold time TOKEC 0.3 ns
Global Set/Reset
Delay from GSR input to any Pad TRPO
XQ4013XL 20.5 ns
XQ4036XL 27.1 ns 10
XQ4062XL 33.7 ns
For output SLOW option add TSLOW 3.0 ns
Note 1: Output timing is measured at ~50% VCC threshold, with 50 pF external capacitive loads.

R
Pinouts
CB228 Package for PIN_NAME CB228
XQ4013XL/4036XL/4062XL IO P43
IO P44
PIN_NAME CB228
IO P45
VTT
IO P46
VSS P1
IO P47
BUFGP_TL_A16_GCK1_IO P2
IO P48
A17_IO P3
IO P49
IO P4
IO P50
IO P5
IO P51
TDI_IO P6
IO P52
TCK_IO P7
IO P53
IO P8
BUFGS_BL_GCK2_IO P54
IO P9
M1 P55
IO P10
VSS P56
IO P11
M0 P57
IO P12
VCC P58
IO P13
M2 P59
VSS P14
BUFGP_BL_GCK3_IO P60
IO_FCLK1 P15
HDC_IO P61
IO P16
IO P62
TMS_IO P17
IO P63
IO P18
IO P64
IO P19
LDC_IO P65
IO P20
IO P66
IO P21
IO P67
IO P22
IO P68
IO P23
IO P69
IO P24
IO P70
IO P25
IO P71
IO P26
VSS P72
VSS P27
IO P73
VCC P28
IO P74
IO P29
IO P75
IO P30
IO P76
IO P31
IO P77
IO P32
IO P78
IO P33
IO P79
IO P34
IO P80
IO P35
IO P81
IO P36
IO P82
VCC P37
IO P83
IO P38
/ERR_INIT_IO P84
IO P39
VCC P85
IO P40
VSS P86
IO_FCLK2 P41
IO P87
VSS P42
IO P88

R
PIN_NAME CB228 PIN_NAME CB228

IO P89 IO P137
IO P90 IO P138
IO P91 IO P139
IO P92 D4_IO P140
IO P93 IO P141
IO P94 VCC P142
VCC P95 VSS P143
IO P96 D3_IO P144
IO P97 /RS_IO P145
IO P98 IO P146
IO P99 IO P147
VSS P100 IO P148
IO P101 IO P149
IO P102 D2_IO P150
IO P103 IO P151
IO P104 VCC P152
IO P105 IO P153
IO P106 IO_FCLK4 P154
IO P107 IO P155
IO P108 IO P156
IO P109 VSS P157
IO P110 IO P158
IO P111 IO P159
BUFGS_BR_GCK4_IO P112 IO P160
VSS P113 IO P161
DONE P114 IO P162
VCC P115 IO P163
/PROG P116 D1_IO P164
D7_IO P117 BUSY_/RDY_RCLK_IO P165
10
BUFGP_BR_GCK5_IO P118 IO P166
IO P119 IO P167
IO P120 D0_DIN_IO P168
IO P121 BUFGS_TR_GCK6_DOUT_IO P169
IO P122 CCLK P170
D6_IO P123 VCC P171
IO P124 TDO P172
IO P125 VSS P173
IO P126 A0_/WS_IO P174
IO P127 BUFGP_TR_GCK7_A1_IO P175
IO P128 IO P176
VSS P129 IO P177
IO P130 CSI_A2_IO P178
IO P131 A3_IO P179
IO_FCLK3 P132 IO P180
IO P133 IO P181
D5_IO P134 IO P182
/CS0_IO P135 IO P183
IO P136 IO P184

R
PIN_NAME CB228
IO P185
VSS P186
IO P187
IO P188
IO P189
IO P190
VCC P191
A4_IO P192
A5_IO P193
IO P194
IO P195
A21_IO P196
A20_IO P197
A6_IO P198
A7_IO P199
VSS P200
VCC P201
A8_IO P202
A9_IO P203
A19_IO P204
A18_IO P205
IO P206
IO P207
A10_IO P208
A11_IO P209
VCC P210
IO P211
IO P212
IO P213
IO P214
VSS P215
IO P216
IO P217
IO P218
IO P219
A12_IO P220
A13_IO P221
IO P222
IO P223
IO P224
IO P225
A14_IO P226
BUFGS_TL_GCK8_A15_IO P227
VCC P228

R
XQ 4062XL -3 PG 475 M
Mil-PRF-38535
Temperature Range
(QML) Processed
M = Military Ceramic (TC = -55oC to +125 oC)
N = Military Plastic (TJ = -55°C to +125°C)
Device Type
XQ4062XL
XQ4036XL
Number of Pins
XQ4013XL

PG = Ceramic Pin Grid Array
PQ/HQ = Plastic Quad Flat Back
BG = Plastic Ball Grid Array
Revision History
Date Version Description
5/98 1.0 Original document release.
12/98 1.1 Addition of new packages, clarification of parameters.
10

R

1
R
QPRO XQR4000XL Radiation Hardened
™
October 5, 1998 (Version 1.0) 1 10* Preliminary Product Specification
XQR4000XL Series Features - IEEE 1149.1-compatible boundary scan logic

support
• Radiation Hardened FPGAs for space and satellite - Individually programmable output slew rate
applications - Programmable input pull-up or pull-down resistors
• Guaranteed Total Ionizing Dose - 12-mA sink current per output
• Latch-up Immune • Configured by Loading Binary File
• Low Soft Upset Rate - Unlimited reprogrammability
• Guaranteed to meet full electrical specifications over • Readback Capability
-55oC to +125oC - Program verification
• Available in -3 speed - Internal node observability
• System featured Field-Programmable Gate Arrays • Development System runs on most common computer
- Select-RAMTM memory: on-chip ultra-fast RAM with platforms
- synchronous write option - Interfaces to popular design environments
- dual-port RAM option - Fully automatic mapping, placement and routing
- Abundant flip-flops - Interactive design editor for design optimization
- Flexible function generators • Highest capacity - over 130,000 usable gates
- Dedicated high-speed carry logic • Buffered Interconnect for Maximum Speed
- Wide edge decoders on each edge • New Latch Capability in Configurable Logic Blocks
- Hierarchy of interconnect lines • Improved VersaRing™ I/O Interconnect for Better Fixed
- Internal 3-state bus capability Pinout Flexibility
- 8 global low-skew clock or signal distribution - Virtually unlimited number of clock signals
networks • Optional Multiplexer or 2-input Function Generator on
• System Performance beyond 60 MHz Device Outputs
• Flexible Array Architecture • 5V tolerant I/Os
• Low Power Segmented Routing Architecture • Advanced 0.35µ process
• Systems-Oriented Features • Processed on Xilinx’s QML Line
10
Table 2: XQR4000X Series Radiation Hardened Field Programmable Gate Arrays
Max. Typical
Logic Max. RAM Gate Range
Logic Gates Bits (Logic and CLB Total Number of Max.
Device Cells (No RAM) (No Logic) RAM)* Matrix CLBs Flip-Flops User I/O Packages
XQR4013XL 1,368 13,000 18,432 10,000 - 30,000 24 x 24 576 1,536 192 CB228
XQR4036XL 3,078 36,000 41,472 22,000 - 65,000 36 x 36 1,296 3,168 288 CB228
XQR4062XL 5,472 62,000 73,728 40,000 - 130,000 48 x 48 2,304 5,376 384 CB228
Note: Max values of Typical Gate Range include 20-30% of CLBs used as RAM.

R
Radiation Specifications
TID Total Ionizing Dose 60K RAD(Si)
SEL Single Event Latch-up LET> 100 MeV CM2/mg. @ +125oC 0
SEU Single Event Upset Galactic p+ (Note 1) 2.43E-8 Upsets/
Bit-Day
SEU Single Event Upset Galactic Heavy Ion (Note 1) 9.54E-8 Upsets/
Bit-Day
SEU Single Event Upset Trapped p+ (Note 1) 2.50E-7 Upsets/
Bit-Day
SEU Single Event Upset Galactic p+ (Note 2) 5.62E-8 Upsets/
Bit-Day
SEU Single Event Upset Galactic Heavy Ion (Note 2) 2.43E-7 Upsets/
Bit-Day
Note 1: 680 Km LEO, 98o Inclination, 100 Mil Al Shielding
Note 2: 35,000 Km GEO, 0o Inclination, 100 Mil Al Shielding
Note 3: Simulations done using Space Radiation Version 2.5 code from Severn Communication Corp.
XQR4000XL Switching Characteristics

Definition of Terms
follows:

R

TJ Junction temperature +150 °C

VCC Supply voltage relative to GND, TC = -55°C to +125°C 3.0 3.6 V
XQR4000XL DC Characteristics Over Recommended Operating Conditions

VOH
High-level output voltage @ IOH = -500 µA, (LVCMOS) 90% VCC V 10

R
XQR4000XL Global Buffer Switching Characteristic Guidelines

temperature).
Speed Grade -3
Units
From pad through Global Low Skew buffer, to any clock K TGLS XQR4013XL 3.6 ns
XQR4036XL 4.8 ns
XQR4062XL 6.3 ns
From pad through Global Early buffer, to any IOB clockK. Values are for TGE XQR4013XL 2.4 ns
BUFGE #s 1, 2, 5 and 6. Add 1 - 2 ns for BUFGE #s 3, 4, 7 and 8 and for XQR4036XL 3.1 ns
all CLB clock Ks driven from any of the 8 BUFGEs, or consult TRCE. XQR4062XL 4.9 ns

R
XQR4000XL CLB Switching Characteristic Guidelines

XQR4000XL devices and expressed in nanoseconds unless otherwise noted.
Speed Grade -3
Units
C inputs via SR/H0 via H to X/Y outputs THH0O 2.5 ns
C inputs via H1 via H to X/Y outputs THH1O 2.4 ns
C inputs via DIN/H2 via H to X/Y outputs THH2O 2.5 ns
Sequential Delays
F/G inputs via H TIHCK 2.3 ns
CIN input via F/G TCCK 2.3 ns
CIN input via F/G and H TCHCK 3.4 ns
10
F/G inputs via H TCKIH 0 ns
C inputs via SR/H0 through H TCKHH0 0 ns
C inputs via H1 through H TCKHH1 0 ns
C inputs via DIN/H2 through H TCKHH2 0 ns
Clock
Set/Reset Direct
Global Set/Reset
Delay from GSR input to any Q TMRQ See page 86 for TRRI
values per device.
Toggle Frequency (MHz) (for export control) FTOG 166 MHz

R
XQR4000XL RAM Synchronous (Edge-Triggered) Write Operation Guidelines

XQR4000XL devices and are expressed in nanoseconds unless otherwise noted.
Speed Grade -3
Size Symbol Min Max
Write Operation
32x1 TWCTS 9.0 ns
32x1 TWPTS 4.5 ns
32x1 TASTS 2.2 ns
32x1 TAHTS 0 ns
32x1 TDSTS 2.5 ns
DIN hold time after clock K 16x2 TDHS 0 ns
32x1 TDHTS 0 ns
32x1 TWSTS 1.8 ns
32x1 TWHTS 0 ns
32x1 TWOTS 8.1 ns
Read Operation
32x1 TRCT 6.5 ns
Data Valid after address change (no Write Enable) 16x2 TILO 1.6 ns
32x1 TIHO 2.7 ns
32x1 TIHCK 2.3 ns

R
Speed Grade -3
Dual Port RAM Units
Size Symbol Min Max
Write Operation
Note 1: Timing for16 x1 RAM option is identical to16 x 2 RAM.
10

R
XQR4000XL CLB RAM Synchronous (Edge-Triggered) Write Timing
TWPS
WCLK (K)
TWSS TWHS
WE
TDSS TDHS
DATA IN
TASS TAHS
ADDRESS
TILO TILO
TWOS
DATA OUT OLD NEW
X6461
XQR4000XL CLB Dual-Port RAM Synchronous (Edge-Triggered) Write Timing
TWPDS
WCLK (K)
TWSDS TWHDS
WE
TDSDS TDHDS
DATA IN
TASDS TAHDS
ADDRESS
TILO TILO
TWODS
DATA OUT OLD NEW
X6474

R
XQR4000XL Pin-to-Pin Output Parameter Guidelines

XQR4000XL Output Flip-Flop, Clock to Out
Speed Grade -3
Units
Global Low Skew Clock to Output using OFF TICKOF XQR4013XL 8.6 ns
XQR4036XL 9.8 ns
XQR4062XL 11.3 ns
Global Early Clock to Output using OFF TICKEOF XQR4013XL 7.4 ns
Values are for BUFGE #s 3, 4, 7, and 8. Add XQR4036XL 8.1 ns
1.4 ns for BUFGE #s 1, 2, 5, and 6. XQR4062XL 9.9 ns
XQR4000XL Output Mux, Clock to Out

Speed Grade -3
Units
Global Low Skew Clock to Output using OFF TICKOF XQR4013XL 8.8 ns
XQR4036XL 10.0 ns 10
XQR4062XL 11.4 ns
Global Early Clock to Output using OFF. Val- TICKEOF XQR4013XL 7.6 ns
ues are for BUFGE #s 3, 4, 7, and 8. Add 1.4 XQR4036XL 8.2 ns
ns for BUFGE #s 1, 2, 5, and 6. XQR4062XL 10.0 ns

R
Capacitive Load Factor

Delta Delay (ns)

delay. 0
-1
voltage and temperature and is independent of the output
slew rate control.
-2
0 20 40 60 80 100 120 140
Capacitance (pF)
X8257

R
XQR4000XL Pin-to-Pin Input Parameter Guidelines

XQR4000XL Global Low Skew Clock, Set-Up and Hold

Speed Grade -3
Units
Input Setup and Hold Times Using
Global Low Skew Clock and IFF
No Delay TPSN/TPHN XQR4013XL 1.2 / 3.2 ns
XQR4036XL 1.2 / 5.5 ns
XQR4062XL 1.2 / 7.0 ns
Partial Delay TPSP/TPHP XQR4013XL 6.1 / 0.0 ns
XQR4036XL 6.4 / 1.0 ns
XQR4062XL 6.7 / 1.2 ns
Full Delay TPSD/TPHD XQR4013XL 6.4 / 0.0 ns
XQR4036XL 6. 6 / 0.0 ns
XQR4062XL 6.8 / 0.0 ns
using the furthest distance and a reference load of one clock pin per two IOBs. Use the static
timing analyzer (TRCE) to determine the setup and hold times under given design conditions.
Note 2: The XQ4013XL, XQ4036XL, and 4062XL have significantly faster partial and full delay setup
times than other devices.
10

R
XQR4000XL BUFGE #s 3, 4, 7, & 8 Global Early Clock, Set-up and Hold for IFF and FCL
Speed Grade -3
No Delay XQR4013XL 1.2 / 4.7
Global Early Clock and IFF TPSEN/TPHEN XQR4036XL 1.2 / 6.7
Global Early Clock and FCL TPFSEN/TPFHEN XQR4062XL 1.2 / 8.4
Partial Delay XQR4013XL 5.4 / 0.0
Global Early Clock and IFF TPSEP/TPHEP XQR4036XL 6.4 / 0.8
Global Early Clock and FCL TPFSEP/TPFHEP XQR4062XL 8.4 / 1.5
Full Delay XQR4013XL 12.0 / 0.0
Global Early Clock and IFF TPSED/TPHED XQR4036XL 13.8 / 0.0
XQR4062XL 13.1 / 0.0
Note 1: Setup time is measured with the fastest route and the lightest load. Hold time is measured using
the furthest distance and a reference load of one clock pin per two IOBs. Use the static timing
analyzer(TRCE) to determine the setup and hold times under given design conditions.

R
XQR4000XL BUFGE #s 1, 2, 5, & 6 Global Early Clock, Set-up and Hold for IFF and FCL
Speed Grade -3
No Delay XQR4013XL 1.2 / 4.7
Global Early Clock and IFF TPSEN/TPHEN XQR4036XL 1.2 / 6.7
Global Early Clock and FCL TPFSEN/TPFHEN XQR4062XL 1.2 / 8.4
Partial Delay XQR4013XL 6.4 / 0.0
Global Early Clock and IFF TPSEP/TPHEP XQR4036XL 7.0 / 0.0
Global Early Clock and FCL TPFSEP/TPFHEP XQR4062XL 9.0 / 0.8
Full Delay XQR4013XL 10.0 / 0.0
Global Early Clock and IFF TPSED/TPHED XQR4036XL 12.2 / 0.0
XQR4062XL 13.1 / 0.0
using the furthest distance and a reference load of one clock pin per two IOBs. Use the
static timing analyzer(TRCE) to determine the setup and hold times under given design con-
ditions.
10

R
XQR4000XL IOB Input Switching Characteristic Guidelines

Speed Grade -3
Units
Clocks
Clock Enable (EC) to Clock (IK) TECIK All devices 0.3 ns
Delay from FCL enable (OK) active edge to IFF clock (IK) TOKIK All devices 1.7 ns
active edge
Setup Times
Pad to Clock (IK), no delay TPICK All devices 1.7 ns
Pad to Clock (IK), via transparent Fast Capture Latch, no TPICKF All devices 2.3 ns
delay
Pad to Fast Capture Latch Enable (OK), no delay TPOCK All devices 0.7 ns
Hold Times
All Hold Times All devices 0 ns
Global Set/Reset
Minimum GSR Pulse Width TMRW All devices 19.8 ns
Delay from GSR input to any Q TRRI XQR4013XL 15.9 ns
XQR4036XL 22.5 ns
XQR4062XL 29.1 ns
Pad to I1, I2 TPID All devices 1.6 ns
Pad to I1, I2 via transparent input latch, no delay TPLI All devices 2.6 ns
Pad to I1, I2 via transparent FCL and input latch, no delay TPFLI All devices 3.1 ns
Clock (IK) to I1, I2 (flip-flop) TIKRI All devices 1.8 ns
Clock (IK) to I1, I2 (latch enable, active Low) TIKLI All devices 1.9 ns
FCL Enable (OK) active edge to I1, I2 TOKLI All devices 3.6 ns

R
XQR4000XL IOB Output Switching Characteristic Guidelines

-3
Units
Clocks
Propagation Delays
Clock (OK) to Pad TOKPOF 5.0 ns
Output (O) to Pad TOPF 4.1 ns
3-state to Pad hi-Z (slew-rate independent) TTSHZ 4.4 ns
3-state to Pad active and valid TTSONF 4.1 ns
Output (O) to Pad via Fast Output MUX TOFPF 5.5 ns
Select (OK) to Pad via Fast MUX TOKFPF 5.1 ns
Output (O) to clock (OK) hold time TOKO 0.0 ns
Clock Enable (EC) to clock (OK) setup time TECOK 0.0 ns
Clock Enable (EC) to clock (OK) hold time TOKEC 0.3 ns
Global Set/Reset
Delay from GSR input to any Pad TRPO
XQR4013XL 20.5 ns
XQR4036XL 27.1 ns 10
XQR4062XL 33.7 ns
For output SLOW option add TSLOW 3.0 ns
Note 1: Output timing is measured at ~50% VCC threshold, with 50 pF external capacitive loads.

R
Pinouts
CB228 Package for PIN_NAME CB228
XQR4013XL/4036XL/4062XL IO P44
IO P45
PIN_NAME CB228
IO P46
VSS P1
IO P47
BUFGP_TL_A16_GCK1_IO P2
IO P48
A17_IO P3
IO P49
IO P4
IO P50
IO P5
IO P51
TDI_IO P6
IO P52
TCK_IO P7
IO P53
IO P8
BUFGS_BL_GCK2_IO P54
IO P9
M1 P55
IO P10
VSS P56
IO P11
M0 P57
IO P12
VCC P58
IO P13
M2 P59
VSS P14
BUFGP_BL_GCK3_IO P60
IO P15
HDC_IO P61
IO P16
IO P62
TMS_IO P17
IO P63
IO P18
IO P64
IO P19
LDC_IO P65
IO P20
IO P66
IO P21
IO P67
IO P22
IO P68
IO P23
IO P69
IO P24
IO P70
IO P25
IO P71
IO P26
VSS P72
VSS P27
IO P73
VCC P28
IO P74
IO P29
IO P75
IO P30
IO P76
IO P31
IO P77
IO P32
IO P78
IO P33
IO P79
IO P34
IO P80
IO P35
IO P81
IO P36
IO P82
VCC P37
IO P83
IO P38
/ERR_INIT_IO P84
IO P39
VCC P85
IO P40
VSS P86
IO P41
IO P87
VSS P42
IO P88
IO P43
IO P89

R
PIN_NAME CB228 PIN_NAME CB228

IO P90 IO P138
IO P91 IO P139
IO P92 D4_IO P140
IO P93 IO P141
IO P94 VCC P142
VCC P95 VSS P143
IO P96 D3_IO P144
IO P97 /RS_IO P145
IO P98 IO P146
IO P99 IO P147
VSS P100 IO P148
IO P101 IO P149
IO P102 D2_IO P150
IO P103 IO P151
IO P104 VCC P152
IO P105 IO P153
IO P106 IO P154
IO P107 IO P155
IO P108 IO P156
IO P109 VSS P157
IO P110 IO P158
IO P111 IO P159
BUFGS_BR_GCK4_IO P112 IO P160
VSS P113 IO P161
DONE P114 IO P162
VCC P115 IO P163
/PROG P116 D1_IO P164
D7_IO P117 BUSY_/RDY_RCLK_IO P165
BUFGP_BR_GCK5_IO P118 IO P166
10
IO P119 IO P167
IO P120 D0_DIN_IO P168
IO P121 BUFGS_TR_GCK6_DOUT_IO P169
IO P122 CCLK P170
D6_IO P123 VCC P171
IO P124 TDO P172
IO P125 VSS P173
IO P126 A0_/WS_IO P174
IO P127 BUFGP_TR_GCK7_A1_IO P175
IO P128 IO P176
VSS P129 IO P177
IO P130 CSI_A2_IO P178
IO P131 A3_IO P179
IO P132 IO P180
IO P133 IO P181
D5_IO P134 IO P182
/CS0_IO P135 IO P183
IO P136 IO P184
IO P137 IO P185

R
PIN_NAME CB228
VSS P186
IO P187
IO P188
IO P189
IO P190
VCC P191
A4_IO P192
A5_IO P193
IO P194
IO P195
A21_IO P196
A20_IO P197
A6_IO P198
A7_IO P199
VSS P200
VCC P201
A8_IO P202
A9_IO P203
A19_IO P204
A18_IO P205
IO P206
IO P207
A10_IO P208
A11_IO P209
VCC P210
IO P211
IO P212
IO P213
IO P214
VSS P215
IO P216
IO P217
IO P218
IO P219
A12_IO P220
A13_IO P221
IO P222
IO P223
IO P224
IO P225
A14_IO P226
BUFGS_TL_GCK8_A15_IO P227
VCC P228

R
XQR 4062XL -3 CB 228 M

QPRO™ Radiation Hardened
Temperature Range
M = Military Ceramic (TC = -55oC
to +125 oC)
Device Type
XQR4062XL
XQR4036XL
Number of Pins
XQR4013XL

10

Packages and Thermal
Characteristics
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

Characteristics
Table of Contents

Package Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-1
Thermal Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-3
Package Thermal Characterization Methods & Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-4
Some Power Management Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-10
Package Electrical Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-11
Component Mass (Weight) by Package Type. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-13
Xilinx Thermally Enhanced Packaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-14
Moisture Sensitivity of PSMCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-15
Tape and Reel. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-17
Reflow Soldering Process Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-19
Sockets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-21
0
R
Characteristics
Package Information
Inches vs. Millimeters packages have a lead spacing of 0.5 mm, 0.65 mm, or 0.8
mm.
The JEDEC standards for PLCC, CQFP, and PGA pack-
ages define package dimensions in inches. The lead spac- Because of the potential for measurement discrepancies,
ing is specified as 25, 50, or 100 mils (0.025", 0.050" or this Data Book provides measurements in the controlling
0.100"). standard only, either inches or millimeters. (See Table 1 for
package dimensions.)
The JEDEC standards for PQFP, HQFP, TQFP, and VQFP
packages define package dimensions in millimeters. These
EIA Standard Board Layout of Soldered Pads for QFP Devices
M ID
e e
M
IE
10
b2
l2 e
Table 1: Dimensions for Xilinx Quad Flat Packs1
HQ160 HQ208 VQ100 HQ240

Dim. VQ44 VQ64 PQ100 TQ144 TQ176 HQ304
PQ160 PQ208 TQ100 PQ240
MID 9.80 9.80 20.40 28.40 28.20 13.80 19.80 23.80 32.20 40.20
MIE 9.80 9.80 14.40 28.40 28.20 13.80 19.80 23.80 32.20 40.20
e 0.80 0.50 0.65 0.65 0.50 0.50 0.50 0.50 0.50 0.50
b2 0.4 - 0.6 0.3 - 0.4 0.3 - 0.5 0.3 - 0.5 0.3 - 0.4 0.3 - 0.4 0.3 - 0.4 0.3 - 0.4 0.3 - 0.4 0.3 - 0.4
I2 1.60 1.60 1.802 1.80 1.60 1.60 1.60 1.60 1.60 1.60
Notes: 1. Dimensions in millimeters

2. For 3.2 mm footprint per MS022, JEDEC Publication 95.

R
Suggested Board Layout of Soldered Pads for BGA, CS and FG Packages
99020200
Figure 1: Suggested Board Layout of Soldered Pads for BGA, CS and FG Packages
Table 2: Soldering Dimensions for BG Packages
BG225 BG256 BG352 BG432 BG560

Solder Land (L) diameter 0.89 0.79 0.79 0.79 0.79
Opening in Solder Mask (M) diameter 0.65 0.58 0.58 0.58 0.58
Solder (Ball) Land Pitch (e) 1.5 1.27 1.27 1.27 1.27
Land Width between Via and Land (w) 0.3 0.3 0.3 0.3 0.3
Distance between Via and Land (D) 1.06 0.9 0.9 0.9 0.9
Via Land (VL) diameter 0.65 0.65 0.65 0.65 0.65
Through Hole (VH), plated diameter 0.3 0.3 0.3 0.3 0.3
Pad Array Full Periphery Periphery Periphery Perihpery
Matrix or External Row 15 x 15 20 x 20 26 x 26 31 x 31 33 x 33
Periphery rows - 4 4 4 5
Notes:
2. Dimensions in millimeters.
3. 3 x 3 matrix for illustration only, one land pad shown with via connection.
4. Reference J-STD-013, use ‘dog-bone’ design via connection to land pad.

R
Table 3: Soldering Dimensions for CS and FG Packages
CS144 FG256 FG456 FG556 FG600 FG680

Solder Land (L) diameter 0.33 0.54 0.54 0.54 0.54 0.54
Opening in Solder Mask (M) diameter 0.44 0.45 0.45 00.45 0.45 0.45
Solder (Ball) Land Pitch (e) 0.8 1.0 1.0 1.0 1.0 1.0
Land Width between Via and Land (w) 0.25 0.3 0.3 0.3 0.3 0.3
Distance between Via and Land (D) 0.56 0.7 0.7 0.7 0.7 0.7
Via Land (VL) diameter 0.51 0.56 0.56 0.56 0.56 0.56
Through Hole (VH), plated diameter 0.25 0.3 0.3 0.3 0.3 0.3
Pad Array Periphery Full Periphery Periphery Perihpery Perihpery
Matrix or External Row 13 x 13 16 x 16 22 x 22 26 x 26 30 x 30 39 x 39
Periphery rows 4 4 7* 4* 6* 5
Notes:
5. Dimensions in millimeters.
6. 3 x 3 matrix for illustration only, one land pad shown with via connection.
7. Reference J-STD-013, use ‘dog-bone’ design via connection to land pad.
8. *FG has more solder balls in the center periphery rows of balls.
Cavity Up or Cavity Down to the surrounding air is impaired if the logo is mounted
down.
Most Xilinx devices attach the die against the inside bottom
of the package (the side that does not carry the Xilinx logo). Thermal Management
This is called cavity-up, and has been the standard IC
assembly method for over 25 years. This method does not Modern high speed logic devices consume an appreciable
provide the best thermal characteristics. Pin Grid Arrays amount of electrical energy. This energy invariably turns
(greater than 130 pins) and Ceramic Quad Flat Packs are into heat. Higher device integration drives technologies to
assembled “Cavity Down”, with the die attached to the produce smaller device geometry and interconnections.
inside top of the package, for optimal heat transfer to the With smaller chip sizes and higher circuit densities, heat
ambient air. generation on a fast switching CMOS circuit can be very
significant. The heat removal needs for these modern
For most packages this information does not affect how the devices must be addressed.
package is used because the user has no choice in how the
package is mounted on a board. For Ceramic Quad Flat Managing heat generation in a modern CMOS logic device 10
Pack (CQFP) packages however, the leads can be formed is an industry-wide pursuit. However, unlike the power
to either side. Therefore, for best heat transfer to the sur- needs of a typical Application Specific Integrated Circuit
rounding air, CQFP packages should be mounted with the (ASIC) gate array, the power requirements for FPGAs are
logo up, facing away from the PC board. not determined as the device leaves the factory. Designs
vary in power needs.
Clockwise or Counterclockwise There is no way of anticipating the power needs of an
The orientation of the die in the package and the orientation FPGA device short of depending on compiled data from
of the package on the PC board affect the PC board layout. previous designs. For each device type, primary packages
PLCC and PQFP packages specify pins in a counterclock- are chosen to handle ‘typical’ designs and gate utilization
wise direction, when viewed from the top of the package requirements. For the most part the choice of a package as
(the surface with the Xilinx logo). PLCCs have pin 1 in the the primary heat removal casing works well.
center of the beveled edge while all other packages have Occasionally designers exercise an FPGA device, particu-
pin 1 in one corner, with one exception: The 100- and larly the high gate count variety, beyond “typical” designs.
165-pin CQFPs (CB100 and CB164) for the XC3000 The use of the primary package without enhancement may
devices have pin 1 in the center of one edge. not adequately address the device’s heat removal needs.
CQFP packages specify pins in a clockwise direction, when Heat removal management through external means or an
viewed from the top of the package. The user can make the alternative enhanced package should be considered.
pins run counterclockwise by forming the leads such that Removing heat ensures the functional and maximum
the logo mounts against the PC board. However, heat flow design temperature limits are maintained. The device may
go outside the temperature limits if heat build up becomes

R
excessive. As a consequence, the device may fail to meet Junction-to-Reference General Setup
electrical performance specifications. It is also necessary
to satisfy reliability objectives by operating at a lower tem-
perature. Failure mechanisms and the failure rate of
devices depend on device operating temperature. Control
of the package and the device temperature ensures prod-
uct reliability.
Package Thermal Characterization

Methods & Conditions
Method and Calibration
Xilinx uses the indirect electrical method for package ther-
mal resistance characterization. The forward-voltage drop
of an isolated diode residing on a special test die is cali-
brated at constant forcing current of 0.520mA with respect
to temperature over a correlation temperature range of
22°C to 125°C (degree Celsius). The calibrated device is
Figure 2: Thermal Measurement Set-Up (Schematic
then mounted in an appropriate environment (still air,
for Junction to Reference)
forced convection, circulating FC-40, etc.) Depending on
the package, between 0.5 to 4 watts of power (Pd) is
applied. Power (Pd) is applied to the device through dif- Junction-to-Case Measurement — ΘJC
fused resistors on the same thermal die. The resulting rise
ΘJC is measured in a 3M Flourinert (FC-40) isothermal cir-
in junction temperature is monitored with the forward-volt-
culating fluid stabilized at 25°C. The Device Under Test
age drop of the precalibrated diode. Typically, three identi-
(DUT) is completely immersed in the fluid and initial stable
cal samples are tested at each data point. The
conditions are recorded. Pd is then applied. Case temper-
reproducibility error in the set-up is within 6%.
ature (TC) is measured at the primary heat-flow path of the
Definition of Terms particular package. Junction temperature (TJ) is calculated
from the diode forward-voltage drop from the initial stable
TJ Junction Temperature — the maximum temperature condition before power was applied.
on the die, expressed in °C (degree Celsius)
ΘJC = (TJ - TC)/Pd
TA Ambient Temperature — expressed in °C.
The junction-to-isothermal-fluid measurement (ΘJI) is also
TC The temperature of the package body taken at a calculated from the same data.
defined location on the body. This is taken at the pri-
mary heat flow path on the package and represents ΘJL = (TJ - TI)/Pd
the hottest part on the package — expressed in °C. The latter data is considered as the ideal ΘJA data for the
Tl The isothermal fluid temperature when junction to package that can be obtained with the most efficient heat
case temperature is taken — expressed in °C. removal scheme. Other schemes such as airflow, heat-
sinks, use of copper clad board, or some combination of all
Pd The total device power dissipation — expressed in
these will tend towards this ideal figure. Since this is not a
watts.
widely used parameter in the industry, and it is not very
realistic for normal application of Xilinx packages, the ΘJI
data is not published. The thermal lab keeps such data for
package comparisons.
Junction-to-Ambient Measurement — ΘJA

ΘJA is measured on FR4 based PC boards measuring 4.5”
x 6.0” x .0625” (114.3mm x 152.4mm x 1.6mm) with edge
connectors. There are two main board types.
Type I, 2L/0P board, is single layer with 2 signal planes
(one on each surface) and no internal Power/GND planes.
The trace density on this board is less than 10% per side.
Type II, the 4L/2P board, has 2 internal copper planes (one

R
power, one ground) and 2 signal trace layers on both sur- the data from these tests shall always be qualified with the
faces. board mounting information.
Data may be taken with the package mounted in a socket or
Data Acquisition and Package Thermal
with the package mounted directly on the board. Socket
measurements typically use the 2L/0P boards. SMT Database
devices may use either board. Published data always Xilinx gathers data for a package type in die sizes, power
reflects the board and mount conditions used. levels and cooling modes (air flow and sometimes heatsink
Data is taken at the prevailing temperature and pressure effects) with a Data Acquisition and Control system (DAS).
conditions (22°C to 25°C ambient). The board with the DUT The DAS controls the power supplies and other ancillary
is mounted in a cylindrical enclosure. The power applica- equipment for hands-free data taking. Different setups
tion and signal monitoring are the same as ΘJC measure- within the DAS software are used to run calibration, ΘJA,
ments. The enclosure (ambient) thermocouple is ΘJC, fan tests, as well as the power effect characteristics of
substituted for the fluid thermocouple and two extra ther- a package.
mocouples brought in to monitor room and board tempera- A package is characterized with respect to the major vari-
tures. The junction to ambient thermal resistance is ables that influence the thermal resistance. The results are
calculated as follows: stored in a database. Thermal resistance data is interpo-
ΘJA = (TJ - TA)/Pd lated as typical values for the individual Xilinx devices that
are assembled in the characterized package. Table 4
The setup described herein lends itself to the application of shows the typical values for different packages. Specific
various airflow velocities from 0 - 800 Linear Feet per device data may not be the same as the typical data. How-
Minute (LFM), i.e., 0 - 4.06 m/s. Since the board selection ever, the data will fall within the given minimum and maxi-
(copper trace density, absence or presence of ground mum ranges. The more widely used packages will have a
planes, etc.) affects the results of the thermal resistance, wider range. Customers may contact the Xilinx application
group for specific device data.
10

R
Table 4: Summary of Thermal Resistance for Packages
ΘJA ΘJA ΘJA ΘJA ΘJA ΘJA

ΘJC
PKG-CODE still air still air still air 250 LFM 500 LFM 750 LFM Comments
(Typ)
(Max) (Typ) (Min) (Typ) (Typ) (Typ)
°C/Watt °C/Watt °C/Watt °C/Watt °C/Watt °C/Watt °C/Watt
BG225 37 30 24 19 17 16 3.3 Various
BG256 32 29 24 19 17 16 3.2 4L/2P-SMT
BG352 14 12 10 8 7 6 0.8 4L/2P-SMT
BG432 13 11 9 8 6 6 0.8 4L/2P-SMT
BG560 10 9 8 7 6 5 0.8 Estimated
CB100 44 41 38 25 19 17 5.1 Socketed
CB164 29 26 25 17 12 11 3.6 Socketed
CB196 25 24 24 15 11 10 1.8 Socketed
CB228 19 18 17 11 8 7 1.3 Socketed
DD8 114 109 97 90 73 60 8.2 Socketed
HQ160 14 14 14 10 8 7 1.0 4L/2P-SMT
HQ208 15 14 14 10 8 7 1.7 4L/2P-SMT
HQ240 13 12 12 9 7 6 1.5 4L/2P-SMT
HQ304 11 11 10 7 5 5 0.9 4L/2P-SMT
HT144 - 10.9 - 7.3 5.7 5.0 0.9 4L/2P-SMT
HT176 - 16.0 - - - - 2.0 Estimated
PC20 86 84 76 63 56 53 25.8 2L/0P-SMT
PC44 51 46 42 35 31 29 13.7 2L/0P-SMT
PC68 46 42 38 31 28 26 9.3 2L/0P-SMT
PC84 41 33 28 25 21 17 5.3 2L/0P-SMT
PD8 82 79 73 60 54 50 22.2 Socketed
PG84 37 34 31 24 18 16 5.8 Socketed
PG120 32 27 25 19 15 13 3.6 Socketed
PG132 32 28 24 20 17 15 2.8 Socketed
PG156 25 23 21 15 11 10 2.6 Socketed
PG175 25 23 20 14 11 10 2.6 Socketed
PG191 24 21 18 15 12 11 1.5 Socketed
PG223 24 20 18 15 12 11 1.5 Socketed
PG299 18 17 16 10 9 8 1.9 Socketed
PG411 16 15 14 9 8 7 1.2 Socketed
PG475 14 13 12 9 8 7 1.2 Socketed
PG559 - 12.00 - - - - - Estimated
PP132 35 34 33 23 18 17 6.0 Socketed
PP175 29 29 28 19 15 13 2.5 Socketed
PQ100 35 33 32 29 28 27 5.5 4L/2P-SMT
PQ160 37 32 22 24 21 20 4.6 2L/0P-SMT
PQ208 35 32 26 23 21 19 4.3 2L/0P-SMT
PQ240 28 23 19 17 15 14 2.8 2L/0P-SMT
SO8 147 147 147 112 105 98 48.3 IEEE-(Ref)
TQ100 37 31 31 26 24 23 7.5 4L/2P-SMT
TQ144 35 32 30 25 21 20 5.3 4L/2P-SMT
TQ176 29 28 27 21 18 17 5.3 4L/2P-SMT
VO8 162 162 162 123 116 108 48.3 Estimated

R
Table 4: Summary of Thermal Resistance for Packages (Continued)
ΘJA ΘJA ΘJA ΘJA ΘJA ΘJA

ΘJC
PKG-CODE still air still air still air 250 LFM 500 LFM 750 LFM Comments
(Typ)
(Max) (Typ) (Min) (Typ) (Typ) (Typ)
°C/Watt °C/Watt °C/Watt °C/Watt °C/Watt °C/Watt °C/Watt
CS48 45 Estimated
CS144 65 Estimated
VQ44 44 44 44 36 34 33 8.2 4L/2P-SMT
VQ64 44 41 39 34 32 31 8.2 4L/2P-SMT
VQ100 47 38 32 32 30 29 9.0 4L/2P-SMT
Notes: 1. Maximum, typical and minimum numbers are based on numbers for all the devices in the specific package at the time of
compilation. The numbers do not necessarily reflect the absolute limits of that packages. Specific device data should lie
within the limits. Packages used for a broader spectrum of devices have a wider range in the table. Specific device data in a
package may be obtained from the factory.
2. Package configurations and drawings are in the package section of the data book.
3. 2L/0P - SMT: the data is from a surface mount type I board -- no internal planes on the board.
4. 4L/2P - SMT: the data is from a 4 layer SMT board incorporating 2 internal planes. Socketed data is taken in socket.
5. Air flow is given Linear Feet per Minute (LFM). 500 LFM = 2.5 Meters per Second
Application of Thermal Resistance Data

Thermal resistance data gauges the IC package thermal Example 1:
performance. ΘJC measures the internal package resis- The manufacturer’s goal is TJ (max) < 100°C
tance to heat conduction from the die surface, through the A module is designed for a TA = 45°C max.
die mount material to the package exterior. ΘJC strongly A XC3042 in a PLCC 84 has a ΘJA = 32°C/watt.
depends on the package’s heat conductivity, architecture
Given a XC3042 with a logic design with a rated power
and geometrical considerations.
Pd of 0.75watt.
ΘJA measures the total package thermal resistance includ-
With this information, the maximum die temperature
ing ΘJC. ΘJA depends on the package material properties
can be calculated as:
and such external conditions as convective efficiency and
board mount conditions. For example, a package mounted TJ = 45 + (32 x .75) ==> 69°C.
on a socket may have a ΘJA value 20% higher than the
The system manufacturer’s goal of TJ < 100°C is met.
same package mounted on a 4 layer board with power and
ground planes. Example 2:
By specifying a few constraints, devices are ensured to A module has a TA = 55°C max. 10
operate within the intended temperature range. This also The Xilinx XC4013E is in a PQ240 package (HQ240 is
ensures device reliability and functionality. The system also considered).
ambient temperature needs to be specified. A maximum TJ A XC4013E, in an example logic design, has a rated
also needs to be established for the system. The following power of 2.50 watts. The module manufacturers goal is
inequality will hold. TJ(max.) < 100°C.
TJ(max) > ΘJA* Pd +TA Table 5 shows the package and thermal enhancement
combinations required to meet the goal of TJ < 100°C.
The following two examples illustrates the use of this ine-
quality.
Table 5: Thermal Resistance for XC4013E in PQ240 and HQ240 Packages
Device ΘJA ΘJA ΘJA ΘJA

Name Package still air (250 LFM) (500 LFM) (750 LFM) ΘJC Comments
XC4013E PQ240 23.7 17.5 15.4 14.3 2.7 Cu, SMT 2L/0P
XC4013E HQ240 12.5 8.6 6.9 6.2 1.5 4 Layer Board data
Notes: Possible Solutions to meet the module requirements of 100°C :
1a.Using the standard PQ240; TJ = 55 + (23.7 x 2.50) ==> 114.25 °C.
1b.Using standard PQ240 with 250LFM forced air; TJ = 55 + (17.5 x 2.50) ==> 98.75 °C
2a.Using standard HQ240TJ = 55 + (12.5 x 2.50) ==> 86.25 °C
2b.Using HQ240 with 250 LFM forced air; TJ = 55 + (8.6 x 2.50) ==> 76.5 °C

R
For all solutions, the junction temperature is calculated as: ent and board temperatures conditions, and most impor-
TJ = Power x ΘJA + TA. All solutions meet the module tantly the total power dissipation, thermal enhancements --
requirement of less than 100°C, with the exception of the such as forced air cooling, heat sinking, etc. may be neces-
PQ240 package in still air. In general, depending on ambi- sary to meet the TJ(max) conditions set.
PQ/HQ Thermal Data Comparison

HQ/PQ Thermal Data
Size effect on ΘJA
35
30
25
ΘJA (°C/watt)
20 HQ208
HQ240
15
HQ304
10 PQ208
PQ240
5
200 300 400 500 600 700
Die size (mils)
HQ/PQ Thermal Data

Effect of Forced Air on ΘJA
30
25
20
ΘJA (°C/watt)
15
10
0
0 200 400 600 800
Airflow - LFM
XC4010E-HQ208 XC4013E-HQ240 XC4025E-HQ304

XC4010E-PQ208 XC4013E-PQ240

R
PGA 299 Thermal Resistance

Effect of Air Flow on ΘJA
25
20
ΘJA (°C/watt)
15
10
0
0 100 200 300 400 500 600 700
Air Flow - LFM
PG191-XC4010E PG223-XC4013E
PG299-XC4025E PG299-FHS(XC4025E)
PG299 Thermal Resistance

Effects of Active & Passive Heat sinks
20
10
15
ΘJA (°C/watt)
10
0
A B C D E F
PG299 - Various Enhancements
A Standard Pkg D Pkg+Active Fan (V=12)

B Pkg+Finned HS (Passive) E Std Pkg +250LFM
C Pkg+Active Fan (V=0) F Pkg+Finned HS+ 250LFM

R
BGA Thermal Resistance

Effect of Air Flow on ΘJA
40
35
30
ΘJA (°C/watt)
25
20
15
10
0 200 400 600 800
Air Flow - LFM
XC4010E-BG225 (2L) XC4013E-BG225(4L) XC73144-BG225(4L)

XC73108-BG225(2L) XC5210-BG225(2L)
Some Power Management Options

FPGA devices are usually not the dominating power con- equivalent HQ package if available may resolve the
sumers in a system, and do not have a big impact on power issue. The heat enhanced packages are pin to pin
supply designs. There are obvious exceptions. When the compatible and they use the same board layout.
actual or estimated power dissipation appears to be more • The use of forced air is an effective way to improve
than the specification of the chosen package, some options thermal performance. As seen on the graphs and the
can be considered. Details on the engineering designs and calculations above, forced air (200 -- 300 LFM) can
analysis of some of these suggested considerations may reduce junction to ambient thermal resistance by 30%.
be obtained from the references listed at the end of the sec- • If space will allow, the use of finned external heatsinks
tion. The options include: can be effective. If implemented with forced air as well,
the benefit can be a 40% to 50% reduction. The
• A Xilinx low power (L) version of the circuit in the same
HQ304, all cavity down PGAs, and the BG352 with
package. With the product and speed grade of choice,
exposed heatsink lend themselves to the application of
up to a 40% power reduction can be anticipated. For
external heatsinks for further heat removal efficiency.
more information, contact the Xilinx Hotline group.
• Outside the package itself, the board on which the
• Explore thermally enhanced package options available
package sits can have a significant impact. Board
for the same device. As illustrated above, the HQ240
designs may be implemented to take advantage of this.
package has a thermal impedance of about 50% of the
Heat flows to the outside of a board mounted package
equivalent PQ240. Besides, the 240 lead, the 208 lead
and is sunk into the board to radiate. The effect of the
and the 304 lead Quad packages have equivalent
board will be dependent on the size and how it
heatsink enhanced versions. Typically 25% to 40%
conducts heat. Board size, the level of copper traces on
improvement in thermal performance can be expected
it, the number of buried copper planes all lower the
from these heatsink enhanced packages. Most of the
junction-to-ambient thermal resistance for a package.
high gate count devices above the XC4013 level come
Some of the heatsink packages with the exposed
either exclusively in heat enhanced packages or have
heatsink on the board side can be glued to the board
these packages as options. If the use of a standard PQ
with thermal compound to enhance heat removal.
appears to be a handicap in this respect, a move to the

R
References
Thermalloy, Inc.
Forced Air Cooling Application Engineering 2021 W. Valley View Lane
COMAIR ROTRON Box 810839
2675 Custom House Court Dallas, TX 75381-0839
San Ysidro, CA 92173 1-214-243-4321
1-619-661-6688
Wakefield Engineering, Inc.
Heatsink Application Engineering
60 Audubon Road
The following facilities provide heatsink solutions for indus- Wakefield MA 01880-1255
try standard packages. 1-617-245-5900
AAVID Thermal Technologies Xilinx does not endorse these vendors nor their products.
1 Kool Path They are listed here for reference only. Any materials or
Box 400 services received from the vendors should be evaluated for
Laconia, NH 03247-0400 compatibility with Xilinx components.
1-603-528-3400
Package Electrical Characterization

In high-speed systems, the effects of electrical package the larger the current spikes are: I = C * dv/dt. Current
parasitics become very critical when optimizing for system spikes through the IC pin and bondwire induces a voltage
performance. Such problems as ground bounce and drop across the leads and bondwires: V = L * di/dt. The
crosstalk can occur due to the inductance, capacitance, result is a momentary voltage difference between the inter-
and resistance of package interconnects. In digital sys- nal IC ground and system ground, which show up as volt-
tems, such phenomena can cause logic error, delay, and age spikes and unswitched outputs.
reduced system speed. A solid understanding and proper
Factors that affect ground bounce:
usage of package characterization data during system
design simulation can help prevent such problems. - rise and fall times
- load capacitance
Theoretical Background - package inductance
- number of output drivers sharing the same ground
There are three major electrical parameters which are used
path
to describe the package performance: resistance, capaci-
- device type
tance, and inductance. Also known as interconnect parasit-
ics, they can cause many serious problems in digital Analytical Formulas for Lead Inductance
systems. For example, a large resistance can cause RC & 10
RL off-chip delays, power dissipation, and edge-rate degra- 1. Rectangular Leadframe/Trace (straight)
dation. Large capacitance can cause RC delays, crosstalk,
edge-rate degradation, and signal distortion. The lead
5l ln  ------------  + ---
Lself = 2l 1
inductance, perhaps the most damaging parasitic in digital nH
w+t 2
circuitry, can cause such problems as ground bounce (also (no ground)
known as simultaneous switching noise or delta-I noise),
RL delays, crosstalk, edge rate degradation, and signal dis-
5l ln  ------------  +  ------------ 
8h w+t
Lself =
tortion.  w + t   4h  nH
Ground bounce is the voltage difference between any two (above ground)
grounds (typically between an IC and circuit board ground) l = lead/trace length
induced by simultaneously switching current through bond- w = lead/trace width
wire, lead, or other interconnect inductance. t = lead/trace thickness
When IC outputs change state, large current spikes result h = ground height
from charging or discharging the load capacitance. The unit = inches
larger the load capacitance and faster the rise/fall times,

R
2. Bondwire (gold wire) self-inductance. The parasitic effects of the return path are
small enough to ignore in the context of this method. For
mutual-inductance measurement, two adjacent leads are
5l ln  -----  – ---
2l 3
Lwire = nH
r 4 probed. A fast risetime step waveform is sent through one
of the leads. The current travels through the lead/bondwire
and returns by the path of the low-impedance ground. On
L = wire length
the adjacent “quiet” lead, a waveform is induced due to
r = wire radius mutual coupling. This waveform is measured as the mutual
unit = inches inductance.
General Measurement Procedure For capacitance measurements, all external leads except
for the lead(s) under test are grounded to the DUT fixture.
Xilinx uses the Time-Domain Reflectometry (TDR) method
For QFP, PLCC, and Power Quad-type of packages, the
for parasitic inductance and capacitance measurements.
die-paddle and the heat slug are left floating. Self-capaci-
The main components of a TDR setup includes: a digitizing
tance is measured by sending a fast risetime step wave-
sampling oscilloscope, a fast rise time step generator (<17
form through the lead under test. The reflection waveform
ps), a device-under-test (DUT) interface, and an imped-
from the lead, which includes the sum of all capacitive cou-
ance-profile analysis software to extract parasitic models
pling with respect to the lead under test, is then measured.
from the TDR reflection waveforms. In this method, a volt-
Appropriately, the self-capacitance can also be called the
age step is propagated down the package under test, and
“bulk” capacitance since the measured value includes the
the incident and reflected voltage waves are monitored by
capacitance between the lead under test and all surround-
the oscilloscope at a particular point on the line. The result-
ing metal, including the ground plane and the heat slug. For
ing characteristic impedance of the package interconnect
mutual-capacitance measurement, two adjacent leads are
shows the nature (resistive, inductive, and capacitive) of
probed. An incident waveform is sent through one lead, and
each discontinuity.
the induced waveform on the neighboring lead is measured
Package & Fixture Preparation as the mutual capacitance.
Before performing the measurements, the package and the In order to de-embed the electrical parasitics of the DUT
DUT interface must be fixtured. Proper fixturing ensures fixture and the measuring probes, the short and the open
accurate and repeatable measurements. The mechanical compensation waveforms are also measured after each
sample for all inductance (self & mutual) measurements package measurement. This procedure compensates the
are finished units with all leads shorted to the internal DUT fixture to the very tip of the probes.
ground. For packages without an internal ground (i.e. QFP,
Inductance & Capacitance Model Extraction
PLCC, etc.) the die-paddle is used instead. The mechanical
sample for all capacitance (self & mutual) measurements All measured reflection waveforms are downloaded to a PC
are finished units with all internal leads floating. The DUT running the analysis software for package parasitic model
interface provides a physical connection between the oscil- extraction. The software uses a method called the Z-profile
loscope and the DUT with minimum crosstalk and algorithm, or the impedance-profile algorithm, for parasitic
probe/DUT reflection. It also provides small ground loop to analysis. This method translates the downloaded reflection
minimize ground inductance of the fixture. waveforms into true impedance waveforms, from which
package models for inductance and capacitance are
Inductance & Capacitance Measurement Proce- extracted.
dure
For inductance measurements, a minimum of 25% and
Data Acquisition and Package Electrical
maximum of 50% of packages leads, including all leads Database
that are adjacent to the lead(s) under test, are insulated Xilinx acquires electrical parasitic data only on the longest
from the DUT fixture ground. All other leads, except for the and the shortest lead/traces of the package. This provides
lead(s) under test, are grounded. This insulation forces the the best and the worst case for each package type (defined
current to return through a low impedance path created on by package design, lead/ball count, pad size, and vendor).
the opposite side of the package. It also eliminates mutual For convenience, the corner interconnects are usually
coupling from the neighboring leads. Self-inductance is selected as the longest interconnect, while the center inter-
measured by sending a fast risetime step waveform connects are usually selected as the shortest.
through the lead under test. The inductive reflection wave-
For symmetrical quad packages, all four sides of the pack-
form through the lead and the bondwire is then obtained.
age are measured and averaged. Three to five samples are
This reflection waveform, which includes the inductance of
usually measured for accuracy and continuity purposes.
the die-paddle (for QFP and PLCC-type packages) and
The average of these samples is then kept as the official
parallel combination of leads in the return path, is the

R
measured parasitic data of that package type in the data-

base.
Component Mass (Weight) by Package Type

Package Description JEDEC Outline # Xilinx # Mass (g)
BG225 MOLDED BGA 27 mm FULL MATRIX MO-151-CAL OBG0001 2.2
BG256 MOLDED BGA 27 mm SQ MO-151-CAL OBG0011 2.2
BG352 SUPERBGA - 35 X 35 mm PERIPHERAL MO-151-BAR OBG0008 7.1
BG432 SUPERBGA - 40 X 40 mm PERIPHERAL MO-151-BAU OBG0009 9.1
BG560 SUPERBGA - 42.5 X 42.5 mm SQ MO-192-BAV OBG0010 11.5
CB100 NCTB TOP BRAZE 3K VER MO-113-AD3 OCQ0008 10.8
CB164 NCTB TOP BRAZE 3K VER MO-113-AA-AD3 OCQ0003 11.5
CB164 NCTB TOP BRAZE 4K VER MO-113-AA-AD3 OCQ0007 11.5
CB196 NCTB TOP BRAZE 4K VER MO-113-AB-AD3 OCQ0005 15.3
DD8 .300 CERDIP PACKAGE MO-036-AA OPD0005 1.1
HQ160 METRIC 28 28 -.65 mm 1.6H/S DIE UP MO-108-DDI OPQ0021 10.8
HQ208 METRIC 28 X 28 - H/S DIE UP MO-143-FA1 OPQ0020 10.8
HQ240 METRIC QFP 32 32 - H/S DIE UP MO-143-GA OPQ0019 15.0
HQ304 METRIC QFP 40 40-H/S DIE DOWN MO-143-JA OPQ0014 26.2
PC20 PLCC JEDEC MO-047 MO-047-AA OPC0006 0.8
PC44 PLCC JEDEC MO-047 MO-047-AC OPC0005 1.2
PC68 PLCC JEDEC MO-047 MO-047-AE OPC0001 4.8
PC84 PLCC JEDEC MO-047 MO-047-AF OPC0001 6.8
PD8 DIP .300 STANDARD MO-001-AA OPD0002 0.5
PG84 CERAMIC PGA CAV UP 11X11 MO-067-AC OPG0003 7.2
PG120 CERAMIC PGA 13 X 13 MATRIX MO-067-AE OPG0012 11.5
PG132 CERAMIC PGA 14 X 14 MATRIX MO-067-AF OPG0004 11.8
PG156 CERAMIC PGA 16 X 16 MATRIX MO-067-AH OPG0007 17.1
PG175 CERAMIC PGA 16 X 16 STD VER. MO-067-AH OPG0009 17.7 10
PG191 CERAMIC PGA 18 X 18 STD - ALL MO-067-AK OPG0008 21.8
PG223 CERAMIC PGA 18 X 18 TYPE MO-067-AK OPG0016 26.0
PG299 CERAMIC PGA 20 X 20 HEATSINK MO-067-AK OPG0022 37.5
PG299 CERAMIC PGA 20 X 20 TYPE MO-067-AK OPG0015 29.8
PG411 CERAMIC PGA 39 X 39 STAGGER MO-128-AM OPG0019 36.7
PG475 CERAMIC PGA 41 X 41 STAGGER MO-128-AM OPG0023 39.5
PG559 CERAMIC PGA 43 x 43 MO-128 OPG0025 44.50
PP132 PLASTIC PGA 14 X 14 MATRIX MO-83-AF OPG0001 8.1
PP175 PLASTIC PGA 16 X 16 BURIED MO-83-AH OPG0006 11.1
PQ100 EIAJ 14 X 20 QFP - 1.60 MO-108-CC1 OPQ0013 1.6
PQ160 EIAJ 28 X 28 .65 mm 1.60 MO-108-DD1 OPQ0002 5.8
PQ208 EIAJ 28 X 28 .5 mm 1.30 MO-143-FAI OPQ0003 5.3
PQ240 EIAJ 32 X 32 .5 mm MO-143-GA OPQ0010 7.1
SO8 VERSION 1 - .150/55MIL MO-150 OPD0006 0.1
TQ100 THIN QFP 1.4 mm thick MS-026-BDE OPQ0004 0.7
TQ144 THIN QFP 1.4 mm thick MS-026-BFB OPQ0007 1.4
TQ176 THIN QFP 1.4 mm thick MS-026-BGA OPQ0008 1.9
CS48 Chip Scale 7 x 7 mm, 0.8 mm ball pitch - OBG0012 0.2

R
Component Mass (Weight) by Package Type (Continued)

Package Description JEDEC Outline # Xilinx # Mass (g)
CS144 Chip Scale 12 x 12 mm, 0.8 mm ball pitch MO-205-BE OBG0015 0.3
FG256 Fine Pitch BGA 17 x 17 mm, 1.0 mm ball pitch MO-151-AAF-1 OBG0021 0.8
FG456 Fine PItch BGA 23 x 23 mm, 1.0 mm ball pitch MO-151-AAJ-1 OBG0019 1.6
FG556 Fine Pitch BGA 31 x 31 mm, 1.0 mm ball pitch MO-151-AAN-1 OBG0020 3.2
FG600 Fine PItch BGA, 27 x 27 mm, 1.0 mm ball pitch MO-151-AAL-1 OBG0018 2.3
FG680 Fine PItch BGA 40 x 40 mm, 1.0 mm ball pitch MO-151-AAU-1 OBG0022 9.7
VO8 THIN SOIC-II N/A OPD0007 0.1
VQ44 THIN QFP 1.0 thick MS-026-ACB OPQ0017 0.4
VQ64 THIN QFP 1.0 thick MS-026-ACD OPQ0009 0.5
VQ100 THIN QFP 1.0 thick MS-026-AED OPQ0012 0.6
Notes: 1. Data represents average values for typical packages with typical devices. The accuracy is between 7% to 10%.
2. More precise numbers (below 5% accuracy) for specific devices may be obtained from Xilinx through a factory
representative or by calling the Xilinx Hotline.
3. Tie-bar details are specific to Xilinx package. Lead width minimum is 0.056”.
Xilinx Thermally Enhanced Packaging

The Package Offering
Heatsink
Xilinx Code Body (mm) THK (mm) Mass (gm) JEDEC No. Xilinx No.
Location
HQ160 28x28 3.40 10.8 DOWN MO-108-DD1 OPQ0021
HQ208 28x28 3.40 10.0 DOWN MO-143-FA OPQ0020
HQ240 32x32 3.40 15.0 DOWN MO-143-GA OPQ0019
HQ304 40x40 3.80 26.2 TOP MO-143-JA OPQ0014
Overview package. This was done to ensure pin to pin

compatibility with the existing PQ and MQ packages.
Xilinx offers thermally enhanced quad flat pack packages
- At the 304 pin count level, the HQ is offered with the
on certain devices. This section discusses the performance heatsink up. This arrangement offers a better potential
and usage of these packages (designated HQ). In sum- for further thermal enhancement by the designer.
mary:
A Die Up/Heatsink Down
• The HQ-series and the regular PQ packages conform
to the same JEDEC drawings.
• The HQ and PQ packages use the same PCB land
patterns. B Die Down/Heatsink Up
• The HQ packages have more mass
• Thermal performance is better for the HQ packages
Where and When Offered A – Heatsink down orientation

B – Heatsink up orientation X5962
- HQ packages are offered as the thermally enhanced

equivalents of PQ packages. They are used for high Mass Comparison
gate count or high l/O count devices in packages,
where heat dissipation without the enhancement Because of the copper heatsink, the HQ series of packages
may be a handicap for device performance. Such are about twice as heavy as the equivalent PQ. Here is a
devices include XC4013E, XC4020E, XC4025E, and quick comparison.
XC5215.
HQ (gm) PQ (gm)
- They are also being used in place of MQUAD (MQ)
160 Pin 10.8 5.8
packages of the same lead count for new devices.
- The HQ series at the 240 pin count level or below 208 Pin 10.8 5.3
are offered with the heatsink at the bottom of the 240 Pin 15.0 7.1
304 Pin 26.2 N/A

R
Thermal Data for the HQ nants to the die surface and increasing the potential for
early device failure.
The data for individual devices may be obtained from Xilinx.
How the effects of moisture in plastic packages and the crit-
Still Air Data Comparison ical moisture content result in package damage or failure is
HQ PQ a complex function of several variables. Among them are
ΘJA (°C/Watt) ΘJA (°C/watt) package construction details -- materials, design, geome-
160 Pin 13.5-14.5 20.5-38.5 try, die size, encapsulant thickness, encapsulant proper-
ties, TCE, and the amount of moisture absorbed. The
208 Pin 14-15 26-35
PSMC moisture sensitivity has, in addition to package
240 Pin 12-13 19-28 cracking, been identified as a contributor to delamina-
304 Pin 10-11 N/A tion-related package failure artifacts. These package failure
Note: ΘJC is typically between 1 and 2 °C/Watt for HQ artifacts include bond lifting and breaking, wire neckdown,
and MQ Packages. For PQ’s, it is between 2 and 7 bond cratering, die passivation, and metal breakage.
°C/Watt.
Because of the importance of the PSMC moisture sensitiv-
ity, both device suppliers and device users have ownership
and responsibility. The background for present conditions,
Data Comparison at Airflow - 250 LFM moisture sensitivity standardized test and handling proce-
HQ PQ dures have been published by two national organizations.
ΘJA (°C/watt) ΘJA (°C/watt) Users and suppliers are urged to obtain copies of both doc-
160 Pin 9-10 15-28.5 uments (listed below) and use them rigorously. Xilinx
adheres to both.
208 Pin 9-10 14-26
240 Pin 8-9 11-21 • JEDEC STANDARD JESD22-A112. Test Method A112
304 Pin 6.5-8 N/A “Moisture-Induced Stress Sensitivity for Plastic Surface
Mounted Devices”.
Other Information
Available through Global Engineering Documents
- Leadframe: Copper EFTEC-64 or C7025
Phone: USA and Canada 800-854-7179, International
- Heat Slug: Copper - Nickel plated → Heatsink metal
1-303-792-2181
is Grounded
- Lead Finish 85/15 Sn/Pb 300 microinches minimum
• IPC Standard IPC-SM-786A “Procedures for
- D/A material - Same as PQ; Epoxy 84-1LMISR4
Characterizing and Handling of Moisture/Reflow
- Mold Cpd. Same as PQ - EME7304LC
Sensitive ICs”.
- Packed in the same JEDEC trays
Available through IPC 10
Moisture Sensitivity of PSMCs Phone: 1-708-677-2850
Moisture Induced Cracking During Solder None of the previously stated or following recommenda-
Reflow tions apply to parts in a socketed application. For board
mounted parts careful handling by the supplier and the user
The surface mount reflow processing step subjects the
is vital. Each of the above publications has addressed the
Plastic Surface Mount Components (PSMC) to high ther-
sensitivity issue and has established 6 levels of sensitivity
mal exposure and chemicals from solder fluxes and clean-
(based on the variables identified). A replication of those
ing fluids during user’s board mount assembly. The plastic
listings, including the preconditioning and test require-
mold compounds used for device encapsulation are, uni-
ments, and the factory floor life conditions for each level are
versally, hygroscopic and absorb moisture at a level deter-
outlined in Table 6. Xilinx devices are characterized to their
mined by storage environment and other factors.
proper level as listed. This information is conveyed to the
Entrapped moisture can vaporize during rapid heating in
user via special labeling on the Moisture Barrier Bag
the solder reflow process generating internal hydrostatic
(MBB).
pressure. Additional stress is added due to thermal mis-
match, and the Thermal Coefficient of Expansion (TCE) of In Table 6, the level number is entered on the MBB prior to
plastic, metal lead frame, and silicon die. The resultant shipment. This establishes the user’s factory floor life con-
pressure may be sufficient to cause delamination within the ditions as listed in the time column. The soak requirement
package, or worse, an internal or external crack in the plas- is the test limit used by Xilinx to determine the level number.
tic package. Cracks in the plastic package can allow high This time includes manufacturer’s exposure time or the
moisture penetration, inducing transport of ionic contami- time it will take for Xilinx to bag the product after baking.

R
Table 6: Package Moisture Sensitivity Levels per J-STD-020
Level Factory Floor Life Soak Requirements (Preconditioning)

Conditions Time Time Conditions
1 ≤30°C / 90% Unlimited 168 hours 85°C / 85% RH
RH
2 ≤30°C / 60% 1 year 168 hours 85°C / 60% RH
RH
Time (hours)
X + Y = Z
3 ≤30°C / 60% 168 hours 24 168 192 30°C / 60% RH
RH
4 ≤30°C / 60% 72 hours 24 72 96 30°C / 60% RH
RH
5 ≤30°C / 60% 24/28 hours 24 24/48 48/72 30°C / 60% RH
RH
6 ≤30°C / 60% 6 hours 0 6 6 30°C / 60% RH
RH
Notes: X =Default value of semiconductor manufacturer’s time between bake and bag. If the semiconductor manufacturer’s
actual time between bake and bag is different from the default value, use the actual time.
Y = Floor life of package after it is removed from dry pack bag.
Z = Total soak time for evaluation.
Factory Floor Life the internal humidity level. The loaded bag is then sealed
shut under a partial vacuum with an impulse heat sealer.
Factory floor life conditions for Xilinx devices are clearly
stated on MBB containing moisture sensitive PSMCs. Artwork on the bags provides storage, handling and use
These conditions have been ascertained by following Test information. There are areas to mark the seal date, quan-
Methods outlined in JEDEC JESD22-A112 and are repli- tity, and moisture sensitivity level and other information.
cated in Table 6. If factory floor conditions are outside the The following paragraphs contain additional information on
stated environmental conditions (30°C/90% RH for level 1, handling PSMCs.
and 30°C/60% RH for Levels 2-6) or if time limits have been
exceeded, then recovery can be achieved by baking the Handling Parts in Sealed Bags
devices before the reflow step. Identified in the next section Inspection
are two acceptable bake schedules. Either can be used for
recovery to the required factory floor level. Note the seal date and all other printed or hand entered
notations. Review the content information against what was
Dry Bake Recommendation and Dry Bag ordered. Thoroughly inspect for holes, tears, or punctures
Policy that may expose contents. Xilinx strongly recommends that
the MBB remain closed until it reaches the actual work sta-
Xilinx recommends, as do the mentioned publications and tion where the parts will be removed from the factory ship-
other industry studies, that all moisture sensitive PSMCs ping form.
be baked prior to use in surface mount applications, or
comply strictly with requirements as specified on the MBB. Storage
Tape and Reeled parts are universally dry packed. Level 1 The sealed MBB should be stored, unopened, in an envi-
parts are shipped without the need for, or use of, an MBB.
ronment of not more than 90% RH and 40°C. The enclosed
Two bake schedules have been identified as acceptable HIC is the only verification to show if the parts have been
and equivalent. The first is 24 hours in air at 125°C., in ship- exposed to moisture. Nothing in part appearance can verify
ping media capable of handling that temperature. The sec- moisture levels.
ond bake schedule is for 192 hours in a controlled
atmosphere of 40°C, equal to or less than 5% RH. Expiration Date
Dry Devices are sealed in special military specification The seal date is indicated on the MBB. The expiration date
Moisture Barrier Bags (MBB). Enough desiccant pouches is 12 months from the seal date. If the expiration date has
are enclosed in the MBB to maintain contents at less than been exceeded or HIC shows exposure beyond 20% upon
20% RH for up to 12 months from the date of seal. A revers- opening the bag bake the devices per the earlier stated
ible Humidity Indicator Card (HIC) is enclosed to monitor

R
bake schedules. The three following options apply after provides room for possible resealing and adhering to the
baking: reseal conditions outlined above. After opening, strictly
adhere to factory floor life conditions to ensure that devices
Use the devices within time limits stated on the MBB.
are maintained below critical moisture levels.
Reseal the parts completely under a partial vacuum Bags opened for less than one hour (strongly dependent on
with an impulse sealer (hot bar sealer) in an approved environment) may be resealed with the original desiccant. If
MBB within 12 hours, using fresh desiccant and HIC, the bag is not resealed immediately, new desiccant or the
and label accordingly. Partial closures using staples, old one that has been dried out may be used to reseal, if
plastic tape, or cloth tape are unacceptable. the factory floor life has not been exceeded. Note that
factory floor life is cumulative. Any period of time when
Store the out-of-bag devices in a controlled atmosphere MBB is opened must be added to all other opened periods.
at less than 20% RH. A desiccator cabinet with con- Both the desiccant pouches and the HIC are reversible.
trolled dry air or dry nitrogen is ideal. Restoration to dry condition is accomplished by baking at
Other Conditions 125°C for 10-16 hours, depending on oven loading condi-
tions.
Open the MBB when parts are to be used. Open the bag by
cutting across the top as close to the seal as possible. This
Tape and Reel

Xilinx offers a tape & reel packing for PLCC, BGA, QFP, and Material and Construction
SO packages. The packing material is made of black con-
ductive Polystyrene and protects the packages from Carrier Tape
mechanical and electrical damage. The reel material pro- • The pocketed carrier Tape is made of conductive
vides a suitable medium for pick and place equipment. polystyrene material, or equivalent, with a surface
resistivity level of less than 106 ohms per square inch.
The tape & reel packaging consists of a pocketed carrier
• Devices are loaded ‘live bug’ or leads down, into a
tape, sealed with a protective cover. The device sits on ped-
device pocket.
estals (for PLCC, QFP packages) to protect the leads from
• Each carrier pocket has a hole in the center for
mechanical damage. All devices loaded into the tape carri-
automated sensing of whether a unit is in the pocket or
ers are baked, lead scanned before the cover tape is
not.
attached and sealed to the carrier. In-line mark inspection
• Sprocket holes along the edge of the carrier tape
for mark quality and package orientation is used to ensure
enable direct feeding into an automated board
shipping quality.
assembly equipment.
Benefits Cover Tape 10
• Increased quantity of devices per reel versus tubes • An anti-static, transparent, polyester cover tape, with
improves cycle time and reduces the amount of time to heat activated adhesive coating, sealed to the carrier
index spent tubes. edges to hold the devices in the carrier pockets.
• Tape & reel packaging enables automated pick and • Surface resistivity on both sides is less than 1011 ohms
place board assembly. per square inch.
• Reels are uniform in size enabling equipment flexibility.
• Transparent cover tape allows device verification and Reel
orientation. • The reel is made of anti-static Polystyrene material. The
• Anti-static reel materials provides ESD protection. loaded carrier tape is wound onto this conductive
• Carrier design include a pedestal to protect package plastic reel.
leads during shipment. • A protective strip made of conductive Polystyrene
• Bar code labels on each reel facilitate automated material is placed on the outer part of the reel to protect
inventory control and component traceability. the devices from external pressure in shipment.
• All tape & reel shipments include desiccant pouches • Surface resistivity is less than 1011 ohms per square
and humidity indicators to insure products are safe from inch.
moisture. • Device loading orientation is in compliance with EIA
• Compliant to Electronic Industries Association (EIA) Standard 481.
481. Bar Code Label
• The bar code label on each reel provides customer
identification, device part number, date code of the

R
product and quantity in the reel. management and data input accuracy.
• Print quality are in accordance with ANSI X3.182-1990
Bar Code Print Quality Guidelines. Presentation of Data Shipping Box
on labels are EIA-556-A compliant. • The shipping container for the reels are in a 13” x 13” x
• The label is an alphanumeric, medium density Code 39 3” C-flute, corrugated, # 3 white ‘pizza’ box, rated to
labels. 200 lb test.
• This machine-readable label enhances inventory
l
Table 7: Tape & Reel Packaging
Carrier Cover Qty per

Package Type Pin Count Width Width Pitch Reel Size Reel
PLCC (Plastic Leaded Chip Carrier) 20 16 mm 13.3 mm 12 mm 7 inch 250
20 16 mm 13.3 mm 12 mm 13 inch 750
44 32 mm 25.5 mm 24 mm 13 inch 500
68 44 mm 37.5 mm 32 mm 13 inch 250
84 44 mm 37.5 mm 36 mm 13 inch 250
SO (Plastic Small Outline) 8 12 mm 9.2 mm 8 mm 7 inch 750
QFP (Plastic Quad Flat Pack) PQ, VQ, 100 44 mm 37.5 mm 32 mm 13 inch 250
TQ, HA 160 44 mm 37.5 mm 40 mm 13 inch 200
BGA (Plastic Ball Grid Array) 225/256 44 mm 37.5 mm 32 mm 13 inch 500
Notes: 1.A minimum of 230mm of empty pockets are provided at the beginning (leader) of each reel.
2.A minimum of 160mm of empty pockets are provided at the end (trailer) of each reel.
3.Tape Leader/Trailer requirements are in compliance to EIA Standards 481.
4.Peel Strength between 20 and 120 grams ensures consistency during de-reeling operations and is compliant to EIA
Standard 481.
5.Each reel is subject to peel back strength tests.
6.For packages not listed above, please contact your Xilinx sales representative for updated information.
Standard Bar Code Label Locations

R
Reflow Soldering Process Guidelines

In order to implement and control the production of surface Each phase of a surface mount reflow profile has min/max
mount assemblies, the dynamics of the solder reflow pro- limits that should be viewed as a process window. The pro-
cess, and how each element of the process is related to the cess requires a careful selection and control of the materi-
end result, must be thoroughly understood. als, geometries of the mating surfaces (package footprint
The primary phases of the reflow process are as follows: vs. PCB land pattern geometries) and the time temperature
of the profile. If all of the factors of the process are suffi-
1. Melting the particles in the solder paste ciently optimized, there will be good solder wetting and fillet
2. Wetting the surfaces to be joined formation (between component leads and the land patterns
on the substrate). If factors are not matched and optimized
3. Solidifying the solder into a strong metallurgical bond there can be potential problems as summarized in
The sequence of five actions that occur during this process Figure 4.
Potential Reflow Soldering Issues

Reflow Soldering Phases
6
200
4
Temperature
5
Temperature (°C)
Surface Tension Takes Over

Wetting and Wicking Begin
2
Solder Melting Completes,
150
3
Cool Down Phase
Solder Balls Melt,
100
1
Flux Reduces
Metal Oxides
Evaporation
50 Time
Solvent
1. Insufficient Temperature to Evaporate Solvent

2. Component Shock and Solder Splatter
Time X5975
3. Insufficient Flux Activation
4. Excessive Flux Activity and Oxidation
5. Trapping of Solvent and Flux, Void Formation
6. Component and/or Board Damage X5976
Figure 3: Soldering Sequence Figure 4: Soldering Problems Summary

10

R
Figure 5 and Figure 6 show typical conditions for solder (PSMCs) must be verified prior to surface mount flow.See
reflow processing using Vapor Phase or IR Reflow. The the preceding sections for a more complete discussion on
moisture sensitivity of Plastic Surface Mount Components PSMC moisture sensitivity.
T-Max (leads)
220° - 235°C 215 - 219°C
2
≅ 45 s max
2 - 4°C/s Ramp down
Temperature °C
2 - 4°C/s
Ramp down
Temperature °C
2 - 4°C/s t183
Temp = 183°C Dwell = 30 - 60 s
t183 Preheat & drying dwell 3

120 s min between
60s < t183 < 120s 95° - 180°C
Preheat & drying dwell applies to lead area
120 - 180 s
between 95° - 180°C 3
2
2
Time (s) Time (s) X5974

X5973
Figure 5: Typical Conditions for IR Reflow Soldering Figure 6: Typical Conditions for Vapor Phase Reflow
Soldering
Notes:
1. Max temperature range = 220°C-235°C (leads) Notes:
Time at temp 30-60 seconds 1. Solvent - FC5312 or equivalent - ensures temperature
2. Preheat drying transition rate 2-4°C/s range of leads @ 215-219°C
3. Preheat dwell 95-180°C for 120-180 seconds 2. Transition rate 4-5°C/s
4. IR reflow shall be performed on dry packages 3. Dwell is intended for partial dryout and reduces the
difference in temperature between leads and PCB
land patterns.
The IR process is strongly dependent on equipment and
4. These guidelines are for reference. They are based on
loading differences. Components may overheat due to lack
laboratory runs using dry packages. It is
of thermal constraints. Unbalanced loading may lead to
recommended that actual packages with known loads
significant temperature variation on the board. This
be checked with the commercial equipment prior to
guideline is intended to assist users in avoiding damage to
mass production.
the components; the actual profile should be determined by
the users using these guidelines.

R
Sockets
Table 8 lists manufacturers known to offer sockets for Xilinx ment by Xilinx. Each user has the responsibility to evaluate
Package types. This summary does not imply an endorse- and approve a particular socket manufacturer.
Table 8: Socket Manufacturers
Packages
PQ
Manufacturer DIP HQ
SO PC TQ PG BG
VO WC VQ PP CB CG
AMP Inc.
470 Friendship Road
X X X
Harrisburg, PA 17105-3608
(800) 522-6752
Augat Inc.
452 John Dietsch Blvd.
P.O. Box 2510 X X X
Attleboro Falls, MA 02763-2510
(508) 699-7646
McKenzie Socket Division
910 Page Avenue
X X X
Fremont, CA 94538
(510) 651-2700
3M Textool
6801 River Place Blvd.
Austin, TX 78726-9000 X X X
(800) 328-0411
(612) 736-7167
Wells Electronics
1701 South Main Street
X
South Bend, IN 46613-2299
(219) 287-5941
Yamaichi Electronics Inc. 10
2235 Zanker Road
X X X X
San Jose, CA 95131
(408) 456-0797

R

0
Package Drawings
Package Drawings
Ceramic DIP Package - DD8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-24
Plastic DIP Package - PD8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-25
SOIC and TSOP Packages - SO8, VO8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-26
SOIC Package - SO20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-27
PLCC Packages - PC20, PC28, PC44, PC68, PC84 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-28
Ceramic Leaded Chip Carrier Package - CC44 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-29
Ball Chip Scale Package - CS48 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-30
Ball Chip Scale Package - CS144 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-31
VQFP Packages - VQ44, VQ64, VQ100. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-32
TQFP/HTQFP Packages - TQ100, TQ144, TQ176, HT100, HT144, HT176 . . . . . . . . . . . . . . . 11-33
PQ/HQFP Packages - PQ100, HQ100 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-34
PQ/HQFP Packages - PQ44, PQ160, PQ208, PQ240, HQ160, HQ208, HQ240 . . . . . . . . . . . . 11-35
PQ/HQFP Packages - PQ304, HQ304 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-36
BGA Packages - BG225 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-38
BGA Packages - BG256 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-39
BGA Packages - BG352, BG432 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-40
BGA Packages - BG560 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-41
Ceramic PGA Packages - PG68, PG84 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-42
Ceramic PGA Packages - PG120, PG132, PG156 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-43
Ceramic PGA Packages - PG175 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-44
Ceramic PGA Packages - PG223, PG299 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-46
Ceramic PGA Packages - PG475, PG559 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-48
Ceramic Brazed QFP Packages - CB100 (XC3000 Version) . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-50
Ceramic Brazed Packages - CB164 (XC3000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-51
Ceramic Brazed QFP Packages - CB100, CB164, CB196 (XC4000 Version) . . . . . . . . . . . . . . 11-52
Ceramic Brazed QFP Packages - CB228 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-53
Ball Fine Pitch Packages - FG256 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-55

Package Drawings February 2, 1999 (Version 1.3) 11
Package Drawings
Ceramic DIP Package - DD8

R
Package Drawings
Plastic DIP Package - PD8
11

R
Package Drawings
SOIC and TSOP Packages - SO8, VO8

R
Package Drawings
SOIC Package - SO20
11

R
Package Drawings
PLCC Packages - PC20, PC28, PC44, PC68, PC84

R
Package Drawings
Ceramic Leaded Chip Carrier Package - CC44
11

R
Package Drawings
Ball Chip Scale Package - CS48

R
Package Drawings
Ball Chip Scale Package - CS144
11

Package Drawings
VQFP Packages - VQ44, VQ64, VQ100

R
Package Drawings
TQFP/HTQFP Packages - TQ100, TQ144, TQ176, HT100, HT144, HT176
11

R
Package Drawings
PQ/HQFP Packages - PQ100, HQ100

R
Package Drawings
PQ/HQFP Packages - PQ44, PQ160, PQ208, PQ240, HQ160, HQ208, HQ240
11

R
Package Drawings
PQ/HQFP Packages - PQ304, HQ304

Package Drawings
BGA Packages - BG225

R
Package Drawings
11

R
Package Drawings
BGA Packages - BG352, BG432

R
Package Drawings
11

Package Drawings
Ceramic PGA Packages - PG68, PG84

R
Package Drawings
Ceramic PGA Packages - PG120, PG132, PG156
11

R
Package Drawings
Ceramic PGA Packages - PG175

Package Drawings
11

R
Package Drawings

R
Package Drawings
11

R
Package Drawings

R
Package Drawings
Ceramic Brazed QFP Packages - CB100 (XC3000 Version)

R
Package Drawings
Ceramic Brazed Packages - CB164 (XC3000)
11

Package Drawings
Ceramic Brazed QFP Packages - CB100, CB164, CB196 (XC4000 Version)

R
Package Drawings
Ceramic Brazed QFP Packages - CB228
11

Package Drawings
Ball Fine Pitch Packages - FG256
11

R
Package Drawings

R
Package Drawings
11

R
Package Drawings

R
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

R
Table of Contents
Quality Assurance and Reliability

Quality Assurance Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-3
Device Reliability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-4
Description of Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-4
Die Qualification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-4
Package Integrity and Assembly Qualification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-4
Testing Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-5
Data Integrity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-9
Memory Cell Design in the FPGA Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-9
Electrostatic Discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-10
Latchup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-10
High Temperature Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-10
0
R
Quality Assurance Program control and capabilities that our customers require. All
key Xilinx subcontract partners are ISO certified.
All aspects of the Quality Assurance Program at Xilinx have • Product Identification & Traceability: is maintained
been designed to eliminate the root cause of defects, rather throughout the manufacturing process. Traceability
than to try to remove them by inspection. A quality system back to the starting materials is available through
was put in place which is in full compliance with the require- unique product identification techniques and markings
ments of ISO9001. Xilinx was found to be in full compliance throughout the manufacturing process.
of the requirements of ISO9001:1994 by an independent • Process Control: is assured by identifying and
auditor in November, 1997. At that time Xilinx was regis- controlling those processes that directly affect the
tered for “the design, manufacturing and testing of pro- quality of our products, whether those processes are
grammable logic devices”. In January 1997 Xilinx was performed directly by Xilinx, or by our subcontract
formally granted transitional QML approval by DSCC. In partners.
November of 1997, Xilinx was audited by DSCC and found • Inspection & Test: is performed to ensure that
in full compliance with the requirements of MIL 38535 for a incoming product is not used or processed until it has
QML supplier. been verified as conforming to required specifications.
The aspects of ISO compliance in place at Xilinx include This inspection is done jointly by Xilinx and by its
the following seventeen points: subcontract partners.
• Inspection, Measuring and Test Equipment: is
• Management Review: a comprehensive system of
calibrated in conformance with the requirements of Mil
management attention and direction for all aspects of
Ref 45662 and/or other international standards.
company performance that directly affect our
Equipment is maintained in such a manner to ensure
customers. These include (among others) Xilinx
that measurement uncertainty is known and is
performance in the areas of Quality, Reliability and
consistent with specification requirements.
On-Time Delivery. Management assures that this
• Inspection & Test Status: of product is uniquely
quality policy is understood, implemented and
identified throughout the manufacturing process both at
maintained at all levels in the organization.
Xilinx and at our subcontract partners. Records are kept
• Quality Systems: are in place to ensure that product
to identify the authority responsible for the release of
conforms to customer specifications. These systems
conforming production.
facilitate, measure and continuously improve Xilinx
• Control of Non-Conforming Product: is assured
performance in those areas that affect customer
through disposition procedures that are defined in such
satisfaction. Xilinx remains committed to achieving
a manner as to prevent the shipping of non-conforming
100% customer satisfaction.
products. The responsibility and authority for the
• Contract Review: is conducted to ensure each
disposition of such products are well defined.
contract adequately defines and documents 12
• Corrective Action: processes are documented and
requirements, that differences between customer and
implemented to prevent the recurrence of
Xilinx standard specifications are mutually satisfactorily
nonconforming product. These processes are the key to
resolved, and that Xilinx has the capability to meet
implementing the Xilinx strategy of eliminating the root
contract requirements.
causes of nonconformity, rather that to apply inspection
• Document Control: procedures are established and
to try to remove nonconformity.
maintained to control all documents and data that relate
• Handling, Storage, Packing & Delivery: procedures
to the performance of Xilinx business and processing
are defined and implemented to prevent damage or
requirements. All organizations who need access to
deterioration of product once the manufacturing
such documentation during the performance of their
process is complete.
functions are assured availability of the latest,
• Quality Records: procedures are established and
controlled versions of that documentation.
maintained for the identification, collection, indexing,
• Purchasing: procedures are in place to ensure that all
filing, storage, maintenance and disposition of quality
purchased products conform to the specified
records.
requirements. As Xilinx is a “fabless” manufacturing
• Internal Quality Audits: are carried out to verify
company, special attention is paid to our subcontract
whether quality activities comply with planned
partners. They are required to demonstrate the type of

R
arrangements and to determine the effectiveness of the erating the failure rate by testing at a higher junction tem-
quality system. These audits are regularly perature (usually 125°C or 145°C). Extensive testing of
supplemented by quality audits performed by our Xilinx devices (performed on actual production devices
customers, and by our independent ISO auditors. taken directly from finished goods) has been accomplished
• Training: procedures have been established and are continuously since 1989 and reported quarterly. Quarterly
maintained to identify the training needs of all personnel reports on the reliability of Xilinx products are available
affecting quality during the production of Xilinx through your Xilinx sales representative and at the
products. Personnel performing such activities are WebLINX web site (www.xilinx.com). During the last two
qualified based upon appropriate education, training years, over 20,000 devices have accumulated a total of
and/or experience. over 36,000,000 hours of both static and dynamic operation
• Statistical Techniques: are in place at Xilinx and at at 125°C (equivalent) to yield the FIT rates shown in
our subcontract partners for verifying the acceptability Figure 1.
of process capabilities and product characteristics.
These key requirements are in place at Xilinx and at our
Description of Tests
subcontract partners to ensure our ability to achieve cus-
Die Qualification
tomer satisfaction through the on-time delivery of quality
products that meet customer requirements and are reli- 1. High Temperature Life: This test is performed to evalu-
able. ate the long-term reliability and life characteristics of the
die. It is defined by the Military Standard from which it is
Device Reliability derived as a “Die-Related Test” and is contained in the
Group C Quality Conformance Tests. Because of the
Device reliability is often expressed in a measurement
acceleration factor induced by higher temperatures,
called Failures in Time (FITs). In this measure one FIT
(typically 125°C and/or 145°C) data representing a large
equals one failure per billion (109) device operating hours.
number of equivalent hours at a normal temperature of
A failure rate in FITS must include the operating tempera-
25°C can be accumulated in a reasonable period of
ture to be meaningful. Hence failure rates are often
time.
expressed in FITS at 70°C (or some other temperature in
excess of the application). 2. Biased Moisture Life: This test is performed to evalu-
ate the reliability of the die under conditions of long-term
Since one billion hours is well in excess of 100,000 years,
exposure to severe, high-moisture environments that
the FIT rate of modern ICs can only be measured by accel-
could cause corrosion. Although it clearly stresses the
Figure 1: Failure Rates in FITs

R
package as well, this test is typically grouped under the from alternating exposure to temperature extremes. The
die-related tests. The device is operated at maxi- range of temperatures is -65°C to +150°C (condition
mum-rated voltage, 5.5 Vdc, and is exposed to a tem- “C”). The transition time is longer than that in the Ther-
perature of 85°C and a relative humidity of 85% mal Shock test but the test is conducted for many more
throughout the test. cycles.
4. Salt Atmosphere: This test was originally designed by
Package Integrity and Assembly
the US Navy to evaluate resistance of military-grade
Qualification ship-board electronics to corrosion from sea water. It is
1. Unbiased Pressure Pot: This test is performed at a used more generally for non-hermetic industrial and
temperature of 121°C and a pressure of 2 atm of satu- commercial products as a test of corrosion resistance of
rated steam to evaluate the ability of the plastic encap- the package marking and finish.
sulating material to resist water vapor. Moisture 5. Resistance to Solvents: This test is performed to eval-
penetrating the package could induce corrosion of the uate the integrity of the package marking during expo-
bonding wires and nonglassivated metal areas of the die sure to a variety of solvents. This is an especially
(bonding pads only for FPGA devices). Under extreme important test, since an increasing number of
conditions, moisture could cause drive-in and corrosion board-level assemblies are subjected to severe condi-
under the glassivation. Although it is difficult to correlate tions of automated cleaning before system assembly.
this test to actual field conditions, it provides a This test is performed according to the methods speci-
well-established method for relative comparison of plas- fied by MIL-STD-883.
tic packaging materials and assembly and molding tech-
niques. 6. Solderability: This test is performed to evaluate the sol-
derability of the leads under conditions of low soldering
2. Thermal Shock: This test is performed to evaluate the temperature following exposure to the aging effects of
resistance of the package to cracking and resistance of water vapor.
the bonding wires and lead frame to separation or dam-
age. It involves nearly instantaneous change in temper- 7. Lead Fatigue: This test is performed to evaluate the
ature from -65°C to +150°C (condition “C”). resistance of the completed assembly to vibrations dur-
ing storage, shipping, and operation.
3. Temperature Cycling: This test is performed to evalu-
ate the long-term resistance of the package to damage
12

R
Testing Facilities requirements for new and/or changed process flows.

Table 3 is a list of current failure analysis capabilities.
Xilinx has complete capability to perform High Temperature These laboratories are dedicated exclusively to increasing
Life Testing, Thermal Shock, Temperature Cycling, Biased customer satisfaction through continuous improvements in
Moisture Life Test, Unbiased Pressure Pot, Solderability our processes and technologies.
and Hermeticity, as well as complete Failure Analysis in
house. Table 1 and Table 2 show typical qualification

R
Table 1: Plastic Package/Product Qualification Requirements

New Assy Techniques (Mat’l/Process/Method
Test Test Description Acc# New New New New Lead Die Die Wire Mold Lead New New Full
Seq (note 1) S.Size Assy Pkg Pkg Pkg Frame Attach Coat Bond CLP Finish Device Fab Qual
(note 2) Plant Type I Type II Type Mask Proc
(note3) (note4) III LF (note6)
Design
(note5)
B1 * Phy. Dimension 0/5 X X X X X
B2 * Resist. to Solvents 0/3 X X X X
B3 * Solderability Test (note 7) 0/5 X X X X
B4 Solder Heat Test (Optn’l) 0/15 X X X X
B5 Auto Clave (SPP)(Optn’l) 0/76 0/76 X X X X X X X X X
B6 * Ball Shear/Bond Pull (note 7) 0/5 X X X X X X X X
B7 ** X-Ray (note 7) 0/5 X X X X X X X X X
B8 * S.A.T/Dye Pen Test (note 7) 0/10 X X X X X X X X
B9 * Adhesion of L/Finish (Optn’l) 0/3 X X X X
B10 * External Visual (note 7) 0/25 X X X X X X X
B11 Internal Visual (note 7) 0/5 X X X X X X X X X X
B12 * Die Shear (note 7) 0/5 X X X X X
B13 Flammability Test (note 7) Per lot X X
C1-A High Temp Life Test 0/76 X X X X
C1-B Low Temp Life Test (note 7) 0/22 X X X
C2 C2-A:HAST (0/22) or C2-B: 85/85 0/76 X X X X X X X X X
C3 ESD (HBM) 0/3 X X X
C4 High Temp Storage (Optn’l) 0/77 X X X X
D1 * Lead Integrity 0/3 X X X X X
D2 Thermal Shock (Optn’l) 0/76 X
D3 Temp Cycle 0/76 X X X X X X X X X X X
E1 Electrical Test & Data Log 0/30 X X X
E2 Electrical Characterization 0/30 X X X
E3 T.D.D.B (note 7) – X X X
E4 Latch-up 0/9 X X X
E5 Electromigration (note 7) – X X X
E6 Photosensitivity (Optn’l) 0/11 X X X
E7 Data Retention Bake EPLD & EPR 0/22 X X X
E8 Input/Output Capacitance 0/5 X X X
E9 Power Cycling (Optn’l) 0/22 X X X
E.Good 239 238 162 248 248 157 314 86 325 0 393 464 636
Qty required per lot E.Reject 63 48 43 35 43 5 5 5 43 29 10 10 64
Total 302 286 205 283 291 162 319 91 368 29 403 474 700
Notes: 1) Test method and stress conditions available upon request. 12
2) For any QUAL which does not meet the standard requirements, approval from Product Engineering and Product QA is
required.
3) Any new package which has not been qualified in the qualified assembly facility.
4) Any new package where the same body size with different lead pitch has been qualified.
5) New leadframe design whereby the paddle size is larger than the existing leadframe paddle size used in the same qualified
package.
6) For new mask from same device family, only high temp life test, ESD, Latch & Capacitance are required.
7) In-process monitor data may be used to satisfy this requirement.
*) Electrical rejects can be used as test sample.
**) This is a non-destructive test, sample can be re-used.

R
Table 2: Hermetic Package/Product Qualification Requirements (Commercial)

New Assy Techniques (Mat’l/Process/Method
Test Test Description Acc# New New New Lead Die Die Wire Type of Lead New New New Full
Seq (note 1) S.Size Assy Pkg Pkg Frame Attach Coat Bond Seal Finish Cavity Device Fab Qual
(note 2) Plant Family Qual Size (note6) Proc
(note3) Family (note6)
(note4)
B1 Solder Heat Test (Optn’l) 0/15 X X X X X

B2 * Resist. to Solvents (note 7) 0/3 X X X X
B3 * Solderability Test (note 7) 0/3 X X X X X
B4 * Die Shear/Stud Pull (note 7) 0/5 X X X X X X X
B5 * Bond Pull (note 7) 0/2 X X X X X X X X X
B6 * External Visual (note 7) 0/25 X X X X X X X
B7 Internal Visual (note 7) 0/5 X X X X X X X X X X
C1-A High Temp Life Test 0/76 X X X X X X X
C1-B Low Temp Life Test (note 7) 0/22 X X X
C2 High Temp Storage (Optn’l) 0/77 X X X X
C3 ESD (HBM) 0/3 X X X
D1 * Phy. Dimension 0/15 X X X X X X X
D2 * Lead Integrity 0/3 X X X X X X X
D3 Thermal Shock + Temp Cycl + 0/32 X X X X X X X X X X X X X
Moisture Resistance
D4 Mech. Shock + Vibration + Con- 0/32 X X X X X X X X X X X
stant Acceleration
D5 * Salt Atmosphere 0/15 X X X X X X
D6 * Internal Vapor Content (note 7) 0/3 X X X X X X X X X
D7 * Adhesion of L/Finish (Optn’l) 0/2 X X X X X X X
D8 * Lid Torque 0/5 X X X X X X X
D9 Temp Cycle 0/45 X X X X X X X X X X X
E1 Electrical Test & Data Log 0/30 X X X
E2 Electrical Characterization 0/30 X X X
E3 T.D.D.B (note 7) – X X X
E4 Latch-up 0/9 X X X
E5 Electromigration (note 7) – X X X
E6 Photosensitivity (Optn’l) 0/11 X X X
E7 Data Retention Bake 0/22 X X X
E8 Input/Output Capacitance 0/5 X X X
E.Good 190 205 129 69 114 235 190 124 32 124 399 399 414
Qty required per lot E.Reject 81 81 75 50 8 5 2 33 41 48 7 50 81
Total 271 286 204 119 122 240 192 157 73 172 406 449 495
Notes: 1) Test method and stress conditions available upon request.
2) For any QUAL which does not meet the standard requirements, approval from Product Engineering and Product QA is
required.
3) Package Family - A set of package type with the same package, material, Package construction techniques, terminal pitch,
lead shape, row spacing and with identical package assembly tech.
4) Package Type - A package with a unique case outline, configuration, material, piece parts and assembly process.
5) Application to new piece parts or leadframe where cavity size is larger than the largest cavity size for the same package.
6) For new mask from same device family, only high temp life test, ESP, Latch & Capacitance are required.
7) In-process monitor data may be used to satisfy this requirement, for Qual data, data from Assy. lot traveler maybe used.
*) Electrical rejects can be used as test samples

R
Table 3: Failure Analysis Equipment List

Model Model
Item Equipment Vendor Number Item Equipment Vendor Number
1 Scanning Electron Microscope JEOL JMS-6401F 17 Die-Shear Tester KELLER see #7
2 Gold Sputter (SEM Sample ANATECH Hummer VIII 18 Steam Aging System Robotic ST2D
Prep) Systems
3 Energy Dispersive X-Ray OXFORD LINK 19 Solder Wave/Pot Robotic RPS-202
INST. ISIS-L200C Systems
4 F.I.B. - Focused Ion Beam F.E.I. FIB-600 20 Lead Fatigue Tester B&G 004-012-00
Workstation
5 Real-Time X-Ray Imaging Sys- FEIN FOCUS FXS-100.10 21 Conventional Oven (C.D.A.) BID Services
tem
6 Scanning Acoustic Microscopy Sonix Micro-Scan 22 Drill-bit to open MQUADS
4HF-200 + Decapping vise
7 Ball Shear Strength Tester KELLER MBS-200 23 Color Printer Tektronic Tektronic
Phaser IISD
8 XRF Lead Finish/Composition Twin City, Inc. XRF-5500 24 Stud Pull Tester B&G 003-010-00
Measurement System
9 Liquid Crystal Hot Spot Detec- Technology P/N 4330 25 Work Benches
tion System/Kit, with 3 temp. Associates
10 Emission Microscope for Hypervision Visionary 26 Cabinets
Multilayer Inspection (EMMI) 2000
11 Curve Tracer BID Services 27 Facilities (Lab Area and
Equipment Installation Costs)
12 Metallurgical High Power Scientific see quote 28 Tool Maker Microscope
Microscope Instrument (various)
Company
13 Stereozoom Low Power Scientific see quote 29 Flowhood & Rinse Station
Microscope - video camera + Instrument (various)
monitor Company
14 Micro-Etcher System TM 30 Precision X-Sectioning Equip-
Associates ment
15 Viseco Camera Interface with Computer 31 Plasma Etcher March CS-1701
High Power Microscope Modules Instruments
16 Hermeticity Test System BID Services -Trio-tech 486 32 E-Beam IDS-3000
- Fine Leak - Veeco
- Gross Leak MS-170
12

R
VCC
Configuration Data Shift Regiater VCC
Q N-1 DS Q DS Q
Read
DR DR
DK DK
SEL SEL
Data Clock
Clock Address QN-1

WR/RD
D Q
CK
Precharge
Word N
Memory
Memory Word Configuration
Cell Line
Circuit Cell Driver Address
Shift Regiater
D Q
Memory Cell
Word
Word N+1 Line
Driver CK
Memory Memory
Cell Cell
Bit M Bit M+1 X3124
Figure 2: Configuration Memory Cell
Data Integrity
Memory Cell Design in the FPGA Device are physically adjacent, supply transients result in only
small relative differences in voltages. Each inverter is truly
An important aspect of SRAM-based FPGA device reliabil-
a complementary pair of transistors. Therefore, whether the
ity is the robustness of the static memory cells used to store output is High or Low, a low-impedance path exists to the
the configuration program. supply rail, resulting in extremely high noise immunity.
The basic cell is a single-ended 5-transistor memory ele- Power supply or ground transients of several volts have no
ment (Figure 2). By eliminating a sixth transistor, which effect on stored data.
would have been used as a pass transistor for the comple- The transistor driving the bit line has been carefully
mentary bit line, a higher circuit density is achieved. During designed so that whenever the data to be written is oppo-
normal operation, the outputs of these cells are fixed, since
site the data stored, it can easily override the output of the
they determine the user configuration. Write and readback
feedback inverter. The reliability of the Write operation is
times, which have no relation to the device performance guaranteed within the tolerances of the manufacturing pro-
during normal operation, will be slower without the extra cess.
transistor. In return, the user receives more functionality
per unit area. In the Read mode, the bit line, which has a significant
amount of parasitic capacitance, is precharged to a logic
This explains the basic cell, but how is the FPGA user one. The pass transistor is then enabled by driving the word
assured of high data integrity in a noisy environment? Con- line High. If the stored value is a zero, the line is then dis-
sider three different situations: normal operation, a Write
charged to ground. Reliable reading of the memory cell is
operation and a Read operation. In the normal operation,
achieved by reducing the word line High level during read-
the data in the basic memory element is not changed. ing to a level that insures that the cell will not be disturbed.
Since the two circularly linked inverters that hold the data

R
Electrostatic Discharge Geometries and doping levels are chosen to provide ESD
protection on all pads for both positive and negative volt-
Electrostatic-discharge (ESD) protection for each pad is ages.
provided by circuitry that uses distributed transistors and/or
diodes, represented by the circles in Figure 3. In older Latchup
devices, these protection circuits are conventional diffused
structures. In newer designs, Xilinx utilizes proprietary Latchup is a condition in which parasitic bipolar transistors
device structures which exhibit substantially enhanced form a positive feedback loop (Figure 4), which quickly
ESD performance (see Table 4). reaches current levels that permanently damage the
device. Xilinx uses techniques based on doping levels and
circuit placement to avoid this phenomenon. The beta of
VCC
each parasitic transistor is minimized by increasing the
ROUT base width. This is achieved with large physical spacings.
Output The butting contacts effectively short the n+ and p+ regions
for both wells, which makes the VBE of each parasitic very
Ground close to zero. This also makes the parasitic transistors very
Pad
VCC hard to forward bias. Finally, each well is surrounded by a
dummy collector, which forces the VCE of each parasitic
Input RIN almost to zero and creates a structure in which the base
width of each parasitic is large, thus making latchup
extremely difficult to induce.
Ground
= Symbol for electrostatic discharge protection circuit
X3132 VCC
Figure 3: Input/Output Protection Circuity
Table 4: ESD Performance of Xilinx Components

Pad
Circuit Human Body Machine Charged
Family Model Model Device
Method 3015 EIAJ 20 Model CDM
XC1700D >6,000v 500– 900v >2,000v

XC2000 1,500–2,500v 250–325v pending
XC3000A 4,500-7,000v 325-600v >2,000v X1825
XC3100A 1,750-5,000v 700-800v >2,000v Figure 4: SCR Model

XC4000 4,000-8,000v 800-900v >1,000v
At elevated temperatures, 100 mA will not cause latchup. At
XC4000E 4,000-8,000v pend >2,000v
room temperature, the FPGA can withstand more than 300
XC4000E 4,000-6,000v pend >2,000v mA without latchup; the EPLD device can withstand more
XC5200 3,000-5,000v pend >2,000v than 200 mA without latchup. However, to avoid
12
XC7000 2,000-4,000v 250-300v >2,000v metal-migration problems, continuous currents in excess of
XC9000 2,000-5,000v pend >2,000v 10 mA are not recommended.
Whenever the voltage on a pad approaches a dangerous
level, current flows through the protective structures to or
High Temperature Performance
from a power supply rail (VCC or ground). In addition, the Although Xilinx guarantees parts to perform only within the
capacitances in these structures integrate the pulse to pro- specifications of the data sheet, extensive high tempera-
vide sufficient time for the protection networks to clamp the ture life testing has been done at 145°C with excellent
input, avoiding damage to the circuit being protected. results.
Revision Table
Version Date Revision
2.1 12/14/98 Revised certification dates, updated Figure 1.

1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

Table of Contents
Technical Support And Services

Support.xilinx.com Web Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
Technical and Applications Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
File Access and Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
Hotline Telephone Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
Technical Literature. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
AppLINX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
XCell Newsletter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-3
Xilinx Programmable Logic Training Courses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-3
Who Should Attend a Training Course? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-3
What You Will Learn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-3
Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-4
Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-4
Time and Cost Savings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-4
Course Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-4
Hands-On Experience. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-4
Platforms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-4
Instructors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-4
Course Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
Product Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
FPGA Tools Course . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
FPGA Tools Course Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
Foundation Interface Course. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
Foundation Interface Course Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
Foundation Express Course . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
Foundation Express Course Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-5
FPGA Synthesis Methodology Workshops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-6
VHDL Methodology Course Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-6
Verilog Methodology Workshop Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-6
Training Locations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-6
Domestic and International Locations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-6
On-Site Courses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-6
Costs: North America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-7
Costs: International . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-7
Scheduling an On-site Course . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-7
Money-back Guarantee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-7
Enrollment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-7
Student Cancellations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-7
Course Cancellations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-8
0
R
A complete and uniquely accessible offering of worldwide problems and answer questions right on the spot, and are
technical support services is available to Xilinx users. contributors to, as well as, users of the Answers database,
accessible at support.xilinx.com.
Xilinx Field Application Engineers, located at sales offices
and technical support centers worldwide, provide local Many different publications assist users in completing
engineering support, including design evaluation of new designs quickly and efficiently, including technical manuals,
projects, close consultation throughout the design process, data sheets, application notes, the AppLINX CD-ROM (a
special training assignments, and new product presenta- regularly-updated collection of the latest application notes
tions. Because their role as advisors and troubleshooters and design hints), and the quarterly XCell newsletter. Most
keeps them constantly on the go, they are best used not for of these publications are available on the support.xil-
general questions, but for more targeted queries such as inx.com web site.
those related to architectural recommendations. The For more in-depth support and instruction, a dedicated
world-wide network of Xilinx sales representatives and dis-
training organization conducts technical training classes
tributors also provide local technical support for Xilinx
worldwide. Courses geared for both novice and experi-
users. enced users are available.
Technical and applications queries can be directed to sup-
The following Technical Support Services are discussed in
port.xilinx.com, the Xilinx technical services web site, or the more detail in this chapter:
telephone “hotlines”. Xilinx provides 24-hour access to the
Answers database, product and applications information, • Support.xilinx.com World Wide Web site
and a variety of files and utilities via support.xilinx.com and • Internet File Download area
the file download areas. Hotline telephone support pro- • Hotline telephone support
vides access to permanent teams of expert Application • Technical literature
Engineers located in the United States, United Kingdom, • Training Courses
France, Germany, and Japan. These engineers can handle
13
Figure 1: HTTP://support.xilinx.com - For the Designing Engineer

R
Support.xilinx.com Web Site • The Software Update Area for access to patches,
utilities, and updates.
support.xilinx.com, the Xilinx World Wide Web site, provides • Expert Journals that provide flow-specific collections of
instant access to the latest information, ranging from Product information including FAQs, Tips, and Hot Topics.
Overviews, Application Notes, and Data Sheets to investor • Documents and applications material.
information and employment opportunities. Designed to pro- • Information about Worldwide Hotline access and
vide users with quick, easy, and intuitive access to the desired training course availability.
information.
Support.xilinx.com holds a wealth of Xilinx information, readily File Access and Transfer
available at your fingertips. What’s more, SmartSearch, our Through the file download areas, users have on-line access to
industry-wide search engine, is the definitive resource for all a variety of useful files, including user manuals, automated
Programmable Logic information on the web. SmartSearch tutorials, design examples, and utilities. Data files can be
searches over 50 different web sites rich in Programmable exchanged with Application Engineers through a secure area
Logic content, providing central access to a vast amount of of the file download area.
data. SmartSearch Agents will watch the Web for you and
inform you via e-mail when new or updated information is Hotline Telephone Support
added to any of the sites served by SmartSearch. Smart-
Search Agents allow you to stay up-to-date in the rapidly A network of Technical Support Hotlines provides Xilinx
changing world of Programmable Logic. users with direct telephone access to Xilinx Application
Engineers dedicated to providing resolutions to problems
New information is constantly being added to the Xilinx site. that may arise during the design process. Xilinx Application
The following is a list of some of the technical information now Engineers use many of the same resources and databases
available on support.xilinx.com (as of July, 1997): that are now directly available to users via the support.xil-
• Over 100 Application Notes organized by system type inx.com web site. Technical questions also can be submit-
(e.g., PCI, DSP, and PCMCIA), function (e.g., memory ted via fax or E-mail. See Table 1 on page 3 for more
functions, arithmetic functions, and busses), component information.
product family, and application.
• Complete and detailed data sheets on all Xilinx products. Technical Literature
• Over 2500 records in our Technical Answers database Xilinx offers many different publications to assist users in
that contains answers to frequently-asked technical completing designs quickly and efficiently. These include
questions. technical manuals, Data Books, data sheets, application
• Xilinx Product Change Notices and Xilinx Customer notes, the AppLINX CD, the XCell newsletter, and The
Updates Answers Database. Most of these publications are avail-
• Access to XCell, our quarterly journal for programmable able on-line at the support.xilinx.com web site.
logic users.
• Software updates and patches. As part of the development system products, Xilinx pro-
• Links to technical Xilinx presentations via Marshall vides manuals and supporting documents for the develop-
Electronics’ NetSeminarTM archives. ment system tools, libraries, CAE tool interfaces, and
related software tools. Many of these manuals are avail-
Technical and Applications able on the CD that holds the software as well as in hard-
copy format. On-line help facilities also are an integral part
Information of the development system products.
The Answers area of support.xilinx.com provides access to
technical and applications information that assists design AppLINX
engineers in solving problems. The Answers area is acces- AppLINX is a collection of current application notes and
sible from the Xilinx home page either through the other new technical documentation provided on a CD-ROM
“Answers” icon or by selecting the “Support” topic. Further, for easy reference by the design engineer. All the material
this collection of technical and applications information is on the CD is provided in Adobe Acrobat format for easy
immediately accessible through the button bar that is viewing and printing. The AppLINX CD is updated regularly
located at the bottom of every Web page. as new material becomes available.
The Answers area provides access to a variety of technical
and applications resources including:
• Over 2500 technical solutions and frequently asked
questions.

R
Table 1: Hotline Telephone Support (Continued)

Table 1: Hotline Telephone Support
Hotline: (82) 2-761-4277
All Regions of the World (support.xilinx.com): Fax: (82) 2-761-4278
web site: http://support.xilinx.com E-mail: korea@xilinx.com
North American Support: Hong Kong Support:

Hours: Mon. - Wed., Fri. Hotline: (85) 2-2424-5200
6:30 AM - 5:00 PM
Fax: (85) 2-2424-7159
Thur.
E-mail: hongkong@xilinx.com
6:30 AM - 4:00 PM Pacific Time
Hotline: 800-255-7778 or 408-879-5199 XCell Newsletter
Fax: 408-879-4442
XCell, the quarterly journal for Xilinx programmable logic
E-mail: hotline@xilinx.com users, is dedicated to supplying up-to-date information for
system designers. A typical issue includes descriptions of
United Kingdom Support: new products, updates on component and software avail-
Hours: Mon. - Thur. ability and revision levels, application ideas, design hints
9:00 AM - 12:00 PM, 1:00 PM - 5:30 PM and techniques, and answers to frequently-asked ques-
Fri. tions.
9:00 AM - 12:00 PM, 1:00 PM - 3:30 PM To add your name to the XCell subscription list, please
Hotline: (44) 1932-820821 send your name, company affiliation, and mailing address
Fax: (44) 1932-828522 to XCell editor, via FAX at 408-879-4676.
E-mail: ukhelp@xilinx.com
Xilinx Programmable Logic Training
France Support: Courses
Hours: Mon. - Fri. Who Should Attend a Training Course?
9:30 AM - 12:30 PM, 2:00 PM - 5:30 PM
All users of Xilinx products should attend one of our training
Hotline: (33) 1-3463-0100
courses. Attending a Xilinx training course is one of the
Fax: (33) 1-3463-0959
fastest and most efficient ways to learn how to design with
E-mail: frhelp@xilinx.com FPGA devices from Xilinx. Hands-on expert instruction with
the latest information and software will allow you to imple-
Germany Support: ment your own designs in less time with more effective use
Hours: Mon. - Thur. of the devices.
8:00 AM - 12:00 PM, 1:00PM - 5:00 PM Classes are held regularly in centers around the world and
Fri. can even be brought to your own facility.
8:00 AM - 12:00 PM, 1:00 PM - 3:00 PM
Not only design engineers, but also test engineers, compo-
nent engineers, CAD engineers, technicians, and engi-
Hotline: (49) 89-93088-130
neering managers may want to attend the course in order
Fax: (49) 89-93088-188 to understand the Xilinx products.
E-mail: dlhelp@xilinx.com
13
What You Will Learn
Japan Support: Not only will you learn about our products, but we will also
Hours: Mon., Tue., Thur., Fri. recommend the best ways to use the software based on
9:00 AM - 5:00 PM our years of experience with thousands of designs. You will
Wed. learn how to efficiently enter, implement, and verify your
9:00 AM - 4:00 PM design. You can use the Xilinx automatic mode or take a
Hotline: (81) 3-3297-9163 power-user approach and guide the automatic tools to the
best implementation of your design.
Fax: (81) 3-3297-0067
E-mail: jhotline@xilinx.com
Korea Support:

R
will quickly become your company’s expert in Xilinx

designs.
Lower Production Costs

By learning how to use the device effectively, you may be
able to get more logic into a smaller device, and/or operate
at a higher speed. As a result, you may be able to save on
the cost of the device itself and the surrounding logic on
your board.
Increase Quality
Effective verification techniques will prove the quality of
your Xilinx-based design. Higher quality leads to less main-
tenance and repair costs, and improved customer satisfac-
tion.
Time and Cost Savings

Attending a Xilinx training course is an investment that will
pay for itself with the first Xilinx design that you begin. The
courses are fast-paced, each providing as much informa-
tion as possible in the short time available. Hands-on expe-
riences throughout the courses make sure that the
Prerequisites information is retained and applied to practical applications.
Just as Xilinx products reduce your development time,
Students need only have a background in digital logic
attending a training course can reduce your design time.
design. Basic familiarity with the PC or workstation is help-
The person attending the course will be an in-house expert
ful but not required. It will benefit you to learn your design
who can be utilized by other members of your company.
entry tool of choice before attending the Xilinx course.
You can reduce your travel costs by attending a class
If you would like to prepare for the training course to maxi-
scheduled in your area or having the course brought right
mize your learning, you should complete the tutorials avail-
to your facility. The tuition pays for the course notes and
able in the development system. There are tutorials
expert, in-person instruction, which can be priceless when
available for the third-party tools for which Xilinx sells the
trying to meet a schedule.
interface.
Benefits Course Descriptions

Make Fewer Design Iterations Hands-On Experience
By learning the proper approach, you will save time and Each course includes hands-on labs. There is at least one
expense in prototyping and debugging designs. However, if computer for every two people in the class.
you do need to make changes to your design, you will learn
how to do this quickly and efficiently. Platforms
The courses use PC platforms with either WIN95 or NT
Get to Market Faster operating systems.
Getting your product to market faster is probably one of the
key reasons you are using Xilinx products in the first place. Instructors
Studies have shown that time-to-market often has a greater Xilinx training courses have been successfully held world-
effect on profits than development costs. Training will allow wide for over eight years. The instructors are Xilinx experts
you to get your product to market on schedule, allowing who are skilled at passing that knowledge on to fellow engi-
your company to reap the rewards that follow. neers. A dedicated Customer Education organization at Xil-
Reduce Your Learning Time inx works closely with the Applications and Engineering
groups to keep the courses up-to-date with the latest
Extensive Xilinx documentation and tutorials provide the improvements to Xilinx and third-party tools.
information you need to complete your design. But attend-
ing the training course for focused, interactive learning is
faster than a question-and-answer approach on your own.
Instead of interruptions and piecemeal self-education, you

R
Course Materials Day 2

Hardware Optimization
All course materials are supplied by Xilinx. Every student
Floorplanner Layout Lab
gets an excellent reference tool in the form of course notes,
Advanced Timing Constraints
which include all the material presented during the course.
Advanced Constraints Lab
The course notes are bound for easy use and include addi-
Implementation Options
tional reference material beyond what is covered in the
Advanced PAR Options
course.
Advanced Optimization Features Lab
Product Coverage EPIC
Xilinx courses (See Table 2) will cover the latest released Foundation Interface Course
versions of our devices and development systems. While
The one-day Foundation Schematic Entry course will pro-
all available products will be covered, emphasis will be
vide customers with the knowledge to fully utilize the tools
placed on the more popular and/or recommended solu-
capabilities. The tool contains a schematic editor, HDL edi-
tions. New products are added to the courses as they
tor, state editor, and a simulator that are tightly integrated
become available. If you have any questions on coverage
with the M1 Implementation tools. The course is heavily
of a particular product, please call the Xilinx Customer Edu-
focused on the labs. Each lab individually addresses a spe-
cation department.
cific piece of the tool and its use in effectively designing an
Table 2: Xilinx Courses FPGA.
Course Title Course Length Foundation Interface Course Outline
FPGA Tools 2 days
The one-day Foundation course addresses the following
Foundation Interface 1 day
topics:
Foundation Express 1 day
VHDL Methodology 3 days Foundation Introduction
Basic FPGA Architecture and Libraries
Verilog Methodology 3 days
Project Manager
Schematic Editor and Lab
FPGA Tools Course LogiBLOX and Lab
This Xilinx training course is two days in length. It is HDL Editor and Lab
designed to be an introduction to Xilinx’ products and tools State Editor and Lab
for those customers who have little or no experience with Simulator and Lab
the Xilinx environment. All North American training sites Advanced Simulator Usage and Lab
and most international locations teach this same course.
The FPGA Tools course is heavily focused on the labs, Foundation Express Course
which feature Xilinx’ Foundation Software. The one-day course is an introduction to Foundation using
Foundation Express. The attendee will learn how to enter a
FPGA Tools Course Outline
design and create an optimized FPGA; the design will be
The following topics will be discussed during the Tools verified in silicon. Customers should have a basic under-
course: standing of VHDL or Verilog before attending this course.
This course supports both VHDL and Verilog.
Day 1
Course Introduction Foundation Express Course Outline 13
Basic FPGA Architecture
Design Entry The agenda for the Foundation Express course is as fol-
LogiBLOX GUI and the Core Generator lows:
LogiBLOX Lab • Design Entry: Foundation Tools for Easy Design
Design Manager and Flow Engine Entry
General Design Flow Lab - Design Wizard
Report Browser - Language Assistant
Timing - LogiBLOX
Basic Constraints • Lab
Timing Analyzer • Synthesis and Implementation
Basic Timing Constraints Lab - Push-button Design Flow
On-line Documentation - Analyzing the Design Performance
• Lab

R
• Constraining the Design Verilog Methodology Workshop

- Methods for Constraining Synthesis and
Implementation
Outline
- Pin Locking, Control of I/O, Timing Constraints The Verilog Methodology workshop includes the following
• HDL Coding Styles topics:
- Design Tips for Small, Fast Designs (VHDL &
Verilog) Day 1
• Advanced Implementation Options (Optional-Time - Verilog HDL Introduction
Dependant) - Hardware Modeling Overview
- Software Techniques for Implementation of Designs - Language Concepts
with Tight Timing Constraints - Verilog Modules & Ports
- Gate Level Modeling
FPGA Synthesis Methodology - Operators & Expressions
- Lab Exercises:
Workshops - Write Basic Modules
These comprehensive courses are an effective introduction - Write Gate Level Model
to the VHDL and Verilog languages with particular empha- - Build Hierarchical Structures with Instantiation
sis on targeting Xilinx and FPGA devices in general. The
information gained here can be applied to any digital Day 2
design using a Top-Down synthesis design approach. - Data-Flow Modeling
These courses are a combination of insightful lecture, cou- - Behavioral Modeling
pled with practical and interesting lab exercises to reinforce - Advanced Process Statements
the key concepts. - Verilog Timing Models
- Targeting Xilinx
They are a great way to quickly gain proficiency in one of - Lab Exercises:
today’s fastest growing hardware design techniques - Write Behavioral Model
- Write & Compile 4 State-Machine
VHDL Methodology Course Outline
The following topics will be discussed during the VHDL Day 3
Methodology course: - Introduction to the Verilog Testbench
- Tasks & Functions
Day 1 - Xilinx Synthesis Tools
- Introduction to VHDL & Hardware Modeling - Lab Exercises:
- Language Concepts - Build Complex Address Decoder
- Signals & Datatypes - Download & In-circuit Verification of Synthesized
- Operators / Expressions Design
- Concurrent & Sequential Statements
- Lab Exercises: Training Locations
- Write Entities
- Build Hierarchical Structures with Instantiation Domestic and International Locations
Day 2 Xilinx courses are held not only in North America but also
- Advanced Process Statements throughout Europe, Asia, India, Israel, Australia, South
Africa, and South America. The classes that are held in
- Behavioral to RTL
international locations vary in length and tuition but are
- VHDL Timing Models
- Targeting Xilinx based on the same material used in North America. The
- Lab Exercises: schedule lists the courses as well as the contact’s name
- Write Code for 4 State State-Machine and phone number.
- Build One Hot Encoded State-Machine
On-Site Courses
Day 3 Xilinx can bring the training course to your own facility for
- Introduction to the VHDL Testbench the greatest convenience to your company.
- Functions / Procedures
On-Site Courses Provide Additional Benefits:
- Xilinx Synthesis Tools
- Lab Exercises: No Travel Costs
- Build Testbench for Module Verification • On-site Xilinx training courses eliminate travel time and
- Download & In-circuit Verification of Synthesized expenses:
Design

R
- No airfare Costs: International

- No hotel bills
• Prices vary; contact your local Xilinx sales
- No car rental
representative.
Courses Tailored To Your Needs Included in course fees:
On-site courses can be tailored to meet the specific needs • A Xilinx-certified instructor
of your company: • Training materials for each student
• Convenient class time and location
• Projects of a proprietary nature can be discussed Scheduling an On-site Course
openly To schedule a training class at your facility and determine
• Students can use their own equipment and begin an pricing, call 877-XLX-CLASS. On-site training courses are
actual design during the course popular, so the more advanced notice we have, the better
our ability to schedule your class exactly when you want it.
Costs: North America (Please see Table 3.)
• Prices start at $5,000 for a minimum class size of six
students. (Prices are subject to change without notice.)
Table 3: Course/Tuition Table
Course Location Tuition Length Benefits

North America
FPGA Tools $595 2 days Basic implementation of Xilinx products
Foundation Interface $195 1 day A complete design-entry flow
Foundation Express $195 1 day Tool usage with VHDL or Verilog design examples
VHDL Methodology $795 3 days Introduction/Intermediate IEEE 1076
VHDL Targeting FPGA and simulation
Verilog Methodology $795 3 days Introduction/Intermediate IEEE 1369
Verilog targeting FPGA and simulation
On-Site Locations
Call for Quote Varies Convenience
Focus on specific issues
International:
Multiple Locations Varies Varies Offered in over 21 countries
Native language
On-Site Locations Varies Varies Can focus on specific issues Convenience
(Prices and class schedules are subject to change without notice. Prices listed are in U.S. dollars)
Money-back Guarantee Express, and Discover) as well as Purchase Orders and

checks.
We are so confident you will be satisfied with the benefits of
a Xilinx training course that we offer the following guaran- 2. Confirmation: Following enrollment in a class, the regis-
tee: trar will send you a confirmation letter.
13
• Full refund of the course cost if you are not completely Please call 877-XLX-CLASS if you have any questions or
satisfied. do not receive a confirmation letter.
If you are unsatisfied or have further questions, call 877-
Enrollment XLX-CLASS (toll-free).
Class size is limited, so early enrollment is recommended.
In order to register for a class, please follow the steps listed Student Cancellations
below: Course tuition is fully refundable up to two weeks before
1. Enroll: Contact the registrar at 877-XLX-CLASS to enroll the class starts. Cancellations within two weeks (7-13
yourself in a class. You must have your tuition payment days) of the class start date will incur a 50% cancellation
information available at the time of enrollment. We now fee. Those who cancel fewer than 7 days prior to the class
accept credit cards (Visa, MasterCard, American will be billed for the full amount of the tuition.

R
Course Cancellations
Xilinx reserves the right to cancel any class with up to 14
days notice. Typically our process is to look at a class’
enrollment number 14 days before a its scheduled start
date and determine if the number of students enrolled is
sufficient for us to hold or cancel the class. Please be
aware of this when booking your travel arrangements.

1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

1

Table of Contents
1 14*
Xilinx FPGAs: A Technical Overview for the First-Time User

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
User Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
Logic Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
Special Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
Inputs/Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
Global Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
Power Consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
Programming or Configuring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
Design Entry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
Design Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
Configuring the FPGA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
Quality and Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
Choosing a Xilinx Product Family

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-7
SRAM-Based FPGAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-7
Overview of SRAM-Based FPGA Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-8
FLASH-Based CPLDs (XC9500) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-9
Overview of CPLD Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-9
Selecting the Xilinx Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-10
Type of Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-10
Special Features Required . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-10
Further Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-12
I/O Characteristics of the ‘XL FPGAs

Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-13
Xilinx Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-13
Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-13
5-V Tolerant Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-13
PCI-Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-13
Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-14
Sink and Source Capability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-14
Effect of Additional Capacitive Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-14
XC4000 Series Technical Information

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-15 14
Voltage/Current Characteristics of XC4000-Family Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-15
Additional Output Delays When Driving Capacitive Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-16
Ground Bounce in XC4000 Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-16
Test Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-16
Interpretation of the Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-17
Guidelines for Reducing Ground-Bounce Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-17
Ground-Bounce vs Delay Trade-Off . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-17
XC4000 and XC4000E Power Consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-18
14-1
Product Technical Information Table of Contents
Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-19
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-19
Configurable Logic Blocks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-19
Function Generator Avoids Glitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-20
Input/Output Blocks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-21
Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-21
Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-22
I/O Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-23
Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-23
Horizontal Longlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-23
Internal Bus Contention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-23
Vertical Longlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-24
Clock Buffers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-24
Power Dissipation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-24
Crystal-Oscillator Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-25
CCLK Frequency Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-27
CCLK Low-Time Restriction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-27
Battery Back-up. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-27
Powerdown Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-28
Configuration and Start-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-28
Start-Up. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-28
Beware of a Slow-Rising XC3000 Series RESET Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-29
FPGA Configuration Guidelines

Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-30
Protection Against Data or Format Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-31
Daisy-Chain Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-31
Start-Up Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-31
Configuration Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-33
Selecting the Best Configuration Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-34
When Configuration Fails . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-34
General Debugging Hints for all Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-34
General Debugging Hints for the XC2000 and XC3000 Families . . . . . . . . . . . . . . . . . . . . . . . . 14-35
General Debugging Hints for the XC4000 and XC5000 Families . . . . . . . . . . . . . . . . . . . . . . . . 14-35
Additional Mode-Specific Debugging Hints for All Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-35
Daisy Chain Debugging Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-36
Potential Length-Count Problem in Parallel or Peripheral Modes. . . . . . . . . . . . . . . . . . . . . . . . 14-37
Miscellaneous Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-37
Configuring Mixed FPGA Daisy Chains

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-38
Configuration Issues: Power-up, Volatility, Security, Battery Back-up

Power-Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-40
Sensitivity to VCC Glitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-40
Design Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-41
Design Security when Configuration Data is Accessible . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-41
Design Security by Hiding the Configuration Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-42
Battery Back-up and Powerdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-42
Powerdown Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-43
Dynamic Reconfiguration
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-44
Important Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-44
14-2
Reconfiguration Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-45
Initiating Reconfiguration in Different Xilinx Device Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-45
XC3000 Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-45
XC4000 Series and XC5200 Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-45
FPGAs Can Control Their Own Reconfiguration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-45
Metastable Recovery
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-46
Metastability Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-47
Metastability Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-47
Set-up and Hold Times

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-49
Overshoot and Undershoot

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-50
Boundary Scan in XC4000 and XC5200 Series Devices

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-51
Overview of XC4000/XC5200 Boundary-Scan Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-51
Deviations from the IEEE Standard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-52
Boundary-Scan Hardware. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-53
Test Access Port. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-53
TAP Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-53
The Boundary-Scan Data Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-53
Bit Sequence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-54
The Bypass Register. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-55
User Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-55
Using Boundary Scan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-56
Boundary Scan Instructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-57
EXTEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-57
SAMPLE/PRELOAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-58
BYPASS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-58
USER1, USER2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-58
CONFIGURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-58
READBACK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-60
Boundary Scan Description Language Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-61
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-61
Virtex I/V Curves for VariousOutput Options

Virtex Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-62
I/V Curves for VariousDevice Families

XC4000 Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-64
Spartan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-65 14
14-3
Product Technical Information Table of Contents
14-4
APPLICATION NOTE
0
R
Xilinx FPGAs: A Technical Overview
for the First-Time User
XAPP 097 December 12, 1998 (Version 1.3) 0 14* Application Note by Peter Alfke
Summary
This Application Note introduces the reader to the various Xilinx product family’s logic components and provides a general
overview of what the logic components within the devices are used for.
Xilinx Families
Spartan ™, XC3000, XC4000, XC5000, XC9000
Introduction structure, but interconnect delays are layout dependent and

must be analyzed if they are performance critical.
In the Spartan™, XC3000, XC4000, and XC5200 device
families, Xilinx offers several evolutionary and compatible Clocks
generations of Field Programmable Gate Arrays (FPGAs).
Clock lines are well-buffered and can drive all flip-flops with
Here is a short description of their common features.
< 2 ns skew from chip corner to corner, even throughout the
Every Xilinx FPGA performs the function of a custom LSI biggest device. The user need not worry about clock load-
circuit, such as a gate array, but the FPGA is user-program- ing or clock-delay balancing, or about hold time issues on
mable and even reprogrammable in the system. Xilinx sells the chip, if the designated global clock lines are used.
standard off-the-shelf devices in multiple families, and There are eight such global low-skew clock lines in Spartan
many different sizes, speeds, operating-temperature and XC4000 devices, four in XC5200 devices, and two in
ranges, and packages. The user selects the appropriate XC3000 devices.
device and then enters the logic design into the Xilinx
development system software running on a PC or worksta- Special Features
tion, and then loads the resulting configuration file into the
All devices can implement internal bidirectional busses.
Xilinx FPGA.
The Spartan, XC4000 and XC5200 family devices have
This overview describes two aspects of Xilinx FPGAs; what dedicated fast carry circuits that improve the efficiency and
logic resources are available to the user and how the speed of adders, subtractors, comparators, accumulators
devices are programmed. and synchronous counters. These families also support
boundary scan on every pin.
User Logic Spartan and XC4000 series devices can use any of their
Different in structure from traditional logic circuits, or PALs, logic block look-up tables as distributed RAM, with synchro-
EPLDs and even gate arrays, the Xilinx FPGAs implement nous write and dual-port options. This makes FIFOs, shift
combinatorial logic in small look-up tables (16 x 1 ROMs); registers and DSP distributed multipliers very fast and effi-
each such table either feeds the D-input of a flip-flop or cient.
drives other logic or I/O. Each FPGA contains a matrix of
identical logic blocks, usually square, from 8 x 8 in the Inputs/Outputs
XC4002XL to 68 x 68 in the XC40125XV. Metal lines of var- All device pins are available as bidirectional user l/O, with
ious lengths run horizontally and vertically in-between 14
the exception of the supply connections and a few dedi-
these logic blocks, selectively interconnecting them or con- cated configuration pins. The outputs on XC3000 and
necting them to the input/output blocks.
XC5200 devices always swing rail-to-rail. Spartan and
XC4000E/EX outputs have a global choice between “TTL =
Logic Blocks
totem pole” or “CMOS = rail-to-rail” output swing.
This modular architecture is rich in registers and powerful
The original families operate from a 5 V supply, but have
function generators that can implement any function of up
added 3.3 V variants. These 3.3 V devices, designated by
to five variables. For wider inputs, function generators are
an “L” in their product name, have rail-to-rail outputs.
easily concatenated. Generous on-chip buffering makes
logic block delays insensitive to loading by the interconnect Inputs of all 5 V devices can be globally configured for
either TTL input thresholds or CMOS thresholds (50% of
XAPP 097 December 12, 1998 (Version 1.3) 14-5

R
Xilinx FPGAs: A Technical Overview for the First-Time User
Vcc). All 3.3 V devices have CMOS input thresholds. All form the intended digital function. The number of configura-
inputs have hysteresis (Schmitt-trigger action) of 100 to tion bits varies with device type, from 14,819 bits for the
200 mV. SpartanXL and XC4000XL inputs are uncondition- smallest device (XC3020) to 2,797,040 bits for the largest
ally 5 V tolerant, even while their supply voltage is as low as device (XC40125XV). Multiple FPGA devices can be
0 V. This eliminates all power-supply sequencing problems. daisy-chained and configured with a common concate-
nated bitstream. Device utilization does not change the
Global Reset number of configuration bits. Inside the device, these con-
All Xilinx FPGAs have a global asynchronous reset input figuration bits control or define the combinatorial circuitry,
affecting all device flip-flops. In the Spartan, XC4000 and flip-flops, interconnect structure, and the I/O buffers, as well
XC5200 family devices, any pin can be configured as a as their pull-up or pull-down resistors, input threshold and
reset input; in XC3000-families, RESET is a dedicated pin. output slew rate.
Power-up Sequence
Power Consumption
Upon power-up, the device waits for VCC to reach an
Since all Xilinx FPGAs use CMOS SRAM technology, their
acceptable level, then clears the configuration memory,
quiescent or stand-by power consumption is very low, micro-
holds all internal flip-flops reset, and 3-states the outputs
watts for XC3000 devices, max 25 mW to 75 mW for the
but activates their weak pull-up resistors. The device then
other 5 V families. The operational power consumption is
initiates configuration, either as a master, (clocking in a
totally dynamic, proportional to the transition frequency of
data stream from an external source), or as a slave
inputs, outputs, and internal nodes. Typical power consump-
(accepting clock and data stream from an external source).
tion is between 100 mW and 5 W, depending on device size,
clock rate, and the internal logic structure. Bit-Serial Configuration
All devices monitor VCC continuously and shut down when The Xilinx serial PROM is the simplest way to configure the
VCC drops to 3 V (2 V for 3.3 V devices). The device then FPGA, using only three or four device pins. Typical configu-
3-states all outputs and prepares for reconfiguration. ration time is around one microsecond per bit, but this can
be reduced. Configuration thus takes from a few millisec-
Programming or Configuring onds to a several hundred milliseconds. Serial PROMs
come in sizes from 36K to 4M bits and can be daisy
Design Entry chained to store a longer bitstream.
A design usually starts as a schematic, drawn with one of
the popular CAE tools, or as a High-Level Language textual
Byte-Parallel Configuration
description. Most CAE tools have an interface to the Xilinx Xilinx SpartanXL, XC3000, XC4000, and XC5200 FPGA
development system, running on PCs or workstations. devices can also be configured with byte-wide data, either
from a PROM or from a microprocessor. Parallel configura-
Design Implementation tion modes are not faster than serial modes except in Spa-
After schematic or HDL design entry, the design is read by tanXL and XC4000XLA.
the Xilinx software. The software first partitions the design Reconfiguration
into logic blocks, then finds a near-optimal placement for
each block, and finally selects the interconnect routing. The user can reconfigure the device at any time by pulling
This process of partitioning the logic, placing it, and routing the PROGRAM pin Low, to initiate a new configuration
the interconnects runs automatically, but the user may also sequence. During this process, outputs not used for config-
affect the outcome by imposing specific timing constraints, uration are 3-stated. Partial reconfiguration is not possible.
or selectively editing critical portions of the design. The For high-volume high-density applications, Xilinx offers
user thus has a wide range of choices between a fully auto- lower cost, fixed programmed HardWire versions.
matic implementation and detailed involvement in the lay-
Readback of Configuration Data
out process.Once the design is complete, a detailed timing
report is generated and a serial bitstream is produced After the device has been programmed, the content of the
which can be downloaded into the FPGA, into a PROM pro- configuration "shift register" can be read back serially, with-
grammer, or made available as a computer file. out interfering with device operation.
Configuring the FPGA Quality and Reliability

The user then exercises one of several options to load this Since 1985, Xilinx has shipped over 100 million FPGA
file into the Xilinx FPGA device, where it is stored in devices. Industry leading quality and reliability (ESD pro-
latches, arranged to resemble one long shift register. The tection, AQL and FIT) and aggressive price reductions have
data content of these latches customizes the FPGA to per- undoubtedly contributed to this success.
14-6 XAPP 097 December 12, 1998 (Version 1.3)

APPLICATION NOTE
0
R
XAPP100 December 3, 1998 (Version 1.4) 0 14* Application Note by Peter Alfke
Summary
This Application Note describes the various Xilinx product families. Differences between the families are highlighted. The
focus of the discussion is how to choose the appropriate family for a particular application.
Xilinx Families
Spartan ™, XC3000, XC4000, XC5000, XC9000
Introduction SRAM-Based FPGAs

Xilinx offers Field-Programmable Logic circuits, mass pro- These families represent an ongoing evolution of the origi-
duced standard integrated circuits that the user can cus- nal Xilinx FPGA architecture, characterized by structural
tomize for the specific application. flexibility and an abundance of flip-flops. Logic is imple-
mented in look-up tables, and is interconnected by a hierar-
Xilinx products offer the following advantages:
chy of metal lines controlled by pass transistors.
• High integration (less space, lower power, higher
Attractive systems features include on-chip bidirectional
reliability, lower cost) than solutions based on existing
busses and individual output 3-state and slew-rate control,
standard devices like MSI and PALs.
• No non-recurring engineering charges and associated common reset for all flip-flops, and multiple global low skew
risk, typically required for mask-programmed gate array clock networks.
solutions. The configuration can be loaded while the devices are con-
• Fast design time and easy design modification, nected into a system, and can be changed an unlimited
important for early time-to-market. number of times by reloading the “bitstream,” the series of
• Designs can be upgraded in the field for added bits used to program the device. Configuration must be
functionality. reloaded whenever Vcc is re-applied. Reconfiguration
takes 20 to 200 ms, during which time all outputs are inac-
Some potential users might be confused by the wide diver-
sity of Xilinx product offerings. This application note pro- tive.
vides a broad overview from the user’s perspective. Static power consumption is very low, down to microwatts
Xilinx offers programmable logic circuits in two distinctly dif- for some of the families. Dynamic power consumption is
ferent technologies. proportional to the clock frequency, and depends on the
logic activity inside the device and on the outputs.
• SRAM-based FPGAs, the original Xilinx offering, now
encompassing the Spartan, XC3000, XC4000, and The description “SRAM based” refers primarily to the stan-
XC5200 series and their sub-families, like the dard high-volume manufacturing process, and secondarily
XC3000A, XC3000L, XC3100A, XC4000E, XC4000EX, to the fact that configuration data is stored in latches. Differ-
and XC4000XL. ent from typical SRAMs, these latches use low-impedance
active pull-up and pull-down transistors. An on-chip voltage
• Flash-based complex PLDs, the XC9500 family.
monitor 3-states the outputs and initiates reconfiguration
14
when Vcc drops significantly (to 3.2 V in a 5V system).
These FPGAs are available in different sizes and many dif-
ferent packages. Usually each device type is available in
many package types. Any package can accommodate dif-
ferent sized devices with compatible pinouts, so the user
can migrate to a larger or smaller device without changing
the PC-board layout.
XAPP100 December 3, 1998 (Version 1.4) 14-7

R
Overview of SRAM-Based FPGA Families • Accept lack of dedicated carry circuits, resulting in less
efficient and possibly slower arithmetic and counters
XC2000: Obsolete, do not use for new designs. than in XC4000. No on-chip RAM; data storage is thus
The Spartan FPGA families or the XC9500 CPLD family, limited to the available 256 to 1,320 flip-flops.
may be an alternative. XC3100L: 3.3V version of XC3100A
XC2000L: 3.3V version of XC2000; Obsolete, do not use • Use for 3.3V applications.
for new designs. • Accept significantly slower speed at 3.3V, compared to
Use the SpartanXL family instead. XC3100A at 5V, as well as higher quiescent power and
much higher powerdown current than XC3000L at 3.3V.
XC3000: Superseded
XC4000: Superseded
Don’t use this family for new designs, since it has been
superseded by the improved, but fully backwards compati- Don’t use this family for new designs, since it has been
ble, XC3000A family. superseded by the improved, but fully backwards compati-
ble XC4000E family.
XC3000A: Newest version of the XC3000 family
XC4000A: Superseded
Five device types cover a complexity range from 1,300 to
7,500 gates, with 256 to 928 flip-flops. Logic is imple- Don’t use this family for new designs, since it has been
mented in 4-input look-up tables; two tables can be com- superseded by the improved, faster, less expensive, pinout
bined to implement any logic function of five variables with compatible (but not bitstream compatible) XC4000E family.
only one combinatorial delay of 4 or 5 ns. Flip-flop toggle XC4000H: High I/O - count version of XC4000; Obsolete,
rate is over 110 MHz. do not use for new designs.
Global choice of input thresholds (1.2 V or 2.5V), output XC4000E: Enhanced superset of the XC4000 family
slew-rate control, and an on-chip crystal oscillator circuit
are attractive system features. The XC4000E family is recommended for new designs.
• Use for medium speed, medium complexity The ten devices in this family stretch from 2,000 to 25,000
applications. logic gate complexity. The emphasis is on systems features
• Accept lack of dedicated carry circuits, resulting in less and speed. The function generators are more versatile than
efficient and slower arithmetic and counters than in in the XC3000-Series parts, and there is a dedicated carry
XC4000 families. No on-chip RAM; data storage is thus network to speed up arithmetic and counters and make
limited to the available 256 to 928 flip flops. them more efficient. Most importantly, the function genera-
tors can be used as user RAM with asynchronous or syn-
XC3000L: 3.3V version of XC3000A chronous write addressing, even as dual-port RAMs. This
• Use for battery operated applications. capability makes register files, shift registers and especially
• Accept significantly slower speed at 3.3V, compared to FIFOs faster and much more efficient than in any other
XC3000A at 5V. FPGA. Dedicated carry logic can speed up wide arithmetic
• For high-performance battery operated applications and long counters.
review SpartanXL as option. • Use for general-purpose logic and data-path logic that
XC3100: Superseded can take advantage of internal busses and fast
arithmetic carry logic. Use for on-chip distributed RAMs,
Don’t use this family for new designs, since it has been
e.g. >50-MHz FIFOs up to 64 deep, 32 bits wide.
superseded by the improved, but fully backwards compati-
ble XC3100A family. XC4000EX: Larger version of the XC4000E family.
XC3100A: Newest version of the high-speed XC3100 fam- Extension of the XC4000E family from 28k to 36k logic
ily. gates, with greatly increased routing resources, faster
clocking options and more versatile output logic.
XC3100A devices are functionally and bitstream identical
with the XC3000A, and are available in the same packages • Use for designs beyond 20,000 gate complexity.
with the same pinouts. The only difference is the higher XC4000XL: 3.3V FPGA
speed of the XC3100A, with a look-up table delay of 1.5 to
4 ns, and the slightly higher standby current of 8 to 14 mA. Complete family stretching from 5000 gates to >100,000
One additional high-end family member, the XC3195A, can gates. Basic features are identical to the XC4000EX but
implement up to 9,000 gates and 1,320 flip-flops. with 5V tolerant input, even when Vcc is <3.0V.
• Use for high performance design with system clock • Use for 3.3V designs, and highest performance.
rates up to 100 MHz.
14-8 XAPP100 December 3, 1998 (Version 1.4)

R
XC4000XLT: 3.3V FPGA, 5V PCI compatible Performance is similar to XC3000A, but dedicated carry
logic can speed up wide arithmetic and long counters.
The XC4000XLT adds 5V VTT pins to the XC4000XL to
enable the positive input signal clamping function required
by 5V PCI specifications.
FLASH-Based CPLDs (XC9500)
These devices are extensions of the popular PAL architec-
XC4000XV: 2.5V FPGA
ture, implementing logic as wide AND gates, ORed
SImilar architecture to XC4000EX (5V) and XC4000XL/T together, driving either a flip-flop or an output directly. The
(3.3V). Together, these families are referred to as simple logic structure makes these devices easy to under-
“XC4000X”. The XC4000XV extends the family to the larg- stand, and results in both fast design compilation and short
est FPGAs available, in the 250,000-500,000 system gate pin-to-pin delays. Wide input gating and fast system clock
range. The XC4000XV family uses 0.25 micron technology, rates up to 150 MHz are attractive features for state
with a 2.5V core supply and a 3.3V I/O supply for 5V com- machines and complex synchronous counters.
patibility.
The XC9500 in-system programmable family, based on
Spartan: 5V lowest cost, FPGA based on XC4000 FLASH technology, eliminates the need for a separate pro-
The Spartan Series of FPGAs offers the best high volume grammer. These new devices also offer boundary scan
FPGA solution for ASIC replacement. Derived from the (JTAG) to simplify board testing.
highly successful XC4000 architecture and spanning up to
40,000 system gates, the Spartan Series combines high Overview of CPLD Families
performance, on-chip RAM, software cores, and lowest XC7200: Obsoleted
prices to meet all the key requirements of ASIC designs for
Do not use for new designs. Use XC9500 instead.
high volume production, and delivers unmatched benefits
over competing programmable logic solutions. XC7300: Superseded
Use for all high-performance, low cost designs up to 40,000 Do not use for new designs. Use XC9500 instead.
system gates in size. XC9500: FLASH-Based CPLD
SpartanXL™: 3.3 V version of Spartan FPGA Six devices cover the range from 36 to 288 macrocells.
Similar architecture to 5V Spartan family, but providing The new XC9500 family provides advanced in-system pro-
higher speed, lower power, and lower cost using smaller gramming and test capabilities for high performance, gen-
process technology. The Spartan and SpartanXL families eral purpose logic integration.
are together referred to as the Spartan Series.
• Delays are deterministic, and compile times are very
XC5200: Low cost FPGA short.
Architecture optimized for low cost, good routability, and the • Use for high-speed logic, short pin-to-pin delays, for
ability to lock pinout while internal logic is being modified. state machines and flexible address decoding, and as
Dedicated carry structure similar to XC4000, but no RAM. PAL replacement.
Four-input function generators avoid the XC3000 input con- • Accept higher power consumption and fewer available
straints. IOBs are less rigidly coupled to the internal matrix flip-flops compared to SRAM FPGA.
of CLBs and interconnects, which greatly improves the flex-
ibility of pin-locked designs. IOBs have no flip-flops.
14

R
Selecting the Xilinx Family 6. For pinout compatibility within and between families:
It is not always obvious which Xilinx family is the “right” Use XC4000E/X, XC5200.
choice for a particular application. To make a decision, start These families are carefully designed to fit the same pinout
with the known data, the target application. Then address in any given available package. This allows easy migration
the following questions: to different device sizes or families in the same package.
• What type of logic is used in the application? The user can add logic or streamline the design or even
• What special features are required? use a less costly or faster family without any need to
change the existing PC-board layout.
Type of Logic
7. For Digital Signal Processing (multiply-accumulate)
All Xilinx devices are general purpose. Any family can applications:
implement any type of logic. There are, however, some fea-
Use Spartan/XL or XC4000E/X families.
tures that make certain families more appropriate than oth-
ers. The following items should be interpreted as “soft” The look-up-table architecture and the dedicated carry
suggestions, not as absolute, unequivocal choices. structure are very efficient for distributed arithmetic, a fast
and effective way to implement fixed-point multiplication in
1. For shortest pin-to-pin delays and fastest flip-flops: digital filters.
Use XC9500, or, if fan-in is sufficient, Spartan/XL,
XC3100A or XC4000E/X families. Special Features Required
XC9500/XL CPLDs have a PAL-like AND/OR structure that The fourteen items below describe specific features and
is inherently very fast. XC3100A has extremely fast logic characteristics available only in the listed families. These
blocks, but the single-level fan-in is limited to five. are, therefore, “hard” selection criteria.
Spartan/XL, XC4000E/X families have a wider fan-in of 1. For on-chip RAM:

nine. XC4000X FPGAs offer a very fast pin-to-pin path
Use Spartan/XL or XC4000E/X families.
using a fast buffer and a 2-input function generator in the
IOB. Has many 16x1 or 32x1 RAMs with synchronous write and
dual-port capability.
2. For fastest state machines:
2. For on-chip (bidirectional) bussing:
For encoded state machines, use XC9500/XL.
Use Spartan/XL, XC3000A, XC3100A, XC4000E/X, and
For “one-hot” state machines, use Spartan/XL, XC3100A,
XC5200 families which have horizontal Longlines that can
XC4000E/X or XC5200, families.
be driven by internal 3-state drivers.
3. For fast counters, adders, subtractors, XC9500 devices implement busses indirectly using the
accumulators, comparators: wired-AND capability in the switch matrix.
Use Spartan/XL, XC4000E/X, XC5200, or XC9500/XL fam-
3. For non-volatile single-chip solutions:
ilies for wide functions.
Use XC9500 or any HardWire device.
Use XC3100A for fast, but short or simple counters.
The SRAM-based devices require an external configuration
Spartan/XL, XC4000E/X, and XC5200 devices have dedi-
source, which may be contained in the microprocessor’s
cated carry-logic that is most effective over the range of 8 to
memory. SpartanXL, XC3000A and XC3000L devices can
32 bits. be used with a battery backed up supply, thus eliminating
XC3100A achieves high speed for short word-length and the need for external configuration storage.
simple operations (such as non-loadable counters) through
its fast logic blocks. 4. For lowest possible static power consumption at 5V:
Use XC3000A and, to a lesser extent, Spartan, XC5200,
4. For shortest design compilation time: XC4000E, XC4000EX families.
Use XC9500/XL for small designs, Spartan/XL for medium For Icc down to a few microamps, use XC3000A/L in pow-
sized designs (up to 40,000 system gates). erdown. The other families consume a few milliamps.
XC9500/XL achieves fast compilation through the simplicity Configurations for CMOS input thresholds on all inputs
of its PAL-like architecture. reduce supply current significantly.
5. For lowest cost per gate:
Use Spartan/XL families for designs up to 40,000 system
gates.

R
5. For avoiding pin-locking problems with 11. For PCl compatibility:

routing-intensive designs:
Use Spartan/XL or XC4000E/XL families.
Use XC9500, XC4000EX, XC4000XL, XC5200.
Target and Initiator designs are available for the Spartan,
XC9500 devices have special architectural features to SpartanXL, XC4000E, and XC4000XL families.
enable pin locking.
12. For Hi-Rel, military, or mil temperature-range
XC4000EX, XC4000XL, and XC5200 provide additional applications:
routing channels, called VersaRing, between the core logic
and the I/O. Use XQ4000E/X.
6. For Boundary-Scan support: 13. For battery-operated applications requiring low

stand-by current:
Use Spartan, SpartanXL, XC4000E, XC4000X, XC5200, or
XC9500 families. Use XC3000A/L, Spartan/XL, XC4000E/EX, XC5200 fami-
lies.
7. For rail-to-rail output voltage swing at 5V Vcc:
SpartanXL devices can use powerdown to ignore all input
Use Spartan, XC3000A, XC3100A, XC4000E, XC4000EX, activity and tolerate Vcc down to 2.3V, while maintaining
or XC5200 families. configuration.
Spartan/XL and XC4000E/EX devices can be configured XC3000L devices have inherently very low static power
with a global choice of either totem-pole or rail-to-rail out- consumption.
puts.
XC3000A devices can use powerdown to ignore all input
XC9500 devices have a “totem-pole” output structure with activity and tolerate Vcc down to 2.3V, while maintaining
lower Voh. configuration.
8. For 3.3V operation: Spartan or XC4000E/EX devices must be configured for

CMOS input thresholds, and the user must shut down clock
Use SpartanXL, XC3000L,or XC4000XL families. and logic activities externally.
9. For 5V operation Interfacing with 3.3V devices: 14. For best protection against Illegal copying of a
Use Spartan, XC4000E/EX or XC9500 families. design (design security):
Any Spartan or XC4000E/EX “totem-pole” output drives Use XC9500 with security bit activated.
3.3V inputs safely, and the TTL input threshold can be Use SpartanXL, XC3000A or XC3000L with powerdown
driven from 3.3V logic. battery-back-up configuration.
10. For In-system programmability:
Use all Xilinx families.
14

R
Further Information
For further information on any of the Xilinx products discussed in this application note, see the Xilinx WEBLINX at
http://www.xilinx.com, or call your local sales office.
Table 1: Selecting a Xilinx Family
XC4000XLA
XC4000XL/
SpartanXL
XC4000EX
XC4000XV
XC3000A
XC3100A
XC4000E
XC3000L
XC3100L
Spartan
XC5200
XC9500
Feature
1. Shortest pin-to-pin X X X X X X X
2. Fastest state machines X X X X X X X X X
3. Fastest arithmetic counters X X X X X X X
4. Fastest compilation X
5. Lowest cost X X
6. Footprint compatible families X X X X X
7. DSP (multiply/accumulate) X X X X X X
8. RAM X X X X X X
9. Bidirectional busses X X X X X X X X X X X X
10. Non-volatile/single chip X
11. Low power @ 5V X X X X X
12. Tolerates pin-locking X X X X X
13. Boundary scan X X X X X X X X
14. Full-swing 5V output X X option option X option
15. 3.3V operation X X X X X
16. 5V out drives 3.3V option option option X
17. In-system programmable X X X X X X X X X X X X
18. PCI-compatible X X X X X
19. Hi-rel, mil, mil-temp X X X
20. Low standby current X X X X X X X X
21. Design security X X X

APPLICATION NOTE
0

XAPP088 November 24, 1997 (Version 1.0) 0 14* Application Note by PETER ALFKE and BOB CONN
Summary
Data sheets describe I/O parameters in digital terms, providing tested and guaranteed worst-case values. This application
note describes I/O parameters in analog terms, giving the designer a better understanding of the circuit behavior. Such
parameters are, however, not production-tested and are, therefore, not guaranteed.
Xilinx Families
XC4000XL, XC4000XV, and Spartan-XL
Inputs ant, and the user can ignore all interface precautions, and
need not worry about power sequencing.
Input threshold, the voltage where a 0 changes to a 1 and
vice versa, is stable over temperature, but proportional to Excellent ESD protection (up to several thousand volts) is
VCC: achieved by means of a patented diode-transistor structure
that connects to ground, and not to VCC. The structure
37 to 38% of VCC for the falling threshold, 39 to 42% for the behaves like a Zener diode; it becomes conductive at >6 V
rising threshold. There is 50 mV to 150 mV of hysteresis, and diverts the charge or current directly to ground. It can han-
smallest at hot and high VCC, largest at cold and low VCC. dle current spikes of several hundred milliamps, but continu-
ous current must be kept below 20 mA to avoid reliability
5-V Tolerant Inputs problems caused by on-chip metal migration.
Currently, many systems use a mixture of older 5-V devices
See also the application note “Supply-Voltage Migration, 5 V
and newer 3.3-V devices. This can pose a problem when a
to 3.3 V”, XAPP080, available at www.xilinx.com.
5-V logic High drives a 3.3-V input. See Figure 1.
On most CMOS ICs each signal pin has a clamp diode to VCC, PCI-Compliance
to protect the circuit against electrostatic discharge (ESD). The ‘XL-I/O is designed to be PCI compliant and also to be
This diode starts conducting when the pin is driven more than 5-V tolerant.
0.7 V positive with respect to its VCC. In mixed-voltage sys-
tems, this diode presents a problem since it might conduct • 3.3-V PCI compliance requires a clamping diode to VCC.
tens of milliamps whenever a 5-V logic High is connected to a • 5-V PCI compliance does not explicitly require such a
3.3 V input. diode, but requires passing the specified PCI overshoot
test.
In the XC4000XL/XV and SpartanXL devices, Xilinx has over-
• 5-V tolerance does not permit such a diode.
come this difficulty by eliminating the clamp diode between the
device pins and VCC. The pins can thus be driven as High as To satisfy these conflicting requirements, an internal diode
5.5 V, irrespective of the actual supplyvoltage on the receiving is added to each output, with its cathode connected to an
input. These devices are, therefore, unconditionally 5-V toler- internal VTT rail. See Figure 2.
VCC = 5 V
C C VCC
M 5 V Tolerance M
O O
S
3.8 3.8 VCC = 3.0 ¥ ¥ ¥ 3.6 V T S
Output pull-up
well-bias circuit
Global 14
T 3.5 Output VTT Bus
T L Drive
VOH T VOH Transistors
L
VIH I/O
2.4 2.4 Pad
2.0 VIH 2.0 PCI
In Bipolar Clamp Optional
5-V Device 3.3-V Device 5-V Device
Diode Bond
VIL ESD Wire
1.0 Circuit
0.8 VIL 0.8 C
T M
VOL 0.4 0.4 T O
L S VTT Pin
VTT
X7167 Pad
X7168
Figure 1: Interface Levels Figure 2: Simplified ‘XL-I/O Structure
XAPP088 November 24, 1997 (Version 1.0) 14-13

R
In the PCI-compliant XC4000XLT devices, this rail is inter- Effect of Additional Capacitive Load
nally bonded to eight device pins which externally must be
connected to the appropriate VCC supply (5 V or 3.3 V). Transition Time
In all other ‘XL devices, the VTT rail is internally left uncon- At the specified 50 pF external load, the rise time is 2.4 ns,
nected, thus assuring 5-V tolerance. and the fall time is 2.0 ns. For additional capacitive loads,
add 60 ps/pF to the rise time, and 40 ps/pF to the fall time.
Outputs
Delay
Sink and Source Capability Add 30 ps/pF to the rising-edge delay at 3.0 V.
Add 23 ps/pF to the rising-edge delay at 3.6 V.
The IBIS files describe the strength of the CMOS output
Add 25 ps/pF to the falling-edge delay at any voltage.
drivers as black boxes, giving only voltage/current values
without revealing proprietary circuit details. IBIS gives an The values were derived from XC4028XL measurements
unnecessarily large set of numbers, when most users just using the fast output option, but the slew-rate limited output
want to know the strength of the pull-down transistor (sink option behaves almost identically.
capability) and the pull-up transistor (source capability). These results are consistent with the IBIS-derived output
Close to either rail, the outputs are resistive, i.e. voltage is impedance, since the delay increases with approximately
proportional to current. one RC time constant, and the rise and fall times increase
Table 1 condenses the information and expresses it as out- each with approximately two time constants.
put resistance in Ohm for a sink voltage less than 1 V above These are not guaranteed and tested parameters; they are
ground, and a source voltage less than 1 V below VCC. established by measuring a few devices. Xilinx, therefore,
(Data based on SPICE simulation). suggests that the user add a 20% guardband (multiply by
Table 1: Sink and Source Capability 1.20) when calculating additional delay due to capacitive
load above the guaranteed test limit of 50 pF.
Sink Source
For the same reason, subtract 20% (multiply by 0.80) when
Resistance to Resistance to
calculating the delay reduction due to a capacitive load that
Device Family GND VCC
is less than 50 pF external. See Figure 4.
XC4000E 22.1 - 27.7 53.3 - 90.5 Ohm
XC4000EX 14.4 - 18.8 48.0 - 58.7 Ohm When comparing Xilinx numbers to those from other vendors
who use 35 pF as a standard load, reduce the Xilinx-specified
XC4000XL/XV 14.4 - 20.5 28.0 - 41.0 Ohm
delay by 0.4 ns. Reduce the Xilinx-specified rise time by 1.0
Spartan-XL
ns and the fall time by 0.6 ns, thus changing both to 1.4 ns.
Optional on all 8.0 - 12.0* 20.0 - 30.0* Ohm
XC4000XV* Example:
* This per-pin option will also be available on all XC4000XL For an external lumped capacitive load of 200 pF, the ris-
and Spartan-XL devices later in 1998. ing-edge delay at 3.0 V increases by 1.2 • 150 • 30 = 5.4 ns
over the guaranteed data sheet value.
200 The rising-edge transition time increases by an amount of
180 1.2 • 150 pF • 60 ps/pF = 10.8 ns over the 50-pF transition
160 time of 2.4 ns. The rise time is thus 13.2 ns.
140 3
120
mA
100 2
80
Delta Delay (ns)
60 1
40
20 0
0
1 2 3 4 5 -1
Volts X7166
Figure 3: Output Voltage/Current Characteristics
(default for XC4000XL, XC4000XV and Spartan-XL) -2
0 20 40 60 80 100 120 140
Capacitance (pF) X7169
Figure 4: Additional Delay at Various Capactive Loads
14-14 XAPP088 November 24, 1997 (Version 1.0)

APPLICATION NOTE APPLICATION NOTE
0
XC4000 Series
Technical Information
XAPP045 November 24, 1997 (Version 1.1) 0 14* Application Note
Summary
This Application Note contains additional information that may be of use when designing with XC4000 Series devices. This
information supplements the product descriptions and specifications, and is provided for guidance only.
Xilinx Family
XC4000/XC4000E/XC4000EX/XC4000L
Introduction These curves represent typical devices. Measurements

were taken at nominal VCC, TA = 25°C. These characteris-
This application note describes the electrical characteristics vary by manufacturing lot, and will be affected by future
tics of the output drivers, their static output characteristics changes in minimum device geometries. These character-
or I/V curves, the additional delay caused by capacitive istics are not production-tested as part of the normal device
loading, and the ground bounce created when many out- test procedure; they can, therefore, not be guaranteed.
puts switch simultaneously. Although these measurements show that the output sink
and source capability far exceeds the guaranteed data
Voltage/Current Characteristics of sheet limits, continuous high-current operation beyond the
XC4000-Family Outputs data sheet limits can cause metal migration of the on-chip
metal traces, permanently damaging the device. Output
Figures 1 and 2 show the output source and sink currents,
currents in excess of the data-sheet limits are, therefore,
both drawn as absolute values. Note that the XC4000E/EX
not recommended for continuous operation. These output
families offer a configuration choice between an n-channel
only, totem-pole like output structure that pulls a High output characteristics can, however, be used to calculate or model
to a voltage level that is one threshold drop lower than VCC, output transient behavior, especially when driving transmis-
and a conventional complementary output with a p-channel sion lines or large capacitive loads.
transistor pulling to the positive supply rail. When driving
inputs that have a 1.4-V threshold, the lower VOH of the
totem-pole (“TTL”) output offers faster speed and more sym-
metrical switching delays.
200 200
180 180
160 160
140 140
IOL
120 120
mA 100 mA 100
80 80 14
IOL
60 60
IOH CMOS
40 40
IOH
TTL
20 20
0 0
1 2 3 4 5 1 2 3 4 5
Volts Volts
X5291 X5292
Figure 1: Output Voltage/Current Characteristics for Figure 2: Output Voltage/Current Characteristics for
XC4000E XC4000XL

R
Additional Output Delays When Ground Bounce in XC4000 Devices

Driving Capacitive Load Ground-bounce is a problem with high-speed digital ICs,
Xilinx Product Specifications in chapter 4 give guaranteed when multiple outputs change state simultaneously caus-
worst-case output delays with a 50-pF load. ing undesired transient behavior on an output, or in the
internal logic. This is also referred to as the Simultaneous
The values below are based on actual measurements on a Switching Output (SSO) problem. Ground bounce is prima-
small number of mid-93 production XC4005-5, all in PQ208 rily due to current changes in the combined inductance of
packages, measured at room temperature and VCC = 5.5 V. ground pins, bond wires, and ground metallization. The
Listed is the additional output delay, measured crossing 1.5 IC-internal ground level deviates from the external system
V, relative to the delays specified in this Data Book. ground level for a short duration (a few nanoseconds) after
These parameters are not part of the normal production multiple outputs change state simultaneously. Ground
test flow, and can, therefore, not be guaranteed. bounce affects outputs that are supposed to be stable Low,
and it also affects all inputs since they interpret the incom-
Table 1: Increase in Output Delay When Driving Light
ing level by referencing it to the internal ground. If the
Capacitive Loads (<150 pF)
ground bounce amplitude exceeds the actual instanta-
High-to-Low Low-to-High neous noise margin, then a non-changing input will be
Slew interpreted as a short pulse with a polarity opposite to the
10 50 100 10 50 100 pF ground bounce.
Mode
Slow -1.6 0* 1.4 -1.4 0* 1.4 ns VCC bounce is not as important as ground bounce,
XC4000 because it is of lower magnitude due to the weaker pull-up
Fast -1.6 0* 1.2 -1.2 0* 1.1 ns
transistors. Also, the noise immunity in the High state is
Note: *Zero by definition usually better than in the Low state, and input levels are ref-
erenced to ground, not VCC. All this is the result of our
Table 2: Increase in Output Delay When Driving Heavy industry’s TTL heritage.
Capacitive Loads (>150 pF)
Test Method
High-to-L Low-to-H
Slew Mode
ow igh Data was taken on XC4005-5, devices in the PQ208 pack-
Slow 1.7 1.2 ns/100 pF age, soldered to the Xilinx Ground Bounce Test Board. Pin
XC4000 82, two pins away from the nearest ground pin, was config-
Fast 1.5 1.2 ns/100 pF
ured as a permanently Low output driver, effectively moni-
Example: toring the internal ground level. The simultaneously
switching outputs were on pins 80 and 83, for two outputs
∆T High-to-Low for XC4005-5 with Fast-mode output driv-
switching; additionally, pins 80 and 86 were used for four
ing 250 pF:
outputs switching. The closest ground pins are 79 and 90.
1.2 ns (from Table 1) plus (250-100) pF • 1.5 ns/100 pF
Four ground-bounce parameters were measured at room
= 1.2 ns + 2.25 ns = 3.45 ns
temperature, with Vcc set at 5.5 V as shown in Figure 3.
Total propagation delay, clock to pad:
• VOLP-HLPeak ground noise when switching High-to-Low
TOKPOF + 3.45 ns = 7.0 ns + 3.45 ns = 10.45 ns • VOLV-HLValley ground noise when switching High-to-Low
• VOLP-LHPeak ground noise switching Low-to-High
• VOLV-LHValley ground noise switching Low-to-High
All four parameters can affect system reliability.
VOH VOH
Switching
Outputs
VOL VOL
VOLP-HL
VOLP-LH
Non-Switching
VOL
Active-Low Output
VOLV-HL VOLV-LH
X5299
Figure 3: Ground Bounce

R
The two positive peak values can cause problems with a the slew-rate mode of these outputs. Switching outputs
signal leaving the ground bounce chip, driving another chip. closer to the monitoring output also cause larger peaks and
The positive ground bounce voltage is added to the VOL, valleys than outputs further away.
and may exceed the receiving input’s noise margin. A con-
tinuously logic Low input may thus be interpreted as a Guidelines for Reducing Ground-Bounce
short-duration High pulse. Effects
The two negative valley parameters can cause problems • Minimize the impedance of the system ground
with a signal arriving at the ground-bounce chip, reducing distribution network and its connection to the IC pins.
the Low-level noise immunity. The incoming voltage may PQFPs are best suited, PGAs are worst, and PLCCs are
not be Low enough, and may, therefore, be interpreted as a in-between.
short-duration High input pulse. • Use PC-boards with ground- and VCC-planes, connected
Table 3: Ground Bounce, 16 Outputs Switching, Each directly to the ICs’ supply pins. Place decoupling
With 50 or 150 pF Load, VCC = 5.5 V capacitors very close to these ground and VCC pins.
• Keep the ground plane as undisturbed as possible. A row
Slew High-to-Low Low-to-High of vias can easily cause a dynamic ground-voltage drop.
Load Unit
Rate VOLP VOLV VOLP VOLV • Keep the clock inputs physically away from the outputs
Slow 670 480 240 240 mV that create ground bounce, and connect clocks to input
16 x 50 pF pins that are close to a ground pin. Make sure that all
Fast 1,170 710 480 660 mV
Slow 740 330 210 280 mV clock and asynchronous inputs have ample noise
16 x 150 pF margin, especially in the Low state.
Fast 1,180 420 350 710 mV
• If possible, avoid simultaneous switching by staggering
output delays, e.g. through additional local routing of
Interpretation of the Results
signals or clocks.
Ground bounce is a linear phenomenon. When multiple out- • Spread simultaneously switching outputs around the IC
puts switch, the total ground bounce is the sum of the periphery. For a 16-bit bus, use two outputs each on
ground-bounce values caused by individual outputs switch- either side of four ground pins.
ing. Since the actual switching of multiple outputs is usually
not quite simultaneous, small timing differences between the Ground-Bounce vs Delay Trade-Off
switching outputs, caused by routing delays, can indirectly
After the external sources of ground bounce have been
affect the amplitude. With low capacitive loading, < 50 pF, the
reduced or eliminated. the designer can trade reduced
peaks and valleys might even partially cancel each other.
ground bounce for additional delay by selecting between
With larger capacitive loads, the tendency is for valleys to
families and slew-rate options. Figure 4 shows the trade-off
combine with valleys and peaks to combine with peaks.
for 16 outputs switching simultaneously High-to-Low.
In most devices tested, the load capacitance does not
directly affect the ground-bounce amplitude, but it does 1800
affect the duration of the ground-bounce signals. 1600
Ground-Bounce Voltage (mV)
On the fastest outputs, minimal load capacitance created a 1400 FAST SLEW RATE
ground-bounce resonant frequency of 340 MHz, with a 16 x 50 pF 16 x 150 pF
1200
half-cycle time of 1.5 ns. Such a signal exceeds 90% of its
peak amplitude for about 0.4 ns. 1000 SLOW SLEW RATE
16 x 50 pF 16 x 150 pF
With a 50 pF load on the switching outputs, the ground 800
bounce resonant frequency is 90 MHz, with a half-cycle 600

time of 5 ns, staying 1.7 ns above 90% of peak amplitude.
400
With a 150 pF load on the switching outputs, the ground 14
200
bounce resonant frequency is 40 to 60 MHz, with a
half-cycle time of 8 to 12 ns, staying 3 ns above 90% of 0
2 3 4 5 Additional 6
peak amplitude. Delay (ns)
The main problem with large load capacitances is not an X5981
increase in amplitude, but rather an increase in duration of
the ground-bounce signal. The amplitude is mainly affected Figure 4: Ground-Bounce vs. Delay Trade-off for 16
by the number of outputs switching simultaneously, and by Outputs Switching 50 and 150 pF Each

R
XC4000 and XC4000E Power The following elements are obviously device-size depen-
dent:
Consumption
• One Global Clock driving all CLB flip-flops, but no
Below are the dynamic power consumption values for typi- flip-flop changing:
cal design elements in XC4000 and XC4000E. in XC4005: 4 mW/MTps = 8 mW/MHz
The differences between XC4000 and XC4000E are too in XC4010: 8 mW/MTps = 16 mW/MHz
small to be statistically relevant: in XC4013: 12 mW/MTps = 24 mW/MHz
in XC4020: 16 mW/MTps = 32 mW/MHz
Global clocks in XC4000E are 3% higher, and Longlines
in XC4025: 20 mW/MTps = 40 mW/MHz
and unloaded outputs in XC4000E are 5 to 10% lower than
in XC4000.
• One full-length horizontal or vertical Longline with one
Power consumption is given at nominal 5.0-V supply and driving CLB source and one driven CLB load:
25°C. in XC4005: 0.10 mW/MHz = 0.20 mW/MHz
Power is proportional to the square of the supply voltage, in XC4010: 0.15 mW/MTps = 0.30 mW/MHz
but is almost constant over temperature changes. Power is in XC4013: 0.18 mW/MTps = 0.36 mW/MHz
given as “mW per million transitions per second”, since the in XC4020: 0.20 mW/MTps = 0.40 mW/MHz
more commonly used “MHz” can be ambiguous. When a in XC4025: 0.24 mW/MTps = 0.48 mW/MHz
10-MHz clock toggles a flip-flop, the clock line obviously These numbers do not account for the 10 mA of static
makes 20 MTps, the flip-flop output only 10 MTps. power consumption when all device inputs are configured
The first six elements are device-size independent, i.e. they in TTL mode, which is always the default mode, and in
are applicable to all XC4000 or XC4000E devices operating XC4000 is actually the only user-accessible mode.
at 5-V Vcc. These numbers assume short rise and fall times on all
• One CLB flip-flop driving nothing but a neighboring inputs, avoiding the cross-current when both the n-channel
flip-flop in the same or adjacent CLB (a typical shift pull-down and the p-channel pull-up transistor in the input
register design): buffer might conduct simultaneously.
0.1 mW per million transitions per second = Tutorial Comments:
0.1 mW/MTps
In its pure form, a CMOS output driving a capacitive load
has a power consumption that is independent of drive
• One CLB flip-flop driving its neighbor plus 9 lines of
impedance or rise and fall time. For a full-swing signal, the
interconnect:
power consumed when charging the capacitor is C x V2 x f
0.2 mW per million transitions per second =
where f is the frequency of charge operations. In each
0.2 mW/MTps
charge operation, half the total energy consumed ends up
on the capacitor, and the other half of the energy is dissi-
• One unloaded or unbonded TTL-level output:
pated in the current-limiting resistor or transistor, whatever
0.25 mW per million transitions per second =
its value may be.
0.25 mW/MTps
The subsequent discharge cycle does not take any new
• 50 pF on a TTL-level output: add 0.5 mW/MTps = 1.0 energy from the power supply, but dissipates in the cur-
mW/MHz rent-limiting resistor/transistor all the energy that was for-
merly stored in the capacitor.
• One unloaded or unbonded XC4000E CMOS-level
It is assumed here that the frequency is low enough so that
output:
the capacitors are completely charged and discharged in
0.31 mW per million transitions per second = each half-cycle.
0.31 mW/MTps
• 50 pF on a CMOS-level output: add 0.625 mW/MTps =

1.25 mW/MHz

0
XC3000 Series
Technical Information
XAPP024 November 24, 1997 (Version 1.0) 0 14* Application Note By Peter Alfke and Bernie New
Summary
This Application Note contains additional information that may be of use when designing with the XC3000 series of FPGA
devices. This information supplements the data sheets, and is provided for guidance only.
Xilinx Family
XC3000/XC3000A/XC3000L/XC3100/XC3100A/XC3100L
Contents Configurable Logic Blocks

CLBs The XC3000/XC3100 CLB, shown in Figure 1, contains a
Function Generators combinatorial function generator and two D-type flip-flops.
Flip-flops Two output pins may be driven by either the function gener-
Longline Access ators or the flip-flops. The flip-flop outputs may be routed
IOBs directly back to the function generator inputs without going
Inputs outside of the CLB.
Outputs The function generator consists of two 4-input look-up
Routing
tables that may be used separately or combined into a sin-
Horizontal Longlines
gle function. Figure 2 shows the three available options.
Bus contention
Since the CLB only has five inputs to the function genera-
Vertical Longlines
tor, inputs must be shared between the two look-up tables.
Vertical Longlines
Clock Buffers In the FG mode, the function generator provides any two
Vertical Longlines 4-input functions of A, B and C plus D or E; the choice
Clock Buffers between D and E is made separately for each function. In
Power Dissipation the F mode, all five inputs are combined into a single
Crystal Oscillator 5-input function of A, B, C, D and E. Any 5-input function
CCLK Frequency Stability and Low-time restriction may be emulated. The FGM mode is a superset of the F
Powerdown and Battery-Backup mode, where two 4-input functions of A, B, C and D are
Configuration and Start-Up multiplexed together according to the fifth variable, E.
Reset In all modes, either of the B and C inputs may be selectively
Beware of slow rise-time replaced by QX and QY, the flip-flop outputs. In the FG
mode, this selection is made separately for the two look-up
Introduction tables, extending the functionality to any two functions of
The background information provided in this Application four variables chosen from seven, provided two of the vari-
Note supplements the XC3000, XC3000A, XC3000L, ables are stored in the flip-flops. This is particularly useful
XC3100A and XC3100L data sheets. It covers a wide in state-machine-like applications.
range of topics, including a number of electrical parameters In the F mode, the function generators implement a single
not specified in the data sheets, and unless otherwise function of five variables that may be chosen from seven, as 14
noted, applies to all six families. These additional parame- described above. The selection of QX and QY is con-
ters are sufficiently accurate for most design purposes; strained to be the same for both look-up tables. The FGM
unlike the parameters specified in the data sheets, how- mode differs from the F mode in that QX and QY may be
ever, they are not worst-case values over temperature and selected separately for the two look-up tables, as in the FG
voltage, and are not 100% production tested. They can, mode. This added flexibility permits the emulation of
therefore, not be guaranteed. selected functions that can include all seven possible
inputs.

R
DI
Data In
0
MUX D Q
F 1
DIN
G
QX
RD
QX X
A
F F
B
Logic C Combinatorial
Variables D Function CLB Outputs
E
G G
QY Y
F QY
DIN
G 0
MUX D Q
1
EC
Enable Clock
RD
1 (Enable)
K
Clock
Reset RD
Direct
0 (Inhibit) Figure 1: Configurable Logic

(Global Reset) Block (CLB)
X3217
Function Generator Avoids Glitches

The combinatorial logic in all CLBs is implemented as a A
B
function generator in the form of a multiplexer, built out QX
Any Function
of transfer gates. The logic inputs form the select inputs QY of Up To 4 F
Variables
to this multiplexer, while the configuration bits drive the C
D
data inputs to the multiplexer.
E
The Xilinx circuit designers were very careful to achieve
a balanced design with similar (almost equal) propaga- A
B
tion delays from the various select inputs to the data QX
Any Function
output. QY of Up To 4 G
Variables
The delay from the data inputs to the output is, of C
D
course, immaterial, since the data inputs do not change
E FG
dynamically. They are only affected by configuration. 2a
Mode
This balanced design minimizes the duration of possi-
ble decoding glitches when more than one select input A
B
changes. Note that there can never be a decoding glitch QX F
Any Function
when only one select input changes. Even a non-over- QY of 5 Variables
lapping decoder cannot generate a glitch problem, C G
D
since the node capacitance will retain the previous logic E
F
level until the new transfer gate is activated about a 2b Mode
nanosecond later.
When more than one input changes “simultaneously,” A
B
the user should analyze the logic output for any possi- QX
Any Function
ble intermediate code. If any such code permutation QY of Up To 4
Variables
produces a different result, the user must assume that C
such a glitch might occur and must make the system D F

M
design immune to it. The glitch might be only a few U
X
nanoseconds long, but that is long enough to upset an A G
B
asynchronous design. QX
Any Function
If none of the possible address sequences produces a QY of Up To 4
Variables
different result, the user can be sure that there will be C
no glitch. D
FGM
The designer of synchronous systems generally E 2c
Mode
doesn't worry about such glitches, since synchronous X3218
designs are fundamentally immune to glitches on all Figure 2: CLB Logic Options
signals except clocks or direct SET/RESET inputs.

R
The automatic logic-partitioning software in the XACTstep Input/Output Blocks

development system only uses the FG and F modes. How-
ever, all three modes are available with manual partitioning, The XC3000/XC3100 IOB, shown in Figure 3, includes a
which may be performed in the schematic. If FG or F 3-state output driver that may be driven directly or regis-
modes are required, it is simply a matter of including in the tered. The polarities of both the output data and the 3-state
schematic CLBMAP symbols that define the inputs and control are determined by configuration bits. Each output
outputs of the CLB. buffer may be configured to have either a fast or a slow slew
rate.
The FGM mode is only slightly more complicated. Again, a
CLBMAP must be used, with the signal that multiplexes The IOB input may also be direct or registered. Additionally,
between the two 4-input functions locked onto the E pin. the input flip-flop may be configured as a latch. When an
The CLB will be configured in the FGM mode if the logic is IOB is used exclusively as an input, an optional pull-up
drawn such that the gates forming the multiplexer are resistor is available, the value of which is 40-150 kΩ. This
shown explicitly with no additional logic merged into them. resistor cannot be used when the IOB is configured as an
output or as a bidirectional pin.
The two D-type flip-flops share a common clock, a common
clock enable, and a common asynchronous reset signal. Unused IOBs should be left unconfigured. They default to
An asynchronous preset can be achieved using the asyn- inputs pulled High with the internal resistor.
chronous reset if data is stored in active-low form; the Low
created by reset corresponds to the bit being asserted. The
Inputs
flip-flops cannot be used as latches. All inputs have limited hysteresis, typically in excess of 200
mV for TTL input thresholds and in excess of 100 mV for
If input data to a CLB flip-flop is derived directly from an
CMOS thresholds. Exceptions to this are the PWRDWN
input pad, without an intervening flip-flop, the
pin, and the XTL2 pin when it is configured as the crystal
data-pad-to-clock-pad hold time will typically be non-zero.
oscillator input.
This hold time is equal the delay from the clock pad to the
CLB, but may be reduced according to the 70% rule, Experiments show that the input rise and fall times should
described later in the IOB Input section of this Application not exceed 250 ns. This value was established through a
Note. Under this rule, the hold time is reduced by 70% of worst-case test using internal ring oscillators to drive all I/O
the delay from the data pad to the CLB, excluding the CLB pins except two, thus generating a maximum of on-chip
set-up time. The minimum hold time is zero, even when noise. One of the remaining I/O pins was configured as an
applying the 70% rule results in a negative number. input, and tested for single-edge response; the other I/O
was used as an output to monitor the response.
The CLB pins to which Longlines have direct access are
shown in Table 1. Note that the clock enable pin (EC) and These test conditions are, perhaps, overly demanding,
the TBUF control pin are both driven from to the same ver- although it was assumed that the PC board had negligible
tical Long Line. Consequently, EC cannot easily be used to ground noise and good power-supply decoupling. While
enable a register that must be 3-stated onto a bus. Simi- conservative, the resulting specification is, in most
larly, EC cannot easily be used in a register that uses the instances, easily satisfied.
Reset Direct pin (RD).
IOB input flip-flops are guaranteed to operate correctly
Table 1: Longline to CLB Direct Access without data hold times (with respect to the device
clock-input pad) provided that the dedicated CMOS clock
CLB TBUF
input pad and the GCLK buffer are used. The use of a TTL
Longline A B C D E K EC RD T clock or a different clock pad will result in a data-hold-time
Left Most Vertical X requirement. The length of this hold time is equal to the
(GCLK) delay from the actual clock pad to the GCLK buffer minus
Left Middle Vertical X X X X the delay from the dedicated CMOS clock pad to the GCLK
Right Middle Vertical X X buffer. 14
Right Most Vertical X To ensure that the input flip-flop has a zero hold time, delay
(ACLK)
is incorporated in the D input of the flip-flop, causing it to
Upper Horizontal X
have a relatively long set-up time. However, the set-up time
Lower Horizontal X X specified in the data sheet is with respect to the clock
reaching the IOB. Since there is an unavoidable delay
between the clock pad and the IOB, the
input-pad-to-clock-pad set-up time is actually less than the
data sheet number.

R
Program-Controlled Memory Cells VCC
Out 3-State Output Slew Passive

Invert Invert Select Rate Pull Up
3-State T
(OUTPUT ENABLE)
O D Q Output
Out
Buffer
Flip-
Flop
I/O Pad
R
I
Direct In
Q
Registered In Q D
Flip- TTL or
Flop CMOS
or Input
Latch Threshold
(Global Reset)
OK IK
CK1
CK2
Program
Controlled
Multiplexer
= Programmable Interconnection Point or PIP X3216
Figure 3: Input/Output Block (IOB)
the compensating delay is greater than the delay from

Part of the clock delay can be subtracted from the internal which it is deducted, the resulting delay is zero.
set-up time. Ideally, all of the clock delay could be sub- The 70% rule in no way defines the absolute minimum val-
tracted, but it is possible for the clock delay to be less than ues delays that might be encountered from chip to chip,
its maximum while the internal set-up time is at its maxi- and with temperature and power-supply variations. It sim-
mum value. Consequently, it is recommended that, in a ply indicates the relative variations that might be found
worst-case design, only 70% of the clock delay is sub- within a specific chip over the range of operating condi-
tracted. tions.
The clock delay can only be less than 70% of its maximum Typically, all delays will be less than their maximum, with
if the internal set-up time requirement is also less than its some delays being disproportionately faster than others.
maximum. In this case, the pad-to-pad set-up time actually The 70% rule describes the spread in the scaling factors;
required will be less than that calculated. the delay that decreases the most will be no less than 70%
For example, in the XC3000-125, the input set-up time with of what it would have been if it had scaled in proportion to
respect to the clock reaching the IOB is 16 ns. If the delay the delay that decreased the least. In particular, in a
worst-case design where it is assumed that any delay might
from the clock pad to the IOB is 6 ns, then 70% of this delay,
not have scaled at all, and remains at its maximum value,
4.2 ns, can be subtracted to arrive at a maximum
pad-to-pad set-up time of ~12 ns. other delays will be no less than 70% of their maximum.
The 70% rule must be applied whenever one delay is sub- Outputs
tracted from another. However, it is recommended that
All XC3000/XC3100 FPGA outputs are true CMOS with
delay compensation only be used routinely in connection
n-channel transistors pulling down and p-channel transis-
with input hold times. Delay compensation in asynchronous
tors pulling up. Unloaded, these outputs pull rail-to-rail.
circuits is specifically not recommended. In any case, the
Some additional ac characteristics of the output are listed
compensated delay must not become negative. If 70% of

R
* Fast and Slow refer to the output programming option.

200
IOB latches have active-Low Latch Enables; they are trans-
180 parent when the clock input is Low and are closed when it is
160 High. The latch captures data on what would otherwise be
140 the active clock edge, and is transparent in the half clock
120
period before the active clock edge.
IOL
mA 100
Routing
80
60 Horizontal Longlines
IOH
40 As shown in Table 3, there are two horizontal Longlines
20 (HLLs) per row of CLBs. Each HLL is driven by one TBUF
0 for each column of CLBs, plus an additional TBUF at the
1 2 3 4 5
Volts
X5294
left end of the Longline. This additional TBUF is convenient
for driving IOB data onto the Longline. In general, the rout-
ing resources to the T and I pins of TBUFs are somewhat
Figure 4: Output Current/Voltage Characteristics for
limited.
XC3000, XC3000A, XC3100 and XC3100A Devices
Table 3: Number of Horizontal Longlines
in Table 2. Figure 4 and Figure 5 show output current/volt-
age curves for typical XC3000 and XC3100 devices. Part Rows x Horizontal TBUFs
CLBs
Name Columns Longlines per HLL
Output-short-circuit-current values are given only to indi- XC3020 8x8 64 16 9
cate the capability to charge and discharge capacitive
XC3030 10 x 10 100 20 11
loads. In accordance with common industry practice for
XC3042 12 x 12 144 24 13
other logic devices, only one output at a time may be short
XC3064 16 x 14 224 32 15
circuited, and the duration of this short circuit to VCC or
XC3090 20 x 16 320 40 17
ground may not exceed one second. Xilinx does not recom-
XC3195 22 x 22 484 44 23
mend a continuous output or clamp current in excess of 20
mA on any one output pin. The data sheet guarantees the Optionally, HLLs can be pulled up at either end, or at both
outputs for no more than 4 mA at 320 mV to avoid problems ends. The value of each pull-up resistor is 3-10 kΩ.
when many outputs are sinking current simultaneously.
In addition, HLLs are permanently driven by low-powered
The active-High 3-state control (T) is the same as an latches that are easily overridden by active outputs or
active-Low output enable (OE). In other words, a High on pull-up resistors. These latches maintain the logic levels on
the T-pin of an OBUFZ places the output in a high imped- HLLs that are not pulled up and temporarily are not driven.
ance state, and a Low enables the output. The same nam- The logic level maintained is the last level actively driven
ing convention is used for TBUFs within the FPGA device. onto the line.
I/O Clocks When using 3-state HLLs for multiplexing, the use of fewer
than four TBUFs can waste resources. Multiplexers with
Internally, up to eight distinct I/O clocks can be used, two on four or fewer inputs can be implemented more efficiently
each of the four edges of the die. While the IOB does not using CLBs.
provide programmable clock polarity, the two clock lines
serving an IOB can be used for true and inverted clock, and Internal Bus Contention
the appropriate polarity connected to the IOB. This does,
however, limit all IOBs on that edge of the die to using only XC3000 and XC4000 Series devices have internal 3-state
the two edges of the one clock. bus drivers (TBUFs). As in any other bus design, such bus 14
drivers must be enabled carefully in order to avoid, or at
Table 2: Additional AC Output Characteristics least minimize, bus contention. (Bus contention means that
AC Parameters Fast* Slow* one driver tries to drive the bus High while a second driver
tries to drive it Low).
Unloaded Output Slew Rate 2.8 V/ns 0.5 V/ns
Unloaded Transition Time 1.45 ns 7.9 ns Since the potential overlap of the enable signals is lay-out
dependent, bus contention is the responsibility of the FPGA
Additional rise time for 812 pF 100 ns 100 ns
user. We can only supply the following information:
normalized 0.12 ns/pF 0.12 ns/pF
Additional fall time for 812 pF 50 ns 64 ns While two internal buffers drive conflicting data, they create
a current path of typically 6 mA. This current is tolerable,
normalized 0.06 ns/pF 0.08 ns/pF
but should not last indefinitely, since it exceeds our (conser-

R
vative) current density rules. A continuous contention local interconnect should only be considered for individual
could, after thousands of hours, lead to metal migration flip-flops.
problems.
In a typical system, 10 ns of internal bus contention at 5
Power Dissipation
MHz would just result in a slight increase in Icc. As in most CMOS ICs, almost all FPGA power dissipation
16 bits x 6 mA x 10 ns x 5 MHz x 50% probability = 2.5 mA. is dynamic, and is caused by the charging and discharging
of internal capacitances. Each node in the device dissi-
There is a special use of the 3-state control input: When it is pates power according to the capacitance in the node,
directly driven by the same signal that drives the data input which is fixed for each type of node, and the frequency at
of the buffer, i.e. when D and T are effectively tied together, which the particular node is switching, which can be differ-
the 3-state buffer becomes an “open collector” driver. Multi- ent from the clock frequency. The total dynamic power is
ple drivers of this type can be used to implement the the sum of the power dissipated in the individual nodes.
“wired-AND” function, using resistive pull-up.
While the clock line frequency is easy to specify, it is usually
In this situation there cannot be any contention, since the more difficult to estimate the average frequency of other
3-state control input is designed to be slow in activating and nodes. Two extreme cases are binary counters, where half
fast in deactivating the driver. Connecting D to ground is an the total power is dissipated in the first flip-flop, and shift
obvious alternative, but may be more difficult to route. registers with alternating zeros and ones, where the whole
circuit is exercised at the clocking speed.
Vertical Longlines
A popular assumption is that, on average, each node is exer-
There are four vertical Longlines per routing channel: two cised at 20% of the clock rate; a major EPLD vendor uses a
general purpose, one for the global clock net and one for 16-bit counter as a model, where the effective percentage is
the alternate clock net. only 12%. Undoubtedly, there are extreme cases, where the
ratio is much lower or much higher, but 15 to 20% may be a
Clock Buffers valid approximation for most normal designs. Note that glo-
XC3000/XC3100 devices each contain two high-fan-out, bal clock lines must always be entered with their real, and
low-skew clock-distribution networks. The global-clock net obviously well-known, frequency.
originates from the GCLK buffer in the upper left corner of
Consequently, most power consumption estimates only
the die, while the alternate clock net originates from the
serve as guidelines based on gross approximations.
ACLK buffer in the lower right corner of the die.
Table 4 shows the dynamic power dissipation, in mW per
The global and alternate clock networks each have optional MHz, for different types of XC3000 nodes. While not pre-
fast CMOS inputs, called TCLKIN and BCLKIN, respec- cise, these numbers are sufficiently accurate for the calcu-
tively. Using these inputs provides the fastest path from the lations in which they are used, and may be used for any
PC board to the internal flip-flops and latches. Since the XC3000/XC3100 device. Table 5 shows a sample power
signal bypasses the input buffer, well-defined CMOS levels calculation.
must be guaranteed on these clock pins.
Table 4: Dynamic Power Dissipation
To specify the use of TCLKIN or BCLKIN in a schematic,
connect an IPAD symbol directly to the GCLK or ACLK XC3020 XC3090
symbol. Placing an IBUF between the IPAD and the clock One CLB driving three local inter- 0.25 0.25 mW/MHz
connects
buffer will prevent TCLKIN or BCLKIN from being used.
One device output with a 50 1.25 1.25 mW/MHz
The clock buffer output nets only drive CLB and IOB clock pF load
pins. They do not drive any other CLB inputs. In rare cases One Global Clock Buffer and line 2.00 3.50 mW/MHz
where a clock needs to be connected to a logic input or a One Longline without driver 0.10 0.15 mW/MHz
device output, a signal should be tapped off the clock buffer
input, and routed to the logic input. This is not possible with Table 5: Sample Power Calculation for XC3020
clocks using TCLKIN or BCLKIN. Quantity Node MHz mW/MHz mW
The clock skew created by routing clocks through local 1 Clock Buffer 40 2.00 80
interconnect makes safe designs very difficult to achieve, 5 CLBs 40 0.25 50
and this practice is not recommended. In general, the fewer 10 CLBs 20 0.25 50
clocks that are used, the safer the design. High fan-out 40 CLBs 10 0.25 100
clocks should always use GCLK or ACLK. If more than two 8 Longlines 20 0.10 16
clocks are required, the ACLK net can be segmented into 20 Outputs 20 1.25 500
individual vertical lines that can be driven by PIPs at the top Total Power ~800
and bottom of each column. Clock signals routed through

R
Crystal Oscillator
XC3000 and XC3100 devices contain an on-chip crystal
XTAL_OUT
oscillator circuit that connects to the ACLK buffer. This cir-
cuit, Figure 5, comprises a high-speed, high-gain inverting
amplifier with its input connected to the dedicated XTL2
pin, and its output connected to the XTL1 pin. An external XTAL_IN
biasing resistor, R1, with a value of 0.5 to 1 MΩ is required. FPGA
A crystal, Y1, and additional phase-shifting components,

R1
R2, C1 and C2, complete the circuit. The capacitors, C1
and C2, in series form the load on the crystal. This load is R2
specified by the crystal manufacturer, and is typically 20 pF.
Y1
The capacitors should be approximately equal: 40 pF each
for a 20 pF crystal.
C1 C2 L
Either series- or parallel-resonant crystals may be used, 3RD
since they differ only in their specification. Crystals con- Overtone
Only
strain oscillation to a narrow band of frequencies, the width
of which is <<1% of the oscillating frequency; the exact fre- X6128
quency of oscillation within this band depends on the com-

ponents surrounding the crystal. Series-resonant crystals Figure 5: Crystal Oscillator
are specified by their manufacturers according to the lower
edge of the frequency band, parallel-resonant crystals
according to the upper edge. Table 6: Third-Harmonic Crystal Oscillator Tank-Circuit
The resistor R2 controls the loop gain and its value must be
established by experimentation. If it is too small, the oscilla-
Frequency LC Tank
tion will be distorted; if it is too large, the oscillation will fail R2 (Ω) C1 (pF)
to start, or only start slowly. In most cases, the value of R2 (MHz) L (µH) C2 (pF) Freq (MHz)
is non-critical, and typically is 0 to 1 kΩ. 32 1 60 20.6 430 23
Once the component values have been chosen, it is good 35 1 44 24.0 310 23
practice to test the oscillator with a resistor (~1 kΩ) in series 49 1 31 28.6 190 23
with the crystal. If the oscillator still starts reliably, indepen- 72 1 18 37.5 150 12
dent of whether the power supply turns on quickly or slowly,
it will always work without the resistor. Crystal-Oscillator Considerations
For operation above 20 to 25 MHz, the crystal must be There is nothing Xilinx-specific about the oscillator circuit.
operated at its third harmonic. The capacitor C2 is replaced It’s a wide-band inverting amplifier, as used in all popular
by a parallel-resonant LC tank circuit tuned to ~2/3 of the microcontrollers. When a crystal and some passive compo-
desired frequency, i.e., twice the fundamental frequency of nents close the feedback path, this circuit becomes a reli-
the crystal. Table 6 shows typical component values for the able and stable clock source.
tank circuit. The path from XTAL2 to XTAL1 inside the LCA device is a
Crystal operation below 1 MHz is not supported. Low-fre- single-stage inverting amplifier, which means it has a
quency crystals have a high resonant impedance and low-frequency phase response of 180°, increasing by 45° at
require more gain than provided by the single stage inverter the 3-dB frequency.
in the XC3000 devices. Low-frequency applications are Input impedance is 10-15 pF, input threshold is CMOS, but
usually also more power-conscious and would not accept dc bias must be supplied externally through a megohm
14
the power consumption of the fast general-purpose Xilinx resistor from XTAL1 to XTAL2.
oscillator circuit. Inexpensive complete oscillator packages
are often a better choice. Low-frequency gain is about 10, rolling off 3dB at 125 MHz.
Output impedance is between 50 and 100 Ω and the
capacitance on the output pin is 10 to 15 pF.
Pulse response is a delay of about 1.5 ns and a rise/fall
time of about 1.5 ns.

R
For stable oscillation, circuit equals the gain in the FPGA device, and where the
total phase shift, internal plus external, equals 360°.
• the loop gain must be exactly one, i.e., the internal gain
must be matched by external attenuation, and Figure 7 explains the function. At the frequency of oscilla-
• the phase shift around the loop must be 360° or an tion, the series-resonant circuit is effectively an inductor,
integer multiple thereof. The external network must, and the two capacitors act as a capacitive voltage divider,
therefore, provide 180° of phase shift. with the center-point grounded. This puts a virtual ground
somewhere along the inductor and causes the non-driven
A crystal is a piezoelectric mechanical resonator that can
end of the crystal to be 180° out of phase with the driven
be modeled by a very high-Q series LC circuit with a small
end, which is the external phase shift required for oscilla-
resistor representing the energy loss. In parallel with this
tion. This circuit is commonly known as a Pierce oscillator.
series-resonant circuit is unavoidable parasitic capacitance
inside and outside the crystal package, and usually also
discrete capacitors on the board.
XC2000/XC3000
The impedance as a function of frequency of this whole
array starts as a small capacitor at low frequencies
(Figure 6). As the frequency increases, this capacitive
reactance decreases rapidly, until it reaches zero at the
series resonant frequency.
C L R
Inductive
XTAL
jϖL Series Parallel
Resonance Resonance
X5321
Frequency Figure 7: Pierce Oscillator
1
jϖC Practical Considerations
• The series resonance resistor is a critical parameter. To
Capacitive
assure reliable operation with worst-case crystals, the
user should experiment with a discrete series resistor
C L R roughly equal to the max internal resistance specified
by the crystal vendor. If the circuit tolerates this
X2818
additional loss, it should operate reliably with a
Figure 6: Reactance as a Function of Frequency worst-case crystal without the additional resistor.
• The two capacitors affect the frequency of oscillation
At slightly higher frequencies, the reactance is inductive, and the start-up conditions. The series connection of
starting with a zero at series resonance, and increasing the two capacitors is the effective capacitive load seen
very rapidly with frequency. It reaches infinity when the by the crystal, usually specified by the crystal vendor.
effective inductive impedance of the series LC circuit • The two capacitors also determine the minimum gain
equals the reactance of the parallel capacitor. The parallel required for oscillation. If the capacitors are too small,
resonance frequency is a fraction of a percent above the more gain is needed, and the oscillator may be
series-resonance frequency. unstable. If the capacitors are too large, oscillation is
stable but the required gain may again be higher. There
Over this very narrow frequency range between series and is an optimum capacitor value, where oscillation is
parallel resonance, the crystal impedance is inductive and stable, and the required gain is at a minimum. For most
changes all the way from zero to infinity. The energy loss crystals, this capacitive load is around 20 pF, i.e., each
represented by the series resistor prevents the impedance of the two capacitors should be around 40 pF.
from actually reaching zero and infinity, but it comes very • Crystal dissipation is usually around 1 mW, and thus of
close. no concern. Beware of crystals with “drive-level
Microprocessor- and FPGA-based crystal oscillators all dependence” of the series resistor. They may not start
operate in this narrow frequency band, where the crystal up. Proper drive level can be checked by varying Vcc.
impedance can be any inductive value. The circuit oscil- The frequency should increase slightly with an increase
lates at a frequency where the attenuation in the external in Vcc. A decreasing frequency or unstable amplitude
indicate an over-driven crystal. Excessive swing at the

R
XTAL2 input results in clipping near Vcc and ground. An and fastest Xilinx FPGA is compatible with the oldest and
additional 1 to 2 kΩ series resistor at the XTAL1 output slowest device ever manufactured. The CCLK frequency is
usually cures that distortion problem. It increases the fairly insensitive to changes in VCC, varying only 0.6% for a
amplifier output impedance and assures additional 10% change in VCC. It is, however, very temperature
phase margin, but results in slower start-up. dependent, increasing 40% as the temperature drops from
• Be especially careful when designing an oscillator that 25°C to -30°C, (Table 7)
must operate near the specified max frequency. The Table 7: Typical CCLK Frequency Variation
circuit needs excess gain at small signal amplitudes to
supply enough energy into the crystal for rapid start-up. VCC Temp Frequency
High-frequency gain may be marginal, and start-up may 4.5 V 25°C 687 kHz
be impaired. 5.0 V 25°C 691 kHz
• Keep the whole oscillator circuit physically as compact
5.5 V 25°C 695 kHz
as possible, and provide a single ground connection.
Grounding the crystal can is not mandatory but may 4.5 V -30°C 966 kHz
improve stability. 4.5 V +130°C 457 kHz
CCLK Low-Time Restriction

Series Resonant or Parallel Resonant? When used as an input in Slave Serial and Readback
Crystal manufacturers label some crystals as modes, CCLK does not tolerate a Low time in excess of 5
series-resonant, others as parallel-resonant, but there µs. For very low speed operation, the CCLK High time can
really is no difference between these two types of crys- be stretched to any value, but the Low time must be kept
tals, they all operate in the same way. Every crystal has short. XC4000 and XC5200 devices do not have this
a series resonance, where the impedance of the crystal restriction.
is extremely low, much lower than at any other fre-
quency. At a slightly higher frequency, the crystal is Battery Back-up
inductive and in parallel resonance with the unavoid- Since SRAM-based FPGAs are manufactured using a
able stray capacitance or the deliberate capacitance high-performance low-power CMOS process, they can pre-
between its pins. serve the configuration data stored in the internal static
The only difference between the two types of crystal is memory cells even during a loss of primary power. This is
the manufacturer's choice of specifying either of the two accomplished by forcing the device into a low-power
frequencies. If series resonance is specified, the actual non-operational state, while supplying the minimal current
frequency of oscillation is a little higher than the speci- requirement of VCC from a battery.
fied value. If parallel resonance is specified, the fre- Circuit techniques used in XC3100, XC4000 and XC5200
quency of oscillation is a little lower. In most cases, devices prevent ICC from being reduced to the level need
these small deviations are irrelevant. for battery back-up. Consequently, battery back-up should
only be used for XC2000, XC2000L, XC3000, XC3000A
and XC3000L devices.
CCLK Frequency Variation There are two primary considerations for battery backup
The on-chip R-C oscillator that is brought out as CCLK also which must be accomplished by external circuits.
performs several other internal functions. It generates the
• Control of the Power-Down (PWRDWN) pin
power-on delay, 216 = 65,536 periods for a master,
• Switching between the primary VCC supply and the
214 = 16,384 periods for a slave or peripheral device. It gen-
erates the shift pulses for clearing the configuration array, battery.
using one clock period per frame, and it is the clock source Important considerations include the following.
for several small shift registers acting as low-pass filters for 14
• Insure that PWRDWN is asserted logic Low prior to VCC
a variety of input signals.
falling, is held Low while the primary VCC is absent, and
The nominal frequency of this oscillator is 1 MHz with a returned High after VCC has returned to a normal level.
max deviation of +25% to -10%. The clock frequency,
PWRDWN edges must not rise or fall slowly.
therefore, is between 1.25 MHz and 0.5 MHz. In the
• Insure “glitch-free” switching of the power connections
XC4000 family, the 1-MHz clock is derived from an internal
to the FPGA device from the primary VCC to the battery
8-MHz clock that also can be used as CCLK source.
and back.
Xilinx circuit designers make sure that the internal clock fre- • Insure that, during normal operation, the FPGA VCC is
quency does not get faster as devices are migrated to maintained at an acceptable level, 5.0 V ± 5% (±10% for
smaller geometries and faster processes. Even the newest Industrial and Military).

R
Figure 8 shows a power-down circuit developed by Shel by activating internal bus drivers with conflicting data onto
Epstein of Epstein Associates, Wilmette, IL. Two Schottky the same Longline. These two situations are farfetched, but
diodes power the FPGA from either the 5.2 V primary sup- they are possible and will result in considerable power con-
ply or a 3 V Lithium battery. A Seiko S8054 3-terminal sumption. It is quite easy to simulate these conditions since
power all inputs are stable and the internal logic is entirely combi-
monitor circuit monitors VCC and pulls PWRDWN Low natorial, unless latches have been made out of function
generators.
whenever VCC falls below 4 V.
During powerdown, the Vcc monitoring circuit is disabled. It
is then up to the user to prevent Vcc dips below 2.3 V, which
Seiko S8054 Specifications would corrupt the stored configuration.
Detect Voltage 3.995 V min
4.305 V max During configuration, the PWRDWN pin must be High,
Hysteresis 208 mV typ
VCC Temp. Coeff. 0.52 mV/°C since configuration uses the internal oscillator. Whenever
ICC @ + 6V 2.6 µA typ Vcc goes below 4 V, PWRDWN must already be Low in
IN5817 IN5817
order to prevent automatic reconfiguration at low Vcc. For
B35 the same reason, Vcc must first be restored to 4 V or more,
VCC
Lithium before PWRDWN can be made High.
Battery
2 SEIKO 1 PWRDWN
S 8054
FPGA PWRDWN has no pull-up resistor. A pull-up resistor would
draw supply current when the pin is Low, which would
3
defeat the idea of powerdown, where Icc is only microam-
X5997 peres.
Figure 8: Battery Back-up Circuit Configuration and Start-up

Start-Up
Powerdown Operation
Start-up is the transition from the configuration process to
A Low level on the PWRDWN input, while Vcc remains the intended user operation. This means a change from
higher than 2.3 V, stops all internal activity, thus reducing one clock source to another, and a change from interfacing
Icc to a very low level: parallel or serial configuration data where most outputs are
• All internal pull-ups (on Long lines as well as on the I/O 3-stated, to normal operation with I/O pins active in the
pads) are turned off. user-system. Start-up must make sure that the user-logic
• The crystal oscillator is turned off “wakes up” gracefully, that the outputs become active with-
• All package outputs are three-stated. out causing contention with the configuration signals, and
• All package inputs ignore the actual input level, and that the internal flip-flops are released from the global
present a High to the internal logic. Reset or Set at the right time.
• All internal flip-flops or latches are permanently reset. Figure 9 describes Start-up timing for the XC3000 families
• The internal configuration is retained. in detail.
• When PWRDWN is returned High, after VCC is at its
DONE can be programmed to go High one CCLK period
nominal value, the device returns to operation with the
before or after the I/O become active. Independent of
same sequence of buffer enable and D/P as at the
DONE, the internal global Reset is de-activated one CCLK
completion of configuration.
period before or after the I/O become active.
Things to Remember The default option, and the most practical one, is for DONE
Powerdown retains the configuration, but loses all data to go High first, disconnecting the configuration data source
stored in the device. Powerdown three-states all outputs and avoiding any contention when the I/Os become active
and ignores all inputs. No clock signal will be recognized, one clock later. Reset is then released another clock period
and the crystal oscillator is stopped. All internal flip-flops later to make sure that user-operation starts from stable
and latches are permanently reset and all inputs are inter- internal conditions. This is the most common sequence,
preted as High, but the internal combinatorial logic is fully shown with heavy lines in Figure 9, but the designer can
functional. modify it to meet particular requirements.
Until the chip goes active after configuration, all I/O pins not
Things to Watch Out For
involved in the configuration process remain in a
Make sure that the combination of all inputs High and all high-impedance state with weak pull-up resistors; all inter-
internal flip-flop outputs Low in your design will not gener- nal flip-flops and latches are held reset. Multiple FPGA
ate internal oscillations or create permanent bus contention devices hooked up in a daisy chain will all go active simul-

R
Length Count Match unasserted, but D remains High since the function genera-
CCLK Period
tor acts as an R-S latch; Q stays Low, and RESET is still
pulled High by the external resistor. On the first system
CCLK clock after configuration ends, Q is clocked High, resetting
the latch and enabling the output driver. which forces
F
DONE RESET Low. This resets the whole chip until the Low on Q
permits RESET to be pulled High again.
I/O
The whole chip has thus been reset by a short pulse insti-
Global Reset
gated by the system clock. No further pulses are gener-
ated, since the High on LDC prevents the R-S latch from
X5967 becoming set.
Beware of a Slow-Rising XC3000
taneously on the same CCLK edge. This is well docu-
mented in the data sheets.
Series RESET Input
It is a wide-spread habit to drive asynchronous RESET
Not documented, however, is how the internal combinato-
inputs with a resistor-capacitor network to lengthen the
rial logic comes alive during configuration: As configuration
reset time after power-on. This can also be done with Xilinx
data is shifted in and reaches its destination, it activates the
FPGAs, but the user should question the need, and should
logic and also “looks at” the IOB inputs. Even the crystal
beware of certain avoidable problems.
oscillator starts operating as soon as it receives its configu-
ration data. Since all flip-flops and latches are being held Xilinx FPGAs contain an internal voltage-monitoring circuit,
reset, and all outputs are being held in their high-imped- and start their internal housekeeping operation only after
ance state, there is no danger in this “staggered awaken- VCC has reached ~3.5 V. The internal housekeeping and
ing” of the internal logic. The operation of the logic prior to configuration memory clearing operation then takes
the end of configuration is even useful; it ensures that clock between about 10 and 100 ms, depending on configuration
enables and output enables are correctly defined before mode and processing variations. Any RC delay shorter
the elements they control become active. than 40 ms for a device in master configuration mode, or
shorter than 10 ms for a device in slave configuration mode,
Once configuration is complete, the FPGA device is acti-
is clearly redundant.
vated. This occurs on a rising edge of CCLK, when all out-
puts and clocks that are enabled become active A significantly longer RC delay can be used to hold off con-
simultaneously. Since the activation is triggered by CCLK, it figuration. Without the use of an external Schmitt trigger cir-
is an asynchronous event with respect to the system clock. cuit, the rise time on the RESET input will be very slow, and
To avoid start-up problems caused by this asynchronism, is likely to cross the threshold of ~1.4 V several times, due
some designs might require a reset pulse that is synchro- to external or internal noise. This can cause the FPGA to
nized to the system clock. start configuration, then immediately abort it, then start it
again, after having automatically cleared the configuration
The circuit shown in Figure 10 generates a short Global
memory once more.
Reset pulse in response to the first system clock after the
end of configuration. It uses one CLB and one IOB, and This is no problem for the FPGA, but it requires that the
also precludes the use of the LDC pin as I/O. source of configuration data, especially an XC1700 serial
PROM, be reset accordingly. This is another reason to use
During Configuration, LDC is asserted Low and holds the
the INIT output of the lead FPGA, instead of LDC, to drive
D-input of the flip-flop High, while Q is held Low by the inter-
the RESET input of the XC1700 serial PROMs.
nal reset, and RESET is kept High by internal and external
pull-up resistors. At the end of configuration, the LDC pin is
14
VCC
OE = High
T = Low
D Q
System Clock
Low
CLB MR IOB
RESET
High LDC
IOB
X3222 Figure 10: Synchronous Reset

0
XAPP090 November 24, 1997 (Version 1.1) 0 14* Application Note By Peter Alfke
Summary
These guidelines describe the configuration process for all members of the XC2000, XC3000, XC4000 and XC5200 FPGA
devices and their derivatives. The average user need not understand or remember all these details, but should refer to the
debugging hints when problems occur.
The XC2000-, XC3000-, XC4000- and XC5200-family Each device passes the incoming header, including the
FPGAs share a basic configuration concept, and can be length-count value, on to the DOUT pin, delayed by half a
combined in a common configuration bitstream, but there CCLK period, i.e. the bits are clocked out on a falling CCLK
are also small differences among the four families as edge. In this way, the header is passed on to all devices
described below. that might be connected in a daisy-chain. After the
length-count data has been passed on, DOUT goes active
Following their initial power-on configuration-memory ini-
High and stays High until the device has been filled with the
tialization, these Xilinx FPGAs are configured by a serial
appropriate number of configuration frames. After that,
configuration bitstream. The byte-parallel configuration
DOUT again passes all incoming configuration data on to
modes just activate an internal parallel-to-serial converter,
other devices that might be part of the daisy chain.
and then use the serial bitstream internally. (Express mode
in the XC5200 configures eight bits in parallel, but this DOUT is thus the best observation point to see whether the
mode is not covered in this application note.) The software configuration process has started properly.
generates a bitstream that starts with a 40-bit header Immediately following the header, configuration data is
(48-bit header for XC5200), see Figure 1. received, formatted in a device-specific sequence of
Each device uses a few of the leading “ones” to prepare for frames. Each frame starts with a single “zero” as start bit
configuration, then detects the 0010 pattern and stores the (XC5200 starts with a byte of seven leading “ones” and a
following 24 bits as a length-count value in an internal reg- single trailing “zero”), followed by a device-specific number
ister. The content of this register is continuously compared of configuration bits per frame, followed by three “ones” as
against a running counter that increments on every rising stop bits (XC2000, XC3000) or, in XC4000 and XC5200, by
CCLK edge. CCLK is either an output (in Master and Asyn- four bits that are either 0110, or four bits of a running 16-bit
chronous Peripheral modes) or an input (in Slave Serial CRC error-checking code. The choice is made in the bit-
and Synchronous Peripheral modes). In all modes, even in stream generator, where the default is “CRC disabled”. The
Master Serial, it is the externally observable Low-to-High header is excluded from the CRC calculation.
transition on the CCLK pin that causes the internal action. Each frame is physically shifted into a serial shift register
Every CCLK rising edge that occurs while INIT and RESET that had been preset to all ones. When the zero start bit hits
are High is counted, even during the preamble. Note that
the far end of this shift register, the data frame is transferred
XC2000 and XC3000 use quasi-static circuitry which
in parallel into the configuration memory, as addressed by
imposes a 5 ms max limit on the CCLK Low time, while
the position of an internal token or pointer. The three stop
XC4000 and XC5200 are completely static and have no
or four error-check bits provide ample time for this transfer,
max CCLK time limit. This is, of course, only of interest in
even at a 10 MHz CCLK rate. After this transfer, the shift-in
XC2000 and XC3000 Slave Serial mode, where CCLK is
generated by the user.
procedure continues with the following frame. Note that 14
there is no counter for the number of bits in the frame nor
While it is permissible, although not meaningful, to modify for the number of frames. The operation is self-synchro-
the number of leading ones by adding additional ones, or nized by detecting the presence of a start bit at the far end
subtracting up to four ones, this would inevitably affect the of the shift register, and by moving the frame pointer.
number of CCLK pulses received by the counter, and thus
change the moment when the internal counter is equal to
the value stored in the length-count register. Don’t add or
delete preamble-leading ones!
Figure 1: 40-Bit Header

R
Each Xilinx FPGA requires a number of configuration bits Daisy-Chain Operation

that is device-dependent, but independent of the configura-
tion content, and independent of the configuration mode. Multiple FPGAs can be configured by a single concate-
The number of configuration bits per device ranges from nated bitstream. The device daisy chain is formed by con-
12,038 for the XC2064 to 1,924,992 for the XC4085XL, necting DOUT to the next device’s DIN, and connecting all
approximately 20 bits per available user gate. Exact values CCLK pins in parallel. DOUT goes active on a falling clock
are listed in the specific family data sheets. edge, and DIN accepts data on the subsequent rising clock
edge. Each DOUT-to-DIN connection adds one extra bit of
Protection Against Data or Format delay to the bitstream. Since the header is passed through
all devices, they all receive the same header information
Errors delayed by one bit per device, but all devices maintain per-
The serial configuration scheme has proven reliable in fect synchronism between their CCLK counters, since all
thousands of designs and millions of devices, but there receive the same CCLK.
have been cases where an erroneous bitstream was
Xilinx recognizes the need for all devices in a daisy chain to
loaded accidentally. The original XC2000 and XC3000
finish configuration and begin user operation simulta-
devices provide no effective protection against this type of
neously, as a result of one common CCLK edge. Therefore,
error. If long enough, any random sequence of 0s and 1s
all devices in a daisy-chain need a common timing refer-
will configure such a device. This inevitably takes additional
ence. They cannot rely on the start pattern received
CCLK pulses, more than specified in the length-count
through the pipelined chain, but must all count the common
value. This means that the CCLK counter already matches
CCLK pulses exactly the same way. This explains the
the length-count value before the last FPGA in the chain is
importance of precise configuration clocking, and the dan-
filled. This comparison is, therefore, ignored, and an addi-
ger of reflections and ringing on the CCLK line.
tional ~16 million CCLK pulses are required to roll the 24-bit
length counter and finish the configuration. Such a configu- Start-Up Procedure
ration will, of course, be wrong and might result in exces-
sive power consumption due tol contentions. During configuration, all outputs that are not involved in the
configuration process are 3-stated, although the crystal
XC3000A, XC3100A, XC3000L and XC3100L devices use oscillator circuit is activated as soon as possible. All internal
a simple and effective method to protect against erroneous flip-flops and latches are held reset (set or reset in
configuration files or against loss (or gain) of CCLK pulses: XC4000), and the DONE output is held Low.
All Xilinx FPGA devices recognize a new frame when its At the end of configuration, these three conditions must
leading zero reaches the end of the shift register. XC2000, change: As shown in detail in Figure 2, the various families
XC3000, and XC3100 devices do not check for the pres- offer different options:
ence of valid stop bits, but XC3000A/XC3100A/XC3000L/
XC3100L devices always check whether the three bits at XC2000 has no options; the I/Os go active one CCLK
the end of the defined frame length are 111. If this check period after length-count match. One CCLK period later,
fails, INIT is pulled Low and the internal configuration is DONE goes active and the global reset is released.
stopped, although a master CCLK keeps running. The user XC3000 makes the I/Os go active two CCLK periods after
must recognize this state and start a new configuration by length-count match; but DONE and the release of the glo-
applying a >6 µs Low level on RESET. bal reset can each occur either one CCLK period before or
This simple check does not protect against single-bit ran- after the I/Os go active. The default is "early DONE and late
dom errors, but it offers almost 100% protection against release of the global reset". This makes the outputs go
erroneous configuration files, defective configuration data active while the internal logic is still held reset. The other
sources, synchronization errors between configuration option, "early release of global reset", lets the internal logic
source and FPGA, as well as PC-board defects, such as be clocked out of its reset state before the outputs go
broken lines or solder bridges. active. 14
The XC4000 and XC5200 devices use, optionally, four bits Normally, there is no defined timing relationship between
of a running 16-bit cyclic redundancy check code at the end the last configuration events triggered by the rising edge of
of each frame, combined with additional CRC bits at the CCLK, and the subsequent events that are controlled by
end of the bit stream. These error-detecting CRC codes the system clock. The user must be aware of the potential
provide excellent protection against errors, even those that timing problems of this asynchronous relationship between
do not change the frame structure. When an error is the two clocks. See the XC4000/XC5200 solution
detected, INIT goes Low and stays Low until the user ini- described below.
tiates a reconfiguration. A master device does, however, XC4000 and XC5200 have more options for the relative
continue generating CCLK pulses and even incrementing timing of I/Os, DONE and GSR, the release of the global
or decrementing the parallel PROM address. set or reset.

R
Length Count Match

CCLK Period
CCLK
F
DONE
I/O
XC2000
Global Reset
F
DONE
XC3000
I/O
Global Reset
F
DONE
C1 C2 C3 C4
XC4000/ I/O
XC5200
CCLK_NOSYNC C2 C3 C4
GSR Active
C2 C3 C4
DONE IN
F
DONE
C1, C2 or C3
XC4000/ I/O
XC5200
CCLK_SYNC Di Di+1
GSR Active
Di Di+1
F
DONE
C1 U2 U3 U4
XC4000/ I/O
XC5200
GSR Active
U2 U3 U4
DONE IN
F
DONE
C1 U2
XC4000/ I/O
XC5200
UCLK_SYNC Di Di+1 Di+2
GSR Active
Di Di+1 Di+2
Synchronization
Note: Thick lines are default option F = Finished, no more

Daisy-chain lead device
must have latest F

default timing X5972

R
These families can also use DONE as an input to hold off

the activation of the I/Os and the release of GSR, until
DONE is no longer pulled Low. The change then takes
place either immediately upon the release of DONE, or as a
result of the next CCLK rising edge. When all DONE pins in
a daisy chain are interconnected, this start-up mode guar-
antees that all devices in the chain go active only when all OE/T
Output
of them have reached the DONE state, an additional pro- Connected
Reset to CCLK
tection against potential configuration errors. 0 0
1 0 Active Low Output
XC4000 and XC5200 can also be configured to employ the 1 1 Active High Output
3-Stated Output
0 1
system (user) clock instead of CCLK, again either using 0 1 3-Stated Output
etc
DONE as an output, or as a bidirectional pin. X5552
The user clock provides a properly synchronized and Figure 3: Additional CCLK-Pulse Generator
race-free transition from the end of configuration to the
beginning of user operation. The unspecified on-chip delay
in the release of GSR (about 100 as in XC4013E) requires figured with late internal reset, which happens to be the
some caution, however, when using a high clock frequency default option.
for configuration.
While devices from different families can be arbitrarily inter- Configuration Modes
spersed in a daisy-chain, there is one restriction: the lead There are six different configuration modes, hard-
device must belong to the highest-numbered family in the ware-selected by applying logic levels to the three mode
chain. If the chain contains XC5200 devices, the lead inputs, M0, M1, and M2. The six modes are: Master Serial,
device cannot be XC4000, XC3000 or XC2000; if the chain Master Parallel Up, Master Parallel Down, Synchronous
contains XC4000 devices, the lead device cannot be Peripheral (XC4000 and XC5200 only), Asynchronous
XC3000 or XC2000; if the chain contains XC3000, then the Peripheral, and Slave Serial. A seventh mode, Express
lead device cannot be XC2000. The reason is shown in Mode, is only available in XC5200 devices, and is not
Figure 2. Since all devices in the chain store the same described here.
length-count value and generate or receive one common
In Master modes, the FPGA addresses an external PROM
sequence of CCLK pulses, they all recognize length-count
or EPROM storage device, and reads data from it. No addi-
match on the same CCLK edge. The master device then
tional timing or control signals are used.
generates additional CCLK pulses until it reaches its finish
point F. As shown inFigure 2, the different families generate In Peripheral mode, the FPGA accepts byte-wide data
and require different numbers of additional CCLK pulses (bit-serial in XC2000), and interacts with the source of data,
until they reach F. Not reaching F means that the device usually a microprocessor, with a Ready/Busy handshake.
has not really finished its configuration process, although In Slave mode, the FPGA receives bit-serial data and a
DONE may have gone High, the outputs have become clock from an external data and timing source, either from a
active, and the internal reset has been released. For microprocessor, or from the lead device in an FPGA-daisy
XC4000 and XC5200, not reaching F means that READ- chain.
BACK cannot be initiated, and most boundary scan instruc-
tions cannot be used. The limitation in daisy-chain order The modes are selected by putting the appropriate logic
has been criticized by designers who want to use an inex- levels on the three mode inputs, M0, M1, and M2 prior to
pensive lead device in Peripheral Mode, and save the more the beginning of configuration. These three pins can be
precious XC4000 I/O pins. Here is a solution for that case hardwired to VCC or Ground, but they can then never be
Figure 3): used as user I/O. It is better to force a mode pin Low with a
3 kΩ pull-down resistor to ground, acting against the 20 to 14
One CLB and one IOB in the lead XC3000 device are used 100 kΩ internal pull-up resistor, and to rely on the built-in
to generate the additional CCLK pulse required by the pull-up resistor to establish a High level on the M1, M2
XC4000 devices. When the lead device releases its internal mode pins, but use a 50 kΩ external pull-up resistor on M0.
reset signal, the 2-bit shift register starts responding to its This eliminates the restrictions on using mode pins for user
clock input, and it generates an active Low output signal for logic or readback.
the duration of one clock period. An external connection
between this IOB pin and the CCLK pin thus creates the When mode pin levels are driven by external logic, these
extra CCLK pulse. This solution requires one CLB, one IOB levels must be established very soon after power-up.
and pin, and an internal clock source with a frequency of up Establishing a mode level too late might eliminate the extra
to 5 MHz. Obviously, the XC3000 lead device must be con- master power-on delay that makes a master wait for slave
devices to be ready after power-on. Delaying mode levels

R
until the beginning of configuration will obviously cause the to the Slave devices. If you don’t see this pattern, you
configuration to fail. Note that some CPLD devices have have a gross error somewhere. Check the following
surprisingly long power-up delays. Be very careful when items:
controlling mode levels in any creative way. • INIT going Low again after configuration start indicates
a configuration bitstream or framing error.
Selecting the Best Configuration Mode • If RESET is used to delay configuration, make sure it
The selection of the most appropriate configuration mode is has a rise time of <100 ns and that it is glitch-free.
influenced by many factors, like • Ringing on the CCLK line, caused by pc-board
reflections, can result in spurious double- clocking and
• the need for interface simplicity, loss of frame synchronization in the FPGA.
• the need for rapid configuration, • Configuration functions can be disrupted by signal
• the need for multiple configuration sources, contention between configuration inputs and the FPGA
• the availability of a microprocessor-based configuration user outputs which become active at the end of
driver. configuration. This change is indicated by I/O pins
The simplest interface is Master Serial, using only two going active and HDC/LDC no longer at their
FPGA pins, CCLK and DIN, and no external timing or con- configuration levels. Contention can be avoided by
trol signals. rearranging pin-outs, maintaining additional 3-state
control of user-I/O outputs, or matching start-up output
The fastest configuration mode is Slave Serial or
levels to the configuration input levels on inputs other
XC4000/XC5200 Synchronous Peripheral. In these modes,
than chip-select. As a last resort, it is also possible to
the user can supply a well-defined CCLK frequency of up to
use a series resistor (1-10 kΩ) to provide isolation
10 MHz for 5-Volt devices. Only Express mode can be
between conflicting signal sources that could occur
faster than that. For prototyping and rapid configuration
after configuration is complete.
change, the PC can configure the FPGA directly in Slave
• If an FPGA heats up significantly, this is usually the
Serial mode, using the Xilinx-provided Download Cable or
result of applying the wrong bitstream, e.g. the
XChecker.
bitstream for a different device, causing contention.
Multiple configuration codes are most conveniently stored Legitimate bitstreams have been screened by the
in a microprocessor memory, using Peripheral mode to Design Rule Checker software, and are guaranteed free
configure the FPGA. Peripheral mode also offers the great- of inherent contention problems, provided the
est flexibility for field upgrades. New files can be supplied configuration is loaded into the designated device. The
via diskette or modem, and can be downloaded by the user can obviously still cause contention on internal
microprocessor. Longlines and on connections outside the device.
• During reprogramming, user logic must generate a
When Configuration Fails time-out that insures all devices have completed the
Clear cycle before any configuration data is sent.
General Debugging Hints for all Families • Removing the FPGA supply voltage while externally
If the DONE output does not go High, there are several powered signals continue to drive input pins, might
things to check. keep the FPGA VCC pins at a 0.5-to-2.0 V level, which
can leave the FPGA in an invalid state. The FPGA
• Checking all supply and configuration-related pins with
input-protection diodes are there to clamp input-voltage
an oscilloscope or logic analyzer can reveal wiring
excursions to the two supply connections. When the
errors, bad socket pins, noisy ground, noisy CCLK, a
FPGA supply voltage falls more than 0.5 V below an
serial configuration PROM‘s VPP pin not connected to
active input signal, this input signal will supply
VCC, PWRDWN not pulled High, poor or noisy RESET,
degenerate VCC levels. If the input signals are not
missing pull-up resistors on DONE (or INIT in the
current-limited, the FPGA inputs can even be damaged
XC3000), bad levels on mode pins, etc. Check all pins:
by the excessive input current.
Any dc voltage between 0.5 V and 3.0 V is a sign of
• If extraneous CCLK pulses are applied after Clear but
serious trouble.
before the beginning of the header, they are counted
• Monitor the DOUT pin of the lead device, i.e. the FPGA
internally, and the internal clock count will then become
that is either configured alone, or forms the beginning of
equal to the stored length-count value before the
a daisy chain. At the start of configuration, you should
configuration data is completely loaded. In this case,
see the 40 (or 48)-bit header shown in Figure 1. After
the DONE output does not become active until the clock
this sequence, the DOUT pin remains High until the
counter equals length count a second time. This
device has received all its data. Then, the device
requires 224 extra clocks, about 20 s at the typical rate of
becomes transparent and passes additional data
0.7 MHz, or about 2 seconds at the nominally 8-MHz
(provided there is a daisy chain) through the DOUT pin
fastest CCLK rate. Whenever configuration takes

R
several or many seconds, this is due to a mismatch • If the PROM is dedicated to the FPGA, the CS and OE
between length count and the number of CCLK pulses PROM inputs should be driven from the DONE or LDC
received. FPGA output.
• XChecker or the XACT Download Cable provide an • Verify that the FPGA is sending addresses to the
alternate method of configuration to verify configuration PROM. If it is not, check the FPGA mode pins.
data and to isolate wiring errors, such as interchanged
or inverted configuration data or control signals. M0 = 0, M1 = 0, M2 = 1 for Master Parallel Up
• Try a different device. Although the chips are 100% M0 = 0, M1 = 1, M2 = 1 for Master Parallel Down
factory-tested, an individual device might have been
damaged after the test. Make sure VCC, RESET and PWRDWN are close to
VCC and all ground pins are at 0 V.
General Debugging Hints for the XC2000
and XC3000 Families • Check that the PROM is receiving addresses and is
• An undefined (floating) or active Low PWRDWN during sending out data. If it is not, check that the PROM is
configuration can disturb the operation. A Low level on enabled and has VCC and ground connected, and verify
PWRDWN immediately before the start of configuration that the PROM is programmed with the correct data.
causes problems in XC2000, forces XC3000 into Slave • Check for contention between the PROM address or
mode, but is acceptable in XC3000A and L. data pins and other signals on the board.
• In the XC2000 and XC3000 families, the • Check that the FPGA is addressing the correct memory
configuration-clock input signal drives quasi-static segment. In Master Parallel Up mode, the FPGA starts
circuitry that does not function correctly with a Low time at address 0000 hex and counts up; in Master Parallel
of more than 5 ms. Down mode it starts at address FFFF hex (3FFFF hex
• At power-up, make sure VCC rises in 25 ms or less. If in XC4000) and counts down. If the PROM requires
this cannot be guaranteed, hold RESET active on the different addressing, that must be taken care of by
FPGAs and on the serial PROMs until VCC has reached external hardware.
4.5 V. • Check for ringing and noise on address and data lines.
• A slowly rising or noisy RESET can cause multiple • Make sure the data in the PROM is correct. You can
FPGAs to get out of synchronization. Always debounce check it against the Rawbits file.
reset switches.
Master Serial Mode
General Debugging Hints for the XC4000 • Review the general debugging hints.
• Verify that the FPGA is generating a clock signal on its
and XC5000 Families CCLK pin and that this signal is reaching the CLK pin of
• At power-up, make sure VCC rises in 25 ms or less. If the XC1700-series Serial-Configuration PROM. If it is
this cannot be guaranteed, hold PROGRAM or INIT not, check the mode pins.
active Low on the FPGAs and hold the serial PROMs
reset until VCC has reached 4.5 V. M0 = 0, M1 = 0, M2 = 0 for Master Serial mode
• The boundary scan input pins are active during
configuration, even if boundary scan is not used in the • Verify that the XC1700-series Serial Configuration
design. Toggling TCK, TMS and TDI during PROM is sending data. If it is not, check that power and
configuration might send the device into EXTEST ground are applied to the Serial PROM, and VPP is
mode, which interferes with configuration. Keeping at connected to VCC.
least one of these three inputs continuously High during
configuration avoids this problem. Do Not Let the VPP Pin Float
A floating VPP pin results in temperature-dependent
Additional Mode-Specific Debugging Hints 14
operation, the most notorious cause of unreliable con-
for All Families figuration.
Master Parallel Up and Down Mode • Check that the DATA pin of the Serial PROM is
• Review the general debugging hints. connected to the DIN pin of the FPGA, and that the
• Check that the PROM data pins are connected to the PROM is enabled with CE Low and OE active. Note that
FPGA input pins D0-D7. Check that the PROM address the OE/RESET pin is programmable for either polarity.
pins are connected to the FPGA output pins A0-A15. Check whether this pin is driven from the INIT output.
Verify that all these connections are in the right order. This is the preferred method of guaranteeing SPROM
Monitor the FPGA pins, not the socket pins. Make sure reset.
the socket is good. • Verify that the PROM is programmed with the correct

R
data. data bits. Beware of contention!

• At power-up, make sure VCC rises from 2.0 V to 4.5 V in • Check for contention between the FPGA pins and other
less than 25 ms. If it does not, hold the FPGA RESET signals on the board. Except in XC2000, data is
and the PROM RESET active until VCC reaches 4.5 V. received as eight bits in parallel. Make sure bit 0 is
A typical result of a slow VCC rise time is that the FPGA connected to the D0 pin, bit 1 to D1 pin, etc. (In XC2000
family, data is received serially. If a PROM
sends out CCLK continuously, the CEO pin on the
file is used as a data source, check that data is properly
PROM(s) goes Low, but the DONE pin never goes
serialized LSB first. Data must be LSB first, although
High.
length count is MSB first. This is not intuitively obvious.)
• If you abort configuration by asserting XC3000 RESET
or by pulling XC4000/XC5000 PROGRAM Low, you Slave Serial Mode
must also reset the serial PROM by asserting its • Review the general debugging hints.
RESET. This occurs automatically if the SROM is reset • Check the mode pin levels.
from INIT.
M0 =1, M1 = 1, M2 = 1 for Slave Serial mode
Asynchronous Peripheral Mode
• Review the general debugging hints.
• See schematics in the data sheet for the FPGA family.
• Check the mode pin levels.
• Make sure Vcc, RESET, and PWRDWN are at 5 V, and
ground pins are at 0 V.
M0 = 1, M1 = 0, M2 = 1 for Peripheral mode
• Verify that the FPGA is receiving data on DIN and that it
is receiving a valid clock signal on CCLK. Check the
• Use an external 1 kilohm resistor from READY/BUSY
device sending the data. Check the device sending the
pin to ground. On power-up, before the FPGA has
clock signal, and make sure the clock meets the timing
interrogated the mode lines, this prevents the pin
requirements specified in the product family
from being pulled High by its internal pull-up, which
documentation. Don’t violate the XC3000 and XC2000
would give an early erroneous READY signal.
CCLK Low time specification of 5.0 µs. A CCLK
• Verify that the FPGA is receiving data at its input pin(s)
generated by a Master FPGA automatically meets the
and that it is receiving valid Write-Strobe and
timing requirements.
Chip-Select signals. If not, check the device driving the
• Make sure the FPGA is ready to receive data.
FPGA. Make sure that these signals meet the timing
XC3000 Family: On power up, make sure the INIT pin
requirements listed in the product family
is High or wait at least 34 ms before you begin sending
documentation. XC3000 Family: Check that the
data to the FPGA.
minimum Write-Strobe active time (TCA min = 100 ns) is
XC2000 Family: On power up, make sure that the
met and observe the RDY/BUSY signal. XC2000
FPGA has had time to “wake up” at least 34 ms, before
Family: Be sure maximum and minimum Write-Strobe
sending it data.
active times (TCA max = 5.0 ms, min = 0.25 ms) are
• At power up, make sure VCC rises from 2.0 V to 4.5 V in
met.
less than 25 ms. If it does not, hold RESET Low until
• Make sure that the FPGA is ready to receive data.
the VCC pins reach 4.5 V.
XC3000 Family: On power up, make sure that the INIT
pin has gone High, or wait at least 34 ms before you Daisy Chain Debugging Hints
begin sending data to the FPGA. Make sure that the
RDY/BUSY signal is High before sending each data • The key to debugging daisy-chain configurations is to
byte. XC2000 Family: On power up, make sure that the isolate the problem and attempt to configure a single
FPGA has had time to "wake up," at least 34 ms, before FPGA. Remove all but the first device from the board
sending it data. and configure it. Then insert the second device and
• Check for contention between the Chip Select and configure both. Repeat as you add one device at a time
Write Strobe signals and monitor the levels on those until they all configure.
pins after configuration. It is safest to use the Chip • The first device in the chain can be in any of the
Select pins only as inputs after configuration. Avoid configuration modes. Debug it first, using the hints
contention if they are used as outputs. With XC2000 provided for the appropriate mode.
family devices, the I/Os become active before the FPGA • All devices after the first one are in Slave Serial mode,
receives its final data bits and clocks, and also before so refer to the Slave Serial mode debugging hints
the DONE pin goes High. In other families, this relative above to solve any problems with Slave device.
timing is programmable. If the user function for any of • Monitor the DOUT pin of each device in the chain and
the Chip Selects or the Write Strobe become outputs verify that the 40-bit header (48-bit with XC5200 as the
after configuration, they might contend and, in effect, lead device) appears at the beginning of configuration,
de-select the FPGA so that it never receives its final staggered by one CCLK period per device.

R
• If the Master device in the chain is an XC2000-family Miscellaneous Notes

device and the Slaves are XC3000-family, make sure
the XC3000-family devices are configured with early CCLK is the most important configuration signal. Once the
DONE. INIT output is High, each device counts every Low-to-High
transition of this configuration clock. In all modes except
Potential Length-Count Problem in Parallel Slave Serial and Synchronous Peripheral, CCLK is a very
or Peripheral Modes fast output that cannot be made slew-rate limited. (it is now
slew-rate limited in the newest XC4000X and XC5200
It is highly desirable that the complete change from config- devices). When distributing this clock, the user should pay
uration to user operation occur as the result of one single special attention to glitches, overshoots, and undershoots.
byte-wide input. The activation of outputs and DONE, the In severe cases, a 33 Ω resistor in series with the CCLK
de-activation of the global reset (set/reset in XC4000), and output might improve the signal integrity. In other cases, it
the progression to the “finished” state F (seeFigure 2) might be better to provide a pull-up resistor at the far end of
should all occur as a result of one common byte input. the CCLK net. Since the clock net has a transmission-line
Under normal circumstances, the software achieves this by characteristic impedance of always less than 100 Ω, the
manipulating the length-count value appropriately, taking limited output drive capability of the CCLK output precludes
into account the additional bits between devices, and proper parallel termination.
adjusting for the fact that byte-wide interfaces always leave
DOUT is an excellent observation point, since every device
the last bit sitting in the P-S converter, shifting it out at the
must output the preamble on this pin, irrespective of the
beginning of the next byte. These complexities, combined
with the many possible daisy-chain arrangements have selected configuration mode, and irrespective of the posi-
occasionally led to problems, where the device outputs go tion in, or the existence of, a daisy chain.
active before the last required byte had been received. This INIT of all devices in a daisy chain should be intercon-
has sometimes lead to contention on the address outputs nected to prevent the configuration from starting before all
or data inputs and might prevent the device from going devices are ready. A 10 kΩ pull-up resistor is recom-
DONE, or reaching the real end of its configuration mended. The parallel INIT of the daisy-chained devices
sequence. Not reaching this “finished” state limits the use must be connected to the INIT of the lead XC4000/XC5200
of readback and boundary scan. A new option solves this device, or to the RESET input of the lead XC3000 device.
problem: This is especially important for re-configuration, where the
master does not have a four-times longer wait period.
The default option is “Length-Count aligned” which adjusts
the length-count value such that length-count match occurs The DONE output indicates the end of the configuration
during the first bit in the last configuration byte. This process. In XC2000 and XC3000 systems, it makes sense
assures sufficient CCLK pulses to complete any selected to ground DONE permanently. The RESET input then
type of start-up sequence. The other option is becomes the reconfiguration input, and cannot be used as
“DONE-aligned”, which adjusts the length count value to the dedicated asynchronous user RESET input. LDC can
make DONE go active at the end of a configuration data be used to indicate end of configuration.
byte, which can cause problems in Peripheral mode.
PWRDWN (on XC2000 and XC3000 devices) must be High
Only Peripheral modes seem to be sensitive to the dif- before and during the configuration process.
ference between these two options.
Don’t let PWRDWN float!
14

0
Configuring Mixed FPGA

Daisy Chains
XAPP091 November 24, 1997 (Version 1.0) 0 14* Application Note by Peter Alfke
Overview Length Count Match

CCLK Period
Xilinx FPGAs can be configured in a common daisy-chain

structure, where the lead device generates CCLK pulses CCLK
F
and feeds serial configuration information into the next DONE
downstream device, which in turn feeds data into the next
I/O
downstream device, etc. There is no limit to the number of XC2000
devices in a daisy chain, and XC2000, XC3000, XC4000, Global Reset

and XC5200 series devices can be mixed freely with only
one constraint: the lead device must be a member of the F
DONE
highest-order family used in the chain. (For the purposes of XC3000
this discussion, there is no difference between the XC4000 I/O
series and the XC5200 family, when XC5200 is used in any

Global Reset
configuration mode except Express Mode). The lead
device must generate a sufficient number of CCLK pulses F
after length-count-match was achieved, but XC3000-series DONE
C1 C2 C3 C4
devices generate fewer CCLK pulses than XC4000-series XC4000/ I/O
or XC5200-family devices require, and XC2000 devices XC5200
CCLK_NOSYNC C2 C3 C4
generate even fewer CCLK pulses after length-count
match. See Figure 1. GSR Active
C2 C3 C4
DONE IN
In a daisy-chain, all CCLK pins are interconnected, and F
DOUT of any upstream device feeds the DIN input of its DONE
downstream neighbor. Those are the basic connections. XC4000/ C1, C2 or C3

I/O
For control purposes, it is advisable to interconnect all the XC5200
CCLK_SYNC Di Di+1
slave INIT pins (the XC2000 does not have this pin) and
connect them to the INIT pin of the lead XC4000/XC5200 GSR Active
Di Di+1
device or the RESET input of the lead XC3000 device.
F
Interconnected INIT pins prevent the master from starting DONE
the configuration process until all slaves are ready. For C1 U2 U3 U4

XC4000/ I/O
power-up this is assured automatically, since the master XC5200
uses four times as many internal clocks for the power-up as
any slave does, but, when re-configuring, master and slave GSR Active
devices consume the same number of clocks to clear a U2 U3 U4

DONE IN
frame, and a fast master might be ready before a slow slave F
is. Interconnecting INITs solves this problem. DONE
C1 U2
The DONE/PROG (D/P) and RESET pins (XC2000, XC4000/ I/O
XC5200
XC3000) and the XC4000/XC5200 PROGRAM pins can be UCLK_SYNC Di Di+1 Di+2
used in different ways, depending on the designer’s prefer- 14
GSR Active
ences regarding reconfiguration, pin utilization, and need
Di Di+1 Di+2
for a global RESET input. Synchronization
If there is no need for a global logic RESET input, then it is Note: Thick lines are default option F = Finished, no more
best to permanently ground the XC2000/3000 D/P pin, Daisy-chain lead device
must have latest F
which means that the RESET input functions as the Recon-
figure input, and should be connected to all default timing X5972
XC4000/XC5200 PROGRAM inputs.


R
Configuring Mixed FPGA Daisy Chains
put) and that, if Serial mode is chosen for the lead device,
VCC the XC1700 device(s) store only one configuration for the
whole daisy chain. The serial PROM(s) must, therefore, be
5K reset before the daisy chain is to be (re)programmed.
REPROGRAM To All D/P
Wired Together
There are three possible types of daisy chains using
GLOBAL To All RESET, Except XC3000 and XC4000/XC5200 devices. Here are the rec-
Lead Device
RESET VCC ommended connections for the configuration control pins.
5K
From All INIT To RESET of Case 1:
Pins Wired Lead Device Daisy chain consists of nothing but XC3000-series
X5982
Together
devices:
Figure 2: Sir Figure Use lead device’s LDC to drive XC1700 CE.
If there is a need for a global logic RESET input that can Use lead device’s INIT to drive XC1700 RESET.
reset all flip-flops in the user logic without causing reconfig- Interconnect all slave INITs and connect them to the lead
uration, then external logic must combine RESET and D/P RESET input.
in such a way, that pulling Low RESET does not affect D/P,
Interconnect all DONE pins.
but pulling Low D/P also pulls down RESET. See Figure 2.
Interconnect all slave RESET inputs
The following simple recommendations guarantee a
well-defined beginning for any FPGA configuration or Instigate Reprogram by pulling the slave RESET net Low
reconfiguration process, after the initialization and clearing for at least 6 µs while all DONE pins are Low.
of the configuration memory in all FPGAs has been com- (DONE can be permanently wired Low, but that sacrifices
pleted, and the address counter in the serial PROM(s) has the use of RESET as a global reset of the user logic. If
been reset. DONE is not wired Low, reprogram must pull DONE Low
The connections described below guarantee reliable oper- with an open-collector or open-drain driver).
ation even under adverse operating conditions such as VCC Case 2:
glitches. Lead device is XC4000-series or XC5200 family, driving
The lead device can use any configuration mode available. any mixture of XC3000, XC4000 and XC5200 devices:
In all modes except Slave Serial, its CCLK pin is the output Use lead device’s LDC to drive XC1700 CE.
that clocks all other devices.
Use lead device’s INIT to drive XC1700 RESET.
Obviously, all CCLK and XC1700 CLK pins must be inter-
connected, the DATA outputs from multiple XC1700 serial Interconnect all INIT pins.
PROMs must be interconnected and connected to the DIN Interconnect all DONE pins.
input of the lead device, and the daisy-chain must be estab-
Interconnect all XC4000/XC5200 PROGRAM inputs.
lished by connecting each DOUT output to the downstream
DIN input. Interconnect all XC3000 RESET inputs.
Configuration control pins are: Combine these two nets into one PROGRAM/RESET net
XC3000A, XC3000L, XC3100, XC3100A: Instigate Reprogram by pulling the combined PRO-
GRAM/RESET Low.
DONE/PROGRAM (open-drain output/input)
RESET (input) Case 3:
INIT (open-drain output) Daisy chain consists of nothing but XC4000/ and
XC5200-type devices:
XC4000 Series (XC4000E, XC4000X) and XC5200 fam-
ily: Use lead device’s LDC to drive XC1700 CE. 14
DONE (open-drain output / input) Use lead device’s INIT to drive XC1700 RESET.
PROGRAM (input) Interconnect all INIT pins.
INIT (open-drain output / input)
Interconnect all DONE pins (only required for UCLK-SYNC
XC1700: option).
RESET (input with programmable polarity) Interconnect all XC4000/XC5200 PROGRAM inputs.
The following recommendations assume that there are no Instigate Reprogram by pulling PROGRAM Low.
XC2000 devices in the daisy chain (they lack the INIT out-

0
Configuration Issues:
Power-up, Volatility, Security,
Battery Back-up
Summary
This application note covers several related subjects: How does a Xilinx FPGA power up, and how does it react to
power-supply glitches? Is there any danger of picking up erroneous data and configuration? What can be done to maintain
configuration during loss of primary power? What can be done to secure a design against illegal reverse-engineering?
Xilinx Families
XC2000, XC3000, XC4000, XC5200
Power-Up is proportional to Vcc), and a reduction in output drive. If Vcc

drops into the 3-V range, it triggers a sensor that forces the
Here is a detailed description of XC3000 Series, XC4000
Series and XC5200 device behavior during supply ramp-up device back to the pre-initialization mode described above.
and ramp-down. All flip-flops are reset, HDC goes High; INIT, LDC and D/P
or DONE go Low, and all other outputs are 3-stated with a
When Vcc is first applied and is still below about 3 V, the weak resistive pull-up. If VCC dips substantially lower, the
device wakes up in the pre-initialization mode. HDC is High; active outputs become weaker, but the device stays in this
INIT, LDC and DONE or DONE/PROG (D/P) are Low, and preinitialization mode. When VCC rises again, a normal
all other outputs are 3-stated with a weak pull-up resistor. configuration process is initiated, as described above.
When VCC has risen to a value above ~3 V, and a 1 and a
0 have been successfully written into two special cells in Sensitivity to VCC Glitches
the configuration memory, the initialization power-on time The user need not be concerned about power supply dips:
delay is started. This delay compensates for differences in The XC3000/XC4000/XC5200 devices stay configured for
VCC detect threshold and internal CCLK oscillator fre- small dips and they are “smart enough” to reconfigure
quency between different devices in a daisy chain. The ini- themselves (if a master) or to ask for reconfiguration by
tialization delay counts clock periods of an on-chip pulling INIT and D/P or DONE Low (if a slave). The devices
oscillator (CCLK) which has a 3:1 frequency uncertainty will not lock up; the user can initiate re-configuration at any
depending on processing, voltage and temperature. time just by pulling D/P or PROGRAM Low or, if D/P is Low,
Time-out, therefore, takes between 11 and 33 ms for a by forcing a High-to-Low transition on RESET.
slave device, four times longer for a master device.
Any digital logic device with internal data storage in latches
This factor of four makes sure that even the fastest master or flip-flops is sensitive to power glitches. This includes
will always take longer than any slave. We assume that the every RAM, microprocessor, microcontroller, and periph-
worst- case difference between 33 ms and 4 x 11 ms is eral circuit. Only purely combinatorial circuits can be guar-
enough to compensate for the VCC rise time spent between anteed to survive a severe power glitch without any
threshold differences (max 2 V) of devices in a daisy chain. problem.
Only in cases of very slow VCC rise time (>25 ms), must the
user hold RESET Low until VCC has reached a proper level. Xilinx SRAM-based FPGAs store their configuration in
Interconnecting the INIT pins of all devices in a daisy-chain latches that lose their data when the supply voltage drops 14
is a better method of synchronizing start-up, but cannot be below a critical value (which is substantially below 3 V for
used with XC2000 devices, since they lack an INIT pin. the 5-V devices), but configuration data is extremely robust
and reliable while VCC stays above 3 V. All Xilinx configura-
After the end of the initialization time-out, each device tion latches are implemented as cross-coupled comple-
clears its configuration memory in a fraction of a millisec- mentary inverters with active pull-down n-channel
ond, then tests for inactive RESET or PROGRAM, stores transistors and active pull-up p-channel transistors. Both
the MODE value and starts the configuration process, as High and Low logic levels have an impedance of less than
described in the Data Sheet. After the device is configured, 5kΩ with respect to their respective supply rail.
the 5-V VCC may dip to about 3.5 V without any significant
consequences beyond an increase in delays (circuit speed Typical SRAM memory devices use passive poly-silicon
pull-up resistors with an impedance of about 5,000 MΩ. A

R
current of one nanoamp (!) is sufficient to upset the typical tor or other control register due to an undetected power
SRAM cell, whereas it takes a million times more current to glitch, with disastrous consequences to the subsequent
upset the Xilinx configuration latch. operation. A Xilinx FPGA detects the power glitch and
always plays it safe by flagging the problem.
This does not mean that SRAMs are unreliable, it just
shows that the levels in Xilinx configuration latches are six No complex system of any kind can function reliably when
orders of magnitude more resistant to upsets caused by Vcc is unreliable. Xilinx FPGAs do the safest thing possible,
external events, like cosmic rays or alpha particles. Xilinx whenever such problems occur.
has never heard about any occurrence of a spontaneous
change in the configuration store in any of its ~50 million Design Security
FPGA devices sold over the past twelve years.
Some Xilinx customers are concerned about the security of
Whereas most digital circuits rely on Vcc staying within their designs. How can they protect their designs against
specification, Xilinx FPGAs have an internal voltage moni- unauthorized copying or reverse-engineering?
toring circuit. For example, in the 5-Volt devices, whenever
We must distinguish between two very different situations:
the supply voltage dips below 3 V, the internal monitoring
circuit causes the Xilinx FPGA to stop normal operation. All • Configuration data in accessible from a serial or parallel
outputs go 3-state, and the device waits for the supply volt- EPROM or in a microprocessor's memory. This is the
age to rise closer to 4 V, when it either demands (slave or normal case.
peripheral mode) or initiates (master mode) a reconfigura- • Configuration data is hidden from the user, since the
tion. In the range between 5.5 and 3 V, all typical CMOS design does not permanently store a source of
devices maintain their functionality and their data storage, configuration data. After the FPGA was configured, the
they just get slower as the voltage goes down. EPROM or other source was removed from the system,
and configuration is kept alive in the FPGA through
Xilinx has made sure that the FPGA cannot be corrupted by
battery-back-up.
a power glitch. The most sensitive circuit is the low-voltage
detector. It kicks in while all other configuration storage and Design Security when Configuration Data is
user logic is still guaranteed to be functional. The volt-
Accessible
age-monitoring feature in the Xilinx device can even be
used to protect other circuitry, or it can be coordinated with In the first case, it is obviously very easy to make an identi-
external monitoring circuits. cal replica of the design by copying the configuration data
and the pc-board interconnect pattern of the standard
There is no possibility of a VCC dip causing the device to
devices, but it is virtually impossible to interpret the bit-
malfunction, i.e., to operate with erroneous configuration
stream in order to understand the design or make intelli-
information.
gent modifications to it. Xilinx keeps the interpretation of
• If VCC stays above the trip point, the device functions the bitstream a closely guarded secret. Reverse-engineer-
normally, albeit at reduced speed, like any other CMOS ing an FPGA would require an enormously tedious analysis
device. of each individual configuration bit, which would still only
• If VCC dips below the trip point, the device 3-states all generate an XACT view of the FPGA, not a usable sche-
outputs and waits for reconfiguration. matic.
Xilinx production-tests the VCC-dip tolerance of all XC3000 The best protection against a mindless copy is legal. The
devices in the following way. bitstream is easily protected by copyright laws that have
proven to be more successfully enforced than the intellec-
After the device is configured, VCC is reduced to 3.5 V, and
tual property rights of circuit designs.
then raised back to 5.0 V. Configuration data is then read
back and compared against the original configuration bit The combination of copyright protection, and the almost
stream. Any discrepancy results in rejection of the device. insurmountable difficulty of creating any design variation for
the intended function, provides good design security. The 14
Subsequently, VCC is reduced to 1.5 V and then raised to
recent successes of small companies in reverse-engineer-
5.0 V. The device must first go 3-state, then respond with a
ing microprocessors and microprocessor support circuits
request for reconfiguration.
show that a non-programmable device can actually be
Both these tests are performed at high temperature (>85°C more vulnerable than an FPGA. For advice on legal protec-
for commercial parts, >100°C for military). Any part failing tion of the configuration bitstream, see the following para-
any of these tests is rejected as a functional failure. graphs.
As a result of these careful precautions, we contend that
Xilinx FPGAs are safer than all other types of circuitry
(except purely combinatorial circuits). A microprocessor
can loose the content of its address register, its accumula-

R
Design Security by Hiding the

Legal Protection of Configuration Bit-Stream Programs Configuration Data
The bit-stream program loaded into the FPGA may If the design does not contain the source of configuration
qualify as a “computer program” as defined in Section data, but relies on battery-back-up of the FPGA configura-
101, Title 17 of the United States Code, and as such tion, then there is no conceivable way of copying this
may be protectable under the copyright law. It may also design. Opening up the package and probing thousands of
be protectable as a trade secret if it is identified as latches in undocumented positions to read out their data
such. We suggest that a user wishing to claim copy- without ever disturbing the configuration is impossible.
right and/or trade secret protection in the bit stream
This mode of operation offers the ultimate design security.
program consider taking the following steps.
It is being used by several Xilinx customers who have rea-
Place an appropriate copyright notice on the FPGA
son to be concerned about illegal pirating of their designs.
device or adjacent to it on the PC board to give notice
to third parties of the copyright. For example, because
of space limitations, this notice on the FPGA device
Battery Back-up and Powerdown
could read “©1996 XYZ Company” or, if on the PC Since SRAM-based FPGAs are manufactured using a
board, could read “Bit Stream ©)1996 XYZ Company”. high-performance low-power CMOS process, they can pre-
serve the configuration data stored in the internal static
File an application to register the copyright claim for the memory cells even during a loss of primary power. This is
bit-stream program with the U.S. Copyright Office. accomplished by forcing the device into a low-power
non-operational state, while supplying the minimal current
If practicable, given the size of the PC board, notice requirement of VCC from a battery.
should also be given that the user is claiming that the Circuit techniques used in XC3100, XC4000 and XC5200
bitstream program is the user's trade secret. A state- devices prevent ICC from being reduced to the level needed
ment could be added to the PC board such as: “Bit- for battery back-up. Consequently, battery back-up should
stream proprietary to XYZ Company. Copying or other only be used for XC2000, XC2000L, XC3000, XC3000A
use of the bitstream program except as expressly au- and XC3000L devices.
thorized by XYZ Company is prohibited.”
There are two primary considerations for battery backup
To the extent that documentation, data books, or other which must be accomplished by external circuits.
literature accompanies the FPGA-based design, appro- • Control of the Power-Down (PWRDWN) pin
priate wording should be added to this literature provid- • Switching between the primary VCC supply and the
ing third parties with notice of the user's claim of battery.
copyright and trade secret in the bit-stream program.
For example, this notice could read: “Bit-Stream©)1996 Important considerations include the following.
XYZ Company. All rights reserved. The bit-stream pro- • Insure that PWRDWN is asserted logic Low prior to VCC
gram is proprietary to XYZ Company and copying or falling, is held Low while the primary VCC is absent, and
other use of the bitstream program except as express- returned High after VCC has returned to a normal level.
ly authorized by XYZ Company is expressly prohibited.”
PWRDWN edges must not rise or fall slowly.
• Insure “glitch-free” switching of the power connections
To help prove unauthorized copying by a third party, ad-
to the FPGA device from the primary VCC to the battery
ditional nonfunctional code should be included at the
end of the bit-stream program. Therefore, should a third and back.
party copy the bit-stream program without proper au- • Insure that, during normal operation, the FPGA VCC is
thorization, if the non-functional code is present in the maintained at an acceptable level, 5.0 V ± 5% (±10% for
copy, the copier cannot claim that the bit-stream pro- Industrial and Military).
gram was independently developed. Figure 1 shows a power-down circuit developed by Shel
These are only suggestions, and Xilinx makes no rep- Epstein of Epstein Associates, Wilmette, IL. Two Schottky
resentations or warranties with respect to the legal diodes power the FPGA from either the 5.2 V primary sup-
effect or consequences of the above suggestions. ply or a 3 V Lithium battery. A Seiko S8054 3-terminal
Each user is advised to consult legal counsel with power monitor circuit monitors VCC and pulls PWRDWN
respect to seeking protection of a bit-stream program Low whenever VCC falls below 4 V.
and to determine the applicability of these suggestions
to the specific circumstances.
If the user has any questions, contact the Xilinx legal
department at 408-879-4984.

R
Things to Remember:
Seiko S8054 Specifications Powerdown retains the configuration, but loses all data
Detect Voltage 3.995 V min stored in the device. Powerdown three-states all outputs
4.305 V max
Hysteresis 208 mV typ and ignores all inputs. No clock signal will be recognized,
VCC Temp. Coeff. 0.52 mV/°C and the crystal oscillator is stopped. All internal flip- flops
ICC @ + 6V 2.6 µA typ
IN5817 IN5817 and latches are permanently reset and all inputs are inter-
preted as High, but the internal combinatorial logic is fully
B35 functional.
Lithium
VCC Battery
2 SEIKO 1 PWRDWN FPGA
Things to Watch Out for:
S 8054
Make sure that the combination of all inputs High and all
3
internal flip-flop outputs Low in your design will not gener-
X5997 ate internal oscillations or create permanent bus contention
by activating internal bus drivers with conflicting data onto
Figure 1: Battery Back-up Circuit the same long line. These two situations are farfetched, but
they are possible and will result in considerable power con-
Powerdown Operation sumption. It is quite easy to simulate these conditions since
all inputs are stable and the internal logic is entirely combi-
A Low level on the PWRDWN input, while VCC remains natorial, unless latches have been made out of function
higher than 2.3 V, stops all internal activity, thus reducing generators.
ICC to a very low level:
During powerdown, the VCC monitoring circuit is disabled. It
• All internal pull-ups (on Long lines as well as on the I/O is then up to the user to prevent VCC dips below 2.3 V,
pads) are turned off. which might corrupt the stored configuration.
• The crystal oscillator is turned off
During configuration, the PWRDWN pin must be High,
• All package outputs are three-stated.
since configuration uses the internal oscillator. Whenever
• All package inputs ignore the actual input level, and
VCC goes below 4 V, PWRDWN must already be Low in
present a High to the internal logic.
order to prevent automatic reconfiguration at low VCC. For
• All internal flip-flops or latches are permanently reset.
the same reason, VCC must first be restored to 4 V or more,
• The internal configuration is retained.
before PWRDWN can be made High.
• When PWRDWN is returned High, after VCC is at its
nominal value, the device returns to operation with the PWRDWN has no pull-up resistor. A pull-up resistor would
same sequence of buffer enable and D/P as at the draw supply current when the pin is Low, which would
completion of configuration. defeat the idea of powerdown, where ICC is only microam-
peres.
14

0
Introduction Important Considerations

All Xilinx SRAM-based FPGAs can be in-system config- • Reconfiguration is “all or nothing”. There is no way to
ured and re-configured an unlimited number of times. The restrict reconfiguration to a part of the chip (Note that
XC6200 family has additional features that allow partial and XC6200 devices do not have this limitation).
very fast (re-)configuration from a microprocessor bus. See • Reconfiguration takes a specific time, determined only
the XC62000 product documentation for details. by device type, size and clock speed, independent of
the particular configuration pattern. Configuration takes
This application note describes the procedures for recon-
figuring the more traditional Xilinx FPGAs of the XC3000, from tens to hundreds of milliseconds. During that time,
all user-outputs of the device, or the whole daisy-chain
XC4000, and XC5200 families.
of devices, are 3-stated with weak internal pull-ups,
All configuration information is stored in latches that are except for HDC and LDC, which are active High or Low
loaded serially, conceptually like a shift register. There are respectively.
several different bit-serial or byte-parallel configuration data • All user-data stored in registers, flip-flops or latches is
interfaces, selected by logic levels on three mode inputs, erased. There is no way to retain data inside the device
but – with the exception of the XC5200 Express mode – from one configuration to the next.
they all result in the bit-serial loading of the configuration
These limitations are absolute. If they are not acceptable,
latches. The byte-parallel interfaces in Master Parallel and
the user must resort to creative solutions, like piggy-back-
Peripheral modes act just as an 8-bit parallel-to-serial con-
ing multiple devices.
verter. Between devices in a daisy-chain, the configuration
information is transmitted bit-serially with a common Con- The designer of reconfigurable applications should be
figuration Clock (CCLK). In Master and Peripheral modes, familiar with the normal configuration process of each
CCLK is generated by the lead FPGA device, in Slave device, as described in the individual product descriptions.
Serial mode, CCLK comes from an external source. There is also pertinent information about daisy-chain oper-
ation, especially about mixed daisy chains, in other applica-
Reconfiguration of an operational device, or a daisy-chain
tion notes.
of devices, goes through the following sequence of events:
Interconnecting the INIT pins of all devices in a daisy-chain
• Reconfiguration is initiated by pulling a specific device
is mandatory for reconfiguration, since this is the only way
pin Low.
to guarantee that the master device does wait for the rest of
• First, all outputs are 3-stated, except HDC = High, LDC
and DONE = Low the daisy-chain to be cleared, before starting the reconfigu-
ration. Only the first configuration after power-up makes the
• Then, all internal registers, flip-flops and latches, as
master device spend four times as many clock periods as
well as the configuration storage latches are cleared.
During this time, the INIT output is being pulled Low. any slave during the initial clear operation, so that the mas-
• Then, the Mode inputs and RESET or PROGRAM ter cannot possibly get ahead of the slaves. Reconfigura-
inputs are sampled to determine the selected tion, however, does not slow down the master this way, so
configuration mode and whether to start the new the interconnection of all INIT pins must serve that same
purpose.
configuration process, or to wait.
• Then configuration data is accepted and loaded into the In Master Serial mode, it is highly recommended that the
internal latches and distributed through the daisy-chain. active Low level of INIT be used to reset the XC1700-family
14
• When all configuration information has been entered, Serial PROM.
the user outputs are activated, DONE goes High and
the internal reset is released, all in the order specified in
the configuration bitstream. All devices in a daisy-chain
perform each of these operations in synchronism.

R
Reconfiguration Time This is the simplest scheme, but it precludes the use of
RESET to clear the flip-flops and latches in the operating
Reconfiguration time is usually more critical than the origi- user-design. RESET must be pulled Low for more than six
nal power-on configuration time, which is often masked by microseconds to overcome its internal low-pass filtering.
the general power-on delays. Configuration starts when RESET has gone High again.
Here are some suggestions to reduce reconfiguration time. 2. Pull DONE Low with an open-drain (“open-collector”)
• A daisy-chain is obviously not conducive to fast output. This assumes that DONE was High, i.e. that the
configuration, it should be broken up into shorter blocks, previous configuration was successful. Reconfiguration
perhaps single devices. Multiple devices can be starts as soon as the internal memory has been cleared.
configured in parallel, but can still use a common DONE can be released anytime.
CCLK, and can also be made to start up together. If the 3. Pull DONE Low with an open-drain (“open-collector”)
devices differ in size or family, they should all be given output and pull RESET Low. Keep RESET Low for at least
the same length count as the largest device in the six microseconds while DONE is Low. DONE can be
group. released anytime after that, or not released at all. See alter-
• Configuration Mode native 1.
Parallel and Peripheral modes are not any faster than
Master Serial mode, since all modes (with the exception XC4000 Series and XC5200 Family
of XC5200 Express mode) internally operate on serial
Pull the PROGRAM input Low for at least 0.3 microseconds
data. The internally generated CCLK frequency is
to initiate clearing the configuration memory, then pull
guard-banded to never approach the upper limit of what
PROGRAM up to start the new configuration process.
the device can tolerate. Therefore, the fastest possible
configuration mode for XC3000 and XC4000-series While PROGRAM is held Low, a Low level on INIT indicates
devices is Slave Serial, with an external well-controlled that the device is continuously clearing the configuration
source for CCLK. Its frequency can be up to 10 MHz for memory. When PROGRAM has been pulled up, INIT stays
all 5-V devices, and there are ways to increase the Low during one more clear operation, then goes High.
average clock rate well beyond that, but they require
All device families, except the original XC4000, have a con-
dynamic clock frequency changes and an intimate
tinuously active pull-up resistor on the PROGRAM pin.
understanding of the configuration frame structure.
At 10 MHz, configuration time per device ranges from FPGAs Can Control Their Own
1.5 ms for the XC3020A to 42 ms for the XC4025E and
192 ms for the XC4085XL. Reconfiguration
• Possible Contention Problems: Pulling PROGRAM, RESET or DONE low can trigger a
Certain user outputs become active during the configu- reconfiguration, as described above. When a user output is
ration process: connected to drive the reconfiguration pin, the FPGA can
Address outputs during Master Parallel mode, Chip trigger its own reconfiguration. Although the triggering out-
Select and Ready/Busy during Peripheral modes. put will go 3-state once reconfiguration is initiated, this trig-
The designer must make sure that these active outputs ger operation is reliable.
do not cause contention with other logic that might use
Such auto-reconfiguration offers interesting opportunities
the same pins as device inputs.
for small systems using a single FPGA in Master Parallel
configuration mode. A manually operated switch selects
Initiating Reconfiguration in the most significant address bits of the PROM, and the
Different Xilinx Device Families FPGA compares the switch settings against a stored value.
Upon detecting a difference, it can trigger reconfiguration
XC3000 Series that is loaded from the newly selected PROM address
There are three alternatives: range. Or an external CMOS register can be loaded with 14
the intended reconfiguration address range and then con-
1. Pull RESET Low while DONE is permanently grounded
trol the upper bits of the PROM address
externally.

0
Metastable Recovery
XAPP094 November 24, 1997 (Version 2.1) 0 14* Application Note By Peter Alfke and Brian Philofsky
Introduction tination might clock in the final data state while the other
does not.
Whenever a clocked flip-flop synchronizes an asynchro-
nous input, there is a small probability that the flip-flop out- With the help of a self-contained circuit, Xilinx evaluated the
put will exhibit an unpredictable delay. This happens when XC4000 and XC3000-series flip-flops. The result of this
the input transition not only violates the setup and hold-time evaluation shows the Xilinx flip-flop to be superior in meta-
specifications, but actually occurs within the tiny timing win- stable performance to many popular MSI and PLD devices.
dow where the flip-flop accepts the new input. Under these Since metastability can only be measured statistically, this
circumstances, the flip-flop can enter a symmetrically bal- data was obtained by configuring several different Xilinx
anced transitory state, called metastable (meta = between). FPGAs with a detector circuit shown in Figure 1. The
While the slightest deviation from perfect balance will flip-flop under test receives the asynchronous ~1-MHz sig-
cause the output to revert to one of its two stable states, the nal on its D input, and is clocked by a much higher manually
delay in doing so depends not only on the gain-bandwidth adjustable frequency. The output QA feeds two flip-flops in
product of the circuit, but also on how perfect the balance parallel, one (QB) being clocked by the same clock edge,
is, and on the noise level within the circuit; the delay can, the other (QC) being clocked by the opposite clock edge.
therefore, only be described in statistical terms. When clocked at a low frequency, each input change gets
captured by the rising clock edge and appears first on QA,
The problem for the system designer is not the illegal logic then, after the falling clock edge, on QC, and finally, after
level in the balanced state (it’s easy enough to translate the subsequent rising clock edge, on QB.
that to either a 0 or a 1), but the unpredictable timing of the
final change to a valid logic state. If the metastable flip-flop If a metastable event in the first flip-flop increases the set-
drives two destinations with differing path delays, one des- tling time on QA so much that QC misses the change, but
QB still captures it on the next rising clock edge, this error
QA QB QC QD 16-Bit
Asynchr, Input D D D D
Counter
16 LEDs
Clock
Clock
Asynchr, Input
14
QA
QB
NO ERROR
QC
ERROR
ERROR
QD
NO ERROR
X5985
Figure 1: Test Circuit and Timing Diagram

R
Metastable Recovery
can be detected by feeding the XOR of QB and QC into a generated a ~1 Hz error rate, FH generated a ~64,000 Hz
falling-edge triggered flip-flop. Its output (QD) is normally error rate.
Low, but goes High for one clock period each time the asyn-
K2 is derived by dividing ln 64,000 by the half-period differ-
chronous input transition caused such a metastable delay
ence.
in QA. The frequency of metastable events can be
observed with a 16-bit counter driven by QD. Table 1: Metastable Measurement Results
By changing the clock frequency, and thus the clock FL FH Half-period K2

half-period, the amount of acceptable metastable delay on Device (MHz) (MHz) Difference (ns) (1/ ns)
the QA output can be varied, and the resulting frequency of XC4005E-3 IOB 111.5 131.6 0.685 16.1
metastable events can be observed on the counter outputs.
XC4005E-3 CLB 109.0 124.4 0.568 19.4
As expected, no metastable events were observed at clock XC4005-6 IOB 73.0 90.0 1.294 8.5
rates below 70 MHz for the XC4005-6, or below 100 MHz XC4005-6 CLB 71.2 88.8 1.392 7.9
for the XC4005E-3, since a half clock period at those fre-
XC5206-5 CLB 70.8 79.8 0.80 13.7
quencies is adequate for almost any metastability-resolu-
tion delay. Increasing the clock rate slightly brought a XC3142A-09 IOB 152.2 206.6 0.87 12.7
sudden burst of metastable events. Careful adjustment of XC3142A-09 CLB 107.4 211.3 2.29 4.8
the clock frequency gave repeatable, reliable measure- XC3042-70 IOB 46.6 61.5 2.60 4.2
ments. XC3042-70 CLB 41.9 64.8 4.22 2.6
Metastability Measurements
The circuit of Figure 1 was implemented in five different Xil- Metastability Calculations
inx devices: two cutting-edge devices using 0.5 micron, The Mean Time Between Failures (MTBF) can only be
3-layer-metal technology, the XC4005E-3 and the defined statistically. It is inversely proportional to the prod-
XC3142A-09, one device, the XC5206 using 0.6 micron, uct of the two frequencies involved, the clock frequency and
3-layer-metal, and, for comparison purposes, also in two the average frequency of the asynchronous data changes,
older-technology devices, the XC4005-6 and the provided that these two frequencies are independent and
XC3042-70. have no correlation.
In each device two different implementations put QA, the The generally accepted equation for MTBF is
flip-flop under test, into an IOB and a CLB (Except for the
XC5200 family which has no flip-flops in the IOB). The eK2 * t
XC4000-series devices showed little difference between MTBF =
F1 * F2 * K1
IOB and CLB behavior, but in the XC3000-series devices,
the IOB flip-flops showed dramatically better metastable K1 represents the metastability-catching set-up time win-
performance than the CLB flip-flops. This difference can be dow, which describes the likelihood of going metastable.
traced to subtle differences in circuit design and layout, and
K2 is an exponent that describes the speed with which the
will guide us to further improvements in metastable perfor-
metastable condition is being resolved. K2 is an indication
mance in future designs.
of the gain-bandwidth product in the feedback path of the
Metastable measurement results are listed in Table 1, and master latch of the master-slave flip-flop. A small increase
are plotted in Figure 2. The results for XC4000E-3 (IOB and in K2 results in an enormous improvement in MTBF.
CLB) and for XC3100A-09 IOB flip-flops are outstanding,
With F1 = 1 MHz, F2 = 10 MHz and K1 = 0.1 ns = 10-10 s:
far superior to most metastable data published anywhere
else. When granted 2 or 3 ns of extra settling delay, these MTBF (in seconds) = 10–3 * eK2*t
devices come close to eliminating the problems caused by
metastability, since their MTBF exceeds millions of years. Experimentally derived (see Table 1): 14
The older-technology devices are obviously less impres- K2 = 16.1 per ns, for the XC4005E-3 IOB flip-flops
sive, but they still show acceptable performance, especially K2 = 19.4 per ns, for the XC4005E-3 CLB flip-flops
in the IOB input flip-flops that are normally used to synchro-
nize asynchronous input signals. K2 = 8.5 per ns, for the XC4005-6 IOB flip-flops
Table 1 lists the experimental results from which the expo- K2 = 7.9 per ns, for the XC4005-6 CLB flip-flops
nential factor K2 was derived. The clock frequency was K2 = 13.7 per ns, for the XC5206-5 CLB flip-flops
adjusted manually, while observing the LSB and the MSB
K2 = 12.7 per ns, for the XC3142A-09 IOB flip-flops
of the 16-bit error counter. FL is the clock frequency that
K2 = 4.8 per ns, for the XC3142A-09 CLB flip-flops

R
Metastable Recovery
XC4005E-3 XC4005E-3 XC5206-5 XC3142A-09 XC4005-6 XC4005-6

MTBF CLB IOB CLB IOB IOB CLB
1 Million Years
13
12
11 XC3142A-09
1,000 Years
CLB
10
9
Log Seconds
8
1 Year
XC3042-70
7
IOB
6
5 1 Day
4
1 Hour
XC3042-70
3
CLB
2
1 Minute
1
-1 1 2 3 4 5 6
-2 Acceptable Extra Delay (ns)

X5986
-3
Figure 2: Mean Time Between Failure for various IOB and CLB flip-flop outputs when synchronizing a ~1 MHz
asynchronous input with a 10 MHz clock.
K2 = 4.2 per ns, for the XC3042-70 IOB flip-flops synchronized by a 40 MHz clock, the MTBF is 40 times
K2 = 2.6 per ns, for the XC3042-70 CLB flip-flops shorter than plotted; for a ~50 kHz signal synchronized by a
1 MHz clock, the MTBF is 200 times longer than plotted
For other operating conditions, divide MTBF by the product here.
of the two frequencies. For a ~10 MHz asynchronous input

0
Set-up and Hold Times
XAPP 095 November 24, 1997 (Version 1.0) 0 14* Application Brief by Peter Alfke
Introduction If the receiving device has a hold time requirement, the

source of data must guarantee an equivalent minimum
Beware of hold-time problems, because they can lead to value for its clock-to-output delay. Almost no IC manufac-
unreliable, temperature-sensitive designs that can fail even turer is willing to do this, and in the few cases where it is
at low clock rates. done, the minimum value is usually a token 1 ns. Any input
“Set-up time” and “hold time” describe the timing require- hold time requirement is, therefore, an invitation to system
ments on the data input of a flip-flop or register with respect failure. Any clock distribution skew on the PC-board can
to the clock input. The set-up and hold times describe a compound this issue and wipe out even the specified short
window of time during which data must be stable in order to minimum delay.
guarantee predictable performance over the full range of Xilinx has addressed this problem by adding a deliberate
operating conditions and manufacturing tolerances. delay to every FPGA data input. In XC3000, and XC3100
A positive set-up time describes the length of time that the FPGAs, this delay is fixed and always present; in XC4000
data must be available and stable before the active clock and XC5200 FPGAs, this delay is optional, and its value is
edge. A positive hold time, on the other hand, describes the tailored to the clock distribution delay (i.e. it is larger for big-
length of time that the data to be clocked into the flip-flop ger devices). As a result we can claim that no Xilinx FPGA
must remain available and stable after the active clock Data input has a hold-time problem (i.e., none has a posi-
edge. A positive set-up time limits the maximum clock rate tive hold time with respect to the externally applied clock),
of a system, but a positive hold time can cause malfunction when the design uses the internal global clock distribution
at any clock rate. Thus, chip designers and system design- network (and, in XC4000 and XC5200, uses the delayed
ers strive to eliminate hold-time requirements. input option). Most competitive devices do not offer this fea-
ture.
The IC design usually guarantees that any individual
flip-flop does not require a positive hold time with respect to
Internal Clock Delay
the clock signal at this flip-flop. External Clock
Hold-time requirements between flip-flops or registers on Internal Clock
the same chip can be avoided by careful design of the Conventional Input Pin
SET-UP
Set-up and Hold Time H
on-chip clock distribution network. If the worst-case
clock-skew value is shorter than the sum of minimum Input Pin Set-up
SET-UP
Time With Delay
clock-to-Q plus minimum interconnect delays, there is X5971
never any on-chip hold-time problem.
It is, however, far more difficult to avoid a hold time problem Figure 1: Set-up and Hold TImes
in the device input flip-flops, with respect to the device clock
input pin. When specifying the data pin-to-clock pin set-up
and hold times, the chip-internal clock distribution delay
must be taken into consideration. It effectively moves the
timing window to the right (seeFigure 1, thus subtracting
from the specified internal set-up time (which is good), but
adding to the hold time (which is very bad). If the clock dis- 14
tribution delay is any longer than the data input delay – and
it easily might be – the device data input has a hold-time
requirement with respect to the clock input.
This means that the data source, usually another IC driven
by the same clock, must guarantee to maintain data beyond
the clock edge. In other words, the data source is not
allowed to be very fast. If it is, the receiver might errone-
ously input the new data instead of the data created by the
previous clock, as it should. This is called a race condition,
and can be a fatal system failure.
XAPP 095 November 24, 1997 (Version 1.0) 14-49

0
Overshoot and Undershoot
XAPP096 September 9, 1997 (Version 1.0) 0 14* Application Note By Peter Alfke
Introduction
The “Absolute Maximum Ratings” table in the Xilinx Data
Book restricts the signal-pin voltage to a maximum 500 mV
excursion above VCC and below ground. The reason for
this tight specification is to prevent uncontrolled current in
the input-clamping ESD-protection diodes. Such tight spec-
ifications are common in the industry; some manufacturers
limit the excursion to 300 mV.
This specification seems to be clean and simple, but it is
violated in almost every practical design. When users put
modern CMOS devices on PC boards, and interconnect
them with unterminated traces, there are reflections, com-
monly called “ringing”, that cause overshoots and under-
shoots of substantial amplitude (2 V and more). The recent
migration to smaller device geometries has made the IC
outputs even faster and increased the slew-rate, causing
more reflections even on short PC-board traces.
Fortunately, this problem has an easy solution:
The concern is not the input voltage, but rather the current
through the input protection diode and other input struc-
tures. Excessive current can cause latch-up if it exceeds
hundreds of milliamps AND if it lasts for microseconds
(shorter duration current spikes do not activate the
SCR-like latch-up mechanism).
PC-board reflections, on the other hand, usually have a
short duration of just a few nanoseconds, and have an
impedance of 40 to 100 Ω, which makes them incapable of
causing latch-up. They don’t drive enough current and they
don’t last long enough to cause any harm.
Here is the new Xilinx specification:
“Maximum DC overshoot or undershoot above VCC or
below GND must be limited to either 0.5 V or 10 mA, which-
ever is easier to achieve. During transitions, the device pins
may undershoot to -2.0 V or overshoot to Vcc + 2.0 V, pro-
vided this over- or undershoot lasts less than 20 ns”.
14
XAPP096 September 9, 1997 (Version 1.0) 14-50

APPLICATION NOTE
0
 Boundary Scan in XC4000 and

XC5200 Series Devices
XAPP017 February 4, 1999 (Version 2.1) 0 14* Application Note
Summary
XC4000 and XC5200 Series FPGA devices contain boundary-scan facilities that are compatible with IEEE Standard 1149.1.
This Application Note describes those facilities in detail, and explains how boundary scan is incorporated into an FPGA
design.
Xilinx Family
XC4000 Series, XC5200
Introduction its internal shift registers, and passing the serial data
directly to the next device.
In production, boards must be tested to assure the integrity
of the components and the interconnections. However, as XC4000/XC5200 FPGA devices contain boundary-scan
integrated circuits have become more complex and registers that are compatible with the IEEE Standard
multi-layer PC-boards have become more dense, it has 1149.1, that was derived from a proposal by the Joint Test
become increasingly difficult to test assembled boards. Action Group (JTAG). External (I/O and interconnect) test-
ing is supported; there is also limited support for internal
Originally, manufacturers used functional tests, applying self-test.
input stimuli to the input connectors of the board, and
observing the results at the output. Later, “bed-of-nails” Overview of XC4000/XC5200
testing became popular, where a customized fixture
presses sharp, nail-like stimulus- and test-probes into the
Boundary-Scan Features
exposed traces on the board. These probes were used to XC4000/XC5200 devices support all the mandatory bound-
force signals onto the traces and observe the response. ary-scan instructions specified in the IEEE Standard
1149.1. A Test Access Port (TAP) and registers are pro-
However, increasingly dense multi-layer PC boards with
vided that implement the EXTEST, SAMPLE/PRELOAD
ICs surface-mounted on both sides have stretched the
and BYPASS instructions. The TAP can also support two
capability of bed-of-nail testing to its limit, and the industry
USERCODE instructions.
is forced to look for a better solution. Boundary-scan tech-
niques provide that solution. Note: If boundary scan is not used after the device is con-
figured, the user can use the special boundary scan pads
The inclusion of boundary-scan registers in ICs greatly
as input or output pins. During configuration, be sure not to
improves the testability of boards. Boundary scan provides
toggle the TAP pins, since inadvertent toggling of the TAP
a mechanism for testing component I/Os and inter-connec-
pins can turn the boundary scan circuitry ‘on.’ The TDI,
tions, while requiring as few as four additional pins and a
TMS, and TCK pads can be used as unrestricted I/O. The
minimum of additional logic in each IC. Component testing
TDO pad can be used as an output pad. In the XC5200
may also be supported in ICs with self-test capability.
family, all four pins have full I/O capability. And like the reg-
Devices containing boundary scan have the capability of ular IOBs, these input and output pins have pullups and
driving or observing the logic levels on I/O pins. To test the pulldowns available. 14
external interconnect, devices drive values onto their out-
Boundary-scan operation is independent of individual IOB
puts and observe input values received from other devices.
configuration and package type. All IOBs are treated as
A central test controller compares the received data with
independently controlled bidirectional pins, including any
expected results. Data to be driven onto outputs is distrib-
unbonded IOBs. Retaining the bidirectional test capability
uted through a chain of shift registers, and observed input
even after configuration affords tremendous flexibility for
data is returned through the same shift-register path.
interconnect testing.
Data is passed serially from one device to the next, thus
Additionally, internal signals can be captured during
forming a boundary-scan path or loop that originates at the
EXTEST by connecting them to unbonded IOBs, or to the
test controller and returns there. Any device can be tempo-
unused outputs in IOBs used as unidirectional input pins.
rarily removed from the boundary-scan path by bypassing
This partially compensates for the lack of INTEST support.
XAPP017 February 4, 1999 (Version 2.1) 14-51

R
The public boundary-scan instructions are always available It should also be noted that the Test Data Register contains
prior to configuration. After configuration, the public instruc- three Xilinx test bits (BSCANT.UPD, TDO.O and TDO.T)
tions and any USER1/USER2 instructions are only avail- and that bits of the register may correspond to unbonded or
able if boundary scan specified in the schematic/HDL code. unused pins.
While SAMPLE and BYPASS are available during configu-
Additionally, the EXTEST instruction incorporates
ration, it is recommended that boundary-scan operations
INTEST-like functionality that is not specified in the stan-
not be performed during this transitory period.
dard, and system clock inputs are not disabled during
In addition to the test instructions outlined above, the EXTEST, as recommended in the standard.
boundary-scan circuitry can also be used to configure the
The TAP pins (TMS, TCK, TDI and TDO) are scanned, but
FPGA device, and read back the configuration data.
connections to the TAP controller are made before the
The following description assumes that the reader is famil- boundary-scan logic. Consequently, the operation of the
iar with boundary-scan testing and the IEEE Standard. TAP controller cannot be affected by boundary-scan test
Only issues specific to the XC4000/XC5200 implementa- data.
tion are discussed in detail. For general information on
When the TAP is in the shift-DR state the contents of all
boundary scan, please refer to the bibliography.
data registers are shifted; if you are in the middle of shifting
out data from the data register, complete shifting out of all
Deviations from the IEEE Standard data first, before switching to the instruction or bypass reg-
The XC4000/XC5200 boundary scan implementation devi- ister.
ates from the IEEE standard in that three dedicated pins
(CCLK, PROGRAM and DONE) are not scanned.
1 TEST-LOGIC-RESET
1 1 1
0 RUN-TEST/IDLE SELECT-DR-SCAN SELECT-IR-SCAN
0 0
1 1
CAPTURE-DR CAPTURE-IR
0 0
SHIFT-DR 0 SHIFT-IR 0
1 1
1 1
EXIT1-DR EXIT1-IR
0 0
PAUSE-DR 0 PAUSE-IR 0
1 1
0 0
EXIT2-DR EXIT2-IR
1 1
UPDATE-DR UPDATE-IR
1 0 1 0
NOTE: The value shown adjacent to each state transition in this figure
represents the signal present at TMS at the time of a rising edge at TCK. X2680
Figure 1: State Diagram for the TAP Controller
14-52 XAPP017 February 4, 1999 (Version 2.1)

R
Boundary-Scan Hardware Table 1: Boundary Scan Instructions
Instruction Test TDO I/O Data

Test Access Port
I2 I1 I0 Selected Source Source
The boundary-scan logic is accessed through the Test 0 0 0 EXTEST DR DR
Access Port (TAP), which comprises four semi-dedicated 0 0 1 SAMPLE/ DR Pin/Logic
pins: Test Mode Select (TMS), Test Clock (TCK), Test Data PRELOAD
Input (TDI) and Test Data Output (TDO), as defined in the
0 1 0 USER 1 BSCAN.TDO1 User Logic
IEEE specification.
0 1 1 USER 2 BSCAN.TDO2 User Logic
The TAP pins are permanently connected to the bound- 1 0 0 READBACK Readback Data Pin/Logic
ary-scan circuitry. However, once the device is configured,
1 0 1 CONFIGURE DOUT Disabled
the connections may be ignored unless the use of bound-
ary scan is specified in the design. See “Using Boundary 1 1 0 RESERVED — —
Scan” on page 56. 1 1 1 BYPASS Bypass Register —
I0 is closest to TDO
If the use of boundary scan is specified, the TAP input pins
(TMS, TCK and TDI) may still be shared with other logic, Note: Whenever the TAP Controller is in the Shift-DR state,
subject to limitations imposed by external connections and all data registers are shifted, regardless of the instruction.
the operation of the TAP Controller. In designs that do not DR data is modified even if a BYPASS instruction is exe-
use boundary scan after configuration, the TAP pins can be cuted.
used as inputs or outputs from the user logic in the FPGA
The instruction register is used not only to hold the current
device. TMS, TCK and TDI are available as unrestricted
instruction. If the TAP is in the capture-IR state and TCK
I/Os, while TDO only provides a 3-state output. In the
goes high, the instruction register captures the current
XC5200 family, all four pins are available as I/O.
boundary-scan state of the device. I0 is 1 by default. I1 is 0
Before the FPGA is configured, it is important not to toggle by default. I2 is 0 if the device is in configure by boundary
the TAP pins (TDI, TMS, TCK), since these pins ‘turn-on’ scan mode. Before and after configure by boundary scan
boundary scan. Before an FPGA is configured, at a mini- mode, I2 will capture 1. Note that I0 is shifted out of TDO
mum, do not toggle TCK. Similarly, if boundary scan is first, then I1, and then I2.
enabled in a design after the FPGA is configured, care
must be taken not to toggle the TAP pins (TDI, TMS, TCK) The Boundary-Scan Data Register
to prevent turning ‘on’ boundary scan by accident.
The Data Register (DR) is a serial shift register imple-
TAP Controller mented in the IOBs of the FPGA device, (Figure 2). Poten-
tially, each IOB can be configured as an independently
The TAP Controller is a 16-state machine that controls the controlled bidirectional pin. Therefore, three data register
operation of the boundary-scan circuitry in response to bits are provided per IOB: for input data, output data and
TMS. This state machine implements the state diagram 3-state control. In practice, many of these bits are redun-
specified by the IEEE standard (Figure 1) and is clocked by dant, but they are not removed from the scan chain.
TCK.
An update latch accompanies each bit of the DR, and is
Upon power-on, or if the boundary scan logic is not used in used to hold injected test data stable during shifting. The
the application, the TAP controller is forced into the update latch is opened during the Update-DR state of the
Test-Logic-Reset state. After configuration, the controller TAP Controller when TCK is Low.
remains disabled, unless its use is explicitly specified in the
In a typical DR instruction, the DR captures data during the
user design. PROGRAM resets the latched decodes for
Capture-DR state (on the rising edge of TCK). This data is
EXTEST, CONFIGURE, and READBACK instructions.
then shifted out and replaced with new test data. Subse-
Loading a 3-bit instruction into the Instruction Register (IR) quently, the update latch opens, and the new test data 14
determines the subsequent operation of the bound- becomes available for injection into the logic or the inter-
ary-scan logic, Table 1. The instruction selects the source connect. The injection of data occurs only if an EXTEST
of the TDO pin, and selects the source of device input and instruction is in progress.
output data (boundary-scan register or input pin/user logic).
Note: The update latch is opened whenever the TAP Con-
troller is in the Update-DR state, regardless of the instruc-
tion. Care must be exercised to ensure that appropriate
data is contained in the update latch prior to initiating an
EXTEST. Any DR instruction, including BYPASS, that is
executed after the test data is loaded, but before the
EXTEST commences, changes the test data.

R
From TDI
Pull-Up
1 sd
Pull-Down
D Q D Q
0
To Global
LE
Clock Buffer
(CLK Pad Only)
1
IOB.I VCC
(To FPGA Interconnect) 0
IOB
1 sd
D Q D Q
0
LE
IOB.O 0
(From FPGA Interconnect)
IOB.T 0
1 sd
D Q D Q 1
0
LE
Towards TDO
Test Logic
Shift/Capture DRCK Update Reset EXTEST
X5998
Figure 2: Boundary Scan Logic in a Typical IOB
The IEEE Standard does not require the ability to inject boundary-scan data. However, if necessary, it is possible to
data into the on-chip system logic and observe the results drive the clock input from boundary scan. The external
during EXTEST. However, this capability helps compensate clock source is 3-stated, and the clock net is driven with
for the lack of INTEST. Logic inputs may be set to specific boundary scan data through the output driver in the
levels by a SAMPLE/PRELOAD or EXTEST instruction and clock-pad IOB. If the clock-pad IOBs are used for non-clock
the resulting logic outputs captured during a subsequent signals, the data may be overwritten normally.
EXTEST. It must be recognized, however, that all DR bits
Figure 3 shows the data-register cell for a TAP pin. An
are captured during an EXTEST and, therefore, may OR-gate permanently disables the output buffer if bound-
change. ary-scan operation is selected. Consequently, it is impossi-
Pull-up and pull-down resistors remain active during ble for the outputs in IOBs used by TAP inputs to conflict
boundary scan. Before and during configuration, all pins with TAP operation. TAP data is taken directly from the pin,
are pulled up. After configuration, the IOB can be config- and cannot be overwritten by injected boundary-scan data.
ured with a pull-up resistor, a pull-down resistor or neither.
Bit Sequence
Note: Internal pull-up/pull-down resistors must be taken
into account when designing test vectors to detect open cir- Table 2 lists, in data-stream order, the boundary-scan cells
cuit PC traces. that make up the DR for the XC4000 Series. The cell clos-
est to TDO corresponds to the first bit of the data-stream,
The primary and secondary global clock inputs (PGCK1-4
and is at the top of the table. This order is consistent with
and SGCK1-4 in XC4000, GCK1-4 in XC5200) are taken
the BSDL description.
directly from the pins, and cannot be overwritten with

R
From TDI
sd Pull-Up
1 Pull-Down
D Q D Q
0
LE To Tap
Controller
1
IOB.I VCC
(To FPGA Interconnect) 0
IOB
1 sd
D Q D Q
0
LE
IOB.O 0
(From FPGA Interconnect)
IOB.T
1
D Q
0
Boundary
Scan Enabled
Device Not
Towards TDO Configured
Test Logic
Shift/Capture DRCK Update Reset EXTEST
X5999
Figure 3: Boundary Scan Logic in a TAP Input (TMS, TCK, and TDI Only)
Each IOB corresponds to three bits in the DR. The 3-state Note: All IOBs remain in the DR, independent of whether
control is first (closest to TDO), the output is next, and the they are actually used, or even bonded. Three bits,
input is last. Other signals correspond to individual register BSCANT.UPD, TDO.O and TDO.T, are included for Xilinx
bits. IOB locations assume that the die is viewed from the test purposes, and may be ignored by other users. CCLK,
top, as in the device-level editors XDE or EPIC. In the PROGRAM and DONE are not included in the boundary
XC4000, the input-only M0 and M2 mode pins contribute scan.
only the In bit to the boundary scan I/O register. Tables in the data sheets show the DR order for all
Table 2: XC4000 Boundary Scan Order XC4000/XC5200 family devices. The DR also includes the
following non-pin bits: TDO.T and TDO.I, which are always
bits 0 and 1 of the DR, respectively, and BSCANT.UPD
which is always the last bit of the DR.
Bit 0 ( TDO end) TDO.T
Bit 1 TDO.O
Bit 2 The Bypass Register
Top-edge IOBs (Right to Left)
This is a 1-bit shift register that passes the serial data
directly to TDO when a BYPASS instruction is executed.
Left-edge IOBs (Top to Bottom)
MD1.T User Registers 14

MD1.O
MD1.I The XC4000 and XC5200 boundary-scan instruction set
MD0.I
MD2.I
includes two USERCODE instructions, USER1 and
USER2. Connections are provided to the TAP and TAP
Bottom-edge IOBs (Left to Right) controller that, together with direct connections to the TAP
pins, permit the user to include boundary-scan self-test fea-
Right-edge IOBs (Bottom to Top) tures in the design.
(TDI end) B SCANT.UPD The XC4000 boundary scan symbol has six connections
for user registers: SEL1, SEL2, TDO1, TDO2, DRCK and
X2674 IDLE. TDI is available directly from the IOB that provides
the TDI pin. The XC5200 boundary scan symbol has three

R
additional pins which make the creation of a user register Figure 4 is a flow chart of the XC4000 FPGA start-up
easier: RESET, UPDATE, and SHIFT. sequence that shows when the boundary-scan instructions
are available. Since PROGRAM resets the TAP controller,
Note: The TDI signal supplied to user test logic is overwrit-
ten by boundary-scan test data during EXTEST. During boundary-scan operations cannot commence until PRO-
user tests, it is not altered. GRAM has been taken High.
SEL1, SEL2 – SEL1 and SEL2 enable user logic. They are .
asserted (High) when the instruction register contains
instructions USER1 and USER2, respectively. VCC No
Boundary Scan >3.5 V
TDO1, TDO2 – TDO1 and TDO2 are inputs to the TDO out- Instructions
Available:
Yes
put multiplexer, permitting user access to the serial bound-
ary-scan output. They are selected when executing the
Test M0 Generate
instructions USER1 and USER2, respectively. Input to user One Time-Out Pulse PROGRAM
= Low
of 16 or 64 ms
data registers can be derived directly from the TDI pin, thus Yes
completing the boundary-scan chain.
Keep Clearing
There is a one flip-flop delay between TDO1/TDO2 and the
TDO output. This flip-flop is clocked on the falling edge of
EXTEST*
TCK. SAMPLE/PRELOAD Completely Clear
BYPASS Configuration Memory ~1.3 µs per Frame
CONFIGURE*
DRCK – Data register clock (DRCK) is a gated and unin- (* if PROGRAM = High)
Once More
verted version of TCK. It is provided to clock user test-data

registers. TDI data should be sampled with the falling edge INIT No
High? if
of DRCK (rising edge of TCK). The TDO output flip-flop Master
accepts data on the rising edge of DRCK (falling edge of Yes Master Waits 50 to 250 µs
Before Sampling Mode Lines
TCK). DRCK is active only during the Capture-DR and
Shift-DR states of the TAP controller. When not active in the Sample
Mode Lines
XC4000, DRCK is Low. In the XC5200, when DRCK is not
Master CCLK
active, it is High. Goes Active

Load One
IDLE – IDLE is a second gated and inverted version of Configuration
Data Frame
TCK. It is active during the RUN-TEST/IDLE state of the
TAP controller, and may be used to clock user test logic a
set number of times, determined through TMS by the cen- Frame Yes Pull INIT Low
tral test controller. Error and Stop
No
RESET - This pin is only available on the XC5200 boundary
scan symbol. Whenever the TAP is in the SAMPLE/PRELOAD
BYPASS
Config-
uration No
TEST-LOGIC-RESET state, the RESET pin is High, in all memory
Full
other cases the RESET pin is Low. Yes
UPDATE - This pin is only available in the XC5200 bound- Pass

Configuration
ary scan symbol. Whenever the USER1 or USER2 instruc- Data to DOUT
tions are used, UPDATE is an inverted version of TCK. In

all other cases, UPDATE is Low. CCLK
Count Equals No
SHIFT - This pin is only available in the XC5200 boundary Length
Count
scan symbol. When the USER1 or USER2 instructions are Yes
used, SHIFT is High, in all other cases SHIFT is Low.

Start-Up
Sequence
Using Boundary Scan F
I/O Active
Operational
Full access to the built-in boundary-scan logic is always EXTEST
SAMPLE PRELOAD
available between power-up and the start of configuration. BYPASS
If Boundary Scan
USER 1
Optionally, the built-in logic is fully available after configura- USER 2 is Selected
CONFIGURE
tion if boundary scan is specified in the design. At this time, READBACK X6076
user test logic is also available, and may be accessed

through the boundary-scan port. During configuration, a
reduced boundary-scan capability remains available: the Figure 4: XC4000 Start-up Sequence
SAMPLE/PRELOAD and BYPASS instructions only.

R
Full boundary-scan capabilities are available until INIT is the TDO pad primitive to an OBUF or OBUFT as required
High. Without external intervention, INIT automatically (see Figure 6.)
goes High after ~1 ms. If more time is required for bound-
From
ary-scan testing, INIT may be held Low beyond this period User
by applying an external Low signal to the INIT pin until test- Logic TDO
ing is complete. Once INIT has gone High, all clocks on the OBUFT
X2676
TCK pin are counted as configuration clocks for data and
length count. See “CONFIGURE” on page 58. for more Figure 6: Typical Non-Boundary-Scan TDO
details. Connection
Boundary scan can be accessed before the FPGA is con-
figured and after the FPGA is configured. If you want to Boundary Scan Instructions
access boundary scan before the device is configured, then The XC4000/XC5200 boundary scan supports three
when you power-up the device, hold the INIT pin Low until IEEE-defined instructions (EXTEST, SAMPLE/PRELOAD
VCC has risen to VCC(min). and BYPASS), two user-definable instructions (USER1 and
If you have already started configuring the device, and data USER2), and two FPGA-specific instructions (CONFIG-
frames are already being sent to the FPGA, then you have URE and READBACK). The instruction codes are shown in
two choices. You can either access full-boundary scan See Table 1 on page 14-53.
mode, or limited boundary scan mode. If you want to
access full-boundary scan mode, then both INIT and PRO- EXTEST
GRAM must be brought Low (Hold INIT and PROG Low for While the EXTEST instruction is present in the IR, the data
over 300 ns and then release PROGRAM.) After releasing presented to the device output buffers is replaced by data
PROGRAM, continue to hold INIT Low while sending sig- previously loaded through the boundary-scan DR and
nals to the TAP. If you can use the limited boundary scan stored in the update latch (Figure 7). SImilarly, the output
mode (which means you only can use the SAMPLE/PRE- 3-state controls are replaced, and the data passed to inter-
LOAD and BYPASS instructions), then just bring INIT Low. nal system logic from input pins is replaced.
Accessing boundary scan after the device is configured When a DR instruction cycle is executed, data arriving at
has one requirement. The BSCAN symbol must be instan- the device input pins is loaded into the DR. The data from
tiated/inserted into your design with the correct syntax (see the system logic that drives output buffers and their 3-state
Figure 5). In this case, activating boundary scan after con- controls is also loaded. This action occurs during the CAP-
figuration amounts to toggling the TAP pins. TURE-DR state of the TAP controller (Figure 1 on page
. 14-52). Data is serially shifted out of the DR during the
BSCAN SHIFT-DR state; simultaneously, new data is shifted in. In
TDI TDI TDO TDO the UPDATE-DR state, the new data is transferred into the
TMS TMS DRCK
update latch for use as replacement data, as described
above.
TCK TCK IDLE
TDO1 SEL1
The replacement of system data with update latch data
starts as soon as the EXTEST instruction is loaded into the
TDO2 SEL2
IR. For this data to be valid, it must have been loaded by a
4k BSCAN Syntax for BSCAN after configure symbol
previous EXTEST or SAMPLE/PRELOAD operation.
BSCAN Since the DR and update latch are modified during any DR
IBUF OBUF
TDI TDI TDO TDO instruction cycle, including BYPASS, the data in the update
IBUF
TMS TMS DRCK latch is only valid if it was loaded in the last DR instruction
IBUF
TCK TCK IDLE cycle executed before EXTEST is asserted.
14
TDO1 SEL1 The IEEE definition of EXTEST only requires that test data
TDO2 SEL2 be driven onto outputs, that 3-state output controls be over-
5k BSCAN Syntax for BSCAN after configure symbol X5966
ridden, and that input data be captured. The capture of out-
put data and 3-state controls and the forcing of test data
Figure 5: Boundary-Scan Schematic Symbols into the system logic is normally performed during INTEST.
The XC4000/XC5200 effectively performs EXTEST and
If the BSCAN symbol is not included, boundary scan is not INTEST simultaneously. This added functionality permits
selected, and the IOBs used by the TAP input pins are the testing of internal logic, and compensates for the
freely available as general purpose IOBs. The TDO output absence of a separate INTEST instruction. However, when
pin may be used as a logic output by explicitly connecting performing an EXTEST, care must be taken as to what sig-

R
Capture – DR
Update – DR
From To
Previous Test Data Register Next
Cell DRCK Cell
Update Latch
EN
Update – DR
System
Logic O
Pad
EXTEST
X2677
Figure 7: EXTEST Data Flow
nals are driven into the system logic. Data captured from Test clocks and paths to TDO are provided, together with
internal system logic must be masked out of the test-data two signals that indicate that user instructions have been
stream before performing check-sum analysis. loaded. See “User Registers” on page 55.
User tests depend upon CLBs and interconnect that must
SAMPLE/PRELOAD
be configured to operate. Consequently, they may only be
The SAMPLE/PRELOAD instruction permits visibility into performed after configuration.
system operation by capturing the state of the I/O. It also
permits valid data to be loaded into the update register CONFIGURE
before commencing an EXTEST. Steps to follow to configure a Xilinx XC4000 or XC5200
The DR and update latch operate exactly as in EXTEST device via JTAG:
(see above). However, data flows through the I/O unmodi- The bitstream format is identical for all configuration
fied.
modes. A user can use a design.BIT file or a design.RBT
file, depending on whether the user wants to read a binary
BYPASS
file (.BIT) or an ASCII file (.RBT).
The BYPASS instruction permits data to be passed syn-
1. Enable the boundary scan circuitry.
chronously to the next device in the boundary-scan path.
There is a 1-bit shift register between the TDI and TDO This can be done one of three ways, either during
flip-flop. power-up, or by configuring the device with boundary
scan enabled, or by pulling the PROGRAM pin low.
USER1, USER2
These instructions permit test logic, designed by the user To enable boundary scan during power-up, hold the
and implemented in CLBs, to be accessed through the TAP. INIT pin Low when power is turned on. When VCC has

R
reached VCC(min), the TAP inputs can be toggled to if configuration failed, the SAMPLE/PRELOAD instruc-
enter JTAG instructions. The INIT pin can be held Low tion can be used to view these IOBs (except PRO-
one of two ways, either manually or with a pulldown. If GRAM and DONE.)
you choose to manually hold the INIT low, then the INIT
pin must be held low until the CONFIGURE instruction LDC is Low during configuration. HDC is High during
is the current instruction. If you choose a pulldown, use configuration. INIT will be high impedance during con-
a pulldown which pulls the INIT pin down to approxi- figuration, but if a CRC error or frame error is detected,
mately 0.5V. The pulldown has the merit of holding INIT INIT will go Low. If a pulldown is present on INIT then
low whenever the FPGA is powered-up, and letting the the user must probe INIT with a meter or scope. With a
user observe the INIT pin during configuration. pulldown (as in step 1) attached to the INIT pin, the
user will see a drop from approximately 0.5V to 0V if
After the FPGA has been configured, if you want to INIT drops Low to indicate a data error. PROGRAM can
reconfigure a configured device that has boundary scan still be used to abort the configuration process. DOUT
enabled after configuration, then just start toggling the and TDO will echo TDI until the preamble and length
boundary scan TAP pins. count are shifted into TDI. After the preamble and
length count have been shifted into the FPGA, DOUT
2. Load the Xilinx CONFIGURE instruction into the Instruc-
will remain High. DONE will go High when configuration
tion Register (IR).
is finished. Until configuration is finished, DONE will
The Xilinx CONFIGURE instruction is 101(I2 I1 I0). I0 is remain Low.
the bit shifted first into the IR.
Additional Notes
3. After shifting in the Xilinx CONFIGURE instruction,
make the CONFIGURE instruction the current JTAG (a) It is possible to configure several XC4000/XC5200
instruction by going to the UPDATE-IR state. When TCK devices in a JTAG chain. But unlike non-JTAG daisy-chain
configuration, this does not necessarily mean merging all
goes low in the UPDATE-IR state, the FPGA is now in
the bitstreams into one bitstream. In the case of JTAG con-
the JTAG configuration mode and will start clearing the
figuration of Xilinx devices in a JTAG chain, all devices,
configuration memory. The CONFIGURE instruction is
except the one being configured, will be placed in BYPASS
now the current instruction, which must be followed by a
mode. The one device in CONFIGURE mode will have its
rising edge on TCK. If you chose to manually hold the
INIT pin Low, then the INIT pin must be held Low until bitstream downloaded to it. After configuring this device it
the CONFIGURE instruction is the current instruction. will be placed in BYPASS, and another device will be taken
out of BYPASS into CONFIGURE.
4. Once the Xilinx CONFIGURE instruction has been
made the current instruction, the user must go to the (b) If you are configuring a long daisy-chain of JTAG
RUN-TEST/IDLE state, and remain in the devices (TDI connected to TDO of the previous device), the
RUN-TEST/IDLE state until the FPGA has finished bitstream for the device with the CONFIGURE instruction
clearing its configuration memory. may need to have its bitstream modified.
The approximate time it takes to clear the FPGA config- For example, assume that the a user has the following
uration memory is: 2 * 1 us * (number of frames per daisy-chain of devices:
device bitstream). source -----> device1 -----> device2 -----> device3
Device1’s TDO pin is connected to device2’s TDI pin, and
When the FPGA has finished clearing its configuration
device2’s TDO pin is connected to device3’s TDI pin.
memory, the open-collector INIT has gone high imped-
ance. At this point, the user should advance to the The way to configure this chain is to place one device in
SHIFT-DR state. Once the TAP is in the SHIFT-DR state CONFIGURE, and the other two in BYPASS. Further
and the INIT pin has been released, clocks on the TCK assume that device1 and device2 configure in this way, but
pin will be considered configuration clocks for data and device3 never configures. Specifically, device3’s DONE pin 14
length count. never goes High. The problem is the bitstream length
count. A possible cause, aside from bitstream corruption,
5. In the SHIFT-DR state, start shifting in the bitstream.
is that the final value of the length count computed by the
Continue shifting in the bitstream until DONE has gone
user/software was reached before the loading was com-
High and the startup sequence has finished.
plete.
During the time you are shifting in the bitstream via the
There are two solutions. One solution involves just continu-
TAP, the configuration pins LDC, HDC, INIT, PRO-
ally clocking TCK (for about 15 seconds) until DONE goes
GRAM, DOUT, and DONE all function as they normally
High. The other solution is to modify the bitstream;
do during non-JTAG configuration. These pins can be
increase the length count by the number of devices ahead
probed by the user. After completion of configuration, or
of the device under configuration.

R
In the preceding example, the user would increase the merge the bitstreams into one bitstream, the user should
length count value by 2. (In a daisy-chain of devices con- configure as in note (a) above. Additionally, the user will
figuring via boundary scan, devices in BYPASS will supply have to tie all INIT pins together. All DONE pins will also
the extra 1s needed at the head of the bitstream.) have to be tied together.
(c) In general for the XC4000 and XC5200, if you are con- NOTE: The intention of configuration for a daisy-chain was
figuring these devices via JTAG, finish configuring the to use either all the devices in boundary scan, or all the
device first before executing any other JTAG instructions. devices in non-boundary scan configuration.
Once configuration through boundary scan is started, the
configuration operation must be finished. READBACK
(d) If boundary scan is not included in the design being Readback through boundary scan allows the user to
configured, then make sure that the release of I/Os is the access the readback features of the device, which would
last event in the startup sequence. normally need to be accessed through user-specified pins.
All limits of ‘normal’ readback are the same with readback
If boundary scan is not available, the FPGA is configured,
through the TAP. Like regular readback, readback through
and the I/Os are released before the startup sequence is
the TAP is at a minimum of 100 KHz and at a maximum of
finished, the FPGA will not respond to input signals and
2 MHz. Like regular readback, the readback bitstream
outputs will not respond at all.
through boundary scan has the same format.
(e) Re-issuing a boundary scan CONFIGURE instruction
Unlike regular readback, which can be done repeatedly,
after the clearing of configuration memory will cancel the
readback through the TAP requires the following circuit:
CONFIGURE instruction.
1. In your schematic, or top-level synthesis design, instan-
The proper method of re-issuing a CONFIGURE instruction
tiate the BSCAN and READBACK symbols.
after the configuration memory is cleared is to issue
another boundary scan instruction, and follow it by the 2. Connect the BSCAN symbol pins TDI, TMS, TCK, and
CONFIGURE instruction. TDO to the boundary scan pads TDI, TMS, TCK, and
TDO, respectively.
(f) If configuration through boundary scan fails, there are
only two boundary scan instructions available: SAM- 3. Next, connect the net between the TCK pad and TCK
PLE/PRELOAD and BYPASS. If another reconfiguration is pin on the BSCAN symbol to an IBUF. Take the output of
to be attempted, then the PROGRAM pin must be pulled the IBUF and connect it to the CLK pin of the READ-
Low, or the FPGA must be repowered. BACK symbol. See Figure 8.
BSCAN
(g) When the CONFIGURE instruction is the current
TDI TDI TDO TDO
instruction, clocks on the TCK pin are not considered con-
TMS TMS DRCK
figuration clocks until the INIT pin has gone high imped-
ance, and the TAP is in the SHIFT-DR state. TCK TCK IDLE READBACK
TDO1 SEL1 CLK DATA

(h) If the user is attempting to configure a chain of devices,
TDO2 SEL2 TRIG RIP
it is recommended that the user only configure the chain in
all boundary scan mode, or use the non-boundary scan IBUF
configuration modes. It is possible to configure a

• 4k BSCAN Symbol setup for multiple READBACKS through TAP
daisy-chain of devices, some in boundary scan and some • For the 5k, add IBUFs to TDI, TMS, and TCK. For TDO, add an OBUF.
(see figure 5)
in non-boundary scan configuration. Configuring in a X5968
mixed mode will not necessarily give the user a continuous

boundary scan chain, which may or may not be a problem Figure 8: Symbol Setup for Multiple Readbacks
for a particular user’s applications.
(j) Currently, there is no software to configure a Xilinx FPGA For the XC5200, the equivalent circuit must be imple-
via the boundary scan pins. The user must provide this. mented using the XACT Design Editor (XDE) program
EditLCA, or EPIC in the M1-based tools. After placing
(k) Configuring a chain of Xilinx FPGAs via boundary scan and routing your XC5200 design, load the design.LCA
does not require merging all the bitstreams into one bit- file into EditLCA, and follow the procedures below:
stream, as in non-boundary scan configuration (<ENTER> means hit the enter key on your keyboard)
daisy-chains. When the FPGA is in boundary scan config-
uration, the same configuration circuitry used for (a) Once EditLCA has displayed the design.LCA file,
non-boundary scan configuration is used. So, if a user type the following:
would like, it is possible to merge all bitstreams into one bit- eb bscan <ENTER>
stream, using the PROM File Formatter or
This will bring up the Editblock window for the XC5200
MakePROM/promgen. In a case where the user wants to
BSCAN symbol.

R
(b) In the Editblock window, select the ‘used’ option, Alternatively, if you do not want to go back to the
which is in the upper left corner of the screen. TEST-LOGIC-RESET state, realize that after shifting out
the readback bitstream, a minimum of three additional
(c) Now type:
clocks are needed on the readback register. So, after doing
endb <ENTER> a readback, instead of going back to TEST-LOGIC-RESET,
This brings you back to the EditLCA screen. a user can opt to execute some other JTAG instruction, and
then perform another readback.
(d) Next type the following:
Also, this procedure is only needed if you intend to do more
addnet username tckpin.i rdbk.ck <ENTER>
than one readback. If you intend only do a readback once,
where tckpin is the pin number of the TCK pin of your then the connection between the BSCAN symbol and the
XC5200 device. ‘username’ is a net name of your READBACK symbol is not needed. In that case, all that is
choIce. For example, if your design used an needed is the BSCAN symbol instantiated with the bound-
XC5202PC84, then the above command line would be: ary scan pads (TDI, TMS, TCK, & TDO) on the top level of
the design.
addnet mynet p16.i rdbk.ck <ENTER>
(e) At this point you should see a net go from the TCK pin Boundary Scan Description
to the CK pin of the Readback symbol.
Language Files
(f) Save your changes to the LCA file and exit XDE.
Boundary Scan Description Language (BSDL) files
4. After entering the above circuit, compile the design to an describe boundary-scan-capable parts in a standard for-
LCA file. mat used by automated test-generation software. The
5. Make the bitstream file for the LCA file by using the fol- order and function of bits in the boundary-scan data regis-
lowing option with makebits, or use the M1 Bitstream ter are included in this description.
Generator: BSDL files are available in the Xilinx File Download area via
-f readclk:rdbk the Xilinx WebLINX web site (www.xilinx.com).
For example, at a unix prompt:
% makebits -f readclk:rdbk design Bibliography
6. Now the FPGA is ready to perform consecutive read- The following publications contain information about the
backs. IEEE Standard 1149.1, and should be consulted for gen-
eral boundary-scan information beyond the scope of this
Readback is performed by loading the IR with the application note.
READBACK instruction and then shifting out the cap-
tured data from the SHIFT-DR state in the TAP. Read- Colin M. Maunder & Rodham E. Tulloss. The Test Access
back data is captured when READBACK is made the Port and Boundary Scan Architecture. IEEE Computer
current instruction in the TAP. Society Press, 10662 Los Vaqueros Circle, P.O. Box 3014,
Los Alamitos, CA 90720-1264. See www.com-
Perform the first readback by loading the IR with the puter.org/cspress/catalog/st01096.htm
READBACK instruction. This first readback must be fin- John Fluke Mfg. Co. Inc. The ABC of Boundary Scan Test.
ished, which means shifting out the *entire* readback John Fluke Mfg. Co. Inc., P.O. Box 9090, Everett, WA
bitstream. To be safe, shift out the entire bitstream and 98206.
then send three additional TCKs.
GenRad Inc. Meeting the Challenge of Boundary Scan.
7. After performing the first readback, another readback GenRad Inc., 300 Baker Ave., Concord, MA 01742-2174.
can be performed by going to the TEST-LOGIC-RESET
Ken Parker. The Boundary Scan Handbook. Kluwer Aca-
state, and re-loading the READBACK instruction and
demic Publications, (617) 871-6600.
performing the Readback as described in the previous 14
paragraph. IEEE Standards, standards.ieee.org
In summary, consecutive readbacks are performed by Texas Instruments, www.ti.com/sc/docs/jtag/jtaghome.htm
starting from TEST-LOGIC-RESET, loading the IR with
the READBACK instruction, shifting out the readback
bitstream plus three additional TCKs, and then going
back to the TEST-LOGIC-RESET state.

APPLICATION NOTE
0
R
Virtex I/V Curves for Various
Output Options
January 4, 1999 (Version 1.0) 0 14* Application Note
These typical curves describe the output sink and source current for average processing, nominal supply voltage and room
temperature. For additional data see the Xilinx IBIS files at: http://www.xilinx.com/techdocs/htm_index/sw_ibis.htm
Note: The Xilinx IBIS files also list the extremes of the processing, temperature and voltage conditions as “min” and “max”
(delay). Some tools rename these values as “fast-strong” and “slow-weak”. This has nothing to do with the FAST or SLOW
slew-rate configuration options, invoked as attributes for each output pin (the default is SLOW).
These slew-rate options are covered in separate IBIS files, with a trailing s for SLOW slew rate, and f for FAST slew rate.
Virtex Series
LVTTL LVCMOS2
100 100
90 90
80 80
IOL
70 70
IOL
60 60
mA
mA
50 50
IOH 40
40
30 30
IOH
20 20
10 10
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Volts Volts 98122804
98122805
SSTL2I
SSTL3I
50 50
45 45
40 40
35 35
30 30 IOL
mA
mA
25 25
20 20
IOH
15 15
10 10
5 5 14
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Volts 98122808
Volts
98122810

R
Virtex I/V Curves for Various Output Options
SSTL3II SSTL2II
100 100
90 90 IOL
80 IOL 80
70 70
60 60
mA
mA
50 50
IOH
40 IOH
40
30 30
20 20
10 10
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Volts 98122811 Volts 98122809
GTL+
GTL
240
240
IOL 220
220
200
200
180
180
160
160 IOL
140
140
mA
mA
120
120
100
100
80
80
60
60
40
40
20
20
0
0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Volts 98122803
Volts 98122802
14

APPLICATION NOTE
0
R
I/V Curves for Various
Device Families
January 4, 1999 (Version 1.0) 0 14* Application Note
These typical curves describe the output sink and source current for average processing, nominal supply voltage and room
temperature. For additional data see the Xilinx IBIS files at: http://www.xilinx.com/techdocs/htm_index/sw_ibis.htm
Note: The Xilinx IBIS files also list the extremes of the processing, temperature and voltage conditions as “min” and “max”
(delay). Some tools rename these values as “fast-strong” and “slow-weak”. This has nothing to do with the FAST or SLOW
slew-rate configuration options, invoked as attributes for each output pin (the default is SLOW).
These slew-rate options are covered in separate IBIS files, with a trailing s for SLOW slew rate, and f for FAST slew rate.
XC4000 Series
XC4000XV
XC4000E
200
200
180
180
160
160
140
140
IOL 120
120 IOL
mA
mA
100
100
80
80
IOH CMOS
60 IOH
60
40
40
20
20
IOH TTL
0
0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Volts 98122817
Volts 98122812
XC4000XL XC4000XLA
200 200
180 180
160 160
140 140
120 120 IOL

mA
IOL
100 100
80 80
60 60
IOH
40 IOH 40
20 20
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 14
Volts 98122816
Volts 98122813

R
I/V Curves for Various Device Families
Spartan
SPARTAN SPARTANXL
200 200
180 180
160 160
IOL
140 140
IOL
120 120
mA
mA
100 100
80 80
IOH CMOS
60 60
IOH
40 40
20 TTL
20
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Volts 98122806 Volts 98122807
XC9500 Series
XC9500 XC9500XL
200 100
180 90
IOL
160 80
IOL
140 70
120 60
mA
mA
100 50
80 40
IOH
60 30
40 20
IOH
20 10
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Volts 98122815
Volts 98122814
14

Index
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

0

Book Index
0 15*
Numerics INIT 4-18, 6-18

LOC 6-38, 6-40, 6-42
3-state buffer MEDDELAY 6-25, 6-26
Spartan Series 4-20 NODELAY 4-8, 6-25, 7-92
XC3000 Series 7-8, 7-17, 14-23 available products
XC4000 Series 6-29 XC1700D 8-12, 8-25, 8-35
XC5200 7-90 XC3000 Series 7-77
3-state net, global. See Global 3-State XC4000 Series 6-154, 6-173, 6-174
5-input function XC9500 5-6, 5-60
XC3000 Series 14-19
XC5200 7-87
70% rule 14-22
B
bar code 11-18
A Base System
Foundation Series (PC) 2-4, 2-5, 2-6, 2-7
A Simple Method of Estimating Power in XC4000XL/EX/E basic moisture life test 12-4
FPGAs 14-13 battery backup in XC3000 Series 14-27, 14-42
absolute maximum ratings BCLKIN 14-24
overshoot and undershoot 14-50 pin description
specifications XC3000 Series 7-39
XC1700D 8-8, 8-21, 8-32 BGA packages
XC3000A 7-42 BG225 package drawings 11-38
XC3000L 7-48 BG256 package drawings 11-39
XC3100A 7-54 BG352, BG432 package drawings 11-40
XC3100L 7-60 BG560 package drawings 11-41
XC4000E 6-100, 10-12 bidirectional (bidi) buffer
XC5200 7-127 XC3000 Series 7-11
ACLK symbol 14-24 bitstream
adaptor selector for programmer 8-38 combining for daisy chain 4-27, 6-50, 7-105
address pins combining with MakePROM 7-20
pin descriptions copyrighting 14-42
XC3000 Series 7-39 format for configuration
XC4000 Series 6-45, 7-103 Spartan Series 4-30, 6-170
advance specifications, definition of 1-1, 6-100, 7-127 XC4000 Series 6-51, 7-107
algorithms, programmer 8-37 boundary scan
Answers database 13-2 access to 14-56
applications Information 13-2 avoiding inadvertent activation 4-23, 6-48, 7-101
AppLINX CD 13-2 bypass register 14-55
asynchronous peripheral configuration mode configuration with 4-34, 6-57, 7-112
debugging hints 14-36 daisy chain configuration 14-59
data register 14-53
15
specifications
XC3000 Series 7-30 effect on GTS 4-21, 6-27, 7-92
XC4000 Series 6-70 implementing in schematic 4-23, 6-47, 7-100
XC3000 Series 7-29 Spartan Series 4-23
XC4000 Series 6-45, 6-69, 7-103, 7-120 XC4000 Series 6-47, 7-100
asynchronous RAM 6-17 library symbol 4-23, 4-57, 6-44, 6-47, 7-100, 7-103
attribute pin descriptions
DECODE 6-30 Spartan Series 4-57
1
Book Index
XC4000 Series 6-43, 7-102 setting frequency

Spartan Series 4-21 Spartan Series 4-30
specifications XC4000 Series 6-51, 7-106
XC4000E 6-110, 6-114, 7-132 use as configuration clock
TAP controller 14-53 Spartan Series 4-26
test access port (TAP) 14-53 XC3000 Series 7-22
user registers 14-55 XC4000 Series 6-49, 7-104
XC4000 Series 6-29, 6-45, 7-93, 14-51 ceramic packages
XC5200 7-98, 14-51 DIP
XC9500 5-16, 5-70 DD8 package drawings 11-24
Boundary Scan Description Language. See BSDL CLB. See configurable logic block
BSCAN symbol 4-23, 4-57, 6-44, 6-47, 7-100, 7-103 clock diagram
BSDL files Spartan 4-14
Spartan Series 4-23 XC4000E 6-39
XC4000 Series 6-47, 7-100, 14-61 XC4000EX 6-39
XC5200 14-61 CMOS input
buffer, 3-state XC3000 Series 7-8
Spartan Series 4-20 XC4000 Series 6-23, 7-91
XC3000 Series 7-8, 7-17, 14-23 XC5200 7-91
XC4000 Series 6-29 CMOS output
XC5200 7-90 Spartan Series 4-10
buffer, bidirectional XC3000 Series 7-8, 14-22
XC3000 Series 7-11 XC4000 Series 6-26, 14-18
buffered switch matrix 6-34 XC5200 7-91
BUFG symbol 7-96 configurable logic block (CLB)
BUFGE symbol 6-42, 6-44 block diagram
BUFGLS symbol 6-40, 6-44 XC3000 Series 7-9
BUFGP symbol 4-15, 6-38 XC4000 Series 6-12
BUFGS symbol 4-15, 6-38 XC5200 7-87
BUFT symbol 6-29 carry logic
bus contention, internal 14-23 Spartan Series 4-18
bypass register 14-55 XC4000 Series 6-20
XC5200 7-88
cascade logic
C XC5200 7-89
flip-flop
carry logic
XC3000 Series 7-9, 14-21
Spartan Series 4-18
XC4000 Series 6-12, 7-89
XC4000 Series 6-20
XC5200 7-87
XC5200 7-88
function generator
cascade logic
XC3000 Series 7-10, 14-19
XC5200 7-89
XC4000 Series 6-11
CBQFP packages
XC5200 7-87
CB100 package drawings 11-50
latch
CB100, CB164, CB196 package drawings 11-52
XC4000EX 6-12, 7-89
XC5200 7-87
RAM 6-14
CCLK
routing associated with
frequency variation
XC3000 Series 14-21
XC3000 Series 14-27
XC4000 Series 6-31
in configuration debug 14-37
Spartan Series 4-5
low-time restriction in XC3000 Series 14-27
specifications
pin description
XC3000A 7-43
Spartan Series 4-56
XC3000L 7-49
XC3000 Series 7-38
XC3100A 7-55
XC4000 Series 6-43, 7-102
2
XC3100L 7-61 XC4000 Series specifications 6-64, 7-115
XC5200 7-129 memory cell 7-6, 14-40
XC3000 Series 7-9, 14-19 mode
XC4000 Series 6-11 selection of 14-34
XC5200 7-86, 7-87 modes, table of
differences from XC4000 and XC3000 7-84 XC3000 Series 7-19
configuration 14-44 XC4000 Series 6-49, 7-104
asynchronous peripheral mode peripheral mode
debugging hints 14-36 XC3000 Series 7-29
XC3000 Series 7-29 XC3000 Series specifications 7-30
XC3000 Series specifications 7-30 peripheral modes, general
XC4000 Series 6-45, 6-69, 7-103, 7-120 debugging hints 14-37
XC4000 Series specifications 6-70 XC3000 Series 7-22
bitstream copyrighting 14-42 XC4000 Series 6-43, 6-49, 7-102, 7-105
bitstream format pin descriptions
Spartan Series 4-30, 6-170 Spartan Series 4-56
XC4000 Series 6-51, 7-107 XC3000 7-38
boundary scan pins, using 4-34, 6-57, 7-112, 14-59 XC4000 Series 6-43, 7-102
clock. See CCLK pin functions during
configuration sequence Spartan Series 4-25, 6-168
Spartan Series 4-32 XC3000 Series 7-40
XC4000 Series 6-53, 7-108 XC4000 Series 6-61, 6-62
control pins 14-39 XC5200 7-124
daisy chain power-on reset
debugging hints 14-36 XC3000 Series 7-19
mixed family 6-50, 7-105, 14-38 reducing time 14-45
Spartan Series 4-27 slave serial mode
XC3000 Series 7-22 debugging hints 14-36
XC4000 Series 6-49, 7-105 Spartan Series 4-27, 4-56
debugging 14-34 Spartan Series specifications 4-28
express mode XC3000 Series 7-22, 7-31
daisy chain 4-28, 6-168, 7-106 XC3000 Series specifications 7-32
XC4000 Series specifications 4-30, 6-170 XC4000 Series 6-43, 6-49, 6-63, 7-102, 7-105,
XC4000EX 4-28, 6-167, 7-102, 7-106, 7-107, 7-114
7-111, 7-122 XC4000 Series specifications 6-63
initiating reconfiguration 14-45 Spartan Series 4-25
length count 4-30, 6-51, 6-57, 7-20, 7-107, 7-111 specifications
master modes, general XC3000 Series 7-25
XC3000 Series 7-19, 7-22 start-up sequence
XC4000 Series 6-49, 6-57, 7-104, 7-111 XC3000 Series 14-28
XC4000 Series specifications 4-37, 6-71 XC4000 Series 6-54, 7-110
XC5200 specifications 7-125 switching characteristics
master parallel mode XC4000 Series 4-37
debugging hints 14-35 XC5200 7-125
XC3000 Series 7-27 synchronous peripheral mode
XC3000 Series specifications 7-28 XC4000 Series 6-43, 6-67, 7-102, 7-118
XC4000 Series 6-43, 6-65, 7-102, 7-115 XC4000 Series specifications 6-68
XC4000 Series specifications 6-66, 7-117 XC3000 Series 7-19, 14-28 15
master serial mode XC4000 Series 6-48, 7-104
debugging hints 14-35 XC5200
Spartan Series 4-26 differences from XC4000 and XC3000 7-84
Spartan Series specifications 4-26 copyrighting bitstream 14-42
XC3000 Series 7-25 CPLD
XC3000 Series specifications 7-26 overview 1-2, 1-4, 14-9
XC4000 Series 6-64, 7-115 product selection guide 14-5, 14-7, 14-10
3
Book Index
CQ100 package of configuration after power-up

pinout table Spartan Series 4-34
CRC error checking optional input delay
XC4000 Series 6-52, 7-107 output
crystal oscillator XC4000 Series 14-16
XC3000 Series 7-18, 7-24, 14-25 with fast capture latch 6-26
CS design security
in configuration debug 14-35 XC9500 5-16, 5-70
CS0, CS1 device literature 13-3
pin descriptions DIN
XC4000 Series 6-45, 7-103 in daisy chain
CS0, CS1, CS2 Spartan Series 4-27
pin descriptions XC3000 Series 7-22
cyclic redundancy check (CRC) pin description
XC4000 Series 6-52, 7-107 XC3000 Series 7-39
XC4000 Series 6-45, 7-103
DIP package
D ceramic 11-24
plastic 11-25
daisy chain
direct interconnect
creating bitstream 7-20
XC3000 Series 7-11
debugging hints 14-36
XC4000EX 6-35
express mode
XC5200 7-86, 7-94
XC4000EX 4-28, 6-168, 7-106
disk space requirements
mixed family 6-50, 7-105, 14-38
programmer 8-37
Spartan Series 4-27
DONE
XC3000 Series 7-22
during power-up 14-40
XC4000 Series 6-49, 7-105
express mode configuration 7-106
data integrity 12-10
going High after configuration
data pins
XC3000 Series 7-23
pin descriptions
XC4000 Series 6-57, 7-111
XC3000 Series 7-39
XC4000 Series 6-45, 7-103
not going High after configuration 14-34
data register 14-53
pin description
data retention, XC9500 5-77, 5-101
Spartan Series 4-56
data stream. See bitstream
XC4000 Series 6-43, 7-102
DC characteristics
DONE/PROG
specifications
XC3000A 7-41
pin description
XC3000L 7-47
XC3000 Series 7-38
XC3100A 7-53
double-length routing
XC3100L 7-59
Spartan Series 4-14
XC4000E 4-39, 6-101, 10-13
XC4000 Series 6-34
XC5200 7-127
XC5200 7-96
DD8 package
DOUT
package drawing 11-24
in daisy chain
debugging
express mode 7-106
configuration 14-34
Spartan Series 4-27
DECODE attribute 6-30
XC3000 Series 7-22
decode logic. See cascade logic, edge decoder
XC4000 Series 6-49, 7-105
delay
pin description
input delay 14-49
Spartan Series 4-58
XC3000 Series 14-21
4
XC4000 Series 6-45, 7-103 in macrocell 5-62
dry bag 11-16 forced air cooling vendors 11-11
dry bake 11-16 Foundation Series 2-3
dual-port RAM 4-16, 6-17 Base System (PC) 2-4, 2-5, 2-6, 2-7
FPGA
advantages of 14-7
E data integrity 12-10
overview 1-2, 1-4
edge decoder 6-30
product selection guide 1-6, 1-7, 1-8, 14-5, 14-7,
edge-triggered RAM 6-15
14-10
advantages of 4-18, 6-14
security 14-41
EIAJ standards 11-1
function block
electrical parameters
XC9500 5-7, 5-61
programmer 8-37
function generator
electrostatic discharge (ESD) 12-11
in CLB
endurance, XC9500 5-16, 5-70
XC3000 Series 7-10, 14-19
error checking, bitstream 4-31, 6-52, 7-24, 7-107
XC4000 Series 6-11
Spartan Series 4-31
XC5200 7-87
XC3000 Series 7-24
in IOB
XC4000 Series 6-52, 7-107
XC4000EX 4-9, 6-28
express configuration mode
using as RAM 6-14
CRC not supported 7-107
specifications
XC4000 Series 4-30, 6-170 G
synchronized to DONE
XC4000EX 7-111 gate array
XC4000EX 4-28, 6-167, 7-102, 7-106, 7-107, 7-111, advantages of FPGAs 1-3
7-122 GCK1 - GCK8
clock diagram 6-39
pin descriptions
F XC4000EX 6-44
GCLK symbol 14-24
factory floor life 11-16
general routing matrix (GRM)
failure analysis 12-6, 12-9
XC5200 7-85, 7-86, 7-93, 7-94
failures in time 12-4
glitch
fast capture latch 6-26
avoidance in XC3000 Series 14-20
fast carry logic. See carry logic
power supply 14-40
FastCONNECT switch matrix 5-67
Global 3-State (GTS)
FastFlash technology
Spartan Series 4-21
XC9500 5-19, 5-73
XC4000 Series 6-27, 6-42, 7-92
FCLK1 - FCLK4
global buffer
clock diagram 6-39
Spartan 4-14
FDCE symbol 6-13, 7-90
specifications
FIFO
XC3000A 7-42
implementing in XC4000 Series RAM 6-14
XC3000L 7-48
FITs 12-4
XC3100A 7-54
flip-flop
XC3100L 7-60
in CLB
XC3000 Series 7-9, 14-21
XC5200 7-128 15
XC3000 Series 7-17, 14-24
XC4000 Series 6-12, 6-13, 7-89, 7-90
XC4000 Series 6-38
XC5200 7-87
XC4000E 6-38
in IOB
XC4000EX 6-40
metastability 14-46
XC5200 7-96
none in XC5200 7-91
Global Early buffer (BUFGE) 6-28, 6-41, 6-44
Spartan Series 4-9, 4-11
fast pin-to-pin path 4-10, 6-28
XC3000 Series 7-8, 14-21
with fast capture latch 6-26
5
Book Index
Global Low-Skew buffer (BUFGLS) 6-40, 6-44 IFD symbol 4-8, 6-23
Global Set/Reset (GSR) ILD symbol 4-8, 6-23
in CLB ILFFX symbol 6-26
Spartan Series 4-21 ILFLX symbol 6-26
XC4000 Series 6-13, 7-90 INIT
in IOB during power-up 14-40
Spartan Series 4-11 in configuration debug 14-37
XC4000 Series 6-28 in daisy chain 14-38
GRM. See general routing matrix pin description
ground bounce Spartan Series 4-57
XC4000 Series 14-16 XC3000 Series 7-38
GSR. See Global Set/Reset XC4000 Series 6-44, 7-103
GTS. See Global 3-state INIT attribute 4-18, 6-18
initializing RAM 4-18, 6-18
input/output block (IOB)
H clock
XC3000 Series 14-23
HardWire
CMOS input
overview 1-4
Spartan Series 4-8
HDC
XC3000 Series 7-8
XC4000 Series 6-23, 7-91
pin description
XC5200 7-91
Spartan Series 4-57
CMOS output
XC3000 Series 7-22, 7-38
Spartan Series 4-10
XC4000 Series 6-44, 7-103
XC3000 Series 7-8, 14-22
heatsink vendors 11-11
XC4000 Series 6-26, 14-18
hermeticity test 12-6
XC5200 7-91
high temperature life test 12-4, 12-11
delay on input 14-49
high-reliability (Hi-Rel)
XC3000 Series 14-21
overview 1-5
XC4000 Series 6-25, 7-91
hold time on data input 14-49
during configuration
XC4000 Series 6-25, 7-91
Spartan Series 4-56
hotline support 13-2
XC3000 Series 7-23
HQ packages 11-14
XC4000 Series 6-42, 7-101
thermal data 11-8
fast capture latch 6-26
HQFP packages
flip-flop
HQ100 package drawings 11-34
metastability 14-46
HQ160, HQ208, HQ240 package drawings 11-35
function generator on output 4-9, 6-28
HQ304 package drawings 11-36
maximum available I/O
HTQFP packages
XC3000 Series 7-65
HT100, HT144, HT176 package drawings 11-33
XC5200 7-154
HW-130 programmer 8-37
multiplexer on output 4-9, 6-28
hysteresis
optional delay with fast capture latch 6-26
XC3000 Series 14-21
pull-down resistor
Spartan Series 4-10
I XC4000 Series 6-28, 7-92
pull-up resistor
I/O block Spartan Series 4-10
see also input/output block XC3000 Series 7-8
see also input/output cell XC4000 Series 6-28, 7-92
XC9500 5-68 rise/fall time
I/O count XC3000 Series 14-21
XC3000 Series 7-65 routing associated with
XC5200 7-154 Spartan Series 4-14
I/V characteristics. See V/I characteristics XC4000 Series 6-35
IEEE Standard 1149.1 14-61 XC5200 7-86, 7-98
6
slew rate control none in XC5200 7-91
XC4000 Series 6-27, 7-92 XC4000 Series 6-23
XC5200 7-91 latchup 12-11
Spartan Series 4-7 LC. See logic cell
specifications LDC
XC3000A 7-45 during power-up 14-40
XC3000L 7-51 in configuration debug 14-35
XC3100A 7-57 pin description
XC3100L 7-63 Spartan Series 4-57
XC5200 7-131 XC3000 Series 7-22, 7-38
TTL input XC4000 Series 6-44, 7-103
Spartan Series 4-8 LDCE symbol 6-13, 7-90
XC3000 Series 7-8 lead fatigue test 12-5
XC4000 Series 6-23, 7-91 length count
XC5200 7-91 configuration debugging hints 14-37
TTL output Spartan Series 4-30
none in XC5200 7-91 XC3000 Series 7-20
Spartan Series 4-10 XC4000 Series 6-51, 6-57, 6-69, 7-107, 7-111, 7-120
XC3000 Series 7-8 level-sensitive RAM 6-17
XC4000 Series 6-26 library symbol
unused I/O 3-state buffer
Spartan Series 4-56 BUFT, XC4000 Series 6-29
XC4000 Series 6-42, 7-101 WAND1 6-29
VersaRing WOR2AND 6-29
Spartan Series 4-14 AND-gate in IOB
XC4000 Series 6-35 OAND2 4-10, 6-28
XC5200 7-86, 7-98 boundary scan
XC3000 Series 7-7, 14-21 BSCAN 4-23, 4-57, 6-44, 6-47, 7-100, 7-103
clock 14-23 TCK, TDI, TDO, TMS 4-23, 4-57, 6-44, 6-47,
XC4000 Series 6-23, 7-91 7-100, 7-103
XC5200 7-91 fast capture latch
XC9500 5-68 ILFFX 6-26
in-system programming ILFLX 6-26
XC9500 5-16, 5-70 flip-flop
interconnect. See routing FDCE, XC4000 Series CLB 6-13, 7-90
ISO9002 1-5, 12-3 IFD, Spartan Series IOB 4-8
IFD, XC4000 Series IOB 6-23
Global 3-State
J STARTUP 4-21, 6-27, 7-92
global buffer
JEDEC standards 11-1
ACLK, XC3000 Series 14-24
JTAG. See boundary scan
BUFG, XC5200 7-96
junction temperature
BUFGE, XC4000EX 6-42, 6-44
junction-to-ambient 11-4
BUFGLS, XC4000EX 6-40, 6-44
junction-to-case 11-4
BUFGP, Spartan 4-15
BUFGP, XC4000E 6-38
15
L BUFGS, Spartan 4-15
BUFGS, XC4000E 6-38
latch GCLK, XC3000 Series 14-24
fast capture latch 6-26 Global Set/Reset
in CLB STARTUP 4-21, 6-13, 7-90
XC4000EX 6-12, 6-13, 7-89, 7-90 latch
XC5200 7-87 ILD, Spartan Series IOB 4-8
in IOB ILD, XC4000 Series IOB 6-23
7
Book Index
LDCE, XC4000EX CLB 6-13, 7-90 manufacturers. See vendors

mode pins mass, package 11-13
MD0, MD1, MD2 6-48, 7-104 master configuration modes, general
oscillator Spartan Series 4-26
OSC, Spartan Series 4-21 XC3000 Series 7-19, 7-22
OSC4, XC4000 Series 6-30 XC4000 Series 6-49, 6-57, 7-104, 7-111
OSC52, XC5200 7-93 XC4000 Series specifications 4-37, 6-71
output multiplexer XC5200 specifications 7-125
OMUX2 4-10, 6-28 master parallel configuration mode
readback debugging hints 14-35
READBACK 4-35, 6-58, 7-113 specifications
resistor XC3000 Series 7-28
PULLDOWN 4-10, 6-28, 7-93 XC4000 Series 6-66, 7-117
PULLUP 4-10, 6-28, 7-93 XC3000 Series 7-27
wide decoder XC4000 Series 6-43, 6-65, 7-102, 7-115
WAND1 6-30 master serial configuration mode
LIM. See local interconnect matrix debugging hints 14-35
literature, technical 13-3 Spartan Series 4-26
LOC attribute 6-38, 6-40, 6-42 specifications
local interconnect matrix (LIM) Spartan Series 4-26
XC5200 7-86, 7-93 XC3000 Series 7-26
local phone support 13-2 XC4000 Series 6-64, 7-115
logic cell (LC) XC3000 Series 7-25
XC5200 7-84, 7-85 XC4000 Series 6-64, 7-115
longline maximum I/O
specifications XC5200 7-154
XC5200 7-128 MD0, MD1, MD2 symbols 6-48, 7-104
XC3000 Series 7-15, 14-21, 14-23 Mean Time Between Failures (MTBF)
XC4000 Series 6-34 metastability 14-47
XC5200 7-96 MEDDELAY attribute 6-25, 6-26
lookup table (LUT). See function generator memory cell, configuration 7-6, 12-10, 14-40
low voltage device memory requirements
XC4000 Series 6-7 programmer 8-37
metastability 14-46
MIL-STD-883B 1-5
M mode pins
pin descriptions
M0, M1, M2. See mode pins
XC4000 Series 6-43, 7-102
M0/RTRIG
XC3000 Series 7-19
pin description
XC4000 Series 6-48, 7-104
XC3000 Series 7-38
moisture sensitivity in surface mount packages 11-15
M1/RDATA
multiplexer
pin description
XC4000EX IOB 4-9, 6-28
XC3000 Series 7-38
M2
pin description N
XC3000 Series 7-38
macrocell newsletter, XCell 13-3
XC9500 5-8, 5-62 NODELAY attribute 4-8, 6-25, 7-92
MakeBits program
bitstream generation
XC3000 Series 7-21 O
crystal oscillator selection 7-24
OAND2 symbol 4-10, 6-28
express mode 4-28, 6-168
octal routing 6-35
MakePROM program
OE
combining bitstreams 7-20
8
in configuration debug 14-35 DD8 package 11-24
OMUX2 symbol 4-10, 6-28 HQ100 package 11-34
on-chip oscillator. See oscillator, on-chip HQ160, HQ208, HQ240 package 11-35
open-drain output 6-27, 7-92 HQ304 package 11-36
operating conditions HT100, HT144, HT176 package 11-33
specifications PC20, PC28, PC44, PC68, PC84 package 11-28
XC1700D 8-8, 8-21, 8-32 PD8 package 11-25
XC3000A 7-41 PG120, PG132, PG156 package 11-43
XC3000L 7-47 PG175 package 11-44
XC3100A 7-53 PG191 package 11-45
XC3100L 7-59 PG223, PG299 package 11-46
XC4000E 4-38, 6-100, 10-12 PG411 package 11-47
XC5200 7-127 PG475 package 11-48
ordering information PG68, PG84 package 11-42
Spartan Series 69 PQ100 package 11-34
XC1700D 8-12, 8-25, 8-35 PQ304 package 11-36
XC4000 Series 6-115, 6-154, 6-167, 6-171, 6-254 PQ44, PQ160, PQ208, PQ240 package 11-35
XC5200 7-155 SO20 package 11-27
OSC symbol 4-21 SO8 package 11-26
OSC4 symbol 6-30 TQ100, TQ144, TQ176 package 11-33
OSC52 symbol 7-93 VO8 package 11-26
oscillator VQ44, VQ64, VQ100 package 11-32
crystal packaging
XC3000 Series 7-18, 7-24, 14-25 bar code 11-18
on-chip 14-40 data acquisition 11-5
Spartan Series 4-21 dimensions 11-1
XC3000 Series 14-27 dry bag 11-16
XC4000 Series 6-30 dry bake 11-16
XC5200 7-85, 7-93 EIAJ standards 11-1
output current EIJ standard board layout 11-3
XC3000 Series 14-23 factory floor life 11-16
XC4000 Series 6-26, 6-42, 7-92, 7-101, 14-15 handling and storage 11-16
XC5200 7-91 JEDEC standards 11-1
output multiplexer in XC4000EX IOB 4-9, 6-28 mass 11-13
output slew rate moisture sensitivity 11-15
Spartan Series 4-10 orientation 11-3
XC3000 Series 7-8 reflow soldering 11-19
XC4000 Series 6-27, 7-92 tape & reel packing 11-17
XC5200 7-91 thermal characteristics 11-4
overshoot 14-50 thermal database 11-5
thermal management 11-3
thermal resistance
P applying data 11-7
table of 11-6
package availability
thermally enhanced 11-14
XC3000 Series 7-77
vendors 11-11, 11-21
XC4000 Series 6-154, 6-173, 6-174
weight 11-13
XC9500 5-61
PC44 package
package drawing
pinout table 15
BG225 package 11-38
XC3000 Series 7-65
BG256 package 11-39
PC68 package
BG352, BG432 package 11-40
pinout table
BG560 package 11-41
XC3000 Series 7-67
CB100 package 11-50
PC84 package
CB100, CB164, CB196 package 11-52
pinout table
CB164 package 11-51
XC3000 Series 7-68
CB228 package 11-53
9
Book Index
PCI compatibility device-specific

PD8 package XC3000 Series 7-76
package drawing 11-25 XC4000 Series 6-115
performance package-specific
XC5200 7-87 PC44 package
peripheral configuration mode XC3000 Series 7-65
specifications PC68 package
XC3000 Series 7-29 PC84 package
peripheral configuration modes, general XC3000 Series 7-68
debugging hints 14-37 PG132 package
specifications XC3000 Series 7-70
XC4000 Series 4-37, 6-71 PG175 package
XC5200 7-125 XC3000 Series 7-73
XC3000 Series 7-22 PG84 package
XC4000 Series 6-43, 6-49, 7-102, 7-105 XC3000 Series 7-67
peripheral configuration modes. See also asynchronous PQ100 package
peripheral, synchronous peripheral XC3000 Series 7-69
PG132 package PQ160 package
pinout table XC3000 Series 7-72
XC3000 Series 7-70 PQ208 package
PG175 package XC3000 Series 7-75
pinout table Spartan Series 4-58
XC3000 Series 7-73 TQ100 package
PG84 package XC3000 Series 7-69
pinout table TQ144 package
PGA packages TQ176 package
PG120, PG132, PG156 package drawings 11-43 XC3000 Series 7-74
PG175 package drawings 11-44 VQ100 package
PG191 package drawings 11-45 XC3000 Series 7-69
PG223, PG299 package drawings 11-46 VQ64 package
PG68, PG84 package drawings 11-42 XC3195 7-76
PGCK1 - PGCK4 XC4000 Series 6-115
clock diagram 4-14, 6-39 XC4003E 6-116
pin descriptions 4-57, 6-44 XC4005E/L 6-117
phone support 13-2 XC4006E 6-118
pin description XC4008E 6-120
Spartan Series 4-56 XC4010E/L 6-121
pin descriptions XC4013E/L 6-123
functions during configuration XC4020E 6-126
Spartan Series 4-25, 6-168 XC4025E 6-128, 6-131
XC3000 Series 7-40 XC4028EX/XL 6-128, 6-131
XC4000 Series 6-61, 6-62 XC4044EX/XL 6-134
XC5200 7-124 XC4052XL 6-138
XC3000 Series 7-38 XC5202 7-133
XC4000 Series 6-42, 7-101 XC5204 7-135
pin locking XC5206 7-139
XC9500 5-15, 5-69 XC5210 7-142
pinout table XC5215 7-147
CQ100 package XC95108 5-96
XC3000 Series 7-69 XC95144 5-104
10
XC95288 5-114 CPLD 14-5, 14-7, 14-10
XC9536 5-80 FPGA 1-6, 1-7, 1-8, 14-5, 14-7, 14-10
XC9572 5-88 product term allocator
XCS05 4-58 XC9500 5-10, 5-64
XCS10 4-59 PROGRAM
XCS20 4-61 during power-up 14-40
XCS30 4-62 inititating reconfiguration 14-45
XCS40 4-65 pin description
pin-to-pin specifications Spartan Series 4-56
XC5200 7-130 XC4000 Series 6-43, 7-102
plastic packages program cycles, XC9500 5-77, 5-101
DIP programmable switch matrix (PSM)
PD8 package drawings 11-25 Spartan Series 4-13
PLCC packages XC3000 Series 7-11
PC20, PC28, PC44, PC68, PC84 package XC4000 Series 6-32
drawings 11-28 programmer 8-37
power consumption algorithms 8-37
reduction of 11-10 software 8-37
XC3000 Series 7-37, 14-24 specifications 8-37
XC4000 Series 14-18 programming, in-system
power distribution XC9500 5-16, 5-70
XC3000 Series 7-36 programming. See configuration
XC4000 Series 6-42, 7-101 PROM
power-down mode bitstream generation
none in XC4000 Series 6-42 Spartan Series 4-27
none in XC5200 7-85 XC4000 Series 6-50, 7-105
XC3000 Series 7-19, 7-37, 14-27, 14-42 configuration
power-on reset Spartan Series
XC3000 Series 7-19 4-4
power-up 14-40 XC3000 Series 7-25, 7-27
power-up. See also start-up after configuration XC4000 Series 6-8, 6-64, 6-65, 7-115
PQ packages in configuration debug 14-35
thermal data 11-8 overview 1-4
PQ100 package programmer 8-37
pinout table size
XC3000 Series 7-69 Spartan 4-32
PQ160 package XC4000E 6-52
pinout table pseudo daisy chain for express mode
XC3000 Series 7-72 XC4000EX 4-28, 6-168, 7-106
PQ208 package pull-down resistor
pinout table IOB
PQFP packages XC4000 Series 6-28, 7-92
PQ100 package drawings 11-34 PULLDOWN symbol 4-10, 6-28, 7-93
PQ304 package drawings 11-36 pull-up resistor
PQ44, PQ160, PQ208, PQ240 package IOB
drawings 11-35 Spartan Series 4-10, 4-56
preliminary specifications, definition of 1-1, 6-100, 7-127 XC3000 Series 7-8
Primary Global Buffer (BUFGP) 4-14, 4-57, 6-38, 6-44 XC4000 Series 6-28, 6-43, 7-92, 7-102 15
product availability longline
Spartan Series 68 none in XC5200 7-96
XC4000 Series 6-154, 6-173, 6-174 XC4000 Series 6-34
XC9500 5-6, 5-60 PULLUP symbol 4-10, 6-28, 7-93
product qualification requirements 12-7 PWRDWN 14-27
product selection guide battery backup mode 14-42
11
Book Index
in configuration debug 14-37 pin description

pin description XC3000 Series 7-38
XC3000 Series 7-38 rise time requirement in XC3000 Series 14-29
reset
Spartan Series
Q See also Global Set/Reset
XC3000 Series 7-24
quad routing 6-34
XC4000 Series 6-42
qualification requirements 12-7
See also Global Set/Reset
quality assurance 12-3
resistance to solvents test 12-5
resistor
R IOB
Spartan Series 4-10
RAM XC3000 Series 7-8
asynchronous 6-17 XC4000 Series 6-28, 7-92
configuration options pull-up on longline
XC4000 Series 6-14 none in XC5200 7-96
dual-port 4-16, 6-17 XC3000 Series 7-17
edge-triggered 6-15 XC4000 Series 6-34
in CLB 6-14 with crystal oscillator 14-26
initialization of 4-18, 6-18 rise/fall time on input
level-sensitive 6-17 XC3000 Series 14-21
readback of contents 4-35, 6-59 routing
setting mode 6-14 bidirectional (bidi) buffer 7-11
synchronous 6-15 buffered switch matrix 6-34
RCLK direct interconnect
pin description XC3000 Series 7-11
XC3000 Series 7-39 XC4000EX 6-35
XC4000 Series 6-43, 7-102 XC5200 7-86, 7-94
RDY/BUSY double-length
pin description Spartan Series 4-14
XC4000 Series 6-43, 7-102 XC5200 7-96
readback global buffer
CRC error checking Spartan 4-14
Spartan Series 4-31 XC4000E 6-38
XC4000 Series 6-52, 7-107 XC4000EX 6-40
Spartan Series 4-35, 4-56 longline
specifications Spartan Series 4-14
Spartan 4-36 XC3000 Series 7-15, 14-21, 14-23
XC4000E 6-60, 7-126 XC5200 7-96
XC3000 Series 7-23 octal 6-35
XC4000 Series 6-43, 6-58, 7-102, 7-112 programmable switch matrix (PSM)
XChecker cable 4-36, 6-59, 7-113 Spartan Series 4-13
READBACK symbol 4-35, 6-58, 7-113 XC3000 Series 7-11
reconfiguration XC4000 Series 6-32
XC4000 Series 6-8 quad 6-34
reconfiguration. See also configuration single-length
reflow soldering 11-19 Spartan Series 4-13
reliability 12-4 XC4000 Series 6-32
RESET XC5200 7-96
during power-up 14-40 Spartan Series 4-11
in configuration debug 14-37 VersaRing (IOB routing)
in daisy chain 14-38 Spartan Series 4-14
inititating reconfiguration 14-45 XC4000 Series 6-35
12
XC5200 7-86, 7-98 programmer 8-37
XC3000 Series 7-10, 14-21 SOIC packages
XC4000 Series 6-31 package drawings 11-26
XC5200 7-86, 7-94 SO20 package drawings 11-27
differences from XC4000 and XC3000 7-84 solderability test 12-5
RS Spartan
pin description clock diagram 4-14
XC4000 Series 6-45, 7-103 global buffer 4-14
Primary Global Buffer (BUFGP) 4-14
Secondary Global Buffer (BUFGS) 4-14
S Spartan Series 4-31
3-state buffer 4-20
salt atmosphere test 12-5
bitstream format 4-30, 6-170
Secondary Global Buffer (BUFGS) 4-14, 4-57, 6-38, 6-44,
boundary scan 4-21
7-103
BSDL files 4-23
security 5-16, 5-70, 14-41
carry logic 4-18
selection guide
CMOS input 4-8
FPGA 1-6, 1-7, 1-8
CMOS output 4-10
service
configurable logic block (CLB) 4-5
overview 1-5
configuration 4-25
setup time on data input 14-47, 14-49
input/output block (IOB) 4-7
XC3000 Series 14-21
interconnect 4-11
SGCK1 - SGCK4
on-chip oscillator 4-21
clock diagram 4-14, 6-39
ordering information 69
pin description 4-57, 6-44, 7-103
pin description 4-56
single-length routing
pinout tables 4-58
Spartan Series 4-13
product availability 68
XC4000 Series 6-32
readback 4-35
XC5200 7-96
routing 4-11
sink current. See output current
soft startup 4-10
slave serial configuration mode
TTL input 4-8
debugging hints 14-36
TTL output 4-10
Spartan Series 4-27, 4-56
specifications
specifications
absolute maximum ratings
Spartan Series 4-28
XC1700D 8-8, 8-21, 8-32
XC3000 Series 7-32
XC3000A 7-42
XC4000 Series 4-37, 6-63, 6-71
XC3000L 7-48
XC5200 7-125
XC3100A 7-54
XC3000 Series 7-22, 7-31
XC3100L 7-60
XC4000 Series 6-43, 6-49, 6-63, 7-102, 7-105, 7-114
XC4000E 6-100, 10-12
slew rate
XC5200E 7-127
Spartan Series 4-10
XC9500 5-77, 5-101
XC3000 Series 7-8
advance, definition of 1-1, 6-100, 7-127
XC4000 Series 6-27, 7-92
boundary scan
XC5200 7-91
XC4000E 6-110, 6-114, 7-132
SmartSearch 13-2
CLB
SMD. See Standard Microcircuit Drawing
XC3000A 7-43
SO8 package
XC3000L 7-49
XC3100A 7-55 15
socket vendors 11-21
XC3100L 7-61
soft startup
XC5200 7-129
Spartan Series 4-10
configuration
XC3000 Series 7-24
XC3000 Series 7-25
XC4000 Series 6-27, 7-92
XC4000 Series 6-63, 7-114
software
DC characteristics
Foundation Series 2-3
XC1700D 8-8, 8-21, 8-32
Base System (PC) 2-4, 2-5, 2-6, 2-7
13
Book Index
XC3000A 7-41 implementing Global Set/Reset 4-21, 6-13, 7-90

XC3000L 7-47 support
XC3100A 7-53 technical support hotline 13-2
XC3100L 7-59 surface mount packages
XC4000E 4-39, 6-101, 10-13 moisture sensitivity 11-15
XC5200 7-127 switch matrix, programmable (PSM)
global buffer Spartan Series 4-13
XC3000A 7-42 XC3000 Series 7-11
XC3000L 7-48 XC4000 Series 6-32
XC3100A 7-54 synchronous peripheral configuration mode
XC3100L 7-60 specifications
XC5200 7-128 XC4000 Series 6-68
IOB XC4000 Series 6-43, 6-67, 7-102, 7-118
XC3000A 7-45 synchronous RAM 6-15
XC3000L 7-51 advantages of 4-18, 6-14
XC3100A 7-57
XC3100L 7-63
XC5200 7-131 T
longline
TAP controller 14-53
XC5200 7-128
tape & reel packing 11-17
operating conditions
TCK,TDI,TDO,TMS. See boundary scan
XC1700D 8-8, 8-21, 8-32
TCLKIN 14-24
XC3000A 7-41
pin description
XC3000L 7-47
XC3000 Series 7-39
XC3100A 7-53
technical Information 13-2
XC3100L 7-59
technical literature 13-3
XC4000E 4-38, 6-100, 10-12
technical support
XC5200 7-127
hotline 13-2
XC9500 5-77, 5-101
overview 1-5
pin-to-pin
telephone support 13-2
XC5200 7-130
temperature cycling test 12-5
XC9500 5-78, 5-102
test access port (TAP) 14-53
preliminary, definition of 1-1, 6-100, 7-127
testing 12-4
programmer 8-37
thermal characteristics 11-4
readback
thermal data
Spartan 4-36
PQ/HQ packages 11-8
XC3000 Series 7-33
thermal management 11-3
XC4000E 6-60, 7-126
thermal resistance
wide edge decoder
applying data 11-7
XC4000E 6-103, 6-104, 10-15, 10-16
table of 11-6
XC3000A 7-41
thermal shock test 12-5
XC3000L 7-47
theta-JA 11-4
XC3100A 7-53
theta-JC 11-4
XC3100L 7-59
time-to-market 1-3
XC4000E 6-100
timing model
XC5200 7-127
XC9500 5-17, 5-71
speed grades available
TQ100 package
XC3000 Series 7-77
pinout table
XC4000 Series 6-154, 6-173, 6-174
XC3000 Series 7-69
SRAM. See FPGA
TQ144 package
Standard Microcircuit Drawing (SMD) 1-5, 10-1
pinout table
start-up after configuration
XC3000 Series 7-71
XC3000 Series 14-28
TQ176 package
XC4000 Series 6-54, 7-110
pinout table
STARTUP symbol
XC3000 Series 7-74
implementing Global 3-State 4-21, 6-27, 7-92
TQFP packages
14
TQ100, TQ144, TQ176 package drawings 11-33 WebLINX 1-1, 1-5, 13-2
training 1-5 Answers database 13-2
TSOP packages SmartSearch 13-2
package drawings 11-26 weight, package 11-13
TTL input wide edge decoder 6-30
Spartan Series 4-8 specifications
XC3000 Series 7-8, 7-23 XC4000E 6-103, 6-104, 10-15, 10-16
XC4000 Series 6-23, 7-91 WOR2AND symbol 6-29
XC5200 7-91 World Wide Web site for Xilinx 1-1, 1-5, 13-2
TTL output WS
none in XC5200 7-91 pin description
XC4000 Series 6-26
X
U
XC1700
unbiased pressure pot test 12-5 programmer 8-37
undershoot 14-50 XC2000 14-7
unused I/O overview 14-8
Spartan Series 4-56 XC3000 14-7
XC4000 Series 6-42, 7-101 overview 14-8
user registers in boundary scan 14-55 XC3000 Series 7-3, 7-24, 14-7, 14-19
3-state buffer 7-8, 7-17, 14-23
5-input function 14-19
V available I/O 7-65
battery backup 14-27, 14-42
V/I characteristics
CCLK frequency variation 14-27
XC3000 Series 14-23
CMOS input 7-8
XC4000 Series 14-15
CMOS output 7-8, 14-22
vendors
configurable logic block (CLB) 7-9, 14-19
forced air cooling 11-11
configuration 7-19, 14-28
heatsink 11-11
specifications 7-25
socket 11-21
crystal oscillator 7-18, 7-24, 14-25
VersaBlock 7-85
feature summary 7-3
VersaRing
glitch avoidance 14-20
Spartan Series 4-14
global buffer 7-17, 14-24
XC4000 Series 6-35
hysteresis 14-21
XC5200 7-86, 7-98
input/output block (IOB) 7-7, 14-21
VO8 package
internal bus contention 14-23
on-chip oscillator 14-27
VPP
output current 14-23
overview 14-8
VQ100 package
performance 7-35
pinout table
pin descriptions 7-38
XC3000 Series 7-69
pinout tables 7-65
VQ64 package
power consumption 7-37, 14-24
pinout table
XC3000 Series 7-66
power distribution 7-36 15
power-down mode 7-19, 7-37
VQFP packages
product availability 7-77
VQ44, VQ64, VQ100 package drawings 11-32
readback 7-23
rise/fall time 14-21
W routing 7-10
soft startup 7-24
WAND1 symbol 6-29, 6-30 specifications
Web site for Xilinx 1-1, 1-5, 13-2 configuration 7-25
15
Book Index
XC3000A 7-41 readback 6-58, 7-112

XC3000L 7-47 routing 6-31
XC3100A 7-53 soft startup 6-27, 7-92
XC3100L 7-59 specifications 4-37
TTL input 7-8 configuration 6-63, 7-114
TTL output 7-8 XC4000E 6-100
V/I characteristics 14-23 TTL input 6-23, 7-91
XC3000A TTL output 6-26
overview 7-4 V/I characteristics 14-15
specifications 7-41 wide edge decoder 6-30
XC3000L XC4000A
overview 7-4 overview 14-8
specifications 7-47 XC4000E
XC3100 14-7 clock diagram 6-39
overview 14-8 compared to XC4000 6-9
XC3100A global buffer 6-38
overview 7-4, 14-8 overview 14-8
specifications 7-53 Primary Global Buffer (BUFGP) 6-38
XC3100L Secondary Global Buffer (BUFGS) 6-38
overview 7-4, 14-8 specifications 6-100
specifications 7-59 XC4000EX
XC3195 buffered switch matrix 6-34
pinout table 7-76 clock diagram 6-39
XC4000 compared to XC4000 6-9, 6-10
overview 14-8 fast capture latch on inputs 6-26
XC4000 Series 6-7, 6-52, 7-107, 14-7 function generator in IOB 4-9, 6-28
3-state buffer 6-29 global buffer 6-40
bitstream format 6-51, 7-107 Global Early buffer (BUFGE) 6-26, 6-41
boundary scan 6-45, 14-51 Global Low-Skew buffer (BUFGLS) 6-40
BSDL files 14-61 interconnect 6-31
carry logic 6-20 latch in CLB 6-12, 7-89
CMOS input 6-23, 7-91 multiplexer in IOB 4-9, 6-28
CMOS output 6-26, 14-18 octal routing 6-35
configurable logic block (CLB) 6-11 overview 14-8
configuration 6-48, 7-104 quad routing 6-34
specifications 6-63, 7-114 routing 6-31
CRC error checking 7-107 VersaRing (IOB routing) 4-14, 6-35
edge decoder 6-30 XC4000H
feature summary 6-7 overview 14-8
global buffer 6-38 XC4000XL 6-7
ground bounce 14-16 XC4000XV Family 6-161
input/output block (IOB) 6-23, 7-91 feature summary 6-161
interconnect 6-31 XC4003E
internal bus contention 14-23 pinout table 6-116
low voltage device 6-7 XC4005E/L
on-chip oscillator 6-30 pinout table 6-117
ordering information 6-115, 6-154, 6-167, 6-171, XC4006E
6-254 pinout table 6-118
output current 6-26, 7-92, 14-15 XC4008E
output delay 14-16 pinout table 6-120
pin descriptions 6-42, 7-101 XC4010E/L
pinout tables 6-115 pinout table 6-121
power consumption 14-18 XC4013E/L
power distribution 6-42, 7-101 pinout table 6-123
product availability 6-154, 6-173, 6-174 XC4020E
RAM 6-14 pinout table 6-126
16
XC4025E overview 14-9
pinout table 6-128, 6-131 XC9500
XC4028EX/XL BSDL files 5-70
pinout table 6-128, 6-131 design security 5-16, 5-70
XC4044EX/XL endurance 5-16, 5-70
pinout table 6-134 FastCONNECT switch matrix 5-67
XC4052XL function block 5-7, 5-61
pinout table 6-138 I/O block 5-68
XC5200 14-7 in-system programming 5-16, 5-70
3-state buffer 7-90 overview 5-5, 5-59, 14-9
5-input function 7-87 package availability and device I/O pins 5-61
available I/O 7-154 pin locking capability 5-15, 5-69
boundary scan 7-98, 14-51 product availability 5-6, 5-60
BSDL files 14-61 product term allocator 5-10, 5-64
carry logic 7-88 timing model 5-17, 5-71
cascade logic 7-89 XC95108
CMOS input 7-91 pinout table 5-96
CMOS output 7-91 XC95144
compared to XC4000 and XC3000 7-84 pinout table 5-104
configurable logic block (CLB) 7-86, 7-87 XC95288
differences from XC4000 and XC3000 7-84 pinout table 5-114
configuration XC9536
general routing matrix (GRM) 7-85, 7-86, 7-93, 7-94 XC9572
global buffer 7-96 pinout table 5-88
input/output block (IOB) 7-91 XCELL newsletter 1-1
interconnect 7-86, 7-94 XCell newsletter 13-3
latch in CLB 7-87 XChecker cable
local interconnect matrix (LIM) 7-86, 7-93 readback
logic cell (LC) 7-84, 7-85 Spartan Series 4-36
on-chip oscillator 7-93 XC4000 Series 6-59, 7-113
ordering information 7-155 XCS05
output current 7-91 pinout table 4-58
overview 14-8 XCS10
performance 7-87 pinout table 4-59
routing 7-86, 7-94 XCS20
specifications 7-125, 7-127 XCS30
XC5200 7-127 pinout table 4-62
TTL input 7-91 XCS40
TTL output not supported 7-91 pinout table 4-65
VersaBlock 7-85 Xilinx
VersaRing (IOB routing) 7-86, 7-98 about the company 1-2
XC5202 quality assurance and reliability 12-3
pinout table 7-133 technical support 1-5
XC5204 Web site 1-1, 1-5, 13-2
pinout table 7-135 XCELL newsletter 1-1
XC5206 XCell newsletter 13-3
pinout table 7-139 xtal oscillator. See crystal oscillator 15
XC5210 XTL1
pinout table 7-142 pin description
XC5215 XC3000 Series 7-39
pinout table 7-147 XTL2
XC7200A pin description
overview 14-9 XC3000 Series 7-39
XC7300
17
Sales Offices, Sales
Representatives, and Distributors
1 Introduction
3 Virtex Products
4 Spartan Products
5 XC9500 Products
6 XC4000 Products
15 Index

Table of Contents

Headquarters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-1
Xilinx Sales Offices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-1
North American Distributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-2
U.S. Sales Representatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-2
International Sales Representatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-4
0
R
December 8, 1998 (Version 1.1) 0 16*
Headquarters XILINX, Inc.

1025 S. Semoran Blvd.
XILINX, Inc.
939 North Plum Grove Road
XILINX, Inc.
2910 South Sheridan Way,
XILINX, Inc. Suite 1093 Suite H Suite 203
2100 Logic Drive Winter Park, FL 32792 Schaumburg, IL 60173 Oakville, Ontario
San Jose, CA 95124 Tel: (407) 673-8661 Tel: (847) 605-1972 Canada L6J7L9
Tel: (408) 559-7778 Fax: 407-673-8663 Fax: (847) 605-1976 Tel: (905) 829-9095
TWX: (510) 600-8750 Fax: (905) 829-3045
Fax: (408) 559-7114 XILINX, Inc. XILINX, Inc.
3280 Pointe Parkway 18283 Minnetonka Blvd. XILINX, Inc.
Xilinx Sales Suite 1600
Norcross, GA 30092
Suite C
Deephaven, MN 55391
34 Hampel Crescent
Stittsville, Ontario
Offices Tel: (770) 448-4733 Tel: (612) 473-4816 Canada K2S 1E4
Fax: 770-448-4857 Fax: 612-473-5060 Tel: (613) 836-5255
NORTH AMERICA Fax: 613-836-5393
XILINX, Inc. XILINX, Inc.
XILINX, Inc. EUROPE
15615 Alton Parkway 12922 Kentbury Dr.
4825 University Square
Suite 280 Clarksville, MD 21029 XILINX, Ltd.
Suite 12
Irvine, CA 92618 Tel: (301) 924-1300 Benchmark House
Huntsville, AL 35816
Tel: (949) 727-0780 Fax: 301-924-1301 203 Brooklands Road
Tel: (256) 721-3370
Fax: (949) 727-3128 Weybridge, Surrey
Fax: 256-721-3371
XILINX, Inc. KT13 ORH
XILINX, Inc. 6010-C Six Forks Road United Kingdom
XILINX, Inc.
6494 Weathers Place Raleigh, NC 27609 Tel: (44) 1932-349401
10235 South 51st St.
Suite 100 Tel: (919) 846-3922 Fax: (44) 1932-349499
Suite 160
San Diego, CA 92121 Fax: (919) 846-8316
Phoenix, AZ 85044
Tel: (619) 597-9855 XILINX, Ltd.
Tel: (602) 753-4503
Fax: 619-597-6418 XILINX, Inc. (Northern European Sales)
Fax: 602-753-4504
4100 McEwen, Suite 237 Suite 1B Cobb House
XILINX, Inc. Dallas, TX 75244 Oyster Lane
XILINX, Inc.
61 Spit Brook Rd. Tel: (972) 960-1043 Byfleet, Surrey
1281 Oakmead Pkwy.
Suite 403 Fax: (972) 960-0927 KT14 7DU
Suite 202
Nashua, NH 03060 United Kingdom
Sunnyvale, CA 94086
Tel: (603) 891-1098 XILINX, Inc. Tel: (44) 1932-349403
Tel: (408) 245-9850
Fax: (603) 891-0890 4765 S. Quail Point Road Fax: (44) 1932-345519
Fax: (408) 245-9865
Salt Lake City, UT 84124
XILINX, Inc. Tel: (801) 273-7338 XILINX Sarl
XILINX, Inc.
39 Surrey Dr. Espace Jouy Technology
1227 Platte Ave.
Belle Mead, NJ 08502 XILINX, Inc. 21, rue Albert Calmette, Bât. C
Ventura, CA 93004
Tel: (908) 359-3136 14575 Bel-Red Road 78353 Jouy en Josas, Cedex
Tel: (805) 647-9221
Fax: 908-359-1253 Suite 102 France
Fax: 805-647-9221
Bellevue, WA 98007 Tel: (33) 1 34 63 01 01
XILINX, Inc. Tel: (425) 603-0102 Fax: (33) 1 34 63 01 09
XILINX, Inc.
30 Two Bridges Rd. Fax: 425-603-0197
5690 DTC Blvd.
Suite 330 XILINX, Sarl
Suite 490W
Fairfield, NJ 07004 XILINX, Inc. 417, Chemin du Cassan
Englewood, CO 80111
Tel: (973) 808-2780 3554 Brecksville Rd 06140 Tourrettes sur Loup
Tel: (303) 220-7541
Fax: 973-808-2738 Richfield, OH 44286 France
Fax: (303) 220-8641
Tel: (330) 659-3131 Tel: (33) 4-9324-1175
XILINX, Inc. Fax: (330) 659-9254 Fax: (33) 4-9324-1007
XILINX, Inc.
14 Mitchell Terrace
1500 Kansas Ave.
West Long Branch, NJ 07764 XILINX, Inc. XILINX GmbH
Suite 1B
Tel: (732) 870-1126 9600 S. W. Oak St. Süskindstr. 4
Longmont, CO 80501
Fax: 732-870-1785 Suite 320 D-81929 München
Tel: (303) 774-1175
Portland, OR 97223 Germany
Fax: 303-774-1198
XILINX, Inc. Tel: (425) 293-9016 Tel: (49) 89-93088-0 16
905 Airport Rd. Fax: 425-293-3858 Tech Support Tel:
Suite 200 (49) 89-93088-130
West Chester, PA 19380 Fax: (49) 89-93088-188
Tel: (610) 430-3300
Fax: (610) 430-0470

R

XILINX AB Insight Electronics COLORADO Thorson Pacific, Inc.
Box 1230 (Locations throughout the U.S.) 14575 Bel-Red Road #102
Torshamnsgatan 35 Tel: (800) 677-7716 Luscombe Engineering, Inc. Bellevue, WA 98007
S-164 28 Kista Fax: (619) 587-1380 1500 Kansas Ave. Suite 1B Tel: (425) 603-9393
Sweden Longmont, CO 80501 Fax: (425) 603-9380
Tel: (46) 8-752-2470 Nu Horizons Tel: (303) 772-3342
Fax: (46) 8-750-6260 Electronics Corp. Fax: (303) 772-8783 ILLINOIS
E-mail: xilinx-nordic@xilinx.com Locations throughout the U.S. CONNECTICUT Advanced Technical Sales
Tel: (516) 396-5000 13755 St. Charles Rock Rd.
XILINX Italia Fax: (516) 396-7576 John E. Boeing, Co., Inc. Bridgeton, MO 63044
Via Zamagna 19 - Scala A 123 South Main Street Tel: (314) 291-5003
20148 Milano
Italy
U.S. Sales Wallingford, CT 06492
Tel: (203) 265-1318
Fax: (314) 291-7958
Tel: (39) 02 487 12 101 Representatives Fax: 203-265-0235 Beta Technology Sales, Inc.
Fax: (39) 02 400 94 700 1009 Hawthorn Drive
ALABAMA DELAWARE
Itasca, IL 60143
XILINX Benelux bvba Electro Source, Southeast Delta Technical Sales, Inc. Tel: (708) 250-9586
Oude Wichelsesteenweg 27 4825 University Sq., Ste.12 122 N. York Rd., Suite 9 Fax: (708) 250-9592
9340 Lede Huntsville, AL 35816 Hatboro, PA 19040
Belgium Tel: (256) 830-2533 Tel: (215) 957-0600 INDIANA
Tel: (32) 3 205 56 65 Fax: (256) 830-5567 Fax: (215) 957-0920 Gen II Marketing,Inc.
Fax: (32) 5381 0472
FLORIDA 31 E. Main St.
ARIZONA Carmel, IN 46032
JAPAN
Quatra Associates Semtronic Assoc., Inc. Tel: (317) 848-3083
XILINX K. K. 10235 S. 51st St. Suite #160 (Disti Office) Fax: (317-848-1264
Shinjuku Square Tower 18F Phoenix, AZ 85044 600 S. North Lake Blvd.
6-22-1 Nishi-Shinjuku Tel: (602) 753-5544 Suite 270 Gen II Marketing, Inc.
Shinjuku-ku, Tokyo 163-1118 Fax: (602) 753-0640 Altamonte Springs, FL 32701 1415 Magnavox Way
Japan E-mail: quatra@earthlink.net Tel: (407) 831-0451 Suite 130
Tel: (81) 3-5321-7711 Fax: (407) 831-6055 Ft. Wayne, IN 46804
Fax: (81) 3-5321-7765 ARKANSAS Tel: (219) 436-4485
Semtronic Assoc., Inc. Fax: (219) 436-1977
ASIA PACIFIC Bonser-Philhower Sales
(OEM Sales)
689 W. Renner Road IOWA
XILINX Asia Pacific 600 S. North Lake Blvd.
Suite 101
Unit 4312, Tower II Suite 220 Advanced Technical Sales
Richardson, TX 75080
Metroplaza Altamonte, Springs, FL 32701 375 Collins Road NE
Tel: (972) 234-8438
Hing Fong Road Tel: (407) 831-8233 Cedar Rapids, IA 52402
Fax: (972) 437-0897
Kwai Fong, N.T. Fax: (407) 831-2844 Tel: (319) 393-8280
Hong Kong CALIFORNIA Fax: (319) 393-7258
Tel: (852) 2-424-5200 Semtronic Assoc., Inc.
Fax: (852) 2-494-7159 Norcomp 3471 NW 55th Street KANSAS
E-mail: hongkong@xilinx.com 1267 Oakmead Pkwy Ft. Lauderdale, FL 33309
Sunnyvale, CA 94086 Tel: (954) 731-2384 Advanced Technical Sales
Tel: (408) 733-7707 Fax: (954) 731-1019 2012 Prairie Cir. Suite A
XILINX Korea
Fax: (408) 774-1947 Olathe, KS 66062
Room #901
Semtronic Assoc., Inc. Tel: (913) 782-8702
Sambo-Hojung Bldg.,
Norcomp 14004 Roosevelt Blvd. Fax: (913) 782-8641
14-24, Yoido-Dong
Youngdeungpo-Ku 8880 Wagon Way Suite 604
Granite Bay, CA 95746 KENTUCKY
Seoul, South Korea Clearwater, FL 33762
Tel: (82) 2-761-4277 Tel: (916) 791-7776 Tel: (727) 507-0504 Gen II Marketing, Inc.
Fax: (82) 2-761-4278 Fax: (916) 791-2223 Fax: (727) 539-0601 861 Corporate Dr. #210
Lexington, KY 40503
XILINX Taiwan Norcomp GEORGIA Tel: (606) 223-9181
Rm. 1006, 10F, No. 2, Lane 150 30101 Agoura Ct. #234 Fax: (606) 223-2864
Electro Source, Southeast
Sec. 5, Hsin Yin Rd. Agoura, CA 91301
3280 Pointe Parkway, LOUISIANA (Northern)
Taipei, 105 Taiwan, R.O.C. Tel: (818) 865-8330
Suite 1500
Tel: (886) 2-2758-8373 Fax: (818) 865-2167 Bonser-Philhower Sales
Norcross, GA 30092
Fax: (886) 2-2758-8367 Tel: (770) 734-9898 689 W. Renner Rd., Suite 101
Norcomp Richardson, TX 75080
Fax: (770) 734-9977
30 Corporate Park #200
North American Irvine, CA 92714 IDAHO (Southwest)
Tel: (972) 234-8438
Fax: (972) 437-0897
Distributors Tel: (949) 260-9868
Fax: (949) 260-9659 Luscombe Engineering, Inc. LOUISIANA (Southern)
Hamilton Hallmark 6901 Emerald , Suite 206
(Locations throughout Quest-Rep Inc. Boise, ID 83704 Bonser-Philhower Sales
the U.S. and Canada) 6494 Weathers Pl, Suite 200 Tel: (208) 377-1444 10700 Richmond, Suite 150
Tel: (800) 332-8638 San Diego, CA 92121 Fax: (208) 377-0282 Houston, TX 77042
Fax: (800) 257-0568 Tel: (619) 622-5040 Tel: (713) 782-4144
Fax: (619) 622-9007 Fax: (713) 789-3072
E-mail: questrep@questrep.com

R

MAINE NEBRASKA NORTH CAROLINA PUERTO RICO
Genesis Associates Advanced Technical Sales Electro Source, Southeast Semtronic Assoc., Inc
128 Wheeler Road 375 Collins Road NE 5964-A Six Forks Rd. Crown Hills
Burlington, MA 01803 Cedar Rapids, IA 52402 Raleigh, NC 27609 125 Carite St.
Tel: (781) 270-9540 Tel: (319) 393-8280 Tel: (919) 846-5888 Esq. Avenue Parana
Fax: (781) 229-8913 Fax: (319) 393-7258 Fax: (919) 846-0408 Rio Piedras, P.R. 00926
NEVADA Tel: (787) 766-0700/0701
MARYLAND Electro Source Fax: (787) 763-8071
Norcomp 12411 Angle Oak Drive
Micro Comp, Inc. Huntersville, NC 28078
8880 Wagon Way RHODE ISLAND
1421 S. Caton Avenue Tel: (704) 948-8905
Baltimore, MD 21227-1082 Granite Bay, CA 95748 Genesis Associates
Tel: (916) 393-8280 Fax: (704) 948-5829
Tel: (410) 644-5700 128 Wheeler Road
Fax: (410) 644-5707 Fax: (916) 393-7258 Burlington, MA 01803
NORTH DAKOTA
Tel: (781) 270-9540
MASSACHUSETTS Quatra Associates Beta Technology Fax: (781) 229-8913
(Las Vegas) 18283 Minnetonka Blvd.
Genesis Associates 4645 S. Lakeshore Dr., Suite 1 Suite C SOUTH CAROLINA
128 Wheeler Road Tempe, AZ 85282 Deephaven, MN 55391
Burlington, MA 01803 Tel: (602) 820-7050 Tel: (612) 473-2680 Electro Source, Southeast
Tel: (781) 270-9540 Fax: (602) 820-7054 Fax: (612) 473-2690 5964-A Six Forks Rd.
Fax: (781) 229-8913 Raleigh NC 27609
NEW HAMPSHIRE OHIO Tel: (919) 846-5888
MICHIGAN Fax: (919) 846-0408
Genesis Associates Bear Marketing, Inc.
Miltimore Sales Inc. 128 Wheeler Road 3554 Brecksville Road SOUTH DAKOTA
22765 Heslip Drive Burlington, MA 01803 PO Box 427
Novi, MI 48375 Tel: (781) 270-9540 Richfield, OH 44286-0427 Beta Technology
Tel: (248) 349-0260 Fax: (781) 229-8913 Tel: (216) 659-3131 18283 Minnetonka Blvd.
Fax: (248) 349-0756 Fax: (216) 659-4823 Suite C
NEW JERSEY (Northern) Deephaven, MN 55391
Miltimore Sales Inc. Bear Marketing, Inc. Tel: (612) 473-2680
3680 44th St., Suite 100-J Parallax Fax: (612) 473-2690
734 Walt Whitman Road 270 Regency Ridge Drive
Kentwood, MI 49512 Suite 115
Tel: (616) 554-9292 Melville, NY 11747 TENNESSEE
Tel: (516) 351-1000 Dayton, OH 45459
Fax: (616) 554-9210 Tel: (513) 436-2061 Electro Source, Southeast
Fax: (516) 351-1606
Fax: (513) 436-9137 4825 University Square,
MINNESOTA
NEW JERSEY (Southern) Suite 12
Beta Technology OKLAHOMA Huntsville, AL 35816
18283 Minnetonka Blvd. Delta Technical Sales, Inc. Tel: (205) 830-2533
122 N. York Road, Suite 9 Bonser-Philhower Sales
Suite C 689 W. Renner Rd., Suite 101 Fax: (205)-830-5567
Deephaven, MN 55391 Hatboro, PA 19040
Tel: (215) 957-0600 Richardson, TX 75080 TEXAS
Tel: (612) 473-2680 Tel: (972) 234-8438
Fax: (612) 473-2690 Fax: (215) 957-0920
Fax: (972) 437-0897 Bonser-Philhower Sales
NEW MEXICO 8240 MoPac Expwy.
MISSISSIPPI OREGON Suite 295
Electro Source, Southeast Quatra Associates Austin, TX 78759
600 Autumnwood Place, SE Thorson Pacific, Inc.
4825 University Sq., Ste.12 9600 SW Oak Street, Tel: (512) 346-9186
Huntsville, AL 35816 Albuquerque, NM 87123 Fax: (512) 346-2393
Tel: (505) 296-6781 Suite 320
Tel: (256) 830-2533 Portland, OR 97223
Fax: (256) 830-5567 Fax: (505) 292-2092 Bonser-Philhower Sales
Tel: (503) 293-9001
NEW YORK (Metro) Fax: (503) 293-9007 10700 Richmond, Suite 150
MISSOURI Houston, TX 77042
Advanced Technical Sales Parallax PENNSYLVANIA Tel: (713) 782-4144
2012 Prairie Cir. Suite A 734 Walt Whitman Road Fax: (713) 789-3072
Melville, NY 11747 Bear Marketing, Inc.
Olathe, KS 66062 4284 Rt. 8, Suite 211
Tel: (913) 782-8702 Tel: (516) 351-1000 Bonser-Philhower Sales
Fax: (516) 351-1606 Allison Park, PA 15101 689 W. Renner Rd., Suite 101
Fax: (913) 782-8641 Tel: (412) 492-1150 Richardson, TX 75080
NEW YORK Fax: (412) 492-1155 Tel: (972) 234-8438
Advanced Technical Sales
13755 St. Charles Rock Rd. Electra Sales Corp. Fax: (972) 437-0897
Delta Technical Sales, Inc.
Bridgeton, MO 63044 333 Metro Park
Rochester, NY 14623
122 N. York Rd., Suite 9 TEXAS (El Paso County)
Tel: (314) 291-5003 Hatboro, PA 19040
Fax: (314) 291-7958 Tel: (716) 427-7860 Quatra Associates
Tel: (215) 957-0600
Fax: (716) 427-0614
Fax: (215) 957-0920 600 Autumnwood Place SE 16
MONTANA Albuquerque, NM 87123
Luscombe Engineering, Inc. Electra Sales Corp. Tel: (505) 296-6781
670 East 3900 South #103 6057 Corporate Drive Fax: (505) 292-2092
Salt Lake City, UT 84107 E. Syracuse, NY 13057
Tel: (801) 268-3434 Tel: (315) 463-1248
Fax: (801) 266-9021 Fax: (315) 463-1717

R

UTAH WYOMING Advanced Component Dist. CANADA (ALBERTA)
Ste. 1, 1048 Beaudesert Rd.
Luscombe Engineering Co. Luscombe Engineering, Inc. Cooper Plains Electro Source
670 East 3900 South #103 1500 Kansas Ave. Suite 1B Queensland 4108 2635 37th Ave NE #245
Salt Lake City, UT 84107 Longmont, CO 80501 Australia Calgary, Alberta T1Y 5Z6
Tel: (801) 268-3434 Tel: (303) 772-3342 Tel: (61) 7-3246-5214 Canada
Fax: (801) 266-9021 Fax: (303) 772-8783 Fax: (61) 7-3275-3662 Tel: (403) 735-6230
Fax: (403) 735-0599
VERMONT
International EDA Solutions Pty. Ltd. CANADA
Genesis Associates Level 3, South Tower
128 Wheeler Road Sales 1-5 Railway Street (BRITISH COLUMBIA)
Burlington, MA 01803
Tel: (781) 270-9540
Representatives Chatswood NSW 2067 Thorson Pacific, Inc.
Australia 4170 Still Creek Dr. #200
Fax: (781) 229-8913 ALGERIA Tel: (61) 02-9413-4611 Burnaby BC V5C 6C6
Fax: (61) 02-9413-4622 Canada
VIRGINIA Development Centre of
Tel: (604) 294-3999
Advanced Technologies EDA Solutions Pty. Ltd.
Microcomp, Inc. Fax: (604) 473-7755
128 Chemin Mohamed GACEM Level 2, 854 Glenferrie Road
1421 S. Caton Avenue
16075 El-Madania Hawthorn VIC 3122 CANADA (OTTAWA)
Baltimore, MD 21227
Algiers Australia
Tel: (410) 644-5700 Electro Source, Inc.
Algeria Tel: (61) 03-9819-0000
Fax: (410) 644-5707 50 Hines Road, Suite 220
Tel: (213) 2-67-73-25 Fax: (61) 03-9818-8870
Fax: (213) 2-66-26-89 Kanata, Ontario K2K 2M5
WASHINGTON
Canada
Thorson Pacific, Inc. ARGENTINA AUSTRIA Tel: (613) 592-3214
14575 Bel-Red Rd. Fax: (613) 592-4256
Reycom Electronica S.A. Metronik GmbH
Suite 102
Bdo. de Irigoyen Diefenbachgasse 35 CANADA (QUEBEC)
Bellevue, WA 98007
972 Piso 2do “B” A-1150 Wien
Tel: (206) 603-9393 Electro Source
1304 Buenos Aires Austria
Fax: (206) 603-9380 6600 TransCanada Hwy
Argentina Tel: (43) 1-89-5762652
Tel: (54) 1-304-2018 Suite 420
WASHINGTON Fax: (43) 1-89-5762650
Fax: (54) 1-304-2010 Pointe Claire, Quebec H9R 4S2
(Vancouver, WA only) Canada
BELGIUM & LUXEMBURG
Thorson Pacific, Inc. Insight Argentina Tel: (514) 630-7486
9600 SW Oak Street Av. Adolfo Davila 550, 2nd Floor SEI Rodelco NV Fax: (514) 630-7421
Suite 320 1107 Buenos Aires Limburg Stirum 243
1780 Wemmel CANADA (TORONTO)
Portland, OR 97223 Argentina
Tel: (503) 293-9001 Tel: (54) 1-310-0052 Belgium Electro Source, Inc.
Fax: (503) 993-9007 Fax: (54) 1-310-0053 Tel: (32) 2-456-0757 230 Galaxy Blvd.
Fax: (32) 2-460-0271 Rexdale Ontario M9W 5R8
WASHINGTON D.C. AUSTRALIA Canada
BRASIL
Micro Comp, Inc. Advanced Component Dist. Tel: (416) 675-4490
1421 S. Caton Avenue Suite 5, Level 1, "Metro Centre" Hitech Fax: (416)-675-6871
Baltimore, MD 21227-1082 124 Forest Rd. Rua Branco de Moraes 489
Ch. Santo Antonio CHILE
Tel: (410) 644-5700 Hurstville 2220
Fax: (410) 644-5707 Australia Sao Paulo DTS Ltda.
Tel: (61) 2-9585-5533 04718-010 - SP - Brazil Rosas 1444
WEST VIRGINIA Fax: (61) 2-9585-5534 Tel: (55) 11-882-4000 Santiago
Fax: (55) 11-882-4100 Chile
Bear Marketing, Inc.
4284 Rt. 8 Suite 211 Advanced Component Dist. Tel: (56) 2-6970991
Unit 2, 17-19 Melrich Road Insight Brasil Fax: (56) 2-6993316
Allison Park, PA 15101 Rua Alcides Ricardini Neves
Tel: (412) 492-1150 Bayswater VIC 3153
Melbourne, Australia 12 - 13o andar CHINA PEOPLE’S
Fax: (412) 492-1155 Conjunto 1306 - Brooklin
Tel: (61) 3-9760-4250 REPUBLIC
WISCONSIN (Western) Fax: (61) 3-9760-4255 04575-050 Sao Paulo SP
Tel: (55) 11-5505-6501/2 Insight
Beta Technology Fax: (55) 11-5505-6702 Rm. 692, Pana Tower
Advanced Component Dist.
18283 Minnetonka Blvd. E-Mail: insbrz@uol.com.br No. 128 Zhichun Rd.
20D William Street
Suite C Haidian District
Norwood SA 5067 BULGARIA
Deephaven, MN 55391 Beijing 10086
Australia
Tel: (612) 473-2680 Petrex 92 Ltd. P.R. China
Tel: (61) 8-8364-2844
Fax: (612) 473-2690 Philip Kutev Str. 1 Tel: (86) 10-6262-8985
Fax: (61) 8-8364-2811
BG-1407 Sofia Fax: (86) 10-6262-0393
WISCONSIN (Eastern)
Bulgaria
Beta Technology Sales, Inc. Tel: (359) 2-626987
9401 N. Beloit, Suite 409 Fax: (359) 2-627099
Milwaukee, WI 53227
Tel: (414) 543-6609
Fax: (414) 543-9288

R

Insight CZECH REPUBLIC AVNET Composants GERMANY
Rm. 2015-2018, Sud-Ouest
Tong Mei Mansion MES Praha s.r.o. Technoparc Avnet EMG GmbH
No. 76, Section 1, Platonova 3287/26 Bât.4, Voie 5, BP 404 Stahlgruberring 12
Jianshe North Rd., CZ 14300 Praha 31314 Labège Cedex D-81829 München
Chengdu, Sichuan 610051 Tel: (420) 2-90059190 France Germany
P.R. China Fax: (420) 2-90059190 Tel: (33) 5 61 39 21 12 Tel: (49) 89-45110-01
Tel: (86) 28-3399-629 E-mail: mespraha@ms.anet.cz Fax: (33) 5 61 39 21 40 Fax: (49) 89-45110-129
Fax: (86) 28-3398-829 DENMARK
AVNET Composants Avnet EMG GmbH
Insight Micronor A/S Rhône-Auvergne Kurfürstenstr. 130
Rm. 705, P.O. Box 929 Parc Club du Moulin à Vent D-10785 Berlin
Wuhan Computer City Torvet 1 Bât 32-33, rue du Dr. G. Levy Germany
39 Luo Yu Road, DK-8600 Silkeborg 69693 Venissieux Cedex Tel: (49) 30-214882-0
Hongshan District Denmark France Fax: (49) 30-2141728
Wuhan, 430079 Tel: (45) 8681-6522 Tel: (33) 4 78 00 1280
PR China Fax: (45) 8681-2827 Fax: (33) 4 78 75 95 97 Avnet EMG GmbH
Tel: (86) 27-8787-4319 E-Mail: e-mail@micronor.dk Wolfenbüttler Str. 33
Fax: (86) 27-8786-3102 WWW: http://www.micronor.dk AVNET EMG D-38102 Braunschweig
E-mail: Parc Club du Moulin à Vent Germany
memecnh@public.wh.hb.cn Avnet Nortec Bât. 40 Tel: (49) 531-22073-0
Transformervej 17 33, av. du Dr. Georges Levy Fax: (49) 531-22073-35
Insight Dek-2730 Herlev 69693 Venissieux Cedex
Rm. 715, Bao Hua Bldg., Denmark France Avnet EMG GmbH
#1016, Hua Qiang North Rd., Tel: (45) 44-88-08-00 Tel: (33) 4 78 77 13 60 Friedrich-Ebert-Damm 145
Shenzhen 518031 Fax: (45) 44-88-08-88 Fax: (33) 4 78 77 13 99 D-22047 Hamburg
P.R. China Germany
Tel: (86) 755-377-9548 EGYPT AVNET Composants Tel: (49) 40-696952-0
Fax: (86) 755-377-9026 Ouest Fax: (49) 40-6962787
Guide Systems Integrators
27 Mokhles Al-Alfi St. Technoparc-Bât. E
Insight 4 Av. des Peupliers, BP 43 Avnet EMG GmbH
First - Zone Nasr City
Room. 1407-1409 35511 Cesson Sévigné Cedex Benzstr. 1
Cairo
China Venturetech Plaza France D-70839 Gerlingen
Egypt
819 Nanjing Road (W) Tel: (33) 2 99 83 84 85 Stuttgart, Germany
Tel: (20) 2-401-4085
Shanghai 200041 Fax: (33) 2 99 83 80 83 Tel: (49) 7156-4390
Fax: (20) 2-401-4997
P.R. China Fax: (49) 7156-439-111
E-Mail: obadr@idsc.gov.eg
Tel: (86) 21-6215-9935 Compress
Fax: (86) 21-6215-9938 FINLAND 47, rue de l’Estérel Avnet EMG GmbH
Silic 539 Max-Planck-Str. 15b
Memec Finland OY D-40699 Erkrath
Insight Kauppakaarre 1 94633 Rungis Cedex
Rm. 805, Genius Xin Jie Kou France Düsseldorf, Germany
00700 Helsinki Tel: (49) 211-92003-0
Commercial Centre Finland Tel: (33) 1 41 80 29 00
#219 Zhongshan South Road. Fax: (33) 1 46 86 67 63 Fax: (49) 211-9200399
Tel: (358) 9-350-8880
Nanjing, 210005, P.R. China Fax: (358) 9-350-88828
Tel: (86) 25-4549-807 / Compress Avnet EMG GmbH
4513-183 Rhône-Alpesl Schmidtstr. 49
FRANCE
Fax: (86) 25-4549-585 19 Chemin du Goyet D-60326 Frankfurt/M.
REP’TRONIC 38300 Bourgoin-Jallieu Germany
Insight 1 Bis, rue Marcel Paul France Tel: (49) 69-973804-0
Rm. 703, Chongqing (Yuzhou) Z.I. La Bonde Tel: (33) 4 74 43 53 33 Fax: (49) 69-7380712
Computer City 91742 Massy Cedex Fax: (33) 4 74 43 53 30
3 Ke Yuan Yi Rd., Shiqiaopu France Avnet EMG GmbH
Chongqing, 400039, P.R. Tel: (33) 1 69 53 67 20 Compress Fürther Str. 212
China Fax: (33) 1 60 13 91 98 Bretagne D-90429 Nürnberg
Tel: (86) 23-6879-0845 E-Mail: 19 rue de Kerjean Germany
Fax: (86) 23-6879-0845 100745.605@compuserve.com 22700 Louannec Tel: (49) 911-93149-0
France Fax: (49) 911-320821
Insight AVNET EMG. Tel: (33) 2 96 49 09 52
Rm 35G, Tower 3, Xiangjiang 79 Rue Pierre Sémard Fax: (33) 2 96 49 09 56 Intercomp
Great Garden 92320 Châtillon Am Hochwald 42
Xiamen, 361009, P.R. China France Compress D-82319 Starnberg
Tel: (86) 592-513-7850 Tel: (33) 1 49 65 27 00 Ouest Germany
Fax: (86) 592-513-7850 Fax: (33) 1 49 65 27 39 5 Impasse Guenot Tel: (49) 8151-16044
31100 Toulouse Fax: (49) 8151-79270 16
France E-Mail: intercomp.tiefenthaler
Tel: (33) 5 62 87 75 08 @t-online.de
Fax: (33) 5 62 87 75 09

R

Intercomp Metronik GmbH INDIA Avnet EMG
Heerstr. 167 Äussere Grossweiden- Centro Direzionale
D-78628 Rottweil mühlstr. 45 CG-CoreEl Logic System Ltd. Via Novara, 570
Germany D-90419 Nürnberg First Floor, Surya Bhavan C.A.P 20153 Milan
Tel: (49) 741-14845 Germany 1181 Fergusson College Rd. Italy
Fax: (49) 741-15220 Tel: (49) 911-338802 Pune 411 005 Tel: (39) 02-381-901
E-Mail: Fax: (49) 911-338827 India Fax: (39) 02-380-02988
intercomp.klink@t-online.de Tel: (91) 212-323982, 328074
Metronik GmbH Fax: (91) 212-323985 Avnet EMG
Intercomp Löwenstrasse 37 Email: xsupport@cromp.ernet.in Ancona
Schustergasse 25 D-70597 Stuttgart Via Adriatica, 13
D-55278 Köngernheim Germany CG-CoreEL Logic System Ltd. 60022 Castelfidardo
Germany Tel: (49) 711-769641-0 961 Main 12th Italy
Tel: (49) 6737-9881 Fax: (49) 711-765-5181 HAL 2nd Stage Tel: (39) 071-781-9644
Fax: (49) 6737-9882 Bangalore 560 008 Fax: (39) 071-781-9699
E-Mail: Metronik GmbH India
intercomp.harkam@t-online.de Franz-Schubert-Str.41 Tel: (91) 80-527-9726 Avnet EMG
D-16548 Glienicke Fax: (91) 80-527-3073 Firenze
Intercomp Berlin, Germany E-Mail: Via Panciatichi, 40
Heidbergstr 30 Tel: (49) 33056-62510 vishwa@giasbg01.vsni.net.in 50127 Firenze
D-22846 Norders

Databk

Uploaded by

Copyright:

Available Formats

Databk

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Databk

Uploaded by

Copyright:

Available Formats

R

The Programmable Logic

United Kingdom Hotline: (44) 1932-820821

France Hotline: (33) 1-3463-0100

Germany Hotline: (49) 89-93088-130

Japan Hotline: (81) 3-3297-9163

Korea Hotline: (82) 2-761-4277

Hong Kong Hotline: (85) 2-2424-5200

2100 Logic Drive

2 Development System Products and CORE Solutions Products

7 XC3000 & XC5200 Products

8 SPROM Products and Programming Support

9 HardWire FpgASIC Products

10 High-Reliability and QML Military Products

11 Packages and Thermal Characteristics

12 Testing, Quality, and Reliability

13 Technical Support and Services

14 Product Technical Information

16 Sales Offices, Sales Representatives, and Distributors

An Introduction to Xilinx Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

Development System Products and CORE Solutions Products

Development Systems: Products Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

Development Systems: Product Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3

CORE Solutions Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13

Spartan and SpartanXL Families

Spartan and SpartanXL Families Field Programmable Gate Arrays . . . . . . . . . . . . . . . . 4-38

Spartan and SpartanXL Families Field Programmable Gate Arrays . . . . . . . . . . . . . . . . 4-56

Spartan and SpartanXL Families Field Programmable Gate Arrays . . . . . . . . . . . . . . . . 68

XC9500/XL Series Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

FastFLASH™ XC9500XL High-Performance CPLD Family . . . . . . . . . . . . . . . . . . . . . . . 5-5

XC9536XL High Performance CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-21

XC9572XL High Performance CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-29

XC95144XL High Performance CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-37

XC9500 In-System Programmable CPLD Family. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-59

XC9536 In-System Programmable CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-75

XC9572 In-System Programmable CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-83

XC95108 In-System Programmable CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-91

XC95144 In-System Programmable CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-99

XC95288 In-System Programmable CPLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-109

XC4000E and XC4000X Series

XC4000E and XC4000X Series Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . 6-7

XC4000E and XC4000X Series Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . 6-72

XC4000E and XC4000X Series Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . 6-115

XC4000E and XC4000X Series Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . 6-154

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-203

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-207

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-211

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-215

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-221

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-227

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-233

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-239

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-246

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-255

XC4000XLA/XV Field Programmable Gate Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-263

XC3000 & XC5200 Products

XC3000 Series Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1

XC5200 Series Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-79

SPROM Products and