Dspic30F Family Reference Manual: High-Performance Digital Signal Controllers

dsPIC30F Family Reference Manual
High-Performance Digital Signal Controllers
2006 Microchip Technology Inc.
DS70046E
Note the following details of the code protection feature on Microchip devices: Microchip products meet the specification contained in their particular Microchip Data Sheet. Microchip believes that its family of products is one of the most secure families of its kind on the market today, when used in the intended manner and under normal conditions. There are dishonest and possibly illegal methods used to breach the code protection feature. All of these methods, to our knowledge, require using the Microchip products in a manner outside the operating specifications contained in Microchips Data Sheets. Most likely, the person doing so is engaged in theft of intellectual property. Microchip is willing to work with the customer who is concerned about the integrity of their code. Neither Microchip nor any other semiconductor manufacturer can guarantee the security of their code. Code protection does not mean that we are guaranteeing the product as unbreakable.
Code protection is constantly evolving. We at Microchip are committed to continuously improving the code protection features of our products. Attempts to break Microchips code protection feature may be a violation of the Digital Millennium Copyright Act. If such acts allow unauthorized access to your software or other copyrighted work, you may have a right to sue for relief under that Act.
Information contained in this publication regarding device applications and the like is provided only for your convenience and may be superseded by updates. It is your responsibility to ensure that your application meets with your specifications. MICROCHIP MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND WHETHER EXPRESS OR IMPLIED, WRITTEN OR ORAL, STATUTORY OR OTHERWISE, RELATED TO THE INFORMATION, INCLUDING BUT NOT LIMITED TO ITS CONDITION, QUALITY, PERFORMANCE, MERCHANTABILITY OR FITNESS FOR PURPOSE. Microchip disclaims all liability arising from this information and its use. Use of Microchip devices in life support and/or safety applications is entirely at the buyers risk, and the buyer agrees to defend, indemnify and hold harmless Microchip from any and all damages, claims, suits, or expenses resulting from such use. No licenses are conveyed, implicitly or otherwise, under any Microchip intellectual property rights.
Trademarks The Microchip name and logo, the Microchip logo, Accuron, dsPIC, KEELOQ, microID, MPLAB, PIC, PICmicro, PICSTART, PRO MATE, PowerSmart, rfPIC and SmartShunt are registered trademarks of Microchip Technology Incorporated in the U.S.A. and other countries. AmpLab, FilterLab, Migratable Memory, MXDEV, MXLAB, SEEVAL, SmartSensor and The Embedded Control Solutions Company are registered trademarks of Microchip Technology Incorporated in the U.S.A. Analog-for-the-Digital Age, Application Maestro, dsPICDEM, dsPICDEM.net, dsPICworks, ECAN, ECONOMONITOR, FanSense, FlexROM, fuzzyLAB, In-Circuit Serial Programming, ICSP, ICEPIC, Linear Active Thermistor, MPASM, MPLIB, MPLINK, MPSIM, PICkit, PICDEM, PICDEM.net, PICLAB, PICtail, PowerCal, PowerInfo, PowerMate, PowerTool, Real ICE, rfLAB, rfPICDEM, Select Mode, Smart Serial, SmartTel, Total Endurance, UNI/O, WiperLock and Zena are trademarks of Microchip Technology Incorporated in the U.S.A. and other countries. SQTP is a service mark of Microchip Technology Incorporated in the U.S.A. All other trademarks mentioned herein are property of their respective companies. 2006, Microchip Technology Incorporated, Printed in the U.S.A., All Rights Reserved. Printed on recycled paper.
Microchip received ISO/TS-16949:2002 quality system certification for its worldwide headquarters, design and wafer fabrication facilities in Chandler and Tempe, Arizona and Mountain View, California in October 2003. The Companys quality system processes and procedures are for its PICmicro 8-bit MCUs, KEELOQ code hopping devices, Serial EEPROMs, microperipherals, nonvolatile memory and analog products. In addition, Microchips quality system for the design and manufacture of development systems is ISO 9001:2000 certified.
DS70046E-page ii
M
Table of Contents
PAGE
SECTION 1. INTRODUCTION 1-1
Introduction ...................................................................................................................................................... 1-2 Manual Objective ............................................................................................................................................. 1-2 Device Structure ............................................................................................................................................... 1-3 Development Support ...................................................................................................................................... 1-4 Style and Symbol Conventions ........................................................................................................................ 1-4 Related Documents .......................................................................................................................................... 1-6 Revision History ............................................................................................................................................... 1-7
SECTION 2. CPU
2-1
Introduction ...................................................................................................................................................... 2-2 Programmers Model ........................................................................................................................................ 2-4 Software Stack Pointer ..................................................................................................................................... 2-8 CPU Register Descriptions ............................................................................................................................. 2-11 Arithmetic Logic Unit (ALU) ............................................................................................................................ 2-17 DSP Engine .................................................................................................................................................... 2-18 Divide Support ................................................................................................................................................ 2-27 Instruction Flow Types ................................................................................................................................... 2-27 Loop Constructs ............................................................................................................................................. 2-30 Address Register Dependencies .................................................................................................................... 2-35 Register Maps ................................................................................................................................................ 2-38 Related Application Notes .............................................................................................................................. 2-40 Revision History ............................................................................................................................................. 2-41
SECTION 3. DATA MEMORY
3-1
Introduction ...................................................................................................................................................... 3-2 Data Space Address Generator Units (AGUs) ................................................................................................. 3-5 Modulo Addressing .......................................................................................................................................... 3-7 Bit-Reversed Addressing ............................................................................................................................... 3-14 Control Register Descriptions ......................................................................................................................... 3-18 Related Application Notes .............................................................................................................................. 3-23 Revision History ............................................................................................................................................. 3-24
SECTION 4. PROGRAM MEMORY
4-1
Program Memory Address Map ....................................................................................................................... 4-2 Program Counter .............................................................................................................................................. 4-4 Data Access from Program Memory ................................................................................................................ 4-4 Program Space Visibility from Data Space ...................................................................................................... 4-8 Program Memory Writes ................................................................................................................................ 4-10 PSV Code Examples ...................................................................................................................................... 4-11 Related Application Notes .............................................................................................................................. 4-12 Revision History ............................................................................................................................................. 4-13
70046E-page iii
M
Table of Contents
PAGE
SECTION 5. FLASH AND EEPROM PROGRAMMING 5-1
Introduction ...................................................................................................................................................... 5-2 Table Instruction Operation .............................................................................................................................. 5-2 Control Registers ............................................................................................................................................. 5-5 Run-Time Self-Programming (RTSP) ............................................................................................................. 5-10 Data EEPROM Programming ........................................................................................................................ 5-15 Design Tips .................................................................................................................................................... 5-21 Related Application Notes .............................................................................................................................. 5-22 Revision History ............................................................................................................................................. 5-23
SECTION 6. RESET INTERRUPTS
6-1
Introduction ...................................................................................................................................................... 6-2 Non-Maskable Traps ........................................................................................................................................ 6-6 Interrupt Processing Timing ........................................................................................................................... 6-11 Interrupt Control and Status Registers ........................................................................................................... 6-14 Interrupt Setup Procedures ............................................................................................................................ 6-42 Design Tips .................................................................................................................................................... 6-44 Related Application Notes .............................................................................................................................. 6-45 Revision History ............................................................................................................................................. 6-46
SECTION 7. OSCILLATOR
7-1
Introduction ...................................................................................................................................................... 7-2 Device Clocking and MIPS ............................................................................................................................... 7-5 Oscillator Configuration .................................................................................................................................... 7-6 Oscillator Control Registers OSCCON and OSCTUN ................................................................................. 7-13 Primary Oscillator ........................................................................................................................................... 7-20 Crystal Oscillators/Ceramic Resonators ......................................................................................................... 7-22 Determining Best Values for Crystals, Clock Mode, C1, C2 and Rs .............................................................. 7-24 External Clock Input ....................................................................................................................................... 7-25 External RC Oscillator .................................................................................................................................... 7-26 Phase Locked Loop (PLL) .............................................................................................................................. 7-30 Low-Power 32 kHz Crystal Oscillator ............................................................................................................. 7-31 Oscillator Start-up Timer (OST) ...................................................................................................................... 7-31 Internal Fast RC Oscillator (FRC) .................................................................................................................. 7-31 Internal Low-Power RC (LPRC) Oscillator ..................................................................................................... 7-32 Fail-Safe Clock Monitor (FSCM) .................................................................................................................... 7-32 Programmable Oscillator Postscaler .............................................................................................................. 7-33 Clock Switching Operation ............................................................................................................................. 7-34 Design Tips .................................................................................................................................................... 7-38 Related Application Notes .............................................................................................................................. 7-39 Revision History ............................................................................................................................................. 7-40
70046E-page iv
M
Table of Contents
PAGE
SECTION 8. RESET 8-1
Introduction ...................................................................................................................................................... 8-2 Clock Source Selection at Reset ...................................................................................................................... 8-5 POR: Power-on Reset ...................................................................................................................................... 8-5 External Reset (EXTR) ..................................................................................................................................... 8-7 Software RESET Instruction (SWR) ................................................................................................................. 8-7 Watchdog Time-out Reset (WDTR) ................................................................................................................. 8-7 Brown-out Reset (BOR) ................................................................................................................................... 8-8 Using the RCON Status Bits .......................................................................................................................... 8-10 Device Reset Times ....................................................................................................................................... 8-11 Device Start-up Time Lines ............................................................................................................................ 8-13 Special Function Register Reset States ......................................................................................................... 8-16 Design Tips .................................................................................................................................................... 8-17 Related Application Notes .............................................................................................................................. 8-18 Revision History ............................................................................................................................................. 8-19
SECTION 9. LOW VOLTAGE DETECT (LVD)
9-1
Introduction ...................................................................................................................................................... 9-2 LVD Operation ................................................................................................................................................. 9-5 Design Tips ...................................................................................................................................................... 9-6 Related Application Notes ................................................................................................................................ 9-7 Revision History ............................................................................................................................................... 9-8
SECTION 10. WATCHDOG TIMER AND POWER SAVING MODES
10-1
Introduction .................................................................................................................................................... 10-2 Power Saving Modes ..................................................................................................................................... 10-2 Sleep Mode .................................................................................................................................................... 10-2 Idle Mode ....................................................................................................................................................... 10-4 Interrupts Coincident with Power Save Instructions ....................................................................................... 10-5 Watchdog Timer ............................................................................................................................................. 10-6 Peripheral Module Disable (PMD) Registers .................................................................................................. 10-9 Design Tips .................................................................................................................................................. 10-10 Related Application Notes ............................................................................................................................ 10-11 Revision History ........................................................................................................................................... 10-12
SECTION 11. I/O PORTS
11-1
Introduction .................................................................................................................................................... 11-2 I/O Port Control Registers .............................................................................................................................. 11-3 Peripheral Multiplexing ................................................................................................................................... 11-4 Port Descriptions ............................................................................................................................................ 11-6 Change Notification (CN) Pins ....................................................................................................................... 11-7 CN Operation in Sleep and Idle Modes .......................................................................................................... 11-8 Related Application Notes ............................................................................................................................ 11-11 Revision History ........................................................................................................................................... 11-12
70046E-page v
M
Table of Contents
PAGE
SECTION 12. TIMERS 12-1
Introduction .................................................................................................................................................... 12-2 Timer Variants ................................................................................................................................................ 12-3 Control Registers ........................................................................................................................................... 12-6 Modes of Operation ........................................................................................................................................ 12-9 Timer Prescalers .......................................................................................................................................... 12-14 Timer Interrupts ............................................................................................................................................ 12-14 Reading and Writing 16-bit Timer Module Registers .................................................................................... 12-15 Low Power 32 kHz Crystal Oscillator Input .................................................................................................. 12-15 32-bit Timer Configuration ............................................................................................................................ 12-16 32-bit Timer Modes of Operation ................................................................................................................. 12-18 Reading and Writing into 32-bit Timers ........................................................................................................ 12-21 Timer Operation in Power Saving States ..................................................................................................... 12-21 Peripherals Using Timer Modules ................................................................................................................ 12-22 Design Tips .................................................................................................................................................. 12-24 Related Application Notes ............................................................................................................................ 12-25 Revision History ........................................................................................................................................... 12-26
SECTION 13. INPUT CAPTURE
13-1
Introduction .................................................................................................................................................... 13-2 Input Capture Registers ................................................................................................................................. 13-3 Timer Selection .............................................................................................................................................. 13-4 Input Capture Event Modes ........................................................................................................................... 13-4 Capture Buffer Operation ............................................................................................................................... 13-8 Input Capture Interrupts ................................................................................................................................. 13-9 UART Autobaud Support ............................................................................................................................... 13-9 Input Capture Operation in Power Saving States ........................................................................................ 13-10 I/O Pin Control .............................................................................................................................................. 13-10 Special Function Registers Associated with the Input Capture Module ....................................................... 13-11 Design Tips .................................................................................................................................................. 13-12 Related Application Notes ............................................................................................................................ 13-13 Revision History ........................................................................................................................................... 13-14
SECTION 14. OUTPUT COMPARE
14-1
Introduction .................................................................................................................................................... 14-2 Output Compare Registers ............................................................................................................................ 14-3 Modes of Operation ........................................................................................................................................ 14-4 Output Compare Operation in Power Saving States .................................................................................... 14-23 I/O Pin Control .............................................................................................................................................. 14-23 Design Tips .................................................................................................................................................. 14-26 Related Application Notes ............................................................................................................................ 14-27 Revision History ........................................................................................................................................... 14-28
70046E-page vi
M
Table of Contents
PAGE
SECTION 15. MOTOR CONTROL PWM 15-1
Introduction .................................................................................................................................................... 15-2 Control Registers ........................................................................................................................................... 15-4 PWM Time Base .......................................................................................................................................... 15-16 PWM Duty Cycle Comparison Units ............................................................................................................. 15-20 Complementary PWM Output Mode ............................................................................................................ 15-26 Dead Time Control ....................................................................................................................................... 15-27 Independent PWM Output Mode .................................................................................................................. 15-30 PWM Output Override .................................................................................................................................. 15-31 PWM Output and Polarity Control ................................................................................................................ 15-34 PWM Fault Pins ........................................................................................................................................... 15-34 PWM Update Lockout .................................................................................................................................. 15-37 PWM Special Event Trigger ......................................................................................................................... 15-38 Operation in Device Power Saving Modes ................................................................................................... 15-38 Special Features for Device Emulation ........................................................................................................ 15-39 Related Application Notes ............................................................................................................................ 15-42 Revision History ........................................................................................................................................... 15-43
SECTION 16. QUADRATURE ENCODER INTERFACE (QEI)
16-1
Module Introduction ........................................................................................................................................ 16-2 Control and Status Registers ......................................................................................................................... 16-4 Programmable Digital Noise Filters ................................................................................................................ 16-9 Quadrature Decoder .................................................................................................................................... 16-10 16-bit Up/Down Position Counter ................................................................................................................. 16-12 Using QEI as an Alternate 16-bit Timer/Counter .......................................................................................... 16-16 Quadrature Encoder Interface Interrupts ..................................................................................................... 16-17 I/O Pin Control .............................................................................................................................................. 16-18 QEI Operation During Power Saving Modes ................................................................................................ 16-19 Effects of a Reset ......................................................................................................................................... 16-19 Design Tips .................................................................................................................................................. 16-21 Related Application Notes ............................................................................................................................ 16-22 Revision History ........................................................................................................................................... 16-23
70046E-page vii
M
Table of Contents
PAGE
SECTION 17. 10-BIT A/D CONVERTER 17-1
Introduction .................................................................................................................................................... 17-2 Control Registers ........................................................................................................................................... 17-4 A/D Result Buffer ........................................................................................................................................... 17-4 A/D Terminology and Conversion Sequence ............................................................................................... 17-11 A/D Module Configuration ............................................................................................................................ 17-13 Selecting the Voltage Reference Source ..................................................................................................... 17-13 Selecting the A/D Conversion Clock ............................................................................................................ 17-13 Selecting Analog Inputs for Sampling .......................................................................................................... 17-14 Enabling the Module .................................................................................................................................... 17-16 Specifying the Sample/Conversion Sequence ............................................................................................. 17-16 How to Start Sampling ................................................................................................................................. 17-17 How to Stop Sampling and Start Conversions ............................................................................................. 17-18 Controlling Sample/Conversion Operation ................................................................................................... 17-29 Specifying How Conversion Results are Written Into the Buffer .................................................................. 17-30 Conversion Sequence Examples ................................................................................................................. 17-31 A/D Sampling Requirements ........................................................................................................................ 17-45 Reading the A/D Result Buffer ..................................................................................................................... 17-46 Transfer Function ......................................................................................................................................... 17-47 A/D Accuracy/Error ...................................................................................................................................... 17-47 Connection Considerations .......................................................................................................................... 17-47 Initialization .................................................................................................................................................. 17-48 A/D Conversion Speeds ............................................................................................................................... 17-49 Operation During Sleep and Idle Modes ...................................................................................................... 17-55 Effects of a Reset ......................................................................................................................................... 17-55 Special Function Registers Associated with the 10-bit A/D Converter ......................................................... 17-56 Design Tips .................................................................................................................................................. 17-57 Related Application Notes ............................................................................................................................ 17-58 Revision History ........................................................................................................................................... 17-59
70046E-page viii
M
Table of Contents
PAGE
SECTION 18. 12-BIT A/D CONVERTER 18-1
Introduction .................................................................................................................................................... 18-2 Control Registers ........................................................................................................................................... 18-4 A/D Result Buffer ........................................................................................................................................... 18-4 A/D Terminology and Conversion Sequence ............................................................................................... 18-10 A/D Module Configuration ............................................................................................................................ 18-11 Selecting the Voltage Reference Source ..................................................................................................... 18-11 Selecting the A/D Conversion Clock ............................................................................................................ 18-12 Selecting Analog Inputs for Sampling .......................................................................................................... 18-12 Enabling the Module .................................................................................................................................... 18-14 How to Start Sampling ................................................................................................................................. 18-14 How to Stop Sampling and Start Conversions ............................................................................................. 18-14 Controlling Sample/Conversion Operation ................................................................................................... 18-19 Specifying How Conversion Results are Written into the Buffer .................................................................. 18-19 Conversion Sequence Examples ................................................................................................................. 18-21 A/D Sampling Requirements ........................................................................................................................ 18-26 Reading the A/D Result Buffer ..................................................................................................................... 18-27 Transfer Function ......................................................................................................................................... 18-28 A/D Accuracy/Error ...................................................................................................................................... 18-28 Connection Considerations .......................................................................................................................... 18-28 Initialization .................................................................................................................................................. 18-29 A/D Conversion Speeds ............................................................................................................................... 18-30 Operation During Sleep and Idle Modes ...................................................................................................... 18-33 Effects of a Reset ......................................................................................................................................... 18-33 Special Function Registers Associated with the 12-bit A/D Converter ......................................................... 18-34 Design Tips .................................................................................................................................................. 18-35 Related Application Notes ............................................................................................................................ 18-36 Revision History ........................................................................................................................................... 18-37
SECTION 19. UART
19-1
Introduction .................................................................................................................................................... 19-2 Control Registers ........................................................................................................................................... 19-3 UART Baud Rate Generator (BRG) ............................................................................................................... 19-8 UART Configuration ..................................................................................................................................... 19-10 UART Transmitter ........................................................................................................................................ 19-11 UART Receiver ............................................................................................................................................ 19-14 Using the UART for 9-bit Communication .................................................................................................... 19-18 Receiving Break Characters ........................................................................................................................ 19-19 Initialization .................................................................................................................................................. 19-20 Other Features of the UART ........................................................................................................................ 19-21 UART Operation During CPU Sleep and Idle Modes ................................................................................... 19-21 Registers Associated with UART Module ..................................................................................................... 19-22 Design Tips .................................................................................................................................................. 19-23 Related Application Notes ............................................................................................................................ 19-24 Revision History ........................................................................................................................................... 19-25
70046E-page ix
M
Table of Contents
PAGE
SECTION 20. SERIAL PERIPHERAL INTERFACE (SPI) 20-1
Introduction .................................................................................................................................................... 20-2 Status and Control Registers ......................................................................................................................... 20-4 Modes of Operation ........................................................................................................................................ 20-7 SPI Master Mode Clock Frequency .............................................................................................................. 20-19 Operation in Power Save Modes ................................................................................................................. 20-20 Special Function Registers Associated with SPI Modules ........................................................................... 20-22 Related Application Notes ............................................................................................................................ 20-23 Revision History ........................................................................................................................................... 20-24
SECTION 21. INTER-INTEGRATED CIRCUIT (I2C)
21-1
Overview ........................................................................................................................................................ 21-2 I2C Bus Characteristics .................................................................................................................................. 21-4 Control and Status Registers ......................................................................................................................... 21-7 Enabling I2C Operation ................................................................................................................................ 21-13 Communicating as a Master in a Single Master Environment ...................................................................... 21-15 Communicating as a Master in a Multi-Master Environment ........................................................................ 21-29 Communicating as a Slave .......................................................................................................................... 21-32 Connection Considerations for I2C Bus ....................................................................................................... 21-47 Module Operation During PWRSAV Instruction ........................................................................................... 21-49 Effects of a Reset ......................................................................................................................................... 21-49 Design Tips .................................................................................................................................................. 21-50 Related Application Notes ............................................................................................................................ 21-51 Revision History ........................................................................................................................................... 21-52
SECTION 22. DATA CONVERTER INTERFACE (DCI)
22-1
Introduction .................................................................................................................................................... 22-2 Control Register Descriptions ......................................................................................................................... 22-2 Codec Interface Basics and Terminology ....................................................................................................... 22-8 DCI Operation .............................................................................................................................................. 22-10 Using the DCI Module .................................................................................................................................. 22-17 Operation in Power Saving Modes ............................................................................................................... 22-28 Registers Associated with DCI ..................................................................................................................... 22-28 Design Tips .................................................................................................................................................. 22-30 Related Application Notes ............................................................................................................................ 22-31 Revision History ........................................................................................................................................... 22-32
70046E-page x
M
Table of Contents
PAGE
SECTION 23. CAN MODULE 23-1
Introduction .................................................................................................................................................... 23-2 Control Registers for the CAN Module ........................................................................................................... 23-2 CAN Module Features .................................................................................................................................. 23-28 CAN Module Implementation ....................................................................................................................... 23-29 CAN Module Operation Modes .................................................................................................................... 23-36 Message Reception ..................................................................................................................................... 23-39 Transmission ................................................................................................................................................ 23-49 Error Detection ............................................................................................................................................. 23-58 CAN Baud Rate ............................................................................................................................................ 23-60 Interrupts ...................................................................................................................................................... 23-64 CAN Capture ................................................................................................................................................ 23-65 CAN Module I/O ........................................................................................................................................... 23-65 Operation in CPU Power Saving Modes ...................................................................................................... 23-66 CAN Protocol Overview ............................................................................................................................... 23-68 Related Application Notes ............................................................................................................................ 23-72 Revision History ........................................................................................................................................... 23-73
SECTION 24. DEVICE CONFIGURATION
24-1
Introduction .................................................................................................................................................... 24-2 Device Configuration Registers ...................................................................................................................... 24-2 Configuration Bit Descriptions ........................................................................................................................ 24-6 Device Identification Registers ....................................................................................................................... 24-7 Related Application Notes .............................................................................................................................. 24-8 Revision History ............................................................................................................................................. 24-9
SECTION 25. DEVELOPMENT TOOL SUPPORT
25-1
Introduction .................................................................................................................................................... 25-2 Microchip Hardware and Language Tools ...................................................................................................... 25-2 Third Party Hardware/Software Tools and Application Libraries ................................................................... 25-6 dsPIC30F Hardware Development Boards .................................................................................................. 25-11 Related Application Notes ............................................................................................................................ 25-15 Revision History ........................................................................................................................................... 25-16
SECTION 26. APPENDIX
26-1
70046E-page xi
M
Table of Contents
PAGE NOTES:
70046E-page xii
1
Introduction
Section 1. Introduction
HIGHLIGHTS
This section of the manual contains the following topics: 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Introduction .................................................................................................................... 1-2 Manual Objective ........................................................................................................... 1-2 Device Structure............................................................................................................. 1-3 Development Support .................................................................................................... 1-4 Style and Symbol Conventions ...................................................................................... 1-4 Related Documents ....................................................................................................... 1-6 Revision History ............................................................................................................. 1-7
DS70048C-page 1-1

1.1 Introduction
Microchip is a leading provider of microcontrollers and analog semiconductors. The companys focus is on products that meet the needs of the embedded control market. We are a leading supplier of: 8-bit general purpose microcontrollers (PICmicro MCUs) dsPIC30F 16-bit microcontrollers Speciality and standard non-volatile memory devices Security devices (KEELOQ) Application specific standard products
Please request a Microchip Product Line Card for a listing of all the interesting products that we have to offer. This literature can be obtained from your local sales office, or downloaded from the Microchip web site (www.microchip.com).
1.2
Manual Objective
PICmicro and dsPIC30F devices are grouped by the size of their Instruction Word and Data Path. The current device families are: 1. 2. 3. 4. 5. Base-Line: Mid-Range: High-End: Enhanced: dsPIC30F: 12-bit Instruction Word length, 8-bit Data Path 14-bit Instruction Word length, 8-bit Data Path 16-bit Instruction Word length, 8-bit Data Path 16-bit Instruction Word length, 8-bit Data Path 24-bit Instruction Word length, 16-bit Data Path
This manual describes the dsPIC30F 16-bit MCU family of devices. This manual explains the operation of the dsPIC30F MCU family architecture and peripheral modules, but does not cover the specifics of each device. The user should refer to the data sheet for device specific information. The information that can be found in the data sheet includes: Device memory map Device pinout and packaging details Device electrical specifications List of peripherals included on the device
Code examples are given throughout this manual. These examples sometimes need to be written as device specific as opposed to family generic, though they are valid for most other devices. Some modifications may be required for devices with variations in register file mappings.
DS70048C-page 1-2
1
1.3 Device Structure
Introduction
Each part of the dsPIC30F device can be placed into one of three groups: 1. 2. 3. CPU Core System Integration Peripherals
1.3.1
CPU Core
The CPU core pertains to the basic features that are required to make the device operate. The sections of the manual related to the CPU core include: 1. 2. 3. 4. 5. CPU Data Memory Program Memory DSP Engine Interrupts
1.3.2
System Integration
System integration functions help to: Decrease system cost Increase system reliability Increase design flexibility The following sections of the manual discuss dsPIC30F system integration functions: 1. 2. 3. 4. 5. 6. Oscillator Reset Low Voltage Detect Watchdog Timer and Power Saving Modes Flash and EEPROM Programming Device Configuration
1.3.3
Peripherals
The dsPIC30F has many peripherals that allow the device to be interfaced to the external world. The peripherals discussed in this manual include: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. I/O Ports Timers Input Capture Module Output Compare Module Quadrature Encoder Interface (QEI) 10-bit A/D Converter 12-bit A/D Converter UART Module SPITM Module I2CTM Module Data Converter Interface (DCI) Module CAN Module
1.3.4
Memory Technology
At the time of this writing, all dsPIC30F devices use Flash program memory technology. The Flash program memory can be electrically erased or programmed.
DS70048C-page 1-3

1.4 Development Support
Microchip offers a wide range of development tools that allow users to efficiently develop and debug application code. Microchips development tools can be broken down into four categories: 1. 2. 3. 4. Code generation Hardware/Software debug Device programmer Product evaluation boards
A full description of each of Microchips development tools is discussed in Section 25. Development Tool Support. As new tools are developed, the latest product briefs and user guides can be obtained from the Microchip web site (www.microchip.com) or from your local Microchip Sales Office. Microchip offers other reference tools to speed the development cycle. These include: Application Notes Reference Designs Microchip web site Local Sales Offices with Field Application Support Corporate Support Line
The Microchip web site lists other sites that may be useful references.
1.5
Style and Symbol Conventions

Throughout this document, certain style and font format conventions are used. Most format conventions imply a distinction should be made for the emphasized text. The MCU industry has many symbols and non-conventional word definitions/abbreviations. Table 1-1 provides a description for many of the conventions contained in this document. Located at the rear of this document, a glossary provides additional word and abbreviation definitions used throughout this manual.
DS70048C-page 1-4
1
1.5.1 Document Conventions
Introduction
Table 1-1 defines some of the symbols and terms used throughout this manual. Table 1-1: Document Conventions Description To force a bit/register to a value of logic 1. To force a bit/register to a value of logic 0. 1) To force a register/bit to its default state. 2) A condition in which the device places itself after a device Reset occurs. Some bits will be forced to 0 (such as interrupt enable bits), while others will be forced to 1 (such as the I/O data direction bits). Designates the number nn in the hexadecimal number system. These conventions are used in the code examples. For example, 0x13F or 13Fh. Designates the number bbbbbbbb in the binary number system. This convention is used in the text and in figures and tables. For example, B10100000. Read-Modify-Write. This is when a register or port is read, then the value is modified, and that value is then written back to the register or port. This action can occur from a single instruction (such as bit set, BSET), or a sequence of instructions. Used to specify a range or the concatenation of registers/bits/pins. One example is TMR3:TMR2, which is the concatenation of two 16-bit registers to form a 32-bit timer value. Concatenation order (left-right) usually specifies a positional relationship (MSb to LSb, higher to lower). Specifies bit(s) locations in a particular register. One example is PTCON<PTMOD1:PTMOD0> (or PTMOD<1:0>), which specifies the register and associated bits or bit positions. Indicates the Least Significant or Most Significant bit in a field. Indicates the Least/Most Significant Byte or Word in a field of bits. Used for code examples, binary numbers and for instruction mnemonics in the text. Used for equations and variables. Used in explanatory text for items called out from a graphic/ equation/example. A Note presents information that we wish to re-emphasize, either to help you avoid a common pitfall, or make you aware of operating differences between some device family members. A Note is always in a shaded box (as below), unless used in a table, where it is at the bottom of the table (as in this table). Note: This is a Note in a shaded note box.
Symbol or Term set clear Reset
0xnn or nnh
Bbbbbbbbb
R-M-W
: (colon)
<>
MSb, MSbit, LSb, LSbit MSByte, MSWord, LSByte, LSWord Courier Font Times Font Times, Bold Font, Italics Note
DS70048C-page 1-5

1.5.2 Electrical Specifications
Throughout this manual, there will be references to electrical specifications and their parameter numbers. Table 1-2 shows the parameter numbering convention for dsPIC30F devices. A parameter number represents a unique set of characteristics and conditions that is consistent between every data sheet, though the actual parameter value may vary from device to device. This manual describes a family of devices and therefore, does not specify the parameter values. The user should refer to the Electrical Specifications section of the device data sheet for the actual parameter values for that device. Table 1-2: Electrical Specification Parameter Numbering Convention Comment DC Specification DC Specification for Analog Peripherals Timing (AC) Specification Device Programming DC Specification Device Programming Timing (AC) Specification
Parameter Number Format DXXX AXXX XXX PDXXX PXXX Legend:
XXX represents a number.
1.6
Related Documents
Microchip, as well as other sources, offers additional documentation which can aid in your development with dsPIC30F MCUs. These lists contain the most common documentation, but other documents may also be available. Please check the Microchip web site (www.microchip.com) for the latest published technical documentation.
1.6.1
Microchip Documentation
The following dsPIC30F documentation is available from Microchip at the time of this writing. Many of these documents provide application specific information that gives actual examples of using, programming and designing with dsPIC30F MCUs. 1. dsPIC30F Programmers Reference Manual (DS70030) The dsPIC30F Programmers Reference Manual provides information about the dsPIC30F programmers model and instruction set. A description of each instruction and syntax examples are provided in this document. 2. dsPIC30F Family Overview (DS70043) This document provides a summary of the available dsPIC30F family variants, including device pinouts, memory sizes and available peripherals. 3. dsPIC30F Data Sheets (DS70082 and DS70083) The data sheets contain device specific information, such as pinout and packaging details, electrical specifications and memory maps.
1.6.2
Third Party Documentation

There are several documents available from third party sources around the world. Microchip does not review these documents for technical accuracy. However, they may be a helpful source for understanding the operation of Microchip dsPIC30F devices. Please refer to the Microchip web site for third party documentation related to the dsPIC30F.
DS70048C-page 1-6
1
1.7 Revision History
Introduction
Revision A
This is the initial released revision of this document.
Revision B
There were no technical content or editorial revisions to this section of the manual, however, this section was updated to reflect Revision B throughout the manual.
Revision C
There were no technical content revisions to this section of the manual, however, this section was updated to reflect Revision C throughout the manual.
DS70048C-page 1-7

NOTES:
DS70048C-page 1-8
2
CPU
Section 2. CPU
HIGHLIGHTS
This section of the manual contains the following topics: 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13 Introduction .................................................................................................................... 2-2 Programmers Model...................................................................................................... 2-4 Software Stack Pointer................................................................................................... 2-8 CPU Register Descriptions .......................................................................................... 2-11 Arithmetic Logic Unit (ALU).......................................................................................... 2-17 DSP Engine ................................................................................................................. 2-18 Divide Support ............................................................................................................. 2-27 Instruction Flow Types ................................................................................................. 2-27 Loop Constructs........................................................................................................... 2-30 Address Register Dependencies ................................................................................. 2-35 Register Maps .............................................................................................................. 2-38 Related Application Notes............................................................................................ 2-40 Revision History ........................................................................................................... 2-41
DS70049C-page 2-1

2.1 Introduction
The dsPIC30F CPU module has a 16-bit (data) modified Harvard architecture with an enhanced instruction set, including significant support for DSP. The CPU has a 24-bit instruction word, with a variable length opcode field. The program counter (PC) is 24-bits wide and addresses up to 4M x 24 bits of user program memory space. A single cycle instruction pre-fetch mechanism is used to help maintain throughput and provides predictable execution. All instructions execute in a single cycle, with the exception of instructions that change the program flow, the double-word move (MOV.D) instruction and the table instructions. Overhead free program loop constructs are supported using the DO and REPEAT instructions, both of which are interruptible at any point. The dsPIC30F devices have sixteen 16-bit working registers in the programmers model. Each of the working registers can act as a data, address, or address offset register. The 16th working register (W15) operates as a software stack pointer for interrupts and calls. The dsPIC30F instruction set has two classes of instructions: the MCU class of instructions and the DSP class of instructions. These two instruction classes are seamlessly integrated into the architecture and execute from a single execution unit. The instruction set includes many Addressing modes and was designed for optimum C compiler efficiency. The data space can be addressed as 32K words or 64 Kbytes and is split into two blocks, referred to as X and Y data memory. Each memory block has its own independent Address Generation Unit (AGU). The MCU class of instructions operate solely through the X memory AGU, which accesses the entire memory map as one linear data space. Certain DSP instructions operate through the X and Y AGUs to support dual operand reads, which splits the data address space into two parts. The X and Y data space boundary is device specific. The upper 32 Kbytes of the data space memory map can optionally be mapped into program space at any 16K program word boundary defined by the 8-bit Program Space Visibility Page (PSVPAG) register. The program to data space mapping feature lets any instruction access program space as if it were data space. Furthermore, RAM may be connected to the program memory bus on devices with an external bus and used to extend the internal data RAM. Overhead free circular buffers (modulo addressing) are supported in both X and Y address spaces. The modulo addressing removes the software boundary checking overhead for DSP algorithms. Furthermore, the X AGU circular addressing can be used with any of the MCU class of instructions. The X AGU also supports bit-reverse addressing to greatly simplify input or output data reordering for radix-2 FFT algorithms. The CPU supports Inherent (no operand), Relative, Literal, Memory Direct, Register Direct and Register Indirect Addressing modes. Each instruction is associated with a predefined Addressing mode group depending upon its functional requirements. As many as 6 Addressing modes are supported for each instruction. For most instructions, the dsPIC30F is capable of executing a data (or program data) memory read, a working register (data) read, a data memory write and a program (instruction) memory read per instruction cycle. As a result, 3 operand instructions can be supported, allowing A+B=C operations to be executed in a single cycle. The DSP engine features a high speed, 17-bit by 17-bit multiplier, a 40-bit ALU, two 40-bit saturating accumulators and a 40-bit bi-directional barrel shifter. The barrel shifter is capable of shifting a 40-bit value up to 15 bits right, or up to 16 bits left, in a single cycle. The DSP instructions operate seamlessly with all other instructions and have been designed for optimal real-time performance. The MAC instruction and other associated instructions can concurrently fetch two data operands from memory while multiplying two W registers. This requires that the data space be split for these instructions and linear for all others. This is achieved in a transparent and flexible manner through dedicating certain working registers to each address space. The dsPIC30F has a vectored exception scheme with up to 8 sources of non-maskable traps and 54 interrupt sources. Each interrupt source can be assigned to one of seven priority levels. A block diagram of the CPU is shown in Figure 2-1.
DS70049C-page 2-2
Section 2. CPU
2
Figure 2-1: dsPIC30F CPU Core Block Diagram
X Address Bus Y Data Bus X Data Bus 16 Interrupt Controller PSV & Table Data Access 24 Control Block 24 16 16 16 Data Latch X Data RAM (4 Kbytes) Address Latch 16
CPU
16
Data Latch Y Data RAM (4 Kbytes) Address Latch Y Address Bus
16
16
24
Address Latch Program Memory (144 Kbytes) Data EEPROM (4 Kbytes) Data Latch 16
PCU PCH PCL Program Counter Loop Stack Control Control Logic Logic
16
16 X RAGU X WAGU
Y AGU
16 ROM Latch
EA MUX
16 IR Literal Data
16
24
16
16 x 16 W Reg Array Instruction Decode & Control Control Signals to Various Blocks Timing Generation 16 16 Divide Support
16
Power-up Timer Oscillator Start-up Timer POR/BOR Reset Watchdog Timer Low Voltage Detect
DSP Engine
OSC1/CLKI
16-bit ALU 16 16
MCLR
VDD, VSS AVDD, AVSS
CAN1, CAN2
10-bit or 12-bit ADC
Input Capture Module
Output Compare Module
I2C
I/O Ports
Timers
DCI
SPI1, SPI2
UART1, UART2
DS70049C-page 2-3

2.2 Programmers Model
The programmers model for the dsPIC30F is shown in Figure 2-2. All registers in the programmers model are memory mapped and can be manipulated directly by instructions. A description of each register is provided in Table 2-1. Table 2-1: Programmers Model Register Descriptions Description Working register array 40-bit DSP Accumulators 23-bit Program Counter ALU and DSP Engine Status register Stack Pointer Limit Value register Table Memory Page Address register Program Space Visibility Page Address register REPEAT Loop Count register DO Loop Count register DO Loop Start Address register DO Loop End Address register Contains DSP Engine and DO Loop control bits
Register(s) Name W0 through W15 ACCA, ACCB PC SR SPLIM TBLPAG PSVPAG RCOUNT DCOUNT DOSTART DOEND CORCON
In addition to the registers contained in the programmers model, the dsPIC30F contains control registers for modulo addressing, bit-reversed addressing and interrupts. These registers are described in subsequent sections of this document. All registers associated with the programmers model are memory mapped, as shown in Table 2-8 on page 2-38.
DS70049C-page 2-4
Section 2. CPU
2
Figure 2-2: Programmers Model
15 0 W0 (WREG) W1 W2 W3 W4 DSP Operand Registers W5 W6 W7 W8 DSP Address Registers W9 W10 W11 W12 PUSH.S and POP.S Shadows W13 Frame Pointer/W14 Stack Ptr/W15 SPLIM 39 DSP Accumulators 22 ACCA ACCB ACCAU ACCBU 31 ACCAH ACCBH 0 0 7 TBLPAG 7 PSVPAG 15 RCOUNT 15 DCOUNT 22 DOSTART 22 DOEND SRH SB OAB SAB DA SRL DC IPL<2:0> 15 CORCON RA N OV 0 Core Control Register SZ C Status Register 0 0 0 0 DO Loop Start Address 0 DO Loop Counter 0 REPEAT Loop Counter 0 Program Space Visibility Page Address 0 Data Table Page Address 0 Stack Pointer Limit 15 ACCAL ACCBL 0 Working/Address Registers
CPU
Program Counter
DO Loop End Address
OA
OB
SA
Note:
DCOUNT, DOSTART and DOEND have one level of shadow registers (not shown) for nested DO loops.
DS70049C-page 2-5

2.2.1 Working Register Array
The 16 working (W) registers can function as data, address or address offset registers. The function of a W register is determined by the Addressing mode of the instruction that accesses it. The dsPIC30F instruction set can be divided into two instruction types: register and file register instructions. Register instructions can use each W register as a data value or an address offset value. For example: MOV MOV ADD 2.2.1.1 W0,W1 ; move contents of W0 to W1 W0,[W1] ; move W0 to address contained in W1 W0,[W4],W5 ; add contents of W0 to contents pointed ; to by W4. Place result in W5.
W0 and File Register Instructions W0 is a special working register because it is the only working register that can be used in file register instructions. File register instructions operate on a specific memory address contained in the instruction opcode and W0. W1-W15 cannot be specified as a target register in file register instructions. The file register instructions provide backward compatibility with existing PICmicro devices which have only one W register. The label WREG is used in the assembler syntax to denote W0 in a file register instruction. For example: MOV ADD Note: WREG,0x0100 0x0100,WREG ; move contents of W0 to address 0x0100 ; add W0 to address 0x0100, store in W0
For a complete description of Addressing modes and instruction syntax, please refer to the dsPIC30F Programmers Reference Manual (DS70032).
2.2.1.2
W Register Memory Mapping Since the W registers are memory mapped, it is possible to access a W register in a file register instruction as shown below: MOV 0x0004, W10 ; equivalent to MOV W2, W10 where 0x0004 is the address in memory of W2. Further, it is also possible to execute an instruction that will attempt to use a W register as both an address pointer and operand destination. For example: MOV where: W1 = 0x1234 W2 = 0x0004 ;[W2] addresses W2 W1,[W2++]
In the example above, the contents of W2 are 0x0004. Since W2 is used as an address pointer, it points to location 0x0004 in memory. W2 is also mapped to this address in memory. Even though this is an unlikely event, it is impossible to detect until run-time. The dsPIC30F ensures that the data write will dominate, resulting in W2 = 0x1234 in the example above. 2.2.1.3 W Registers and Byte Mode Instructions Byte instructions which target the W register array only affect the Least Significant Byte of the target register. Since the working registers are memory mapped, the Least and Most Significant Bytes can be manipulated through byte wide data memory space accesses.
2.2.2
Shadow Registers
Many of the registers in the programmers model have an associated shadow register as shown in Figure 2-2. None of the shadow registers are accessible directly. There are two types of shadow registers: those utilized by the PUSH.S and POP.S instructions and those utilized by the DO instruction.
DS70049C-page 2-6
Section 2. CPU
2
2.2.2.1 PUSH.S and POP.S Shadow Registers The PUSH.S and POP.S instructions are useful for fast context save/restore during a function call or Interrupt Service Routine (ISR). The PUSH.S instruction will transfer the following register values into their respective shadow registers: W0...W3 SR (N, OV, Z , C, DC bits only) The POP.S instruction will restore the values from the shadow registers into these register locations. A code example using the PUSH.S and POP.S instructions is shown below: MyFunction: PUSH.S MOV ADD BTSC BSET POP.S RETURN ; #0x03,W0 ; RAM100 ; SR,#Z ; Flags,#IsZero ; ; Save W registers, MCU status load a literal value into W0 add W0 to contents of RAM100 is the result 0? Yes, set a flag Restore W regs, MCU status
CPU
The PUSH.S instruction will overwrite the contents previously saved in the shadow registers. The shadow registers are only one level in depth, so care must be taken if the shadow registers are to be used for multiple software tasks. The user must ensure that any task using the shadow registers will not be interrupted by a higher priority task that also uses the shadow registers. If the higher priority task is allowed to interrupt the lower priority task, the contents of the shadow registers saved in the lower priority task will be overwritten by the higher priority task. 2.2.2.2 DO Loop Shadow Registers The following registers are automatically saved in shadow registers when a DO instruction is executed: DOSTART DOEND DCOUNT The DO shadow registers are one level in depth, permitting two loops to be automatically nested. Refer to Section 2.9.2.2 DO Loop Nesting for further details.
2.2.3
Uninitialized W Register Reset

The W register array (with the exception of W15) is cleared during all Resets and is considered uninitialized until written to. An attempt to use an uninitialized register as an address pointer will reset the device. A word write must be performed to initialize a W register. A byte write will not affect the initialization detection logic.
DS70049C-page 2-7

2.3 Software Stack Pointer
W15 serves as a dedicated software stack pointer and is automatically modified by exception processing, subroutine calls and returns. However, W15 can be referenced by any instruction in the same manner as all other W registers. This simplifies reading, writing and manipulating the stack pointer (e.g., creating stack frames). Note: In order to protect against misaligned stack accesses, W15<0> is fixed to 0 by the hardware.
W15 is initialized to 0x0800 during all Resets. This address ensures that the stack pointer (SP) will point to valid RAM in all dsPIC30F devices and permits stack availability for non-maskable trap exceptions, which may occur before the SP is initialized by the user software. The user may reprogram the SP during initialization to any location within data space. The stack pointer always points to the first available free word and fills the software stack working from lower towards higher addresses. It pre-decrements for a stack pop (read) and post-increments for a stack push (writes), as shown in Figure 2-3. When the PC is pushed onto the stack, PC<15:0> is pushed onto the first available stack word, then PC<22:16> is pushed into the second available stack location. For a PC push during any CALL instruction, the MSByte of the PC is zero-extended before the push as shown in Figure 2-3. During exception processing, the MSByte of the PC is concatenated with the lower 8 bits of the CPU status register, SR. This allows the contents of SRL to be preserved automatically during interrupt processing. Figure 2-3: Stack Operation for a CALL Instruction 15 Stack Grows Towards Higher Address 0
CALL
SUBR
PC<15:0> B000000000 PC<22:16> <Free Word>
W15 (before CALL) W15 (after CALL)
DS70049C-page 2-8
Section 2. CPU
2
2.3.1 Software Stack Examples
The software stack is manipulated using the PUSH and POP instructions. The PUSH and POP instructions are the equivalent of a MOV instruction with W15 used as the destination pointer. For example, the contents of W0 can be pushed onto the stack by: PUSH W0 This syntax is equivalent to: MOV W0,[W15++] The contents of the top-of-stack can be returned to W0 by: POP W0 This syntax is equivalent to: MOV [--W15],W0 Figure 2-4 through Figure 2-7 show examples of how the software stack is used. Figure 2-4 shows the software stack at device initialization. W15 has been initialized to 0x0800. Furthermore, this example assumes the values 0x5A5A and 0x3636 have been written to W0 and W1, respectively. The stack is pushed for the first time in Figure 2-5 and the value contained in W0 is copied to the stack. W15 is automatically updated to point to the next available stack location (0x0802). In Figure 2-6, the contents of W1 are pushed onto the stack. In Figure 2-7, the stack is popped and the top-of-stack value (previously pushed from W1) is written to W3. Figure 2-4: Stack Pointer at Device Reset 0x0000 W15 0x0800
CPU
0xFFFE W15 = 0x0800 W0 = 0x5A5A W1 = 0x3636 Figure 2-5: Stack Pointer After the First PUSH Instruction 0x0000 0x5A5A W15 0x0800 0x0802 PUSH W0
0xFFFE W15 = 0x0802 W0 = 0x5A5A W1 = 0x3636 Figure 2-6: Stack Pointer After the Second PUSH Instruction 0x0000 0x5A5A 0x3636 W15 0x0800 0x0802 0x0804 0xFFFE W15 = 0x0804 W0 = 0x5A5A W1 = 0x3636 PUSH W1
DS70049C-page 2-9

Figure 2-7: Stack Pointer After a POP Instruction 0x0000 W15 0x05A5A 0x03636 0x0800 0x0802 POP W3
0xFFFE W15 = 0x0802 0x3636 W3
2.3.2
W14 Software Stack Frame Pointer

A frame is a user defined section of memory in the stack that is used by a single subroutine. W14 is a special working register because it can be used as a stack frame pointer with the LNK (link) and ULNK (unlink) instructions. W14 can be used in a normal working register by instructions when it is not used as a frame pointer. Refer to the dsPIC30F Programmers Reference Manual (DS70030) for software examples that use W14 as a stack frame pointer.
2.3.3
Stack Pointer Overflow

There is a stack limit register (SPLIM) associated with the stack pointer that is reset to 0x0000. SPLIM is a 16-bit register, but SPLIM<0> is fixed to 0 because all stack operations must be word aligned. The stack overflow check will not be enabled until a word write to SPLIM occurs, after which time it can only be disabled by a device Reset. All effective addresses generated using W15 as a source or destination are compared against the value in SPLIM. If the contents of the Stack Pointer (W15) are greater than the contents of the SPLIM register by 2 and a push operation is performed, a Stack Error Trap will not occur. The Stack Error Trap will occur on a subsequent push operation. Thus, for example, if it is desirable to cause a Stack Error Trap when the stack grows beyond address 0x2000 in RAM, initialize the SPLIM with the value, 0x1FFE. Note: A Stack Error Trap may be caused by any instruction that uses the contents of the W15 register to generate an effective address (EA). Thus, if the contents of W15 are greater than the contents of the SPLIM register by 2, and a CALL instruction is executed, or if an interrupt occurs, a Stack Error Trap will be generated.
If stack overflow checking has been enabled, a stack error trap will also occur if the W15 effective address calculation wraps over the end of data space (0xFFFF). Note: A write to the Stack Pointer Limit register, SPLIM, should not be followed by an indirect read operation using W15.
Refer to Section 6. Reset Interrupts for more information on the stack error trap.
2.3.4
Stack Pointer Underflow

The stack is initialized to 0x0800 during Reset. A stack error trap will be initiated should the stack pointer address ever be less than 0x0800. Note: Locations in data space between 0x0000 and 0x07FF are, in general, reserved for core and peripheral special function registers.
DS70049C-page 2-10
Section 2. CPU
2
2.4 2.4.1 CPU Register Descriptions SR: CPU Status Register
CPU
The dsPIC30F CPU has a 16-bit status register (SR), the LSByte of which is referred to as the lower status register (SRL). The upper byte of SR is referred to as SRH. A detailed description of SR is shown in Register 2-1. SRL contains all the MCU ALU operation status flags, plus the CPU interrupt priority status bits, IPL<2:0> and the REPEAT loop active status bit, RA (SR<4>). During exception processing, SRL is concatenated with the MSByte of the PC to form a complete word value, which is then stacked. SRH contains the DSP Adder/Subtractor status bits, the DO loop active bit, DA (SR<9>) and the Digit Carry bit, DC (SR<8>). The SR bits are readable/writable with the following exceptions: 1. 2. 3. 4. The DA bit (SR<8>): DA is a read only bit. The RA bit (SR<4>): RA is a read only bit. The OA, OB (SR<15:14>) and OAB (SR<11>) bits: These bits are read only and can only be modified by the DSP engine hardware. The SA, SB (SR<13:12>) and SAB (SR<10>) bits: These are read and clear only and can only be set by the DSP engine hardware. Once set, they remain set until cleared by the user, irrespective of the results from any subsequent DSP operations. Note: Clearing the SAB bit will also clear both the SA and SB bits.
Note:
A description of the SR bits affected by each instruction is provided in the dsPIC30F Programmers Reference Manual (DS70030).
2.4.2
CORCON: Core Control Register

The CORCON register contains bits that control the operation of the DSP multiplier and DO loop hardware. The CORCON register also contains the IPL3 status bit, which is concatenated with IPL<2:0> (SR<7:5>), to form the CPU Interrupt Priority Level.
DS70049C-page 2-11

Register 2-1: Upper Byte: R-0 OA bit 15 SR: CPU Status Register R-0 OB R/C-0 SA R/C-0 SB R-0 OAB R/C-0 SAB R -0 DA R/W-0 DC bit 8
Lower Byte: (SRL) R/W-0(2) R/W-0(2) IPL<2:0> bit 7 bit 15
R/W-0(2)
R-0 RA
R/W-0 N
R/W-0 OV
R/W-0 Z
R/W-0 C bit 0
OA: Accumulator A Overflow Status bit 1 = Accumulator A overflowed 0 = Accumulator A has not overflowed OB: Accumulator B Overflow Status bit 1 = Accumulator B overflowed 0 = Accumulator B has not overflowed SA: Accumulator A Saturation Sticky Status bit 1 = Accumulator A is saturated or has been saturated at some time 0 = Accumulator A is not saturated Note: This bit may be read or cleared (not set). SB: Accumulator B Saturation Sticky Status bit 1 = Accumulator B is saturated or has been saturated at some time 0 = Accumulator B is not saturated Note: This bit may be read or cleared (not set). OAB: OA || OB Combined Accumulator Overflow Status bit 1 = Accumulators A or B have overflowed 0 = Neither Accumulators A or B have overflowed SAB: SA || SB Combined Accumulator Sticky Status bit 1 = Accumulators A or B are saturated or have been saturated at some time in the past 0 = Neither Accumulator A or B are saturated Note: This bit may be read or cleared (not set). Clearing this bit will clear SA and SB. DA: DO Loop Active bit 1 = DO loop in progress 0 = DO loop not in progress DC: MCU ALU Half Carry/Borrow bit 1 = A carry-out from the 4th low order bit (for byte-sized data) or 8th low order bit (for word-sized data) of the result occurred 0 = No carry-out from the 4th low order bit (for byte-sized data) or 8th low order bit (for word-sized data) of the result occurred
bit 14
bit 13
bit 12
bit 11
bit 10
bit 9
bit 8
DS70049C-page 2-12
Section 2. CPU
2
Register 2-1: bit 7-5 SR: CPU Status Register (Continued) IPL<2:0>: CPU Interrupt Priority Level Status bits(1) 111 = CPU Interrupt Priority Level is 7 (15). User interrupts disabled. 110 = CPU Interrupt Priority Level is 6 (14) 101 = CPU Interrupt Priority Level is 5 (13) 100 = CPU Interrupt Priority Level is 4 (12) 011 = CPU Interrupt Priority Level is 3 (11) 010 = CPU Interrupt Priority Level is 2 (10) 001 = CPU Interrupt Priority Level is 1 (9) 000 = CPU Interrupt Priority Level is 0 (8) Note 1: The IPL<2:0> bits are concatenated with the IPL<3> bit (CORCON<3>) to form the CPU Interrupt Priority Level. The value in parentheses indicates the IPL if IPL<3> = 1. User interrupts are disabled when IPL<3> = 1. 2: The IPL<2:0> status bits are read only when NSTDIS = 1 (INTCON1<15>). bit 4 RA: REPEAT Loop Active bit 1 = REPEAT loop in progress 0 = REPEAT loop not in progress N: MCU ALU Negative bit 1 = Result was negative 0 = Result was non-negative (zero or positive) OV: MCU ALU Overflow bit This bit is used for signed arithmetic (2s complement). It indicates an overflow of the magnitude which causes the sign bit to change state. 1 = Overflow occurred for signed arithmetic (in this arithmetic operation) 0 = No overflow occurred Z: MCU ALU Zero bit 1 = An operation which effects the Z bit has set it at some time in the past 0 = The most recent operation which effects the Z bit has cleared it (i.e., a non-zero result) C: MCU ALU Carry/Borrow bit 1 = A carry-out from the Most Significant bit of the result occurred 0 = No carry-out from the Most Significant bit of the result occurred Legend: R = Readable bit C = Clear only bit 1 = Bit is set W = Writable bit S = Set only bit 0 = Bit is cleared U = Unimplemented bit, read as 0 -n = Value at POR x = Bit is unknown
CPU
bit 3
bit 2
bit 1
bit 0
DS70049C-page 2-13

Register 2-2: Upper Byte: U-0 bit 15 CORCON: Core Control Register U-0 U-0 R/W-0 US R/W-0 EDT R-0 R-0 DL<2:0> R-0 bit 8 Lower Byte: R/W-0 SATA bit 7
R/W-0 SATB
R/W-1 SATDW
R/W-0 ACCSAT
R/C-0 IPL3
R/W-0 PSV
R/W-0 RND
R/W-0 IF bit 0
bit 15-13 Unimplemented: Read as '0 bit 12 US: DSP Multiply Unsigned/Signed Control bit 1 = DSP engine multiplies are unsigned 0 = DSP engine multiplies are signed EDT: Early DO Loop Termination Control bit 1 = Terminate executing DO loop at end of current loop iteration 0 = No effect Note: bit 10-8 This bit will always read as 0. DL<2:0>: DO Loop Nesting Level Status bits 111 = 7 DO loops active 001 = 1 DO loop active 000 = 0 DO loops active SATA: AccA Saturation Enable bit 1 = Accumulator A saturation enabled 0 = Accumulator A saturation disabled SATB: AccB Saturation Enable bit 1 = Accumulator B saturation enabled 0 = Accumulator B saturation disabled SATDW: Data Space Write from DSP Engine Saturation Enable bit 1 = Data space write saturation enabled 0 = Data space write saturation disabled ACCSAT: Accumulator Saturation Mode Select bit 1 = 9.31 saturation (super saturation) 0 = 1.31 saturation (normal saturation) IPL3: CPU Interrupt Priority Level Status bit 3 1 = CPU interrupt priority level is greater than 7 0 = CPU interrupt priority level is 7 or less Note: The IPL3 bit is concatenated with the IPL<2:0> bits (SR<7:5>) to form the CPU interrupt priority level.
bit 11
bit 7
bit 6
bit 5
bit 4
bit 3
DS70049C-page 2-14
Section 2. CPU
2
Register 2-2: bit 2 CORCON: Core Control Register (Continued) PSV: Program Space Visibility in Data Space Enable bit 1 = Program space visible in data space 0 = Program space not visible in data space RND: Rounding Mode Select bit 1 = Biased (conventional) rounding enabled 0 = Unbiased (convergent) rounding enabled IF: Integer or Fractional Multiplier Mode Select bit 1 = Integer mode enabled for DSP multiply ops 0 = Fractional mode enabled for DSP multiply ops Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared C = Bit can be cleared x = Bit is unknown
CPU
bit 1
bit 0
DS70049C-page 2-15

2.4.3 Other dsPIC30F CPU Control Registers
The registers listed below are associated with the dsPIC30F CPU core, but are described in further detail in other sections of this manual. 2.4.3.1 TBLPAG: Table Page Register The TBLPAG register is used to hold the upper 8 bits of a program memory address during table read and write operations. Table instructions are used to transfer data between program memory space and data memory space. Refer to Section 4. Program Memory for further details. 2.4.3.2 PSVPAG: Program Space Visibility Page Register Program space visibility allows the user to map a 32-Kbyte section of the program memory space into the upper 32 Kbytes of data address space. This feature allows transparent access of constant data through dsPIC30F instructions that operate on data memory. The PSVPAG register selects the 32 Kbyte region of program memory space that is mapped to the data address space. Refer to Section 4. Program Memory for more information on the PSVPAG register. 2.4.3.3 MODCON: Modulo Control Register The MODCON register is used to enable and configure modulo addressing (circular buffers). Refer to Section 3. Data Memory for further details on modulo addressing. 2.4.3.4 XMODSRT, XMODEND: X Modulo Start and End Address Registers The XMODSRT and XMODEND registers hold the start and end addresses for modulo (circular) buffers implemented in the X data memory address space. Refer to Section 3. Data Memory for further details on modulo addressing. 2.4.3.5 YMODSRT, YMODEND: Y Modulo Start and End Address Registers The YMODSRT and YMODEND registers hold the start and end addresses for modulo (circular) buffers implemented in the Y data memory address space. Refer to Section 3. Data Memory for further details on modulo addressing. 2.4.3.6 XBREV: X Modulo Bit-Reverse Register The XBREV register is used to set the buffer size used for bit-reversed addressing. Refer to Section 3. Data Memory for further details on bit-reversed addressing. 2.4.3.7 DISICNT: Disable Interrupts Count Register The DISICNT register is used by the DISI instruction to disable interrupts of priority 1-6 for a specified number of cycles. See Section 6. Reset Interrupts for further information.
DS70049C-page 2-16
Section 2. CPU
2
2.5 Arithmetic Logic Unit (ALU)
The dsPIC30F ALU is 16-bits wide and is capable of addition, subtraction, single bit shifts and logic operations. Unless otherwise mentioned, arithmetic operations are 2s complement in nature. Depending on the operation, the ALU may affect the values of the Carry (C), Zero (Z), Negative (N), Overflow (OV) and Digit Carry (DC) status bits in the SR register. The C and DC status bits operate as a Borrow and Digit Borrow bits, respectively, for subtraction operations. The ALU can perform 8-bit or 16-bit operations, depending on the mode of the instruction that is used. Data for the ALU operation can come from the W register array or data memory depending on the Addressing mode of the instruction. Likewise, output data from the ALU can be written to the W register array or a data memory location. Refer to the dsPIC30F Programmers Reference Manual (DS70030) for information on the SR bits affected by each instruction, Addressing modes and 8-bit/16-bit Instruction modes. Note 1: Byte operations use the 16-bit ALU and can produce results in excess of 8 bits. However, to maintain backward compatibility with PICmicro devices, the ALU result from all byte operations is written back as a byte (i.e., MSByte not modified), and the SR register is updated based only upon the state of the LSByte of the result. 2: All register instructions performed in Byte mode only affect the LSByte of the W registers. The MSByte of any W register can be modified by using file register instructions that access the memory mapped contents of the W registers.
CPU
2.5.1
Byte to Word Conversion

The dsPIC30F has two instructions that are helpful when mixing 8-bit and 16-bit ALU operations. The sign-extend (SE) instruction takes a byte value in a W register or data memory and creates a sign-extended word value that is stored in a W register. The zero-extend (ZE) instruction clears the 8 MSbs of a word value in a W register or data memory and places the result in a destination W register.
DS70049C-page 2-17

2.6 DSP Engine
The DSP engine is a block of hardware which is fed data from the W register array but contains its own specialized result registers. The DSP engine is driven from the same instruction decoder that directs the MCU ALU. In addition, all operand effective addresses (EAs) are generated in the W register array. Concurrent operation with MCU instruction flow is not possible, though both the MCU ALU and DSP engine resources may be shared by all instructions in the instruction set. The DSP engine consists of the following components: high speed 17-bit x 17-bit multiplier barrel shifter 40-bit adder/subtractor two target accumulator registers rounding logic with Selectable modes saturation logic with Selectable modes
Data input to the DSP engine is derived from one of the following sources: 1. Directly from the W array (registers W4, W5, W6 or W7) for dual source operand DSP instructions. Data values for the W4, W5, W6 and W7 registers are pre-fetched via the X and Y memory data buses. From the X memory data bus for all other DSP instructions.
2.
Data output from the DSP engine is written to one of the following destinations: 1. 2. The target accumulator, as defined by the DSP instruction being executed. The X memory data bus to any location in the data memory address space.
The DSP engine has the capability to perform inherent accumulator to accumulator operations which require no additional data. The MCU shift and multiply instructions use the DSP engine hardware to obtain their results. The X memory data bus is used for data reads and writes in these operations. A block diagram of the DSP engine is shown in Figure 2-8. Note: For detailed code examples and instruction syntax related to this section, refer to the dsPIC30F Programmers Reference Manual (DS70030).
DS70049C-page 2-18
Section 2. CPU
2
Figure 2-8: DSP Engine Block Diagram
CPU
40
Saturation Logic
Round Logic
40-bit Accumulator A 40-bit Accumulator B Saturate Adder Negate
40
16
40
40
40
Barrel Shifter
16
Sign-Extend
Y Data Bus
32 Zero Backfill 32 32
16
17-bit x 17-bit Multiplier/Scaler 16-bit to 17-bit Conversion 16 16
To/From W Array
DS70049C-page 2-19
X Data Bus
40

2.6.1 Data Accumulators
There are two 40-bit data accumulators, ACCA and ACCB, that are the result registers for the DSP instructions listed in Table 2-3. Each accumulator is memory mapped to three registers, where x denotes the particular accumulator: ACCxL: ACCx<15:0> ACCxH: ACCx<31:16> ACCxU: ACCx<39:32> For fractional operations that use the accumulators, the radix point is located to the right of bit 31. The range of fractional values that be stored in each accumulator is -256.0 to (256.0 2-31). For integer operations that use the accumulators, the radix point is located to the right of bit 0. The range of integer values that can be stored in each accumulator is -549,755,813,888 to 549,755,813,887.
2.6.2
Multiplier
The dsPIC30F features a 17-bit x 17-bit multiplier which is shared by both the MCU ALU and the DSP engine. The multiplier is capable of signed or unsigned operation and can support either 1.31 fractional (Q.31) or 32-bit integer results. The multiplier takes in 16-bit input data and converts the data to 17-bits. Signed operands to the multiplier are sign-extended. Unsigned input operands are zero-extended. The 17-bit conversion logic is transparent to the user and allows the multiplier to support mixed sign and unsigned/unsigned multiplication. The IF control bit (CORCON<0>) determines integer/fractional operation for the instructions listed in Table 2-3. The IF bit does not affect MCU multiply instructions listed in Table 2-4, which are always integer operations. The multiplier scales the result one bit to the left for fractional operation. The LSbit of the result is always cleared. The multiplier defaults to Fractional mode for DSP operations at a device Reset. The representation of data in hardware for each of these modes is as follows: Integer data is inherently represented as a signed twos complement value, where the MSbit is defined as a sign bit. Generally speaking, the range of an N-bit twos complement integer is -2N-1 to 2N-1 1. Fractional data is represented as a twos complement fraction where the MSbit is defined as a sign bit and the radix point is implied to lie just after the sign bit (Q.X format). The range of an N-bit twos complement fraction with this implied radix point is -1.0 to (1 21-N). Figure 2-9 and Figure 2-10 illustrate how the multiplier hardware interprets data in Integer and Fractional modes. The range of data in both Integer and Fractional modes is listed in Table 2-2.
DS70049C-page 2-20
Section 2. CPU
2
Figure 2-9: Integer and Fractional Representation of 0x4001
Different Representations of 0x4001
CPU
Integer: 0 -2 15 1 214 0 213 0 212 0 .... 0 0 0 0 0 0 0 0 0 0 1 20
0x4001 = 214 + 20 = 16385
1.15 Fractional: 0 -20 1 0 2-2 0 2-3 0 ... 0 0 0 0 0 0 0 0 0 0 1 2-15
2-1
Implied Radix Point 0x4001 = 2-1 + 2-15 = 0.500030518
Figure 2-10:
Integer and Fractional Representation of 0xC002
Different Representations of 0xC002 Integer: 1 -215 1 214 0 213 0 212 0 .... 0 0 0 0 0 0 0 0 0 1 0 20
0xC002 = -215 + 214 + 20 = -32768 + 16384 + 2 = -16382
1.15 Fractional: 1 -20 1 0

-1
0 2
-3
0 ...
0 2-15
-2
Implied Radix Point 0xC002 = -20 + 2-1 + 2-14 = -1 + 0.5 + 0.000061035 = -0.499938965
DS70049C-page 2-21

Table 2-2: Register Size 16-bit 32-bit 40-bit dsPIC30F Data Ranges Integer Range -32768 to 32767 -2,147,483,648 to 2,147,483,647 -549,755,813,888 to 549,755,813,887 Fraction Range -1.0 to (1.0 2-15) (Q.15 Format) -1.0 to (1.0 2-31) (Q.31 Format) -256.0 to (256.0 2-31) (Q.31 Format with 8 Guard bits) Fraction Resolution 3.052 x 10-5 4.657 x 10-10 4.657 x 10-10
2.6.2.1
DSP Multiply Instructions The DSP instructions that utilize the multiplier are summarized in Table 2-3. Table 2-3: DSP Instructions that Utilize the Multiplier Description Multiply and Add to Accumulator OR Square and Add to Accumulator Multiply and Subtract from Accumulator Multiply Multiply and Negate Result Partial Euclidean Distance Add Partial Euclidean Distance to the Accumulator Algebraic Equivalent a = a + b*c a = a + b2 a = a b*c a = b*c a = -b*c a = (b c)2 a = a + (b c)2
DSP Instruction MAC MSC MPY MPY.N ED EDAC Note:
DSP instructions using the multiplier can operate in Fractional (1.15) or Integer modes.
The US control bit (CORCON<12>) determines whether DSP multiply instructions are signed (default) or unsigned. The US bit does not influence the MCU multiply instructions which have specific instructions for signed or unsigned operation. If the US bit is set, the input operands for instructions shown in Table 2-3 are considered as unsigned values which are always zero-extended into the 17th bit of the multiplier value. 2.6.2.2 MCU Multiply Instructions The same multiplier is used to support the MCU multiply instructions, which include integer 16-bit signed, unsigned and mixed sign multiplies as shown in Table 2-4. All multiplications performed by the MUL instruction produce integer results. The MUL instruction may be directed to use byte or word sized operands. Byte input operands will produce a 16-bit result and word input operands will produce a 32-bit result to the specified register(s) in the W array. Table 2-4: MCU Instructions that Utilize the Multiplier Description Multiply two unsigned integers Multiply two signed integers Multiply a signed integer with an unsigned integer
MCU Instruction MUL/MUL.UU MUL.SS MUL.SU/MUL.US
Note 1: MCU instructions using the multiplier operate only in Integer mode. 2: Result of an MCU multiply is 32-bits long and is stored in a pair of W registers.
DS70049C-page 2-22
Section 2. CPU
2
2.6.3 Data Accumulator Adder/Subtractor
The data accumulators have a 40-bit adder/subtractor with automatic sign extension logic for the multiplier result (if signed). It can select one of two accumulators (A or B) as its pre-accumulation source and post-accumulation destination. For the ADD (accumulator) and LAC instructions, the data to be accumulated or loaded can optionally be scaled via the barrel shifter prior to accumulation. The 40-bit adder/subtractor may optionally negate one of its operand inputs to change the sign of the result (without changing the operands). The negate is used during multiply and subtract (MSC), or multiply and negate (MPY.N) operations. The 40-bit adder/subtractor has an additional saturation block which controls accumulator data saturation, if enabled. 2.6.3.1 Accumulator Status Bits Six Status register bits have been provided to support saturation and overflow. They are located in the CPU Status register, SR, and are listed below: Table 2-5: Status Bit OA OB SA Accumulator Overflow and Saturation Status Bits Location SR<15> SR<14> SR<13> Description Accumulator A overflowed into guard bits (ACCA<39:32>) Accumulator B overflowed into guard bits(ACCB<39:32>) ACCA saturated (bit 31 overflow and saturation) or ACCA overflowed into guard bits and saturated (bit 39 overflow and saturation) ACCB saturated (bit 31 overflow and saturation) or ACCB overflowed into guard bits and saturated (bit 39 overflow and saturation) OA logically ORed with OB SA logically ORed with SB. Clearing SAB will also clear SA and SB.
CPU
SB
SR<12>
OAB SAB
SR<11> SR<10>
The OA and OB bits are read only and are modified each time data passes through the accumulator add/subtract logic. When set, they indicate that the most recent operation has overflowed into the accumulator guard bits (bits 32 through 39). This type of overflow is not catastrophic; the guard bits preserve the accumulator data. The OAB status bit is the logically ORed value of OA and OB. The OA and OB bits, when set, can optionally generate an arithmetic error trap. The trap is enabled by setting the corresponding overflow trap flag enable bit OVATE:OVBTE (INTCON1<10:9>). The trap event allows the user to take immediate corrective action, if desired. The SA and SB bits can be set each time data passes through the accumulator saturation logic. Once set, these bits remain set until cleared by the user. The SAB status bit indicates the logically ORed value of SA and SB. The SA and SB bits will be cleared when SAB is cleared. When set, these bits indicate that the accumulator has overflowed its maximum range (bit 31 for 32-bit saturation or bit 39 for 40-bit saturation) and will be saturated (if saturation is enabled). When saturation is not enabled, the SA and SB bits indicate that a catastrophic overflow has occurred (the sign of the accumulator has been destroyed). If the COVTE (INTCON1<8>) bit is set, SA and SB bits will generate an arithmetic error trap when saturation is disabled. Note: See Section 6. Reset Interrupts for further information on arithmetic warning traps. The user must remember that SA, SB and SAB status bits can have different meanings depending on whether accumulator saturation is enabled. The Accumulator Saturation mode is controlled via the CORCON register.
Note:
DS70049C-page 2-23

2.6.3.2 Saturation and Overflow Modes The device supports three Saturation and Overflow modes. 1. Accumulator 39-bit Saturation: In this mode, the saturation logic loads the maximally positive 9.31 value (0x7FFFFFFFFF), or maximally negative 9.31 value (0x8000000000), into the target accumulator. The SA or SB bit is set and remains set until cleared by the user. This Saturation mode is useful for extending the dynamic range of the accumulator. To configure for this mode of saturation, the ACCSAT(CORCON<4>) bit must be set. Additionally, the SATA and/or SATB (CORCON<7 and/or 6>) bits must be set to enable accumulator saturation. 2. Accumulator 31-bit Saturation: In this mode, the saturation logic loads the maximally positive 1.31 value (0x007FFFFFFF) or maximally negative 1.31 value (0xFF80000000) into the target accumulator. The SA or SB bit is set and remains set until cleared by the user. When this Saturation mode is in effect, the guard bits 32 through 39 are not used, except for sign-extension of the accumulator value. Consequently, the OA, OB or OAB bits in SR will never be set. To configure for this mode of overflow and saturation, the ACCSAT (CORCON<4>) bit must be cleared. Additionally, the SATA and/or SATB (CORCON<7 and/or 6>) bits must be set to enable accumulator saturation. 3. Accumulator Catastrophic Overflow: If the SATA and/or SATB (CORCON<7 and/or 6>) bits are not set, then no saturation operation is performed on the accumulator and the accumulator is allowed to overflow all the way up to bit 39 (destroying its sign). If the COVTE bit (INTCON1<8>) is set, a catastrophic overflow will initiate an arithmetic error trap.
Note that accumulator saturation and overflow detection can only result from the execution of a DSP instruction that modifies one of the two accumulators via the 40-bit DSP ALU. Saturation and overflow detection will not take place when the accumulators are accessed as memory mapped registers via MCU class instructions. Furthermore, the accumulator status bits shown in Table 2-5 will not be modified. However, the MCU status bits (Z, N, C, OV, DC) will be modified depending on the MCU instruction that accesses the accumulator. Note: 2.6.3.3 See Section 6. Reset Interrupts for further information on arithmetic error traps.
Data Space Write Saturation In addition to adder/subtractor saturation, writes to data space can be saturated without affecting the contents of the source accumulator. This feature allows data to be limited while not sacrificing the dynamic range of the accumulator during intermediate calculation stages. Data space write saturation is enabled by setting the SATDW control bit (CORCON<5>). Data space write saturation is enabled by default at a device Reset. The data space write saturation feature works with the SAC and SAC.R instructions. The value held in the accumulator is never modified when these instructions are executed. The hardware takes the following steps to obtain the saturated write result: 1. 2. 3. The read data is scaled based upon the arithmetic shift value specified in the instruction. The scaled data is rounded (SAC.R only). The scaled/rounded value is saturated to a 16-bit result based on the value of the guard bits. For data values greater than 0x007FFF, the data written to memory is saturated to the maximum positive 1.15 value, 0x7FFF. For input data less than 0xFF8000, data written to memory is saturated to the maximum negative 1.15 value, 0x8000.
DS70049C-page 2-24
Section 2. CPU
2
2.6.3.4 Accumulator Write Back The MAC and MSC instructions can optionally write a rounded version of the accumulator that is not the target of the current operation into data space memory. The write is performed across the X-bus into combined X and Y address space. This accumulator write back feature is beneficial in certain FFT and LMS algorithms. The following Addressing modes are supported by the accumulator write back hardware: 1. W13, register direct: The rounded contents of the non-target accumulator are written into W13 as a 1.15 fractional result. [W13]+=2, register indirect with post-increment: The rounded contents of the non-target accumulator are written into the address pointed to by W13 as a 1.15 fraction. W13 is then incremented by 2.
CPU
2.
2.6.4
Round Logic
The round logic can perform a conventional (biased) or convergent (unbiased) round function during an accumulator write (store). The Round mode is determined by the state of the RND (CORCON<1>) bit. It generates a 16-bit, 1.15 data value, which is passed to the data space write saturation logic. If rounding is not indicated by the instruction, a truncated 1.15 data value is stored. The two Rounding modes are shown in Figure 2-11. Conventional rounding takes bit 15 of the accumulator, zero-extends it and adds it to the MSWord excluding the guard or overflow bits (bits 16 through 31). If the LSWord of the accumulator is between 0x8000 and 0xFFFF (0x8000 included), the MSWord is incremented. If the LSWord of the accumulator is between 0x0000 and 0x7FFF, the MSWord is left unchanged. A consequence of this algorithm is that over a succession of random rounding operations, the value will tend to be biased slightly positive. Convergent (or unbiased) rounding operates in the same manner as conventional rounding except when the LSWord equals 0x8000. If this is the case, the LSbit of the MSWord (bit 16 of the accumulator) is examined. If it is 1, the MSWord is incremented. If it is 0, the MSWord is not modified. Assuming that bit 16 is effectively random in nature, this scheme will remove any rounding bias that may accumulate. The SAC and SAC.R instructions store either a truncated (SAC) or rounded (SAC.R) version of the contents of the target accumulator to data memory via the X-bus (subject to data saturation, see Section 2.6.3.3 Data Space Write Saturation). Note that for the MAC class of instructions, the accumulator write back data path is always subject to rounding.
Figure 2-11:
Conventional and Convergent Rounding Modes

Conventional (Biased) 16 15 MSWord 0 MSWord Convergent (Unbiased) 16 15 0
1XXX XXXX XXXX XXXX
1 1000 0000 0000 0000
Round Up (add 1 to MSWord) when: LSWord >= 0x8000
Round Up (add 1 to MSWord) when: 1. LSWord = 0x8000 and bit 16 = 1 2. LSWord > 0x8000
16 15 MSWord
0 MSWord
16 15
0XXX XXXX XXXX XXXX
0 1000 0000 0000 0000
Round Down (add nothing) when: LSWord < 0x8000
Round Down (add nothing) when: 1. LSWord = 0x8000 and bit 16 = 0 2. LSWord < 0x8000
DS70049C-page 2-25

2.6.5 Barrel Shifter
The barrel shifter is capable of performing up to a 16-bit arithmetic right shift, or up to a 16-bit left shift, in a single cycle. The barrel shifter can be used by DSP instructions or MCU instructions for multi-bit shifts. The shifter requires a signed binary value to determine both the magnitude (number of bits) and direction of the shift operation: A positive value will shift the operand right A negative value will shift the operand left A value of 0 will not modify the operand The barrel shifter is 40-bits wide to accommodate the width of the accumulators. A 40-bit output result is provided for DSP shift operations, and a 16-bit result for MCU shift operations. A summary of instructions that use the barrel shifter is provided below in Table 2-6. Table 2-6: Instructions that Utilize the DSP Engine Barrel Shifter Description Arithmetic multi-bit right shift of data memory location Logical multi-bit right shift of data memory location Multi-bit shift left of data memory location Store DSP accumulator with optional shift Shift DSP accumulator
Instruction ASR LSR SL SAC SFTAC
2.6.6
DSP Engine Mode Selection

The various operational characteristics of the DSP engine discussed in previous sub-sections can be selected through the CPU Core Configuration register (CORCON). These are listed below: Fractional or integer multiply operation. Conventional or convergent rounding. Automatic saturation on/off for ACCA. Automatic saturation on/off for ACCB. Automatic saturation on/off for writes to data memory. Accumulator Saturation mode selection.
2.6.7
DSP Engine Trap Events

The various arithmetic error traps that can be generated for handling exceptions in the DSP engine are selected through the Interrupt Control register (INTCON1). These are listed below: Trap on ACCA overflow enable, using OVATE (INTCON1<10>). Trap on ACCB overflow enable, using OVBTE (INTCON1<9>). Trap on catastrophic ACCA and/or ACCB overflow enable, using COVTE (INTCON1<8>). An arithmetic error trap will also be generated when the user attempts to shift a value beyond the maximum allowable range (+/- 16 bits) using the SFTAC instruction. This trap source cannot be disabled. The execution of the instruction will complete, but the results of the shift will not be written to the target accumulator. For further information on bits in the INTCON1 register and arithmetic error traps, please refer to Section 6. Reset Interrupts.
DS70049C-page 2-26
Section 2. CPU
2
2.7 Divide Support
The dsPIC30F supports the following types of division operations: DIVF: 16/16 signed fractional divide DIV.SD: 32/16 signed divide DIV.UD: 32/16 unsigned divide DIV.SW: 16/16 signed divide DIV.UW: 16/16 unsigned divide
CPU
The quotient for all divide instructions is placed in W0, and the remainder in W1. The 16-bit divisor can be located in any W register. A 16-bit dividend can be located in any W register and a 32-bit dividend must be located in an adjacent pair of W registers. All divide instructions are iterative operations and must be executed 18 times within a REPEAT loop. The user is responsible for programming the REPEAT instruction. A complete divide operation takes 19 instruction cycles to execute. The divide flow is interruptible, just like any other REPEAT loop. All data is restored into the respective data registers after each iteration of the loop, so the user will be responsible for saving the appropriate W registers in the ISR. Although they are important to the divide hardware, the intermediate values in the W registers have no meaning to the user. The divide instructions must be executed 18 times in a REPEAT loop to produce a meaningful result. Refer to the dsPIC30F Programmers Reference Manual (DS70030) for more information and programming examples for the divide instructions.
2.8
Instruction Flow Types

Most instructions in the dsPIC30F architecture occupy a single word of program memory and execute in a single cycle. An instruction pre-fetch mechanism facilitates single cycle (1 TCY) execution. However, some instructions take 2 or 3 instruction cycles to execute. Consequently, there are seven different types of instruction flow in the dsPIC architecture. These are described below: 1. 1 Instruction Word, 1 Instruction Cycle: These instructions will take one instruction cycle to execute as shown in Figure 2-12. Most instructions are 1-word, 1-cycle instructions.
Figure 2-12:
Instruction Flow 1-Word, 1-Cycle TCY0 TCY1 Execute 1 Fetch 2 Execute 2 Fetch 3 1 Instruction Word, 2 Instruction Cycles: In these instructions, there is no pre-fetch flush. The only instructions of this type are the MOV.D instructions (load and store double-word). Two cycles are required to complete these instructions, as shown in Figure 2-13. Execute 3 TCY2 TCY3 TCY4 TCY5
1. MOV 2. MOV 3. MOV
#0x55AA,W0 W0,PORTA W0,PORTB 2.
Fetch 1
Figure 2-13:
Instruction Flow 1-Word, 2-Cycle (MOV.D Operation) TCY0 TCY1 Execute 1 Fetch 2 Execute 2 R/W Cycle 1 Fetch 3 Execute 2 R/W Cycle2 No Fetch Execute 3 Fetch 4 Execute 4 TCY2 TCY3 TCY4 TCY5
1. MOV 2. MOV.D 3. MOV
#0x1234,W0 [W0++],W1 #0x00AA,W1
Fetch 1
4. MOV
#0x00CC,W0
DS70049C-page 2-27

3. 1 Instruction Word, 2 or 3 Instruction Cycle Program Flow Changes: These instructions include relative call and branch instructions, and skip instructions. When an instruction changes the PC (other than to increment it), the program memory pre-fetch data must be discarded. This makes the instruction take two effective cycles to execute, as shown in Figure 2-14. Figure 2-14: Instruction Flow 1-Word, 2-Cycle (Program Flow Change) TCY0 1. MOV.B #0x55,W0 2. BTSC PORTA,#3 Fetch 1 TCY1 Execute 1 Fetch 2 (executed as FNOP) (executed as FNOP) Execute 2 Skip Taken Fetch 3 Forced NOP Fetch 4 Execute 4 Fetch 5 Forced NOP Fetch SUB_1 TCY2 TCY3 TCY4 TCY5
3. ADD.B PORTA 4. BRA SUB_1 5. ADD.B PORTB
6. SUB_1: Instruction @ address SUB_1
Three cycles will be taken when a two-word instruction is skipped. In this case, the program memory pre-fetch data is discarded and the second word of the two-word instruction is fetched. The second word of the instruction will be executed as a NOP, as shown in Figure 2-15. Figure 2-15: Instruction Flow 1-Word, 3-Cycle (2-Word Instruction Skipped) TCY0 1. BTSC 2. GOTO SR,#Z LABEL Fetch 1 TCY1 Execute 1, Skip Taken Fetch 2 Forced NOP Fetch 2nd word of GOTO 2nd word executed as a NOP Fetch 3 Execute 3 Fetch 4 Execute 4 TCY2 TCY3 TCY4 TCY5
(GOTO 2nd word)
3. BCLR 4. MOV
PORTB,#3 W0,W1
4.
1 Instruction Word, 3 Instruction Cycles (RETFIE, RETURN, RETLW): The RETFIE, RETURN and RETLW instructions, that are used to return from a subroutine call or an Interrupt Service Routine, take 3 instruction cycles to execute, as shown in Figure 2-16.
Figure 2-16:
Instruction Flow 1-Word, 3-Cycle (RETURN, RETFIE, RETLW) TCY0 TCY1 Execute 1 Fetch 2 Execute 2 Fetch 3 Execute 2 No Fetch Execute 2 Fetch 4 Execute 4 Fetch 5 TCY2 TCY3 TCY4 TCY5
1. MOV
#0x55AA,W0
Fetch 1
2. RETURN 4. MOV W0, W3 5. MOV W3, W5
3. (instruction in old program flow)
(instruction in new program flow)
DS70049C-page 2-28
Section 2. CPU
2
5. Table Read/Write Instructions: These instructions will suspend fetching to insert a read or write cycle to the program memory. The instruction fetched while executing the table operation is saved for 1 cycle and executed in the cycle immediately after the table operation as shown in Figure 2-17. Figure 2-17: Instruction Pipeline Flow Table Operations TCY0 1. MOV #0x1234,W0 2. TBLRDL.w [W0++],W1 3. MOV #0x00AA,W1 Fetch 1 TCY1 Execute 1 Fetch 2 Execute 2 Fetch 3 PM Data Read Cycle Bus Read 4. MOV #0x00CC,W0 Execute 3 Fetch 4 Execute 4 TCY2 TCY3 TCY4 TCY5
CPU
6.
2 Instruction Words, 2 Instruction Cycles: In these instructions, the fetch after the instruction contains data. This results in a 2-cycle instruction as shown in Figure 2-18. The second word of a two-word instruction is encoded so that it will be executed as a NOP, should it be fetched by the CPU without first fetching the first word of the instruction. This is important when a two-word instruction is skipped by a skip instruction (see Figure 2-15).
Figure 2-18:
Instruction Pipeline Flow 2-Word, 2-Cycle TCY0 TCY1 Execute 1 Fetch 2L Update PC Fetch 2H Forced NOP Fetch 3 Execute 3 Fetch 4 Execute 4 TCY2 TCY3 TCY4 TCY5
1. MOV #0xAA55,W0 2. GOTO LABEL 3. LABEL: MOV W0,W2 4. BSET PORTA, #3
Fetch 1
7.
Address Register Dependencies: These are instructions that are subjected to a stall due to a data address dependency between the X-data space read and write operations. An additional cycle is inserted to resolve the resource conflict as discussed in Section 2.10 Address Register Dependencies.
Figure 2-19:
Instruction Pipeline Flow 1-Word, 1-Cycle (With Instruction Stall) TCY0 TCY1 Execute 1 Fetch 2 Execute 1 Stall Execute 2 Fetch 3 Execute 3 TCY2 TCY3 TCY4 TCY5
1. MOV W0,W1 2. MOV [W1],[W4] 3. MOV W2,W1
Fetch 1
DS70049C-page 2-29

2.9 Loop Constructs
The dsPIC30F supports both REPEAT and DO instruction constructs to provide unconditional automatic program loop control. The REPEAT instruction is used to implement a single instruction program loop. The DO instruction is used to implement a multiple instruction program loop. Both instructions use control bits within the CPU Status register, SR, to temporarily modify CPU operation.
2.9.1
Repeat Loop Construct

The REPEAT instruction causes the instruction that follows it to be repeated a number of times. A literal value contained in the instruction or a value in one of the W registers can be used to specify the repeat count value. The W register option enables the loop count to be a software variable. An instruction in a REPEAT loop will be executed at least once. The number of iterations for a repeat loop will be the 14-bit literal value + 1, or Wn + 1. The syntax for the two forms of the REPEAT instruction is given below: REPEAT #lit14 ; RCOUNT <-- lit14 (Valid target Instruction) or REPEAT Wn ; RCOUNT <-- Wn (Valid target Instruction)
2.9.1.1
Repeat Operation The loop count for Repeat operations is held in the 14-bit RCOUNT register, which is memory mapped. RCOUNT is initialized by the REPEAT instruction. The REPEAT instruction sets the Repeat Active, or RA (SR<4>) status bit to 1, if the RCOUNT is a non-zero value. RA is a read only bit and cannot be modified through software. For repeat loop count values greater than 0, the PC is not incremented. Further PC increments are inhibited until RCOUNT = 0. See Figure 2-20 for an instruction flow example of a Repeat loop. For a loop count value equal to 0, REPEAT has the effect of a NOP and the RA (SR<4>) bit is not set. The Repeat loop is essentially disabled before it begins, allowing the target instruction to execute only once while pre-fetching the subsequent instruction (i.e., normal execution flow). Note: The instruction immediately following the REPEAT instruction (i.e., the target instruction) is always executed at least one time. It is always executed one time more than the value specified in the 14-bit literal or the W register operand.
Figure 2-20:
REPEAT Instruction Pipeline Flow TCY0 TCY1 Execute 1 Fetch 2 Execute 2 No Fetch Execute 2 No Fetch Execute 2 Fetch 3 PC X 0 PC+2 2 1 PC+2 1 1 PC+2 0 0 PC+4 0 0 Execute 3 PC+6 0 0 TCY2 TCY3 TCY4 TCY5
1.REPEAT #0x2
Fetch 1
2.MAC W4*W5,A,[W8]+=2,W4
3.BSET
PORTA,#3
PC (at end of instruction) RCOUNT (at end of instruction) RA (at end of instruction)
DS70049C-page 2-30
Section 2. CPU
2
2.9.1.2 Interrupting a REPEAT Loop A REPEAT instruction loop may be interrupted at any time. The RA state is preserved on the stack during exception processing to allow the user to execute further REPEAT loops from within (any number) of nested interrupts. After SRL is stacked, the RA status bit is cleared to restore normal execution flow within the ISR. Note: If a Repeat loop has been interrupted and an ISR is being processed, the user must stack the RCOUNT (Repeat Count register) prior to executing another REPEAT instruction within an ISR.
CPU
Note: If Repeat was used within an ISR, the user must unstack RCOUNT prior to executing RETFIE. Returning into a Repeat loop from an ISR using RETFIE requires no special handling. Interrupts will pre-fetch the repeated instruction during the third cycle of the RETFIE. The stacked RA bit will be restored when the SRL register is popped and, if set, the interrupted Repeat loop will be resumed. Note: Should the repeated instruction (target instruction in the Repeat loop) be accessing data from PS using PSV, the first time it is executed after a return from an exception will require 2 instruction cycles. Similar to the first iteration of a loop, timing limitations will not allow the first instruction to access data residing in PS in a single instruction cycle.
2.9.1.2.1
Early Termination of a Repeat Loop An interrupted Repeat loop can be terminated earlier than normal in the ISR by clearing the RCOUNT register in software.
2.9.1.3
Restrictions on the REPEAT Instruction Any instruction can immediately follow a REPEAT except for the following: 1. 2. 3. 4. Program Flow Control instructions (any branch, compare and skip, subroutine calls, returns, etc.). Another REPEAT or DO instruction. DISI, ULNK, LNK, PWRSAV, RESET. MOV.D instruction. Note: There are some instructions and/or Instruction Addressing modes that can be executed within a Repeat loop, but make little sense when repeated.
DS70049C-page 2-31

2.9.2 DO Loop Construct
The DO instruction can execute a group of instructions that follow it a specified number of times without software overhead. The set of instructions up to and including the end address will be repeated. The repeat count value for the DO instruction can be specified by a 14-bit literal or by the contents of a W register declared within the instruction. The syntax for the two forms of the DO instruction is given below: DO #lit14,LOOP_END Instruction1 Instruction2 : : LOOP_END: Instruction n DO Wn,LOOP_END Instruction1 Instruction2 : : LOOP_END: Instruction n ; DCOUNT <-- lit14
; DCOUNT <-- Wn<13:0>
The following features are provided in the DO loop construct: A W register can be used to specify the loop count. This allows the loop count to be defined at run-time. The instruction execution order need not be sequential (i.e., there can be branches, subroutine calls, etc.). The loop end address does not have to be greater than the start address. 2.9.2.1 DO Loop Registers and Operation The number of iterations executed by a DO loop will be the (14-bit literal value +1) or the (Wn value + 1). If a W register is used to specify the number of iterations, the two MSbits of the W register are not used to specify the loop count. The operation of a DO loop is similar to the do-while construct in the C programming language because the instructions in the loop will always be executed at least once. The dsPIC30F has three registers associated with DO loops: DOSTART, DOEND and DCOUNT. These registers are memory mapped and automatically loaded by the hardware when the DO instruction is executed. DOSTART holds the starting address of the DO loop while DOEND holds the end address of the DO loop. The DCOUNT register holds the number of iterations to be executed by the loop. DOSTART and DOEND are 22-bit registers that hold the PC value. The MSbits and LSbits of these registers is fixed to 0. Refer to Figure 2-2 for further details. The LSbit is not stored in these registers because PC<0> is always forced to 0. The DA status bit (SR<9>) indicates that a single DO loop (or nested DO loops) is active. The DA bit is set when a DO instruction is executed and enables a PC address comparison with the DOEND register on each subsequent instruction cycle. When PC matches the value in DOEND, DCOUNT is decremented. If the DCOUNT register is not zero, the PC is loaded with the address contained in the DOSTART register to start another iteration of the DO loop. The DO loop will terminate when DCOUNT = 0. If there are no other nested DO loops in progress, then the DA bit will also be cleared. Note: The group of instructions in a DO loop construct is always executed at least one time. The DO loop is always executed one time more than the value specified in the literal or W register operand.
DS70049C-page 2-32
Section 2. CPU
2
2.9.2.2 DO Loop Nesting The DOSTART, DOEND and DCOUNT registers each have a shadow register associated with them, such that the DO loop hardware supports one level of automatic nesting. The DOSTART, DOEND and DCOUNT registers are user accessible and they may be manually saved to permit additional nesting, where required. The DO Level bits, DL<2:0> (CORCON<10:8>) indicate the nesting level of the DO loop currently being executed. When the first DO instruction is executed, DL<2:0> is set to B001 to indicate that one level of DO loop is underway. The DA (SR<9>) is also set. When another DO instruction is executed within the first DO loop, the DOSTART, DOEND and DCOUNT registers are transferred into the shadow registers, prior to being updated with the new loop values. The DL<2:0> bits are set to B010 indicating that a second, nested DO loop is in progress. The DA (SR<9>) bit also remains set. If no more than one level of DO loop nesting is required in the application, no special attention is required. Should the user require more than one level of DO loop nesting, this may be achieved through manually saving the DOSTART, DOEND and DCOUNT registers prior to executing the next DO instruction. These registers should be saved whenever DL<2:0> is B010 or greater. The DOSTART, DOEND and DCOUNT registers will automatically be restored from their shadow registers when a DO loop terminates and DL<2:0> = B010. Note: The DL<2:0> (CORCON<10:8>) bits are combined (logically OR-ed) to form the DA (SR<9>) bit. If nested DO loops are being executed, the DA bit is cleared only when the loop count associated with the outer most loop expires.
CPU
2.9.2.3
Interrupting a DO Loop DO loops may be interrupted at any time. If another DO loop is to be executed during the ISR, the user must check the DL<2:0> status bits and save the DOSTART, DOEND and DCOUNT registers as required. No special handling is required if the user can ensure that only one level of DO loop will ever be executed in: both background and any one ISR handler (if interrupt nesting is enabled) or both background and any ISR (if interrupt nesting is disabled) Alternatively, up to two (nested) DO loops may be executed in either background or within any one ISR handler (if interrupt nesting is enabled) or in any ISR (if interrupt nesting is disabled) It is assumed that no DO loops are used within any trap handlers. Returning to a DO loop from an ISR, using the RETFIE instruction, requires no special handling.
2.9.2.4
Early Termination of the DO loop There are two ways to terminate a DO loop, earlier than normal: 1. The EDT (CORCON<11>) bit provides a means for the user to terminate a DO loop before it completes all loops. Writing a 1 to the EDT bit will force the loop to complete the iteration underway and then terminate. If EDT is set during the penultimate or last instruction of the loop, one more iteration of the loop will occur. EDT will always read as a 0; clearing it has no effect. After the EDT bit is set, the user can optionally branch out of the DO loop. Alternatively, the code may branch out of the loop at any point except from the last instruction, which cannot be a flow control instruction. Although the DA bit enables the DO loop hardware, it will have no effect unless the address of the penultimate instruction is encountered during an instruction pre-fetch. This is not a recommended method for terminating a DO loop. Note: Exiting a DO loop without using EDT is not recommended because the hardware will continue to check for DOEND addresses.
2.
DS70049C-page 2-33

2.9.2.5 DO Loop Restrictions DO loops have the following restrictions imposed: choice of last instruction in the loop the loop length (offset from the first instruction) reading of the DOEND register All DO loops must contain at least 2 instructions because the loop termination tests are performed in the penultimate instruction. REPEAT should be used for single instruction loops. The special function register, DOEND, cannot be read by user software in the instruction that immediately follows either a DO instruction, or a file register write operation to the DOEND SFR. The instruction that is executed two instructions before the last instruction in a DO loop should not modify any of the following: CPU priority level governed by the IPL (SR<7:5>) bits Peripheral Interrupt Enable bits governed by the IEC0, IEC1 and IEC2 registers Peripheral Interrupt Priority bits governed by the IPC0 through IPC11 registers If the restrictions above are not followed, the DO loop may execute incorrectly. 2.9.2.5.1 Last Instruction Restrictions There are restrictions on the last instruction executed in a DO loop. The last instruction in a DO loop should not be: 1. 2. 3. 4. 5. 6. 2.9.2.5.2 Flow control instruction (for e.g., any branch, compare and skip, GOTO, CALL, RCALL, TRAP). RETURN, RETFIE and RETLW will work correctly as the last instruction of a DO loop, but the user must be responsible for returning into the loop to complete it. Another REPEAT or DO instruction. Target instruction within a REPEAT loop. This restriction implies that the penultimate instruction also cannot be a REPEAT. Any instruction that occupies two words in program space. DISI instruction
Loop Length Restrictions Loop length is defined as the signed offset of the last instruction from the first instruction in the DO loop. The loop length when added to the address of the first instruction in the loop forms the address of the last instruction of the loop.There are some loop length values that should be avoided. 1. Loop Length = -2 Execution will start at the first instruction in the loop (i.e., at [PC]) and will continue until the loop end address (in this case [PC 4]) is pre-fetched. As this is the first word of the DO instruction, it will execute the DO instruction again, re-initializing the DCOUNT and pre-fetching [PC]. This will continue forever as long as the loop end address [PC 4] is pre-fetched. This value of n has the potential of creating an infinite loop (subject to a Watchdog Timer Reset). end_loop: DO #33, end_loop ;DO is a two-word instruction NOP ;2nd word of DO executes as a NOP ADD W2,W3,W4 ;First instruction in DO loop([PC])
DS70049C-page 2-34
Section 2. CPU
2
2. Loop Length = -1 Execution will start at the first instruction in the loop (i.e., at [PC]) and will continue until the loop end address ([PC 2]) is pre-fetched. Since the loop end address is the second word of the DO instruction, it will execute as a NOP but will still pre-fetch [PC]. The loop will then execute again. This will continue as long as the loop end address [PC 2] is pre-fetched and the loop does not terminate. Should the value in the DCOUNT register reach zero and on a subsequent decrement generate a borrow, the loop will terminate. However, in such a case the initial instruction outside the loop will once again be the first loop instruction. DO #33, end_loop ;DO is a two-word instruction end_loop: NOP ;2nd word of DO executes as a NOP ADD W2,W3,W4 ;First instruction in DO loop([PC]) Loop Length = 0 Execution will start at the first instruction in the loop (i.e., at [PC]) and will continue until the loop end address ([PC]) is pre-fetched. If the loop is to continue, this pre-fetch will cause the DO loop hardware to load the DOEND address ([PC]) into the PC for the next fetch (which will be [PC] again). After the first true iteration of the loop, the first instruction in the loop will be executed repeatedly until the loop count underflows and the loop terminates. When this occurs, the initial instruction outside the loop will be the instruction after [PC]. DO #33, end_loop ;DO is a two-word instruction NOP ;2nd word of DO executes as a NOP end_loop: ADD W2,W3,W4 ;First instruction in DO loop([PC])
CPU
3.
2.10
Address Register Dependencies

The dsPIC30F architecture supports a data space read (source) and a data space write (destination) for most MCU class instructions. The effective address (EA) calculation by the AGU and subsequent data space read or write, each take a period of 1 instruction cycle to complete. This timing causes the data space read and write operations for each instruction to partially overlap, as shown in Figure 2-21. Because of this overlap, a Read-After-Write (RAW) data dependency can occur across instruction boundaries. RAW data dependencies are detected and handled at run-time by the dsPIC30F CPU. Figure 2-21: Data Space Access Timing
ADD MOV 1 Instruction Cycle (TCY) TCY0 Instruction Register Contents X-Space RAGU [W7] TCY1 TCY2 W0, [W7], [W10] [W8], [W9]++
ADD
MOV
[W8]
X-Space WAGU
[W10]
[W9]++
X-Space Address
W7
W10
W8
W9
DS70049C-page 2-35

2.10.1 Read-After-Write Dependency Rules
If the W register is used as a write operation destination in the current instruction and the W register being read in the pre-fetched instruction are the same, the following rules will apply: 1. If the destination write (current instruction) does not modify the contents of Wn, no stalls will occur. or If the source read (pre-fetched instruction) does not calculate an EA using Wn, no stalls will occur.
2.
During each instruction cycle, the dsPIC30F hardware automatically checks to see if a RAW data dependency is about to occur. If the conditions specified above are not satisfied, the CPU will automatically add a one instruction cycle delay before executing the pre-fetched instruction. The instruction stall provides enough time for the destination W register write to take place before the next (pre-fetched) instruction has to use the written data. Table 2-7: Read-After-Write Dependency Summary Source Addressing Mode using Wn Direct Indirect Indirect with modification Direct Indirect Indirect with modification Direct Indirect Status Allowed Stall Stall Allowed Allowed Allowed Allowed Stall ADD.w MOV.w ADD.w MOV.w ADD.w MOV.w ADD.w MOV.w ADD.w MOV.w ADD.w MOV.w ADD.w MOV.w Examples (Wn = W2) W0, W1, W2 W2, W3 W0, W1, W2 [W2], W3 W0, W1, W2 [W2++], W3 W0, W1, [W2] W2, W3 W0, W1, [W2] [W2], W3 W0, W1, [W2] [W2++], W3 W0, W1, [W2++] W2, W3
Destination Addressing Mode using Wn Direct Direct Direct Indirect Indirect Indirect Indirect with modification Indirect
ADD.w W0, W1, [W2] MOV.w [W2], W3 ; W2=0x0004 (mapped W2) ADD.w W0, W1, [W2] MOV.w [W2++], W3 ; W2=0x0004 (mapped W2) ADD.w MOV.w ADD.w MOV.w W0, W1, [W2++] [W2], W3 W0, W1, [W2++] [W2++], W3
Indirect
Indirect with modification Indirect Indirect with modification
Stall
Indirect with modification Indirect with modification
Stall Stall
2.10.2
Instruction Stall Cycles

An instruction stall is essentially a one instruction cycle wait period appended in front of the read phase of an instruction, in order to allow the prior write to complete before the next read operation. For the purposes of interrupt latency, it should be noted that the stall cycle is associated with the instruction following the instruction where it was detected (i.e., stall cycles always precede instruction execution cycles).
DS70049C-page 2-36
Section 2. CPU
2
If a RAW data dependency is detected, the dsPIC30F will begin an instruction stall. During an instruction stall, the following events occur: 1. 2. 3. 4. 2.10.2.1 The write operation underway (for the previous instruction) is allowed to complete as normal. Data space is not addressed until after the instruction stall. PC increment is inhibited until after the instruction stall. Further instruction fetches are inhibited until after the instruction stall.
CPU
Instruction Stall Cycles and Interrupts When an interrupt event coincides with two adjacent instructions that will cause an instruction stall, one of two possible outcomes could occur: 1. The interrupt could be coincident with the first instruction. In this situation, the first instruction will be allowed to complete and the second instruction will be executed after the ISR completes. In this case, the stall cycle is eliminated from the second instruction because the exception process provides time for the first instruction to complete the write phase. The interrupt could be coincident with the second instruction. In this situation, the second instruction and the appended stall cycle will be allowed to execute prior to the ISR. In this case, the stall cycle associated with the second instruction executes normally. However, the stall cycle will be effectively absorbed into the exception process timing. The exception process proceeds as if an ordinary two-cycle instruction was interrupted.
2.
2.10.2.2
Instruction Stall Cycles and Flow Change Instructions The CALL and RCALL instructions write to the stack using W15 and may, therefore, force an instruction stall prior to the next instruction, if the source read of the next instruction uses W15. The RETFIE and RETURN instructions can never force an instruction stall prior to the next instruction because they only perform read operations. However, the user should note that the RETLW instruction could force a stall, because it writes to a W register during the last cycle. The GOTO and branch instructions can never force an instruction stall because they do not perform write operations.
2.10.2.3
Instruction Stalls and DO and REPEAT Loops Other than the addition of instruction stall cycles, RAW data dependencies will not affect the operation of either DO or REPEAT loops. The pre-fetched instruction within a REPEAT loop does not change until the loop is complete or an exception occurs. Although register dependency checks occur across instruction boundaries, the dsPIC30F effectively compares the source and destination of the same instruction during a REPEAT loop. The last instruction of a DO loop either pre-fetches the instruction at the loop start address or the next instruction (outside the loop). The instruction stall decision will be based on the last instruction in the loop and the contents of the pre-fetched instruction.
2.10.2.4
Instruction Stalls and Program Space Visibility (PSV) When program space (PS) is mapped to data space by enabling the PSV (CORCON<2>) bit, and the X space EA falls within the visible program space window, the read or write cycle is redirected to the address in program space. Accessing data from program space takes up to 3 instruction cycles. Instructions operating in PSV address space are subject to RAW data dependencies and consequent instruction stalls, just like any other instruction. Consider the following code segment: ADD MOV W0,[W1],[W2++] [W2],[W3] ; PSV = 1, W1=0x8000, PSVPAG=0xAA
This sequence of instructions would take 5 instruction cycles to execute. 2 instruction cycles are added to perform the PSV access via W1. Furthermore, an instruction stall cycle is inserted to resolve the RAW data dependency caused by W2.
DS70049C-page 2-37
2.11
A summary of the registers associated with the dsPIC30F CPU core is provided in Table 2-8.
Bit 13
W0 (WREG) W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 W13 W14 W15 SPLIM ACCAL ACCAH Sign-extension of ACCA<39> ACCBL ACCBH Sign-extension of ACCB<39> PCL RCOUNT DCOUNT DOSTARTL SA SB OAB SAB DA DOENDL DC IPL2 IPL1 IPL0 RA DOENDH N OV Z C DOSTARTH 0 0 PCH TBLPAG PSVPAG 0 ACCBU ACCAU
Register Maps
Table 2-8:
Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
dsPIC30F Core Register Map

Reset State
0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxx0 0000 0000 00xx xxxx xxxx xxxx xxxx xxx0 0000 0000 00xx xxxx 0000 0000 0000 0000
Name
Addr
Bit 15
Bit 14
DS70049C-page 2-38
W0
0000
W1
0002
W2
0004
W3
0006
W4
0008
W5
000A
W6
000C
W7
000E
W8
0010
W9
0012
W10
0014
W11
0016
W12
0018
W13
001A
W14
001C
W15
001E
SPLIM
0020
ACCAL
0022
ACCAH
0024
ACCAU
0026
ACCBL
0028
ACCBH
002A
ACCBU
002C
PCL
002E
PCH
0030
TBLPAG
0032
PSVPAG
0034
RCOUNT
0036
DCOUNT
0038
DOSTARTL
003A
DOSTARTH
003C
DOENDL
003E
DOENDH
0040
SR
0042
OA
OB
Table 2-8:
Bit 13
XMODSRT<15:0> XMODEND<15:0> YMODSRT<15:0> YMODEND<15:0> XBREV<14:0> DISICNT<13:0> 1 0 1 0 BWM<3:0> YWM<3:0> XWM<3:0> US EDT DL2 DL<1:0> SATA SATB SATDW ACCSAT IPL3 PSV RND IF
dsPIC30F Core Register Map (Continued)

Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Reset State
0000 0000 0010 0000 0000 0000 0000 0000 xxxx xxxx xxxx xxx0 xxxx xxxx xxxx xxx1 xxxx xxxx xxxx xxx0 xxxx xxxx xxxx xxx1 xxxx xxxx xxxx xxxx 0000 0000 0000 0000 0000 0000 0000 0000
Name
Addr
Bit 15
Bit 14
CORCON
0044
MODCON
0046
XMODEN YMODEN
XMODSRT
0048
XMODEND
004A
YMODSRT
004C
YMODEND
004E
XBREV
0050
BREN
CPU
DISICNT
0052
Reserved
0054 007E
Legend: x = uninitiated Note: Refer to the device data sheet for specific Core Register Map details.
Section 2. CPU
DS70049C-page 2-39

2.12 Related Application Notes
This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the dsPIC30F CPU module are: Title No related application notes at this time. Application Note #
Note:
Please visit the Microchip web site (www.microchip.com) for additional Application Notes and code examples for the dsPIC30F Family of devices.
DS70049C-page 2-40
Section 2. CPU
2
2.13 Revision History Revision A
CPU
Revision B
This revision incorporates additional technical content for the dsPIC30F CPU module.
Revision C
This revision incorporates all known errata at the time of this document update.
DS70049C-page 2-41

NOTES:
DS70049C-page 2-42
Section 3. Data Memory

HIGHLIGHTS
This section of the manual contains the following topics: 3.1 3.2 3.3 3.4 3.5 3.6 3.7 Introduction .................................................................................................................... 3-2 Data Space Address Generator Units (AGUs)............................................................... 3-5 Modulo Addressing ........................................................................................................ 3-7 Bit-Reversed Addressing ............................................................................................. 3-14 Control Register Descriptions ...................................................................................... 3-18 Related Application Notes............................................................................................ 3-23 Revision History ........................................................................................................... 3-24
3
Data Memory
DS70050C-page 3-1

3.1 Introduction
The dsPIC30F data width is 16-bits. All internal registers and data space memory are organized as 16-bits wide. The dsPIC30F features two data spaces. The data spaces can be accessed separately (for some DSP instructions) or together as one 64-Kbyte linear address range (for MCU instructions). The data spaces are accessed using two Address Generation Units (AGUs) and separate data paths. An example data space memory map is shown in Figure 3-1. Data memory addresses between 0x0000 and 0x07FF are reserved for the device special function registers (SFRs). The SFRs include control and status bits for the CPU and peripherals on the device. The RAM begins at address 0x0800 and is split into two blocks, X and Y data space. For data writes, the X and Y data spaces are always accessed as a single, linear data space. For data reads, the X and Y memory spaces can be accessed independently or as a single, linear space. Data reads for MCU class instructions always access the the X and Y data spaces as a single combined data space. Dual source operand DSP instructions, such as the MAC instruction, access the X and Y data spaces separately to support simultaneous reads for the two source operands. MCU instructions can use any W register as an address pointer for a data read or write operation. During data reads, the DSP class of instructions isolates the Y address space from the total data space. W10 and W11 are used as address pointers for reads from the Y data space. The remaining data space is referred to as X space, but could more accurately be described as X minus Y space. W8 and W9 are used as address pointers for data reads from the X data space in DSP class instructions. Figure 3-2 shows how the data memory map functions for both MCU class and DSP class instructions. Note that it is the W register number and type of instruction that determines how address space is accessed for data reads. In particular, MCU instructions treat the X and Y memory as a single combined data space. The MCU instructions can use any W register as an address pointer for reads and writes. The DSP instructions that can simultaneously pre-fetch two data operands, split the data memory into two spaces. Specific W registers must be used for read address pointers in this case. Some DSP instructions have the ability to store the accumulator that is not targeted by the instruction to data memory. This function is called accumulator write back. W13 must be used as an address pointer to the combined data memory space for accumulator write back operations. For DSP class instructions, W8 and W9 should point to implemented X memory space for all memory reads. If W8 or W9 points to Y memory space, zeros will be returned. If W8 or W9 points to an unimplemented memory address, an address error trap will be generated. For DSP class instructions, W10 and W11 should point to implemented Y memory space for all memory reads. If W10 or W11 points to implemented X memory space, all zeros will be returned. If W10 or W11 points to an unimplemented memory address, an address error trap will be generated. For additional information on address error traps, refer to Section 6. Reset Interrupts. Note: The data memory map and the partition between the X and Y data spaces is device specific. Refer to the specific dsPIC30F device data sheet for further details.
DS70050C-page 3-2

Figure 3-1: Example Data Memory Map LSByte Address LSByte 0x0000 SFR Space 0x07FF 0x0801 Near Data Memory 0x17FF 0x1801 X Data RAM 0x17FE 0x1800 Y Data RAM 0x07FE 0x0800
MSByte Address MSByte 0x0001
16-bits
0x1FFF
0x27FF 0x2801
0x27FE 0x2800
3
0x8001 0x8000
Data Memory
X Data RAM Unimplemented Provides Program Space Visibility
0xFFFF
Note 1: The partition between the X and Y data spaces is device specific. Refer to the appropriate device data sheet for further details. The data space boundaries indicated here are used for example purposes only. 2: Near data memory can be accessed directly via file register instructions that encode a 13-bit address into the opcode. At a minimum, the near data memory region overlaps all of the SFR space and a portion of X memory space. All of X memory space and some or all of Y memory space may be included in the near data memory region, depending on the device variant. 3: All data memory can be accessed indirectly via W registers or directly using the MOV instruction. 4: Upper half of data memory map can be mapped into a segment of program memory space for program space visibility.
DS70050C-page 3-3

Figure 3-2: Data Spaces for MCU and DSP Instructions
Indirect EA from W10, W11 Indirect EA from W8, W9
UNUSED
X SPACE
(Y SPACE)
Y SPACE
UNUSED
UNUSED
MCU Class Instructions (Read/Write) DSP Instructions (Write)
Dual Source Operand DSP Instructions (Read)
Note:
Data writes for DSP instructions consider the entire data memory as one combined space. DSP instructions that perform an accumulator write back use W13 as an address pointer for writes to the combined data spaces.
3.1.1
Near Data Memory

An 8-Kbyte address space, referred to as near data memory, is reserved in the data memory space between 0x0000 and 0x1FFF. Near data memory is directly addressable via a 13-bit absolute address field within all file register instructions. The memory regions included in the near data region will depend on the amount of data memory implemented for each dsPIC30F device variant. At a minimum, the near data region will include all of the SFRs and some of the X data memory. For devices that have smaller amounts of data memory, the near data region may include all of X memory space and possibly some or all of Y memory space. Refer to Figure 3-1 for more details. Note: The entire 64K data space can be addressed directly using the MOV instruction. Refer to the dsPIC30F Programmers Reference Manual (DS70030) for further details.
DS70050C-page 3-4
X SPACE
X SPACE

3.2 Data Space Address Generator Units (AGUs)
The dsPIC30F contains an X AGU and a Y AGU for generating data memory addresses. Both X and Y AGUs can generate any effective address (EA) within a 64-Kbyte range. However, EAs that are outside the physical memory provided will return all zeros for data reads and data writes to those locations will have no effect. Furthermore, an address error trap will be generated. For more information on address error traps, refer to Section 6. Reset Interrupts.
3.2.1
X Address Generator Unit

The X AGU is used by all instructions and supports all Addressing modes. The X AGU consists of a read AGU (X RAGU) and a write AGU (X WAGU), which operate independently on separate read and write buses during different phases of the instruction cycle. The X read data bus is the return data path for all instructions that view data space as combined X and Y address space. It is also the X address space data path for the dual operand read instructions (DSP instruction class). The X write data bus is the only write path to the combined X and Y data space for all instructions. The X RAGU starts its effective address calculation during the prior instruction cycle, using information derived from the just pre-fetched instruction. The X RAGU EA is presented to the address bus at the beginning of the instruction cycle. The X WAGU starts its effective address calculation at the beginning of the instruction cycle. The EA is presented to the address bus during the write phase of the instruction. Both the X RAGU and the X WAGU support modulo addressing. Bit-reversed addressing is supported by the X WAGU only.
3.2.2
Y Address Generator Unit

The Y data memory space has one AGU that supports data reads from the Y data memory space. The Y memory bus is never used for data writes. The function of the Y AGU and Y memory bus is to support concurrent data reads for DSP class instructions. The Y AGU timing is identical to that of the X RAGU, in that its effective address calculation starts prior to the instruction cycle, using information derived from the pre-fetched instruction. The EA is presented to the address bus at the beginning of the instruction cycle. The Y AGU supports Modulo Addressing and Post-modification Addressing modes for the DSP class of instructions that use it. Note: The Y AGU does not support data writes. All data writes occur via the X WAGU to the combined X and Y data spaces. The Y AGU is only used during data reads for dual source operand DSP instructions.
3
Data Memory
DS70050C-page 3-5

Figure 3-3: Data Space Access Timing
ADD.W MOV.W MAC SUB.W
TCY
W0, [W7], [W10] W10, [W9++] W4*W5, A, W4, [W8]+=2, W5, [W10]+=2, [W13]+=2 W4, [--W9], [W6++]
IR X RAGU X WAGU Y AGU X Address X Data Read X Data Write Y Address Y Data (Read) [W7] [W7]
ADD ALU OP
MOV [W8]+=2 Stall Check [W9++]
MAC [--W9]
SUB
[W10]
W10 [W7] [W10]
W9
W8
W13 [W8]
W9-2
[W9-2] [W13]
[W9] W10
[W10]
3.2.3
Address Generator Units and DSP Class Instructions

The Y AGU and Y memory data path are used in concert with the X RAGU by the DSP class of instructions to provide two concurrent data read paths. For example, the MAC instruction can simultaneously pre-fetch two operands to be used in the next multiplication. The DSP class of instructions dedicates two W register pointers, W8 and W9, to always operate through the X RAGU and address X data space independently from Y data space, plus two W register pointers, W10 and W11, to always operate through the Y AGU and address Y data space independently from X data space. Any data write performed by a DSP class instruction will take place in the combined X and Y data space and the write will occur across the X-bus. Consequently, the write can be to any address irrespective of where the EA is directed. The Y AGU only supports Post-modification Addressing modes associated with the DSP class of instructions. For more information on Addressing modes, please refer to the dsPIC30F Programmers Reference Manual. The Y AGU also supports modulo addressing for automated circular buffers. All other (MCU) class instructions can access the Y data address space through the X AGU when it is regarded as part of the composite linear space.
DS70050C-page 3-6
ALU OP During Q3 W6
[W13] Stall Check [W10]+=2
Stall Check [W6++]

3.2.4 Data Alignment
The ISA supports both word and byte operations for all MCU instructions that access data through the X memory AGU. The LSb of a 16-bit data address is ignored for word operations. Word data is aligned in the little-endian format with the LSByte at the even address (LSB = 0) and the MSByte at the odd address (LSB = 1). For byte operations, the LSB of the data address is used to select the byte that is accessed. The addressed byte is placed on the lower 8 bits of the internal data bus. All effective address calculations are automatically adjusted depending on whether a byte or a word access is performed. For example, an address will be incremented by 2 for a word operation that post-increments the address pointer. Note: All word accesses must be aligned to an even address (LSB = 0). Misaligned word data fetches are not supported, so care must be taken when mixing byte and word operations or translating from existing PICmicro code. Should a misaligned word read or write be attempted, an address error trap will occur. A misaligned read operation will complete, but a misaligned write will not take place. The trap will then be taken, allowing the system to examine the machine state prior to execution of the address Fault. Data Alignment MSByte Byte 1 Byte 3 Byte 5 Word 0 Word 1 Long Word<15:0> Long Word<31:16> LSByte Byte 0 Byte 2 Byte 4
Figure 3-4:
15 0001 0003 0005
8 7
0 0000 0002
3
Data Memory
0004 0006 0008 000A 000C
3.3
Modulo Addressing
Modulo, or circular addressing provides an automated means to support circular data buffers using hardware. The objective is to remove the need for software to perform data address boundary checks when executing tightly looped code as is typical in many DSP algorithms. Any W register, except W15, can be selected as the pointer to the modulo buffer. The modulo hardware performs boundary checks on the address held in the selected W register and automatically adjusts the pointer value at the buffer boundaries, when required. dsPIC30F modulo addressing can operate in either data or program space (since the data pointer mechanism is essentially the same for both). One circular buffer can be supported in each of the X (which also provides the pointers into Program space) and Y data spaces. The modulo data buffer length can be any size up to 32K words. The modulo buffer logic supports buffers using word or byte sized data. However, the modulo logic only performs address boundary checks at word address boundaries, so the length of a byte modulo buffer must be even. In addition, byte-sized modulo buffers cannot be implemented using the Y AGU because byte access is not supported via the Y memory data bus.
DS70050C-page 3-7

3.3.1 Modulo Start and End Address Selection
Four address registers are available for specifying the modulo buffer start and end addresses: XMODSRT: X AGU Modulo Start Address Register XMODEND: X AGU Modulo End Address Register YMODSRT: Y AGU Modulo Start Address Register YMODEND: Y AGU Modulo End Address Register
The start address for a modulo buffer must be located at an even byte address boundary. The LSB of the XMODSRT and YMODSRT registers is fixed at 0 to ensure the correct modulo start address. The end address for a modulo buffer must be located at an odd byte address boundary. The LSB of the XMODEND and YMODEND registers is fixed to 1 to ensure the correct modulo end address. The start and end address selected for each modulo buffer have certain restrictions, depending on whether an incrementing or decrementing buffer is to be implemented. For an incrementing buffer, a W register pointer is incremented through the buffer address range. When the end address of the incrementing buffer is reached, the W register pointer is reset to point to the start of the buffer. For a decrementing buffer, a W register pointer is decremented through the buffer address range. When the start address of a decrementing buffer is reached, the W register pointer is reset to point to the end of the buffer. Note: The user must decide whether an incrementing or decrementing modulo buffer is required for the application. There are certain address restrictions that depend on whether an incrementing or decrementing modulo buffer is to be implemented.
3.3.1.1
Modulo Start Address The data buffer start address is arbitrary, but must be at a zero power of two boundary for incrementing modulo buffers. The modulo start address can be any value for decrementing modulo buffers and is calculated using the chosen buffer end address and buffer length. For example, if the buffer length for an incrementing buffer is chosen to be 50 words (100 bytes), then the buffer start byte address must contain 7 Least Significant zeros. Valid start addresses may, therefore, be 0xNN00 and 0xNN80, where N is any hexadecimal value.
3.3.1.2
Modulo End Address The data buffer end address is arbitrary but must be at a ones boundary for decrementing buffers. The modulo end address can be any value for an incrementing buffer and is calculated using the chosen buffer start address and buffer length. For example, if the buffer size (modulus value) is chosen to be 50 words (100 bytes), then the buffer end byte address for decrementing modulo buffer must contain 7 Least Significant ones. Valid end addresses may, therefore, be 0xNNFF and 0xNN7F, where x is any hexadecimal value. Note: If the required modulo buffer length is an even power of 2, modulo start and end addresses can be chosen that satisfy the requirements for incrementing and decrementing buffers.
DS70050C-page 3-8

3.3.1.3 Modulo Address Calculation The end address for an incrementing modulo buffer must be calculated from the chosen start address and the chosen buffer length in bytes. Equation 3-1 may be used to calculate the end address. Equation 3-1: Modulo End Address for Incrementing Buffer End Address = Start Address + Buffer Length 1 The start address for a decrementing modulo buffer is calculated from the chosen end address and the buffer length, as shown in Equation 3-2. Equation 3-2: Modulo Start Address for Decrementing Buffer Start Address = End Address Buffer Length + 1 3.3.1.4 Data Dependencies Associated with Modulo Addressing SFRs A write operation to the Modulo Addressing Control register, MODCON, should not be immediately followed by an indirect read operation using any W register. The code segment shown in Example 3-1 will thus lead to unexpected results. Note 1: Using a POP instruction to pop the contents of the top-of-stack (TOS) location into MODCON, also constitutes a write to MODCON. The instruction immediately following a write to MODCON cannot be any instruction performing an indirect read operation. 2: The user should note that some instructions perform an indirect read operation, implicitly. These are: POP, RETURN, RETFIE, RETLW and ULNK. Example 3-1:
MOV MOV MOV
3
Data Memory
Incorrect MODCON Initialization

;Initialize MODCON ;Incorrect EA generated here
#0x8FF4, w0 w0, MODCON [w1], w2
To work around this problem of initialization, use any Addressing mode other than indirect reads in the instruction that immediately follows the initialization of MODCON. A simple work around to the problem is achieved by adding a NOP after initializing MODCON, as shown in Example 3-2. Example 3-2:
MOV MOV NOP MOV
Correct MODCON Initialization

;Initialize MODCON ;See Note below ;Correct EA generated here
#0x8FF4, w0 w0, MODCON [w1], w2
DS70050C-page 3-9

An additional condition exists for indirect read operations performed immediately after writing to the modulo address SFRs: XMODSRT XMODEND YMODSRT YMODEND
If modulo addressing has already been enabled in MODCON, then a write to the X (or Y) modulo address SFRs should not be immediately followed by an indirect read, using the W register designated for modulo buffer access from X-data space (or Y-data space). The code segment in Example 3-3 shows how initializing the modulo SFRs associated with the X-data space, could lead to unexpected results. A similar example can be made for initialization in Y-data space. Example 3-3:
MOV MOV MOV MOV MOV MOV MOV
Incorrect Modulo Addressing Setup

;Modulo addressing enabled ;in X-data space using w4 ;for buffer access ;XMODSRT is initialized ;XMODEND is initialized ;Incorrect EA generated
#0x8FF4, w0 w0, MODCON #0x1200, w4 w4, XMODSRT #0x12FF, w0 w0, XMODEND [w4++], w5
To work around this issue, insert a NOP, or perform any operation other than an indirect read that uses the W register designated for modulo buffer access, after initializing the modulo address SFRs. This is demonstrated in Example 3-4. Another alternative would be to enable modulo addressing in MODCON after initializing the modulo start and end address SFRs. Example 3-4:
MOV MOV MOV MOV MOV MOV NOP MOV
Correct Modulo Addressing Setup

;Modulo addressing enabled ;in X-data space using w4 ;for buffer access ;XMODSRT is initialized ;XMODEND is initialized ;See Note below ;Correct EA generated here
#0x8FF4, w0 w0, MODCON #0x1200, w4 w4, XMODSRT #0x12FF, w0 w0, XMODEND [w4++], w5
Note: Alternatively, execute other instructions that do not perform indirect read operations, using the W register designated for modulo buffer access.
DS70050C-page 3-10

3.3.2 W Address Register Selection
The X address space pointer W register (XWM) to which modulo addressing is to be applied, is stored in MODCON<3:0> (see Register 3-1). The XMODSRT, XMODEND, and the XWM register selection are shared between the X RAGU and X WAGU. Modulo addressing is enabled for X data space when XWM is set to any value other than 15 and the XMODEN bit is set (MODCON<15>). W15 cannot be used as the pointer for modulo addressing because it is the dedicated software stack pointer. The Y address space pointer W register (YWM) to which modulo addressing is to be applied, is stored in MODCON<7:4> (see Register 3-2). Modulo addressing is enabled for Y data space when YWM is set to any value other than 15 and the YMODEN bit is set (MODCON<14>). Note: A write to the MODCON register should not be followed by an instruction that performs an indirect read operation using a W register. Unexpected results may occur. Some instructions perform an implicit indirect read. These are:POP, RETURN, RETFIE, RETLW and ULNK.
3.3.3
Modulo Addressing Applicability

Modulo addressing can be applied to the effective address (EA) calculation associated with the selected W register. It is important to realize that the address boundary tests look for addresses equal to or greater than the upper address boundary for incrementing buffers and equal to or less than the lower address boundary for decrementing buffers. Address changes may, therefore, jump over boundaries and still be adjusted correctly. Remember that the automatic adjustment of the W register pointer by the modulo hardware is uni-directional. That is, the W register pointer may not be adjusted correctly by the modulo hardware when the W register pointer for an incrementing buffer is decremented and vice versa. The exception to this rule is when the buffer length is an even power of 2 and the start and end addresses can be chosen to meet the -boundary requirements for both incrementing and decrementing modulo buffers. A new EA can exceed the modulo buffer boundary by up to the length of the buffer and still be successfully corrected. This is important to remember when the Register Indexed ([Wb + Wn]) and Literal Offset ([Wn + lit10]) Addressing modes are used. The user should remember that the Register Indexed and Literal Offset Addressing modes do not change the value held in the W register. Only the indirect with Pre- and Post-modification Addressing modes ([Wn++], [Wn--], [++Wn], [--Wn]) will modify the W register address value.
3
Data Memory
DS70050C-page 3-11

3.3.4 Modulo Addressing Initialization for Incrementing Modulo Buffer
The following steps describe the setup procedure for an incrementing circular buffer. The steps are similar whether the X AGU or Y AGU is used. 1. 2. Determine the buffer length in 16-bit data words. Multiply this value by 2 to get the length of the buffer in bytes. Select a buffer starting address that is located at a binary zeros boundary based on the desired length of the buffer. Remember that the buffer length in words must be multiplied by 2 to obtain the byte address range. For example, a buffer with a length of 100 words (200 bytes) could use 0xXX00 as the starting address. Calculate the buffer end address using the buffer length chosen in Step 1 and the buffer start address chosen in Step 2. The buffer end address is calculated using Equation 3-1. Load the XMODSRT (YMODSRT) register with the buffer start address chosen in Step 2. Load the XMODEND (YMODEND) register with the buffer end address calculated in Step 3. Write to the XWM<3:0> (YWM<3:0>) bits in the MODCON register to select the W register that will be used to access the circular buffer. Set the XMODEN (YMODEN) bit in the MODCON register to enable the circular buffer. Load the selected W register with address that points to the buffer. The W register address will be adjusted automatically at the end of the buffer when an indirect access with pre/post increment is performed (see Figure 3-5).
3. 4. 5. 6. 7. 8. 9.
Figure 3-5: Byte Address
Incrementing Buffer Modulo Addressing Operation Example

MOV MOV MOV MOV MOV MOV MOV MOV DO FILL: MOV #0x1100,W0 W0,XMODSRT #0x1163,W0 W0,XMODEND #0x8001,W0 W0,MODCON #0x0000,W0 #0x1100,W1 #49,FILL W0,[W1++]
;set modulo start address ;set modulo end address ;enable W1, X AGU for modulo ;W0 holds buffer fill value ;point W1 to buffer ;fill the 50 buffer locations ;fill the next location
0x1100
;W1 = 0x1100 when DO loop completes
0x1163
Start Addr = 0x1100 End Addr = 0x1163 Length = 50 Words
DS70050C-page 3-12

3.3.5 Modulo Addressing Initialization for Decrementing Modulo Buffer
The following steps describe the setup procedure for a decrementing circular buffer. The steps are similar whether the X AGU or Y AGU is used. 1. 2. Determine the buffer length in 16-bit data words. Multiply this value by 2 to get the length of the buffer in bytes. Select a buffer end address that is located at a binary ones boundary, based on the desired length of the buffer. Remember that the buffer length in words must be multiplied by 2 to obtain the byte address range. For example, a buffer with a length of 128 words (256 bytes) could use 0xXXFF as the end address. Calculate the buffer start address using the buffer length chosen in Step 1 and the end address chosen in Step 2. The buffer start address is calculated using Equation 3-2. Load the XMODSRT (YMODSRT) register with the buffer start address chosen in Step 3. Load the XMODEND (YMODEND) register with the buffer end address chosen in Step 2. Write to the XWM<3:0> (YWM<3:0>) bits in the MODCON register to select the W register that will be used to access the circular buffer. Set the XMODEN (YMODEN) bit in the MODCON register to enable the circular buffer. Load the selected W register with address that points to the buffer. The W register address will be adjusted automatically at the end of the buffer when an indirect access with pre/post-decrement is performed (see Figure 3-6).
3. 4. 5. 6. 7. 8. 9.
Figure 3-6: Byte Address
Decrementing Buffer Modulo Addressing Operation Example
3
Data Memory
0x11E0
MOV MOV MOV MOV MOV MOV MOV MOV DO MOV FILL: DEC
#0x11E0,W0 W0,XMODSRT #0x11FF,W0 W0,XMODEND #0x8001,W0 W0,MODCON #0x000F,W0 #0x11FE,W1 #15,FILL W0,[W1--]
;set modulo start address ;set modulo end address ;enable W1, X AGU for modulo ;W0 holds buffer fill value ;point W1 to buffer ;fill the 16 buffer locations ;fill the next location
W0,W0
;decrement the fill value
0x11FF
; W1 = 0x11FE when DO loop completes
Start Addr = 0x11E0 End Addr = 0x11FF Length = 16 Words
DS70050C-page 3-13

3.4 3.4.1 Bit-Reversed Addressing Introduction to Bit-Reversed Addressing
Bit-reversed addressing simplifies data re-ordering for radix-2 FFT algorithms. It is supported through the X WAGU only. Bit-reversed addressing is accomplished by effectively creating a mirror image of an address pointer by swapping the bit locations around the center point of the binary value, as shown in Figure 3-7. An example bit-reversed sequence for a 4-bit address field is shown in Table 3-1. Figure 3-7: Bit-Reversed Address Example b3 b2 b1 b0 Bit locations swapped left-to-right around center of binary value. b0 b1 b2 b3 Bit-Reversed Result Table 3-1: Bit-Reversed Address Sequence (16-Entry) Normal Address A3 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 A2 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1 A1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 A0 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 decimal 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 A3 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 A2 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 Bit-Reversed Address A1 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1 A0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 decimal 0 8 4 12 2 10 6 14 1 9 5 13 3 11 7 15
DS70050C-page 3-14

3.4.2 Bit-Reversed Addressing Operation
Bit-reversed addressing is only supported by the X WAGU and is controlled by the MODCON and XBREV special function registers. Bit-reversed addressing is invoked as follows: 1. 2. 3. Bit-reversed addressing is assigned to one of the W registers using the BWM control bits (MODCON<11:8>). Bit-reversed addressing is enabled by setting the BREN control bit (XBREV<15>). The X AGU bit-reverse modifier is set via the XB control bits (XBREV<14:0>).
When enabled, the bit-reversed addressing hardware will generate bit-reversed addresses, only when the register indirect with Pre- or Post-increment Addressing modes are used ([Wn++], [++Wn]). Furthermore, bit-reverse addresses are only generated for Word mode instructions. It will not function for all other Addressing modes or Byte mode instructions (normal addresses will be generated). Note: A write to the MODCON register should not be followed by an instruction that performs an indirect read operation using a W register. Unexpected results may occur. Some instructions perform an implicit indirect read. These are: POP, RETURN, RETFIE, RETLW and ULNK.
3.4.2.1
Modulo Addressing and Bit-Reversed Addressing Modulo addressing and bit-reversed addressing can be enabled simultaneously using the same W register, but bit-reversed addressing operation will always take precedence for data writes when enabled. As an example, the following setup conditions would assign the same W register to modulo and bit-reversed addressing: X modulo addressing is enabled (XMODEN = 1) Bit-reverse addressing is enabled (BREN = 1) W1 assigned to modulo addressing (XWM<3:0> = 0001) W1 assigned to bit-reversed addressing (BWM<3:0> = 0001)
3
Data Memory
For data reads that use W1 as the pointer, modulo address boundary checking will occur. For data writes using W1 as the destination pointer, the bit-reverse hardware will correct W1 for data re-ordering. 3.4.2.2 Data Dependencies Associated with XBREV If bit-reversed addressing has already been enabled by setting the BREN (XBREV<15>) bit, then a write to the XBREV register should not be followed by an indirect read operation using the W register, designated as the bit reversed address pointer.
DS70050C-page 3-15

3.4.3 Bit-Reverse Modifier Value
The value loaded into the XBREV register is a constant that indirectly defines the size of the bit-reversed data buffer. The XB modifier values used with common bit-reversed buffers are summarized in Table 3-2. Table 3-2: Bit-Reversed Address Modifier Values Buffer Size (Words) 32768 16384 8192 4096 2048 1024 512 256 128 64 32 16 8 4 2 Note: XB Bit-Reversed Address Modifier Value 0x4000 0x2000 0x1000 0x0800 0x0400 0x0200 0x0100 0x0080 0x0040 0x0020 0x0010 0x0008 0x0004 0x0002 0x0001
Only the the bit-reversed modifier values shown will produce valid bit-reversed address sequences.
The bit-reverse hardware modifies the W register address by performing a reverse-carry addition of the W contents and the XB modifier constant. A reverse-carry addition is performed by adding the bits from left-to-right instead of right-to-left. If a carry-out occurs in a bit location, the carry out bit is added to the next bit location to the right. Example 3-5 demonstrates the reverse-carry addition and subsequent W register values using 0x0008 as the XB modifier value. Note that the XB modifier is shifted one bit location to the left to generate word address values.
DS70050C-page 3-16

Example 3-5: XB Address Calculation 0000 0000 0000 0000 +1 0000 Wn points to word 0 Wn = Wn + XB
0000 0000 0001 0000 +1 0000
Wn points to word 8 Wn = Wn + XB
0000 0000 0000 1000 +1 0000
0000 0000 0001 1000 +1 0000
0000 0000 0000 0100 +1 0000
0000 0000 0001 0100
Wn points to word 10
3
When XB<14:0> = 0x0008, the bit-reversed buffer size will be 16 words. Bits 1-4 of the W register will be subject to bit-reversed address correction, but bits 5-15 (outside the pivot point) will not be modified by the bit-reverse hardware. Bit 0 is not modified because the bit-reverse hardware only operates on word addresses. The XB modifier controls the pivot point for the bit-reverse address modification. Bits outside of the pivot point will not be subject to bit-reversed address corrections. Figure 3-8:
15
Data Memory
Bit-Reversed Address Modification for 16-Word Buffer

10 9 8 7 6 5 4 3 2 1 0
XB<14:0> = 0x0008 Bits 1-4 of address are modified.
14 13 12 11
15
14 13 12 11
10
Bit-Reversed Result
Pivot Point
DS70050C-page 3-17

3.4.4 Bit-Reversed Addressing Code Example
The following code example reads a series of 16 data words and writes the data to a new location in bit-reversed order. W0 is the read address pointer and W1 is the write address pointer subject to bit-reverse modification.
; Set XB for 16-word buffer, enable bit reverse addressing MOV #0x8008,W0 MOV W0,XBREV ; Setup MODCON to use W1 for bit reverse addressing MOV #0x01FF,W0 MOV W0,MODCON ; W0 points to input data buffer MOV #Input_Buf,W0 ; W1 points to bit reversed data MOV #Bit_Rev_Buf,W1 ; Re-order the data from Input_Buf into Bit_Rev_Buf REPEAT #15 MOV [W0++],[W1++]
3.5
Control Register Descriptions

The following registers are used to control modulo and bit-reversed addressing: MODCON: Modulo Addressing Control Register XMODSRT: X AGU Modulo Start Address Register XMODEND: X AGU Modulo End Address Register YMODSRT: Y AGU Modulo Start Address Register YMODEND: Y AGU Modulo End Address Register XBREV: X AGU Bit-Reverse Addressing Control Register
A detailed description of each register is provided on subsequent pages.
DS70050C-page 3-18

Register 3-1: MODCON: Modulo and Bit-Reversed Addressing Control Register U-0 U-0 R/W-0 R/W-0 R/W-0 BWM<3:0> R/W-0 bit 8
Upper Byte: R/W-0 R/W-0 XMODEN YMODEN bit 15
Lower Byte: R/W-0 bit 7 bit 15
R/W-0 R/W-0 YWM<3:0>
R/W-0
R/W-0
R/W-0 R/W-0 XWM<3:0>
R/W-0 bit 0
XMODEN: X RAGU and X WAGU Modulus Addressing Enable bit 1 = X AGU modulus addressing enabled 0 = X AGU modulus addressing disabled YMODEN: Y AGU Modulus Addressing Enable bit 1 = Y AGU modulus addressing enabled 0 = Y AGU modulus addressing disabled BWM<3:0>: X WAGU Register Select for Bit-Reversed Addressing bits 1111 = Bit-reversed addressing disabled 1110 = W14 selected for bit-reversed addressing 1101 = W13 selected for bit-reversed addressing 0000 = W0 selected for bit-reversed addressing YWM<3:0>: Y AGU W Register Select for Modulo Addressing bits 1111 = Modulo addressing disabled 1010 = W10 selected for modulo addressing 1011 = W11 selected for modulo addressing Note: All other settings of the YWM<3:0> control bits are reserved and should not be used.
bit 14
bit 13-12 Unimplemented: Read as 0 bit 11-8
3
Data Memory
bit 7-4
bit 3-0
XWM<3:0>: X RAGU and X WAGU W Register Select for Modulo Addressing bits 1111 = Modulo addressing disabled 1110 = W14 selected for modulo addressing 0000 = W0 selected for modulo addressing Note: A write to the MODCON register should not be followed by an instruction that performs an indirect read operation using a W register. Unexpected results may occur. Some instructions perform an implicit indirect read. These are: POP, RETURN, RETFIE, RETLW and ULNK.
Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70050C-page 3-19

Register 3-2: Upper Byte: R/W-0 bit 15 Lower Byte: R/W-0 bit 7 bit 15-1 bit 0 XS<15:1>: X RAGU and X WAGU Modulo Addressing Start Address bits Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown XMODSRT: X AGU Modulo Addressing Start Register R/W-0 R/W-0 R/W-0 R/W-0 XS<15:8> R/W-0 R/W-0 R/W-0 bit 8
R/W-0
R/W-0
R/W-0 XS<7:1>
R/W-0
R/W-0
R/W-0
R-0 0 bit 0
Register 3-3: Upper Byte: R/W-0 bit 15
XMODEND: X AGU Modulo Addressing End Register R/W-0 R/W-0 R/W-0 R/W-0 XE<15:8> R/W-0 R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 bit 7
R/W-0
R/W-0
R/W-0 XE<7:1>
R/W-0
R/W-0
R/W-0
R-1 1 bit 0
bit 15-1 bit 0
XE<15:1>: X RAGU and X WAGU Modulo Addressing End Address bits Unimplemented: Read as 1 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70050C-page 3-20

Register 3-4: Upper Byte: R/W-0 bit 15 Lower Byte: R/W-0 bit 7 bit 15-1 bit 0 YS<15:1>: Y AGU Modulo Addressing Start Address bits Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown YMODSRT: Y AGU Modulo Addressing Start Register R/W-0 R/W-0 R/W-0 R/W-0 YS<15:8> R/W-0 R/W-0 R/W-0 bit 8
R/W-0
R/W-0
R/W-0 YS<7:1>
R/W-0
R/W-0
R/W-0
R-0 0 bit 0
YMODEND: Y AGU Modulo Addressing End Register R/W-0 R/W-0 R/W-0 R/W-0 YE<15:8> R/W-0 R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 bit 7
3
Data Memory
R/W-0
R/W-0
R/W-0 YE<7:1>
R/W-0
R/W-0
R/W-0
R-1 1 bit 0
bit 15-1 bit 0
YE<15:1>: Y AGU Modulo Addressing End Address bits Unimplemented: Read as 1 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70050C-page 3-21

Register 3-6: Upper Byte: R/W-0 BREN bit 15 XBREV: X Write AGU Bit-Reversal Addressing Control Register R/W-0 R/W-0 R/W-0 R/W-0 XB<14:8> R/W-0 R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 bit 7 bit 15 BREN: Bit-Reversed Addressing (X AGU) Enable bit 1 = Bit-reversed addressing enabled 0 = Bit-reversed addressing disabled XB<14:0>: X AGU Bit-Reversed Modifier bits 0x4000 = 32768 word buffer 0x2000 = 16384 word buffer 0x1000 = 8192 word buffer 0x0800 = 4096 word buffer 0x0400 = 2048 word buffer 0x0200 = 1024 word buffer 0x0100 = 512 word buffer 0x0080 = 256 word buffer 0x0040 = 128 word buffer 0x0020 = 64 word buffer 0x0010 = 32 word buffer 0x0008 = 16 word buffer 0x0004 = 8 word buffer 0x0002 = 4 word buffer 0x0001 = 2 word buffer Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
R/W-0
R/W-0
R/W-0 R/W-0 XB<7:0>
R/W-0
R/W-0
R/W-0 bit 0
bit 14-0
DS70050C-page 3-22

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Data Memory module are: Title No related application notes at this time. Application Note #
Note:
3
Data Memory
DS70050C-page 3-23

Revision B
This revision incorporates additional technical content for the dsPIC30F Data Memory module.
Revision C
DS70050C-page 3-24
Section 4. Program Memory

HIGHLIGHTS
This section of the manual contains the following topics: 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 Program Memory Address Map ..................................................................................... 4-2 Program Counter ........................................................................................................... 4-4 Data Access from Program Memory.............................................................................. 4-4 Program Space Visibility from Data Space .................................................................... 4-8 Program Memory Writes .............................................................................................. 4-10 PSV Code Examples ................................................................................................... 4-11 Related Application Notes............................................................................................ 4-12 Revision History ........................................................................................................... 4-13
4
Program Memory
DS70051D-page 4-1

4.1 Program Memory Address Map
The dsPIC30F devices have a 4M x 24-bit program memory address space, shown in Figure 4-1. There are three available methods for accessing program space. 1. 2. 3. Via the 23-bit PC. Via table read (TBLRD) and table write (TBLWT) instructions. By mapping a 32-Kbyte segment of program memory into the data memory address space.
The program memory map is divided into the user program space and the user configuration space. The user program space contains the Reset vector, interrupt vector tables, program memory and data EEPROM memory. The user configuration space contains non-volatile configuration bits for setting device options and the device ID locations.
DS70051D-page 4-2

Figure 4-1: Example Program Space Memory Map
Reset - GOTO Instruction Reset - Target Address Level 15 Trap Vector Level 14 Trap Vector Level 13 Trap Vector Level 12 Trap Vector Level 11 Trap Vector Level 10 Trap Vector Level 9 Trap Vector Level 8 Trap Vector Interrupt 0 Vector Interrupt 1 Vector 000000 000002 000004
Interrupt Vector Table 000014
Interrupt 52 Vector Interrupt 53 Vector User Memory Space Reserved Level 15 Trap Vector Level 14 Trap Vector Level 13 Trap Vector Level 12 Trap Vector Level 11 Trap Vector Level 10 Trap Vector Level 9 Trap Vector Level 8 Trap Vector Interrupt 0 Vector Interrupt 1 Vector
00007E 000080 000082 000084
Alternate Interrupt Vector Table
Interrupt 52 Vector Interrupt 53 Vector
0000FE 000100
User Flash Program Memory (48K Instructions) 017FFE 018000 7FEFFE Data EEPROM (4 Kbytes) Configuration Memory Space 7FF000
Reserved (Read 0s)
7FFFFE 800000
4
Reserved
Program Memory
8005BE UNITID 8005C0 8005FE 800600 Reserved Device Configuration Registers F7FFFE F80000 F8000E F80010
Reserved
DEVID (2)
FEFFFE FF0000 FFFFFE
Note:
The address boundaries for user Flash program memory and data EEPROM memory will depend on the dsPIC30F device variant that is selected. Refer to the appropriate device data sheet for further details.
DS70051D-page 4-3

4.2 Program Counter
The PC increments by 2 with the LSb set to 0 to provide compatibility with data space addressing. Sequential instruction words are addressed in the 4M program memory space by PC<22:1>. Each instruction word is 24-bits wide. The LSb of the program memory address (PC<0>) is reserved as a byte select bit for program memory accesses from data space that use Program Space Visibility or table instructions. For instruction fetches via the PC, the byte select bit is not required. Therefore, PC<0> is always set to 0. An instruction fetch example is shown in Figure 4-2. Note that incrementing PC<22:1> by one is equivalent to adding 2 to PC<22:0>. Figure 4-2: Instruction Fetch Example
24-bits 0x000000 24
23 +1(1) 23
Instruction
Program Counter 22
0 0
0x7FFFFE Note 1: Increment of PC<22:1> is equivalent to PC<22:0>+2.
4.3
Data Access from Program Memory

There are two methods by which data can be transferred between the program memory and data memory spaces: via special table instructions, or through the remapping of a 32-Kbyte program space page into the upper half of data space. The TBLRDL and TBLWTL instructions offer a direct method of reading or writing the LSWord of any address within program space without going through data space, which is preferable for some applications. The TBLRDH and TBLWTH instructions are the only method whereby the upper 8-bits of a program word can be accessed as data.
DS70051D-page 4-4
Instruction Latch
User Space

4.3.1 Table Instruction Summary
A set of table instructions is provided to move byte or word-sized data between program space and data space. The table read instructions are used to read from the program memory space into data memory space. The table write instructions allow data memory to be written to the program memory space. Note: Detailed code examples using table instructions can be found in Section 5. Flash and EEPROM Programming.
The four available table instructions are listed below: TBLRDL: Table Read Low TBLWTL: Table Write Low TBLRDH: Table Read High TBLWTH: Table Write High
For table instructions, program memory can be regarded as two 16-bit word wide address spaces residing side by side, each with the same address range as shown in Figure 4-3. This allows program space to be accessed as byte or aligned word addressable, 16-bit wide, 64-Kbyte pages (i.e., same as data space). TBLRDL and TBLWTL access the LS Data Word of the program memory, and TBLRDH and TBLWTH access the upper word. As program memory is only 24-bits wide, the upper byte from this latter space does not exist, though it is addressable. It is, therefore, termed the phantom byte. Figure 4-3: High and Low Address Regions for Table Operations
23 00000000 00000000 00000000 00000000 16 8 0
PC Address 0x000100 0x000102 0x000104 0x000106
HIGH Table Address Range Program Memory Phantom Byte (Read as 0)
LOW Table Address Range
4
Program Memory
DS70051D-page 4-5

4.3.2 Table Address Generation
For all table instructions, a W register address value is concatenated with the 8-bit Data Table Page register, TBLPAG, to form a 23-bit effective program space address plus a byte select bit, as shown in Figure 4-4. As there are 15 bits of program space address provided from the W register, the data table page size in program memory is, therefore, 32K words. Figure 4-4: Address Generation for Table Operations
EA<0> Selects Byte
TBLPAG<7> Selects User/Configuration Space 7 TBLPAG 0 15 EA
8 bits from TBLPAG
16 bits from Wn
24-bit EA
4.3.3
Program Memory Low Word Access

The TBLRDL and TBLWTL instructions are used to access the lower 16 bits of program memory data. The LSb of the W register address is ignored for word-wide table accesses. For byte-wide accesses, the LSb of the W register address determines which byte is read. Figure 4-5 demonstrates the program memory data regions accessed by the TBLRDL and TBLWTL instructions. Figure 4-5:
PC Address 0x000100 0x000102 0x000104 0x000106 00000000 00000000 00000000 00000000
Program Data Table Access (LSWord)

23 16 8 0
Program Memory Phantom Byte (Read as 0)
TBLRDL.W
TBLRDL.B (Wn<0> = 0) TBLRDL.B (Wn<0> = 1)
DS70051D-page 4-6

4.3.4 Program Memory High Word Access
The TBLRDH and TBLWTH instructions are used to access the upper 8 bits of the program memory data. These instructions also support Word or Byte Access modes for orthogonality, but the high byte of the program memory data will always return 0, as shown in Figure 4-6. Figure 4-6: Program Data Table Access (MS Byte)
TBLRDH.W
PC Address 0x000100 0x000102 0x000104 0x000106 00000000 00000000 00000000 00000000
23
16
TBLRDH.B (Wn<0> = 0) Program Memory Phantom Byte (Read as 0)
TBLRDH.B (Wn<0> = 1)
4.3.5
Data Storage in Program Memory

It is assumed that for most applications, the high byte (P<23:16>) will not be used for data, making the program memory appear 16-bits wide for data storage. It is recommended that the upper byte of program data be programmed either as a NOP, or as an illegal opcode value, to protect the device from accidental execution of stored data. The TBLRDH and TBLWTH instructions are primarily provided for array program/verification purposes and for those applications that require compressed data storage.
4
Program Memory
DS70051D-page 4-7

4.4 Program Space Visibility from Data Space
The upper 32 Kbytes of the dsPIC30F data memory address space may optionally be mapped into any 16K word program space page. This mode of operation is called Program Space Visibility (PSV) and provides transparent access of stored constant data from X data space without the need to use special instructions (i.e., TBLRD, TBLWT instructions).
4.4.1
PSV Configuration
Program Space Visibility is enabled by setting the PSV bit (CORCON<2>). A description of the CORCON register can be found in Section 2. CPU. When PSV is enabled, each data space address in the upper half of the data memory map will map directly into a program address (see Figure 4-7). The PSV window allows access to the lower 16 bits of the 24-bit program word. The upper 8 bits of the program memory data should be programmed to force an illegal instruction, or a NOP, to maintain machine robustness. Note that table instructions provide the only method of reading the upper 8 bits of each program memory word. Figure 4-8 shows how the PSV address is generated. The 15 LSbs of the PSV address are provided by the W register that contains the effective address. The MSb of the W register is not used to form the address. Instead, the MSb specifies whether to perform a PSV access from program space or a normal access from data memory space. If a W register effective address of 0x8000 or greater is used, the data access will occur from program memory space when PSV is enabled. All accesses will occur from data memory when the W register effective address is less than 0x8000. The remaining address bits are provided by the PSVPAG register (PSVPAG<7:0>), as shown in Figure 4-8. The PSVPAG bits are concatenated with the 15 LSbs of the W register, holding the effective address to form a 23-bit program memory address. PSV can only be used to access values in program memory space. Table instructions must be used to access values in the user configuration space. The LSb of the W register value is used as a byte select bit, which allows instructions using PSV to operate in Byte or Word mode.
4.4.2
PSV Mapping with X and Y Data Spaces

The Y data space is located outside of the upper half of data space for most dsPIC30F variants, such that the PSV area will map into X data space. The X and Y mapping will have an effect on how PSV is used in algorithms. As an example, the PSV mapping can be used to store coefficient data for Finite Impulse Response (FIR) filter algorithms. The FIR filter multiplies each value of a data buffer containing historical filter input data with elements of a data buffer that contains constant filter coefficients. The FIR algorithm is executed using the MAC instruction within a REPEAT loop. Each iteration of the MAC instruction pre-fetches one historical input value and one coefficient value to be multiplied in the next iteration. One of the pre-fetched values must be located in X data memory space and the other must be located in Y data memory space. To satisfy the PSV mapping requirements for the FIR filter algorithm, the user must locate the historical input data in the Y memory space and the filter coefficients in X memory space.
DS70051D-page 4-8

Figure 4-7: Program Space Visibility Operation
Program Space 0x000100
Data Space 0x0000
0x8000 EA<15> = 1 15
PSVPAG 0x01 8 23
23
15
0 0x008000
0xFFFF
Upper 8 bits of Program Memory Data cannot be read using Program Space Visibility.
0x017FFF
Data Read
Figure 4-8:
Program Space Visibility Address Generation
4
23 bits
Program Memory
Select
Wn
PSVPAG Reg 8 bits 15 bits
Wn<0> is Byte Select 23-bit EA
DS70051D-page 4-9

4.4.3 PSV Timing
Instructions that use PSV will require two extra instruction cycles to complete execution, except the following instructions that require only one extra cycle to complete execution: - The MAC class of instructions with data pre-fetch operands - All MOV instructions including the MOV.D instruction The additional instruction cycles are used to fetch the PSV data on the program memory bus. 4.4.3.1 Using PSV in a Repeat Loop Instructions that use PSV within a REPEAT loop eliminate the extra instruction cycle(s) required for the data access from program memory, hence incurring no overhead in execution time. However, the following iterations of the REPEAT loop will incur an overhead of two instruction cycles to complete execution: 4.4.3.2 The first iteration The last iteration Instruction execution prior to exiting the loop due to an interrupt Instruction execution upon re-entering the loop after an interrupt is serviced
PSV and Instruction Stalls Refer to Section 2. CPU for more information about instruction stalls using PSV.
4.5
Program Memory Writes

The dsPIC30F family of devices contains internal program Flash memory for executing user code. There are two methods by which the user can program this memory: 1. 2. Run-Time Self Programming (RTSP) In-Circuit Serial Programming (ICSP)
RTSP is accomplished using TBLWT instructions. ICSP is accomplished using the SPI interface and integral bootloader software. Refer to Section 5. Flash and EEPROM Programming for further details about RTSP. ICSP specifications can be downloaded from the Microchip Technology web site (www.microchip.com).
DS70051D-page 4-10

4.6 4.6.1 PSV Code Examples PSV Code Example in C:
// PSV code example in C // When defined as below the const string uses the PSV feature of dsPIC const unsigned char hello[] = {"Hello World:\r\n"}; unsigned char *TXPtr; // Transmit pointer int main(void) { // Initialize the UART1 U1MODE = 0x8000; U1STA = 0x0000; U1BRG = ((FCY/16)/BAUD) - 1; TXPtr = &hello[0]; U1STAbits.UTXEN = 1; while (1) { while (*TXPtr) if (!U1STAbits.UTXBF) U1TXREG = *TXPtr++; DelayNmSec(500); TXPtr = &hello[0]; } }
// set baud rate = BAUD // point to first char in string // Initiate transmission
// while valid char in string ... // and buffer not full ... // transmit string via UART // delay for 500 mS // re-initialize pointer to first char
// end main
4.6.2
PSV code Example in Assembly:

.equ CORCONL, CORCON .section .const, "r" hello: .ascii "Hello World:\n\r\0" .global __reset .text __reset: clr mov mov mov mov bset Again: rcall mov mov bset.b mov TxSend: mov.b cp bra BufferTest: btsc bra mov bra [w0++], w1 w1,#0 Z,Again U1STA,#UTXBF BufferTest w1,U1TXREG TxSend ; get char in string ; if Null ; then re-initialize ; ; ; ; see if buffer full wait till empty load value in TX buffer repeat for next char. Delay500mSec ; delay for 500 mS #psvpage(hello),w0 w0, PSVPAG CORCONL,#PSV #psvoffset(hello),w0 U1STA #0x8000,W0 W0,U1MODE #BR,W0 W0, U1BRG U1STA,#UTXEN ;Declare the label for the start of code ;Start of Code section
4
Program Memory
; enable UART module ; set baudrate using formula value ; / ; initiate transmission
DS70051D-page 4-11

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Program Memory module are: Title No related application notes at this time. Application Note #
Note:
DS70051D-page 4-12

Revision B
Revision C
Revision D
Section 4.6 PSV Code Examples, has been added.
4
Program Memory
DS70051D-page 4-13

NOTES:
DS70051D-page 4-14
Section 5. Flash and EEPROM Programming

HIGHLIGHTS
This section of the manual contains the following topics: 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 Introduction .................................................................................................................... 5-2 Table Instruction Operation ............................................................................................ 5-2 Control Registers ........................................................................................................... 5-5 Run-Time Self-Programming (RTSP) .......................................................................... 5-10 Data EEPROM Programming ...................................................................................... 5-15 Design Tips .................................................................................................................. 5-21 Related Application Notes............................................................................................ 5-22 Revision History ........................................................................................................... 5-23
5
Flash and EEPROM Programming
DS70052D-page 5-1

5.1 Introduction
This section describes programming techniques for Flash program memory and data EEPROM memory. The dsPIC30F family of devices contains internal program Flash memory for executing user code. There are two methods by which the user can program this memory: 1. 2. Run-Time Self Programming (RTSP) In-Circuit Serial Programming (ICSP)
RTSP is performed by the users software. ICSP is performed using a serial data connection to the device and allows much faster programming times than RTSP. RTSP techniques are described in this chapter. The ICSP protocol is described in the dsPIC30F Programming Specification document, which may be downloaded from the Microchip web site. The data EEPROM is mapped into the program memory space. The EEPROM is organized as 16-bit wide memory and the memory size can be up to 2K words (4 Kbytes). The amount of EEPROM is device dependent. Refer to the device data sheet for further information. The programming techniques used for the data EEPROM are similar to those used for Flash program memory RTSP. The key difference between Flash and data EEPROM programming operations is the amount of data that can be programmed or erased during each program/erase cycle.
5.2
Table Instruction Operation

The table instructions provide one method of transferring data between the program memory space and the data memory space of dsPIC30F devices. A summary of the table instructions is provided here since they are used during programming of the Flash program memory and data EEPROM. There are four basic table instructions: TBLRDL: Table Read Low TBLRDH: Table Read High TBLWTL: Table Write Low TBLWTH: Table Write High
The TBLRDL and the TBLWTL instructions are used to read and write to bits <15:0> of program memory space. TBLRDL and TBLWTL can access program memory in Word or Byte mode. The TBLRDH and TBLWTH instructions are used to read or write to bits <23:16> of program memory space. TBLRDH and TBLWTH can access program memory in Word or Byte mode. Since the program memory is only 24-bits wide, the TBLRDH and TBLWTH instructions have the ability to address an upper byte of program memory that does not exist. This byte is called the phantom byte. Any read of the phantom byte will return 0x00 and a write to the phantom byte has no effect. Always remember that the 24-bit program memory can be regarded as two side-by-side 16-bit spaces, with each space sharing the same address range. Therefore, the TBLRDL and TBLWTL instructions access the low program memory space (PM<15:0>). The TBLRDH and TBLWTH instructions access the high program memory space (PM<31:16>). Any reads or writes to PM<31:24> will access the phantom (unimplemented) byte. When any of the table instructions are used in Byte mode, the LSb of the table address will be used as the byte select bit. The LSb determines which byte in the high or low program memory space is accessed. Figure 5-1 shows how the program memory is addressed using the table instructions. A 24-bit program memory address is formed using bits <7:0> of the TBLPAG register and the effective address (EA) from a W register, specified in the table instruction. The 24-bit program counter is shown in Figure 5-1 for reference. The upper 23 bits of the EA are used to select the program memory location. For the Byte mode table instructions, the LSb of the W register EA is used to pick which byte of the 16-bit program memory word is addressed. A 1 selects bits <15:8>, a 0 selects bits <7:0>. The LSb of the W register EA is ignored for a table instruction in Word mode. In addition to the program memory address, the table instruction also specifies a W register (or a W register pointer to a memory location) that is the source of the program memory data to be written, or the destination for a program memory read. For a table write operation in Byte mode, bits <15:8> of the source working register are ignored.
DS70052D-page 5-2

Figure 5-1: Addressing for Table Instructions
EA<0> Selects Byte
TBLPAG<7> Selects User/Configuration Space 7 TBLPAG 0
15 EA
8 bits from TBLPAG
16 bits from Wn
24-bit EA
5.2.1
Using Table Read Instructions

Table reads require two steps. First, an address pointer is setup using the TBLPAG register and one of the W registers. Then, the program memory contents at the address location may be read.
5.2.1.1
Read Word Mode The following code example shows how to read a word of program memory using the table instructions in Word mode:
; Setup the address pointer to program space MOV #tblpage(PROG_ADDR),W0 ; get table page value MOV W0,TBLPAG ; load TBLPAG register MOV #tbloffset(PROG_ADDR),W0 ; load address LS word ; Read the program memory location TBLRDH [W0],W3 ; Read high byte to W3 TBLRDL [W0],W4 ; Read low word to W4
5.2.1.2
Read Byte Mode

; Setup the address pointer to program space MOV #tblpage(PROG_ADDR),W0 ; get table page value MOV W0,TBLPAG ; load TBLPAG register MOV #tbloffset(PROG_ADDR),W0 ; load address LS word ; Read the program memory location TBLRDH.B [W0],W3 ; Read high byte to W3 TBLRDL.B [W0++],W4 ; Read low byte to W4 TBLRDL.B [W0++],W5 ; Read middle byte to W5
In the code example above, the post-increment operator on the read of the low byte causes the address in the working register to increment by one. This sets EA<0> to a 1 for access to the middle byte in the third write instruction. The last post-increment sets W0 back to an even address, pointing to the next program memory location. Note: The tblpage() and tbloffset() directives are provided by the Microchip assembler for the dsPIC30F. These directives select the appropriate TBLPAG and W register values for the table instruction from a program memory address value. Refer to the MPLAB ASM 30, MPLAB LINK30 and Utilities Users Guide (DS51317) for further details.
5
DS70052D-page 5-3

5.2.2 Using Table Write Instructions
The effect of a table write instruction will depend on the type of memory technology that is present in the device program memory address space. The program memory address space could contain volatile or non-volatile program memory, non-volatile data memory, and an External Bus Interface (EBI). If a table write instruction occurs within the EBI address region, for example, the write data will be placed onto the EBI data lines. 5.2.2.1 Table Write Holding Latches Table write instructions do not write directly to the non-volatile program and data memory. Instead, the table write instructions load holding latches that store the write data. The holding latches are not memory mapped and can only be accessed using table write instructions. When all of the holding latches have been loaded, the actual memory programming operation is started by executing a special sequence of instructions. The number of holding latches will determine the maximum memory block size that can be programmed and may vary depending on the type of non-volatile memory and the device variant. For example, the number of holding latches could be different for program memory, data EEPROM memory and Device Configuration registers for a given device. In general, the program memory is segmented into rows and panels. Each panel will have its own set of table write holding latches. This allows multiple memory panels to be programmed at once, reducing the overall programming time for the device. For each memory panel, there are generally enough holding latches to program one row of memory at a time. The memory logic automatically decides which set of write latches to load based on the address value used in the table write instruction. Please refer to the specific device data sheet for further details. 5.2.2.2 Write Word Mode The following sequence can be used to write a single program memory latch location in Word mode:
; Setup the address pointer to program space MOV #tblpage(PROG_ADDR),W0 ; get table page value MOV W0,TBLPAG ; load TBLPAG register MOV #tbloffset(PROG_ADDR),W0 ; load address LS word ; Load write data into W registers MOV #PROG_LOW_WORD,W2 MOV #PROG_HI_BYTE,W3 ; Perform the table writes to load the latch TBLWTL W2,[W0] TBLWTH W3,[W0++]
In this example, the contents of the upper byte of W3 does not matter because this data will be written to the phantom byte location. W0 is post-incremented by 2, after the second TBLWTH instruction, to prepare for the write to the next program memory location.
DS70052D-page 5-4

5.2.2.3 Write Byte Mode To write a single program memory latch location in Byte mode, the following code sequence can be used:
; Setup the address pointer to program space MOV #tblpage(PROG_ADDR),W0 ; get table page value MOV W0,TBLPAG ; load TBLPAG register MOV #tbloffset(PROG_ADDR),W0 ; load address LS word ; Load data into working registers MOV #LOW_BYTE,W2 MOV #MID_BYTE,W3 MOV #HIGH_BYTE,W4 ; Write data to the latch ; write high byte TBLWTH.B W4,[W0] TBLWTL.B W2,[W0++] ; write low byte ; write middle byte TBLWTL.B W3,[W0++]
In the code example above, the post-increment operator on the write to the low byte causes the address in W0 to increment by one. This sets EA<0> = 1 for access to the middle byte in the third write instruction. The last post-increment sets W0 back to an even address pointing to the next program memory location.
5.3
Control Registers
Flash and data EEPROM programming operations are controlled using the following Non-Volatile Memory (NVM) control registers: NVMCON: Non-Volatile Memory Control Register NVMKEY: Non-Volatile Memory Key Register NVMADR: Non-Volatile Memory Address Register
5.3.1
NVMCON Register
The NVMCON register is the primary control register for Flash and EEPROM program/erase operations. This register selects Flash or EEPROM memory, whether an erase or program operation will be performed, and is used to start the program or erase cycle. The NVMCON register is shown in Register 5-1. The lower byte of NVMCOM configures the type of NVM operation that will be performed. For convenience, a summary of NVMCON setup values for various program and erase operations is given in Table 5-1. Table 5-1: NVMCON Register Values NVMCON Register Values for RTSP Program and Erase Operations Memory Type Flash PM Operation Erase Program Erase Data EEPROM Program Configuration Register Write(1) Data Size 1 row (32 instr. words) 1 row (32 instr. words) 1 data word 16 data words Entire EEPROM 1 data word 16 data words 1 config. register NVMCON Value 0x4041 0x4001 0x4044 0x4045 0x4046 0x4004 0x4005 0x4008
Note 1: The Device Configuration registers, except for FG5, may be written to a new value without performing an erase cycle.
5
DS70052D-page 5-5

5.3.2 NVM Address Register
There are two NVM Address Registers - NVMADRU and NVMADR. These two registers when concatenated form the 24-bit effective address (EA) of the selected row or word for programming operations. The NVMADRU register is used to hold the upper 8 bits of the EA, while the NVMADR register is used to hold the lower 16 bits of the EA. The register pair, NVMADRU:NVMADR, capture the EA<23:0> of the last table-write instruction that has been executed and select the row of Flash or EEPROM memory to write/erase. Figure 5-2 shows how the program memory EA is formed for programming and erase operations. Although the NVMADRU and NVMADR registers are automatically loaded by the table-write instructions, the user can also directly modify their contents before the programming operation begins. A write to these registers will be required prior to an erase operation, because no table-write instructions are required for any erase operation. Figure 5-2: NVM Addressing with TBLPAG and NVM Address Registers
24-bit PM address
EA<0> is Byte Select
TBLPAG Reg
W Register EA
NVMADRU Register Using NVMADR Addressing 8 bits
NVMADR Register
16 bits
TBLPAG<7> selects User or Configuration Space
NVMADR register loaded with contents of W register EA used during last table-write instruction. NVMADRU register loaded with contents of TBLPAG register during last table-write instruction
5.3.3
NVMKEY Register
NVMKEY is a write only register that is used to prevent accidental writes/erasures of Flash or EEPROM memory. To start a programming or an erase sequence, the following steps must be taken in the exact order shown: 1. 2. 3. Write 0x55 to NVMKEY. Write 0xAA to NVMKEY. Execute two NOP instructions.
After this sequence, a write will be allowed to the NVMCON register for one instruction cycle. In most cases, the user will simply need to set the WR bit in the NVMCON register to start the program or erase cycle. Interrupts should be disabled during the unlock sequence. The code example below shows how the unlock sequence is performed:
PUSH MOV IOR MOV MOV MOV MOV BSET NOP NOP POP SR #0x00E0,W0 SR #0x55,W0 #0xAA,W0 W0,NVMKEY W0,NVMKEY NVMCON,#WR ; Disable interrupts, if enabled
; NOP not required ; Start the program/erase cycle
SR
; Re-enable interrupts
Refer to Section 5.4.2 Flash Programming Operations for further programming examples.
DS70052D-page 5-6

Register 5-1: NVMCON: Non-Volatile Memory Control Register R/W-0 WRERR U-0 U-0 U-0 U-0 U-0 bit 8
Upper Byte: R/S-0 R/W-0 WR WREN bit 15
Lower Byte: R/W-0 bit 7 bit 15
R/W-0
R/W-0
R/W-0 R/W-0 PROGOP<7:0>
R/W-0
R/W-0
R/W-0 bit 0
WR: Write (Program or Erase) Control bit 1 = Initiates a data EEPROM or program Flash erase or write cycle (the WR bit can be set but not cleared in software) 0 = Write cycle is complete WREN: Write (Erase or Program) Enable bit 1 = Enable an erase or program operation 0 = No operation allowed (Device clears this bit on completion of the write/erase operation) WRERR: Flash Error Flag bit 1 = A write operation is prematurely terminated (any MCLR or WDT Reset during programming operation) 0 = The write operation completed successfully Reserved: User code should write 0s to these locations PROGOP<7:0>: Programming Operation Command Byte bits Erase Operations: 0x41 = Erase 1 row (32 instruction words) of program Flash 0x44 = Erase 1 data word from data EEPROM 0x45 = Erase 1 row (16 data words) from data EEPROM 0x46 = Erase entire data EEPROM Programming Operations: 0x01 = Program 1 row (32 instruction words) into Flash program memory 0x04 = Program 1 data word into data EEPROM 0x05 = Program 1 row (16 data words) into data EEPROM 0x08 = Program 1 data word into device configuration register Legend: R = Readable bit S = Settable bit 0 = Bit is cleared W = Writable bit -n = Value at POR x = Bit is unknown U = Unimplemented bit, read as 0 1 = Bit is set
bit 14
bit 13
bit 12-8 bit 7-0
5
DS70052D-page 5-7

Register 5-2: Upper Byte: R/W-x bit 15 Lower Byte: R/W-x bit 7 bit 15-0 NVMADR<15:0>: NV Memory Write Address bits Selects the location to program or erase in program or data Flash memory. This register may be read or written by user. This register will contain the address of EA<15:0> of the last table write instruction executed, until written by the user. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown NVMADR: Non-Volatile Memory Address Register R/W-x R/W-x R/W-x R/W-x NVMADR<15:8> R/W-x R/W-x R/W-x bit 8
R/W-x
R/W-x
R/W-x R/W-x NVMADR<7:0>
R/W-x
R/W-x
R/W-x bit 0
DS70052D-page 5-8

Register 5-3: Upper Byte: U-0 bit 15 NVMADRU: Non-Volatile Memory Upper Address Register U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8
Lower Byte: R/W-x bit 7 bit 15-8 bit 7-0
R/W-x
R/W-x
R/W-x R/W-x NVMADRU<7:0>
R/W-x
R/W-x
R/W-x bit 0
Unimplemented: Read as 0 NVMADRU<7:0>: NV Memory Upper Write Address bits Selects the upper 8 bits of the location to program or erase in program or data Flash memory. This register may be read or written by the user. This register will contain the value of the TBLPAG register when the last table write instruction executed, until written by the user. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
Register 5-4: Upper Byte: U-0 bit 15
NVMKEY: Non-Volatile Memory Key Register U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8
Lower Byte: W-0 bit 7
W-0
W-0
W-0 W-0 NVMKEY<7:0>
W-0
W-0
W-0 bit 0
bit 15-8 Unimplemented: Read as 0 bit 7-0 NVMKEY<7:0>: Key Register (Write Only) bits Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
5
DS70052D-page 5-9

5.4 Run-Time Self-Programming (RTSP)
RTSP allows the user code to modify Flash program memory contents. RTSP is accomplished using TBLRD (table read) and TBLWT (table write) instructions, and the NVM Control registers. With RTSP, the user may erase program memory, 32 instructions (96 bytes) at a time and can write program memory data, 32 instructions (96 bytes) at a time.
5.4.1
RTSP Operation
The dsPIC30F Flash program memory is organized into rows and panels. Each row consists of 32 instructions or 96 bytes. The panel size may vary depending on the dsPIC30F device variant. Refer to the device data sheet for further information. Typically, each panel consists of 128 rows, or 4K x 24 instructions. RTSP allows the user to erase one row (32 instructions) at a time and to program 32 instructions at one time. Each panel of program memory contains write latches that hold 32 instructions of programming data. These latches are not memory mapped. The only way for the user to access the write latches is through the use of table write instructions. Prior to the actual programming operation, the write data must be loaded into the panel write latches with table write instructions. The data to be programmed into the panel is typically loaded in sequential order into the write latches: instruction 0, instruction 1, etc. The instruction words loaded must always be from an even group of four address boundaries (e.g., loading of instructions 3, 4, 5, 6 is not allowed). Another way of stating this requirement is that the starting program memory address of the four instructions must have the 3 LSbs equal to 0. All 32 write latches must be written during a programming operation to ensure that any old data held in the latches is overwritten. The basic sequence for RTSP programming is to setup a table pointer, then do a series of TBLWT instructions to load the write latches. Programming is performed by setting special bits in the NVMCON register. 32 TBLWTL and 32 TBLWTH instructions are required to load the four instructions. If multiple, discontinuous regions of program memory need to be programmed, the table pointer should be changed for each region and the next set of write latches written. All of the table write operations to the Flash program memory take 2 instruction cycles each, because only the table latches are written. The actual programming operation is initiated using the NVMCON register.
5.4.2
Flash Programming Operations

A program/erase operation is necessary for programming or erasing the internal Flash program memory in RTSP mode. The program or erase operation is automatically timed by the device and is nominally 2 msec in duration. Setting the WR bit (NVMCON<15>) starts the operation and the WR bit is automatically cleared when the operation is finished. The CPU stalls (waits) until the programming operation is finished. The CPU will not execute any instruction or respond to interrupts during this time. If any interrupts do occur during the programming cycle, then they will remain pending until the cycle completes.
DS70052D-page 5-10

5.4.2.1 Flash Program Memory Programming Algorithm The user can erase and program Flash Program Memory by rows (32 instruction words). The general process is as follows: 1. Read one row of program Flash (32 instruction words) and store into data RAM as a data image. The RAM image must be read from an even 32-word program memory address boundary. Update the RAM data image with the new program memory data. Erase program Flash row. Setup NVMCON register to erase 1 row of Flash program memory. Write address of row to be erased into NVMADRU and NVMADR registers. Disable interrupts. Write the key sequence to NVMKEY to enable the erase. Set the WR bit. This will begin erase cycle. CPU will stall for the duration of the erase cycle. The WR bit is cleared when erase cycle ends. Re-enable interrupts. Write 32 instruction words of data from RAM into the Flash program memory write latches. Program 32 instruction words into program Flash. Setup NVMCON to program one row of Flash program memory. Disable interrupts. Write the key sequence to NVMKEY to enable the program cycle. Set the WR bit. This will begin the program cycle. CPU will stall for duration of the program cycle. The WR bit is cleared by the hardware when program cycle ends. Re-enable interrupts. Repeat steps 1 through 6, as needed, to program the desired amount of Flash program memory Note: The user should remember that the minimum amount of program memory that can be modified using RTSP is 32 instruction word locations. Therefore, it is important that an image of these locations be stored in general purpose RAM before an erase cycle is initiated. An erase cycle must be performed on any previously written locations before any programming is done.
2. 3.
4. 5.
6.
5
DS70052D-page 5-11

5.4.2.2 Erasing a Row of Program Memory The following is a code sequence that can be used to erase a row (32 instructions) of program memory. The NVMCON register is configured to erase one row of program memory. The NVMADRU and NVMADR registers are loaded with the address of the row to be erased. The program memory must be erased at even row boundaries. Therefore, the 6 LSbits of the value written to the NVMADR register have no effect when a row is erased. The erase operation is initiated by writing a special unlock, or key sequence to the NVMKEY register before setting the WR control bit (NVMCON<15>). The unlock sequence needs to be executed in the exact order shown without interruption. Therefore, interrupts should be disabled prior to writing the sequence. Two NOP instructions should be inserted in the code at the point where the CPU will resume operation. Finally, interrupts can be enabled (if required).
; Setup NVMCON to erase one row of Flash program memory MOV #0x4041,W0 MOV W0,NVMCON ; Setup address pointer to row to be ERASED MOV #tblpage(PROG_ADDR),W0 MOV W0,NVMADRU MOV #tbloffset(PROG_ADDR),W0 MOV W0,NVMADR ; Disable interrupts, if enabled PUSH SR MOV #0x00E0,W0 IOR SR ; Write the KEY sequence MOV #0x55,W0 MOV W0, NVMKEY MOV #0xAA, W0 MOV W0, NVMKEY ; Start the erase operation BSET NVMCON,#WR ; Insert two NOPs after the erase cycle (required) NOP NOP ; Re-enable interrupts, if needed POP SR
Note:
When erasing a row of program memory, the user writes the upper 8 bits of the erase address directly to the NVMADRU and NVMADR registers. Together, the contents of the NVMADRU and NVMADR registers form the complete address of the program memory row to be erased. The NVMADRU and NVMADR registers specify the address for all Flash erase and program operations. However, these two registers do not have to be directly written by the user for Flash program operations. This is because the table write instructions used to write the program memory data automatically transfers the TBLPAG register contents and the table write address into the NVMADRU and NVMADR registers. The above code example could be modified to perform a dummy table write operation to capture the program memory erase address.
DS70052D-page 5-12

5.4.2.3 Loading Write Latches The following is a sequence of instructions that can be used to load the 768-bits of write latches (32 instruction words). 32 TBLWTL and 32 TBLWTH instructions are needed to load the write latches selected by the table pointer. The TBLPAG register is loaded with the 8 MSbits of the program memory address. The user does not need to write the NVMADRU:NVMADR register-pair for a Flash programming operation. The 24-bits of the program memory address are automatically captured into the NVMADRU:NVMADR register-pair when each table write instruction is executed. The program memory must be programmed at an even 32 instruction word address boundary. In effect, the 6 LSbits of the value captured in the NVMADR register are not used during the programming operation. The row of 32 instruction words do not necessarily have to be written in sequential order. The 6 LSbits of the table write address determine which of the latches will be written. However, all 32 instruction words should be written for each programming cycle to overwrite old data. Note: The following code example is the Load_Write_Latch code referred to in subsequent examples.
; Set up a pointer to the first program memory location to be written. MOV #tblpage(PROG_ADDR),W0 MOV W0,TBLPAG MOV #tbloffset(PROG_ADDR),W0 ; Perform the TBLWT instructions to write the latches ; W0 is incremented in the TBLWTH instruction to point to the ; next instruction location. MOV MOV TBLWTL TBLWTH MOV MOV TBLWTL TBLWTH MOV MOV TBLWTL TBLWTH MOV MOV TBLWTL TBLWTH ........ ........ MOV MOV TBLWTL TBLWTH #LOW_WORD_0,W2 #HIGH_BYTE_0,W3 W2,[W0] W3,[W0++] ; #LOW_WORD_1,W2 #HIGH_BYTE_1,W3 W2,[W0] W3,[W0++] ; #LOW_WORD_2,W2 #HIGH_BYTE_2,W3 W2, [W0] ; W3, [W0++] #LOW_WORD_3,W2 #HIGH_BYTE_3,W3 W2,[W0] W3,[W0++] ;
1st_program_word
2nd_program_word
3rd_program_word
4th_program_word
#LOW_WORD_31,W2 #HIGH_BYTE_31,W3 W2,[W0] W3,[W0++] ; 32nd_program_word
5
DS70052D-page 5-13

5.4.2.4 Single Row Programming Example An example of single row programming code is:
; Setup NVMCON to write 1 row of program memory MOV #0x4001,W0 MOV W0,NVMCON ; Load the 32 program memory write latches CALL Load_Write_Latch(1)
; Disable interrupts, if enabled

PUSH SR MOV #0x00E0,W0 IOR SR Write the KEY sequence MOV #0x55,W0 MOV W0,NVMKEY MOV #0xAA,W0 MOV W0,NVMKEY Start the programming sequence BSET NVMCON,#WR Insert two NOPs after programming NOP NOP Re-enable interrupts, if required POP SR
; ;
Note
1: See Section 5.4.2.3 Loading Write Latches
5.4.3
Writing to Device Configuration Registers

RTSP may be used to write to the Device Configuration registers. RTSP allows each Configuration register, except the FG5, to be individually rewritten without first performing an erase cycle. Caution must be exercised when writing the Configuration registers since they control critical device operating parameters, such as the system clock source, PLL multiplication ratio and WDT enable. The procedure for programming a Device Configuration register is similar to the procedure for Flash program memory, except that only TBLWTL instructions are required. This is because the upper 8 bits are unused in each Device Configuration register. Furthermore, bit 23 of the table write address must be set to access the Configuration registers. Refer to Section 24. Device Configuration and the device data sheet for a full description of the Device Configuration registers.
5.4.3.1
Configuration Register Write Algorithm 1. 2. 3. 4. 5. 6. 7. Write the new configuration value to the table write latch using a TBLWTL instruction. Configure NVMCON for a Configuration register write (NVMCON = 0x4008). Disable interrupts, if enabled. Write the key sequence to NVMKEY. Start the write sequence by setting WR (NVMCON<15>). CPU execution will resume when the write is finished. Re-enable interrupts, if needed.
DS70052D-page 5-14

5.4.3.2 Configuration Register Write Code Example The following code sequence can be used to modify a Device Configuration register:
; Set up a pointer to the location to be written. MOV #tblpage(CONFIG_ADDR),W0 MOV W0,TBLPAG MOV #tbloffset(CONFIG_ADDR),W0 ; Get the new data to write to the configuration register MOV #ConfigValue,W1 ; Perform the table write to load the write latch TBLWTL W1,[W0] ; Configure NVMCON for a configuration register write MOV #0x4008,W0 MOV W0,NVMCON ; Disable interrupts, if enabled PUSH SR MOV #0x00E0,W0 IOR SR ; Write the KEY sequence MOV #0x55,W0 MOV W0,NVMKEY MOV #0xAA,W0 MOV W0,NVMKEY ; Start the programming sequence BSET NVMCON,#WR ; Insert two NOPs after programming NOP NOP ; Re-enable interrupts, if required POP SR
5.5
Data EEPROM Programming

The EEPROM block is accessed using table read and write operations similar to the program memory. The TBLWTH and TBLRDH instructions are not required for EEPROM operations since the memory is only 16-bits wide. The program and erase procedures for the data EEPROM are similar to those used for the Flash program memory, except they are optimized for fast data access. The following programming operations can be performed on the data EEPROM: Erase one word Erase one row (16 words) Erase entire data EEPROM Program one word Program one row (16 words)
The data EEPROM is readable and writable during normal operation (full VDD operating range). Unlike the Flash program memory, normal program execution is not stopped during an EEPROM program or erase operation. EEPROM erase and program operations are performed using the NVMCON and NVMKEY registers. The programming software is responsible for waiting for the operation to complete. The software may detect when the EEPROM erase or programming operation is complete by one of three methods: Poll the WR bit (NVMCON<15>) in software. The WR bit will be cleared when the operation is complete. Poll the NVMIF bit (IFS0<12>) in software. The NVMIF bit will be set when the operation is complete. Enable NVM interrupts. The CPU will be interrupted when the operation is complete. Further programming operations can be handled in the ISR. Note: Unexpected results will be obtained should the user attempt to read the EEPROM while a programming or erase operation is underway.
5
DS70052D-page 5-15

5.5.1 EEPROM Single Word Programming Algorithm
1. Erase one EEPROM word. Setup NVMCON register to erase one EEPROM word. Write address of word to be erased into NVMADRU, NVMADR registers. Clear NVMIF status bit and enable NVM interrupt (optional). Write the key sequence to NVMKEY. Set the WR bit. This will begin erase cycle. Either poll the WR bit or wait for the NVM interrupt. Write data word into data EEPROM write latch. Program the data word into the EEPROM. Setup the NVMCON register to program one EEPROM word. Clear NVMIF status bit and enable NVM interrupt (optional). Write the key sequence to NVMKEY. Set the WR bit. This will begin the program cycle. Either poll the WR bit or wait for the NVM interrupt.
2. 3.
5.5.2
EEPROM Row Programming Algorithm

If multiple words need to be programmed into the EEPROM, it is quicker to erase and program 16 words (1 row) at a time. The process to program 16 words of EEPROM is: 1. 2. 3. Read one row of data EEPROM (16 words) and store into data RAM as a data image. The section of EEPROM to be modified must fall on an even 16-word address boundary. Update the data image with the new data. Erase the EEPROM row. Setup the NVMCON register to erase one row of EEPROM. Write starting address of row to be erased into NUMADRU and NVMADR registers. Clear NVMIF status bit and enable NVM interrupt (optional). Write the key sequence to NVMKEY. Set the WR bit. This will begin the erase cycle. Either poll the WR bit or wait for the NVM interrupt. Write the 16 data words into the data EEPROM write latches. Program a row into data EEPROM. Setup the NVMCON register to program one row of EEPROM. Clear NVMIF status bit and enable NVM interrupt (optional). Write the key sequence to NVMKEY. Set the WR bit. This will begin the program cycle. Either poll the WR bit or wait for the NVM interrupt.
4. 5.
DS70052D-page 5-16

5.5.3 Erasing One Word of Data EEPROM Memory
The NVMADRU and NVMADR registers must be loaded with the data EEPROM address to be erased. Since one word of the EEPROM is accessed, the LSB of the NVMADR has no effect on the erase operation. The NVMCON register must be configured to erase one word of EEPROM memory. Setting the WR control bit (NVMCON<15>) initiates the erase. A special unlock or key sequence should be written to the NVMKEY register before setting the WR control bit. The unlock sequence needs to be executed in the exact order shown without interruption. Therefore, interrupts should be disabled prior to writing the sequence.
; Set up a pointer to the EEPROM location to be erased. MOV #tblpage(EE_ADDR),W0 MOV W0,NVMADRU MOV #tbloffset(EE_ADDR),W0 MOV W0,NVMADR ; Setup NVMCON to erase one word of data EEPROM MOV #0x4044,W0 MOV W0,NVMCON ; Disable interrupts while the KEY sequence is written PUSH SR MOV #0x00E0,W0 IOR SR ; Write the KEY sequence MOV #0x55,W0 MOV W0,NVMKEY MOV #0xAA,W0 MOV W0,NVMKEY ; Start the erase cycle BSET NVMCON,#WR ; Re-enable interrupts POP SR
5
DS70052D-page 5-17

5.5.4 Writing One Word of Data EEPROM Memory
Assuming the user has erased the EEPROM location to be programmed, use a table write instruction to write one write latch. The TBLPAG register is loaded with the 8 MSBs of the EEPROM address. The 16 LSBs of the EEPROM address are automatically captured into the NVMADR register when the table write is executed. The LSB of the NVMADR register has no effect on the programming operation. The NVMCON register is configured to program one word of data EEPROM. Setting the WR control bit (NVMCON<15>) initiates the programming operation. A special unlock or key sequence should be written to the NVMKEY register before setting the WR control bit. The unlock sequence needs to be executed in the exact order shown without interruption. Therefore, interrupts should be disabled prior to writing the sequence.
; Setup a pointer to data EEPROM MOV #tblpage(EE_ADDR),W0 MOV W0,TBLPAG MOV #tbloffset(EE_ADDR),W0 ; Write data value to holding latch MOV EE_DATA,W1 TBLWTL W1,[ W0] ; NVMADR captures write address from the TBLWTL instruction. ; Setup NVMCON for programming one word to data EEPROM MOV #0x4004,W0 MOV W0,NVMCON ; Disable interrupts while the KEY sequence is written PUSH SR MOV #0x00E0,W0 IOR SR ; Write the key sequence MOV #0x55,W0 MOV W0,NVMKEY MOV #0xAA,W0 MOV W0,NVMKEY ; Start the write cycle BSET NVMCON,#WR ;Re-enable interrupts, if needed POP SR
DS70052D-page 5-18

5.5.5 Erasing One Row of Data EEPROM
The NVMCON register is configured to erase one row of EEPROM memory. The NVMADRU and NVMADR registers must point to the row to be erased. The data EEPROM must be erased at even address boundaries. Therefore, the 5 LSBs of the NVMADR register will have no effect on the row that is erased. Setting the WR control bit (NVMCON<15>) initiates the erase. A special unlock or key sequence should be written to the NVMKEY register before setting the WR control bit. The unlock sequence needs to be executed in the exact order shown without interruption. Therefore, interrupts should be disabled prior to writing the sequence.
; Set up a pointer to the EEPROM row to be erased. MOV #tblpage(EE_ADDR),W0 MOV W0,NVMADRU MOV #tbloffset(EE_ADDR),W0 MOV W0,NVMADR ; Setup NVMCON to erase one row of EEPROM MOV #0x4045,W0 MOV W0,NVMCON ; Disable interrupts while the KEY sequence is written PUSH SR MOV #0x00E0,W0 IOR SR ; Write the KEY Sequence MOV #0x55,W0 MOV W0,NVMKEY MOV #0xAA,W0 MOV W0,NVMKEY ; Start the erase operation BSET NVMCON,#WR ;Re-enable interrupts, if needed POP SR
5
DS70052D-page 5-19

5.5.6 Write One Row of Data EEPROM Memory
To write a row of data EEPROM, all sixteen write latches must be written before the programming sequence is initiated. The TBLPAG register is loaded with the 8 MSbs of the EEPROM address. The 16 LSbs of the EEPROM address are automatically captured into the NVMADR register when each table write is executed. Data EEPROM row programming must occur at even address boundaries, so the 5 LSbs of the NVMADR register have no effect on the row that is programmed. Setting the WR control bit (NVMCON<15>) initiates the programming operation. A special unlock or key sequence should be written to the NVMKEY register before setting the WR control bit. The unlock sequence needs to be executed in the exact order shown without interruption. Therefore, interrupts should be disabled prior to writing the sequence.
; Set up a pointer to the EEPROM row to be programmed. MOV #tblpage(EE_ADDR),W0 MOV W0,TBLPAG MOV #tbloffset(EE_ADDR),W0 ; Write the data to the programming latches. MOV data_ptr,W1 ; Use W1 as pointer to the data. ; Write 1st data word TBLWTL [W1++],[W0++] TBLWTL [W1++],[W0++] ; Write 2nd data word ; Write 3rd data word TBLWTL [W1++],[W0++] ; Write 4th data word TBLWTL [W1++],[W0++] TBLWTL [W1++],[W0++] ; Write 5th data word ; Write 6th data word TBLWTL [W1++],[W0++] ; Write 7th data word TBLWTL [W1++],[W0++] TBLWTL [W1++],[W0++] ; Write 8th data word ; Write 9th data word TBLWTL [W1++],[W0++] ; Write 10th data word TBLWTL [W1++],[W0++] TBLWTL [W1++],[W0++] ; Write 11th data word ; Write 12th data word TBLWTL [W1++],[W0++] ; Write 13th data word TBLWTL [W1++],[W0++] TBLWTL [W1++],[W0++] ; Write 14th data word ; Write 15th data word TBLWTL [W1++],[W0++] ; Write 16th data word TBLWTL [W1++],[W0++] ; The NVMADR captures last table access address. ; Setup NVMCON to write one row of EEPROM MOV #0x4005,W0 MOV W0,NVMCON ; Disable interrupts while the KEY sequence is written PUSH SR MOV #0x00E0,W0 IOR SR ; Write the KEY sequence MOV #0x55,W0 MOV W0,NVMKEY MOV #0xAA,W0 MOV W0,NVMKEY ; Start the programming operation BSET NVMCON,#WR ;Re-enable interrupts, if needed POP SR
Note:
Sixteen table write instructions have been used in this code segment to provide clarity in the example. The code segment could be simplified by using a single table write instruction in a REPEAT loop.
DS70052D-page 5-20

5.5.7 Reading the Data EEPROM Memory
A TBLRD instruction reads a word at the current program word address. This example uses W0 as a pointer to data Flash. The result is placed into register W4.
; Setup pointer to EEPROM memory MOV #tblpage(EE_ADDR),W0 MOV W0,TBLPAG MOV #tbloffset(EE_ADDR),W0 ; Read the EEPROM data TBLRDL [W0],W4
Note:
Program Space Visibility (PSV) can also be used to read locations in the program memory address space. See Section 4. Program Memory for further information about PSV.
5.6
Design Tips
Question 1: I cannot get the device to program or erase properly. My code appears to be correct. What could be the cause?
Answer: Interrupts should be disabled when a program or erase cycle is initiated to ensure that the key sequence executes without interruption. Interrupts can be disabled by raising the current CPU priority to level 7. The code examples in this chapter disable interrupts by saving the current SR register value on the stack, then ORing the value 0x00E0 with SR to force IPL<2:0> = 111. If no priority level 7 interrupts are enabled, then the DISI instruction provides another method to temporarily disable interrupts, while the key sequence is executed. Question 2: What is an easy way to read data EEPROM without using table instructions?
Answer: The data EEPROM is mapped into the program memory space. PSV can be used to map the EEPROM region into data memory space. See Section 4. Program Memory for further information about PSV.
5
DS70052D-page 5-21

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Flash and EEPROM Programming module are: Title Using the dsPIC30F for Sensorless BLDC Control Note: Application Note # AN901
DS70052D-page 5-22

Revision B
This revision incorporates technical content changes for the dsPIC30F Flash and EEPROM Programming module.
Revision C
Revision D
This revision incorporates technical content changes for the dsPIC30F Flash and EEPROM Programming module.
5
DS70052D-page 5-23

NOTES:
DS70052D-page 5-24
6
Interrupts
Section 6. Reset Interrupts

HIGHLIGHTS
This section of the manual contains the following topics: 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 Introduction .................................................................................................................... 6-2 Non-Maskable Traps...................................................................................................... 6-6 Interrupt Processing Timing ......................................................................................... 6-11 Interrupt Control and Status Registers......................................................................... 6-14 Interrupt Setup Procedures.......................................................................................... 6-42 Design Tips .................................................................................................................. 6-44 Related Application Notes............................................................................................ 6-45 Revision History ........................................................................................................... 6-46
DS70053C-page 6-1

6.1 Introduction
The dsPIC30F interrupt controller module reduces the numerous peripheral interrupt request signals to a single interrupt request signal to the dsPIC30F CPU and has the following features: Up to 8 processor exceptions and software traps 7 user selectable priority levels Interrupt Vector Table (IVT) with up to 62 vectors A unique vector for each interrupt or exception source Fixed priority within a specified user priority level Alternate Interrupt Vector Table (AIVT) for debug support Fixed interrupt entry and return latencies
6.1.1
Interrupt Vector Table

The Interrupt Vector Table (IVT) is shown in Figure 6-1. The IVT resides in program memory, starting at location 0x000004. The IVT contains 62 vectors consisting of 8 non-maskable trap vectors plus up to 54 sources of interrupt. In general, each interrupt source has its own vector. Each interrupt vector contains a 24-bit wide address. The value programmed into each interrupt vector location is the starting address of the associated Interrupt Service Routine (ISR).
6.1.2
Alternate Vector Table

The Alternate Interrupt Vector Table (AIVT) is located after the IVT as shown in Figure 6-1. Access to the AIVT is provided by the ALTIVT control bit (INTCON2<15>). If the ALTIVT bit is set, all interrupt and exception processes will use the alternate vectors instead of the default vectors. The alternate vectors are organized in the same manner as the default vectors. The AIVT supports emulation and debugging efforts by providing a means to switch between an application and a support environment without requiring the interrupt vectors to be reprogrammed. This feature also enables switching between applications for evaluation of different software algorithms at run-time. If the AIVT is not needed, the AIVT should be programmed with the same addresses used in the IVT.
6.1.3
Reset Sequence
A device Reset is not a true exception because the interrupt controller is not involved in the Reset process. The dsPIC30F device clears its registers in response to a Reset which forces the PC to zero. The processor then begins program execution at location 0x000000. The user programs a GOTO instruction at the Reset address which redirects program execution to the appropriate start-up routine. Note: Any unimplemented or unused vector locations in the IVT and AIVT should be programmed with the address of a default interrupt handler routine that contains a RESET instruction.
DS70053C-page 6-2
Section 6. Interrupts
6
Figure 6-1: Interrupt Vector Table
Interrupts
IVT
AIVT
Reset GOTO Instruction Reset GOTO Address Reserved Oscillator Fail Trap Vector Address Error Trap Vector Stack Error Trap Vector Arithmetic Error Trap Vector Reserved Reserved Reserved Interrupt Vector 0 Interrupt Vector 1 ~ ~ ~ Interrupt Vector 52 Interrupt Vector 53 Reserved Reserved Reserved Oscillator Fail Trap Vector Address Error Trap Vector Stack Error Trap Vector Arithmetic Error Trap Vector Reserved Reserved Reserved Interrupt Vector 0 Interrupt Vector 1 ~ ~ ~ Interrupt Vector 52 Interrupt Vector 53
0x000000 0x000002 0x000004
Decreasing Natural Order Priority
0x000014
0x00007E 0x000080 0x000082 0x000084
See Table 6-2 for Interrupt Vector details.
0x000094
0x0000FE
Table 6-1: Vector Number 0 1 2 3 4 5 6 7
Trap Vector Details IVT Address 0x000004 0x000006 0x000008 0x00000A 0x00000C 0x00000E 0x000010 0x000012 AIVT Address 0x000084 0x000086 0x000088 0x00008A 0x00008C 0x00008E 0x000090 0x000092 Reserved Oscillator Failure Address Error Stack Error Arithmetic Error Reserved Reserved Reserved Trap Source
DS70053C-page 6-3

Table 6-2:
Vector Number 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53-61
Interrupt Vector Details

IVT Address 0x000014 0x000016 0x000018 0x00001A 0x00001C 0x00001E 0x000020 0x000022 0x000024 0x000026 0x000028 0x00002A 0x00002C 0x00002E 0x000030 0x000032 0x000034 0x000036 0x000038 0x00003A 0x00003C 0x00003E 0x000040 0x000042 0x000044 0x000046 0x000048 0x00004A 0x00004C 0x00004E 0x000050 0x000052 0x000054 0x000056 0x000058 0x00005A 0x00005C 0x00005E 0x000060 0x000062 0x000064 0x000066 0x000068 0x00006A 0x00006C 0x00006E-0x00007E AIVT Address 0x000094 0x000096 0x000098 0x00009A 0x00009C 0x00009E 0x0000A0 0x0000A2 0x0000A4 0x0000A6 0x0000A8 0x0000AA 0x0000AC 0x0000AE 0x0000B0 0x0000B2 0x0000B4 0x0000B6 0x0000B8 0x0000BA 0x0000BC 0x0000BE 0x0000C0 0x0000C2 0x0000C4 0x0000C6 0x0000C8 0x0000CA 0x0000CC 0x0000CE 0x0000D0 0x0000D2 0x0000D4 0x0000D6 0x0000D8 0x0000DA 0x0000DC 0x0000DE 0x0000E0 0x0000E2 0x0000E4 0x0000E6 0x0000E8 0x0000EA 0x0000EC Interrupt Source INT0 External Interrupt 0 IC1 Input Compare 1 OC1 Output Compare 1 T1 Timer 1 IC2 Input Capture 2 OC2 Output Compare 2 T2 Timer 2 T3 Timer 3 SPI1 U1RX UART1 Receiver U1TX UART1 Transmitter ADC ADC Convert Done NVM NVM Write Complete I2C Slave Operation Message Detect I2C Master Operation Message Event Complete Change Notice Interrupt INT1 External Interrupt 1 IC7 Input Capture 7 IC8 Input Capture 8 OC3 Output Compare 3 OC4 Output Compare 4 T4 Timer 4 T5 Timer 5 INT2 External Interrupt 2 U2RX UART2 Receiver U2TX UART2 Transmitter SPI2 CAN1 IC3 Input Capture 3 IC4 Input Capture 4 IC5 Input Capture 5 IC6 Input Capture 6 OC5 Output Compare 5 OC6 Output Compare 6 OC7 Output Compare 7 OC8 Output Compare 8 INT3 External Interrupt 3 INT4 External Interrupt 4 CAN2 PWM PWM Period Match QEI Position Counter Compare DCI Codec Transfer Done LVD Low Voltage Detect FLTA MCPWM Fault A FLTB MCPWM Fault B
0x00006E-0x00007E Reserved
DS70053C-page 6-4
6
6.1.4 CPU Priority Status
Interrupts
The CPU can operate at one of sixteen priority levels, 0-15. An interrupt or trap source must have a priority level greater than the current CPU priority in order to initiate an exception process. Peripheral and external interrupt sources can be programmed for level 0-7, while CPU priority levels 8-15 are reserved for trap sources. A trap is a non-maskable interrupt source intended to detect hardware and software problems (see Section 6.2 Non-Maskable Traps). The priority level for each trap source is fixed and only one trap is assigned to a priority level. Note that an interrupt source programmed to priority level 0 is effectively disabled, since it can never be greater than the CPU priority. The current CPU priority level is indicated by the following four status bits: IPL<2:0> status bits located in SR<7:5> IPL3 status bit located in CORCON<3> The IPL<2:0> status bits are readable and writable, so the user may modify these bits to disable all sources of interrupts below a given priority level. If IPL<2:0> = 3, for example, the CPU would not be interrupted by any source with a programmed priority level of 0, 1, 2 or 3. Trap events have higher priority than any user interrupt source. When the IPL3 bit is set, a trap event is in progress. The IPL3 bit can be cleared, but not set by the user. In some applications, it may be desirable to clear the IPL3 bit when a trap has occurred and branch to an instruction other than the instruction after the one that originally caused the trap to occur. All user interrupt sources can be disabled by setting IPL<2:0> = 111. Note: The IPL<2:0> bits become read only bits when interrupt nesting is disabled. See Section 6.2.4.2 Interrupt Nesting for more information.
6.1.5
Interrupt Priority
Each peripheral interrupt source can be assigned to one of seven priority levels. The user assignable interrupt priority control bits for each individual interrupt are located in the Least Significant 3 bits of each nibble within the IPCx register(s). Bit 3 of each nibble is not used and is read as a 0. These bits define the priority level assigned to a particular interrupt. The usable priority levels start at 1 as the lowest priority and level 7 as the highest priority. If the IPC bits associated with an interrupt source are all cleared, then the interrupt source is effectively disabled. Since more than one interrupt request source may be assigned to a specific priority level, a means is provided to resolve priority conflicts within a given user assigned level. Each source of interrupt has a natural order priority based on its location in the IVT. Table 6-2 shows the location of each interrupt source in the IVT. The lower numbered interrupt vectors have higher natural priority, while the higher numbered vectors have lower natural priority. The overall priority level for any pending source of interrupt is determined first by the user assigned priority of that source in the IPCx register, then by the natural order priority within the IVT. Natural order priority is used only to resolve conflicts between simultaneous pending interrupts with the same user assigned priority level. Once the priority conflict is resolved and the exception process begins, the CPU can only be interrupted by a source with higher user assigned priority. Interrupts with the same user assigned priority but a higher natural order priority, that become pending after the exception process begins, will remain pending until the current exception process completes. The ability for the user to assign each interrupt source to one of seven priority levels means that the user can give an interrupt with a low natural order priority a very high overall priority level. For example: the PLVD (Programmable Low Voltage Detect) can be given a priority of 7 and the INT0 (External Interrupt 0) may be assigned to priority level 1, thus giving it a very low effective priority. Note: The peripherals and sources of interrupt available in the IVT will vary depending on the specific dsPIC30F device. The sources of interrupt shown in this document represent a comprehensive listing of all interrupt sources found on dsPIC30F devices. Refer to the specific device data sheet for further details.
DS70053C-page 6-5

6.2 Non-Maskable Traps
Traps can be considered as non-maskable, nestable interrupts which adhere to a fixed priority structure. Traps are intended to provide the user a means to correct erroneous operation during debug and when operating within the application. If the user does not intend to take corrective action in the event of a trap error condition, these vectors must be loaded with the address of a software routine that will reset the device. Otherwise, the trap vector is programmed with the address of a service routine that will correct the trap condition. The dsPIC30F has four implemented sources of non-maskable traps: Oscillator Failure Trap Stack Error Trap Address Error Trap Arithmetic Error Trap
Note that many of these trap conditions can only be detected when they happen. Consequently, the instruction that caused the trap is allowed to complete before exception processing begins. Therefore, the user may have to correct the action of the instruction that caused the trap. Each trap source has a fixed priority as defined by its position in the IVT. An oscillator failure trap has the highest priority, while an arithmetic error trap has the lowest priority (see Figure 6-1). In addition, trap sources are classified into two distinct categories: Hard traps and Soft traps.
6.2.1
Soft Traps
The arithmetic error trap (priority level 11) and stack error trap (priority level 12) are categorized as soft trap sources. Soft traps can be treated like non-maskable sources of interrupt that adhere to the priority assigned by their position in the IVT. Soft traps are processed like interrupts and require 2 cycles to be sampled and Acknowledged prior to exception processing. Therefore, additional instructions may be executed before a soft trap is Acknowledged.
6.2.1.1
Stack Error Trap (Soft Trap, Level 12) The stack is initialized to 0x0800 during Reset. A stack error trap will be generated should the stack pointer address ever be less than 0x0800. There is a Stack Limit register (SPLIM) associated with the stack pointer that is uninitialized at Reset. The stack overflow check is not enabled until a word write to SPLIM occurs. All Effective Addresses (EA) generated using W15 as a source or destination pointer are compared against the value in SPLIM. Should the EA be greater than the contents of the SPLIM register, then a stack error trap is generated. In addition, a stack error trap will be generated should the EA calculation wrap over the end of data space (0xFFFF). A stack error can be detected in software by polling the STKERR status bit (INTCON1<2>). To avoid re-entering the Trap Service Routine, the STKERR status flag must be cleared in software prior to returning from the trap with a RETFIE instruction.
DS70053C-page 6-6
6
6.2.1.2 Arithmetic Error Trap (Soft Trap, Level 11)
Interrupts
Any of the following events will cause an arithmetic error trap to be generated: Accumulator A Overflow Accumulator B Overflow Catastrophic Accumulator Overflow Divide by Zero Shift Accumulator (SFTAC) operation exceeding +/-16 bits
There are three enable bits in the INTCON1 register that enable the three types of accumulator overflow traps. The OVATE control bit (INTCON1<10>) is used to enable traps for an Accumulator A overflow event. The OVBTE control bit (INTCON1<9>) is used to enable traps for an Accumulator B overflow event. The COVTE control bit (INTCON1<8>) is used to enable traps for a catastrophic overflow of either accumulator. An Accumulator A or Accumulator B overflow event is defined as a carry-out from bit 31. Note that no accumulator overflow can occur if the 31-bit Saturation mode is enabled for the accumulator. A catastrophic accumulator overflow is defined as a carry-out from bit 39 of either accumulator. No catastrophic overflow can occur if accumulator saturation (31-bit or 39-bit) is enabled. Divide-by-zero traps cannot be disabled. The divide-by-zero check is performed during the first iteration of the REPEAT loop that executes the divide instruction. Accumulator shift traps cannot be disabled. The SFTAC instruction can be used to shift the accumulator by a literal value or a value in one of the W registers. If the shift value exceeds +/-16 bits, an arithmetic trap will be generated. The SFTAC instruction will execute, but the results of the shift will not be written to the target accumulator. An arithmetic error trap can be detected in software by polling the MATHERR status bit (INTCON1<4>). To avoid re-entering the Trap Service Routine, the MATHERR status flag must be cleared in software prior to returning from the trap with a RETFIE instruction. Before the MATHERR status bit can be cleared, all conditions that caused the trap to occur must also be cleared. If the trap was due to an accumulator overflow, the OA and OB status bits (SR<15:14>) must be cleared. The OA and OB status bits are read only, so the user software must perform a dummy operation on the overflowed accumulator (such as adding 0) that will cause the hardware to clear the OA or OB status bit.
6.2.2
Hard Traps
Hard traps include exceptions of priority level 13 through level 15, inclusive. The address error (level 13) and oscillator error (level 14) traps fall into this category. Like soft traps, hard traps can also be viewed as non-maskable sources of interrupt. The difference between hard traps and soft traps is that hard traps force the CPU to stop code execution after the instruction causing the trap has completed. Normal program execution flow will not resume until after the trap has been Acknowledged and processed.
6.2.2.1
Trap Priority and Hard Trap Conflicts If a higher priority trap occurs while any lower priority trap is in progress, processing of the lower priority trap will be suspended and the higher priority trap will be Acknowledged and processed. The lower priority trap will remain pending until processing of the higher priority trap completes. Each hard trap that occurs must be Acknowledged before code execution of any type may continue. If a lower priority hard trap occurs while a higher priority trap is pending, Acknowledged, or is being processed, a hard trap conflict will occur. The conflict occurs because the lower priority trap cannot be Acknowledged until processing for the higher priority trap completes. The device is automatically reset in a hard trap conflict condition. The TRAPR status bit (RCON<15> ) is set when the Reset occurs, so that the condition may be detected in software.
DS70053C-page 6-7

6.2.2.2 Oscillator Failure Trap (Hard Trap, Level 14) An oscillator failure trap event will be generated for any of the following reasons: The Fail-Safe Clock Monitor (FSCM) is enabled and has detected a loss of the system clock source. A loss of PLL lock has been detected during normal operation using the PLL. The FSCM is enabled and the PLL fails to achieve lock at a Power-On Reset (POR). An oscillator failure trap event can be detected in software by polling the OSCFAIL status bit (INTCON1<1>), or the CF status bit (OSCCON<3>). To avoid re-entering the Trap Service Routine, the OSCFAIL status flag must be cleared in software prior to returning from the trap with a RETFIE instruction. Refer to Section 7. Oscillator and Section 24. Device Configuration for more information about the FSCM. 6.2.2.3 Address Error Trap (Hard Trap, Level 13) The following paragraphs describe operating scenarios that would cause an address error trap to be generated: 1. A misaligned data word fetch is attempted. This condition occurs when an instruction performs a word access with the LSb of the effective address set to 1. The dsPIC30F CPU requires all word accesses to be aligned to an even address boundary. A bit manipulation instruction using the Indirect Addressing mode with the LSb of the effective address set to 1. A data fetch from unimplemented data address space is attempted. Execution of a BRA #literal instruction or a GOTO #literal instruction, where literal is an unimplemented program memory address. Executing instructions after modifying the PC to point to unimplemented program memory addresses. The PC may be modified by loading a value into the stack and executing a RETURN instruction.
2. 3. 4. 5.
Data space writes will be inhibited whenever an address error trap occurs, so that data is not destroyed. An address error can be detected in software by polling the ADDRERR status bit (INTCON1<3>). To avoid re-entering the Trap Service Routine, the ADDRERR status flag must be cleared in software prior to returning from the trap with a RETFIE instruction. Note: In the MAC class of instructions, the data space is split into X and Y spaces. In these instructions, unimplemented X space includes all of Y space, and unimplemented Y space includes all of X space.
6.2.3
Disable Interrupts Instruction

The DISI (disable interrupts) instruction has the ability to disable interrupts for up to 16384 instruction cycles. This instruction is useful when time critical code segments must be executed. The DISI instruction only disables interrupts with priority levels 1-6. Priority level 7 interrupts and all trap events still have the ability to interrupt the CPU when the DISI instruction is active. The DISI instruction works in conjunction with the DISICNT register. When the DISICNT register is non-zero, priority level 1-6 interrupts are disabled. The DISICNT register is decremented on each subsequent instruction cycle. When the DISICNT register counts down to 0, priority level 1-6 interrupts will be re-enabled. The value specified in the DISI instruction includes all cycles due to PSV accesses, instruction stalls, etc. The DISICNT register is readable and writable. The user can terminate the effect of a previous DISI instruction early by clearing the DISICNT register. The amount of time that interrupts are disabled can also be increased by writing to or adding to DISICNT.
DS70053C-page 6-8
6
Note that if the DISICNT register is zero, interrupts cannot be disabled by simply writing a non-zero value to the register. Interrupts must first be disabled by using the DISI instruction. Once the DISI instruction has executed and DISICNT holds a non-zero value, the interrupt disable time can be extended by modifying the contents of DISICNT. Note: Software modification of the DISICNT register is not recommended.
Interrupts
The DISI status bit (INTCON2<14>) is set whenever interrupts are disabled as a result of the DISI instruction. Note: The DISI instruction can be used to quickly disable all user interrupt sources if no source is assigned to CPU priority level 7.
6.2.4
Interrupt Operation
All interrupt event flags are sampled during each instruction cycle. A pending Interrupt Request (IRQ) is indicated by the flag bit being equal to a 1 in an IFSx register. The IRQ will cause an interrupt to occur if the corresponding bit in the Interrupt Enable (IECx) registers is set. For the rest of the instruction cycle in which the IRQ is sampled, the priorities of all pending interrupt requests are evaluated. No instruction will be aborted when the CPU responds to the IRQ. The instruction that was in progress when the IRQ is sampled will be completed before the ISR is executed. If there is a pending IRQ with a user assigned priority level greater than the current processor priority level, indicated by the IPL<2:0> status bits (SR<7:5>), an interrupt will be presented to the processor. The processor then saves the following information on the software stack: the current PC value the low byte of the Processor Status register (SRL) the IPL3 status bit (CORCON<3>) These three values that are saved on the stack allow the return PC address value, MCU status bits, and the current processor priority level to be automatically saved. After the above information is saved on the stack, the CPU writes the priority level of the pending interrupt into the IPL<2:0> bit locations. This action will disable all interrupts of less than, or equal priority, until the Interrupt Service Routine (ISR) is terminated using the RETFIE instruction. Figure 6-2:
15 Stack Grows Towards Higher Address
Stack Operation for Interrupt Event

0 This stack location used to store the IPL3 status bit (CORCON<3>).
PC<15:0> PC<22:16> SR<7:0> <Free Word>
W15 (before IRQ) W15 (after IRQ)
6.2.4.1
Return from Interrupt The RETFIE (Return from Interrupt) instruction will unstack the PC return address, IPL3 status bit, and SRL register to return the processor to the state and priority level prior to the interrupt sequence.
DS70053C-page 6-9

6.2.4.2 Interrupt Nesting Interrupts, by default, are nestable. Any ISR that is in progress may be interrupted by another source of interrupt with a higher user assigned priority level. Interrupt nesting may be optionally disabled by setting the NSTDIS control bit (INTCON1<15>). When the NSTDIS control bit is set, all interrupts in progress will force the CPU priority to level 7 by setting IPL<2:0> = 111. This action will effectively mask all other sources of interrupt until a RETFIE instruction is executed. When interrupt nesting is disabled, the user assigned interrupt priority levels will have no effect, except to resolve conflicts between simultaneous pending interrupts. The IPL<2:0> bits become read only when interrupt nesting is disabled. This prevents the user software from setting IPL<2:0> to a lower value, which would effectively re-enable interrupt nesting.
6.2.5
Wake-up from Sleep and Idle

Any source of interrupt that is individually enabled, using its corresponding control bit in the IECx registers, can wake-up the processor from Sleep or Idle mode. When the interrupt status flag for a source is set and the interrupt source is enabled via the corresponding bit in the IEC Control registers, a wake-up signal is sent to the dsPIC30F CPU. When the device wakes from Sleep or Idle mode, one of two actions may occur: 1. 2. If the interrupt priority level for that source is greater than the current CPU priority level, then the processor will process the interrupt and branch to the ISR for the interrupt source. If the user assigned interrupt priority level for the source is less than or equal the current CPU priority level, then the processor will simply continue execution, starting with the instruction immediately following the PWRSAV instruction that previously put the CPU in Sleep or Idle mode. Note: User interrupt sources that are assigned to CPU priority level 0 cannot wake the CPU from Sleep or Idle mode, because the interrupt source is effectively disabled. To use an interrupt as a wake-up source, the CPU priority level for the interrupt must be assigned to CPU priority level 1 or greater.
6.2.6
A/D Converter External Conversion Request

The INT0 external interrupt request pin is shared with the A/D converter as an external conversion request signal. The INT0 interrupt source has programmable edge polarity, which is also available to the A/D converter external conversion request feature.
6.2.7
External Interrupt Support

The dsPIC30F supports up to 5 external interrupt pin sources (INT0-INT4). Each external interrupt pin has edge detection circuitry to detect the interrupt event. The INTCON2 register has five control bits (INT0EP-INT4EP) that select the polarity of the edge detection circuitry. Each external interrupt pin may be programmed to interrupt the CPU on a rising edge or falling edge event. See Register 6-4 for further details.
DS70053C-page 6-10
6
6.3 6.3.1 Interrupt Processing Timing
Interrupts
Interrupt Latency for One-Cycle Instructions

Figure 6-3 shows the sequence of events when a peripheral interrupt is asserted during a one-cycle instruction. The interrupt process takes four instruction cycles. Each cycle is numbered in the Figure for reference. The interrupt flag status bit is set during the instruction cycle after the peripheral interrupt occurs. The current instruction completes during this instruction cycle. In the second instruction cycle after the interrupt event, the contents of the PC and SRL registers are saved into a temporary buffer register. The second cycle of the interrupt process is executed as a NOP to maintain consistency with the sequence taken during a two-cycle instruction (see Section 6.3.2 Interrupt Latency for Two-Cycle Instructions). In the third cycle, the PC is loaded with the vector table address for the interrupt source and the starting address of the ISR is fetched. In the fourth cycle, the PC is loaded with the ISR address. The fourth cycle is executed as a NOP while the first instruction in the ISR is fetched.
Figure 6-3:
Interrupt Timing During a One-Cycle Instruction

TCY 1 2 3 4
PC
PC
PC+2
Vector#
2000 (ISR)
2002
2004
2006
INST Executed
INST(PC-2)
INST(PC)
FNOP
Fetch Vector
FNOP
ISR
ISR + 2
ISR + 4
Interrupt Flag Status bit CPU Priority 4 4 4 4 6 6 6 6
Save PC in temporary buffer. Peripheral interrupt event occurs at or before midpoint of this cycle.
PUSH SRL and High 8 bits of PC (from temporary buffer). PUSH Low 16 bits of PC (from temporary buffer).
DS70053C-page 6-11

6.3.2 Interrupt Latency for Two-Cycle Instructions
The interrupt latency during a two-cycle instruction is the same as during a one-cycle instruction. The first and second cycle of the interrupt process allow the two-cycle instruction to complete execution. The timing diagram in Figure 6-5 shows the case when the peripheral interrupt event occurs in the instruction cycle prior to execution of the two-cycle instruction. Figure 6-5 shows the timing when a peripheral interrupt is coincident with the first cycle of a two-cycle instruction. In this case, the interrupt process completes as for a one-cycle instruction (see Section 6.3.1 Interrupt Latency for One-Cycle Instructions). Figure 6-4: Interrupt Timing During a Two-Cycle Instruction
TCY 1 2 3 4
PC
PC
PC+2
Vector#
2000 (ISR)
2002
2004
2006
INST Executed
INST(PC-2)
INST(PC) 1st cycle
INST(PC) 2nd cycle
Fetch Vector
FNOP
ISR
ISR + 2
ISR + 4
Peripheral interrupt event occurs at or before midpoint of this cycle.
Save PC in temporary buffer.
Figure 6-5:
Interrupt Timing, Interrupt Occurs During 1st Cycle of a 2-Cycle Instruction

TCY 1 2 3 4
PC
PC
PC + 2
Vector#
2000 (ISR)
2002
2004
2006
INST Executed
INST(PC) 1st cycle
INST(PC) 2nd cycle
FNOP
Fetch Vector
FNOP
ISR
ISR + 2
ISR + 4
Peripheral interrupt event occurs at or before midpoint of this cycle.
Save PC in temporary buffer.
DS70053C-page 6-12
6
6.3.3 Returning from Interrupt
Interrupts
The Return from Interrupt instruction, RETFIE, exits an interrupt or trap routine. During the first cycle of a RETFIE instruction, the upper bits of the PC and the SRL register are popped from the stack. The lower 16 bits of the stacked PC value are popped from the stack during the second cycle. The third instruction cycle is used to fetch the instruction addressed by the updated program counter. This cycle executes as a NOP. Figure 6-6: Return from Interrupt Timing
TCY
INST Executed PC
ISR last instruction ISR
RETFIE
RETFIE 2nd cycle
FNOP
PC
PC + 2
PC + 4
ISR + 2
PC
PC + 2
PC + 4
PC + 6
CPU Priority
POP SRL and High 8 bits of PC. POP Low 16 bits of PC to RAM Stack.
6.3.4
Special Conditions for Interrupt Latency

The dsPIC30F allows the current instruction to complete when a peripheral interrupt source becomes pending. The interrupt latency is the same for both one and two-cycle instructions. However, there are certain conditions that can increase interrupt latency by one cycle, depending on when the interrupt occurs. The user should avoid these conditions if a fixed latency is critical to the application. These conditions are as follows: A MOV.D instruction is executed that uses PSV to access a value in program memory space. An instruction stall cycle is appended to any two-cycle instruction. An instruction stall cycle is appended to any one-cycle instruction that performs a PSV access. A bit test and skip instruction (BTSC, BTSS) uses PSV to access a value in the program memory space.
DS70053C-page 6-13

6.4 Interrupt Control and Status Registers
The following registers are associated with the interrupt controller: INTCON1, INTCON2 Registers Global interrupt control functions are derived from these two registers. INTCON1 contains the Interrupt Nesting Disable (NSTDIS) bit, as well as the control and status flags for the processor trap sources. The INTCON2 register controls the external interrupt request signal behavior and the use of the alternate vector table. IFSx: Interrupt Flag Status Registers All interrupt request flags are maintained in the IFSx registers, where x denotes the register number. Each source of interrupt has a Status bit, which is set by the respective peripherals or external signal and is cleared via software. IECx: Interrupt Enable Control Registers All Interrupt Enable Control bits are maintained in the IECx registers, where x denotes the register number. These control bits are used to individually enable interrupts from the peripherals or external signals. IPCx: Interrupt Priority Control Registers Each user interrupt source can be assigned to one of eight priority levels. The IPC registers are used to set the interrupt priority level for each source of interrupt. SR: CPU Status Register The SR is not specifically part of the interrupt controller hardware, but it contains the IPL<2:0> Status bits (SR<7:5>) that indicate the current CPU priority level. The user may change the current CPU priority level by writing to the IPL bits. CORCON: Core Control Register The CORCON is not specifically part of the interrupt controller hardware, but it contains the IPL3 Status bit which indicates the current CPU priority level. IPL3 is a Read Only bit so that trap events cannot be masked by the user software. Each register is described in detail on the following pages. Note: The total number and type of interrupt sources will depend on the device variant. Refer to the specific device data sheet for further details.
6.4.1
Assignment of Interrupts to Control Registers

The interrupt sources are assigned to the IFSx, IECx and IPCx registers in the same sequence that they are listed in Table 6-2. For example, the INT0 (External Interrupt 0) is shown as having vector number and a natural order priority of 0. Thus, the INT0IF Status bit is found in IFS0<0>. The INT0 interrupt uses bit 0 of the IEC0 register as its Enable bit and the IPC0<2:0> bits assign the interrupt priority level for the INT0 interrupt.
DS70053C-page 6-14
6
Register 6-1: Upper Byte: R-0 OA bit 15 SR: Status Register (In CPU)
Interrupts
R-0 OB
R/C-0 SA
R/C-0 SB
R-0 OAB
R/C-0 SAB
R-0 DA
R-0 DC bit 8
Lower Byte: R/W-0 bit 7 bit 7-5
R/W-0 IPL<2:0>
R/W-0
R-0 RA
R/W-0 N
R/W-0 OV
R/W-0 Z
R/W-0 C bit 0
IPL<2:0>: CPU Interrupt Priority Level Status bits 111 = CPU interrupt priority level is 7 (15). User interrupts disabled. 110 = CPU interrupt priority level is 6 (14) 101 = CPU interrupt priority level is 5 (13) 100 = CPU interrupt priority level is 4 (12) 011 = CPU interrupt priority level is 3 (11) 010 = CPU interrupt priority level is 2 (10) 001 = CPU interrupt priority level is 1 (9) 000 = CPU interrupt priority level is 0 (8) Note 1: The IPL<2:0> bits are concatenated with the IPL<3> bit (CORCON<3>) to form the CPU interrupt priority level. The value in parentheses indicates the IPL if IPL<3> = 1. 2: The IPL<2:0> status bits are read only when NSTDIS = 1 (INTCON1<15>). Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared C = Bit can be cleared x = Bit is unknown
CORCON: Core Control Register U-0 U-0 R/W-0 US R/W-0 EDT R-0 R-0 DL<1:0> R-0 bit 8 Lower Byte: R/W-0 SATA bit 7
R/W-0 SATB
R/W-1 SATDW
R/W-0 ACCSAT
R/C-0 IPL3
R/W-0 PSV
R/W-0 RND
R/W-0 IF bit 0
bit 3
IPL3: CPU Interrupt Priority Level Status bit 3 1 = CPU interrupt priority level is greater than 7 0 = CPU interrupt priority level is 7 or less Note 1: The IPL3 bit is concatenated with the IPL<2:0> bits (SR<7:5>) to form the CPU interrupt priority level. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared C = Bit can be cleared x = Bit is unknown
DS70053C-page 6-15

Register 6-3: Upper Byte: R/W-0 NSTDIS bit 15 INTCON1: Interrupt Control Register 1 U-0 U-0 U-0 U-0 R/W-0 OVATE R/W-0 OVBTE R/W-0 COVTE bit 8
Lower Byte: U-0 bit 7 bit 15
U-0
U-0
R/W-0 MATHERR
R/W-0 ADDRERR
R/W-0 STKERR
R/W-0 OSCFAIL
U-0 bit 0
NSTDIS: Interrupt Nesting Disable bit 1 = Interrupt nesting is disabled 0 = Interrupt nesting is enabled OVATE: Accumulator A Overflow Trap Enable bit 1 = Trap overflow of Accumulator A 0 = Trap disabled OVBTE: Accumulator B Overflow Trap Enable bit 1 = Trap overflow of Accumulator B 0 = Trap disabled COVTE: Catastrophic Overflow Trap Enable bit 1 = Trap on catastrophic overflow of Accumulator A or B enabled 0 = Trap disabled Unimplemented: Read as 0 MATHERR: Arithmetic Error Status bit 1 = Overflow trap has occurred 0 = Overflow trap has not occurred ADDRERR: Address Error Trap Status bit 1 = Address error trap has occurred 0 = Address error trap has not occurred STKERR: Stack Error Trap Status bit 1 = Stack error trap has occurred 0 = Stack error trap has not occurred OSCFAIL: Oscillator Failure Trap Status bit 1 = Oscillator failure trap has occurred 0 = Oscillator failure trap has not occurred Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 14-11 Unimplemented: Read as 0 bit 10
bit 9
bit 8
bit 7-5 bit 4
bit 3
bit 2
bit 1
bit 0
DS70053C-page 6-16
6
Register 6-4: Upper Byte: R/W-0 ALTIVT bit 15 INTCON2: Interrupt Control Register 2
Interrupts
R-0 DISI
U-0
U-0
U-0
U-0
U-0
U-0 bit 8
U-0
U-0
R/W-0 INT4EP
R/W-0 INT3EP
R/W-0 INT2EP
R/W-0 INT1EP
R/W-0 INT0EP bit 0
ALTIVT: Enable Alternate Interrupt Vector Table bit 1 = Use alternate vector table 0 = Use standard (default) vector table DISI: DISI Instruction Status bit 1 = DISI instruction is active 0 = DISI is not active Unimplemented: Read as 0 INT4EP: External Interrupt #4 Edge Detect Polarity Select bit 1 = Interrupt on negative edge 0 = Interrupt on positive edge INT3EP: External Interrupt #3 Edge Detect Polarity Select bit 1 = Interrupt on negative edge 0 = Interrupt on positive edge INT2EP: External Interrupt #2 Edge Detect Polarity Select bit 1 = Interrupt on negative edge 0 = Interrupt on positive edge INT1EP: External Interrupt #1 Edge Detect Polarity Select bit 1 = Interrupt on negative edge 0 = Interrupt on positive edge INT0EP: External Interrupt #0 Edge Detect Polarity Select bit 1 = Interrupt on negative edge 0 = Interrupt on positive edge Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 14
bit 13-5 bit 4
bit 3
bit 2
bit 1
bit 0
DS70053C-page 6-17

Register 6-5: IFS0: Interrupt Flag Status Register 0 R/W-0 SI2CIF R/W-0 NVMIF R/W-0 ADIF R/W-0 U1TXIF R/W-0 U1RXIF R/W-0 SPI1IF bit 8
Upper Byte: R/W-0 R/W-0 CNIF MI2CIF bit 15
Lower Byte: R/W-0 T3IF bit 7 bit 15
R/W-0 T2IF
R/W-0 OC2IF
R/W-0 IC2IF
R/W-0 T1IF
R/W-0 OC1IF
R/W-0 IC1IF
R/W-0 INT0IF bit 0
CNIF: Input Change Notification Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred MI2CIF: I2C Bus Collision Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred SI2CIF: I2C Transfer Complete Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred NVMIF: Non-Volatile Memory Write Complete Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred ADIF: A/D Conversion Complete Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred U1TXIF: UART1 Transmitter Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred U1RXIF: UART1 Receiver Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred SPI1IF: SPI1 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred T3IF: Timer3 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred T2IF: Timer2 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred OC2IF: Output Compare Channel 2 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred IC2IF: Input Capture Channel 2 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred T1IF: Timer1 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred OC1IF: Output Compare Channel 1 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred
bit 14
bit 13
bit 12
bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
DS70053C-page 6-18
6
Register 6-5: bit 1 IFS0: Interrupt Flag Status Register 0 (Continued)
Interrupts
IC1IF: Input Capture Channel 1 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred INT0IF: External Interrupt 0 Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 0
DS70053C-page 6-19

Register 6-6: Upper Byte: R/W-0 IC6IF bit 15 IFS1: Interrupt Flag Status Register 1 R/W-0 IC5IF R/W-0 IC4IF R/W-0 IC3IF R/W-0 C1IF R/W-0 SPI2IF R/W-0 U2TXIF R/W-0 U2RXIF bit 8
Lower Byte: R/W-0 INT2IF bit 7 bit 15
R/W-0 T5IF
R/W-0 T4IF
R/W-0 OC4IF
R/W-0 OC3IF
R/W-0 IC8IF
R/W-0 IC7IF
R/W-0 INT1IF bit 0
IC6IF: Input Capture Channel 6 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred IC5IF: Input Capture Channel 5 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred IC4IF: Input Capture Channel 4 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred IC3IF: Input Capture Channel 3 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred C1IF: CAN1 (Combined) Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred SPI2IF: SPI2 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred U2TXIF: UART2 Transmitter Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred U2RXIF: UART2 Receiver Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred INT2IF: External Interrupt 2 Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred T5IF: Timer5 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred T4IF: Timer4 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred OC4IF: Output Compare Channel 4 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred OC3IF: Output Compare Channel 3 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred IC8IF: Input Capture Channel 8 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred
bit 14
bit 13
bit 12
bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
DS70053C-page 6-20
6
Interrupts
IC7IF: Input Capture Channel 7 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred INT1IF: External Interrupt 1 Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 0
DS70053C-page 6-21

Register 6-7: Upper Byte: U-0 bit 15 IFS2: Interrupt Flag Status Register 2 U-0 U-0 R/W-0 FLTBIF R/W-0 FLTAIF R/W-0 LVDIF R/W-0 DCIIF R/W-0 QEIIF bit 8
Lower Byte: R/W-0 PWMIF bit 7
R/W-0 C2IF
R/W-0 INT4IF
R/W-0 INT3IF
R/W-0 OC8IF
R/W-0 OC7IF
R/W-0 OC6IF
R/W-0 OC5IF bit 0
bit 15-13 Unimplemented: Read as 0 bit 12 FLTBIF: Fault B Input Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred FLTAIF: Fault A Input Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred LVDIF: Programmable Low Voltage Detect Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred DCIIF: Data Converter Interface Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred QEIIF: Quadrature Encoder Interface Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred PWMIF: Motor Control Pulse Width Modulation Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred C2IF: CAN2 (Combined) Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred INT4IF: External Interrupt 4 Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred INT3IF: External Interrupt 3 Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred OC8IF: Output Compare Channel 8 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred OC7IF: Output Compare Channel 7 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred
bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
DS70053C-page 6-22
6
Interrupts
OC6IF: Output Compare Channel 6 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred OC5IF: Output Compare Channel 5 Interrupt Flag Status bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 0
DS70053C-page 6-23

Register 6-8: IEC0: Interrupt Enable Control Register 0 R/W-0 SI2CIE R/W-0 NVMIE R/W-0 ADIE R/W-0 U1TXIE R/W-0 U1RXIE R/W-0 SPI1IE bit 8
Upper Byte: R/W-0 R/W-0 CNIE MI2CIE bit 15
Lower Byte: R/W-0 T3IE bit 7 bit 15
R/W-0 T2IE
R/W-0 OC2IE
R/W-0 IC2IE
R/W-0 T1IE
R/W-0 OC1IE
R/W-0 IC1IE
R/W-0 INT0IE bit 0
CNIE: Input Change Notification Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled MI2CIE: I2C Bus Collision Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled SI2CIE: I2C Transfer Complete Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled NVMIE: Non-Volatile Memory Write Complete Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled ADIE: A/D Conversion Complete Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled U1TXIE: UART1 Transmitter Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled U1RXIE: UART1 Receiver Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled SPI1IE: SPI1 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled T3IE: Timer3 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled T2IE: Timer2 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled OC2IE: Output Compare Channel 2 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled IC2IE: Input Capture Channel 2 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled T1IE: Timer1 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled OC1IE: Output Compare Channel 1 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled
bit 14
bit 13
bit 12
bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
DS70053C-page 6-24
6
Register 6-8: bit 1 IEC0: Interrupt Enable Control Register 0 (Continued)
Interrupts
IC1IE: Input Capture Channel 1 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled INT0IE: External Interrupt 0 Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 0
DS70053C-page 6-25

Register 6-9: Upper Byte: R/W-0 IC6IE bit 15 IEC1: Interrupt Enable Control Register 1 R/W-0 IC5IE R/W-0 IC4IE R/W-0 IC3IE R/W-0 C1IE R/W-0 SPI2IE R/W-0 U2TXIE R/W-0 U2RXIE bit 8
Lower Byte: R/W-0 INT2IE bit 7 bit 15
R/W-0 T5IE
R/W-0 T4IE
R/W-0 OC4IE
R/W-0 OC3IE
R/W-0 IC8IE
R/W-0 IC7IE
R/W-0 INT1IE bit 0
IC6IE: Input Capture Channel 6 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled IC5IE: Input Capture Channel 5 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled IC4IE: Input Capture Channel 4 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled IC3IE: Input Capture Channel 3 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled C1IE: CAN1 (Combined) Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled SPI2IE: SPI2 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled U2TXIE: UART2 Transmitter Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled U2RXIE: UART2 Receiver Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled INT2IE: External Interrupt 2 Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled T5IE: Timer5 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled T4IE: Timer4 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled OC4IE: Output Compare Channel 4 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled OC3IE: Output Compare Channel 3 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled IC8IE: Input Capture Channel 8 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled
bit 14
bit 13
bit 12
bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
DS70053C-page 6-26
6
Interrupts
IC7IE: Input Capture Channel 7 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled INT1IE: External Interrupt 1 Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 0
DS70053C-page 6-27

Register 6-10: Upper Byte: U-0 bit 15 IEC2: Interrupt Enable Control Register 2 U-0 U-0 R/W-0 FLTBIE R/W-0 FLTAIE R/W-0 LVDIE R/W-0 DCIIE R/W-0 QEIIE bit 8
Lower Byte: R/W-0 PWMIE bit 7
R/W-0 C2IE
R/W-0 INT4IE
R/W-0 INT3IE
R/W-0 OC8IE
R/W-0 OC7IE
R/W-0 OC6IE
R/W-0 OC5IE bit 0
bit 15-13 Unimplemented: Read as 0 bit 12 FLTBIE: Fault B Input Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled FLTAIE: Fault A Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled LVDIE: Programmable Low Voltage Detect Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled DCIIE: Data Converter Interface Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled QEIIE: Quadrature Encoder Interface Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled PWMIE: Motor Control Pulse Width Modulation Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled C2IE: CAN2 (Combined) Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled INT4IE: External Interrupt 4 Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled INT3IE: External Interrupt 3 Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled OC8IE: Output Compare Channel 8 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled OC7IE: Output Compare Channel 7 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled
bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
DS70053C-page 6-28
6
Interrupts
OC6IE: Output Compare Channel 6 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled OC5IE: Output Compare Channel 5 Interrupt Enable bit 1 = Interrupt request enabled 0 = Interrupt request not enabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 0
DS70053C-page 6-29

Register 6-11: Upper Byte: U-0 bit 15 IPC0: Interrupt Priority Control Register 0 R/W-1 R/W-0 T1IP<2:0> R/W-0 U-0 R/W-1 R/W-0 R/W-0 OC1IP<2:0> bit 8
R/W-1
R/W-0 IC1IP<2:0>
R/W-0
U-0
R/W-1
R/W-0 R/W-0 INT0IP<2:0> bit 0
Unimplemented: Read as 0
bit 14-12 T1IP<2:0>: Timer1 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 OC1IP<2:0>: Output Compare Channel 1 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 IC1IP<2:0>: Input Capture Channel 1 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 INT0IP<2:0>: External Interrupt 0 Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-30
6
Register 6-12: Upper Byte: U-0 bit 15 IPC1: Interrupt Priority Control Register 1
Interrupts
R/W-1
R/W-0 T3IP<2:0>
R/W-0
U-0
R/W-1
R/W-0 T2IP<2:0>
R/W-0 bit 8
R/W-1
R/W-0 OC2IP<2:0>
R/W-0
U-0
R/W-1
R/W-0 IC2IP<2:0>
R/W-0 bit 0
bit 14-12 T3IP<2:0>: Timer3 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 T2IP<2:0>: Timer2 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 OC2IP<2:0>: Output Compare Channel 2 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 IC2IP<2:0>: Input Capture Channel 2 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-31

Register 6-13: Upper Byte: U-0 bit 15 IPC2: Interrupt Priority Control Register 2 R/W-1 R/W-0 ADIP<2:0> R/W-0 U-0 R/W-1 R/W-0 R/W-0 U1TXIP<2:0> bit 8
R/W-1
R/W-0 U1RXIP<2:0>
R/W-0
U-0
R/W-1
R/W-0 R/W-0 SPI1IP<2:0> bit 0
bit 14-12 ADIP<2:0>: A/D Conversion Complete Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 U1TXIP<0>: UART1 Transmitter Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 U1RXIP<2:0>: UART1 Receiver Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 SPI1IP<2:0>: SPI1 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-32
6
Interrupts
R/W-1
R/W-0 CNIP<2:0>
R/W-0
U-0
R/W-1
R/W-0 R/W-0 MI2CIP<2:0> bit 8
R/W-1
R/W-0 SI2CIP<2:0>
R/W-0
U-0
R/W-1
R/W-0 R/W-0 NVMIP<2:0> bit 0
bit 14-12 CNIP<2:0>: Input Change Notification Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 MI2CIP<2:0>: I2C Bus Collision Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 SI2CIP<2:0>: I2C Transfer Complete Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 NVMIP<2:0>: Non-Volatile Memory Write Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-33

Register 6-15: Upper Byte: U-0 bit 15 IPC4: Interrupt Priority Control Register 4 R/W-1 R/W-0 OC3IP<2:0> R/W-0 U-0 R/W-1 R/W-0 IC8IP<2:0> R/W-0 bit 8 Lower Byte: U-0 bit 7 bit 15
R/W-1
R/W-0 IC7IP<2:0>
R/W-0
U-0
R/W-1
bit 14-12 OC3IP<2:0>: Output Compare Channel 3 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 IC8IP<2:0>: Input Capture Channel 8 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 IC7IP<2:0>: Input Capture Channel 7 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 INT1IP<2:0>: External Interrupt 1 Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-34
6
Interrupts
R/W-1
R/W-0 INT2IP<2:0>
R/W-0
U-0
R/W-1
R/W-0 T5IP<2:0>
R/W-0 bit 8
R/W-1
R/W-0 T4IP<2:0>
R/W-0
U-0
R/W-1
R/W-0 R/W-0 OC4IP<2:0> bit 0
bit 14-12 INT2IP<2:0>: External Interrupt 2 Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 T5IP<2:0>: Timer5 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 T4IP<2:0>: Timer4 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 OC4IP<2:0>: Output Compare Channel 4 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-35

Register 6-17: Upper Byte: U-0 bit 15 IPC6: Interrupt Priority Control Register 6 R/W-1 R/W-0 C1IP<2:0> R/W-0 U-0 R/W-1 R/W-0 R/W-0 SPI2IP<2:0> bit 8
R/W-1
R/W-0 U2TXIP<2:0>
R/W-0
U-0
R/W-1
R/W-0 R/W-0 U2RXIP<2:0> bit 0
bit 14-12 C1IP<2:0>: CAN1 (Combined) Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 SPI2IP<2:0>: SPI2 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 U2TXIP<2:0>: UART2 Transmitter Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 U2RXIP<2:0>: UART2 Receiver Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-36
6
Interrupts
R/W-1
R/W-0 IC6IP<2:0>
R/W-0
U-0
R/W-1
R/W-0 IC5IP<2:0>
R/W-0 bit 8
R/W-1
R/W-0 IC4IP<2:0>
R/W-0
U-0
R/W-1
R/W-0 IC3IP<2:0>
R/W-0 bit 0
bit 14-12 IC6IP<2:0>: Input Capture Channel 6 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 IC5IP<2:0>: Input Capture Channel 5 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 IC4IP<2:0>: Input Capture Channel 4 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 IC3IP<2:0>: Input Capture Channel 3 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-37

Register 6-19: Upper Byte: U-0 bit 15 IPC8: Interrupt Priority Control Register 8 R/W-1 R/W-0 OC8IP<2:0> R/W-0 U-0 R/W-1 R/W-0 R/W-0 OC7IP<2:0> bit 8
R/W-1
R/W-0 OC6IP<2:0>
R/W-0
U-0
R/W-1
R/W-0 R/W-0 OC5IP<2:0> bit 0
bit 14-12 OC8IP<2:0>: Output Compare Channel 8 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 OC7IP<2:0>: Output Compare Channel 7 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 OC6IP<2:0>: Output Compare Channel 6 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 OC5IP<2:0>: Output Compare Channel 5 Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-38
6
Interrupts
R/W-1
R/W-0 PWMIP<2:0>
R/W-0
U-0
R/W-1
R/W-0 C2IP<2:0>
R/W-0 bit 8
R/W-1
R/W-0 INT4IP<2:0>
R/W-0
U-0
R/W-1
bit 14-12 PWMIP<2:0>: Motor Control Pulse Width Modulation Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 C2IP<2:0>: CAN2 (Combined) Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 INT4IP<2:0>: External Interrupt 4 Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 INT3IP<2:0>: External Interrupt 3 Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-39

Register 6-21: Upper Byte: U-0 bit 15 IPC10: Interrupt Priority Control Register 10 R/W-1 R/W-0 FLTAIP<2:0> R/W-0 U-0 R/W-1 R/W-0 R/W-0 LVDIP<2:0> bit 8
R/W-1
R/W-0 DCIIP<2:0>
R/W-0
U-0
R/W-1
R/W-0 QEIIP<2:0>
R/W-0 bit 0
bit 14-12 FLTAIP<2:0>: Fault A Input Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled bit 11 bit 10-8 Unimplemented: Read as 0 LVDIP<2:0>: Programmable Low Voltage Detect Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 DCIIP<2:0>: Data Converter Interface Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Unimplemented: Read as 0 QEIIP<2:0>: Quadrature Encoder Interface Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7 bit 6-4
bit 3 bit 2-0
DS70053C-page 6-40
6
Interrupts
U-0
U-0
U-0
U-0
U-0
U-0
U-0 bit 8
Lower Byte: U-0 bit 7 bit 15-3 bit 2-0
U-0
U-0
U-0
U-0
R/W-1
R/W-0 R/W-0 FLTBIP<2:0> bit 0
Unimplemented: Read as 0 FLTBIP<2:0>: Fault B Input Interrupt Priority bits 111 = Interrupt is priority 7 (highest priority interrupt) 001 = Interrupt is priority 1 000 = Interrupt source is disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70053C-page 6-41

6.5 6.5.1 Interrupt Setup Procedures Initialization
The following steps describe how to configure a source of interrupt: 1. 2. Set the NSTDIS Control bit (INTCON1<15>) if nested interrupts are not desired. Select the user assigned priority level for the interrupt source by writing the control bits in the appropriate IPCx Control register. The priority level will depend on the specific application and type of interrupt source. If multiple priority levels are not desired, the IPCx register control bits for all enabled interrupt sources may be programmed to the same non-zero value. Note: 3. 4. At a device Reset, the IPC registers are initialized, such that all user interrupt sources are assigned to priority level 4.
Clear the interrupt flag status bit associated with the peripheral in the associated IFSx Status register. Enable the interrupt source by setting the interrupt enable control bit associated with the source in the appropriate IECx Control register.
6.5.2
Interrupt Service Routine

The method that is used to declare an ISR and initialize the IVT with the correct vector address will depend on the programming language (i.e., C or assembler) and the language development tool suite that is used to develop the application. In general, the user must clear the interrupt flag in the appropriate IFSx register for the source of interrupt that the ISR handles. Otherwise, the ISR will be re-entered immediately after exiting the routine. If the ISR is coded in assembly language, it must be terminated using a RETFIE instruction to unstack the saved PC value, SRL value, and old CPU priority level.
6.5.3
Trap Service Routine

A Trap Service Routine (TSR) is coded like an ISR, except that the appropriate trap status flag in the INTCON1 register must be cleared to avoid re-entry into the TSR.
6.5.4
Interrupt Disable
All user interrupts can be disabled using the following procedure: 1. 2. Push the current SR value onto the software stack using the PUSH instruction. Force the CPU to priority level 7 by inclusive ORing the value 0xE0 with SRL.
To enable user interrupts, the POP instruction may be used to restore the previous SR value. Note that only user interrupts with a priority level of 7 or less can be disabled. Trap sources (level 8-level 15) cannot be disabled. The DISI instruction provides a convenient way to disable interrupts of priority levels 1-6, for a fixed period of time. Level 7 interrupt sources are not disabled by the DISI instruction.
DS70053C-page 6-42
Table 6-3:
Bit 13 SI2CIF IC4IF SI2CIE IC4IE T1IP<2:0> LVDIP<2:0> C2IP<2:0> INT41IP<2:0> DCIIP<2:0> OC7IP<2:0> OC6IP<2:0> IC5IP<2:0> IC4IP<2:0> SPI2IP<2:0> U2TXIP<2:0> T5IP<2:0> T4IP<2:0> IC8IP<2:0> IC7IP<2:0> MI2CIP<2:0> SI2CIP<2:0> U1TXIP<2:0> U1RXIP<2:0> T2IP<2:0> OC2IP<2:0> OC1IP<2:0> IC1IP<2:0> FLTBIE FLTAIE LVDIE DCIIE QEIIE PWMIE C2IE INT4IE INT3IE OC8IE OC7IE IC3IE C1IE SPI2IE U2TXIE U2RXIE INT2IE T5IE T4IE OC4IE OC3IE IC8IE IC7IE OC6IE INT0IP<2:0> IC2IP<2:0> SPI1IP<2:0> NVMIP<2:0> INT1IP<2:0> OC4IP<2:0> U2RXIP<2:0> IC3IP<2:0> OC5IP<2:0> INT3IP<2:0> QEIIP<2:0> FLTBIP<2:0> NVMIE ADIE U1TXIE U1RXIE SPI1IE T3IE T2IE OC2IE IC2IE T1IE OC1IE IC1IE FLTBIF FLTAIF LVDIF DCIIF QEIIF PWMIF C2IF INT4IF INT3IF OC8IF OC7IF OC6IF OC5IF INT0IE INT1IE OC5IE IC3IF C1IF SPI2IF U2TXIF U2RXIF INT2IF T5IF T4IF OC4IF OC3IF IC8IF IC7IF INT1IF NVMIF ADIF U1TXIF U1RXIF SPI1IF T3IF T2IF OC2IF IC2IF T1IF OC1IF IC1IF INT0 INT4EP INT3EP INT2EP INT1EP INT0EP OVATE OVBTE COVTE MATHERR ADDRERR STKERR OSCFAIL Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Reset State
Special Function Registers Associated with Interrupt Controller
SFR Name
ADR
Bit 15
Bit 14
INTCON1
0080 NSTDIS
0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0000 0000 0000 0100
INTCON2
0082 ALTIVT
IFT0IF
0084
CNIF
MI2CIF
IFS1
0086
IC6IF
IC5IF
IFS2
0088
Interrupts
IEC0
008C
CNIE
MI2CIE
IEC1
008E
IC6IE
IC5IE
IEC2
0090
IPC0
0094
IPC1
0096
T31P<2:0>
IPC2
0098
ADIP<2:0>
IPC3
009A
CNIP<2:0>
IPC4
009C
OC3IP<2:0>
IPC5
009E
INT2IP<2:0>
IPC6
00A0
C1IP<2:0>
IPC7
00A2
IC6IP<2:0>
IPC8
00A4
OC8IP<2:0>
IPC9
00A6
PWMIP<2:0>
IPC10
00A8
FLTAIP<2:0>
IPC11
00AA
Note:
All interrupt sources and their associated control bits may not be available on a particular device. Refer to the device data sheet for details.
DS70053C-page 6-43

6.6 Design Tips
Question 1: What happens when two sources of interrupt become pending at the same time and have the same user assigned priority level?
Answer: The interrupt source with the highest natural order priority will take precedence. The natural order priority is determined by the Interrupt Vector Table (IVT) address for that source. Interrupt sources with a smaller IVT address have a higher natural order priority.
Question 2:
Can the DISI instruction be used to disable all sources of interrupt and traps?
Answer: The DISI instruction does not disable traps or priority level 7 interrupt sources. However, the DISI instruction can be used as a convenient way to disable all interrupt sources if no priority level 7 interrupt sources are enabled in the users application.
DS70053C-page 6-44
6
This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Interrupts module are: Title No related application notes at this time. Application Note #
Interrupts
Note:
DS70053C-page 6-45

Revision B
This revision incorporates additional technical content for the dsPIC30F Interrupts module.
Revision C
DS70053C-page 6-46
Section 7. Oscillator
HIGHLIGHTS
This section of the manual contains the following topics: 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11 7.12 7.13 7.14 7.15 7.16 7.17 7.18 7.19 7.20 Introduction .................................................................................................................... 7-2 Device Clocking and MIPS ............................................................................................ 7-5 Oscillator Configuration.................................................................................................. 7-6 Oscillator Control Registers OSCCON and OSCTUN .............................................. 7-13 Primary Oscillator......................................................................................................... 7-20 Crystal Oscillators/Ceramic Resonators ...................................................................... 7-22 Determining Best Values for Crystals, Clock Mode, C1, C2 and Rs ............................ 7-24 External Clock Input..................................................................................................... 7-25 External RC Oscillator.................................................................................................. 7-26 Phase Locked Loop (PLL) ........................................................................................... 7-30 Low-Power 32 kHz Crystal Oscillator........................................................................... 7-31 Oscillator Start-up Timer (OST).................................................................................... 7-31 Internal Fast RC Oscillator (FRC) ................................................................................ 7-31 Internal Low-Power RC (LPRC) Oscillator................................................................... 7-32 Fail-Safe Clock Monitor (FSCM) .................................................................................. 7-32 Programmable Oscillator Postscaler............................................................................ 7-33 Clock Switching Operation........................................................................................... 7-34 Design Tips .................................................................................................................. 7-38 Related Application Notes............................................................................................ 7-39 Revision History ........................................................................................................... 7-40
7
Oscillator
DS70054D-page 7-1

7.1 Introduction
This section describes the operation of the oscillator system for dsPIC30F devices in the General Purpose, Sensor and Motor Control families. The oscillator system has the following modules and features: Various external and internal oscillator options as clock sources An on-chip PLL to boost internal operating frequency Clock switching between various clock sources Programmable clock postscaler for system power savings A Fail-Safe Clock Monitor (FSCM) that detects clock failure and takes fail-safe measures Device clocking controlled by Special Function Registers as well as nonvolatile Configuration bits
A simplified diagram of the oscillator system is shown in Figure 7-1.
7.1.1
Oscillator System Features Summary

dsPIC30F devices in the General Purpose, Sensor and Motor Control families feature one of three versions of the oscillator system VERSION 1, VERSION 2 and VERSION 3. The features of the three versions of the oscillator system are summarized in Table 7-1. . Note: Refer to the device data sheet to determine the version of the oscillator system featured on the dsPIC30F device you are using.
Table 7-1: Oscillator System VERSION 1
Device-Specific Oscillator System Feature Summary dsPIC30F Device 30F6010, 30F6011, 30F6012, 30F6013, 30F6014 Feature Summary Oscillator Sources: Primary oscillator with Multiple Clock modes XT, EC, HS Secondary oscillator (Low-Power 32 kHz Crystal oscillator) FRC oscillator: Fast Internal RC (7.37 MHz) LPRC oscillator: Low-Power Internal RC (512 kHz)
PLL Clock Multiplier: 4 MHz-10 MHz input frequency range 4x Multiplier mode (FOUT = 16 MHz-40 MHz) 8x Multiplier mode (FOUT = 32 MHz-80 MHz) 16x Multiplier mode (FOUT = 64 MHz-120 MHz) PLL VCO lock indication plus out of lock trap option PLL input provided by the following sources: - XT or EC Primary oscillator
Clock Scaling Options: Generic postscaler for device clock (divide by 4, 16, 64) Fail-Safe Clock Monitor (FSCM): Detects clock failure and switches over to internal FRC oscillator VERSION 2 30F2010, 30F4011, 30F4012, 30F5011, 30F5013 Oscillator System VERSION 2 adds the following capabilities to VERSION 1: Internal FRC oscillator may also be provided as an input to the PLL to allow fast execution while eliminating the need for an external clock source (This feature is applicable to all devices other than the 30F2010) User tuning capability added for the Internal FRC oscillator
DS70054D-page 7-2
Table 7-1: Oscillator System VERSION 3 Device-Specific Oscillator System Feature Summary dsPIC30F Device 30F2011, 30F2012, 30F3010, 30F3011, 30F3012, 30F3013, 30F3014, 30F4013, 30F5015, 30F5016, 30F6010A, 30F6011A, 30F6012A, 30F6013A, 30F6014A, 30F6015 Feature Summary Oscillator System VERSION 3 adds the following capabilities to VERSION 2: HS oscillator may also be provided as an input to the PLL to allow greater choices of crystal frequency
7
Oscillator
DS70054D-page 7-3

Figure 7-1: Oscillator System Block Diagram
Oscillator Configuration bits PWRSAV Instruction Wake-up Request OSC1 OSC2 Primary Oscillator PLL x4, x8, x16
PLL
Primary Osc
(2)
Primary Oscillator Stability Detector
POR
Oscillator Start-up Timer Secondary Osc
Clock Switching and Control Block
Programmable Clock Divider
FOSC(1)
SOSCO SOSCI
Secondary Oscillator 32 kHz
Secondary Oscillator Stability Detector
Internal Fast RC Oscillator (FRC)
FRC
Internal Low Power RC Oscillator (LPRC)
LPRC
Fail-Safe Clock Monitor (FSCM)
Oscillator Trap
to Timer1 Note 1: The system clock output, FOSC, is divided by 4 to get the instruction cycle clock. 2: Devices that feature VERSION 2 or VERSION 3 of the Oscillator System allow the internal FRC oscillator to be connected to the PLL.
DS70054D-page 7-4
7.2 Device Clocking and MIPS
Referring to Figure 7-1, the system clock source can be provided by one of four sources. These sources are the Primary oscillator, Secondary oscillator, Internal Fast RC (FRC) oscillator or the Low-Power RC (LPRC) oscillator. The Primary oscillator source has the option of using the internal PLL. The frequency of the selected clock source can optionally be reduced by the programmable postscaler (clock divider). The output from the programmable postscaler becomes the system clock source, FOSC. The system clock source is divided by four to produce the internal instruction cycle clock, FCY. In this document, the instruction cycle clock is also denoted by FOSC/4. The timing diagram in Figure 7-2 shows the relationship between the system clock source and instruction execution. The internal instruction cycle clock, FCY, can be provided on the OSC2 I/O pin for some Operating modes of the Primary oscillator (see Section 7.3 Oscillator Configuration). Figure 7-2: Clock/Instruction Cycle Timing
TCY FOSC FCY PC PC Fetch INST (PC) Execute INST (PC - 2) PC + 2 PC + 4
7
Oscillator
Fetch INST (PC + 2) Execute INST (PC)
Fetch INST (PC + 4) Execute INST (PC + 2)
Equation 7-1: MIPS and Source Oscillator Frequency Relationship FCY = FOSC 4 =
( SOURCE OSCILLATOR FREQUENCY * PLL MULTIPLIER ) PROGRAMMABLE POSTSCALER * 4
DS70054D-page 7-5

7.3 Oscillator Configuration
The oscillator source (and Operating mode) that is used at a device Power-on Reset event is selected using nonvolatile Configuration bits. The oscillator Configuration bits are located in the FOSC Configuration register. The FOS bits in the FOSC nonvolatile Configuration register select the oscillator source that is used at a Power-on Reset. The Primary oscillator is the default (unprogrammed) selection. The Secondary oscillator, or one of the internal oscillators, may be chosen by programming these bit locations. The FPR bits in the FOSC nonvolatile Configuration register select the Operating mode of the Primary oscillator. dsPIC30F devices in the General Purpose, Sensor and Motor Control families may feature one of three versions of the Oscillator system. The definition of the FOSC nonvolatile Configuration register varies between these versions, as described in the sub-sections below.
7.3.1
Oscillator System VERSION 1 Configuration

For devices that feature the Oscillator system VERSION 1, the FOSC nonvolatile Configuration register is shown in Register 7-1. The operating modes for the FPR bits may be selected as shown in Table 7-2.
7.3.2

7.3.3

7.3.4
Clock Switching Mode Configuration Bits

The FCKSM<1:0> Configuration bits (FOSC<15:14>) are used to enable/disable device clock switching and the Fail-Safe Clock Monitor (FSCM). When these bits are unprogrammed (default), clock switching and the FSCM are disabled. These bits carry the same definition and functionality across all versions of the Oscillator system.
DS70054D-page 7-6
Register 7-1: Upper Byte: U bit 23 Middle Byte: R/P bit 15 Lower Byte: R/P U U U U R/P R/P bit 8 FCKSM<1:0> FOS<1:0> U U U U U U U bit 16 FOSC: Oscillator Configuration Register for Oscillator System VERSION 1
7
Oscillator
R/P bit 0
U bit 7 bit 23-16 bit 15-14
R/P
R/P
R/P
FPR<3:0>
Unimplemented: Read as 0 FCKSM<1:0>: Clock Switching Mode Selection Fuses bits 1x = Clock switching is disabled, Fail-Safe Clock Monitor is disabled 01 = Clock switching is enabled, Fail-Safe Clock Monitor is disabled 00 = Clock switching is enabled, Fail-Safe Clock Monitor is enabled Unimplemented: Read as 0 FOS<1:0>: Oscillator Source Selection on POR bits 11 = Primary Oscillator (Primary Oscillator mode selected by FPR<3:0>) 10 = Internal Low-Power RC Oscillator 01 = Internal Fast RC Oscillator 00 = Low-Power 32 kHz Oscillator (Timer1 oscillator) Unimplemented: Read as 0 FPR<3:0>: Oscillator Selection within Primary Group bits, See Table 7-2 Legend: R = Readable bit P = Programmable bit U = Unimplemented bit
bit 13-10 bit 9-8
bit 7-4 bit 3-0
DS70054D-page 7-7

Table 7-2: Oscillator System VERSION 1: Configuration Bit Values for Clock Selection Oscillator Source Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Secondary Internal Internal FOS<1:0> 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 1 1 1 1 0 0 0 0 0 0 FPR<3:0> 1 1 1 1 0 0 0 0 1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 x x OSC2 Pin Function I/O (Note 4) I/O I/O I/O FOSC/4 n/a FOSC/4 I/O (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) (Notes 1, 2) (Notes 1, 2) (Notes 1, 2)
Oscillator Mode EC w/ PLL 16x EC w/ PLL 8x EC w/ PLL 4x ECIO EC Reserved ERC ERCIO XT w/ PLL 16x XT w/ PLL 8x XT w/ PLL 4x XT HS XTL LP FRC LPRC
Note 1: OSC2 pin function is determined by the Primary Oscillator mode selection (FPR<3:0> Configuration bits). 2: Note that OSC1 pin cannot be used as an I/O pin, even if the Secondary oscillator or an internal clock source is selected at all times. 3: In these Oscillator modes, a crystal is connected between the OSC1 and OSC2 pins. 4: This is the default Oscillator mode for an unprogrammed (erased) device. An unprogrammed Configuration bit has a value of 1. 5: XTL XTL Crystal Oscillator mode (200 kHz-4 MHz crystal) 6: XT XT Crystal Oscillator mode (4 MHz-10 MHz crystal) 7: HS HS Crystal Oscillator mode (10 MHz-25 MHz crystal)
DS70054D-page 7-8
Register 7-2: Upper Byte: U bit 23 Middle Byte: R/P bit 15 Lower Byte: R/P U U U U R/P R/P bit 8 FCKSM<1:0> FOS<1:0> U U U U U U U bit 16 FOSC: Oscillator Configuration Register for Oscillator System VERSION 2
7
Oscillator
R/P bit 0
U bit 7 bit 23-16 bit 15-14
R/P
R/P
R/P
FPR<3:0>
Unimplemented: Read as 0 FCKSM<1:0>: Clock Switching Mode Selection Fuses bits 1x = Clock switching is disabled, Fail-Safe Clock Monitor is disabled 01 = Clock switching is enabled, Fail-Safe Clock Monitor is disabled 00 = Clock switching is enabled, Fail-Safe Clock Monitor is enabled Unimplemented: Read as 0 FOS<1:0>: Oscillator Source Selection on POR bits 11 = Primary Oscillator (Primary Oscillator mode selected by FPR<3:0>) 10 = Internal Low-Power RC Oscillator 01 = Internal Fast RC Oscillator 00 = Low-Power 32 kHz Oscillator (Timer1 oscillator) Unimplemented: Read as 0 FPR<3:0>: Oscillator Mode Selection within Primary Group bits, See Table 7-3 Legend: R = Readable bit P = Programmable bit U = Unimplemented bit
bit 13-10 bit 9-8
bit 7-4 bit 3-0
DS70054D-page 7-9

Table 7-3: Oscillator System VERSION 2: Configuration Bit Values for Clock Selection: Oscillator Source Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Primary Secondary Internal FRC Internal LPRC FOS<1:0> 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 1 1 1 0 0 0 0 0 0 0 1 0 FPR<3:0> 0 1 1 1 1 0 0 1 1 1 1 0 0 0 0 0 1 0 0 1 1 0 0 0 0 1 1 0 1 0 1 1 1 0 1 0 1 1 0 0 1 0 1 0 0 1 0 1 OSC2 Pin Function CLKO I/O I/O I/O I/O (Note 4) CLKO I/O (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) I/O I/O I/O (Notes 1, 2) (Notes 1, 2) (Notes 1, 2)
Oscillator Mode EC ECIO EC w/PLL 4x EC w/PLL 8x EC w/PLL 16x ERC ERCIO XT XT w/PLL 4x XT w/PLL 8x XT w/PLL 16x XTL HS FRC w/PLL 4x FRC w/PLL 8x FRC w/PLL 16x LP FRC LPRC
Note 1: OSC2 pin function is determined by the Primary Oscillator mode selection (FPR<3:0>). 2: Note that OSC1 pin cannot be used as an I/O pin, even if the secondary oscillator or an internal clock source is selected at all times. 3: In these Oscillator modes, a crystal is connected between the OSC1 and OSC2 pins. 4: This is the default Oscillator mode for an unprogrammed (erased) device. An unprogrammed Configuration bit has a value of 1. 5: XTL XTL Crystal Oscillator mode (200 kHz-4 MHz crystal) 6: XT XT Crystal Oscillator mode (4 MHz-10 MHz crystal) 7: HS HS Crystal Oscillator mode (10 MHz-25 MHz crystal)
DS70054D-page 7-10
Register 7-3: Upper Byte: U bit 23 Middle Byte: R/P bit 15 Lower Byte: R/P U U U R/P R/P FOS<2:0> bit 8 R/P FCKSM<1:0> U U U U U U U bit 16 FOSC: Oscillator Configuration Register for Oscillator System VERSION 3
7
Oscillator
R/P bit 0
U bit 7
R/P
R/P
R/P FPR<4:0>
R/P
bit 23-16 Unimplemented: Read as 0 bit 15-14 FCKSM<1:0>: Clock Switching and Monitor Selection Configuration bits 1x = Clock switching is disabled, Fail-Safe Clock Monitor is disabled 01 = Clock switching is enabled, Fail-Safe Clock Monitor is disabled 00 = Clock switching is enabled, Fail-Safe Clock Monitor is enabled bit 13-10 Unimplemented: Read as 0 bit 9-8 FOS<2:0>: Oscillator Group Selection on POR bit 111 = PLL Oscillator; PLL source selected by FPR<4:0> bits 011 = EXT: External Oscillator; OSC1/OSC2 pins; External Oscillator configuration selected by FPR<4:0> bits 010 = LPRC: Internal Low-Power RC 001 = FRC: Internal Fast RC 000 = LPOSC: Low-Power Crystal Oscillator; SOSCI/SOSCO pins Unimplemented: Read as 0 FPR<4:0>: Oscillator Selection within Primary Group bits, See Table 7-4. Legend: R = Readable bit W = Writable bit U = Unimplemented bit, read as 0
bit 7-4 bit 3-0
DS70054D-page 7-11

Table 7-4: Oscillator System VERSION 3: Configuration Bit Values for Clock Selection Oscillator Source PLL PLL PLL PLL PLL PLL PLL PLL PLL PLL PLL PLL PLL PLL PLL External External External External External External External Secondary Internal FRC Internal LPRC 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 FOS<2:0> 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 0 x x x 1 1 1 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 1 0 x x x FPR<4:0> 1 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 1 0 0 0 0 0 x x x 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 0 1 1 0 0 0 x x x 1 0 1 1 0 1 1 0 1 1 0 1 1 0 1 0 0 0 1 1 0 0 x x x OSC2 Pin Function I/O I/O I/O I/O I/O I/O (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) (Note 3) I/O (Note 3) (Note 3) CLKOUT CLKOUT I/O (Note 3) (Note 1, 2) (Note 1, 2) (Note 1, 2)
Oscillator Mode ECIO w/ PLL 4x ECIO w/ PLL 8x ECIO w/ PLL 16x FRC w/ PLL 4x FRC w/ PLL 8x FRC w/ PLL 16x XT w/ PLL 4x XT w/ PLL 8x XT w/ PLL 16x HS2 w/ PLL 4x HS2 w/ PLL 8x HS2 w/ PLL 16x HS3 w/ PLL 4x HS3 w/ PLL 8x HS3 w/ PLL 16x ECIO XT HS EC ERC ERCIO XTL LP FRC LPRC
Note 1: OSC2 pin function is determined by (FPR<4:0>). 2: OSC1 pin cannot be used as an I/O pin even if the secondary oscillator or an internal clock source is selected at all times. 3: In these Oscillator modes, a crystal is connected between the OSC1 and OSC2 pins. 4: XTL XTL Crystal Oscillator mode (200 kHz-4 MHz crystal) 5: XT XT Crystal Oscillator mode (4 MHz-10 MHz crystal) 6: HS HS Crystal Oscillator mode (10 MHz-25 MHz crystal)
DS70054D-page 7-12
7.4 Oscillator Control Registers OSCCON and OSCTUN
Run-time control and status of the Oscillator system is provided to the user via Special Function Registers. Table 7-5 summarizes the run-time control features provided in VERSION 1, VERSION 2 and VERSION 3 of the Oscillator System. Refer to the device data sheet to determine the version of the oscillator system featured on the dsPIC30F device you are using. Table 7-5: Oscillator Control SFRs Oscillator Control SFRs Feature Summary Control via OSCCON SFR. Refer to Register 7-4. Control via OSCCON SFR. User may tune the FRC oscillator via TUN<3:0> bits in OSCCON. Refer to Register 7-5. Control via OSCCON and OSCTUN SFRs. User may tune the FRC oscillator via TUN<3:0> bits in OSCTUN. Refer to Register 7-6 and Register 7-7.
Oscillator System VERSION 1 VERSION 2 VERSION 3
7
Oscillator
The OSCCON Control register provides control of clock switching and clock source status information. The COSC Status bits in OSCCON are read-only bits that indicate the oscillator source that the device is operating from. The COSC bits are set to the FOS Configuration bit values at a Power-on Reset and will change to indicate the new oscillator source at the end of a clock switch operation. The NOSC Status bits in OSCCON are control bits that select the new clock source for a clock switch operation. The NOSC bits are set to the FOS Configuration bit values at a Power-on Reset or Brown-out Reset and are modified by the user software during a clock switch operation. The POST<1:0> control bits (OSCCON<8:7>) control the system clock divide ratio. The LOCK Status bit (OSCCON<5>) is read-only and indicates the status of the PLL circuit. The CF Status bit (OSCCON<3>) is a readable/writable Status bit that indicates a clock failure. The LPOSCEN control bit (OSCCON<1>) is used to enable or disable the 32 kHz Low-Power Crystal oscillator. The OSWEN control bit (OSCCON<0>) is used to initiate a clock switch operation. The OSWEN bit is cleared automatically after a successful clock switch. The TUN<3:0> bits allow the user to tune the internal FRC oscillator to frequencies higher and lower than the nominal value of 7.37 MHz. Note: The OSCCON register is write-protected because it controls the device clock switching mechanism. See Section 7.4.1 Protection Against Accidental Writes to OSCCON for instructions on writing to OSCCON.
7.4.1
Protection Against Accidental Writes to OSCCON

A write to the OSCCON register is intentionally made difficult, because it controls clock switching and clock scaling. To write to the OSCCON low byte, the following code sequence must be executed without any other instructions in between: Byte Write 0x46 to OSCCONL Byte Write 0x57 to OSCCONL After this sequence, a byte write to OSCCONL is allowed for one instruction cycle. Write the desired value or use a bit manipulation instruction. To write to the OSCCON high byte, the following instructions must be executed without any other instructions in between: Byte Write 0x78 to OSCCONH Byte Write 0x9A to OSCCONH After this sequence, a byte write is allowed to OSCCONH for one instruction cycle. Write the desired value or use a bit manipulation instruction.
DS70054D-page 7-13

Register 7-4: Upper Byte: U-0 bit 15 OSCCON: Oscillator Control Register Oscillator System VERSION 1 U-0 R-y R-y COSC<1:0> U-0 U-0 R/W-y R/W-y NOSC<1:0> bit 8
Lower Byte: R/W-0 R/W-0 POST<1:0> bit 7 bit 15-14 bit 13-12
R-0 LOCK
U-0
R/W-0 CF
U-0
R/W-0 LPOSCEN
R/W-0 OSWEN bit 0
bit 11-10 bit 9-8
bit 7-6
bit 5
bit 4 bit 3
bit 2 bit 1
bit 0
Unimplemented: Read as 0 COSC<1:0>: Current Oscillator Source Status bits 11 = Primary oscillator 10 = Internal LPRC oscillator 01 = Internal FRC oscillator 00 = Low-Power 32 kHz Crystal oscillator (Timer1) Unimplemented: Read as 0 NOSC<1:0>: New Oscillator Group Selection bits 11 = Primary oscillator 10 = Internal LPRC oscillator 01 = Internal FRC oscillator 00 = Low-Power 32 kHz Crystal oscillator (Timer1) POST<1:0>: Oscillator Postscaler Selection bits 11 = Oscillator postscaler divides clock by 64 10 = Oscillator postscaler divides clock by 16 01 = Oscillator postscaler divides clock by 4 00 = Oscillator postscaler does not alter clock LOCK: PLL Lock Status bit 1 = Indicates that PLL is in lock 0 = Indicates that PLL is out of lock (or disabled) Reset on POR or BOR Reset when a valid clock switching sequence is initiated Set when PLL lock is achieved after a PLL start Reset when lock is lost Read zero when PLL is not selected as a system clock Unimplemented: Read as 0 CF: Clock Fail Status bit 1 = FSCM has detected a clock failure 0 = FSCM has not detected a clock failure Reset on POR or BOR Reset when a valid clock switching sequence is initiated Set when clock fail detected Unimplemented: Read as 0 LPOSCEN: 32 kHz LP Oscillator Enable bit 1 = LP oscillator is enabled 0 = LP oscillator is disabled Reset on POR or BOR OSWEN: Oscillator Switch Enable bit 1 = Request oscillator switch to selection specified by NOSC<1:0> bits 0 = Oscillator switch is complete Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
y = Value set from Configuration bits on POR or BOR
DS70054D-page 7-14
Register 7-5: Upper Byte: R/W-0 TUN3 bit 15 OSCCON: Oscillator Control Register Oscillator System VERSION 2 R/W-0 TUN2 R-y R-y COSC<1:0> R/W-0 TUN1 R/W-0 TUN0 R/W-y R/W-y NOSC<1:0> bit 8
Lower Byte: R/W-0 R/W-0 POST<1:0> bit 7 bit 15-14 bit 13-12
R-0 LOCK
U-0
R/W-0 CF
U-0
R/W-0 LPOSCEN
R/W-0 OSWEN bit 0
7
Oscillator
TUN<3:2>: Upper 2 bits of the TUN bit-field. Refer to the description of TUN<1:0> (OSCCON<11:10>) bits for details. COSC<1:0>: Current Oscillator Source Status bits 11 = Primary oscillator 10 = Internal LPRC oscillator 01 = Internal FRC oscillator 00 = Low-Power 32 kHz Crystal oscillator (Timer1) TUN<1:0>: Lower 2 bits of the TUN bit-field. The four bit field specified by TUN<3:0> allows the user to tune the Internal Fast RC oscillator which has a nominal frequency of 7.37 MHz. For example, the user may be able to tune the frequency of the FRC oscillator within a range of +/- 12% (or 960 kHz) in steps of 1.5% around the factory-calibrated frequency setting, as follows: TUN<3:0> = 0111 provides the highest frequency ...... TUN<3:0> = 0000 provides the factory-calibrated frequency ...... TUN<3:0> = 1000 provides the lowest frequency Note: Refer to the device-specific data sheet for the exact tuning range and tuning step size for the FRC oscillator on your device.
bit 11-10
bit 9-8
NOSC<1:0>: New Oscillator Group Selection bits 11 = Primary oscillator 10 = Internal LPRC oscillator 01 = Internal FRC oscillator 00 = Low-Power 32 kHz Crystal oscillator (Timer1) POST<1:0>: Oscillator Postscaler Selection bits 11 = Oscillator postscaler divides clock by 64 10 = Oscillator postscaler divides clock by 16 01 = Oscillator postscaler divides clock by 4 00 = Oscillator postscaler does not alter clock LOCK: PLL Lock Status bit 1 = Indicates that PLL is in lock 0 = Indicates that PLL is out of lock (or disabled) Reset on POR or BOR Reset when a valid clock switching sequence is initiated Set when PLL lock is achieved after a PLL start Reset when lock is lost Read zero when PLL is not selected as a system clock Unimplemented: Read as 0 CF: Clock Fail Status bit 1 = FSCM has detected a clock failure 0 = FSCM has not detected a clock failure Reset on POR or BOR Reset when a valid clock switching sequence is initiated Set when clock fail detected Unimplemented: Read as 0
bit 7-6
bit 5
bit 4 bit 3
bit 2
DS70054D-page 7-15

OSCCON: Oscillator Control Register Oscillator System VERSION 2 (Continued) bit 1 LPOSCEN: 32 kHz LP Oscillator Enable bit 1 = LP oscillator is enabled 0 = LP oscillator is disabled Reset on POR or BOR OSWEN: Oscillator Switch Enable bit 1 = Request oscillator switch to selection specified by NOSC<1:0> bits 0 = Oscillator switch is complete Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 0
y = Value set from Configuration bits on POR or BOR
DS70054D-page 7-16
Register 7-6: Upper Byte: U-0 bit 15 Lower Byte: R/W-0 bit 7 bit 15 Unimplemented: Read as 0 R/W-0 R-0 LOCK U-0 R/W-0 CF U-0 R/W-0 LPOSCEN R/W-0 OSWEN bit 0 POST<1:0> R-y R-y COSC<2:0> R-y U-0 R/W-y R/W-y NOSC<2:0> bit 8 R/W-y OSCCON: Oscillator Control Register Oscillator System VERSION 3
7
Oscillator
bit 14-12 COSC<2:0>: Current Oscillator Group Selection bits (Read-Only) 111 = PLL Oscillator; PLL source selected by FPR<4:0> bits 011 = External Oscillator; OSC1/OSC2 pins; External Oscillator configuration selected by FPR<4:0> bits 010 = LPRC internal low-power RC 001 = FRC internal fast RC 000 = LP crystal oscillator; SOSCI/SOSCO pins Set to FOS<2:0> values on POR or BOR Loaded with NOSC<2:0> at the completion of a successful clock switch Set to FRC value when FSCM detects a failure and switches clock to FRC bit 11 bit 10-8 Unimplemented: Read as 0 NOSC<2:0>: New Oscillator Group Selection bits 111 = PLL Oscillator; PLL source selected by FPR<4:0> bits 011 = External Oscillator; OSC1/OSC2 pins; External Oscillator configuration selected by FPR<4:0> bits 010 = LPRC internal low-power RC 001 = FRC internal fast RC 000 = LP crystal oscillator; SOSCI/SOSCO pins Set to FOS<2:0> values on POR or BOR POST<1:0>: Oscillator Postscaler Selection bits 11 = Oscillator postscaler divides clock by 64 10 = Oscillator postscaler divides clock by 16 01 = Oscillator postscaler divides clock by 4 00 = Oscillator postscaler does not alter clock LOCK: PLL Lock Status bit (Read-Only) 1 = Indicates that PLL is in lock 0 = Indicates that PLL is out of lock (or disabled) Reset on POR or BOR Reset when a valid clock switching sequence is initiated Set when PLL lock is achieved after a PLL start Reset when lock is lost Read zero when PLL is not selected as a system clock Unimplemented: Read as 0
bit 7-6
bit 5
bit 4
DS70054D-page 7-17

OSCCON: Oscillator Control Register Oscillator System VERSION 3 (Continued) bit 3 CF: Clock Fail Detect bit (Read/Clearable by application) 1 = FSCM has detected clock failure 0 = FSCM has NOT detected clock failure Reset on POR or BOR Reset when a valid clock switching sequence is initiated Set when clock fail detected Unimplemented: Read as 0 LPOSCEN: 32 kHz Secondary (LP) Oscillator Enable bit 1 = Secondary Oscillator is enabled 0 = Secondary Oscillator is disabled Reset on POR or BOR OSWEN: Oscillator Switch Enable bit 1 = Request Oscillator switch to selection specified by NOSCG<2:0> bits 0 = Oscillator switch is complete Reset on POR or BOR Reset after a successful clock switch Reset after a redundant clock switch (i.e., a clock switch operation is requested to the current oscillator) Reset after FSCM switches the oscillator to (Group 1) FRC Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 2 bit 1
bit 0
y = Value set from Configuration bits on POR
DS70054D-page 7-18
Register 7-7: Upper Byte: U-0 bit 15 Lower Byte: U-0 bit 7 bit 15-4 bit 3-0 Unimplemented: Read as 0 U-0 U-0 U-0 R/W-0 R/W-0 R/W-0 R/W-0 bit 0 TUN<3:0> U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8 OSCTUN: FRC Oscillator Tuning Register Oscillator System VERSION 3 Only
7
Oscillator
TUN<3:0>: The four bit field specified by TUN<3:0> allows the user to tune the Internal Fast RC oscillator which has a nominal frequency of 7.37 MHz. TUN<3:0> = 0111 provides the highest frequency ...... TUN<3:0> = 0000 provides the factory-calibrated frequency ...... TUN<3:0> = 1000 provides the lowest frequency Note 1: Refer to the device-specific data sheet for the exact tuning range and tuning step size for the FRC oscillator on your device. 2: Certain devices may have more than four TUN bits. Refer to the device-specific data sheet to identify the number of TUN bits available to the user for tuning the FRC oscillator. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
y = Value set from Configuration bits on POR
DS70054D-page 7-19

7.5 Primary Oscillator
The Primary oscillator is available on the OSC1 and OSC2 pins of the dsPIC30F device family. The Primary oscillator has a wide variety of operation modes summarized in Table 7-6. In general, the Primary oscillator can be configured for an external clock input, external RC network, or an external crystal. Further details of the Primary Oscillator Operating modes are described in subsequent sections. The FPR bits in the FOSC nonvolatile Configuration register select the Operating mode of the Primary oscillator. Table 7-6: Primary Oscillator Operating Modes Description 200 kHz-4 MHz crystal on OSC1:OSC2. 4 MHz-10 MHz crystal on OSC1:OSC2. 4 MHz-10 MHz crystal on OSC1:OSC2, 4x PLL enabled. 4 MHz-10 MHz crystal on OSC1:OSC2, 8x PLL enabled. 4 MHz-10 MHz crystal on OSC1:OSC2, 16x PLL enabled(1). 32 kHz crystal on SOSCO:SOSCI(2). 10 MHz-25 MHz crystal. 10 MHz -25 MHz crystal, divide by 2, 4x PLL enabled. 10 MHz-25MHz crystal, divide by 2, 8x PLL enabled. 10 MHz-25MHz crystal, divide by 2, 16x PLL enabled(1). 10 MHz-25 MHz crystal, divide by 3, 4x PLL enabled. 10 MHz-25MHz crystal, divide by 3, 8x PLL enabled. 10 MHz-25MHz crystal, divide by 3, 16x PLL enabled(1). External clock input (0-40 MHz). External clock input (0-40 MHz), OSC2 pin is I/O. External clock input (4-10 MHz), OSC2 pin is I/O, 4x PLL enabled(1). External clock input (4-10 MHz), OSC2 pin is I/O, 8x PLL enabled(1). External clock input (4-10 MHz), OSC2 pin is I/O, 16x PLL enabled(1). External RC oscillator, OSC2 pin is FOSC/4 output(3). External RC oscillator, OSC2 pin is I/O(3). 7.37 MHz internal Fast RC oscillator. 7.37 MHz Internal Fast RC oscillator, 4x PLL enabled. 7.37 MHz Internal Fast RC oscillator, 8x PLL enabled. 7.37 MHz Internal Fast RC oscillator, 16x PLL enabled. 512 kHz internal Fast RC oscillator. dsPIC30F maximum operating frequency of 120 MHz must be met. LP oscillator can be conveniently shared as system clock, as well as real-time clock for Timer1. Requires external R and C. Frequency operation up to 4 MHz. This table lists a cumulative set of operating modes featured in Oscillator system VERSION 1, VERSION 2 and VERSION 3.
Oscillator Mode(4) XTL XT XT w/ PLL 4x XT w/ PLL 8x XT w/ PLL 16x LP HS HS/2 w/PLL 4x HS/2 w/ PLL 8x HS/2 w/ PLL 16x HS/3 w/PLL 4x HS/3 w/ PLL 8x HS/3 w/ PLL 16x EC ECIO EC w/ PLL 4x EC w/ PLL 8x EC w/ PLL 16x ERC ERCIO FRC FRC w/ PLL 4x FRC w/ PLL 8x FRC w/ PLL 16x LPRC Note 1: 2: 3: 4:
DS70054D-page 7-20
7.5.1 Oscillator Mode Selection Guidelines
The main difference between the XT, XTL and HS modes is the gain of the internal inverter of the oscillator circuit, which allows the different frequency ranges. In general, use the oscillator option with the lowest possible gain that still meets specifications. This will result in lower dynamic currents (IDD). The frequency range of each Oscillator mode is the recommended frequency cutoff, but the selection of a different Gain mode is acceptable as long as a thorough validation is performed (voltage, temperature and component variations, such as resistor, capacitor and internal oscillator circuitry). The oscillator feedback circuit is disabled in all EC and ECIO modes. The OSC1 pin is a high impedance input and can be driven by a CMOS driver. The ERC and ERCIO modes provide the least expensive solution for device oscillation (only an external resistor and capacitor is required). These modes also provide the most variation in the oscillation frequency. If the Primary oscillator is configured for an external clock input or an external RC network, the OSC2 pin is not required to support the oscillator function. For these modes, the OSC2 pin can be used as an additional device I/O pin or a clock output pin. When the OSC2 pin is used as a clock output pin, the output frequency is FOSC/4. The XTL mode is a Low Power/Low Frequency mode. This mode of the oscillator consumes the least amount of power of the three Crystal modes. The XT mode is a Medium Power/Medium Frequency mode and HS mode provides the highest oscillator frequencies with a crystal. The EC and XT modes that use the PLL circuit provide the highest device operating frequencies. The oscillator circuit will consume the most current in these modes because the PLL is enabled to multiply the frequency of the oscillator.
7
Oscillator
DS70054D-page 7-21

7.6 Crystal Oscillators/Ceramic Resonators
In XT, XTL and HS modes, a crystal or ceramic resonator is connected to the OSC1 and OSC2 pins to establish oscillation (Figure 7-3). The dsPIC30F oscillator design requires the use of a parallel cut crystal. Using a series cut crystal may give a frequency out of the crystal manufacturers specifications. Figure 7-3: Crystal or Ceramic Resonator Operation (XT, XT or HS Oscillator Mode)
OSC1 C1(3) XTAL OSC2 RS(1) C2(3) dsPIC30FXXXX Note 1: A series resistor, Rs, may be required for AT strip cut crystals. 2: The internal feedback resistor, RF, is typically in the range of 2 to 10 M. 3: See Section 7.7 Determining Best Values for Crystals, Clock Mode, C1, C2 and Rs. RF(2) To Internal Logic Sleep
7.6.1
Oscillator/Resonator Start-up
As the device voltage increases from VSS, the oscillator will start its oscillations.The time required for the oscillator to start oscillating depends on many factors. These include: Crystal/resonator frequency Capacitor values used (C1 and C2 in Figure 7-3) Device VDD rise time System temperature Series resistor value and type if used (Rs in Figure 7-3) Oscillator mode selection of device (selects the gain of the internal oscillator inverter) Crystal quality Oscillator circuit layout System noise
Figure 7-4 shows a plot of a typical oscillator/resonator start-up. Figure 7-4: Example Oscillator/Resonator Start-up Characteristics
Maximum VDD of System Device VDD VIH Voltage
VIL
0V Crystal Start-up Time Time
DS70054D-page 7-22
7.6.2 Tuning the Oscillator Circuit
Since Microchip devices have wide operating ranges (frequency, voltage and temperature; depending on the part and version ordered) and external components (crystals, capacitors,...) of varying quality and manufacture, validation of operation needs to be performed to ensure that the component selection will comply with the requirements of the application. There are many factors that go into the selection and arrangement of these external components. These factors include: amplifier gain desired frequency resonant frequency(s) of the crystal temperature of operation supply voltage range start-up time stability crystal life power consumption simplification of the circuit use of standard components component count
7
Oscillator
7.6.3
Oscillator Start-up from Sleep Mode

The most difficult time for the oscillator to start-up is when waking up from Sleep mode. This is because the load capacitors have both partially charged to some quiescent value, and phase differential at wake-up is minimal. Thus, more time is required to achieve stable oscillation. Remember also that low voltage, high temperatures and the Lower Frequency Clock modes also impose limitations on loop gain, which in turn affects start-up. Each of the following factors increases the start-up time: a low frequency design (with a Low Gain Clock mode) a quiet environment (such as a battery operated device) operating in a shielded box (away from the noisy RF area) low voltage high temperature wake-up from Sleep mode
Noise actually helps lower the oscillator start-up time since it provides a kick start to the oscillator. Prior to entering Sleep mode, the application may switch to the Internal FRC(+PLL) oscillator in order to reduce the time taken by the device to wake-up from Sleep.
DS70054D-page 7-23

7.7 Determining Best Values for Crystals, Clock Mode, C1, C2 and RS
The best method for selecting components is to apply a little knowledge and a lot of trial, measurement and testing. Crystals are usually selected by their parallel resonant frequency only, however, other parameters may be important to your design, such as temperature or frequency tolerance. Application Note AN588 PICmicro Microcontroller Oscillator Design Guide is an excellent reference to learn more about crystal operation and their ordering information. The dsPIC30F internal oscillator circuit is a parallel oscillator circuit, which requires that a parallel resonant crystal be selected. The load capacitance is usually specified in the 22 pF to 33 pF range. The crystal will oscillate closest to the desired frequency with a load capacitance in this range. It may be necessary to alter these values, as described later, in order to achieve other benefits. The Clock mode is primarily chosen based on the desired frequency of the crystal oscillator. The main difference between the XT, XTL and HS Oscillator modes is the gain of the internal inverter of the oscillator circuit, which allows the different frequency ranges. In general, use the oscillator option with the lowest possible gain that still meets specifications. This will result in lower dynamic currents (IDD). The frequency range of each Oscillator mode is the recommended frequency cutoff, but the selection of a different Gain mode is acceptable, as long as a thorough validation is performed (voltage, temperature and component variations, such as resistor, capacitor and internal oscillator circuitry). C1 and C2 (see Figure 7-3) should also be initially selected based on the load capacitance as suggested by the crystal manufacturer and the tables supplied in the device data sheet. The values given in the device data sheet can only be used as a starting point since the crystal manufacturer, supply voltage, and other factors already mentioned may cause your circuit to differ from the one used in the factory characterization process. Ideally, the capacitance is chosen so that it will oscillate at the highest temperature and the lowest VDD that the circuit will be expected to perform under. High temperature and low VDD both have a limiting effect on the loop gain, such that if the circuit functions at these extremes, the designer can be more assured of proper operation at other temperatures and supply voltage combinations. The output sine wave should not be clipped in the highest gain environment (highest VDD and lowest temperature) and the sine output amplitude should be large enough in the lowest gain environment (lowest VDD and highest temperature) to cover the logic input requirements of the clock as listed in the device data sheet. A method for improving start-up is to use a value of C2 greater than C1. This causes a greater phase shift across the crystal at power-up, which speeds oscillator start-up. Besides loading the crystal for proper frequency response, these capacitors can have the effect of lowering loop gain if their value is increased. C2 can be selected to affect the overall gain of the circuit. A higher C2 can lower the gain if the crystal is being over driven (also, see discussion on Rs). Capacitance values that are too high can store and dump too much current through the crystal, so C1 and C2 should not become excessively large. Unfortunately, measuring the wattage through a crystal is difficult, but if you do not stray too far from the suggested values you should not have to be concerned with this. A series resistor, Rs, is added to the circuit if, after all other external components are selected to satisfaction, the crystal is still being overdriven. This can be determined by looking at the OSC2 pin, which is the driven pin, with an oscilloscope. Connecting the probe to the OSC1 pin will load the pin too much and negatively affect performance. Remember that a scope probe adds its own capacitance to the circuit, so this may have to be accounted for in your design (i.e., if the circuit worked best with a C2 of 22 pF and scope probe was 10 pF, a 33 pF capacitor may actually be called for). The output signal should not be clipping or flattened. Overdriving the crystal can also lead to the circuit jumping to a higher harmonic level or even crystal damage.
DS70054D-page 7-24
The OSC2 signal should be a clean sine wave that easily spans the input minimum and maximum of the clock input pin (4V to 5V peak-to-peak for a 5V VDD is usually good). An easy way to set this is to again test the circuit at the minimum temperature and maximum VDD that the design will be expected to perform in, then look at the output. This should be the maximum amplitude of the clock output. If there is clipping or the sine wave is distorted near VDD and VSS, increasing load capacitors may cause too much current to flow through the crystal or push the value too far from the manufacturers load specification. To adjust the crystal current, add a trimmer potentiometer between the crystal inverter output pin and C2 and adjust it until the sine wave is clean. The crystal will experience the highest drive currents at the low temperature and high VDD extremes. The trimmer potentiometer should be adjusted at these limits to prevent overdriving. A series resistor, Rs, of the closest standard value can now be inserted in place of the trimpot. If Rs is too high, perhaps more than 20 kOhms, the input will be too isolated from the output, making the clock more susceptible to noise. If you find a value this high is needed to prevent overdriving the crystal, try increasing C2 to compensate or changing the Oscillator Operating mode. Try to get a combination where Rs is around 10k or less and load capacitance is not too far from the manufacturer specification.
7
Oscillator
7.8
External Clock Input

Two of the Primary Oscillator modes use an external clock. These modes are EC and ECIO. In the EC mode (Figure 7-5), the OSC1 pin can be driven by CMOS drivers. In this mode, the OSC1 pin is high-impedance and the OSC2 pin is the clock output (FOSC/4). This output clock is useful for testing or synchronization purposes. Figure 7-5: External Clock Input Operation (EC Oscillator Configuration)
OSC1 dsPIC30F OSC2
Clock from Ext. System FOSC/4
In the ECIO mode (Figure 7-6), the OSC1 pin can be driven by CMOS drivers. In this mode, the OSC1 pin is high-impedance and the OSC2 pin becomes a general purpose I/O pin. The feedback device between OSC1 and OSC2 is turned off to save current. Figure 7-6: External Clock Input Operation (ECIO Oscillator Configuration)
OSC1 dsPIC30F I/O I/O (OSC2)
Clock from Ext. System
DS70054D-page 7-25

7.9 External RC Oscillator
For timing insensitive applications, the ERC and ERCIO modes of the Primary oscillator offer additional cost savings. The RC oscillator frequency is a function of the: Supply voltage External resistor (REXT) values External capacitor (CEXT) values Operating temperature
In addition to this, the oscillator frequency will vary from unit-to-unit due to normal process parameter variation. Furthermore, the difference in lead frame capacitance between package types will also affect the oscillation frequency, especially for low CEXT values. The user also needs to take into account variation due to tolerance of external REXT and CEXT components used. Figure 7-7 shows how the RC combination is connected. For REXT values below 2.2 k, oscillator operation may become unstable or stop completely. For very high REXT values (e.g., 1 M), the oscillator becomes sensitive to noise, humidity and leakage. Thus, it is recommended that a REXT value between 3 k and 100 k is used. Figure 7-7: ERC Oscillator Mode
V DD REXT OSC1 CEXT VSS OSC2 FOSC/4 Internal Clock dsPIC30F
Although the oscillator will operate with no external capacitor (CEXT = 0 pF), a value above 20 pF should be used for noise and stability reasons. With no or a small external capacitance, the oscillation frequency can vary dramatically due to changes in external capacitances, such as PCB trace capacitance and package lead frame capacitance. The oscillator frequency, divided by 4, is available on the OSC2/CLKO pin, and can be used for test purposes or to synchronize other logic. Note: An external clock source should not be connected to the OSC1 pin when the oscillator is configured for ERC or ERCIO modes.
DS70054D-page 7-26
7.9.1 External RC Oscillator with I/O Enabled
The ERCIO Oscillator mode functions in the exact same manner as the ERC Oscillator mode. The only difference is that the OSC2 pin is configured as an I/O pin. As in the RC mode, the user needs to take into account any variation of the clock frequency due to tolerance of external REXT and CEXT components used, process variation, voltage and temperature. Figure 7-8 shows how the RC with the I/O pin combination is connected. Figure 7-8: ERCIO Oscillator Mode
VDD
7
REXT OSC1 CEXT VSS I/O (OSC2) dsPIC30F Internal Clock
Oscillator
7.9.2
External RC Start-up
There is no start-up delay associated with the RC oscillator. Oscillation will begin when VDD is applied. Note: The user should verify that VDD is within specifications before the device begins to execute code.
7.9.3
RC Operating Frequency
The following graphs show the external RC oscillator frequency as a function of device voltage for a selection of RC component values. Note: The following graphs should be used only as approximate guidelines for RC component selection. The actual frequency will vary based on the system temperature and device. Please refer to the specific device data sheet for further RC oscillator characteristic data.
DS70054D-page 7-27

Figure 7-9:
5.0
Typical External RC Oscillator Frequency vs. VDD, CEXT = 20 pF
4.5
Operation above 4 MHz is not recommended.
4.0
3.5 REXT = 10k 3.0 Freq (MHz)
2.5
2.0
1.5
1.0 REXT = 100k 0.5
0.0 2.5 3.0 3.5 4.0 VDD (V) 4.5 5.0 5.5
Figure 7-10:
5.0
Operation above 4 MHz is not recommended.

4.0 REXT = 5.1k
3.0 Freq (MHz)
REXT = 10k 2.0
1.0
REXT = 100k 0.0 2.5 3.0 3.5 4.0 VDD (V) 4.5 5.0 5.5
DS70054D-page 7-28
Figure 7-11:
300
250 REXT = 3.3k
200 REXT = 5.1k Freq (kHz)
7
Oscillator
150
REXT = 10k 100
50 REXT = 100k 0 2.5 3.0 3.5 4.0 VDD (V) 4.5 5.0 5.5
DS70054D-page 7-29

7.10 Phase Locked Loop (PLL)
The PLL can be enabled for x4, x8 or x16 Operation modes using the FPR<3:0> oscillator Configuration bits. The input and output frequency ranges for each Operating mode are summarized in Table 7-7. Note: Some PLL output frequency ranges can be achieved that exceed the maximum operating frequency of the dsPIC30F device. Refer to the Electrical Specifications in the specific device data sheet for further details. PLL Frequency Range FIN 4 MHz-10 MHz 4 MHz-10 MHz 4 MHz-7.5 MHz PLL Multiplier x4 x8 x16 FOUT 16 MHz-40 MHz 32 MHz-80 MHz 64 MHz-120 MHz
Table 7-7:
7.10.1
PLL Lock Status

The PLL circuit is able to detect when the PLL enters a phase locked state. It can also detect when the PLL loses lock. The time delay for the PLL to achieve lock is designated as TLOCK. The TLOCK value is nominally 20 s. Refer to the Electrical Specifications in the specific device data sheet for further information. The LOCK bit is a read-only Status bit (OSCCON<5>) that reflects the LOCK status of the PLL. The LOCK bit is cleared at a Power-on Reset.
7.10.1.1
Loss of PLL Lock During Clock Switching When the PLL is selected as a destination clock source in a clock switch operation (including a Power-on Reset), the LOCK bit is cleared. The LOCK bit is set after phase lock has been achieved. If the PLL fails to achieve lock, then the clock switching circuit will NOT switch to the PLL output for system clock; instead, it will continue to run with the old clock source.
7.10.1.2
Loss of PLL Lock During a Power-on Reset If the PLL fails to achieve lock at a Power-on Reset (POR) and the Fail-Safe Clock Monitor (FSCM) is enabled, the FRC oscillator will become the device clock source and a clock failure trap will occur.
7.10.1.3
Loss of PLL Lock During Normal Device Operation If the PLL loses lock during normal operation for at least 4 input clock cycles, then the LOCK bit is cleared, indicating a loss of PLL lock. Furthermore, a clock failure trap will be generated. In this situation, the processor continues to run using the PLL clock source. The user can switch to another clock source in the Trap Service Routine, if desired. Note: Refer to Section 6. Reset Interrupts for further details about oscillator failure traps. A loss of PLL lock during normal device operation will generate a clock failure trap, but the system clock source will not be changed. The FSCM does not need to be enabled to detect the loss of lock.
Note:
DS70054D-page 7-30
7.11 Low-Power 32 kHz Crystal Oscillator
The LP or Secondary oscillator is designed specifically for low power operation with a 32 kHz crystal. The LP oscillator is located on the SOSCO and SOSCI device pins and serves as a secondary crystal clock source for low power operation. The LP oscillator can also drive Timer1 for a real-time clock application.
7.11.1
LP Oscillator Enable
The following control bits affect the operation of the LP oscillator: 1. 2. The COSC<1:0> bits in the OSCCON register (OSCCON<13:12>). The LPOSCEN bit in the OSCCON register (OSCCON<1>).
When the LP Oscillator is enabled, the SOSCO and SOSCI I/O pins are controlled by the oscillator and cannot be used for other I/O functions. 7.11.1.1 LP Oscillator Continuous Operation The LP oscillator will always be enabled if the LPOSCEN control bit (OSCCON<1>) is set. There are two reasons to leave the LP oscillator running. First, keeping the LP oscillator ON at all times allows a fast switch to the 32 kHz system clock for lower power operation. Returning to the faster main oscillator will still require an oscillator start-up time if it is a crystal type source (see Section 7.12 Oscillator Start-up Timer (OST)). Second, the oscillator should remain ON at all times when using Timer1 as a real-time clock. 7.11.1.2 LP Oscillator Intermittent Operation When the LPOSCEN control bit (OSCCON<1>) is cleared, the LP oscillator will only operate when it is selected as the current device clock source (COSC<1:0> = 00). The LP oscillator will be disabled if it is the current device clock source and the device enters Sleep mode.
7
Oscillator
7.11.2
LP Oscillator Operation with Timer1

The LP oscillator can be used as a clock source for Timer1 in a real-time clock application. Refer to Section 12. Timers for further details.
7.12
Oscillator Start-up Timer (OST)

In order to ensure that a crystal oscillator (or ceramic resonator) has started and stabilized, an Oscillator Start-up Timer is provided. It is a simple 10-bit counter that counts 1024 TOSC cycles before releasing the oscillator clock to the rest of the system. The time-out period is designated as TOST. The amplitude of the oscillator signal must reach the VIL and VIH thresholds for the oscillator pins before the OST can begin to count cycles (see Figure 7-4). The TOST time is involved every time the oscillator has to restart (i.e., on POR, BOR and wake-up from Sleep mode). The Oscillator Start-up Timer is applied to the LP oscillator and the XT, XTL and HS modes for the Primary oscillator.
7.13
Internal Fast RC Oscillator (FRC)

The FRC oscillator is a fast (7.37 MHz nominal) internal RC oscillator. This oscillator is intended to provide a range of device operating speeds without the use of an external crystal, ceramic resonator or RC network. Devices featuring the Oscillator system VERSIONs 2 or 3 may optionally provide the FRC oscillator as an input frequency to the PLL.
DS70054D-page 7-31

7.14 Internal Low-Power RC (LPRC) Oscillator
The LPRC oscillator is a component of the Watchdog Timer (WDT) and oscillates at a nominal frequency of 512 kHz. The LPRC oscillator is the clock source for the Power-up Timer (PWRT) circuit, WDT and clock monitor circuits. It may also be used to provide a low frequency clock source option for applications where power consumption is critical, and timing accuracy is not required. Note: The oscillation frequency of the LPRC oscillator will vary depending on the device voltage and operating temperature. Refer to the Electrical Specifications in the specific device data sheet for further details.
7.14.1
Enabling the LPRC Oscillator

The LPRC oscillator is always enabled at a Power-on Reset because it is the clock source for the PWRT. After the PWRT expires, the LPRC oscillator will remain ON if one of the following is TRUE: The Fail-Safe Clock Monitor is enabled. The WDT is enabled. The LPRC oscillator is selected as the system clock (COSC<1:0> = 10). If none of the above conditions is true, the LPRC will shut-off after the PWRT expires.
7.15
Fail-Safe Clock Monitor (FSCM)

The Fail-Safe Clock Monitor (FSCM) allows the device to continue to operate even in the event of an oscillator failure. The FSCM function is enabled by programming the FCKSM bits (Clock Switch and Monitor bits) in the FOSC Device Configuration register. Refer to 7.3 Oscillator Configuration for further details. If the FSCM function is enabled, the LPRC internal oscillator will run at all times (except during Sleep mode). In the event of an oscillator failure, the FSCM will generate a clock failure trap and will switch the system clock to the FRC oscillator. The user will then have the option to either attempt to restart the oscillator or execute a controlled shutdown. The FSCM module will take the following actions when switching to the FRC oscillator: 1. 2. 3. The COSC<1:0> bits are loaded with 01. The CF bit is set to indicate the clock failure. The OSWEN control bit is cleared to cancel any pending clock switches. Note: For more information about the oscillator failure trap, please refer to Section 6. Reset Interrupts.
7.15.1
FSCM Delay
On a POR, BOR or wake-up event from Sleep mode, a nominal 100 s delay (TFSCM) may be inserted before the FSCM begins to monitor the system clock source. The purpose of the FSCM delay is to provide time for the oscillator and/or PLL to stabilize when the Power-up Timer (PWRT) is not utilized. The FSCM delay will be generated after the internal System Reset signal, SYSRST, has been released. Refer to Section 8. Reset for FSCM delay timing information. The FSCM delay, TFSCM, is applied when the FSCM is enabled and any of the following device clock sources is selected as the system clock: EC+PLL XT+PLL XT HS HS/2 or HS/3 + PLL XTL LP Note: Please refer to the Electrical Specifications section of the device data sheet for TFSCM specification values.
DS70054D-page 7-32
7.15.2 FSCM and Slow Oscillator Start-up
If the chosen device oscillator has a slow start-up time coming out of POR, BOR or Sleep mode, it is possible that the FSCM delay will expire before the oscillator has started. In this case, the FSCM will initiate a clock failure trap. As this happens, the COSC<1:0> bits (OSCCON<13:12>) are loaded with the FRC oscillator selection. This will effectively shut-off the original oscillator that was trying to start. The user can detect this situation and initiate a clock switch back to the desired oscillator in the Trap Service Routine.
7.15.3
FSCM and WDT

In the event of a clock failure, the WDT is unaffected and continues to run on the LPRC clock.
7.16
Programmable Oscillator Postscaler

The postscaler allows the user to save power by lowering the frequency of the clock which feeds the CPU and the peripherals. Postscale values can be changed at any time via the POST<1:0> control bits (OSCCON<7:6>). To ensure a clean clock transition, there is some delay before a clock change occurs. The clock postscaler does not change the clock selection multiplexer until a falling edge on the divide-by-64 output occurs. In effect, the switching delay could be up to 64 system clock cycles depending on when the POST<1:0> control bits are written. Figure 7-13 shows the postscaler operation for three different postscaler changes. Figure 7-12: Programmable Oscillator Postscaler
7
Oscillator
System Clock Input (from Clock Switch and Control Logic)
div. by 4 Counter div. by 16 div. by 64
00 01 10 11
Postscaled System Clock
POST1
POST0 Note: The system clock input can be any available source.
DS70054D-page 7-33

Figure 7-13: Postscaler Update Timing
System Clock Divide by 4 Divide by 16
Divide by 64
POST<1:0> Postscaled System Clock
01
10
00
11
1:4
1:16
1:1
1:64
Note: This diagram demonstrates the clock postscaler function only. The divide ratios shown in the timing diagram are not correct.
7.17
Clock Switching Operation

The selection of clock sources available for clock switching during device operation are as follows: Primary oscillator on OSC1/OSC2 pins Low-Power 32 kHz Crystal (Secondary) oscillator on SOSCO/SOSCI pins Internal Fast RC (FRC) oscillator Internal Low-Power RC (LPRC) oscillator Note: The Primary oscillator has multiple operating modes (EC, RC, XT, FRC etc.). The Operating mode of the Primary oscillator is determined by the FPR Configuration bits in the FOSC device Configuration register. (Refer to 7.3 Oscillator Configuration for further details.)
7.17.1
Clock Switching Enable

To enable clock switching, the FCKSM1 Configuration bit in the FOSC Configuration register must be programmed to a 0. (Refer to 7.3 Oscillator Configuration for further details.) If the FCKSM1 Configuration bit is a 1 (unprogrammed), then the clock switching function is disabled. The Fail-Safe Clock Monitor function is also disabled. This is the default setting. The NOSC control bits in OSCCON do not control the clock selection when clock switching is disabled. However, the COSC bits in OSCCON will reflect the clock source selected by the FPR and FOS Configuration bits in the FOSC Configuration register. The OSWEN control bit (OSCCON<0>) has no effect when clock switching is disabled. It is held at 0 at all times.
DS70054D-page 7-34
7.17.2 Oscillator Switching Sequence
The following steps are taken by the hardware and software to change the device clock source. (The steps shown below use the OSCCON register definition for the Oscillator system VERSION 1. For a description of the OSCCON register for the Oscillator system VERSION 2 and VERSION 3, refer to 7.4 Oscillator Control Registers OSCCON and OSCTUN): 1. Read the COSC<1:0> Status bits (OSCCON<13:12>), if desired, to determine current oscillator source. 2. Perform the unlock sequence to allow a write to the OSCCON register high byte. 3. Write the appropriate value to the NOSC<1:0> control bits (OSCCON<9:8>) for the new oscillator source. 4. Perform the unlock sequence to allow a write to the OSCCON register low byte. 5. Set the OSWEN bit (OSCCON<0>). This will INITIATE the oscillator switch. 6. The clock switching hardware compares the COSC<1:0> Status bits with the new value of the NOSC<1:0> control bits. If they are the same, then the clock switch is a redundant operation. In this case, the OSWEN bit is cleared automatically and the clock switch is aborted. 7. If a valid clock switch has been initiated, the LOCK (OSCCON<5>) and the CF (OSCCON<3>) Status bits are cleared. 8. The new oscillator is turned on by the hardware if it is not currently running. If a crystal oscillator must be turned on, the hardware will wait until the OST expires. If the new source is using the PLL, then the hardware waits until a PLL lock is detected (LOCK = 1). 9. The hardware waits for 10 clock cycles from the new clock source and then performs the clock switch. 10. The hardware clears the OSWEN bit to indicate a successful clock transition. In addition, the NOSC<1:0> bit values are transferred to the COSC<1:0> Status bits. 11. The clock switch is completed. The old clock source will be turned off at this time, with the following exceptions: The LPRC oscillator will stay on if the WDT or FSCM is enabled. The LP oscillator will stay on if LPOSCEN = 1 (OSCCON<1>). Note: The processor will continue to execute code throughout the clock switching sequence. Timing sensitive code should not be executed during this time. Clock Transition Timing Diagram
New Source Enabled New Source Stable Old Source Disabled
7
Oscillator
Figure 7-14:
Old Clock Source New Clock Source System Clock
1 2 3 4 5 6 7 8 9 10
OSWEN
Both Oscillators Active
Note: The system clock can be any selected source Primary, Secondary, FRC or LPRC.
DS70054D-page 7-35

7.17.3 Clock Switching Tips
If the destination clock source is a crystal oscillator, the clock switch time will be dominated by the oscillator start-up time. If the new clock source does not start, or is not present, then the clock switching hardware will simply wait for the 10 synchronization cycles to occur. The user can detect this situation because the OSWEN bit (OSCCON<0>) remains set indefinitely. If the new clock source uses the PLL, a clock switch will not occur until lock has been achieved. The user can detect a loss of PLL lock because the LOCK bit will be cleared and the OSWEN bit is set. The user may wish to consider the settings of the POST<1:0> control bits (OSCCON<7:6>) when executing a clock switch. Switching to a low frequency clock source, such as the LP oscillator with a postscaler ratio greater than 1:1, will result in very slow device operation. Note: The application should not attempt to switch to a clock of frequency lower than 100 kHz when the Fail-Safe Clock Monitor is enabled. If such clock switching is performed, the device may generate an oscillator fail trap and switch to the Fast RC oscillator.
7.17.4
Aborting a Clock Switch

In the event the clock switch did not complete, the clock switch logic can be reset by clearing the OSWEN bit. Clearing the OSWEN bit (OSCCON<0>) will: 1. 2. 3. Abandon the clock switch Stop and reset the OST, if applicable Stop the PLL, if applicable
A clock switch procedure can be aborted at any time.
7.17.5
Entering Sleep Mode During a Clock Switch

If the device enters Sleep mode during a clock switch operation, the clock switch operation is aborted. The processor keeps the old clock selection and the OSWEN bit is cleared. The PWRSAV instruction is then executed normally.
7.17.6
Recommended Code Sequence for Clock Switching

The following steps should be taken to change the oscillator source: Disable interrupts during the OSCCON register unlock and write sequence. Execute unlock sequence for OSCCON high byte. Write new oscillator source to NOSC control bits. Execute unlock sequence for OSCCON low byte. Set OSWEN bit. Continue to execute code that is not clock sensitive (optional). Invoke an appropriate amount of software delay (cycle counting) to allow for oscillator and/or PLL start-up. Check to see if OSWEN is 0. If it is, we are DONE SUCCESSFULLY. If OSWEN is still set, then check LOCK bit to determine cause of failure.
DS70054D-page 7-36
7.17.7
7.17.7.1
Clock Switch Code Examples

Starting a Clock Switch The following code sequence shows how to unlock the OSCCON register and begin a clock switch operation:
;Place the new oscillator selection in W0 ;OSCCONH (high byte) Unlock Sequence MOV #OSCCONH, w1 MOV #0x78, w2 MOV #0x9A, w3 MOV.B w2, [w1] MOV.B w3, [w1] ;Set new oscillator selection MOV.B WREG, OSCCONH
7
Oscillator
;OSCCONL MOV MOV.B MOV MOV MOV.B MOV.B
(low byte) unlock sequence #OSCCONL, w1 #0x01, w0 #0x46, w2 #0x57, w3 w2, [w1] w3, [w1]
;Start oscillator switch operation MOV.b w0, [w1]
7.17.7.2
Aborting a Clock Switch The following code sequence would be used to ABORT an unsuccessful clock switch:
MOV BCLR MOV MOV.B MOV.B MOV.B MOV.B MOV.B OSCCON,W0 W0, #OSWEN #OSCCON,W1 #0x46,W2 #0x57,W3 W2, [W1] W3, [W1] W0, [W1] ; ; ; ; ; ; ; ; Read OSCCON into W0 Clear bit 0 in W0 pointer to OSCCON first unlock code second unlock code write first unlock code write second unlock code ABORT the switch
DS70054D-page 7-37

7.18 Design Tips
Question 1: Answer: 1. Entering Sleep mode with no source for wake-up (such as, WDT, MCLR, or an interrupt). Verify that the code does not put the device to Sleep without providing for wake-up. If it is possible, try waking it up with a low pulse on MCLR. Powering up with MCLR held low will also give the crystal oscillator more time to start-up, but the Program Counter will not advance until the MCLR pin is high. The wrong Clock mode is selected for the desired frequency. For a blank device, the default oscillator is EC + 16x PLL. Most parts come with the clock selected in the Default mode, which will not start oscillation with a crystal or resonator. Verify that the Clock mode has been programmed correctly. The proper power-up sequence has not been followed. If a CMOS part is powered through an I/O pin prior to power-up, bad things can happen (latch-up, improper start-up, etc.). It is also possible for brown-out conditions, noisy power lines at start-up, and slow VDD rise times to cause problems. Try powering up the device with nothing connected to the I/O, and power-up with a known, good, fast rise, power supply. Refer to the power-up information in the device data sheet for considerations on brown-out and power-up sequences. The C1 and C2 capacitors attached to the crystal have not been connected properly or are not the correct values. Make sure all connections are correct. The device data sheet values for these components will usually get the oscillator running; however, they just might not be the optimal values for your design. The device starts, but runs at a frequency much higher than the resonant frequency of the crystal. When looking at the OSC2 pin after power-up with an oscilloscope, there is no clock. What can cause this?
2.
3.
4.
Question 2:
Answer: The gain is too high for this oscillator circuit. Refer to Section 7.6 Crystal Oscillators/Ceramic Resonators to aid in the selection of C2 (may need to be higher), Rs (may be needed) and Clock mode (wrong mode may be selected). This is especially possible for low frequency crystals, like the common 32.768 kHz. Question 3: The design runs fine, but the frequency is slightly off. What can be done to adjust this?
Answer: Changing the value of C1 has some effect on the oscillator frequency. If a SERIES resonant crystal is used, it will resonate at a different frequency than a PARALLEL resonant crystal of the same frequency call-out. Ensure that you are using a PARALLEL resonant crystal. Question 4: The board works fine, then suddenly quits or loses time.
Answer: Other than the obvious software checks that should be done to investigate losing time, it is possible that the amplitude of the oscillator output is not high enough to reliably trigger the oscillator input. Look at the C1 and C2 values and ensure that the device Configuration bits are correct for the desired oscillator mode. Question 5: If I put an oscilloscope probe on an oscillator pin, I dont see what I expect. Why?
Answer: Remember that an oscilloscope probe has capacitance. Connecting the probe to the oscillator circuitry will modify the oscillator characteristics. Consider using a low capacitance (active) probe.
DS70054D-page 7-38
This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Oscillator module are: Title
Application Note # AN588 AN606 AN826
PICmicro Microcontroller Oscillator Design Guide Low Power Design using PICmicro Microcontrollers Crystal Oscillator Basics and Crystal Selection for rfPIC and PICmicro Devices
7
Oscillator
Note:
DS70054D-page 7-39

Revision B
This revision incorporates technical content changes for the dsPIC30F Oscillator module.
Revision C
Revision D
This revision incorporates details on the three versions (VERSION 1, VERSION 2 and VERSION 3) of the Oscillator system implemented in dsPIC30F devices in the General Purpose, Sensor and Motor Control families.
DS70054D-page 7-40
Section 8. Reset
HIGHLIGHTS
This section of the manual contains the following topics: 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 8.11 8.12 8.13 8.14 Introduction .................................................................................................................... 8-2 Clock Source Selection at Reset ................................................................................... 8-5 POR: Power-on Reset ................................................................................................... 8-5 External Reset (EXTR) .................................................................................................. 8-7 Software Reset Instruction (SWR) ................................................................................. 8-7 Watchdog Time-out Reset (WDTR) ............................................................................... 8-7 Brown-out Reset (BOR) ................................................................................................. 8-8 Using the RCON Status Bits ........................................................................................ 8-10 Device Reset Times ..................................................................................................... 8-11 Device Start-up Time Lines .......................................................................................... 8-13 Special Function Register Reset States....................................................................... 8-16 Design Tips .................................................................................................................. 8-17 Related Application Notes............................................................................................ 8-18 Revision History ........................................................................................................... 8-19
8
Reset
DS70055C-page 8-1

8.1 Introduction
The Reset module combines all Reset sources and controls the device Master Reset Signal, SYSRST. The following is a list of device Reset sources: POR: Power-on Reset EXTR: Pin Reset (MCLR) SWR: RESET Instruction WDTR: Watchdog Timer Reset BOR: Brown-out Reset TRAPR: Trap Conflict Reset IOPR: Illegal Opcode Reset UWR: Uninitialized W Register Reset
A simplified block diagram of the Reset module is shown in Figure 8-1. Any active source of Reset will make the SYSRST signal active. Many registers associated with the CPU and peripherals are forced to a known Reset state. Most registers are unaffected by a Reset; their status is unknown on POR and unchanged by all other Resets. Note: Refer to the specific peripheral or CPU section of this manual for register Reset states.
All types of device Reset will set a corresponding status bit in the RCON register to indicate the type of Reset (see Register 8-1). A POR will clear all bits except for the POR and BOR bits (RCON<2:1>), which are set. The user may set or clear any bit at any time during code execution. The RCON bits only serve as status bits. Setting a particular Reset status bit in software will not cause a device Reset to occur. The RCON register also has other bits associated with the Low Voltage Detect module, Watchdog Timer, and device power saving states. The function of these bits is discussed in other sections of this manual. Figure 8-1: Reset System Block Diagram
RESET Instruction Glitch Filter MCLR Sleep or Idle WDT Module VDD Rise Detect VDD Brown-out Reset BOR BOREN POR SYSRST
Trap Conflict Illegal Opcode Uninitialized W Register
DS70055C-page 8-2
Section 8. Reset
Register 8-1: RCON: Reset Control Register R-0 BGST R/W-0 LVDEN R/W-0 R/W-1 R/W-0 LVDL<3:0> R/W-1 bit 8
Upper Byte: R/W-0 R/W-0 TRAPR IOPUWR bit 15
Lower Byte: R/W-0 EXTR bit 7 bit 15
R/W-0 SWR
R/W-0 SWDTEN
R/W-0 WDTO
R/W-0 SLEEP
R/W-0 IDLE
R/W-1 BOR
R/W-1 POR bit 0
TRAPR: Trap Reset Flag bit 1 = A Trap Conflict Reset has occurred 0 = A Trap Conflict Reset has not occurred IOPUWR: Illegal Opcode or Uninitialized W Access Reset Flag bit 1 = An illegal opcode detection, an illegal Address mode, or uninitialized W register used as an address pointer caused a Reset 0 = An illegal opcode or uninitialized W Reset has not occurred BGST: Bandgap Stable bit 1 = The bandgap has stabilized 0 = Bandgap is not stable and LVD interrupts should be disabled LVDEN: Low Voltage Detect Power Enable bit 1 = Enables LVD, powers up LVD circuit 0 = Disables LVD, powers down LVD circuit LVDL<3:0>: Low Voltage Detection Limit bits Refer to Section 9. Low Voltage Detect (LVD) for further details.
bit 14
bit 13
bit 12
8
Reset
bit 11-8 bit 7
EXTR: External Reset (MCLR) Pin bit 1 = A Master Clear (pin) Reset has occurred 0 = A Master Clear (pin) Reset has not occurred SWR: Software RESET (Instruction) Flag bit 1 = A RESET instruction has been executed 0 = A RESET instruction has not been executed SWDTEN: Software Enable/Disable of WDT bit 1 = WDT is turned on 0 = WDT is turned off Note: If FWDTEN fuse bit is 1 (unprogrammed), the WDT is ALWAYS ENABLED, regardless of the SWDTEN bit setting.
bit 6
bit 5
bit 4
WDTO: Watchdog Timer Time-out Flag bit 1 = WDT Time-out has occurred 0 = WDT Time-out has not occurred SLEEP: Wake From Sleep Flag bit 1 = Device has been in Sleep mode 0 = Device has not been in Sleep mode IDLE: Wake-up From Idle Flag bit 1 = Device was in Idle mode 0 = Device was not in Idle mode
bit 3
bit 2
DS70055C-page 8-3

Register 8-1: bit 1 RCON: Reset Control Register (Continued)
BOR: Brown-out Reset Flag bit 1 = A Brown-out Reset has occurred. Note that BOR is also set after Power-on Reset. 0 = A Brown-out Reset has not occurred POR: Power-on Reset Flag bit 1 = A Power-up Reset has occurred 0 = A Power-up Reset has not occurred Note: All of the Reset status bits may be set or cleared in software. Setting one of these bits in software does not cause a device Reset.
bit 0
DS70055C-page 8-4
Section 8. Reset
8.2 Clock Source Selection at Reset
If clock switching is enabled, the system clock source at device Reset is chosen as shown in Table 8-1. If clock switching is disabled, the system clock source is always selected according to the oscillator configuration fuses. Refer to Section 7. Oscillator for further details. Table 8-1: Oscillator Selection vs. Type of Reset (Clock Switching Enabled) Reset Type POR BOR EXTR WDTR SWR Clock Source Selected Based On Oscillator Configuration Fuses Oscillator Configuration Fuses COSC Control bits (OSCCON<13:12>) COSC Control bits (OSCCON<13:12>) COSC Control bits (OSCCON<13:12>)
8.3
POR: Power-on Reset

There are two threshold voltages associated with a Power-on Reset (POR). The first voltage is the device threshold voltage, VPOR. The device threshold voltage is the voltage at which the device logic circuits become operable. The second voltage associated with a POR event is the POR circuit threshold voltage which is nominally 1.85V. A power-on event will generate an internal Power-on Reset pulse when a VDD rise is detected. The Reset pulse will be generated at VPOR. The device supply voltage characteristics must meet specified starting voltage and rise rate requirements to generate the POR pulse. In particular, VDD must fall below VPOR before a new POR is initiated. For more information on the VPOR and the VDD rise rate specifications, please refer to the Electrical Specifications section of the device data sheet. The POR pulse will reset a POR timer and place the device in the Reset state. The POR also selects the device clock source identified by the oscillator configuration bits. After the Power-on Reset pulse is generated, the POR circuit inserts a small delay, TPOR, which is nominally 10 s and ensures that internal device bias circuits are stable. Furthermore, a user selected Power-up Time-out (TPWRT) may be applied. The TPWRT parameter is based on device configuration bits and can be 0 ms (no delay), 4 ms, 16 ms or 64 ms. The total delay time at device power-up is TPOR + TPWRT. When these delays have expired, SYSRST will be released on the next leading edge of the instruction cycle clock, and the PC will jump to the Reset vector. The timing for the SYSRST signal is shown in Figure 8-2. A Power-on Reset is initialized when VDD falls below a threshold voltage, VT. The POR delay time is inserted when VDD crosses the POR circuit threshold voltage. Finally, the PWRT delay time, TPWRT, is inserted before SYSRST is released. The power-on event will set the POR and BOR status bits (RCON<1:0>).
8
Reset
DS70055C-page 8-5

Figure 8-2: POR Module Timing Diagram for Rising VDD
POR Circuit Threshold Voltage VDD
VPOR Time Internal Power-on Reset pulse occurs at VPOR and begins POR delay time, TPOR. TPOR
POR Circuit
POR circuit is initialized at VPOR.
Time System Reset is released after Power-up Timer expires. TPWRT SYSRST (0 ms, 4 ms, 16 ms or 64 ms)
Time
Note:
When the device exits the Reset condition (begins normal operation), the device operating parameters (voltage, frequency, temperature, etc.) must be within their operating ranges, otherwise the device will not function correctly. The user must ensure that the delay between the time power is first applied and the time SYSRST becomes inactive is long enough to get all operating parameters within specification.
DS70055C-page 8-6
Section 8. Reset
8.3.1 Using the POR Circuit
To take advantage of the POR circuit, just tie the MCLR pin directly to VDD. This will eliminate external RC components usually needed to create a Power-on Reset delay. A minimum rise time for VDD is required. Refer to the Electrical Specifications section in the specific device data sheet for further details. Depending on the application, a resistor may be required between the MCLR pin and VDD. This resistor can be used to decouple the MCLR pin from a noisy power supply rail. The resistor will also be necessary if the device programming voltage, VPP, needs to be placed on the MCLR pin while the device is installed in the application circuit. VPP is 13 volts for most devices. Figure 8-3 shows a possible POR circuit for a slow power supply ramp up. The external Power-on Reset circuit is only required if the device would exit Reset before the device VDD is in the valid operating range. The diode, D, helps discharge the capacitor quickly when VDD powers down. Figure 8-3: External Power-on Reset Circuit (For Slow VDD Rise Time)
VDD D VDD R R1 MCLR C dsPIC30F
8
Note 1: The value of R should be low enough so that the voltage drop across it does not violate the VIH specification of the MCLR pin. 2: R1 will limit any current flowing into MCLR from external capacitor C in the event of MCLR/VPP pin breakdown, due to Electrostatic Discharge (ESD) or Electrical Overstress (EOS).
Reset
8.3.2
Power-up Timer (PWRT)

The PWRT provides an optional time delay (TPWRT) before SYSRST is released at a device POR or BOR (Brown-out Reset). The PWRT time delay is provided in addition to the POR delay time (TPOR). The PWRT time delay may be 0 ms, 4 ms, 16 ms or 64 ms nominal (see Figure 8-2). The PWRT delay time is selected using the FPWRT<1:0> configuration fuses in the FBORPOR Device Configuration register. Refer to Section 24. Device Configuration for further details.
8.4
External Reset (EXTR)

Whenever the MCLR pin is driven low, the device will asynchronously assert SYSRST, provided the input pulse on MCLR is longer than a certain minimum width. (Refer to the Electrical Specifications in the specific device data sheet for further details.) When the MCLR pin is released, SYSRST will be released on the next instruction clock cycle, and the Reset vector fetch will commence. The processor will maintain the existing clock source that was in use before the EXTR occurred. The EXTR status bit (RCON<7>) will be set to indicate the MCLR Reset.
8.5
Software RESET Instruction (SWR)

Whenever the RESET instruction is executed, the device will assert SYSRST, placing the device in a special Reset state. This Reset state will not re-initialize the clock. The clock source in effect prior to the RESET instruction will remain. SYSRST will be released at the next instruction cycle, and the Reset vector fetch will commence.
8.6
Watchdog Time-out Reset (WDTR)

Whenever a Watchdog time-out occurs, the device will asynchronously assert SYSRST. The clock source will remain unchanged. Note that a WDT time-out during Sleep or Idle mode will wake-up the processor, but NOT reset the processor. For more information, refer to Section 10. Watchdog Timer and Power Saving Modes.
DS70055C-page 8-7

8.7 Brown-out Reset (BOR)
The BOR (Brown-out Reset) module is based on an internal voltage reference circuit. The main purpose of the BOR module is to generate a device Reset when a brown-out condition occurs. Brown-out conditions are generally caused by glitches on the AC mains (i.e., missing waveform portions of the AC cycles due to bad power transmission lines), or voltage sags due to excessive current draw when a large load is energized. The BOR module allows selection of one of the following voltage trip points: VBOR = 2.0V VBOR = 2.7V VBOR = 4.2V VBOR = 4.5V Note: The BOR voltage trip points indicated here are nominal values provided for design guidance only. Refer to the Electrical Specifications in the specific device data sheet for BOR voltage limit specifications.
On a BOR, the device will select the system clock source based on the device configuration bit values (FPR<3:0>, FOS<1:0>). The PWRT time-out (TPWRT), if enabled, will be applied before SYSRST is released. If a crystal oscillator source is selected, the Brown-out Reset will invoke the Oscillator Start-up Timer (OST). The system clock is held until OST expires. If a system clock source is derived from the PLL, then the clock will be held until the LOCK bit (OSCCON<5>) is set. The BOR status bit (RCON<1>) will be set to indicate that a BOR has occurred. The BOR circuit, if enabled, will continue to operate while in Sleep or Idle modes and will reset the device should VDD fall below the BOR threshold voltage. Refer to the Electrical Specifications section of the appropriate device data sheet for the BOR electrical specifications. Typical brown-out scenarios are shown in Figure 8-4. As shown, a PWRT delay (if enabled) will be initiated each time VDD rises above the VBOR trip point. Figure 8-4:
VDD VBOR TPWRT SYSRST
Brown-out Situations
VDD VBOR TPWRT SYSRST VDD dips before PWRT expires VDD TPWRT SYSRST VBOR
DS70055C-page 8-8
Section 8. Reset
8.7.1 BOR Configuration
The BOR module is enabled/disabled and configured via device configuration fuses. The BOR module is enabled by default and may be disabled (to reduce power consumption) by programming the BOREN device configuration fuse to a 0 (FBORPOR<7>). The BOREN configuration fuse is located in the FBORPOR Device Configuration register. The BOR voltage trip point (VBOR) is selected using the BORV<1:0> configuration fuses (FBOR<5:4>). Refer to Section 24. Device Configuration for further details.
8.7.2
Current Consumption for BOR Operation

The BOR circuit relies on an internal voltage reference circuit that is shared with other peripheral devices, such as the Low Voltage Detect module. The internal voltage reference will be active whenever one of its associated peripherals is enabled. For this reason, the user may not observe the expected change in current consumption when the BOR is disabled.
8.7.3
Illegal Opcode Reset

A device Reset will be generated if the device attempts to execute an illegal opcode value that was fetched from program memory. The Illegal Opcode Reset function can prevent the device from executing program memory sections that are used to store constant data. To take advantage of the Illegal Opcode Reset, use only the lower 16 bits of each program memory section to store the data values. The upper 8 bits should be programmed with 0x3F, which is an illegal opcode value. If a device Reset occurs as a result of an illegal opcode value, the IOPUWR status bit (RCON<14>) will be set.
8
Reset
8.7.4
Uninitialized W Register Reset

The W register array (with the exception of W15) is cleared during all Resets and is considered uninitialized until written to. An attempt to use an uninitialized register as an address pointer will reset the device. Furthermore, the IOPUWR status bit (RCON<14>) will be set.
8.7.5
Trap Conflict Reset

A device Reset will occur whenever multiple hard trap sources become pending at the same time. The TRAPR status bit (RCON<15>) will be set. Refer to Section 6. Reset Interrupts for more information on Trap Conflict Resets.
DS70055C-page 8-9

8.8 Using the RCON Status Bits
The user can read the RCON register after any device Reset to determine the cause of the Reset. Note: The status bits in the RCON register should be cleared after they are read so that the next RCON register value after a device Reset will be meaningful.
Table 8-2 provides a summary of the Reset flag bit operation. Table 8-2: Reset Flag Bit Operation Flag Bit TRAPR (RCON<15>) IOPWR (RCON<14>) EXTR (RCON<7>) SWR (RCON<6>) WDTO (RCON<4>) SLEEP (RCON<3>) IDLE (RCON<2>) BOR (RCON<1>) POR (RCON<0>) Note: Set by: Trap conflict event Illegal opcode or uninitialized W register access MCLR Reset RESET instruction WDT time-out PWRSAV #SLEEP instruction PWRSAV #IDLE instruction POR, BOR POR POR POR POR POR PWRSAV instruction, POR POR POR Cleared by:
All RESET flag bits may be set or cleared by the user software.
DS70055C-page 8-10
Section 8. Reset
8.9 Device Reset Times
The Reset times for various types of device Reset are summarized in Table 8-3. Note that the system Reset signal, SYSRST, is released after the POR delay time and PWRT delay times expire. The time that the device actually begins to execute code will also depend on the system oscillator delays, which include the Oscillator Start-up Timer (OST) and the PLL lock time. The OST and PLL lock times occur in parallel with the applicable SYSRST delay times. The FSCM delay determines the time at which the FSCM begins to monitor the system clock source after the SYSRST signal is released. Table 8-3: Reset Type POR Reset Delay Times for Various Device Resets Clock Source EC, EXTRC, FRC, LPRC EC + PLL XT, HS, XTL, LP XT + PLL BOR EC, EXTRC, FRC, LPRC EC + PLL XT, HS, XTL, LP XT + PLL MCLR WDT Software Any Clock Any Clock Any clock SYSRST Delay TPOR + TPWRT TPOR + TPWRT TPOR + TPWRT TPOR + TPWRT TPWRT TPWRT TPWRT TPWRT System Clock Delay TLOCK TOST TOST + TLOCK TLOCK TOST TOST + TLOCK FSCM Delay TFSCM TFSCM TFSCM TFSCM TFSCM TFSCM Notes 1, 2 1, 2, 4, 5 1, 2, 3, 5 1, 2, 3, 4, 5 2 1, 2, 4, 5 1, 2, 3, 5 1, 2, 3, 4, 5
8
Reset
Illegal Opcode Any Clock Uninitialized W Any Clock Trap Conflict Any Clock
Note 1: TPOR = Power-on Reset delay (10 s nominal). 2: TPWRT = Additional power-up delay as determined by the FPWRT<1:0> configuration bits. This delay is 0 ms, 4 ms, 16 ms or 64 ms nominal. 3: TOST = Oscillator Start-up Timer. A 10-bit counter counts 1024 oscillator periods before releasing the oscillator clock to the system. 4: TLOCK = PLL lock time (20 s nominal). 5: TFSCM = Fail-Safe Clock Monitor delay (100 s nominal).
DS70055C-page 8-11

8.9.1 POR and Long Oscillator Start-up Times
The oscillator start-up circuitry and its associated delay timers is not linked to the device Reset delays that occur at power-up. Some crystal circuits (especially low frequency crystals) will have a relatively long start-up time. Therefore, one or more of the following conditions is possible after SYSRST is released: The oscillator circuit has not begun to oscillate. The oscillator start-up timer has NOT expired (if a crystal oscillator is used). The PLL has not achieved a LOCK (if PLL is used). The device will not begin to execute code until a valid clock source has been released to the system. Therefore, the oscillator and PLL start-up delays must be considered when the Reset delay time must be known.
8.9.2
Fail-Safe Clock Monitor (FSCM) and Device Resets

If the FSCM is enabled, it will begin to monitor the system clock source when SYSRST is released. If a valid clock source is not available at this time, the device will automatically switch to the FRC oscillator and the user can switch to the desired crystal oscillator in the Trap Service Routine.
8.9.2.1
FSCM Delay for Crystal and PLL Clock Sources When the system clock source is provided by a crystal oscillator and/or the PLL, a small delay, TFSCM, will automatically be inserted after the POR and PWRT delay times. The FSCM will not begin to monitor the system clock source until this delay expires. The FSCM delay time is nominally 100 s and provides additional time for the oscillator and/or PLL to stabilize. In most cases, the FSCM delay will prevent an oscillator failure trap at a device Reset when the PWRT is disabled.
DS70055C-page 8-12
Section 8. Reset
8.10 Device Start-up Time Lines
Figure 8-5 through Figure 8-8 show graphical time lines of the delays associated with device Reset for several operating scenarios. Figure 8-5 shows the delay time line when a crystal oscillator and PLL are used as the system clock and the PWRT is disabled. The internal Power-on Reset pulse occurs at the VPOR threshold. A small POR delay occurs after the internal Reset pulse. (The POR delay is always inserted before device operation begins.) The FSCM, if enabled, begins to monitor the system clock for activity when the FSCM delay expires. Figure 8-5 shows that the oscillator and PLL delays expire before the Fail-Safe Clock Monitor (FSCM) is enabled. However, it is possible that these delays may not expire until after FSCM is enabled. In this case, the FSCM would detect a clock failure and a clock failure trap will be generated. If the FSCM delay does not provide adequate time for the oscillator and PLL to stabilize, the PWRT could be enabled to allow more delay time before device operation begins and the FSCM starts to monitor the system clock. Figure 8-5: Device Reset Delay, Crystal + PLL Clock Source, PWRT Disabled
VDD
POR Circuit Threshold Voltage
Internal Power-on Reset Pulse POR System Reset released. TPOR
8
Reset
SYSRST
Oscillator Oscillator released to system, device operation begins.
TOST OSC Delay
TLOCK
System OSC FSCM enabled. TFSCM FSCM
Note 1: Delay times shown are not drawn to scale. 2: FSCM, if enabled, monitors system clock at expiration of TPOR + TFSCM. 3: TLOCK not inserted when PLL is disabled.
DS70055C-page 8-13

The Reset time line shown in Figure 8-6 is similar to that shown in Figure 8-5, except that the PWRT has been enabled to increase the amount of delay time before SYSRST is released. The FSCM, if enabled, will begin to monitor the system clock after TFSCM expires. Note that the additional PWRT delay time added to TFSCM provides ample time for the system clock source to stabilize in most cases. Figure 8-6: Device Reset Delay, Crystal + PLL Clock Source, PWRT Enabled
VDD
Internal Power-on Reset Pulse POR
Device operation begins.
TPOR SYSRST
TPWRT
Oscillator released to system. OSC Delay TOST TLOCK
TFSCM FSCM TFSCM short compared to TPWRT.
Note 1: Delay times shown are not drawn to scale. 2: FSCM, if enabled, monitors system clock at expiration of TPOR + TPWRT + TFSCM. 3: TLOCK not inserted when PLL is is disabled.
DS70055C-page 8-14
Section 8. Reset
The Reset time line in Figure 8-7 shows an example when an EC + PLL clock source is used as the system clock and the PWRT is enabled. This example is similar to the one shown in Figure 8-6, except that the oscillator start-up timer delay, TOST, does not occur. Figure 8-7: Device Reset Delay, EC + PLL Clock, PWRT Enabled
VDD
Internal Power-on Reset Pulse POR
Device operation begins.
TPOR SYSRST
TPWRT
Oscillator released to system. OSC Delay TLOCK
TFSCM FSCM TFSCM short compared to TPWRT.
8
Reset
Note 1: Delay times shown are not drawn to scale. 2: FSCM, if enabled, monitors system clock at expiration of TPOR + TPWRT + TFSCM. 3: TLOCK not inserted when PLL is is disabled.
DS70055C-page 8-15

The Reset time line shown in Figure 8-8 shows an example where an EC without PLL, or RC system clock source is selected and the PWRT is disabled. Note that this configuration provides minimal Reset delays. The POR delay is the only delay time that occurs before device operation begins. No FSCM delay will occur if the FSCM is enabled, because the system clock source is not derived from a crystal oscillator or the PLL. Figure 8-8: Device Reset Delay, EC or RC Clock, PWRT Disabled
VDD
Internal Power-on Reset Pulse POR TPOR SYSRST Oscillator released to system. OSC Delay System Reset released.
FSCM
Note 1: Delay times shown are not drawn to scale. 2: If enabled, FSCM will begin to monitor system clock at expiration of TPOR.
8.11
Special Function Register Reset States

Most of the special function registers (SFRs) associated with the dsPIC30F CPU and peripherals are reset to a particular value at a device Reset. The SFRs are grouped by their peripheral or CPU function and their Reset values are specified in each section of this manual. The Reset value for each SFR does not depend on the type of Reset, with the exception of two registers. The Reset value for the Reset Control register, RCON, will depend on the type of device Reset. The Reset value for the Oscillator Control register, OSCCON, will depend on the type of Reset and the programmed values of the oscillator configuration bits in the FOSC Device Configuration register (see Table 8-1).
DS70055C-page 8-16
Section 8. Reset
8.12 Design Tips
Question 1: How do I use the RCON register?
Answer: The initialization code after a Reset should examine RCON and confirm the source of the Reset. In certain applications, this information can be used to take appropriate action to correct the problem that caused the Reset to occur. All Reset status bits in the RCON register should be cleared after reading them to ensure the RCON value will provide meaningful results after the next device Reset. Question 2: How should I use BOR in a battery operated application?
Answer: The BOR feature is not designed to operate as a low battery detect, and should be disabled in battery operated systems (to save current). The Low Voltage Detect peripheral can be used to detect when the battery has reached its end of life voltage. Question 3: The BOR module does not have the programmable trip points that my application needs. How can I work around this?
Answer: There are some applications where the devices programmable BOR trip point levels may still not be at the desired level for the application. Figure 8-9 shows a possible circuit for external brown-out protection, using the MCP100 system supervisor. Figure 8-9: External Brown-out Protection Using the MCP100
VDD VDD MCP100 VSS RST MCLR dsPIC30F
8
Reset
Question 4:
I initialized a W register with a 16-bit address, but the device appears to reset when I attempt to use the register as an address.
Answer: Because all data addresses are 16 bit values, the uninitialized W register logic only recognizes that a register has been initialized correctly if it was subjected to a word load. Two byte moves to a W register, even if successive, will not work, resulting in a device Reset if the W register is used as an address pointer in an operation.
DS70055C-page 8-17

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Reset module are: Title Power-up Trouble Shooting Power-up Considerations Application Note # AN607 AN522
Note:
DS70055C-page 8-18
Section 8. Reset
Revision B
Revision C
8
Reset
DS70055C-page 8-19

NOTES:
DS70055C-page 8-20
Section 9. Low Voltage Detect (LVD)

HIGHLIGHTS
This section of the manual contains the following topics: 9.1 9.2 9.3 9.4 9.5 Introduction .................................................................................................................... 9-2 LVD Operation ............................................................................................................... 9-5 Design Tips .................................................................................................................... 9-6 Related Application Notes.............................................................................................. 9-7 Revision History ............................................................................................................. 9-8
9
Low Voltage Detect (LVD)
DS70056C-page 9-1

9.1 Introduction
The LVD module is applicable to battery operated applications. As the battery drains its energy, the battery voltage slowly drops. The battery source impedance also increases as it loses energy. The LVD module is used to detect when the battery voltage (and therefore, the VDD of the device) drops below a threshold, which is considered near the end of battery life for the application. This allows the application to gracefully shutdown its operation. The LVD module uses an internal reference voltage for comparison. The threshold voltage, VLVD, is programmable during run-time. Figure 9-1 shows a possible application battery voltage curve. Over time, the device voltage decreases. When the device voltage equals voltage VLVD, the LVD logic generates an interrupt. This occurs at time TA. The application software then has until the device voltage is no longer in valid operating range to shutdown the system. Voltage point VB is the minimum valid operating voltage specification. This gives a time TB. The total time for shutdown is TB TA. Figure 9-1: Typical Low Voltage Detect Application
VLVD VMIN Voltage
Time
TA
TB
Legend: VLVD = LVD trip point VMIN = Minimum valid device operating voltage
DS70056C-page 9-2
Section 9. Low Voltage Detect

Figure 9-2 shows the block diagram for the LVD module. A comparator uses an internally generated reference voltage as the set point. When the selected tap output of the device voltage is lower than the reference voltage, the LVDIF bit (IFS2<10>) is set. Each node in the resistor divider represents a trip point voltage. This voltage is software programmable to any one of 16 values. Figure 9-2: Low Voltage Detect (LVD) Block Diagram
External LVD Input pin VDD LVDIN 4 LVDL<3:0>
16 to 1 MUX
LVDIF
LVDEN
Internally Generated Reference Voltage
9.1.1
LVD Control Bits

The LVD module control bits are located in the RCON register. The LVDEN bit (RCON<12>) enables the Low Voltage Detect module. The LVD module is enabled when LVDEN = 1. If power consumption is important, the LVDEN bit can be cleared for maximum power savings.
9
9.1.1.1
LVD Trip Point Selection The LVDL<3:0> bits (RCON<11:8>) will choose the LVD trip point. There are 15 trip point options that may be selected from the internal voltage divider connected to VDD. If none of the trip point options are suitable for the application, there is one option that allows the LVD sample voltage to be applied externally on the LVDIN pin. (Refer to the specific device data sheet for the pin location.) The nominal trip point voltage for the external LVD input is 1.24 volts. The LVD external input option requires that the user select values for an external voltage divider circuit that will generate a LVD interrupt at the desired VDD.
9.1.2
Internal Voltage Reference

The LVD uses an internal bandgap voltage reference circuit that requires a nominal amount of time to stabilize. Refer to the Electrical Specifications in the specific device data sheet for details. The BGST status bit (RCON<13>) indicates when the bandgap voltage reference has stabilized. The user should poll the BGST status bit in software after the LVD module is enabled. At the end of the stabilization time, the LVDIF bit (IFS2<10>) should be cleared. Refer to the LVD module setup procedure in Section 9.2 LVD Operation. The bandgap voltage reference circuit can also be used by other peripherals on the device so it may already be active (and stabilized) prior to enabling the LVD module.
DS70056C-page 9-3

Register 9-1: RCON: Reset Control Register R-0 BGST R/W-0 LVDEN R/W-0 R/W-1 R/W-0 LVDL<3:0> R/W-1 bit 8 Upper Byte: R/W-0 R/W-0 TRAPR IOPUWR bit 15
Lower Byte: R/W-0 EXTR bit 7 bit 13
R/W-0 SWR
R/W-0 SWDTEN
R/W-0 WDTO
R/W-0 SLEEP
R/W-0 IDLE
R/W-1 BOR
R/W-1 POR bit 0
BGST: Bandgap Stable bit 1 = The bandgap has stabilized 0 = Bandgap is not stable and LVD interrupts should be disabled LVDEN: Low Voltage Detect Power Enable bit 1 = Enables LVD, powers up LVD circuit 0 = Disables LVD, powers down LVD circuit LVDL<3:0>: Low Voltage Detection Limit bits 1111 = Input to LVD is the LVDIN pin (1.24V threshold, nominal) 1110 = 4.6V 1101 = 4.3V 1100 = 4.1V 1011 = 3.9V 1010 = 3.7V 1001 = 3.6V 1000 = 3.4V 0111 = 3.1V 0110 = 2.9V 0101 = 2.8V (default value at Reset) 0100 = 2.6V 0011 = 2.5V 0010 = 2.3V 0001 = 2.1V 0000 = 1.9V Note: The voltage threshold values shown here are provided for design guidance only. Refer to the Electrical Specifications in the device data sheet for further details.
bit 12
bit 11-8
Note:
See Section 8. Reset for a description of other bits in the RCON register.
DS70056C-page 9-4

9.2 LVD Operation
The LVD module adds robustness to the application because the device can monitor the state of the device voltage. When the device voltage enters a voltage window near the lower limit of the valid operating voltage range, the device can save values to ensure a clean shutdown. Note: The system design should ensure that the application software is given adequate time to save values before the device exits the valid operating range, or is forced into a Brown-out Reset.
Depending on the power source for the device, the supply voltage may decrease relatively slowly. This means that the LVD module does not need to be constantly operating. To decrease the current requirements, the LVD circuitry only needs to be enabled for short periods where the voltage is checked. After doing the check, the LVD module may be disabled.
9.2.1
LVD Initialization Steps

The following steps are required to setup the LVD module: 1. If the external LVD input pin is used (LVDIN), ensure that all other peripherals multiplexed on the pin are disabled and the pin is configured as an input by setting the appropriate bit in the TRISx registers. Write the desired value to the LVDL control bits (RCON<11:8>), which selects the desired LVD threshold voltage. Ensure that LVD interrupts are disabled by clearing the LVDIE bit (IEC2<10>). Enable the LVD module by setting the LVDEN bit (RCON<12>). Wait for the internal voltage reference to become stable by polling the BGST status bit (RCON<13>), if required (see Section 9.1.2 Internal Voltage Reference). Ensure that the LVDIF bit (IFS2<10>) is cleared before interrupts are enabled. If LVDIF is set, the device VDD may be below the chosen LVD threshold voltage. Set LVD interrupts to the desired CPU priority level by writing the LVDIP<2:0> control bits (IPC10<10:8>). Enable LVD interrupts by setting the LVDIE control bit.
2. 3. 4. 5. 6. 7. 8.
Once the VDD has fallen below the programmed LVD threshold, the LVDIF bit will remain set. When the LVD module has interrupted the CPU, one of two actions may be taken in the ISR: 1. or 2. Decrease the LVD voltage threshold using the LVDL control bits and clear the LDVIF status bit. This technique can be used to track a gradually decreasing battery voltage. Clear the LVDIE control bit to disable further LVD module interrupts and take the appropriate shutdown procedures.
9
9.2.2
Current Consumption for LVD Operation

The LVD circuit relies on an internal voltage reference circuit that is shared with other peripheral devices, such as the Brown-out Reset (BOR) module. The internal voltage reference will be active whenever one of its associated peripherals is enabled. For this reason, the user may not observe the expected change in current consumption when the LVD module is disabled.
9.2.3
Operation in Sleep and Idle Mode

When enabled, the LVD circuitry continues to operate during Sleep or Idle modes. If the device voltage crosses the trip point, the LVDIF bit will be set. The criteria for exiting from Sleep or Idle modes are as follows: If the LVDIE bit (IEC2<10>) is set, the device will wake from Sleep or Idle mode. If the assigned priority for the LVD interrupt is less than or equal to the current CPU priority, the device will wake-up and continue code execution from the instruction following the PWRSAV instruction that initiated the Sleep or Idle mode. If the assigned priority level for the LVD interrupt is greater than the current CPU priority, the device will wake-up and the CPU exception process will begin. Code execution will continue from the first instruction of the LVD ISR.
DS70056C-page 9-5

9.3 Design Tips
Question 1: The LVD circuitry seems to be generating random interrupts?
Answer: Ensure that the internal voltage reference is stable before enabling the LVD interrupt. This is done by polling the BGST status bit (RCON<13>) after the LVD module is enabled. After this time delay, the LVDIF bit should be cleared and then, the LVDIE bit may be set. Question 2: How can I reduce the current consumption of the module?
Answer: Low Voltage Detect is used to monitor the device voltage. The power source is normally a battery that ramps down slowly. This means that the LVD circuity can be disabled for most of the time, and only enabled occasionally to do the device voltage check. Question 3: Should I enable the BOR circuit for a battery powered application?
Answer: The BOR circuit is intended to protect the device from improper operation due to power supply fluctuations caused by the AC line voltage. The BOR is typically not required for battery applications and can be disabled for lower current consumption.
DS70056C-page 9-6

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Low Voltage Detect module are: Title No related application notes at this time. Application Note #
Note:
9
DS70056C-page 9-7

Revision B
Revision C
DS70056C-page 9-8
Section 10. Watchdog Timer and Power Saving Modes

HIGHLIGHTS
This section of the manual contains the following topics: 10.1 Introduction .................................................................................................................. 10-2 10.2 Power Saving Modes ................................................................................................... 10-2 10.3 Sleep Mode.................................................................................................................. 10-2 10.4 Idle Mode ..................................................................................................................... 10-4 10.5 Interrupts Coincident with Power Save Instructions..................................................... 10-5 10.6 Watchdog Timer........................................................................................................... 10-6 10.7 Peripheral Module Disable (PMD) Registers ............................................................... 10-9 10.8 Design Tips ................................................................................................................ 10-10 10.9 Related Application Notes.......................................................................................... 10-11 10.10 Revision History ......................................................................................................... 10-12
10
WDT and Power Saving Modes
DS70057D-page 10-1

10.1 Introduction
This section addresses the Watchdog Timer (WDT) and Power Saving modes of the dsPIC30F device family. The dsPIC DSC devices have two reduced Power modes that can be entered through execution of the PWRSAV instruction: Sleep Mode: The CPU, system clock source, and any peripherals that operate on the system clock source are disabled. This is the lowest Power mode for the device. Idle Mode: The CPU is disabled, but the system clock source continues to operate. Peripherals continue to operate, but can optionally be disabled. The WDT, when enabled, operates from the internal LPRC clock source and can be used to detect system software malfunctions by resetting the device if the WDT has not been cleared in software. Various WDT time-out periods can be selected using the WDT postscaler. The WDT can also be used to wake the device from Sleep or Idle mode.
10.2
Power Saving Modes

The dsPIC30F device family has two special Power Saving modes, Sleep mode and Idle mode, that can be entered through the execution of a special PWRSAV instruction. The assembly syntax of the PWRSAV instruction is as follows:
PWRSAV #SLEEP_MODE PWRSAV #IDLE_MODE ; Put the device into SLEEP mode ; Put the device into IDLE mode
Note:
SLEEP_MODE and IDLE_MODE are constants defined in the assembler include file for the selected device.
The Power Saving modes can be exited as a result of an enabled interrupt, WDT time-out, or a device Reset. When the device exits one of these two Operating modes, it is said to wake-up. The characteristics of the Power Saving modes are described in subsequent sections.
10.3
Sleep Mode
The characteristics of Sleep mode are as follows: The system clock source is shutdown. If an on-chip oscillator is used, it is turned off. The device current consumption will be at a minimum provided that no I/O pin is sourcing current. The Fail-Safe Clock Monitor (FSCM) does not operate during Sleep mode since the system clock source is disabled. The LPRC clock will continue to run in Sleep mode if the WDT is enabled. The Low Voltage Detect circuit, if enabled, remains operative during Sleep mode. The BOR circuit, if enabled, remains operative during Sleep mode. The WDT, if enabled, is automatically cleared prior to entering Sleep mode. Some peripherals may continue to operate in Sleep mode. These peripherals include I/O pins that detect a change in the input signal, or peripherals that use an external clock input. Any peripheral that is operating on the system clock source will be disabled in Sleep mode. The processor will exit, or wake-up, from Sleep on one of the following events: On any interrupt source that is individually enabled On any form of device Reset On a WDT time-out
10.3.1
Clock Selection on Wake-up from Sleep

The processor will restart the same clock source that was active when Sleep mode was entered.
DS70057D-page 10-2
Section 10. WDT and Power Saving Modes

10.3.2 Delay on Wake-up from Sleep
The power-up and oscillator start-up delays associated with waking up from Sleep mode are shown in Table 10-1. In all cases, the POR delay time (TPOR = 10 s nominal) is applied to allow internal device circuits to stabilize before the internal system Reset signal, SYSRST, is released. Table 10-1: Delay Times for Exit from Sleep Mode SYSRST Delay TPOR TPOR TPOR TPOR TPOR TPOR TPOR Oscillator Delay TLOCK TOST + TLOCK TOST TOST FSCM Delay TFSCM TFSCM TFSCM TFSCM 1 1, 3, 4 1, 2, 3, 4 1, 2, 4 1, 2, 4 1 1 Notes
Clock Source EC, EXTRC EC + PLL XT + PLL XT, HS, XTL LP (OFF during Sleep) LP (ON during Sleep) FRC, LPRC
Note 1: TPOR = Power-on Reset delay (10 s nominal). 2: TOST = Oscillator Start-up Timer. A 10-bit counter counts 1024 oscillator periods before releasing the oscillator clock to the system. 3: TLOCK = PLL lock time (20 s nominal). 4: TFSCM = Fail-Safe Clock Monitor delay (100 s nominal). Note: Please refer to the Electrical Specifications section of the dsPIC30F device data sheet for TPOR, TFSCM and TLOCK specification values.
10.3.3
Wake-up from Sleep Mode with Crystal Oscillator or PLL

If the system clock source is derived from a crystal oscillator and/or the PLL, then the Oscillator Start-up Timer (OST) and/or PLL lock times must be applied before the system clock source is made available to the device. As an exception to this rule, no oscillator delays are necessary if the system clock source is the LP oscillator and it was running while in Sleep mode. Note that in spite of various delays applied, the crystal oscillator (and PLL) may not be up and running at the end of the POR delay.
10.3.4
FSCM Delay and Sleep Mode

If the following conditions are true, a nominal 100 s delay (TFSCM) will be applied after the POR delay expires when waking from Sleep mode: The oscillator was shutdown while in Sleep mode. The system clock is derived from a crystal oscillator source and/or the PLL. The FSCM delay provides time for the OST to expire and the PLL to stabilize before device execution resumes in most cases. If the FSCM is enabled, it will begin to monitor the system clock source after the FSCM delay expires.
10.3.5
Slow Oscillator Start-up

The OST and PLL lock times may not have expired when the power-up delays have expired. If the FSCM is enabled, then the device will detect this condition as a clock failure and a clock fail trap will occur. The device will switch to the FRC oscillator and the user can re-enable the crystal oscillator source in the clock failure Trap Service Routine. If FSCM is NOT enabled, then the device will simply not start executing code until the clock is stable. From the users perspective, the device will appear to be in Sleep until the oscillator clock has started.
10
DS70057D-page 10-3

10.3.6 Wake-up from Sleep on Interrupt
User interrupt sources that are assigned to CPU priority level 0 cannot wake the CPU from Sleep mode, because the interrupt source is effectively disabled. To use an interrupt as a wake-up source, the CPU priority level for the interrupt must be assigned to CPU priority level 1 or greater. Any source of interrupt that is individually enabled, using its corresponding IE control bit in the IECx registers, can wake-up the processor from Sleep mode. When the device wakes from Sleep mode, one of two actions may occur: If the assigned priority for the interrupt is less than or equal to the current CPU priority, the device will wake-up and continue code execution from the instruction following the PWRSAV instruction that initiated Sleep mode. If the assigned priority level for the interrupt source is greater than the current CPU priority, the device will wake-up and the CPU exception process will begin. Code execution will continue from the first instruction of the ISR. The Sleep status bit (RCON<3>) is set upon wake-up.
10.3.7
Wake-up from Sleep on Reset

All sources of device Reset will wake the processor from Sleep mode. Any source of Reset (other than a POR) that wakes the processor will set the Sleep status bit (RCON<3>) to indicate that the device was previously in Sleep mode. On a Power-on Reset, the Sleep bit is cleared.
10.3.8
Wake-up from Sleep on Watchdog Time-out

If the Watchdog Timer (WDT) is enabled and expires while the device is in Sleep mode, the processor will wake-up. The Sleep and WDTO status bits (RCON<3>, RCON<4>) are both set to indicate that the device resumed operation due to the WDT expiration. Note that this event does not reset the device. Operation continues from the instruction following the PWRSAV instruction that initiated Sleep mode.
10.4
Idle Mode
User interrupt sources that are assigned to CPU priority level 0 cannot wake the CPU from Idle mode, because the interrupt source is effectively disabled. To use an interrupt as a wake-up source, the CPU priority level for the interrupt must be assigned to CPU priority level 1 or greater. When the device enters Idle mode, the following events occur: The CPU will stop executing instructions. The WDT is automatically cleared. The system clock source will remain active and peripheral modules, by default, will continue to operate normally from the system clock source. Peripherals can optionally be shutdown in Idle mode using their stop-in-idle control bit. (See peripheral descriptions for further details.) If the WDT or FSCM is enabled, the LPRC will also remain active. The processor will wake from Idle mode on the following events: On any interrupt that is individually enabled. On any source of device Reset. On a WDT time-out. Upon wake-up from Idle, the clock is re-applied to the CPU and instruction execution begins immediately starting with the instruction following the PWRSAV instruction, or the first instruction in the ISR.
DS70057D-page 10-4

10.4.1 Wake-up from Idle on Interrupt
Any source of interrupt that is individually enabled using the corresponding IE control bit in the IECx register and exceeds the current CPU priority level, will be able to wake-up the processor from Idle mode. When the device wakes from Idle mode, one of two options may occur: If the assigned priority for the interrupt is less than or equal to the current CPU priority, the device will wake-up and continue code execution from the instruction following the PWRSAV instruction that initiated Idle mode. If the assigned priority level for the interrupt source is greater than the current CPU priority, the device will wake-up and the CPU exception process will begin. Code execution will continue from the first instruction of the ISR. The Idle status bit (RCON<2>) is set upon wake-up.
10.4.2
Wake-up from Idle on Reset

Any Reset, other than a POR, will wake the CPU from Idle mode. On any device Reset, except a POR, the Idle status bit is set (RCON<2>) to indicate that the device was previously in Idle mode. In a Power-on Reset, the Idle bit is cleared.
10.4.3
Wake-up from Idle on WDT Time-out

If the WDT is enabled, then the processor will wake from Idle mode on a WDT time-out and continue code execution with the instruction following the PWRSAV instruction that initiated Idle mode. Note that the WDT time-out does not reset the device in this case. The WDTO and Idle status bits (RCON<4>, RCON<2>) will both be set.
10.4.4
Time Delays on Wake from Idle Mode

Unlike a wake-up from Sleep mode, there are no time delays associated with wake-up from Idle mode. The system clock is running during Idle mode, therefore, no start-up times are required at wake-up.
10.5
Interrupts Coincident with Power Save Instructions

Any interrupt that coincides with the execution of a PWRSAV instruction will be held off until entry into Sleep or Idle mode has completed. The device will then wake-up from Sleep or Idle mode.
10
DS70057D-page 10-5

10.6 Watchdog Timer
The primary function of the Watchdog Timer (WDT) is to reset the processor in the event of a software malfunction. The WDT is a free running timer, which runs on the internal LPRC oscillator requiring no external components. Therefore, the WDT timer will continue to operate even if the system clock source (e.g., the crystal oscillator) fails. A block diagram of the WDT is shown in Figure 10-1. Figure 10-1: WDT Block Diagram
SWDTEN FWDTEN 2
Enable WDT LPRC Control LPRC Oscillator Wake-up from Sleep WDT Overflow Reset 8-bit Watchdog Timer Reset WDT Programmable Prescaler B 1:1, 1:2, 1:3, 1:15, 1:16 4 512 kHz 4 FWC = 128 kHz
FWPSA1 FWPSA0
Programmable Prescaler A 1:1, 1:8, 1:64, 1:512
FWPSB3 FWPSB2 FWPSB1 FWPSB0
CLRWDT Instr. PWRSAV Instr. All Device Resets Sleep or Idle State
10.6.1
Enabling and Disabling the WDT

The WDT is enabled or disabled by the FWDTEN device configuration bit in the FWDT Device Configuration register. The FWDT Configuration register values are written during device programming. When the FWDTEN configuration bit is set, the WDT is enabled. This is the default value for an erased device. Refer to Section 24. Device Configuration for further details on the FWDT Device Configuration register.
10.6.1.1
Software Controlled WDT If the FWDTEN device configuration bit is set, then the WDT is always enabled. However, the WDT can be optionally controlled in the user software when the FWDTEN configuration bit has been programmed to 0. The WDT is enabled in software by setting the SWDTEN control bit (RCON<5>). The SWDTEN control bit is cleared on any device Reset. The software WDT option allows the user to enable the WDT for critical code segments and disable the WDT during non-critical segments for maximum power savings.
DS70057D-page 10-6

10.6.2 WDT Operation
If enabled, the WDT will increment until it overflows or times out. A WDT time-out will force a device Reset, except during Sleep or Idle modes. To prevent a WDT Time-out Reset, the user must periodically clear the Watchdog Timer using the CLRWDT instruction. The CLRWDT instruction also clears the WDT prescalers. If the WDT times out during Sleep or Idle modes, the device will wake-up and continue code execution from where the PWRSAV instruction was executed. In either case, the WDTO bit (RCON<4>) will be set to indicate that the device Reset or wake-up event was due to a WDT time-out. If the WDT wakes the CPU from Sleep or Idle mode, the Sleep status bit (RCON<3>), or Idle status bit (RCON<2>) will also be set to indicate that the device was previously in a Power Saving mode.
10.6.3
WDT Timer Period Selection

The WDT clock source is the internal LPRC oscillator, which has a nominal frequency of 512 kHz. The LPRC clock is further divided by 4 to provide a 128 kHz clock to the WDT. The counter for the WDT is 8-bits wide, so the nominal time-out period for the WDT (TWDT) is 2 milliseconds.
10.6.3.1
WDT Prescalers The WDT has two clock prescalers, Prescaler A and Prescaler B, to allow a wide variety of time-out periods. Prescaler A can be configured for 1:1, 1:8, 1:64 or 1:512 divide ratios. Prescaler B can be configured for any divide ratio from 1:1 through 1:16. Time-out periods that range between 2 ms and 16 seconds (nominal) can be achieved using the prescalers. The prescaler settings are selected using the FWPSA<1:0> (Prescaler A) and FWPSB<3:0> (Prescaler B) configuration bits in the FWDT Device Configuration register. The FWPSA<1:0> and FWPSB<3:0> values are written during device programming. For more information on the WDT prescaler configuration bits, please refer to Section 24. Device Configuration. The time-out period of the WDT is calculated as follows: Equation 10-1: WDT Time-out Period WDT Period = 2 ms Prescale A Prescale B
Note:
The WDT time-out period is directly related to the frequency of the LPRC oscillator. The frequency of the LPRC oscillator will vary as a function of device operating voltage and temperature. Please refer to the specific dsPIC30F device data sheet for LPRC clock frequency specifications.
10
DS70057D-page 10-7

Table 10-2 shows time-out periods for various prescaler selections: Table 10-2: Prescaler B Value WDT Time-out Period vs. Prescale A and Prescale B Settings Prescaler A Value 1 8 64 128 256 384 512 640 768 896 1024 1152 1280 1408 1536 1664 1792 1920 2048 512 1024 2048 3072 4096 5120 6144 7168 8192 9216 10240 11264 12288 13312 14336 15360 16384
1 2 16 2 4 32 3 6 48 4 8 64 5 10 80 6 12 96 7 14 112 8 16 128 9 18 144 10 20 160 11 22 176 12 24 192 13 26 208 14 28 224 15 30 240 16 32 256 Note: All time values are in milliseconds.
10.6.4
Resetting the Watchdog Timer

The WDT and all its prescalers are reset: On ANY device Reset When a PWRSAV instruction is executed (i.e., Sleep or Idle mode is entered) By a CLRWDT instruction during normal execution
10.6.5
Operation of WDT in Sleep and Idle Modes

If the WDT is enabled, it will continue to run during Sleep or Idle modes. When the WDT time-out occurs, the device will wake the device and code execution will continue from where the PWRSAV instruction was executed. The WDT is useful for low power system designs, because it can be used to periodically wake the device from Sleep mode to check system status and provide action if necessary. Note that the SWDTEN bit is very useful in this respect. If the WDT is disabled during normal operation (FWDTEN = 0), then the SWDTEN bit (RCON<5>) can be used to turn on the WDT just before entering Sleep mode.
DS70057D-page 10-8

10.7 Peripheral Module Disable (PMD) Registers
The Peripheral Module Disable (PMD) registers provide a method to disable a peripheral module by stopping all clock sources supplied to that module. When a peripheral is disabled via the appropriate PMD control bit, the peripheral is in a minimum power consumption state. The control and status registers associated with the peripheral will also be disabled so writes to those registers will have no effect and read values will be invalid. A peripheral module will only be enabled if both the associated bit in the the PMD register is cleared and the peripheral is supported by the specific dsPIC variant. If the peripheral is present in the device, it is enabled in the PMD register by default. Note: If a PMD bit is set, the corresponding module is disabled after a delay of 1 instruction cycle. Similarly, if a PMD bit is cleared, the corresponding module is enabled after a delay of 1 instruction cycle (assuming the module control registers are already configured to enable module operation).
Please check individual device data sheet for specific operational details of the PMD register.
10
DS70057D-page 10-9

10.8 Design Tips
Question 1: The device resets even though I have inserted a CLRWDT instruction in my main software loop.
Answer: Make sure that the software loop that contains the CLRWDT instruction meets the minimum specification of the WDT (not the typical value). Also, make sure that interrupt processing time has been accounted for. Question 2: What should my software do before entering Sleep or Idle mode?
Answer: Make sure that the sources intended to wake the device have their IE bits set. In addition, make sure that the particular source of interrupt has the ability to wake the device. Some sources do not function when the device is in Sleep mode. If the device is to be placed in Idle mode, make sure that the stop-in-idle control bit for each device peripheral is properly set. These control bits determine whether the peripheral will continue operation in Idle mode. See the individual peripheral sections of this manual for further details. Question 3: How do I tell which peripheral woke the device from Sleep or Idle mode?
Answer: You can poll the IF bits for each enabled interrupt source to determine the source of wake-up.
DS70057D-page 10-10

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Watchdog Timer and Power Saving Modes are: Title Low Power Design using PICmicro Microcontrollers Application Note # AN606
Note:
10
DS70057D-page 10-11

Revision B
Revision C
Revision D
Section 10.7, Peripheral Module Disable (PMD) Registers, has been added.
DS70057D-page 10-12

NOTES:
10
DS70057D-page 10-13
11
I/O Ports
Section 11. I/O Ports

HIGHLIGHTS
This section of the manual contains the following topics: 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 Introduction .................................................................................................................. 11-2 I/O Port Control Registers............................................................................................ 11-3 Peripheral Multiplexing................................................................................................. 11-4 Port Descriptions.......................................................................................................... 11-6 Change Notification (CN) Pins ..................................................................................... 11-7 CN Operation in Sleep and Idle Modes ....................................................................... 11-8 Related Application Notes...........................................................................................11-11 Revision History ......................................................................................................... 11-12
DS70058D-page 11-1
11
11.1 Introduction
This section provides information on the I/O ports for the dsPIC30F family of devices. All of the device pins (except VDD, VSS, MCLR, and OSC1/CLKI) are shared between the peripherals and the general purpose I/O ports. The general purpose I/O ports allow the dsPIC30F to monitor and control other devices. Most I/O pins are multiplexed with alternate function(s). The multiplexing will depend on the peripheral features on the device variant. In general, when a peripheral is functioning, that pin may not be used as a general purpose I/O pin. Figure 11-1 shows a block diagram of a typical I/O port. This block diagram does not take into account peripheral functions that may be multiplexed onto the I/O pin. Figure 11-1: Dedicated Port Structure Block Diagram
Dedicated Port Module
I/O Ports
Read TRIS
I/O Cell
Data Bus WR TRIS
D CK
TRIS Latch D WR LAT WR PORT CK Data Latch Q I/O pin
Read LAT Read Port
DS70058D-page 11-2

11.2 I/O Port Control Registers
All I/O ports have three registers directly associated with the operation of the port, where x is a letter that denotes the particular I/O port: TRISx: Data Direction register PORTx: I/O Port register LATx: I/O Latch register Each I/O pin on the device has an associated bit in the TRIS, PORT and LAT registers. Note: The total number of ports and available I/O pins will depend on the device variant. In a given device, all of the bits in a port control register may not be implemented. Refer to the specific device data sheet for further details.
11.2.1
TRIS Registers
The TRISx register control bits determine whether each pin associated with the I/O port is an input or an output. If the TRIS bit for an I/O pin is a 1, then the pin is an input. If the TRIS bit for an I/O pin is a 0, then the pin is configured for an output. An easy way to remember is that a 1 looks like an I (input) and a 0 looks like an O (output). All port pins are defined as inputs after a Reset.
11.2.2
PORT Registers
Data on an I/O pin is accessed via a PORTx register. A read of the PORTx register reads the value of the I/O pin, while a write to the PORTx register writes the value to the port data latch. Many instructions, such as BSET and BCLR instructions, are read-modify-write operations. Therefore, a write to a port implies that the port pins are read, this value is modified, and then written to the port data latch. Care should be taken when read-modify-write commands are used on the PORTx registers and when some I/O pins associated with the port are configured as inputs. If an I/O pin configured as an input is changed to an output at some later time, an unexpected value may be output on the I/O pin. This effect occurs because the read-modify-write instruction reads the instantaneous value on the input pin and loads that value into the port data latch.
11.2.3
LAT Registers
The LATx register associated with an I/O pin eliminates the problems that could occur with read-modify-write instructions. A read of the LATx register returns the values held in the port output latches, instead of the values on the I/O pins. A read-modify-write operation on the LAT register, associated with an I/O port, avoids the possibility of writing the input pin values into the port latches. A write to the LATx register has the same effect as a write to the PORTx register. The differences between the PORT and LAT registers can be summarized as follows: A write to the PORTx register writes the data value to the port latch. A write to the LATx register writes the data value to the port latch. A read of the PORTx register reads the data value on the I/O pin. A read of the LATx register reads the data value held in the port latch.
Any bit and its associated data and control registers that are not valid for a particular device will be disabled. That means the corresponding LATx and TRISx registers, and the port pin, will read as zeros.
DS70058D-page 11-3
11
11.3 Peripheral Multiplexing
When a peripheral is enabled the associated pin output drivers are typically module controlled while a few are user settable. The I/O pin may be read through the input data path, but the output driver for the I/O port bit is generally disabled. An I/O port that shares a pin with another peripheral is always subservient to the peripheral. The peripherals output buffer data and control signals are provided to a pair of multiplexers. The multiplexers select whether the peripheral, or the associated port, has ownership of the output data and control signals of the I/O pin. Figure 11-2 shows how ports are shared with other peripherals, and the associated I/O pin to which they are connected. Note: In order to use PORTB pins for digital I/O, the corresponding bits in the ADPCFG register must be set to 1, even if the A/D module is turned off.
I/O Ports
Figure 11-2:
Shared Port Structure Block Diagram

PIO Module Peripheral Multiplexers Read TRIS Peripheral A Enable Peripheral B Enable D CK TRIS Latch D Q Peripheral B Data Peripheral A Data Q Peripheral B o.e. Peripheral A o.e. 0 0 1 1 0 0 1 1 I/O pin
Data Bus WR TRIS
WR LAT WR Port
CK Data Latch
Read LAT PERA/PERB/PIO Read Port
Peripheral A Input
Peripheral B Input
11.3.1
I/O Multiplexing with Multiple Peripherals

For some dsPIC30F devices, especially those with a small number of I/O pins, multiple peripheral functions may be multiplexed on each I/O pin. Figure 11-2 shows an example of two peripherals multiplexed to the same I/O pin. The name of the I/O pin defines the priority of each function associated with the pin. The conceptual I/O pin, shown in Figure 11-2, has two multiplexed peripherals, Peripheral A and Peripheral B and is named PERA/PERB/PIO. The I/O pin name is chosen so that the user can easily tell the priority of the functions assigned to the pin. For the example shown in Figure 11-2, Peripheral A has the highest priority for control of the pin. If Peripheral A and Peripheral B are enabled at the same time, Peripheral A will take control of the I/O pins.
DS70058D-page 11-4

11.3.1.1 Software Input Pin Control Some of the functions assigned to an I/O pin may be input functions that do not take control of the pin output driver. An example of one such peripheral is the Input Capture module. If the I/O pin associated with the Input Capture is configured as an output, using the appropriate TRIS control bit, the user can manually affect the state of the Input Capture pin through its corresponding PORT register. This behavior can be useful in some situations, especially for testing purposes, when no external signal is connected to the input pin. Referring to Figure 11-2, the organization of the peripheral multiplexers will determine if the peripheral input pin can be manipulated in software using the PORT register. The conceptual peripherals shown in this figure disconnect the PORT data from the I/O pin when the peripheral function is enabled. In general, the following peripherals allow their input pins to be controlled manually through the PORT registers: External Interrupt pins Timer Clock Input pins Input Capture pins PWM Fault pins
Most serial communication peripherals, when enabled, take full control of the I/O pin, so that the input pins associated with the peripheral cannot be affected through the corresponding PORT registers. These peripherals include the following: 11.3.1.2 SPITM I2CTM DCI UART CAN
Pin Control Summary When a peripheral is enabled the associated pin output drivers are typically module controlled while a few are user settable. The term "Module Control" means that the associated port pin output driver is disabled and the pin can only be controlled and accessed by the peripheral. The term "User Settable" means that the associated peripheral port pin output driver is user configurable via the associated TRISx SFR. The TRISx register must be set properly for the peripheral to function properly. For "User Settable" peripheral pins, the actual port pin state can always be read via the PORTx SFR. An Input Capture peripheral makes a good example of a User Settable peripheral. The user must write the associated TRIS register to configure the Input Capture pin as an input. Since the I/O pin circuitry is still active when the Input Capture is enabled, a 'trick' can be used to manually produce capture events using software. The Input Capture pin is configured as an output using the associated TRIS register. Then, the software can write values to the corresponding LAT register drive to internally control the Input Capture pin and force capture events. As another example an INTx pin can be configured as an output and then by writing to the associated LATx bit an INTx interrupt, if enabled, can be generated. The UART is an example of a Module Control peripheral. When the UART is enabled, the PORT and TRIS registers have no effect and cannot be used to read or write the RX and TX pins. Most communication peripheral functions available on the dsPIC are Module Control peripherals. For example, the SPI module can be configured for Master mode in which only the SDO pin is required. In this scenario the SDI pin can be configured as a general purpose output pin by clearing (setting to a logic "0") the associated TRISx bit. Table 11-1 presents a summary of the dsPIC peripherals and associated Pin Output Control and Port pin read status.
DS70058D-page 11-5
11
Table 11-1: Port Pin Control Summary Table Peripheral Pins SDOx SDIx SCKx SSx UART (x = 1 or 2) I2C Input Change Notice Input Capture Output Compare Data Converter Interface UxRX UxTX SCL SDA CN0 - CN23 IC1 - IC8 OC1 - OC8 COFS CSCK CSDI CSDO PWMx FLTA/B QEA QEB INDX TRISx - Pin Output Control <DISSDO = 1>, User Settable <DISSDO = 0>, Module Control User Settable Module Control <SSEN = 0>, User Settable <SSEN = 1>, Module Control Module Control <UTXEN = 0>, User Settable <UTXEN = 1>, Module Control Module Control Module Control User Settable User Settable Module Control Module Control Module Control Module Control Module Control Module Control User Settable Module Control (QEI mode) User Settable (16-bit Timer mode) Module Control (QEI mode) User Settable (16-bit Timer mode) Module Control (QEIM<2:0> = 100 or 110) User Settable in all other modes Module Control Module Control User Settable PORTx Pin Read Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Peripheral Module Enabled State SPI (x = 1 or 2)
I/O Ports
Motor Control PWM QEI
CAN (x = 1 or 2) INTx
CxRX CxTX INT0 - INT5
Yes Yes Yes
11.4
Port Descriptions
Refer to the device data sheet for a description of the available I/O ports and peripheral multiplexing details.
DS70058D-page 11-6

11.5 Change Notification (CN) Pins
The Change Notification (CN) pins provide dsPIC30F devices the ability to generate interrupt requests to the processor in response to a change of state on selected input pins. Up to 24 input pins may be selected (enabled) for generating CN interrupts. The total number of available CN inputs is dependent on the selected dsPIC30F device. Refer to the device data sheet for further details. Figure 11-3 shows the basic function of the CN hardware. Figure 11-3: Input Change Notification Block Diagram
CN0PUE (CNPU1<0>) CN0 pin D C Q CN0 Change CN Interrupt Q C CN0IE (CNEN1<0>) CN1 Change CN1-CN23 Details Not Shown
CN23 Change
11.5.1
CN Control Registers
There are four control registers associated with the CN module. The CNEN1 and CNEN2 registers contain the CNxIE control bits, where x denotes the number of the CN input pin. The CNxIE bit must be set for a CN input pin to interrupt the CPU. The CNPU1 and CNPU2 registers contain the CNxPUE control bits. Each CN pin has a weak pull-up device connected to the pin, which can be enabled or disabled using the CNxPUE control bits. The weak pull-up devices act as a current source that is connected to the pin and eliminate the need for external resistors when push button or keypad devices are connected. Refer to the Electrical Specifications section of the device data sheet for CN pull-up device current specifications.
DS70058D-page 11-7
11
11.5.2 CN Configuration and Operation
The CN pins are configured as follows: 1. 2. 3. 4. 5. 6. Ensure that the CN pin is configured as a digital input by setting the associated bit in the TRISx register. Enable interrupts for the selected CN pins by setting the appropriate bits in the CNEN1 and CNEN2 registers. Turn on the weak pull-up devices (if desired) for the selected CN pins by setting the appropriate bits in the CNPU1 and CNPU2 registers. Clear the CNIF (IFS0<15>) interrupt flag. Select the desired interrupt priority for CN interrupts using the CNIP<2:0> control bits (IPC3<14:12>). Enable CN interrupts using the CNIE (IEC0<15>) control bit.
I/O Ports
When a CN interrupt occurs, the user should read the PORT register associated with the CN pin(s). This will clear the mismatch condition and setup the CN logic to detect the next pin change. The current PORT value can be compared to the PORT read value obtained at the last CN interrupt to determine the pin that changed. The CN pins have a minimum input pulse width specification. Refer to the Electrical Specifications section of the device data sheet for further details.
11.6
CN Operation in Sleep and Idle Modes

The CN module continues to operate during Sleep or Idle modes. If one of the enabled CN pins changes states, the CNIF (IFS0<15>) status bit will be set. If the CNIE bit (IEC0<15>) is set, the device will wake from Sleep or Idle mode and resume operation. If the assigned priority level of the CN interrupt is equal to or less than the current CPU priority level, device execution will continue from the instruction immediately following the SLEEP or IDLE instruction. If the assigned priority level of the CN interrupt is greater than the current CPU priority level, device execution will continue from the CN interrupt vector address.
DS70058D-page 11-8

Register 11-1: CNEN1: Input Change Notification Interrupt Enable Register1 R/W-0 CN13IE R/W-0 CN12IE R/W-0 CN11IE R/W-0 CN10IE R/W-0 CN9IE R/W-0 CN8IE bit 8
Upper Byte: R/W-0 R/W-0 CN15IE CN14IE bit 15
Lower Byte: R/W-0 R/W-0 CN7IE CN6IE bit 7 bit 15-0
R/W-0 CN5IE
R/W-0 CN4IE
R/W-0 CN3IE
R/W-0 CN2IE
R/W-0 CN1IE
R/W-0 CN0IE bit 0
CNxIE: Input Change Notification Interrupt Enable bits 1 = Enable interrupt on input change 0 = Disable interrupt on input change Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
CNEN2: Input Change Notification Interrupt Enable Register2 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 CN23IE bit 7 R/W-0 CN22IE R/W-0 CN21IE R/W-0 CN20IE R/W-0 CN19IE R/W-0 CN18IE R/W-0 CN17IE R/W-0 CN16IE bit 0
bit 15-8 bit 7-0
Unimplemented: Read as 0 CNxIE: Input Change Notification Interrupt Enable bits 1 = Enable interrupt on input change 0 = Disable interrupt on input change Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70058D-page 11-9
11
Register 11-3: CNPU1: Input Change Notification Pull-up Enable Register1 Upper Byte: R/W-0 R/W-0 CN15PUE CN14PUE bit 15
I/O Ports
R/W-0 CN13PUE
R/W-0 CN12PUE
R/W-0 R/W-0 CN11PUE CN10PUE
R/W-0 CN9PUE
R/W-0 CN8PUE bit 8
Lower Byte: R/W-0 R/W-0 CN7PUE CN6PUE bit 7 bit 15-0
R/W-0 CN5PUE
R/W-0 CN4PUE
R/W-0 CN3PUE
R/W-0 CN2PUE
R/W-0 CN1PUE
R/W-0 CN0PUE bit 0
CNxPUE: Input Change Notification Pull-up Enable bits 1 = Enable pull-up on input change 0 = Disable pull-up on input change Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
CNPU2: Input Change Notification Pull-up Enable Register2 U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8
Lower Byte: R/W-0 R/W-0 CN23PUE CN22PUE bit 7 bit 15-8 bit 7-0 Unimplemented: Read as 0
R/W-0 CN21PUE
R/W-0 R/W-0 CN20PUE CN19PUE
R/W-0 CN18PUE
R/W-0 CN17PUE
R/W-0 CN16PUE bit 0
CNxPUE: Input Change Notification Pull-up Enable bits 1 = Enable pull-up on input change 0 = Disable pull-up on input change Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70058D-page 11-10

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the I/O Ports module are: Title Implementing Wake-up on Key Stroke Application Note # AN552
Note:
DS70058D-page 11-11
11
I/O Ports
Revision B
This revision incorporates additional technical content for the dsPIC30F I/O Ports module.
Revision C
Revision D
Section 11.3.1.2 Pin Control Summary was added to this revision.
DS70058D-page 11-12
Section 12. Timers

HIGHLIGHTS
This section of the manual contains the following major topics: 12.1 Introduction .................................................................................................................. 12-2 12.2 Timer Variants .............................................................................................................. 12-3 12.3 Control Registers ......................................................................................................... 12-6 12.4 Modes of Operation ..................................................................................................... 12-9 12.5 Timer Prescalers ........................................................................................................ 12-14 12.6 Timer Interrupts.......................................................................................................... 12-14 12.7 Reading and Writing 16-bit Timer Module Registers ................................................. 12-15 12.8 Low Power 32 kHz Crystal Oscillator Input ................................................................ 12-15 12.9 32-bit Timer Configuration.......................................................................................... 12-16 12.10 32-bit Timer Modes of Operation ............................................................................... 12-18 12.11 Reading and Writing into 32-bit Timers...................................................................... 12-21 12.12 Timer Operation in Power Saving States ................................................................... 12-21 12.13 Peripherals Using Timer Modules .............................................................................. 12-22 12.14 Design Tips ................................................................................................................ 12-24 12.15 Related Application Notes.......................................................................................... 12-25 12.16 Revision History ......................................................................................................... 12-26
12
Timers
DS70059D-page 12-1
Section 12. Timers

12.1 Introduction
Depending on the specific variant, the dsPIC30F device family offers several 16-bit timers. These timers are designated as Timer1, Timer2, Timer3, ..., etc. Each timer module is a 16-bit timer/counter consisting of the following readable/writable registers: TMRx: 16-bit timer count register PRx: 16-bit period register associated with the timer TxCON: 16-bit control register associated with the timer Each timer module also has the associated bits for interrupt control: Interrupt Enable Control bit (TxIE) Interrupt Flag Status bit (TxIF) Interrupt Priority Control bits (TxIP<2:0>) With certain exceptions, all of the 16-bit timers have the same functional circuitry. The 16-bit timers are classified into three types to account for their functional differences: Type A time base Type B time base Type C time base Some 16-bit timers can be combined to form a 32-bit timer. This section does not describe the dedicated timers that are associated with peripheral devices. For example, this includes the time bases associated with the Motor Control PWM module and the Quadrature Encoder Interface (QEI) module.
12
Timers
DS70059D-page 12-2

12.2 Timer Variants
All 16-bit timers available on the dsPIC30F devices are functionally identical with certain exceptions. The 16-bit timers are classified into three functional types; Type A timers, Type B timers and Type C timers. Note: Please refer to the device data sheet for the available timers and the type of each.
12.2.1
Type A Timer
At least one Type A timer is available on most dsPIC30F devices. For most dsPIC30F devices, Timer1 is a Type A timer. A Type A timer has the following unique features over other types: can be operated from the device Low Power 32 kHz Oscillator can be operated in an Asynchronous mode from an external clock source In particular, the unique features of a Type A timer allow it to be used for Real-Time Clock (RTC) applications. A block diagram of the Type A timer is shown in Figure 12-1.
Figure 12-1:
Type A Timer Block Diagram
PRx Equal
Comparator x 16
TSYNC 1 Sync
TMRx Reset TxIF Event Flag 0 1 TGATE (Note 1) SOSCO LPOSCEN SOSCI Gate Sync TCY Q Q D CK TCS TGATE TGATE 0
TON 1X 01 00
TCKPS<1:0> 2 Prescaler 1, 8, 64, 256
Note 1: Refer to Section 7. Oscillator for information on enabling the LP Oscillator.
DS70059D-page 12-3
Section 12. Timers

12.2.2 Type B Timer
Timer2 and Timer4, if present, are Type B timers on most dsPIC30F devices. A Type B timer has the following unique features over other types of timers: A Type B timer can be concatenated with a Type C timer to form a 32-bit timer. The TxCON register for a Type B timer has the T32 control bits to enable the 32-bit timer function. The clock synchronization for a Type B timer is performed after the prescale logic. A block diagram of the Type B timer is shown in Figure 12-2. Figure 12-2: Type B Timer Block Diagram
12
PRx Equal Comparator x 16
Timers
Reset 0 1 TGATE
TMRx
Sync
TxIF Event Flag
Q Q
D CK
TGATE TCS TGATE
TxCKI Gate Sync TCY
TON 1X 01 00
TCKPS<1:0> 2 Prescaler 1, 8, 64, 256
DS70059D-page 12-4

12.2.3 Type C Timer
Timer3 and Timer5 are Type C timers on most dsPIC30F devices. A Type C timer has the following unique features over other types of timers: A Type C timer can be concatenated with a Type B timer to form a 32-bit timer. On a given device, at least one Type C timer has the ability to trigger an A/D conversion. A block diagram of the Type C timer is shown in Figure 12-3. Figure 12-3: Type C Timer Block Diagram
PRx ADC Event Trigger Equal Comparator x 16
TMRx Reset 0 1 TGATE Q Q D CK TCS TGATE TGATE
TxIF Event Flag
TxCK
TON Sync 1X 01 TCY 00
TCKPS<1:0> 2 Prescaler 1, 8, 64, 256
Note:
In certain variants of the dsPIC30F family, the TxCK pin may not be available. Refer to the device data sheet for the I/O pin details. In such cases, the timer must use the system clock (FOSC/4) as its input clock, unless it is configured for 32-bit operation.
DS70059D-page 12-5
Section 12. Timers

12.3 Control Registers
TxCON: Type A Time Base Register U-0 R/W-0 TSIDL U-0 U-0 U-0 U-0 U-0 bit 8
Register 12-1: Upper Byte: R/W-0 TON bit 15
R/W-0 TGATE
R/W-0 R/W-0 TCKPS<1:0>
U-0
R/W-0 TSYNC
R/W-0 TCS
U-0 bit 0
12
Timers
TON: Timer On Control bit 1 = Starts the timer 0 = Stops the timer Unimplemented: Read as 0 TSIDL: Stop in Idle Mode bit 1 = Discontinue timer operation when device enters Idle mode 0 = Continue timer operation in Idle mode Unimplemented: Read as 0 TGATE: Timer Gated Time Accumulation Enable bit 1 = Gated time accumulation enabled 0 = Gated time accumulation disabled (TCS must be set to 0 when TGATE = 1. Reads as 0 if TCS = 1) TCKPS<1:0>: Timer Input Clock Prescale Select bits 11 = 1:256 prescale value 10 = 1:64 prescale value 01 = 1:8 prescale value 00 = 1:1 prescale value Unimplemented: Read as 0 TSYNC: Timer External Clock Input Synchronization Select bit When TCS = 1: 1 = Synchronize external clock input 0 = Do not synchronize external clock input When TCS = 0: This bit is ignored. Read as 0. Timer1 uses the internal clock when TCS = 0. TCS: Timer Clock Source Select bit 1 = External clock from pin TxCK 0 = Internal clock (FOSC/4) Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 14 bit 13
bit 12-7 bit 6
bit 5-4
bit 3 bit 2
bit 1
bit 0
DS70059D-page 12-6

Register 12-2: Upper Byte: R/W-0 TON bit 15 TxCON: Type B Time Base Register U-0 R/W-0 TSIDL U-0 U-0 U-0 U-0 U-0 bit 8
R/W-0 TGATE
R/W-0 T32
U-0
R/W-0 TCS
U-0 bit 0
TON: Timer On bit When T32 = 1 (in 32-bit Timer mode): 1 = Starts 32-bit TMRx:TMRy timer pair 0 = Stops 32-bit TMRx:TMRy timer pair When T32 = 0 (in 16-bit Timer mode): 1 = Starts 16-bit timer 0 = Stops 16-bit timer Unimplemented: Read as 0 TSIDL: Stop in Idle Mode bit 1 = Discontinue timer operation when device enters Idle mode 0 = Continue timer operation in Idle mode Unimplemented: Read as 0 TGATE: Timer Gated Time Accumulation Enable bit 1 = Timer gated time accumulation enabled 0 = Timer gated time accumulation disabled (TCS must be set to logic 0 when TGATE = 1) TCKPS<1:0>: Timer Input Clock Prescale Select bits 11 = 1:256 prescale value 10 = 1:64 prescale value 01 = 1:8 prescale value 00 = 1:1 prescale value T32: 32-bit Timer Mode Select bits 1 = TMRx and TMRy form a 32-bit timer 0 = TMRx and TMRy form separate 16-bit timer Unimplemented: Read as 0 TCS: Timer Clock Source Select bit 1 = External clock from pin TxCK 0 = Internal clock (FOSC/4) Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 14 bit 13
bit 12-7 bit 6
bit 5-4
bit 3
bit 2 bit 1
bit 0
DS70059D-page 12-7
Section 12. Timers

Register 12-3: Upper Byte: R/W-0 TON bit 15 TxCON: Type C Time Base Register U-0 R/W-0 TSIDL U-0 U-0 U-0 U-0 U-0 bit 8
R/W-0 TGATE
U-0
U-0
R/W-0 TCS
U-0 bit 0
12
Timers
TON: Timer On bit 1 = Starts 16-bit TMRx 0 = Stops 16-bit TMRx Unimplemented: Read as 0 TSIDL: Stop in Idle Mode bit 1 = Discontinue module operation when device enters Idle mode 0 = Continue module operation in Idle mode Unimplemented: Read as 0 TGATE: Timer Gated Time Accumulation Enable bit 1 = Timer gated time accumulation enabled 0 = Timer gated time accumulation disabled (Read as 0 if TCS = 1) (TCS must be set to logic 0 when TGATE = 1) TCKPS<1:0>: Timer Input Clock Prescale Select bits 11 = 1:256 prescale value 10 = 1:64 prescale value 01 = 1:8 prescale value 00 = 1:1 prescale value Unimplemented: Read as 0 TCS: Timer Clock Source Select bit 1 = External clock from pin TxCK 0 = Internal clock (FOSC/4) Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 14 bit 13
bit 12-7 bit 6
bit 5-4
bit 3-2 bit 1
bit 0
DS70059D-page 12-8

12.4 Modes of Operation
Each timer module can operate in one of the following modes: As a synchronous timer As a synchronous counter As a gated timer As an asynchronous counter (Type A time base only)
The Timer modes are determined by the following bits: TCS (TxCON<1>): Timer Clock Source Control bit TSYNC (T1CON<2>): Timer Synchronization Control bit (Type A time base only) TGATE (TxCON<6>): Timer Gate Control bit Each timer module is enabled or disabled using the TON Control bit (TxCON <15>). Note: Only Type A time bases support the External Asynchronous Clock mode.
12.4.1
Timer Mode
All types of timers have the ability to operate in Timer mode. In Timer mode, the input clock to the timer is provided from the internal system clock (FOSC/4). When enabled, the timer increments once per instruction cycle for a 1:1 prescaler setting. The Timer mode is selected by clearing the TCS control bit (TxCON<1>). The Synchronous mode control bit, TSYNC (T1CON<2>), has no effect, since the system clock source is used to generate the timer clock. Example 12-1:
; ; ; ; ;
Initialization Code for 16-bit Timer Using System Clock
The following code example will enable Timer1 interrupts, load the Timer1 Period register and start Timer1. When a Timer1 period match interrupt occurs, the interrupt service routine must clear the Timer1 interrupt status flag in software. CLR CLR MOV MOV BSET BCLR BCLR BCLR BSET BSET T1CON TMR1 #0xFFFF, w0 w0, PR1 IPC0, #T1IP0 IPC0, #T1IP1 IPC0, #T1IP2 IFS0, #T1IF IEC0, #T1IE T1CON, #TON ; ; ; ; ; ; ; ; ; ; ; ; Stops the Timer1 and reset control reg. Clear contents of the timer register Load the Period register with the value 0xFFFF Setup Timer1 interrupt for desired priority level (this example assigns level 1 priority) Clear the Timer1 interrupt status flag Enable Timer1 interrupts Start Timer1 with prescaler settings at 1:1 and clock source set to the internal instruction cycle
; Example code for Timer1 ISR __T1Interrupt: BCLR IFS0, #T1IF RETFIE
; Reset Timer1 interrupt flag ; User code goes here. ; Return from ISR
DS70059D-page 12-9
Section 12. Timers

12.4.2 Synchronous Counter Mode Using External Clock Input
When the TCS control bit (TxCON<1>) is set, the clock source for the timer is provided externally and the selected timer increments on every rising edge of clock input on the TxCK pin. External clock synchronization must be enabled for a Type A time base. This is accomplished by setting the TSYNC control bit (TxCON<2>). For Type B and Type C time bases, the external clock input is always synchronized to the system instruction cycle clock, TCY. When the timer is operated in the Synchronized Counter mode, there are minimum requirements for the external clock high time and low time. The synchronization of the external clock source with the device instruction clock is accomplished by sampling the external clock signal at two different times within an instruction cycle. A timer operating from a synchronized external clock source will not operate in Sleep mode, since the synchronization circuit is shut-off during Sleep mode. Note: The external input clock must meet certain minimum high time and low time requirements when Timerx is used in the Synchronous Counter mode. Refer to the device data sheet Electrical Specifications section for further details. Initialization Code for 16-bit Synchronous Counter Mode Using an External Clock Input
12
Timers
Example 12-2:
; ; ; ; ;
The following code example will enable Timer1 interrupts, load the Timer1 Period register and start Timer1 using an external clock and a 1:8 prescaler setting. When a Timer1 period match interrupt occurs, the interrupt service routine must clear the Timer1 interrupt status flag in software. CLR CLR MOV MOV BSET BCLR BCLR BCLR BSET MOV MOV T1CON TMR1 #0x8CFF, w0 w0, PR1 IPC0, #T1IP0 IPC0, #T1IP1 IPC0, #T1IP2 IFS0, #T1IF IEC0, #T1IE #0x8016, W0 w0, T1CON ; ; ; ; ; ; ; ; ; ; ; ; Stops the Timer1 and reset control reg. Clear contents of the timer register Load the Period register with the value 0x8CFF Setup Timer1 interrupt for desired priority level (this example assigns level 1 priority) Clear the Timer1 interrupt status flag Enable Timer1 interrupts Start Timer1 with prescaler settings at 1:8 and clock source set to the external clock in the synchronous mode
Example code for Timer1 ISR
__T1Interrupt: BCLR IFS0, RETFIE
#T1IF
DS70059D-page 12-10

12.4.3 Type A Timer Asynchronous Counter Mode Using External Clock Input
A Type A time base has the ability to operate in an Asynchronous Counting mode, using an external clock source connected to the TxCK pin. When the TSYNC control bit (TxCON<2>) is cleared, the external clock input is not synchronized with the device system clock source. The time base continues to increment asynchronously to the internal device clock. The asynchronous operation time base is beneficial for the following applications: The time base can operate during Sleep mode and can generate an interrupt on period register match that will wake-up the processor. The time base can be clocked from the low power 32 kHz oscillator for real-time clock applications Please see Section 12.12.1 Timer Operation in Sleep Mode for more details.. Note 1: Only Type A time bases support the Asynchronous Counter mode. 2: The external input clock must meet certain minimum high time and low time requirements when Timerx is used in the Asynchronous Counter mode. Refer to the device data sheet Electrical Specifications section for further details. 3: Unexpected results may occur when reading Timer1, in asynchronous mode. Example 12-3: Initialization Code for 16-bit Asynchronous Counter Mode Using an External Clock Input
; ; ; ; ;
The following code example will enable Timer1 interrupts, load the Timer1 Period register and start Timer1 using an asynchronous external clock and a 1:8 prescaler setting. When a Timer1 period match interrupt occurs, the interrupt service routine must clear the Timer1 interrupt status flag in software. CLR CLR MOV MOV BSET BCLR BCLR BCLR BSET MOV MOV T1CON TMR1 #0x7FFF, w0 w0, PR1 IPC0, #T1IP0 IPC0, #T1IP1 IPC0, #T1IP2 IFS0, #T1IF IEC0, #T1IE #0x8012, w0 w0, T1CON ; ; ; ; ; ; ; ; ; ; ; ; Stops the Timer1 and reset control reg. Clear contents of the timer register Load the Period register with the value 0x7FFF Setup Timer1 interrupt for desired priority level (this example assigns level 1 priority) Clear the Timer1 interrupt status flag Enable Timer1 interrupts Start Timer1 with prescaler settings at 1:8 and clock source set to the external clock in the asynchronous mode
; Example code for Timer1 ISR __T1Interrupt: BCLR IFS0, RETFIE
#T1IF
DS70059D-page 12-11
Section 12. Timers

12.4.4 Timer Operation with Fast External Clock Source
In some applications, it may be desirable to use one of the timers to count clock edges from a relatively high frequency external clock source. In these situations, Type A and Type B time bases are the most suitable choices for counting the external clock source, because the clock synchronization logic for these timers is located after the timer prescaler (see Figure 12-1 and Figure 12-2). This allows a higher external clock frequency to be used that will not violate the minimum high and low times required by the prescaler. When a timer prescaler ratio other than 1:1 is selected for a Type A or Type B time base, the minimum high and low times for the external clock input are reduced by the chosen prescaler ratio. A Type A time base is unique because it can be operated in an Asynchronous Clock mode, eliminating any prescaler timing requirements. Note that in all cases, there are minimum high and low times for the external clock signal that cannot be exceeded. These minimum times are required to satisfy the I/O pin timing requirements.
12
Timers
Please refer to the device data sheet for the external clock timing specifications associated with the time bases.
12.4.5
Gated Time Accumulation Mode

The Gated Time Accumulation mode allows the internal timer register to increment based upon the duration of the high time applied to the TxCK pin. In the Gated Time Accumulation mode, the timer clock source is derived from the internal system clock. When the TxCK pin state is high, the timer register will count up until a period match has occurred, or the TxCK pin state is changed to a low state. A pin state transition from high to low will set the TxIF interrupt flag. Depending on when the edge occurs, the interrupt flag is asserted 1 or 2 instruction cycles after the falling edge of the signal on the TxCK pin. The TGATE control bit (TxCON<6>) must be set to enable the Gated Time Accumulation mode. The timer must be enabled, TON (TxCON<15>) = 1, and the timer clock source set to the internal clock, TCS (TxCON<1>) = 0. The gate operation starts on a rising edge of the signal applied to the TxCK pin and terminates on the falling edge of the signal applied to the TxCK pin. The respective timer will increment while the external gate signal is high. The falling edge of the gate signal terminates the count operation, but does not reset the timer. The user must reset the timer if it is desired to start from zero on the next rising edge gate input. The falling edge of the gate signal generates an interrupt. Note: The timer will not interrupt the CPU when a timer period match occurs in Gate Time Accumulation mode.
The resolution of the timer count is directly related to the timer clock period. For a timer prescaler of 1:1, the timer clock period is one instruction cycle. For a timer prescaler of 1:256, the timer clock period is 256 times the instruction cycle. The timer clock resolution can be associated to the pulse width of the gate signal. Refer to the Electrical Specifications section in the device data sheet for further details on the gate width pulse requirements.
DS70059D-page 12-12

Figure 12-4: Gated Timer Mode Operation
1 Instruction Cycle (TCY)
TMRx TxCK pin
0000
0001
0002
0003
0004
0005
TxIF bit cleared by user in software. TxIF
Example 12-4:
; ; ; ; ;
Initialization Code for 16-bit Gated Time Accumulation Mode
The following code example will enable Timer2 interrupts, load the Timer2 Period register and start Timer2 using an internal clock and an external gate signal. On the falling edge of the gate signal a Timer2 interrupt occurs. The interrupt service routine must clear the Timer2 interrupt status flag in software . CLR CLR MOV MOV BSET BCLR BCLR BCLR BSET BSET BSET T2CON TMR2 #0xFFFF, w0 w0, PR2 IPC1, IPC1, IPC1, IFS0, IEC0, T2CON, T2CON, #T2IP0 #T2IP1 #T2IP2 #T2IF #T2IE #TGATE #TON ; ; ; ; ; ; ; ; ; ; ; ; Stops the Timer2 and reset control reg. Clear contents of the timer register Load the Period register with the value 0xFFFF Setup Timer2 interrupt for desired priority level (this example assigns level 1 priority) Clear the Timer2 interrupt status flag Enable Timer2 interrupts Set up Timer2 for operation in Gated Time Accumulation mode Start Timer2
#T2IF
DS70059D-page 12-13
Section 12. Timers

12.5 Timer Prescalers
The input clock (FOSC/4 or external clock) to all 16-bit timers has prescale options of 1:1, 1:8, 1:64 and 1:256. The clock prescaler is selected using the TCKPS<1:0> control bits (TxCON<5:4>). The prescaler counter is cleared when any of the following occurs: A write to the TMRx register Clearing TON (TxCON<15>) to 0 Any device Reset Note: The TMRx register is not cleared when TxCON is written.
12.6
Timer Interrupts
A 16-bit timer has the ability to generate an interrupt on a period match or falling edge of the external gate signal, depending on the Operating mode. The TxIF bit is set when one of the following conditions is true: The timer count matches the respective period register and the timer module is not operating in Gated Time Accumulation mode. The falling edge of the gate signal is detected when the timer is operating in Gated Time Accumulation mode. The TxIF bit must be cleared in software. A timer is enabled as a source of interrupt via the respective timer interrupt enable bit, TxIE. Furthermore, the interrupt priority level bits (TxIP<2:0>) must be written with a non-zero value in order for the timer to be a source of interrupt. Refer to Section 6. Reset Interrupts for further details. Note: A special case occurs when the period register is loaded with 0x0000 and the timer is enabled. No timer interrupts will be generated for this configuration.
12
Timers
Figure 12-5:
Interrupt Timing for Timer Period Match

1 Instruction Cycle (TCY)
TMR2 PR2
47FD
47FE
47FF
4800
0000
0001
0002
0003
0004
0005
TMR2 Resets Here 4800 Cleared by User TxIF
DS70059D-page 12-14

12.7 Reading and Writing 16-bit Timer Module Registers
All timer module SFRs can be written to as a byte (8-bits) or as a word (16-bits). All timer module SFRs can only be read as a word (16-bits).
12.7.1
Writing to the 16-bit Timers

The timer and its respective period register can be written to while the module is operating. The user should be aware of the following when byte writes are performed: If the timer is incrementing and the low byte of the timer is written to, the upper byte of the timer is not affected. If 0xFF is written into the low byte of the timer, the next timer count clock after this write will cause the low byte to rollover to 0x00 and generate a carry into the high byte of the timer. If the timer is incrementing and the high byte of the timer is written to, the low byte of the timer is not affected. If the low byte of the timer contains 0xFF when the write occurs, the next timer count clock will generate a carry from the timer low byte and this carry will cause the upper byte of the timer to increment. When the TMRx register is written to (word or byte) via an instruction, the TMRx register increment is masked and does not occur during that instruction cycle. Writes to a timer with an asynchronous clock source should be avoided in a real-timekeeping application. See Section 12.4.1 Timer Mode for more details.
12.7.2
Reading from the 16-bit Timers

All reads of the timers and their associated SFRs must be word reads (16-bits). A byte read will have no effect (0 will be returned). The timer and respective period register can be read while the module is operating. A read of the TMRx register does not prevent the timer from incrementing during the same instruction cycle.
12.8
Low Power 32 kHz Crystal Oscillator Input

In each device variant, the Low Power 32 kHz Crystal Oscillator is available to a Type A timer module for Real-Time Clock (RTC) applications. The LP Oscillator becomes the clock source for the timer when the LP Oscillator is enabled and the timer is configured to use the external clock source. The LP Oscillator is enabled by setting the LPOSCEN control bit in the OSCCON register. The 32 kHz crystal is connected to the SOSCO/SOSCI device pins. Refer to Section 7. Oscillator for further details.
DS70059D-page 12-15
Section 12. Timers

12.9 32-bit Timer Configuration
A 32-bit timer module can be formed by combining a Type B and a Type C 16-bit timer module. The Type C time base becomes the MSWord of the combined timer and the Type B time base is the LSWord. When configured for 32-bit operation, the control bits for the Type B time base control the operation of the 32-bit timer. The control bits in the TxCON register for the Type C time base have no effect. For interrupt control, the combined 32-bit timer uses the interrupt enable, interrupt flag and interrupt priority control bits of the Type C time base. The interrupt control and status bits for the Type B time base are not used during 32-bit timer operation. Note: Refer to the device data sheet for information on the specific Type B and Type C time bases that can be combined.
12
Timers
The following configuration settings assume Timer3 is a Type C time base and Timer2 is a Type B time base: TON (T2CON<15>) = 1. T32 (T2CON<3>) = 1. TCKPS<1:0> (T2CON<5:4>) are used to set the Prescaler mode for Timer2 (Type B time base). The TMR3:TMR2 register pair contains the 32-bit value of the timer module; the TMR3 (Type C time base) register is the Most Significant Word, while the TMR2 (Type B time base) register is the Least Significant Word of the 32-bit timer value. The PR3:PR2 register pair contains the 32-bit period value that is used for comparison with the TMR3:TMR2 timer value. T3IE (IEC0<7>) is used to enable the 32-bit timer interrupt for this configuration. T3IF (IFS0<7>) is used as a status flag for the timer interrupt. T3IP<2:0> (IPC1<14:12>) sets the interrupt priority level for the 32-bit timer. T3CON<15:0> are dont care bits. A block diagram representation of the 32-bit timer module using Timer2 and Timer3 as an example is shown in Figure 12-6.
DS70059D-page 12-16

Figure 12-6: Type B-Type C Timer Pair Block Diagram (32-bit Timer)
Data Bus<15:0>
TMR3HLD 16 Write TMR2 Read TMR2 16 Reset 16
TMR3 MSWord
TMR2 LSWord
Sync
ADC Event Trigger
Equal
Comparator x 32
PR3 0 1 TGATE (T2CON<6>)
PR2
T3IF Event Flag
Q Q
D CK
TGATE (T2CON<6>) TGATE
TCS
TCKPS<1:0> TON 2
T2CKI Gate Sync TCY
1X Prescaler 1, 8, 64, 256
01
00
Note 1: This block diagram assumes Timer3 is a Type C time base, Timer2 is a Type B time base. 2: Timer configuration bit, T32 (T2CON<3>), must be set to 1 for a 32-bit timer/counter operation. All control bits are respective to the T2CON register.
DS70059D-page 12-17
Section 12. Timers

12.10 12.10.1 32-bit Timer Modes of Operation Timer Mode
Example 12-5 shows how to configure a 32-bit timer in Timer mode. This example assumes Timer2 is a Type B time base and Timer3 is a Type C time base. For 32-bit timer operation, the T32 control bit must be set in the T2CON register (Type B time base). When Timer2 and Timer3 are configured for a 32-bit timer, the T3CON control bits are ignored. Only the T2CON control bits are required for setup and control. The Timer2 clock and gate input is utilized for the 32-bit timer module, but an interrupt is generated with the T3IF flag. Timer2 is the LSWord and Timer3 is the MSWord of the 32-bit timer. TMR3 is incremented by an overflow (carry-out) from TMR2. The 32-bit timer increments up to a match value preloaded into the combined 32-bit period register formed by PR2 and PR3, then rolls over and continues. For a maximum 32-bit timer count, load PR3:PR2 with a value of 0xFFFFFFFF. An interrupt is generated on a period match, if enabled. Example 12-5: Initialization Code for 32-bit Timer Using Instruction Cycle as Input Clock
12
Timers
; ; ; ; ;
The following code example will enable Timer3 interrupts, load the Timer3:Timer2 Period Register and start the 32-bit timer module consisting of Timer3 and Timer2. When a 32-bit period match interrupt occurs, the user must clear the Timer3 interrupt status flag in software. CLR CLR CLR CLR MOV MOV MOV BSET BCLR BCLR BCLR BSET BSET BSET T2CON T3CON TMR3 TMR2 #0xFFFF, w0 w0, PR3 w0, PR2 IPC1, IPC1, IPC1, IFS0, IEC0, T2CON, T2CON, #T3IP0 #T3IP1 #T3IP2 #T3IF #T3IE #T32 #TON ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; Stops any 16/32-bit Timer2 operation Stops any 16-bit Timer3 operation Clear contents of the Timer3 timer register Clear contents of the Timer2 timer register Load the Period Register 3 with the value 0xFFFF Load the Period Register2 with value 0xFFFF Setup Timer3 interrupt for desired priority level (this example assigns level 1 priority) Clear the Timer3 interrupt status flag Enable Timer3 interrupts Enable 32-bit Timer operation Start 32-bit timer with prescaler settings at 1:1 and clock source set to the internal instruction cycle
#T3IF
DS70059D-page 12-18

12.10.2 Synchronous Counter Mode
The 32-bit timer operates similarly to a 16-bit timer in Synchronous Counter mode. Example 12-6 shows how to configure a 32-bit timer in Synchronous Counter mode. This example assumes Timer2 is a Type B time base and Timer3 is a Type C time base. Example 12-6: Initialization Code for 32-bit Synchronous Counter Mode Using an External Clock Input
; ; ; ; ;
The following code example will enable Timer2 interrupts, load the Timer3:Timer2 Period register and start the 32-bit timer module consisting of Timer3 and Timer2. When a 32-bit period match interrupt occurs, the user must clear the Timer3 interrupt status flag in the software. CLR CLR CLR CLR MOV MOV MOV BSET BCLR BCLR BCLR BSET MOV MOV T2CON T3CON TMR3 TMR2 #0xFFFF, w0 w0, PR3 w0, PR2 IPC1, #T3IP0 IPC1, #T3IP1 IPC1, #T3IP2 IFS0, #T3IF IEC0, #T3IE #0x801A, w0 w0, T2CON ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; Stops any 16/32-bit Timer2 operation Stops any 16-bit Timer3 operation Clear contents of the Timer3 timer register Clear contents of the Timer2 timer register Load the Period Register3 with the value 0xFFFF Load the Period Register2 with value 0xFFFF Setup Timer3 interrupt for desired priority level (this example assigns level 1 priority) Clear the Timer3 interrupt status flag Enable Timer3 interrupts Enable 32-bit Timer operation and start 32-bit timer with prescaler settings at 1:8 and clock source set to external clock
#T3IF
DS70059D-page 12-19
Section 12. Timers

12.10.3 Asynchronous Counter Mode
Type B and Type C time bases do not support the Asynchronous External Clock mode. Therefore, no 32-bit Asynchronous Counter mode is supported.
12.10.4
Gated Time Accumulation Mode

The 32-bit timer operates similarly to a 16-bit timer in Gated Time Accumulation mode. Example 12-7 shows how to configure a 32-bit timer in Gated Time Accumulation mode. This example assumes Timer2 is a Type B time base and Timer3 is a Type C time base. Example 12-7:
; ; ; ; ; ; ;
Initialization Code for 32-bit Gated Time Accumulation Mode
12
Timers
The following code example will enable Timer2 interrupts, load the Timer3:Timer2 Period register and start the 32-bit timer module consisting of Timer3 and Timer2. When a 32-bit period match occurs the timer will simply roll over and continue counting. However, when at the falling edge of the Gate signal on T2CK an interrupt is generated, if enabled. The user must clear the Timer3 interrupt status flag in the software. CLR CLR CLR CLR MOV MOV MOV BSET BCLR BCLR BCLR BSET MOV MOV T2CON T3CON TMR3 TMR2 #0xFFFF, w0 w0, PR3 w0, PR2 IPC1, #T3IP0 IPC1, #T3IP1 IPC1, #T3IP2 IFS0, #T3IF IEC0, #T3IE #0x804C, w0 w0, T2CON ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; Stops any 16/32-bit Timer2 operation Stops any 16-bit Timer3 operation Clear contents of the Timer3 register Clear contents of the Timer2 register Load the Period Register3 with the value 0xFFFF Load the Period Register2 with value 0xFFFF Setup Timer3 interrupt for desired priority level (this example assigns level 1 priority) Clear the Timer3 interrupt status flag Enable Timer3 interrupts Enable 32-bit Timer operation and Start 32-bit timer in gated time accumulation mode.
#T3IF
DS70059D-page 12-20

12.11 Reading and Writing into 32-bit Timers
In order for 32-bit read/write operations to be synchronized between the LSWord and MSWord of the 32-bit timer, additional control logic and holding registers are utilized (see Figure 12-6). Each Type C time base has a register called TMRxHLD, that is used when reading or writing the timer register pair. The TMRxHLD registers are only used when their respective timers are configured for 32-bit operation. Assuming TMR3:TMR2 form a 32-bit timer pair; the user should first read the LSWord of the timer value from the TMR2 register. The read of the LSWord will automatically transfer the contents of TMR3 into the TMR3HLD register. The user can then read TMR3HLD to get the MSWord of the timer value. This is shown in the example below: Example 12-8:
; ;
Reading from a 32-bit Timer
The following code segment reads the 32-bit timer formed by the Timer3-Timer2 pair into the registers W1(MS Word) and W0(LS Word). MOV TMR2, W0 MOV TMR3HLD, W1 ;Transfer the LSW into W1 ;Transfer the MSW from the holding register to W0
To write a value to the TMR3:TMR2 register pair, the user should first write the MSWord to the TMR3HLD register. When the LSWord of the timer value is written to TMR2, the contents of TMR3HLD will automatically be transferred to the TMR3 register.
12.12 12.12.1
Timer Operation in Power Saving States Timer Operation in Sleep Mode

When the device enters Sleep mode, the system clock is disabled. If the timer module is running from the internal clock source (FOSC/4), it will also be disabled. A Type A timer is different from the other timer modules because it can operate asynchronously from an external clock source. Because of this distinction, the Type A time base module can continue to operate during Sleep mode. To operate in Sleep mode, Type A time base must be configured as follows: The Timer1 module is enabled, TON = 1 (T1CON<15>) and The Timer1 clock source is selected as external, TCS = 1 (T1CON<1> = 1) and The TSYNC bit (T1CON<2>) is set to logic 0 (Asynchronous Counter mode enabled). Note: Asynchronous counter operation is only supported for the Timer1 module.
When all of the above conditions are met, Timer1 will continue to count and detect period matches when the device is in Sleep mode. When a match between the timer and the period register occurs, the TxIF bit will be set and an interrupt can be generated to optionally wake the device from Sleep. Refer to Section 10. Watchdog Timer and Power Saving Modes for further details. When executing the SLEEP instruction in asynchronous mode using a 32.768 kHz real time oscillator to keep track of real time in seconds, it is important to insure the crystal connected to the inputs is operating. This can be done by checking for a non-zero value in TMR1 and then executing the SLEEP instruction. Failure to do so may result in a normal sleep execution with no wake-up from sleep due to timer time out. Example Code:
mov.b mov.b mov mov.b mov.b bset clr mov mov #0x46,w1 #0x57,w2 #OSCCONL,w3 w1,[w3] w2,[w3] OSCCONL,#LPOSCEN TMR1 #0x7FFF,W0 W0,PR1 ; follow write sequence ; for OSCCONL writes.
;enable 32Khz external xtal ; set up TMR1 for ;interrupts every 1.0 Sec
DS70059D-page 12-21
Section 12. Timers

bclr IFS0,#T1IF mov #0x8002,W0 mov W0,T1CON bset IEC0,#T1IE TestLPOperation: mov TMR1,W0 cp W0,#0 bra Z,TestLPOperation pwrsav #0 ; clr interrupt flag ; External clock, 1:1; start TMR1 ; enable interrupt ; check for non-zero value in TMR1 ; if zero then recheck ; non-zero then execute sleep
12.12.2
Timer Operation in Idle Mode

When the device enters Idle mode, the system clock sources remain functional and the CPU stops executing code. The timer modules can optionally continue to operate in Idle mode. The TSIDL bit (TxCON<13>) selects if the timer module will stop in Idle mode, or continue to operate normally. If TSIDL = 0, the module will continue operation in Idle mode. If TSIDL = 1, the module will stop in Idle mode.
12
Timers
12.13 12.13.1
Peripherals Using Timer Modules Time Base for Input Capture/Output Compare
The Input Capture and Output Compare peripherals can select one of two timer modules as their time base. Refer to Section 13. Input Capture, Section 14. Output Compare, and the device data sheet for further details.
12.13.2
A/D Special Event Trigger

On each device variant, one Type C time base has the capability to generate a special A/D conversion trigger signal on a period match, in both 16 and 32-bit modes. The timer module provides a conversion start signal to the A/D sampling logic. If T32 = 0, when a match occurs between the 16-bit timer register (TMRx) and the respective 16-bit period register (PRx), the A/D special event trigger signal is generated. If T32 = 1, when a match occurs between the 32-bit timer (TMRx:TMRy) and the 32-bit respective combined period register (PRx:PRy), the A/D special event trigger signal is generated. The special event trigger signal is always generated by the timer. The trigger source must be selected in the A/D converter control registers. Refer to Section 17. 10-bit A/D Converter, Section 18. 12-bit A/D Converter, and the device data sheet for additional information.
12.13.3
Timer as an External Interrupt Pin

The external clock input pin for each timer can be used as an additional interrupt pin. To provide the interrupt, the timer period register, PRx, is written with a non-zero value and the TMRx register is initialized to a value of 1 less than the value written to the period register. The timer must be configured for a 1:1 clock prescaler. An interrupt will be generated when the next rising edge of the external clock signal is detected.
12.13.4
I/O Pin Control

When a timer module is enabled and configured for external clock or gate operation, the user must ensure the I/O pin direction is configured for an input. Enabling the timer module does not configure the pin direction.
DS70059D-page 12-22
Table 12-1:
Bit 13 Timer1 Register Timer1 Period Register Timer2 Register Timer3 Holding Register (used in 32-bit mode only) Timer3 Register Timer2 Period Register Timer3 Period Register Timer4 Register Timer5 Holding Register (used in 32-bit mode only) Timer5 Register Timer4 Period Register Timer5 Period Register SI2CIF IC4IF IC2IE IC4IE T1IP<2:0> T3IP<2:0> INT2IP<2:0> T5IP<2:0> T2IP<2:0> OC1IP<2:0> IC3IE C1IE SPI2IE U2TXIE U2RXIE INT2IE T5IE NVMIE ADIE U1TXIE U1RXIE SPI1IE T3IE T2IE IC3IF C1IF SPI2IF U2TXIF U2RXIF INT2IF T5IF T4IF OC2IE T4IE IC1IP<2:0> OC2IP<2:0> T4IP<2:0> NVMIF ADIF U1TXIF U1RXIF SPI1IF T3IF T2IF OC2IF TSIDL TGATE TSIDL TGATE TCKPS1 TCKPS0 TCKPS1 TCKPS0 IC2IF OC4IF IC2IE OC4IE T32 T1IF OC3IF T1IE OC3IE OC1IF IC8IF OC1IE IC8IE TCS TCS IC1IF IC7IF IC1IE IC7IE INT0IP<2:0> IC2IP<2:0> OC4IP<2:0> INT01F INT1IF INT0IE INT1IE TSIDL TGATE TCKPS1 TCKPS0 TSIDL TGATE TCKPS1 TCKPS0 T32 TCS TCS TSIDL TGATE TCKPS1 TCKPS0 TSYNC TCS Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Value on All Resets
Special Function Registers Associated with Timer Modules
Name SFR Address
Bit 15
Bit 14
TMR1
0100
0000 0000 0000 0000 1111 1111 1111 1111 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 1111 1111 1111 1111 1111 1111 1111 1111 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 1111 1111 1111 1111 1111 1111 1111 1111 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100
PR1
0102
T1CON
0104
TON
DS70059D-page 12-23
TMR2
0106
TMR3HLD
0108
TMR3
010A
PR2
010C
PR3
010E
T2CON
0110
TON
T3CON
0112
TON
TMR4
0114
TMR5HLD
0116
TMR5
0118
PR4
011A
PR5
011C
T4CON
011E
TON
T5CON
0120
TON
IFS0
0084
CNIF
MI2CIF
IFS1
0086
IC61F
IC5IF
IEC0
008C
CNIE
MI2CIE
IEC1
008E
IC6IE
IC5IE
IPC0
0094
IPC1
0096
IPC5
009E
Note:
Please refer to the device data sheet for specific memory map details.
Section 12. Timers

12.14 Design Tips
Question 1: Can a timer module be used to wake the device from Sleep mode?
Answer: Yes, but only Timer1 has the ability to wake the device from Sleep mode. This is because Timer1 allows the TMR1 register to increment from an external, unsynchronized clock source. When the TMR1 register is equal to the PR1 register, the device will wake from Sleep mode, if Timer1 interrupts have been enabled using the T1IE control bit. Refer to Section 12.12.1 Timer Operation in Sleep Mode for further details.
12.14.1
Example Application
An example application is shown in Figure 12-7, where Timer1 (Type A time base) is driven from an external 32.768 kHz oscillator. The external 32.768 kHz oscillator is typically used in applications where real-time needs to be kept, but it is also desirable to have the lowest possible power consumption. The Timer1 oscillator allows the device to be placed in Sleep while the timer continues to increment. When Timer1 overflows, the interrupt wakes up the device so that the appropriate registers can be updated. Figure 12-7: Timer1 Application
Power-Down Detect dsPIC30FXXX 8 OSC1 Current Sink S1 Backup Battery TMR1 CN1 S3 SOSCO 32.768 kHz SOSCI CN3 VSS CN2 S4 CN0 S2
12
Timers
VDD
In this example, a 32.768 kHz crystal is used as the time base for the Real-Time Clock. If the clock needs to be updated at 1 second intervals, then the period register, PR1, must be loaded with a value to allow the Timer1 to PR1 match at the desired rate. In the case of a 1 second Timer1 match event, the PR1 register should be loaded with a value of 0x8000. Note: The TMR1 register should never be written for correct real-time clock functionality, since the Timer1 clock source is asynchronous to the system clock. Writes to the TMR1 register may corrupt the real-time counter value, resulting in inaccurate timekeeping.
DS70059D-page 12-24

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Timers module are: Title Using Timer1 in Asynchronous Clock Mode Yet Another Clock Featuring the PIC16C924 Application Note # AN580 AN649
Note:
DS70059D-page 12-25
Section 12. Timers

Revision B
This revision incorporates technical content changes for the dsPIC30F Timers module.
Revision C
Revision D
Added example code in Section 12.12.1 Timer Operation in Sleep Mode.
12
Timers
DS70059D-page 12-26
DS70059D-page 12-27
Section 13. Input Capture

HIGHLIGHTS
This section of the manual contains the following major topics: 13.1 Introduction .................................................................................................................. 13-2 13.2 Input Capture Registers ............................................................................................... 13-3 13.3 Timer Selection ............................................................................................................ 13-4 13.4 Input Capture Event Modes ......................................................................................... 13-4 13.5 Capture Buffer Operation............................................................................................. 13-8 13.6 Input Capture Interrupts ............................................................................................... 13-9 13.7 UART Autobaud Support ............................................................................................. 13-9 13.8 Input Capture Operation in Power Saving States ...................................................... 13-10 13.9 I/O Pin Control ........................................................................................................... 13-10 13.10 Special Function Registers Associated with the Input Capture Module..................... 13-11 13.11 Design Tips ................................................................................................................ 13-12 13.12 Related Application Notes.......................................................................................... 13-13 13.13 Revision History ......................................................................................................... 13-14
13
Input Capture
DS70060C-page 13-1

13.1 Introduction
This section describes the Input Capture module and its associated Operational modes. The Input Capture module is used to capture a timer value from one of two selectable time bases, upon an event on an input pin. The Input Capture features are quite useful in applications requiring frequency (Time Period) and pulse measurement. Figure 13-1 depicts a simplified block diagram of the Input Capture module. Refer to the specific device data sheet for further information on the number of channels available in a particular device. All Input Capture channels are functionally identical. In this section, an x in the pin name or register name denotes the specific Input Capture channel. The Input Capture module has multiple Operating modes, which are selected via the ICxCON register. The Operating modes include: Capture timer value on every falling edge of input applied at the ICx pin Capture timer value on every rising edge of input applied at the ICx pin Capture timer value on every fourth rising edge of input applied at the ICx pin Capture timer value on every 16th rising edge of input applied at the ICx pin Capture timer value on every rising and every falling edge of input applied at the ICx pin
The Input Capture module has a four-level FIFO buffer. The number of capture events required to generate a CPU interrupt can be selected by the user. Figure 13-1: Input Capture Block Diagram
From 16-bit Timers TMRy TMRz
16
16
1 Prescaler Counter (1, 4, 16) ICx pin 3 Edge Detection Logic and Clock Synchronizer ICM<2:0>(ICxCON<2:0>) Mode Select ICBNE, ICOV(ICxCON<4:3>) FIFO R/W Logic
ICTMR (ICxCON<7>)
ICxBUF ICxI<1:0> ICxCON Interrupt Logic
System Bus
Set Flag ICxIF (in IFSn Register)
Note: An x in a signal, register or bit name denotes the number of the capture channel.
DS70060C-page 13-2
FIFO

13.2 Input Capture Registers
Each capture channel available on the dsPIC30F devices has the following registers, where x denotes the number of the capture channel: ICxCON: Input Capture Control Register ICxBUF: Input Capture Buffer Register Register 13-1: Upper Byte: U-0
ICxCON: Input Capture x Control Register U-0
R/W-0 ICSIDL
U-0
U-0
U-0
U-0
U-0
bit 15 Lower Byte: R/W-0 ICTMR bit 7
bit 8
R/W-0 R/W-0 ICI<1:0>
R-0, HC ICOV
R-0, HC ICBNE
R/W-0
R/W-0 ICM<2:0>
R/W-0 bit 0
bit 15-14 Unimplemented: Read as 0 bit 13 ICSIDL: Input Capture Module Stop in Idle Control bit 1 = Input capture module will halt in CPU Idle mode 0 = Input capture module will continue to operate in CPU Idle mode Unimplemented: Read as 0 ICTMR: Input Capture Timer Select bits 1 = TMR2 contents are captured on capture event 0 = TMR3 contents are captured on capture event Note: bit 6-5 Timer selections may vary. Refer to the device data sheet for details. ICI<1:0>: Select Number of Captures per Interrupt bits 11 = Interrupt on every fourth capture event 10 = Interrupt on every third capture event 01 = Interrupt on every second capture event 00 = Interrupt on every capture event ICOV: Input Capture Overflow Status Flag (Read Only) bit 1 = Input capture overflow occurred 0 = No input capture overflow occurred ICBNE: Input Capture Buffer Empty Status (Read Only) bit 1 = Input capture buffer is not empty, at least one more capture value can be read 0 = Input capture buffer is empty ICM<2:0>: Input Capture Mode Select bits 111 = Input Capture functions as interrupt pin only, when device is in Sleep or Idle mode (Rising edge detect only, all other control bits are not applicable.) 110 = Unused (module disabled) 101 = Capture mode, every 16th rising edge 100 = Capture mode, every 4th rising edge 011 = Capture mode, every rising edge 010 = Capture mode, every falling edge 001 = Capture mode, every edge (rising and falling) (ICI<1:0> does not control interrupt generation for this mode.) 000 = Input capture module turned off Legend: HC = Cleared in Hardware R = Readable bit -n = Value at POR HS = Set in Hardware W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 12-8 bit 7
13
Input Capture
bit 4
bit 3
bit 2-0
DS70060C-page 13-3

13.3 Timer Selection
Each dsPIC30F device may have one or more input capture channels. Each channel can select between one of two 16-bit timers for the time base. Refer to the device data sheet for the specific timers that can be selected. Selection of the timer resource is accomplished through the ICTMR control bit (ICxCON<7>). The timers can be setup using the internal clock source (FOSC/4), or using a synchronized external clock source applied at the TxCK pin.
13.4
Input Capture Event Modes

The input capture module captures the 16-bit value of the selected time base register when an event occurs at the ICx pin. The events that can be captured are listed below in three categories: 1. Simple Capture Event modes Capture timer value on every falling edge of input at ICx pin Capture timer value on every rising edge of input at ICx pin Capture timer value on every edge (rising and falling) Prescaler Capture Event modes Capture timer value on every 4th rising edge of input at ICx pin Capture timer value on every 16th rising edge of input at ICx pin
2. 3.
These Input Capture modes are configured by setting the appropriate Input Capture mode bits, ICM<2:0> (ICxCON<2:0>).
13.4.1
Simple Capture Events

The capture module can capture a timer count value (TMR2 or TMR3) based on the selected edge (rising or falling defined by mode) of the input applied to the ICx pin. These modes are specified by setting the ICM<2:0> (ICxCON<2:0>) bits to 010 or 011, respectively. In these modes, the prescaler counter is not used. See Figure 13-3 and Figure 13-2 for simplified timing diagrams of a simple capture event. The input capture logic detects and synchronizes the rising or falling edge of the capture pin signal on the internal phase clocks. If the rising/falling edge has occurred, the capture module logic will write the current time base value to the capture buffer and signal the interrupt generation logic. When the number of elapsed capture events matches the number specified by the ICI<1:0> control bits, the respective capture channel interrupt status flag, ICxIF, is asserted 2 instruction cycles after the capture buffer write event. If the capture time base increments every instruction cycle, the captured count value will be the value that was present 1 or 2 instruction cycles past the time of the event on the ICx pin. This time delay is a function of the actual ICx edge event related to the instruction cycle clock and delay associated with the input capture logic. If the input clock to the capture time base is prescaled, then the delay in the captured value can be eliminated. See Figure 13-3 and Figure 13-2 for details. The input capture pin has minimum high time and low time specifications. Refer to the Electrical Specifications section of the device data sheet for further details.
DS70060C-page 13-4

Figure 13-2: Simple Capture Event Timing Diagram, Time Base Prescaler = 1:1
TCY ICxIF Set
TMR2
n-3
n-2
n-1
n+1
n+2
n+3
n+4
n+5
ICx pin See Note 1 Capture Data n+1
Note 1: A capture signal edge that occurs in this region will result in a capture buffer entry value of 1 or 2 timer counts from the capture signal edge.
Figure 13-3:
Simple Capture Event Timing Diagram, Time Base Prescaler = 1:4

TCY ICxIF Set
TMRy
n-1
n+1
13
Input Capture
ICx pin
Capture Data
DS70060C-page 13-5

13.4.2 Prescaler Capture Events
The capture module has two Prescaled Capture modes. The Prescale modes are selected by setting the ICM<2:0> (ICxCON<2:0>) bits to 100 or 101, respectively. In these modes, the capture module counts four or sixteen rising edge pin events before a capture event occurs. The capture prescaler counter is incremented on every valid rising edge applied to the capture pin. The rising edge applied to the pin effectively serves as a clock to a counter. When the prescaler counter equals four or sixteen counts (depending on the mode selected), the counter will output a valid capture event signal, which is then synchronized to the instruction cycle clock. This synchronized capture event signal will trigger a capture buffer write event and signal the interrupt generation logic. The respective capture channel interrupt status flag, ICxIF, is asserted 2 instruction cycles after the capture buffer write event. If the capture time base increments every instruction cycle, the captured count value will be the value that was present 1 or 2 instruction cycles past the time of the synchronized capture event. The input capture pin has minimum high time and low time specifications. Refer to the Electrical Specifications section of the device data sheet for further details. Switching from one prescale setting to another may generate an interrupt. Also, the prescaler counter will not be cleared, therefore, the first capture may be from a non-zero prescaler. Example 13-1 shows the recommended method for switching between capture prescale settings. The prescaler counter is cleared when: The capture channel is turned off (i.e., ICM<2:0> = 000). Any device Reset. The prescaler counter is not cleared when: The user switches from one active Capture mode to another.
DS70060C-page 13-6

Example 13-1:
; ; ; ;
Prescaled Capture Code Example
The following code example will set the Input Capture 1 module for interrupts on every second capture event, capture on every fourth rising edge and select Timer 2 as the time-base. This code example clears ICxCON to avoid unexpected interrupts. BSET BCLR BCLR BCLR BSET CLR MOV MOV MOV MOV IPC0, IPC0, IPC0, IFS0, IEC0, #IC1IP0 #IC1IP1 #IC1IP2 #IC1IF #IC1IE ; ; ; ; ; Setup Input Capture 1 interrupt for desired priority level (this example assigns level 1 priority) Clear the IC1 interrupt status flag Enable IC1 interrupts
IC1CON #0x00A2, w0 w0, IC1CON #IC1BUF, w0 #TEMP_BUFF, w1
; Turn off Input Capture 1 Module. ; Load the working register with the new ; prescaler mode and write to IC1CON ; Create capture data fetch pointer ; Create data storage pointer ; Assumes TEMP_BUFF is already defined
; ; ;
The following code shows how to read the capture buffer when an interrupt is generated. W0 contains the capture buffer address. Example code for Input Capture 1 ISR:
__IC1Interrupt: BCLR IFS0, #IC1IF MOV [w0++], [w1++] MOV [w0], [w1] RETFIE
; ; ; ; ;
Reset respective interrupt flag Read and save off first capture entry Read and save off second capture entry Remaining user code here Return from ISR
13
Input Capture
Note:
It is recommended that the user turn off the capture module (i.e., clear ICM<2:0> (ICxCON<2:0>)) before switching to a new mode. If the user switches to a new Capture mode, the prescaler counter is not cleared. Therefore, it is possible that the first capture event and its associated interrupt is generated due to a non-zero prescaler counter (at the time of switching modes).
DS70060C-page 13-7

13.4.3 Edge Detection Mode
The capture module can capture a time base count value on every rising and falling edge of the input signal applied to the ICx pin. The Edge Detection mode is selected by setting the ICM<2:0> (ICxCON<2:0>) bits to 001. In this mode, the capture prescaler counter is not used. See Figure 13-4 for a simplified timing diagram. When the input capture module is configured for Edge Detection mode, the module will: Set the input capture interrupt flag (ICxIF) on every edge, rising and falling. The Interrupt-on-Capture mode bits, ICI<1:0> (ICxCON<6:5>), are not used in this mode. Every capture event will generate an interrupt. No capture overflow, ICOV (ICxCON<4>), bit is generated. As with the simple Capture Event mode, the input capture logic detects and synchronizes the rising and falling edge of the capture pin signal on the internal phase clocks. If the rising or falling edge has occurred, the capture module logic will write the current timer count to the capture buffer and signal the interrupt generation logic. The respective capture channel interrupt status flag, ICxIF, is asserted 2 instruction cycles after the capture buffer write event. The captured timer count value will be 1 or 2 TCY (instruction cycles) past the time of the occurrence of the edge at the ICx pin (see Figure 13-4). Figure 13-4: Edge Detection Mode Timing Diagram
TCY ICxIF Set ICxIF Set
TMRy
n-3
n-2
n-1
n+1
n+2
n+3
n+4
n+5
n+6
ICx pin
Capture Data
n+4
13.5
Capture Buffer Operation

Each capture channel has an associated four-deep FIFO buffer. The ICxBUF register is the buffer register visible to the user, as it is memory mapped. When the input capture module is reset, ICM<2:0> = 000 (ICxCON<2:0>), the input capture logic will: Clear the overflow condition flag (i.e., clear ICxOV (ICxCON<4>) to 0). Reset the capture buffer to the empty state (i.e., clears ICBNE (ICxCON<3>) to 0). Reading the FIFO buffer under the following conditions will lead to indeterminate results: In the event the input capture module is first disabled and at some later time re-enabled. In the event a FIFO read is performed when the buffer is empty. After a device Reset. There are two status flags which provide status on the FIFO buffer: ICBNE (ICxCON<3>): Input Capture Buffer Not Empty ICOV (ICxCON<4>): Input Capture Overflow
DS70060C-page 13-8

13.5.1 Input Capture Buffer Not Empty (ICBNE)
The ICBNE read only Status bit (ICxCON<3>) will be set on the first input capture event and remain set until all capture events have been read from the capture buffer. For example, if three capture events have occurred, then three reads of the capture buffer are required before the ICBNE (ICxCON<3>) flag will be cleared. If four capture events, then four reads are required to clear the ICBNE (ICxCON<3>) flag. Each read of the capture buffer will allow the remaining word(s) to move to the next available top location. Since the ICBNE reflects the capture buffer state, the ICBNE Status bit will be cleared in the event of any device Reset.
13.5.2
Input Capture Overflow (ICOV)

The ICOV read only Status bit (ICxCON<4>) will be set when the capture buffer overflows. In the event that buffer is full with four capture events and a fifth capture event occurs prior to a read of the buffer, an overrun condition will occur, the ICOV (ICxCON<4>) bit will be set to a logic 1 and the respective capture event interrupt will not be generated. In addition, the fifth capture event is not recorded and all subsequent capture events will not alter the current buffer contents. To clear the overrun condition, the capture buffer must be read four times. Upon the fourth read, the ICOV (ICxCON<4>) status flag will be cleared and the capture channel will resume normal operation. Clearing of the overflow condition can be accomplished in the following ways: Set ICM<2:0> (ICxCON<2:0>) = 000 Read capture buffer until ICBNE (ICxCON<3>) = 0 Any device Reset
13.5.2.1
ICOV and Interrupt Only Mode The input capture module can also be configured to function as an external interrupt pin. For this mode, the ICI<1:0> (ICxCON<6:5>) bits must be set to 00. Interrupts will be generated independently of buffer reads.
13
Input Capture
13.6
Input Capture Interrupts

The input capture module has the ability to generate an interrupt based upon a selected number of capture events. A capture event is defined as a write of a time base value into the capture buffer. This setting is configured by the control bits ICI<1:0> (ICxCON<6:5>). Except for the case when ICI<1:0> = 00, no interrupts will be generated until a buffer overflow condition is removed (see Section 13.5.2 Input Capture Overflow (ICOV)). When the capture buffer has been emptied, either by a Reset condition or a read operation, the interrupt count is reset. This allows for the resynchronization of the interrupt count to the FIFO entry status.
13.6.1
Interrupt Control Bits

Each input capture channel has interrupt flag Status bits (ICxIF), interrupt enable bits (ICxIE) and interrupt priority control bits (ICxIP<2:0>). Refer to Section 6. Reset Interrupts for further information on peripheral interrupts.
13.7
UART Autobaud Support

The input capture module can be used by the UART module when the UART is configured for the Autobaud mode of operation, ABAUD = 1 (UxMODE<5>). When the ABAUD control bit is set, the UART RX pin will be internally connected to the assigned input capture module input. The I/O pin associated with the capture module will be disconnected. The baud rate can be determined by measuring the width of the Start bit when a NULL character is received. Note that the capture module must be configured for the Edge Detection mode (capture on every rising and falling edge) to take advantage of the autobaud feature. The input capture module assignment for each UART will depend on the dsPIC30F device variant that is selected. Refer to the device data sheet for further details on the autobaud support.
DS70060C-page 13-9

13.8 13.8.1 Input Capture Operation in Power Saving States Input Capture Operation in Sleep Mode
When the device enters Sleep mode, the system clock is disabled. In Sleep mode, the input capture module can only function as an external interrupt source. This mode is enabled by setting control bits ICM<2:0> = 111. In this mode, a rising edge on the capture pin will generate device wake-up from Sleep condition. If the respective module interrupt bit is enabled and the module priority is of the required priority, an interrupt will be generated. In the event the capture module has been configured for a mode other than ICM<2:0> = 111 and the dsPIC30F does enter the Sleep mode, no external pin stimulus, rising or falling, will generate a wake-up condition from Sleep.
13.8.2
Input Capture Operation in Idle Mode

When the device enters Idle mode, the system clock sources remain functional and the CPU stops executing code. The ICSIDL bit (ICxCON<13>) selects if the module will stop in Idle mode, or continue to operate in Idle mode. If ICSIDL = 0 (ICxCON<13>), the module will continue operation in Idle mode. Full functionality of the input capture module is provided for, including the 4:1 and 16:1 capture prescale settings, defined by control bits ICM<2:0> (ICxCON<2:0>). These modes require that the selected timer is enabled during Idle mode as well. If the Input Capture mode is configured for ICM<2:0> = 111, the input capture pin will serve only as an external interrupt pin. In this mode, a rising edge on the capture pin will generate device wake-up from Idle mode. A capture time base does not have to be enabled. If the respective module interrupt enable bit is set and the user assigned priority is greater than the current CPU priority level, an interrupt will be generated. If ICSIDL = 1 (ICxCON<13>), the module will stop in Idle mode. The module will perform the same functions when stopped in Idle mode as for Sleep mode (see Section 13.8.1 Input Capture Operation in Sleep Mode).
13.8.3
Device Wake-up on Sleep/Idle

An input capture event can generate a device wake-up or interrupt, if enabled, if the device is in Idle or Sleep mode. Independent of the timer being enabled, the input capture module will wake-up from Sleep or Idle mode when a capture event occurs, if the following are true: Input Capture mode bits, ICM<2:0> = 111 (ICxCON<2:0>) and The interrupt enable bit (ICxIE) is asserted. This same wake-up feature will interrupt the CPU if: The respective interrupt is enabled (ICxIE = 1) and is of the required priority. This wake-up feature is quite useful for adding extra external pin interrupts. The following conditions are true when the input capture module is used in this mode: The capture prescaler counter is not utilized while in this mode. The ICI<1:0>(ICxCON<6:5>) bits are not applicable.
13.9
I/O Pin Control

When the capture module is enabled, the user must ensure the I/O pin direction is configured for an input by setting the associated TRIS bit. The pin direction is not set when the capture module is enabled. Furthermore, all other peripherals multiplexed with the input pin must be disabled.
DS70060C-page 13-10
13.10
Special Function Registers Associated with the Input Capture Module
Table 13-1:
Bit 13 SI2CIF IC4IF SI2CIE IC4IE T1IP<2:0> IC1IP<2:0> OC2IP<2:0> IC7IP<2:0> IC4IP<2:0> T31P<2:0> OC3IP<2:0> IC6IP<2:0> Input 1 Capture Register ICSIDL Input 2 Capture Register ICSIDL Input 3 Capture Register ICSIDL Input 4 Capture Register ICSIDL Input 5 Capture Register ICSIDL Input 6 Capture Register ICSIDL ICTMR ICI<1:0> ICOV ICBNE ICM<2:0> ICTMR ICI<1:0> ICOV ICBNE ICM<2:0> ICTMR ICI<1:0> ICOV ICBNE ICM<2:0> ICTMR ICI<1:0> ICOV ICBNE ICM<2:0> ICTMR ICI<1:0> ICOV ICBNE ICM<2:0> ICTMR ICI<1:0> ICOV ICBNE ICM<2:0> IC5IP<2:0> IC8IP<2:0> T2IP<2:0> OC1IP<2:0> INT0IP<2:0> IC2IP<2:0> INT1IP<2:0> IC3IP<2:0> IC3IE C1IE SPI2IE U2TXIE U2RXIE INT2IE T5IE T4IE OC4IE OC3IE IC8IE IC7IE IR12 ADIE U1TXIE U1RXIE SPI1IE T3IE T2IE OC2IE IC2IE T1IE OC1IE IC1IE INT0IE INT1IE IC3IF C1IF SPI2IF U2TXIF U2RXIF INT2IF T5IF T4IF OC4IF OC3IF IC8IF IC7IF INT1IF NVMIF ADIF U1TXIF U1RXIF SPI1IF T3IF T2IF OC2IF IC2IF T1IF OC1IF IC1IF INT0F Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
Example Memory Map for Input Capture Modules

Reset State 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 uuuu uuuu uuuu uuuu 0000 0000 0000 0000 uuuu uuuu uuuu uuuu 0000 0000 0000 0000 uuuu uuuu uuuu uuuu 0000 0000 0000 0000 uuuu uuuu uuuu uuuu 0000 0000 0000 0000 uuuu uuuu uuuu uuuu 0000 0000 0000 0000 uuuu uuuu uuuu uuuu 0000 0000 0000 0000 uuuu uuuu uuuu uuuu ICTMR ICI<1:0> ICOV ICBNE ICM<2:0> 0000 0000 0000 0000 uuuu uuuu uuuu uuuu ICTMR ICI<1:0> ICOV ICBNE ICM<2:0> 0000 0000 0000 0000
SFR Name
Addr.
Bit 15
Bit 14
IFS0
0084
CNIF
MI2CIF
IFS1
0086
IC6IF
IC5IF
IEC0
008C
CNIE
MI2CIE
IEC1
008E
IC6IE
EI30
13
Input Capture

Input 7 Capture Register ICSIDL Input 8 Capture Register ICSIDL
IPC0
0094
IPC1
0096
IPC4
009C
IPC7
00A2
IC1BUF
0140
IC1CON
0142
IC2BUF
0144
IC2CON
0146
IC3BUF
0148
IC3CON
014A
IC4BUF
014C
IC4CON
014E
IC5BUF
0150
IC5CON
0152
IC6BUF
0154
IC6CON
0156
IC7BUF
0158
IC7CON
015A
IC8BUF
015C
IC8CON
015E
Legend: Note:
u = uninitialized Refer to the device data sheet for specific memory map details.
DS70060C-page 13-11

13.11 Design Tips
Question 1: Can the Input Capture module be used to wake the device from Sleep mode?
Answer: Yes. When the Input Capture module is configured to ICM<2:0> = 111 and the respective channel interrupt enable bit is asserted, ICxIE = 1, a rising edge on the capture pin will wake-up the device from Sleep (see Section 13.8 Input Capture Operation in Power Saving States).
DS70060C-page 13-12

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Input Capture module are: Title Using the CCP Module(s) Implementing Ultrasonic Ranging Application Note # AN594 AN597
Note:
13
Input Capture
DS70060C-page 13-13

Revision B
Revision C
DS70060C-page 13-14
Section 14. Output Compare

HIGHLIGHTS
This section of the manual contains the following major topics: 14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8 Introduction .................................................................................................................. 14-2 Output Compare Registers .......................................................................................... 14-3 Modes of Operation ..................................................................................................... 14-4 Output Compare Operation in Power Saving States.................................................. 14-23 I/O Pin Control ........................................................................................................... 14-23 Design Tips ................................................................................................................ 14-26 Related Application Notes.......................................................................................... 14-27 Revision History ......................................................................................................... 14-28
14
Output Compare
DS70061C-page 14-1

14.1 Introduction
The Output Compare module has the ability to compare the value of a selected time base with the value of one or two compare registers (depending on the Operation mode selected). Furthermore, it has the ability to generate a single output pulse, or a train of output pulses, on a compare match event. Like most dsPIC peripherals, it also has the ability to generate interrupts-on- compare match events. The dsPIC30F device may have up to eight output compare channels, designated OC1, OC2, OC3, etc. Refer to the specific device data sheet for the number of channels available in a particular device. All output compare channels are functionally identical. In this section, an x in the pin, register or bit name denotes the specific output compare channel. Each output compare channel can use one of two selectable time bases. The time base is selected using the OCTSEL bit (OCxCON<3>). Please refer to the device data sheet for the specific timers that can be used with each output compare channel number. Figure 14-1: Output Compare Module Block Diagram
Set Flag bit OCxIF(1)
OCxRS(1)
OCxR(1)
Output Logic 3 OCM<2:0> Mode Select
S R
OCx(1) Output Enable OCFA or OCFB (see Note 2)
Comparator
OCTSEL
16
16
TMR register inputs from time bases (see Note 3).
Period match signals from time bases (see Note 3).
Note 1: Where x is shown, reference is made to the registers associated with the respective output compare channels 1 through 8. 2: OCFA pin controls OC1-OC4 channels. OCFB pin controls OC5-OC8 channels. 3: Each output compare channel can use one of two selectable time bases. Refer to the device data sheet for the time bases associated with the module.
DS70061C-page 14-2

14.2 Output Compare Registers
Each output compare channel has the following registers: OCxCON: the control register for the channel OCxR: a data register for the output compare channel OCxRS: a secondary data register for the output compare channel The control registers for the 8 compare channels are named OC1CON through OC8CON. All 8 control registers have identical bit definitions. They are represented by a common register definition below. The x in OCxCON represents the output compare channel number. Register 14-1: Upper Byte: U-0 bit 15 OCxCON: Output Compare x Control Register U-0 R/W-0 OCSIDL U-0 U-0 U-0 U-0 U-0 bit 8
Lower Byte: U-0 bit 7
U-0
U-0
R-0, HC OCFLT
R/W-0 OCTSEL
R/W-0
R/W-0 OCM<2:0>
R/W-0 bit 0
bit 15-14 Unimplemented: Read as 0 bit 13 OCSIDL: Stop Output Compare in Idle Mode Control bit 1 = Output compare x will halt in CPU Idle mode 0 = Output compare x will continue to operate in CPU Idle mode Unimplemented: Read as 0 OCFLT: PWM Fault Condition Status bit 1 = PWM Fault condition has occurred (cleared in HW only) 0 = No PWM Fault condition has occurred (This bit is only used when OCM<2:0> = 111.) OCTSEL: Output Compare Timer Select bit 1 = Timer3 is the clock source for compare x 0 = Timer2 is the clock source for compare x Note: bit 2-0 Refer to the device data sheet for specific time bases available to the output compare module.
bit 12-5 bit 4
bit 3
14
Output Compare
OCM<2:0>: Output Compare Mode Select bits 111 = PWM mode on OCx, Fault pin enabled 110 = PWM mode on OCx, Fault pin disabled 101 = Initialize OCx pin low, generate continuous output pulses on OCx pin 100 = Initialize OCx pin low, generate single output pulse on OCx pin 011 = Compare event toggles OCx pin 010 = Initialize OCx pin high, compare event forces OCx pin low 001 = Initialize OCx pin low, compare event forces OCx pin high 000 = Output compare channel is disabled Legend: HC = Cleared in Hardware R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70061C-page 14-3

Each output compare module has the following modes of operation: Single Compare Match mode Dual Compare Match mode generating - Single Output Pulse - Continuous Output Pulses Simple Pulse Width Modulation mode - with Fault Protection Input - without Fault Protection Input Note 1: It is recommended that the user turn off the output compare module (i.e., clear OCM<2:0> (OCxCON<2:0>)) before switching to a new mode. 2: In this section, a reference to any SFRs associated with the selected timer source is indicated by a y suffix. For example, PRy is the Period register for the selected timer source, while TyCON is the Timer Control register for the selected timer source.
14.3.1
Single Compare Match Mode

When control bits OCM<2:0> (OCxCON<2:0>) are set to 001, 010 or 011, the selected output compare channel is configured for one of three Single Output Compare Match modes. In the Single Compare mode, the OCxR register is loaded with a value and is compared to the selected incrementing timer register, TMRy. On a compare match event, one of the following events will take place: Compare forces OCx pin high, initial state of pin is low. Interrupt is generated on the single compare match event. Compare forces OCx pin low, initial state of pin is high. Interrupt is generated on the single compare match event. Compare toggles OCx pin. Toggle event is continuous and an interrupt is generated for each toggle event.
DS70061C-page 14-4

14.3.1.1 Compare Mode Output Driven High To configure the output compare module for this mode, set control bits OCM<2:0> = 001. The compare time base should also be enabled. Once this Compare mode has been enabled, the output pin, OCx, will be initially driven low and remain low until a match occurs between the TMRy and OCxR registers. Referring to Figure 14-2, there are some key timing events to note: The OCx pin is driven high one instruction clock after the compare match occurs between the compare time base and the OCxR register. The OCx pin will remain high until a mode change has been made, or the module is disabled. The compare time base will count up to the value contained in the associated period register and then reset to 0x0000 on the next instruction clock. The respective channel interrupt flag, OCxIF, is asserted 2 instruction clocks after the OCx pin is driven high. Figure 14-2: Single Compare Mode: Set OCx High on Compare Match Event
1 Instruction Clock Period
TMRy
3000
3001
3002
3003
3004
3FFF
4000
0000
0001
TMRy Resets Here PRy OCxR 4000 3002 Cleared by User
OCx pin OCxIF
2 TCY
Note: An x represents the output compare channel number. A y represents the time base number.
14
Output Compare
DS70061C-page 14-5

14.3.1.2 Compare Mode Output Driven Low To configure the output compare module for this mode, set control bits OCM<2:0> = 010. The compare time base must also be enabled. Once this Compare mode has been enabled, the output pin, OCx, will be initially driven high and remain high until a match occurs between the Timer and OCxR registers. Referring to Figure 14-3, there are some key timing events to note: The OCx pin is driven low one instruction clock after the compare match occurs between the compare time base and the OCxR register. The OCx pin will remain low until a mode change has been made, or the module is disabled. The compare time base will count up to the value contained in the associated period register and then reset to 0x0000 on the next instruction clock. The respective channel interrupt flag, OCxIF, is asserted 2 instruction clocks after OCx pin is driven low. Figure 14-3: Single Compare Mode: Force OCx Low on Compare Match Event
TMRy
47FE
47FF
4800
4801
4802
4BFF
4C00
0000
0001
TMRy Resets Here PRy OCxR OCx pin 2 TCY OCxIF Cleared by User 4C00 4800
DS70061C-page 14-6

14.3.1.3 Single Compare Mode Toggle Output To configure the output compare module for this mode, set control bits OCM<2:0> = 011. In addition, Timer 2 or Timer 3 must be selected and enabled. Once this Compare mode has been enabled, the output pin, OCx, will be initially driven low and then toggle on each and every subsequent match event between the Timer and OCxR registers. Referring to Figure 14-4 and Figure 14-5, there are some key timing events to note: The OCx pin is toggled one instruction clock after the compare match occurs between the compare time base and the OCxR register. The OCx pin will remain at this new state until the next toggle event, or until a mode change has been made, or the module is disabled. The compare time base will count up to the contents in the period register and then reset to 0x0000 on the next instruction clock. The respective channel interrupt flag, OCxIF, is asserted 2 instruction clocks after the OCx pin is toggled. Note: The internal OCx pin output logic is set to a logic 0 on a device Reset. However, the operational OCx pin state for the Toggle mode can be set by the user software. Example 14-1 shows a code example for defining the desired initial OCx pin state in the Toggle mode of operation.
Figure 14-4:
Single Compare Mode: Toggle Output on Compare Match Event (PR2 > OCxR)
TMRy
0500
0501
0502
0600
0000
0001 TMRy Resets Here
0500
0501
0502
PRy OCxR OCx pin
0600 0500
2 TCY OCxIF
Cleared by User
14
Figure 14-5: Single Compare Mode: Toggle Output on Compare Match Event (PR2 = OCxR)
Output Compare
TMR2
0500
0000
0500
0000
0500
0000
0001
PR2 OCxR OCx pin
0500 0500
2 TCY OCxIF
2 TCY
2 TCY
Cleared by User
Cleared by User
DS70061C-page 14-7

Example 14-1:
; ; ;
Compare Mode Toggle Mode Pin State Setup
The following code example illustrates how to define the initial OC1 pin state for the output compare toggle mode of operation. Toggle mode with initial OC1 pin state set low MOV MOV BSET 0x0001, w0 w0, OC1CON OC1CON, #1 ; ; ; ; load setup value into w0 enable module for OC1 pin low, toggle high set module to toggle mode with initial pin state low
Toggle mode with initial OC1 pin state set high MOV MOV BSET 0x0002, w0 w0, OC1CON OC1CON, #0 ; ; ; ; load setup value into w0 enable module for OC1 pin high, toggle low set module to toggle mode with initial pin state high
Example 14-2 shows example code for the configuration and interrupt service of the Single Compare mode toggle event. Example 14-2:
; ; ; ; ;
Compare Mode Toggle Setup and Interrupt Servicing
The following code example will set the Output Compare 1 module for interrupts on the toggle event and select Timer 2 as the clock source for the compare time-base. It is assumed in that Timer 2 and Period Register 2 are properly configured. Timer 2 will be enabled here. CLR MOV MOV MOV MOV BSET BCLR BCLR BCLR BSET BSET OC1CON #0x0003, w0 w0, OC1CON #0x0500, w0 w0, OC1R IPC0, #OC1IP0 IPC0, #OC1IP1 IPC0, #OC1IP2 IFS0, #OC1IF IEC0, #OC1IE T2CON, #TON ; ; ; ; ; ; ; ; ; ; ; Turn off Output Compare 1 Module. Load the working register with the new compare mode and write to OC1CON Initialize Compare Register 1 with 0x0500 Setup Output Compare 1 interrupt for desired priority level (this example assigns level 1 priority) Clear Output Compare 1 interrupt flag Enable Output Compare 1 interrupts Start Timer2 with assumed settings
Example code for Output Compare 1 ISR:
__OC1Interrupt: BCLR IFS0, RETFIE
#OC1IF
; Reset respective interrupt flag ; Remaining user code here ; Return from ISR
DS70061C-page 14-8

14.3.2 Dual Compare Match Mode
When control bits OCM<2:0> = 100 or 101 (OCxCON<2:0>), the selected output compare channel is configured for one of two Dual Compare Match modes which are: Single Output Pulse mode Continuous Output Pulse mode In the Dual Compare mode, the module uses both the OCxR and OCxRS registers for the compare match events. The OCxR register is compared against the incrementing timer count, TMRy, and the leading (rising) edge of the pulse is generated at the OCx pin, on a compare match event. The OCxRS register is then compared to the same incrementing timer count, TMRy, and the trailing (falling) edge of the pulse is generated at the OCx pin, on a compare match event. 14.3.2.1 Dual Compare Mode: Single Output Pulse To configure the Output Compare module for the Single Output Pulse mode, set control bits OCM<2:0> = 100. In addition, the compare time base must be selected and enabled. Once this mode has been enabled, the output pin, OCx, will be driven low and remain low until a match occurs between the time base and OCxR registers. Referring to Figure 14-6 and Figure 14-7, there are some key timing events to note: The OCx pin is driven high one instruction clock after the compare match occurs between the compare time base and OCxR register. The OCx pin will remain high until the next match event occurs between the time base and the OCxRS register. At this time, the pin will be driven low. The OCx pin will remain low until a mode change has been made, or the module is disabled. The compare time base will count up to the value contained in the associated period register and then reset to 0x0000 on the next instruction clock. If the time base period register contents are less than the OCxRS register contents, then no falling edge of the pulse is generated. The OCx pin will remain high until OCxRS <= PRy, or a mode change or Reset condition has occurred. The respective channel interrupt flag, OCxIF, is asserted 2 instruction clocks after the OCx pin is driven low (falling edge of single pulse). Figure 14-6 depicts the General Dual Compare mode generating a single output pulse. Figure 14-7 depicts another timing example where OCxRS > PRy. In this example, no falling edge of the pulse is generated since the compare time base resets before counting up to 0x4100.
14
Output Compare
DS70061C-page 14-9

Figure 14-6: Dual Compare Mode
1 Instruction Clock Period 4000 TMRy Resets Here PRy OCxR OCxRS 4000 3000 3003 0000
TMRy
3000
3001
3002
3003
3004
3005
3006
OCx pin OCxIF
2 TCY Cleared by User
Note 1: An x represents the output compare channel number. A y represents the time base number. 2: OCxR = Compare Register, OCxRS = Secondary Compare Register.
Figure 14-7:
Dual Compare Mode: Single Output Pulse (OCxRS > PR2)

TMRy
3000
3001
3002
3003
3004
3005
4000
0000
PRy OCxR OCxRS
4000 3000 4100
OCx pin Compare Interrupt does not occur OCxIF
DS70061C-page 14-10

14.3.2.2 Setup for Single Output Pulse Generation When control bits OCM<2:0> (OCxCON<2:0>) are set to 100, the selected output compare channel initializes the OCx pin to the low state and generates a single output pulse. To generate a single output pulse, the following steps are required (these steps assume timer source is initially turned off, but this is not a requirement for the module operation): 1. Determine the instruction clock cycle time. Take into account the frequency of the external clock to the timer source (if one is used) and the timer prescaler settings. 2. Calculate time to the rising edge of the output pulse relative to the TMRy start value (0x0000). 3. Calculate the time to the falling edge of the pulse based on the desired pulse width and the time to the rising edge of the pulse. 4. Write the values computed in steps 2 and 3 above into the compare register, OCxR, and the secondary compare register, OCxRS, respectively. 5. Set timer period register, PRy, to value equal to or greater than value in OCxRS, the secondary compare register. 6. Set OCM<2:0> = 100 and the OCTSEL (OCxCON<3>) bit to the desired timer source. The OCx pin state will now be driven low. 7. Set the TON (TyCON<15>) bit to 1, which enables the compare time base to count. 8. Upon the first match between TMRy and OCxR, the OCx pin will be driven high. 9. When the incrementing timer, TMRy, matches the secondary compare register, OCxRS, the second and trailing edge (high-to-low) of the pulse is driven onto the OCx pin. No additional pulses are driven onto the OCx pin and it remains at low. As a result of the second compare match event, the OCxIF interrupt flag bit set, which will result in an interrupt if it is enabled, by setting the OCxIE bit. For further information on peripheral interrupts, refer to Section 6. Reset Interrupts. 10. To initiate another single pulse output, change the timer and compare register settings, if needed, and then issue a write to set OCM<2:0> (OCxCON<2:0>) bits to 100. Disabling and re-enabling of the timer and clearing the TMRy register are not required, but may be advantageous for defining a pulse from a known event time boundary. The output compare module does not have to be disabled after the falling edge of the output pulse. Another pulse can be initiated by rewriting the value of the OCxCON register.
14
Output Compare
DS70061C-page 14-11

Example 14-3 shows example code for configuration of the single output pulse event. Example 14-3:
; ; ; ; ;
Single Output Pulse Setup and Interrupt Servicing
The following code example will set the Output Compare 1 module for interrupts on the single pulse event and select Timer 2 as the clock source for the compare time base. It is assumed that Timer 2 and Period Register 2 are properly initialized. Timer 2 will be enabled here. CLR MOV MOV MOV MOV MOV MOV BSET BCLR BCLR BCLR BSET BSET OC1CON #0x0004, w0 W0, OC1CON #0x3000, w0 W0, OC1R #0x3003, w0 W0, OC1RS IPC0, #OC1IP0 IPC0, #OC1IP1 IPC0, #OC1IP2 IFS0, #OC1IF IEC0, #OC1IE T2CON, #TON ; ; ; ; ; ; ; ; ; ; ; ; Turn off Output Compare 1 Module. Load the working register with the new compare mode and write to OC1CON Initialize Compare Register 1 with 0x3000 Initialize Secondary Compare Register 1 with 0x3003 Setup Output Compare 1 interrupt for desired priority level (this example assigns level 1 priority) Clear Output Compare 1 interrupt flag Enable Output Compare 1 interrupts
; Start Timer2 with assumed settings
#OC1IF
DS70061C-page 14-12

14.3.2.3 Special Cases for Dual Compare Mode Generating a Single Output Pulse Depending on the relationship of the OCxR, OCxRS and PRy values, the output compare module has a few unique conditions which should be understood. These special conditions are specified in Table 14-1, along with the resulting behavior of the module. Table 14-1: Special Cases for Dual Compare Mode Generating a Single Output Pulse
Special Conditions OCxR = 0 Initialize TMRy = 0 Operation In the first iteration of the TMRy counting from 0x0000 up to PRy, the OCx pin remains low, no pulse is generated. After the TMRy resets to zero (on period match), the OCx pin goes high due to match with OCxR. Upon the next TMRy to OCxRS match, the OCx pin goes low and remains there. The OCxIF bit will be set as a result of the second compare. There are two alternative initial conditions to consider: a] Initialize TMRy = 0 and set OCxR >= 1 b] Initialize TMRy = PRy (PRy > 0) and set OCxR = 0 Output at OCx Pulse will be delayed by the value in the PRy register depending on setup
SFR Logical Relationship PRy >= OCxRS and OCxRS > OCxR
PRy >= OCxR and OCxR >= OCxRS
OCxR >= 1 and PRy >= 1
TMRy counts up to OCxR and on a compare match event (i.e., Pulse TMRy = OCxR), the OCx pin is driven to a high state. TMRy then continues to count and eventually resets on period match (i.e., PRy =TMRy). The timer then restarts from 0x0000 and counts up to OCxRS, and on a compare match event (i.e., TMRy = OCxRS), the OCx pin is driven to a low state. The OCxIF bit will be set as a result of the second compare. Only the rising edge will be generated at the OCx pin. The OCxIF will not be set. Rising edge/ transition to high
OCxRS > PRy and PRy >= OCxR OCxR = OCxRS = PRy = 0x0000
None None
An output pulse delayed 2 instruction clock periods upon the Delayed pulse match of the timer and period register is generated at the OCx pin. The OCxIF bit will be set as a result of the second compare. Unsupported mode, timer resets prior to match condition. Remains low
OCxR > PRy
None
Note 1: In all the cases considered herein, the TMRy register is assumed to be initialized to 0x0000. 2: OCxR = Compare Register, OCxRS = Secondary Compare Register, TMRy = Timery Count, PRy = Timery Period Register.
14
Output Compare
DS70061C-page 14-13

14.3.2.4 Dual Compare Mode: Continuous Output Pulses To configure the output compare module for this mode, set control bits OCM<2:0> = 101. In addition, the compare time base must be selected and enabled. Once this mode has been enabled, the output pin, OCx, will be driven low and remain low until a match occurs between the compare time base and OCxR register. Referring to Figure 14-8 and Figure 14.3.2.5, there are some key timing events to note: The OCx pin is driven high one instruction clock after the compare match occurs between the compare time base and OCxR register. The OCx pin will remain high until the next match event occurs between the time base and the OCxRS register, at which time the pin will be driven low. This pulse generation sequence of a low-to-high and high-to-low edge will repeat on the OCx pin without further user intervention. Continuous pulses will be generated on the OCx pin until a mode change is made, or the module is disabled. The compare time base will count up to the value contained in the associated period register and then reset to 0x0000 on the next instruction clock. If the compare time base period register value is less than the OCxRS register value, then no falling edge is generated. The OCx pin will remain high until OCxRS <= PR2, a mode change is made, or the device is reset. The respective channel interrupt flag, OCxIF, is asserted 2 instruction clocks after the OCx pin is driven low (falling edge of single pulse). Figure 14-8 depicts the General Dual Compare mode generating a continuous output pulse. Figure 14.3.2.5 depicts another timing example where OCxRS > PRy. In this example, no falling edge of the pulse is generated, since the time base will reset before counting up to the contents of OCxRS. Figure 14-8: Dual Compare Mode: Continuous Output Pulse (PR2 = OCxRS)
1 Instruction Clock Period 3002 3003 3000
TMRy
3000
3001
3002
3003
3000
3001
0000
TMRy Resets Here
PRy OCxR OCxRS
3003 3000 3003
OCx pin 2 TCY OCxIF Cleared by User 2 TCY
DS70061C-page 14-14

Figure 14-9: Dual Compare Mode: Continuous Output Pulse (PR2 = OCxRS)
1 Instruction Clock Period TMRy 3000 3001 3002 3003 0000 TMRy Resets Here PRy OCxR OCxRS 3003 3000 3003 3000 3001 3002 3003 0000 3000
TMRy Resets Here
OCx pin Compare Interrupt does not Occur OCxIF
14.3.2.5
Setup for Continuous Output Pulse Generation When control bits OCxM<2:0> (OCxCON<2:0>) are set to 101, the selected output compare channel initializes the OCx pin to the low state and generates output pulses on each and every compare match event. For the user to configure the module for the generation of a continuous stream of output pulses, the following steps are required (these steps assume timer source is initially turned off, but this is not a requirement for the module operation): 1. Determine the instruction clock cycle time. Take into account the frequency of the external clock to the timer source (if one is used) and the timer prescaler settings. 2. Calculate time to the rising edge of the output pulse relative to the TMRy start value (0x0000). 3. Calculate the time to the falling edge of the pulse, based on the desired pulse width and the time to the rising edge of the pulse. 4. Write the values computed in step 2 and 3 above into the compare register, OCxR, and the secondary compare register, OCxRS, respectively. 5. Set timer period register, PRy, to value equal to or greater than value in OCxRS, the secondary compare register. 6. Set OCM<2:0> = 101 and the OCTSEL (OCxCON<3>) bit to the desired timer source. The OCx pin state will now be driven low. 7. Enable the compare time base by setting the TON (TyCON<15>) bit to 1. 8. Upon the first match between TMRy and OCxR, the OCx pin will be driven high. 9. When the compare time base, TMRy, matches the secondary compare register, OCxRS, the second and trailing edge (high-to-low) of the pulse is driven onto the OCx pin. 10. As a result of the second compare match event, the OCxIF interrupt flag bit set. 11. When the compare time base and the value in its respective period register match, the TMRy register resets to 0x0000 and resumes counting. 12. Steps 8 through 11 are repeated and a continuous stream of pulses is generated, indefinitely. The OCxIF flag is set on each OCxRS-TMRy compare match event.
14
Output Compare
DS70061C-page 14-15

Example 14-4 shows example code for configuration of the continuous output pulse event. Example 14-4:
; ; ; ; ;
Continuous Output Pulse Setup and Interrupt Servicing
The following code example will set the Output Compare 1 module for interrupts on the continuous pulse event and select Timer 2 as the clock source for the compare time-base. It is assumed that Timer 2 and Period Register 2 are properly configured. Timer 2 will be enabled here. CLR MOV MOV MOV MOV MOV MOV BSET BCLR BCLR BCLR BSET BSET OC1CON #0x0005, W0 W0, OC1CON #0x3000, W0 W0, OC1R #0x3003, W0 W0, OC1RS IPC0, #OC1IP0 IPC0, #OC1IP1 IPC0, #OC1IP2 IFS0, #OC1IF IEC0, #OC1IE T2CON, #TON ; ; ; ; ; ; ; ; ; ; ; ; Turn off Output Compare 1 Module. Load the working register with the new compare mode and write to OC1CON Initialize Compare Register 1 with 0x3000 Initialize Secondary Compare Register 1 with 0x3003 Setup Output Compare 1 interrupt for desired priority level (this example assigns level 1 priority) Clear Output Compare 1 interrupt flag Enable Output Compare 1 interrupts
; Start Timer2 with assumed settings
#OC1IF
DS70061C-page 14-16

14.3.2.6 Special Cases for Dual Compare Mode Generating Continuous Output Pulses Depending on the relationship of the OCxR, OCxRS and PRy values, the output compare module may not provide the expected results. These special cases are specified in Table 14-2, along with the resulting behavior of the module. Table 14-2: Special Cases for Dual Compare Mode Generating Continuous Output Pulses
Special Conditions OCxR = 0 Initialize TMRy = 0 Operation In the first iteration of the TMRy counting from 0x0000 up to PRy, the OCx pin remains low, no pulse is generated. After the TMRy resets to zero (on period match), the OCx pin goes high. Upon the next TMRy to OCxRS match, the OCx pin goes low. If OCxR = 0 and PRy = OCxRS, the pin will remain low for one clock cycle, then be driven high until the next TMRy to OCxRS match. The OCxIF bit will be set as a result of the second compare. There are two alternative initial conditions to consider: a] Initialize TMRy = 0 and set OCxR >= 1 b] Initialize TMRy = PRy (PRy > 0) and set OCxR = 0 Output at OCx Continuous pulses with the first pulse delayed by the value in the PRy register, depending on setup.
SFR Logical Relationship PRy >= OCxRS and OCxRS > OCxR
PRy >= OCxR and OCxR >= OCxRS
OCxR >= 1 and PRy >= 1
TMRy counts up to OCxR and on a compare match event Continuous pulses (i.e., TMRy = OCxR), the OCx pin is driven to a high state. TMRy then continues to count and eventually resets on period match (i.e., PRy =TMRy). The timer then restarts from 0x0000 and counts up to OCxRS, and on a compare match event (i.e., TMRy = OCxR), the OCx pin is driven to a low state. The OCxIF bit will be set as a result of the second compare. Only one transition will be generated at the OCx pin until the OCxRS register contents have been changed to a value less than or equal to the period register contents (PRy). OCxIF is not set until then. Rising edge/ transition to high
OCxRS > PRy and PRy >= OCxR
None
OCxR = OCxRS = PRy = 0x0000
None
Continuous output pulses are generated at the OCx pin. First pulse is delayed. The first pulse is delayed 2 instruction clock periods upon Continuous pulses the match of the timer and period register. The OCxIF bit are generated. will be set as a result of the second compare. Unsupported mode, Timer resets prior to match condition. Remains low
OCxR > PRy
None
Note 1: In all the cases considered herein, the TMRy register is assumed to be initialized to 0x0000. 2: OCxR = Compare Register, OCxRS = Secondary Compare Register, TMRy = Timery Count, PRy = Timery Period Register.
14
Output Compare
DS70061C-page 14-17

14.3.3 Pulse Width Modulation Mode
When control bits OCM<2:0> (OCxCON<2:0>) are set to 110 or 111, the selected output compare channel is configured for the PWM (Pulse Width Modulation) mode of operation. The following two PWM modes are available: PWM without Fault Protection Input PWM with Fault Protection Input The OCFA or OCFB Fault input pin is utilized for the second PWM mode. In this mode, an asynchronous logic level 0 on the OCFx pin will cause the selected PWM channel to be shutdown. (Described in Section 14.3.3.1, PWM with Fault Protection Input Pin.) In PWM mode, the OCxR register is a read only slave duty cycle register and OCxRS is a buffer register that is written by the user to update the PWM duty cycle. On every timer to period register match event (end of PWM period), the duty cycle register, OCxR, is loaded with the contents of OCxRS. The TyIF interrupt flag is asserted at each PWM period boundary. The following steps should be taken when configuring the output compare module for PWM operation: 1. 2. 3. 4. 5. 6. Set the PWM period by writing to the selected timer period register (PRy). Set the PWM duty cycle by writing to the OCxRS register. Write the OxCR register with the initial duty cycle. Enable interrupts, if required, for the timer and output compare modules. The output compare interrupt is required for PWM Fault pin utilization. Configure the output compare module for one of two PWM Operation modes by writing to the Output Compare mode bits OCM<2:0> (OCxCON<2:0>). Set the TMRy prescale value and enable the time base by setting TON (TxCON<15>) = 1. Note: The OCxR register should be initialized before the Output Compare module is first enabled. The OCxR register becomes a read only duty cycle register when the module is operated in the PWM modes. The value held in OCxR will become the PWM duty cycle for the first PWM period. The contents of the duty cycle buffer register, OCxRS, will not be transferred into OCxR until a time base period match occurs.
An example PWM output waveform is shown in Figure 14-10. Figure 14-10: PWM Output Waveform
Period = (PRy + 1)
Duty Cycle = (OCxRS )
1 1 2 3
Timery is cleared and new duty cycle value is loaded from OCxRS into OCxR. Timer value equals value in the OCxR register, OCx Pin is driven low. Timer overflow, value from OCxRS is loaded into OCxR, OCx pin driven high. TyIF interrupt flag is asserted.
DS70061C-page 14-18

14.3.3.1 PWM with Fault Protection Input Pin When the Output Compare mode bits, OCM<2:0> (OCxCON<2:0>), are set to 111, the selected output compare channel is configured for the PWM mode of operation. All functions described in Section 14.3.3, Pulse Width Modulation Mode apply, with the addition of input Fault protection. Fault protection is provided via the OCFA and OCFB pins. The OCFA pin is associated with the output compare channels 1 through 4, while the OCFB pin is associated with the output compare channels 5 through 8. If a logic 0 is detected on the OCFA/OCFB pin, the selected PWM output pin(s) are placed in the high impedance state. The user may elect to provide a pull-down or pull-up resistor on the PWM pin to provide for a desired state if a Fault condition occurs. The shutdown of the PWM output is immediate and is not tied to the device clock source. This state will remain until: The external Fault condition has been removed and The PWM mode is re-enabled by writing to the appropriate mode bits, OCM<2:0> (OCxCON<2:0>). As a result of the Fault condition, the respective interrupt flag, OCxIF bit, is asserted and an interrupt will be generated, if enabled. Upon detection of the Fault condition, the OCFLT bit (OCx-CON<4>) is asserted high (logic 1). This bit is a read only bit and will only be cleared once the external Fault condition has been removed and the PWM mode is re-enabled, by writing to the appropriate mode bits, OCM<2:0> (OCxCON<2:0>). Note: The external Fault pins, if enabled for use, will continue to control the OCx output pins, while the device is in Sleep or Idle mode.
14.3.3.2
PWM Period The PWM period is specified by writing to PRy, the Timery period register. The PWM period can be calculated using the following formula: Equation 14-1: Calculating the PWM Period PWM Period = [(PRy) + 1] TCY (TMRy Prescale Value) PWM Frequency = 1/[PWM Period]
Note:
A PRy value of N will produce a PWM period of N + 1 time base count cycles. For example: a value of 7 written into the PRy register will yield a period consisting of 8 time base cycles.
14
Output Compare
DS70061C-page 14-19

14.3.3.3 PWM Duty Cycle The PWM duty cycle is specified by writing to the OCxRS register. The OCxRS register can be written to at any time, but the duty cycle value is not latched into OCxR until a match between PRy and TMRy occurs (i.e., the period is complete). This provides a double buffer for the PWM duty cycle and is essential for glitchless PWM operation. In the PWM mode, OCxR is a read only register. Some important boundary parameters of the PWM duty cycle include: If the duty cycle register, OCxR, is loaded with 0x0000, the OCx pin will remain low (0% duty cycle). If OCxR is greater than PRy (timer period register), the pin will remain high (100% duty cycle). If OCxR is equal to PRy, the OCx pin will be low for one time base count value and high for all other count values. See Figure 14-11 for PWM mode timing details. Table 14-3 and Table 14-4 show example PWM frequencies and resolutions for a device operating at 10 and 30 MIPs, respectively. Equation 14-2: Calculation for Maximum PWM Resolution log10 Maximum PWM Resolution (bits) =
log10(2)
(FF ) bits
OSC PWM
Example 14-5:
PWM Period and Duty Cycle Calculation
Desired PWM frequency is 52.08 kHz, FOSC = 10 MHz with x4 PLL (40 MHz device clock rate) (TCY = 4/FOSC)) Timer 2 prescale setting: 1:1 1/52.08 kHz 19.20 s PR2 = = = (PR2+1) TCY (Timer 2 prescale value) (PR2+1) 0.1 s (1) 191
Find the maximum resolution of the duty cycle that can be used with a 48 kHz frequency and a 40 MHz device clock rate. 1/52.08 kHz = 19.20 s = 768 = log10(768) = PWM Resolution= 2PWM RESOLUTION 1/40 MHz 1 2PWM RESOLUTION 25 ns 1 2PWM RESOLUTION (PWM Resolution) log10(2) 9.5 bits
DS70061C-page 14-20

Figure 14-11: PWM Output Timing
1 Instruction Clock Period TMR3 0005 0000 0001 0002 0003 0004 0005 0000 0001 0002 0003 0004 0005
PR3 OCxR OCxRS
0005 0002 0002 0001 0001 New Duty Cycle Loaded Here
New Value Written to OCxRS OCx pin
TyIF is Set OCxR = OCxRS
TyIF is Set OCxR = OCxRS
Table 14-3:
Example PWM Frequencies and Resolutions at 10 MIPs (FOSC = 40 MHz) 19 Hz 8 0xFFFF 16 153 Hz 1 0xFFFF 16 305 Hz 1 0x7FFF 15 2.44 kHz 1 0x0FFF 12 9.77 kHz 1 0x03FF 10 78.1 kHz 1 0x007F 7 313 kHz 1 0x001F 5
PWM Frequency Timer Prescaler Ratio Period Register Value Resolution (bits) Table 14-4:
Example PWM Frequencies and Resolutions at 30 MIPs (FOSC = 120 MHz) 57 Hz 8 0xFFFF 16 458 Hz 1 0xFFFF 16 916 Hz 1 0x7FFF 15 7.32 kHz 1 0x0FFF 12 29.3 kHz 1 0x03FF 10 234 kHz 1 0x007F 7 938 kHz 1 0x001F 5
PWM Frequency Timer Prescaler Ratio Period Register Value Resolution (bits)
14
Output Compare
DS70061C-page 14-21

Example 14-6 shows configuration and interrupt service code for the PWM mode of operation. Example 14-6:
; ; ; ; ; The for PWM the are CLR MOV MOV MOV MOV MOV MOV MOV BSET BCLR BCLR BCLR BSET BSET ;
PWM Mode Pulse Setup and Interrupt Servicing

will set the Output Compare 1 module enabled, a 50% duty cycle and a at Fosc = 40 MHz. Timer2 is selected as base and Timer2 interrupts
following code example PWM mode w/o FAULT pin frequency of 52.08 kHz clock for the PWM time enabled. OC1CON #0x0060, w0 w0, OC1RS w0, OC1R #0x0006, w0 w0, OC1CON #0x00BF w0 w0, PR2 IPC0, IPC0, IPC0, IFS0, IEC0, T2CON, #T2IP0 #T2IP1 #T2IP2 #T21IF #T21IE #TON
; Turn off Output Compare 1 Module. ; Initialize Duty Cycle to 0x0060 ; Write duty cycle buffer register ; Write OC1R to initial duty cycle value ; Load the working register with the new ; compare mode and write to OC1CON ; Initialize PR2 with 0x00BF ; ; ; ; ; ; ; Setup Timer 2 interrupt for desired priority level (this example assigns level 1 priority) Clear Timer 2 interrupt flag Enable Timer 2 interrupts Start Timer2 with assumed settings
Example code for Timer 2 ISR:
#T21IF
DS70061C-page 14-22

14.4 14.4.1 Output Compare Operation in Power Saving States Output Compare Operation in Sleep Mode
When the device enters Sleep mode, the system clock is disabled. During Sleep, the output compare channel will drive the pin to the same active state as driven prior to entering Sleep. The module will then halt at this state. For example, if the pin was high and the CPU entered the Sleep state, the pin will stay high. Likewise, if the pin was low and the CPU entered the Sleep state, the pin will stay low. In both cases when the part wakes up, the output compare module will resume operation.
14.4.2
Output Compare Operation in Idle Mode

When the device enters Idle mode, the system clock sources remain functional and the CPU stops executing code. The OCSIDL bit (OCxCON<13>) selects if the capture module will stop in Idle mode or continue operation in Idle mode. If OCSIDL = 1, the module will discontinue operation in Idle mode. The module will perform the same procedures when stopped in Idle mode (OCxSIDL = 1) as it does for Sleep mode. If OCSIDL = 0, the module will continue operation in Idle only if the selected time base is set to operate in Idle mode. The output compare channel(s) will operate during the CPU Idle mode if the OCSIDL bit is a logic 0. Furthermore, the time base must be enabled with the respective TxSIDL bit set to a logic 0. Note: The external Fault pins, if enabled for use, will continue to control the associated OCx output pins while the device is in Sleep or Idle mode.
14.5
I/O Pin Control

When the output compare module is enabled, the I/O pin direction is controlled by the compare module. The compare module returns the I/O pin control back to the appropriate pin LAT and TRIS control bits when it is disabled. When the PWM with Fault Protection Input mode is enabled, the OCFx Fault pin must be configured for an input by setting the respective TRIS SFR bit. Enabling this special PWM mode does not configure the OCFx Fault pin as an input. Table 14-5: Pin Name OC1 OC2 OC3 OC4 OC5 OC6 OC7 OC8 OCFA OCFB Pins Associated with Output Compare Modules 1- 8 Pin Type O O O O O O O O I I Buffer Type ST ST Description Output Compare/PWM Channel 1 Output Compare/PWM Channel 2
14
Output Compare
Output Compare/PWM Channel 3 Output Compare/PWM Channel 4 Output Compare/PWM Channel 5 Output Compare/PWM Channel 6 Output Compare/PWM Channel 7 Output Compare/PWM Channel 8 PWM Fault Protection A Input (For Channels 1-4) PWM Fault Protection B Input (For Channels 5 -8)
Legend: ST = Schmitt Trigger input with CMOS levels, I = Input, O = Output
DS70061C-page 14-23
Table 14-6:
Bit 13 Timer2 Register Timer3 Register Period Register 2 Period Register 3 TSIDL TSIDL Output Compare 1 Secondary Register Output Compare 1 Register OCSIDL Output Compare 2 Secondary Register Output Compare 2 Register OCSIDL Output Compare 3 Secondary Register Output Compare 3 Register OCSIDL Output Compare 4 Secondary Register Output Compare 4 Register OCSIDL Output Compare 5 Secondary Register Output Compare 5 Register OCSIDL Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Reset State 0000 0000 0000 0000 0000 0000 0000 0000 1111 1111 1111 1111 1111 1111 1111 1111 TCKPS1 TCKPS0 TCKPS1 TCKPS0 T32
Example Register Map Associated with Output Compare Module
SFR Name
Addr.
Bit 15
Bit 14
TMR2
0106
TMR3
010A
DS70061C-page 14-24
PR2
010C
PR3
010E
T2CON
0110
TON
TGATE TGATE TCS TCS
0000 0000 0000 0000 0000 0000 0000 0000 uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu
T3CON
0112
TON
OC1RS
0180
OC1R
0182
OC1CON OCFLT OCTSEL
0184
OCM<2:0>
0000 0000 0000 0000 uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu
OC2RS
0186
OC2R
0188
OC2CON OCFLT OCTSEL
018A
OCM<2:0>
OC3RS
018C
OC3R
018E
OC3CON
0190
OCFLT
OCTSEL
OCM<2:0>
OC4RS
0192
OC4R
0194

0000 0000 0000 0000 OCFLT OCTSEL OCM<2:0> uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu
OC4CON
0196
OC5RS
0198
OC5R
019A
OC5CON
019C
Output Compare 6 Register OCSIDL
OCFLT
OCTSEL
OCM<2:0>
OC6RS
019E
Output Compare 6 Secondary Register
OC6R
01A0
OC6CON
01A2
OCFLT
OCTSEL
OCM<2:0>
OC7RS
01A4
Output Compare 7 Secondary Register Output Compare 7 Register OCSIDL
OC7R
01A6
OC7CON
01A8
OCFLT
OCTSEL
OCM<2:0>
0000 0000 0000 0000 uuuu uuuu uuuu uuuu
OC8RS
01AA
Output Compare 8 Secondary Register Output Compare 8 Register OCSIDL SI2CIF NVMIF ADIF
OC8R
01AC
uuuu uuuu uuuu uuuu
OC8CON
01AE
SPI1IF
T3IF
T2IF
OC2IF
OCFLT IC2IF
OCTSEL T1IF OC1IF
OCM<2:0> IC1IF INT0
0000 0000 0000 0000 0000 0000 0000 0000
IFS0
0084
CNIF
MI2CIF
U1TXIF U1RXIF
Legend: u = uninitialized
Note:
The register map will depend on the number of output compare modules on the device. Please refer to the device data sheet for details.
Table 14-6:
Bit 13 IC4IF IC3IF FLTBIF FLTAIF NVMIE IC3IE FLTBIE FLTAIE LVDIE OC1IP<2:0> T2IP<2:0> IC8IP<2:0> T5IP<2:0> OC7IP<2:0> DCIIE QEIIE PWMIE C2IE IC1IP<2:0> OC2IP<2:0> IC7IP<2:0> T4IP<2:0> OC6IP<2:0> INT4IE INT3IE OC8IE OC7IE INT0IP<2:0> IC2IP<2:0> INT1IP<2:0> OC4IP<2:0> OC5IP<2:0> OC6IE OC5IE C1IE SPI2IE U2TXIE U2RXIE INT2IE T5IE T4IE OC4IE OC3IE IC8IE IC7IE INT1IE ADIE U1TXIE U1RXIE SPI1IE T3IE T2IE OC2IE IC2IE T1IE OC1IE IC1IE INT0IE LVDIF DCIIF QEIIF PWMIF C2IF INT4IF INT3IF OC8IF OC7IF OC6IF OC5IF C1IF SPI2IF U2TXIF U2RXIF INT2IF T5IF T4IF OC4IF OC3IF IC8IF IC7IF INT1IF Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Reset State
Example Register Map Associated with Output Compare Module (Continued)
SFR Name
Addr.
Bit 15
Bit 14
IFS1
0086
IC6IF
IC5IF
0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100 0100
IFS2
0088
IC4IE
IEC0
008C
CNIE
MI2CIE SI2CIE
IEC1
008E
IC6IE
IC5IE
IEC2 T1IP<2:0> T3IP<2:0> OC3IP<2:0> INT2IP<2:0> OC8IP<2:0>
0090
IPC0
0094
Output Compare
IPC1
0096
IPC4
009C
IPC5
009E
IPC8
00A4
Legend: u = uninitialized
Note:
The register map will depend on the number of output compare modules on the device. Please refer to the device data sheet for details.
DS70061C-page 14-25
14

14.6 Design Tips
Question 1: The Output Compare pin stops functioning even when the OCSIDL bit is not set. Why?
Answer: This is most likely to occur when the TSIDL bit (TxCON<13>) of the associated timer source is set. Therefore, it is the timer that actually goes into Idle mode when the PWRSAV instruction is executed. Question 2: Can I use the Output Compare modules with the selected time base configured for 32-bit mode?
Answer: No. The T32 bit (TxCON<3>) should be cleared when the timer is used with an output compare module.
DS70061C-page 14-26

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Output Compare module are: Title No related application notes at this time. Application Note #
Note:
14
Output Compare
DS70061C-page 14-27

Revision B
Revision C
DS70061C-page 14-28
Section 15. Motor Control PWM

HIGHLIGHTS
This section of the manual contains the following topics: 15.1 Introduction .................................................................................................................. 15-2 15.2 Control Registers ......................................................................................................... 15-4 15.3 PWM Time Base ........................................................................................................ 15-16 15.4 PWM Duty Cycle Comparison Units .......................................................................... 15-20 15.5 Complementary PWM Output Mode .......................................................................... 15-26 15.6 Dead Time Control ..................................................................................................... 15-27 15.7 Independent PWM Output Mode ............................................................................... 15-30 15.8 PWM Output Override................................................................................................ 15-31 15.9 PWM Output and Polarity Control.............................................................................. 15-34 15.10 PWM Fault Pins ......................................................................................................... 15-34 15.11 PWM Update Lockout ................................................................................................ 15-37 15.12 PWM Special Event Trigger ....................................................................................... 15-38 15.13 Operation in Device Power Saving Modes ................................................................ 15-38 15.14 Special Features for Device Emulation...................................................................... 15-39 15.15 Related Application Notes.......................................................................................... 15-42 15.16 Revision History ......................................................................................................... 15-43
15
Motor Control PWM
DS70062D-page 15-1

15.1 Introduction
The motor control PWM (MCPWM) module simplifies the task of generating multiple, synchronized pulse width modulated outputs. In particular, the following power and motion control applications are supported: Three-Phase AC Induction Motor Switched Reluctance (SR) Motor Brushless DC (BLDC) Motor Uninterruptable Power Supply (UPS) Dedicated time base supports TCY/2 PWM edge resolution Two output pins for each PWM generator Complementary or independent operation for each output pin pair Hardware dead time generators for complementary mode Output pin polarity programmed by device configuration bits Multiple output modes: - Edge aligned mode - Center aligned mode - Center aligned mode with double updates - Single event mode Manual override register for PWM output pins Duty cycle updates are configurable to be immediate or synchronized to the PWM Hardware fault input pins with programmable function Special Event Trigger for synchronizing A/D conversions Each output pin associated with the PWM can be individually enabled
The PWM module has the following features:
15.1.1
MCPWM Module Variants

There are two versions of the MCPWM module depending on the dsPIC30F device that is selected. There is an 8-output module that is typically found on devices that have 64 or more pins. A 6-output MCPWM module is also available and is typically found on smaller devices that have less than 64 pins. A given dsPIC30F device may have more than one MCPWM module. Please refer to the specific device data sheet for further details. Table 15-1: Feature Summary: 6-Output MCPWM vs. 8-Output MCPWM Feature I/O Pins PWM Generators Fault Input Pins Dead Time Generators 6-Output MCPWM Module 6 3 1 1 8-Output MCPWM Module 8 4 2 2
The 6-output MCPWM module is useful for single or 3-phase power application, while the 8 MCPWM can support 4-phase motor applications. Table 15-1 provides a feature summary for 6- and 8-output MCPWM modules. Both modules can support multiple single phase loads. The 8-output MCPWM also provides increased flexibility in an application because it supports two fault pins and two programmable dead times. These features are discussed in greater detail in subsequent sections. A simplified block diagram of the MCPWM module is shown in Figure 15-1.
DS70062D-page 15-2

Figure 15-1: MCPWM Block Diagram
PTCON PWMCON1
PWM time base control
PWM enable and mode SFRs PWMCON2 DTCON1 Dead time control SFRs DTCON2 FLTACON Fault pin control SFRs FLTBCON OVDCON PWM manual control
PWM Generator #1 PDC1
16-bit data bus
PWM1 duty cycle register
Comparator
Channel 1 Dead Time Generator and Override Logic

Channel 2 Dead Time Generator and Override Logic Channel 3 Dead Time Generator and Override Logic Channel 4 Dead Time Generator and Override Logic
PWM1H PWM1L
PTMR
PWM Generator #2
PWM2H PWM2L
Comparator PWM Generator #3 PTMR period register PWM Generator #4 PTPER PWM3H PWM3L
PWM4H PWM4L
FLTA FLTB Special Event Postscaler
Comparator
Special Event Trigger for A/D converter
SEVTCMP
15
Motor Control PWM
Note
1: 2:
Details of PWM Generator #2, #3 and #4 not shown for clarity. Logic within dashed lines not present on 6-output MCPWM module.
DS70062D-page 15-3

The following registers control the operation of the MCPWM module: PTCON: PWM Time Base Control register PTMR: PWM Time Base register PTPER: PWM Time Base Period register SEVTCMP: PWM Special Event Compare register PWMCON1: PWM Control register #1 PWMCON2: PWM Control register #2 DTCON1: Dead Time Control register #1 DTCON2: Dead Time Control register #2 FLTACON: Fault A Control register FLTBCON: Fault B Control register PDC1: PWM Duty Cycle register #1 PDC2: PWM Duty Cycle register #2 PDC3: PWM Duty Cycle register #3 PDC4: PWM Duty Cycle register #4
In addition, there are three device configuration bits associated with the MCPWM module to set up the initial Reset states and polarity of the I/O pins. These configuration bits are located in the FBORPOR device configuration register. Please refer to Section 24. Device Configuration for further details.
DS70062D-page 15-4

Register 15-1: Upper Byte: R/W-0 PTEN bit 15 PTCON: PWM Time Base Control Register U-0 R/W-0 PTSIDL U-0 U-0 U-0 U-0 U-0 bit 8
Lower Byte: R/W-0 R/W-0 R/W-0 PTOPS<3:0> bit 7 bit 15 PTEN: PWM Time Base Timer Enable bit 1 = PWM time base is ON 0 = PWM time base is OFF Unimplemented: Read as 0 PTSIDL: PWM Time Base Stop in Idle Mode bit 1 = PWM time base halts in CPU Idle mode 0 = PWM time base runs in CPU Idle mode Unimplemented: Read as 0
R/W-0
R/W-0 R/W-0 PTCKPS<1:0>
R/W-0 R/W-0 PTMOD<1:0> bit 0
bit 14 bit 13
bit 12-8 bit 7-4
PTOPS<3:0>: PWM Time Base Output Postscale Select bits 1111 = 1:16 Postscale 0001 = 1:2 Postscale 0000 = 1:1 Postscale PTCKPS<1:0>: PWM Time Base Input Clock Prescale Select bits 11 = PWM time base input clock period is 64 TCY (1:64 prescale) 10 = PWM time base input clock period is 16 TCY (1:16 prescale) 01 = PWM time base input clock period is 4 TCY (1:4 prescale) 00 = PWM time base input clock period is TCY (1:1 prescale) PTMOD<1:0>: PWM Time Base Mode Select bits 11 = PWM time base operates in a continuous up/down mode with interrupts for double PWM updates 10 = PWM time base operates in a continuous up/down counting mode 01 = PWM time base operates in single event mode 00 = PWM time base operates in a free running mode Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 3-2
bit 1-0
15
Motor Control PWM
DS70062D-page 15-5

Register 15-2: Upper Byte: R-0 PTDIR bit 15 PTMR: PWM Time Base Register R/W-0 R/W-0 R/W-0 R/W-0 PTMR <14:8> R/W-0 R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 R/W-0 bit 7 bit 15 PTDIR: PWM Time Base Count Direction Status bit (Read Only) 1 = PWM time base is counting down 0 = PWM time base is counting up PTMR <14:0>: PWM Timebase Register Count Value Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
R/W-0
R/W-0 R/W-0 PTMR <7:0>
R/W-0
R/W-0
R/W-0 bit 0
bit 14-0
PTPER: PWM Time Base Period Register R/W-0 R/W-0 R/W-0 R/W-0 PTPER <14:8> R/W-0 R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 R/W-0 bit 7
R/W-0
R/W-0 R/W-0 PTPER <7:0>
R/W-0
R/W-0
R/W-0 bit 0
bit 15 bit 14-0
Unimplemented: Read as 0 PTPER<14:0>: PWM Time Base Period Value bits Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
DS70062D-page 15-6

Register 15-4: Upper Byte: R/W-0 SEVTDIR bit 15 SEVTCMP: Special Event Compare Register R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 SEVTCMP <14:8> R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 R/W-0 bit 7 bit 15 SEVTDIR: Special Event Trigger Time Base Direction bit(1) 1 = A special event trigger will occur when the PWM time base is counting downwards. 0 = A special event trigger will occur when the PWM time base is counting upwards. SEVTCMP <14:0>: Special Event Compare Value bit(2) Note 1: SEVTDIR is compared with PTDIR (PTMR<15>) to generate the special event trigger. 2: SEVTCMP<14:0> is compared with PTMR<14:0> to generate the special event trigger. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
R/W-0
R/W-0 R/W-0 SEVTCMP <7:0>
R/W-0
R/W-0
R/W-0 bit 0
bit 14-0
PWMCON1: PWM Control Register 1 U-0 U-0 U-0 R/W-0 PMOD4 R/W-0 PMOD3 R/W-0 PMOD2 R/W-0 PMOD1 bit 8
Lower Byte: R/W-1 R/W-1 PEN4H PEN3H bit 7 bit 15-12 Unimplemented: Read as 0 bit 11-8
R/W-1 PEN2H
R/W-1 PEN1H
R/W-1 PEN4L
R/W-1 PEN3L
R/W-1 PEN2L
R/W-1 PEN1L bit 0
PMOD4:PMOD1: PWM I/O Pair Mode bits 1 = PWM I/O pin pair is in the independent output mode 0 = PWM I/O pin pair is in the complementary output mode PEN4H-PEN1H: PWMxH I/O Enable bits(1) 1 = PWMxH pin is enabled for PWM output 0 = PWMxH pin disabled. I/O pin becomes general purpose I/O PEN4L-PEN1L: PWMxL I/O Enable bits(1) 1 = PWMxL pin is enabled for PWM output 0 = PWMxL pin disabled. I/O pin becomes general purpose I/O Note 1: Reset condition of the PENxH and PENxL bits depend on the value of the PWM/PIN device configuration bit in the FBORPOR Device Configuration Register. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7-4
bit 3-0
15
Motor Control PWM
DS70062D-page 15-7

Register 15-6: Upper Byte: U-0 bit 15 PWMCON2: PWM Control Register 2 U-0 U-0 U-0 R/W-0 R/W-0 R/W-0 SEVOPS<3:0> R/W-0 bit 8 Lower Byte: U-0 bit 7
U-0
U-0
U-0
U-0
R/W-0 IUE
R/W-0 OSYNC
R/W-0 UDIS bit 0
bit 15-12 Unimplemented: Read as 0 bit 11-8 SEVOPS<3:0>: PWM Special Event Trigger Output Postscale Select bits 1111 = 1:16 Postscale 0001 = 1:2 Postscale 0000 = 1:1 Postscale Unimplemented: Read as 0 IUE: Immediate Update Enable bit(1) 1 = Updates to the active PDC registers are immediate 0 = Updates to the active PDC registers are synchronized to the PWM time base OSYNC: Output Override Synchronization bit 1 = Output overrides via the OVDCON register are synchronized to the PWM time base 0 = Output overrides via the OVDCON register occur on next TCY boundary UDIS: PWM Update Disable bit 1 = Updates from duty cycle and period buffer registers are disabled 0 = Updates from duty cycle and period buffer registers are enabled Note 1: IUE bit is not implemented on the dsPIC30F6010 device. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7-2 bit 2
bit 1
bit 0
DS70062D-page 15-8

Register 15-7: DTCON1: Dead Time Control Register 1 R/W-0 R/W-0 R/W-0 R/W-0 DTB<5:0> R/W-0 R/W-0 bit 8
Upper Byte: R/W-0 R/W-0 DTBPS<1:0> bit 15
Lower Byte: R/W-0 R/W-0 DTAPS<1:0> bit 7
R/W-0
R/W-0
R/W-0 R/W-0 DTA<5:0>
R/W-0
R/W-0 bit 0
bit 15-14 DTBPS<1:0>: Dead Time Unit B Prescale Select bits 11 = Clock period for Dead Time Unit B is 8 TCY 10 = Clock period for Dead Time Unit B is 4 TCY 01 = Clock period for Dead Time Unit B is 2 TCY 00 = Clock period for Dead Time Unit B is TCY bit 13-8 bit 7-6 DTB<5:0>: Unsigned 6-bit Dead Time Value bits for Dead Time Unit B DTAPS<1:0>: Dead Time Unit A Prescale Select bits 11 = Clock period for Dead Time Unit A is 8 TCY 10 = Clock period for Dead Time Unit A is 4 TCY 01 = Clock period for Dead Time Unit A is 2 TCY 00 = Clock period for Dead Time Unit A is TCY DTA<5:0>: Unsigned 6-bit Dead Time Value bits for Dead Time Unit A Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
bit 5-0
15
Motor Control PWM
DS70062D-page 15-9

Register 15-8: Upper Byte: U-0 bit 15 DTCON2: Dead Time Control Register 2 U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8
Lower Byte: R/W-0 DTS4A bit 7 bit 15-8 bit 7
R/W-0 DTS4I
R/W-0 DTS3A
R/W-0 DTS3I
R/W-0 DTS2A
R/W-0 DTS2I
R/W-0 DTS1A
R/W-0 DTS1I bit 0
Unimplemented: Read as 0 DTS4A: Dead Time Select bit for PWM4 Signal Going Active 1 = Dead time provided from Unit B 0 = Dead time provided from Unit A DTS4I: Dead Time Select bit for PWM4 Signal Going Inactive 1 = Dead time provided from Unit B 0 = Dead time provided from Unit A DTS3A: Dead Time Select bit for PWM3 Signal Going Active 1 = Dead time provided from Unit B 0 = Dead time provided from Unit A DTS3I: Dead Time Select bit for PWM3 Signal Going Inactive 1 = Dead time provided from Unit B 0 = Dead time provided from Unit A DTS2A: Dead Time Select bit for PWM2 Signal Going Active 1 = Dead time provided from Unit B 0 = Dead time provided from Unit A DTS2I: Dead Time Select bit for PWM2 Signal Going Inactive 1 = Dead time provided from Unit B 0 = Dead time provided from Unit A DTS1A: Dead Time Select bit for PWM1 Signal Going Active 1 = Dead time provided from Unit B 0 = Dead time provided from Unit A DTS1I: Dead Time Select bit for PWM1 Signal Going Inactive 1 = Dead time provided from Unit B 0 = Dead time provided from Unit A Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
bit 6
bit 5
bit 4
bit 3
bit 2
bit 1
bit 0
DS70062D-page 15-10

Register 15-9: FLTACON: Fault A Control Register R/W-0 FAOV3H R/W-0 FAOV3L R/W-0 FAOV2H R/W-0 FAOV2L R/W-0 FAOV1H R/W-0 FAOV1L bit 8
Upper Byte: R/W-0 R/W-0 FAOV4H FAOV4L bit 15
Lower Byte: R/W-0 FLTAM bit 7 bit 15-8
U-0
U-0
U-0
R/W-0 FAEN4
R/W-0 FAEN3
R/W-0 FAEN2
R/W-0 FAEN1 bit 0
FAOV4H-FAOV1L: Fault Input A PWM Override Value bits 1 = The PWM output pin is driven ACTIVE on an external fault input event 0 = The PWM output pin is driven INACTIVE on an external fault input event FLTAM: Fault A Mode bit 1 = The Fault A input pin functions in the cycle-by-cycle mode 0 = The Fault A input pin latches all control pins to the programmed states in FLTACON<15:8> Unimplemented: Read as 0 FAEN4: Fault Input A Enable bit 1 = PWM4H/PWM4L pin pair is controlled by Fault Input A 0 = PWM4H/PWM4L pin pair is not controlled by Fault Input A FAEN3: Fault Input A Enable bit 1 = PWM3H/PWM3L pin pair is controlled by Fault Input A 0 = PWM3H/PWM3L pin pair is not controlled by Fault Input A FAEN2: Fault Input A Enable bit 1 = PWM2H/PWM2L pin pair is controlled by Fault Input A 0 = PWM2H/PWM2L pin pair is not controlled by Fault Input A FAEN1: Fault Input A Enable bit 1 = PWM1H/PWM1L pin pair is controlled by Fault Input A 0 = PWM1H/PWM1L pin pair is not controlled by Fault Input A Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7
bit 6-4 bit 3
bit 2
bit 1
bit 0
15
Motor Control PWM
DS70062D-page 15-11

Register 15-10: FLTBCON: Fault B Control Register Upper Byte: R/W-0 R/W-0 FBOV4H FBOV4L bit 15 R/W-0 FBOV3H R/W-0 FBOV3L R/W-0 FBOV2H R/W-0 FBOV2L R/W-0 FBOV1H R/W-0 FBOV1L bit 8
Lower Byte: R/W-0 FLTBM bit 7 bit 15-8
U-0
U-0
U-0
R/W-0 FBEN4
R/W-0 FBEN3
R/W-0 FBEN2
R/W-0 FBEN1 bit 0
FBOV4H:FBOV1L: Fault Input B PWM Override Value bits 1 = The PWM output pin is driven ACTIVE on an external fault input event 0 = The PWM output pin is driven INACTIVE on an external fault input event FLTBM: Fault B Mode bit 1 = The Fault B input pin functions in the cycle-by-cycle mode 0 = The Fault B input pin latches all control pins to the programmed states in FLTBCON<15:8> Unimplemented: Read as 0 FAEN4: Fault Input B Enable bit(1) 1 = PWM4H/PWM4L pin pair is controlled by Fault Input B 0 = PWM4H/PWM4L pin pair is not controlled by Fault Input B FAEN3: Fault Input B Enable bit(1) 1 = PWM3H/PWM3L pin pair is controlled by Fault Input B 0 = PWM3H/PWM3L pin pair is not controlled by Fault Input B FAEN2: Fault Input B Enable bit(1) 1 = PWM2H/PWM2L pin pair is controlled by Fault Input B 0 = PWM2H/PWM2L pin pair is not controlled by Fault Input B FAEN1: Fault Input B Enable bit(1) 1 = PWM1H/PWM1L pin pair is controlled by Fault Input B 0 = PWM1H/PWM1L pin pair is not controlled by Fault Input B Note 1: Fault pin A has priority over Fault pin B, if enabled. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7
bit 6-4 bit 3
bit 2
bit 1
bit 0
DS70062D-page 15-12

Register 15-11: OVDCON: Override Control Register Upper Byte: R/W-1 R/W-1 POVD4H POVD4L bit 15 R/W-1 POVD3H R/W-1 POVD3L R/W-1 POVD2H R/W-1 R/W-1 POVD2L POVD1H R/W-1 POVD1L bit 8
Lower Byte: R/W-0 R/W-0 POUT4H POUT4L bit 7 bit 15-8
R/W-0 POUT3H
R/W-0 POUT3L
R/W-0 POUT2H
R/W-0 POUT2L
R/W-0 POUT1H
R/W-0 POUT1L bit 0
POVD4H-POVD1L: PWM Output Override bits 1 = Output on PWMxx I/O pin is controlled by the PWM generator 0 = Output on PWMxx I/O pin is controlled by the value in the corresponding POUTxx bit POUT4H-POUT1L: PWM Manual Output bits 1 = PWMxx I/O pin is driven ACTIVE when the corresponding POVDxx bit is cleared 0 = PWMxx I/O pin is driven INACTIVE when the corresponding POVDxx bit is cleared Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7-0
Register 15-12: PDC1: PWM Duty Cycle Register 1 Upper Byte: R/W-0 bit 15 Lower Byte: R/W-0 R/W-0 bit 7 bit 15-0 PDC1<15:0>: PWM Duty Cycle #1 Value bits Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 PWM Duty Cycle #1 bits 15-8 R/W-0 R/W-0 bit 8
R/W-0
R/W-0 R/W-0 R/W-0 PWM Duty Cycle #1 bits 7-0
R/W-0
R/W-0 bit 0
15
Motor Control PWM
DS70062D-page 15-13

R/W-0
R/W-0
R/W-0 bit 0
R/W-0
R/W-0
R/W-0 bit 0
DS70062D-page 15-14

R/W-0
R/W-0
R/W-0 bit 0
Register 15-16: FBORPOR: BOR AND POR Device Configuration Register Upper Byte: U-0 bit 23 U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 16
Middle Byte: U-0 bit 15
U-0
U-0
U-0
U-0
R/P PWMPIN
R/P HPOL
R/P LPOL bit 8
Lower Byte: R/P BOREN bit 7 bit 10
U-0
R/P R/P BORV<1:0>
U-0
U-0
R/P R/P FPWRT<1:0> bit 0
PWMPIN: MPWM Drivers Initialization bit 1 = Pin state at reset controlled by I/O Port (PWMCON1<7:0> = 0x00) 0 = Pin state at reset controlled by module (PWMCON1<7:0> = 0xFF) HPOL: MCPWM High Side Drivers (PWMxH) Polarity bit 1 = Output signal on PWMxH pins has active high polarity 0 = Output signal on PWMxH pins has active low polarity LPOL: MCPWM Low Side Drivers (PWMxL) Polarity bit 1 = Output signal on PWMxL pins has active high polarity 0 = Output signal on PWMxL pins has active low polarity Note: See Section 24. Device Configuration for information about other configuration bits on this register.
bit 9
bit 8
15
Motor Control PWM
Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented, read as 0 0 = Bit is cleared x = Bit is unknown
P = Programmable configuration bit
DS70062D-page 15-15

15.3 PWM Time Base
The PWM time base is provided by a 15-bit timer with a prescaler and postscaler (see Figure 15-2). The 15 bits of the time base are accessible via the PTMR register. PTMR<15> is a read-only status bit, PTDIR, that indicates the present count direction of the PWM time base. If the PTDIR status bit is cleared, PTMR is counting upwards. If PTDIR is set, PTMR is counting downwards. The time base is enabled/disabled by setting/clearing the PTEN bit (PTCON<15>). PTMR is not cleared when the PTEN bit is cleared in software. Figure 15-2: PWM Time Base Block Diagram
Zero match Period match PTMOD1 PTMOD0
Clock Control
PTEN
TCY
Prescaler 1:1, 1:4, 1:16, 1:64
PTMR Register
PTMR clock
Timer reset Up/Down
Zero detect
Zero match Timer Direction Control
PTDIR (PTMR<15>)
Comparator
Period match PTMOD1
Time Base period register Gated Period load Period load Immediate Update Enable (IUE)
Gated Duty Cycle Load
PTPER Update disable (UDIS) Zero match
Postscaler 1:1-1:16 Interrupt Control PTIF
Period match PTMOD1 PTMOD0
The PWM time base can be configured for four different modes of operation: 1. 2. 3. 4. Free Running mode Single Event mode Continuous Up/Down Count mode Continuous Up/Down Count mode with interrupts for double-updates.
DS70062D-page 15-16

These four modes are selected by the PTMOD<1:0> control bits (PTCON<1:0>). Note: The mode of the PWM time base determines the type of PWM signal that is generated by the module. (See Section 15.4.2, Section 15.4.3 and Section 15.4.4 for more details.)
15.3.1
Free Running Mode

In the Free Running mode, the time base will count upwards until the value in the PTPER register is matched. The PTMR register is reset on the following input clock edge and the time base will continue counting upwards as long as the PTEN bit remains set.
15.3.2
Single-Event Mode
In the Single Event Counting mode, the PWM time base will begin counting upwards when the PTEN bit is set. When the PTMR value matches the PTPER register, the PTMR register will be reset on the following input clock edge and the PTEN bit will be cleared by the hardware to halt the time base.
15.3.3
Up/Down Counting Modes

For the Continuous Up/Down Counting modes, the PWM time base will count upwards until the value in the PTPER register is matched. The timer will begin counting downwards on the following input clock edge and continue counting down until it reaches 0. The PTDIR bit PTMR<15> is read-only and indicates the counting direction. The PTDIR bit is set when the timer counts downwards.
15.3.4
PWM Time Base Prescaler

The input clock to PTMR, (TCY) has prescaler options of 1:1, 1:4, 1:16 or 1:64 selected by control bits PTCKPS<1:0> (PTCON<3:2>). The prescaler counter is cleared when any of the following occurs: A write to the PTMR register A write to the PTCON register Any device reset The PTMR register is not cleared when PTCON is written.
15.3.5
PWM Time Base Postscaler

The match output of PTMR can optionally be post-scaled through a 4-bit postscaler (which gives a 1:1 to 1:16 scaling inclusive) to generate an interrupt. The postscaler is useful when the PWM duty cycle does not need to be updated every PWM cycle. The postscaler counter is cleared when any of the following occurs: A write to the PTMR register A write to the PTCON register Any device reset The PTMR register is not cleared when PTCON is written.
15
Motor Control PWM
DS70062D-page 15-17

15.3.6 PWM Time Base Interrupts
The interrupt signals generated by the PWM time base depend on the mode selection bits, PTMOD<1:0> (PTCON<1:0>), and the time base postscaler bits, PTOPS<3:0> (PTCON<7:4>). Free Running Mode When the PWM time base is in the Free Running mode (PTMOD<1:0> = 00), an interrupt is generated when the PTMR register is reset to 0, due to a match with the PTPER register. The postscaler selection bits may be used in this mode of the timer to reduce the frequency of the interrupt events. Single Event Mode When the PWM time base is in the Single Event mode (PTMOD<1:0> = 01), an interrupt is generated when the PTMR register is reset to 0 due to a match with the PTPER register. The PTEN bit (PTCON<15>) is also cleared at this time to inhibit further PTMR increments. The postscaler selection bits have no effect in this mode of the timer. Up/Down Counting Mode In the Up/Down Counting mode (PTMOD<1:0> = 10), an interrupt event is generated each time the value of the PTMR register becomes zero and the PWM time base begins to count upwards. The postscaler selection bits may be used in this mode of the timer to reduce the frequency of the interrupt events. Up/Down Counting Mode with Double Updates In the Double Update mode (PTMOD<1:0> = 11), an interrupt event is generated each time the PTMR register is equal to zero and each time a period match occurs. The postscaler selection bits have no effect in this mode of the timer. The Double Update mode allows the control loop bandwidth to be doubled because the PWM duty cycles can be updated twice per period. Every rising and falling edge of the PWM signal can be controlled using the double update mode.
15.3.7
PWM Period
The PTPER register sets the counting period for PTMR. The user must write a 15-bit value to PTPER<14:0>. When the value in PTMR<14:0> matches the value in PTPER<14:0>, the time base will either reset to 0 or reverse the count direction on the next clock input edge. The action taken depends on the operating mode of the time base. The time base period is double buffered to allow on-the-fly period changes of the PWM signal without glitches. The PTPER register serves as a buffer register to the actual time base period register, which is not accessible by the user. The PTPER register contents are loaded into the actual time base period register at the following times: Free Running and Single Event modes: when the PTMR register is reset to zero after a match with the PTPER register. Up/Down Counting modes: When the PTMR register is zero. The value held in the PTPER register is automatically loaded into the time base period register when the PWM time base is disabled (PTEN = 0). Figure 15-3 and Figure 15-4 indicate the times when the contents of the PTPER register are loaded into the time base period register.
DS70062D-page 15-18

Figure 15-3: PWM Period Buffer Updates in Free Running Count Mode
Period value loaded from PTPER buffer register
PTMR Value New PTPER value Old PTPER value
New value written to PTPER buffer
The PWM period can be determined from the following formula: Equation 15-1: PWM Period Calculation for Free Running Count Mode (PTMOD = 10 or 11) FCY FPWM (PTMR Prescaler) -1
PTPER = Example:
FCY = 20 MHz FPWM = 20,000 Hz PTMR Prescaler = 1:1 PTPER = = 1000 -1 = 999 Figure 15-4: PWM Period Buffer Updates in Up/Down Counting Modes
Period value loaded from PTPER Buffer register
20,000,000 20,000 1
-1
New PTPER value PTMR Value Old PTPER value
New value written to PTPER buffer
15
Motor Control PWM
DS70062D-page 15-19

Equation 15-2: PWM Period Calculation in Up/Down Counting Modes (PTMOD = 00 or 01) FCY FPWM (PTMR Prescaler) 2 -1
PTPER = Example:
FCY = 20 MHz FPWM = 20,000 Hz PTMR Prescaler = 1:1 PTPER = = 500 -1 = 499 20,000,000 20,000 1 2 -1
15.4
PWM Duty Cycle Comparison Units

The MCPWM module has four PWM generators. There are four 16-bit special function registers used to specify duty cycle values for the PWM generators: PDC1 PDC2 PDC3 PDC4
In subsequent discussions, PDCx refers to any of the four PWM duty cycle registers.
15.4.1
PWM Duty Cycle Resolution

The maximum resolution (in bits) for a given device oscillator and PWM frequency can be determined from the following formula: Equation 15-3: PWM Resolution 2T PWM log ----------------- T CY Resolution = ------------------------------log ( 2 ) The PWM resolutions and frequencies are shown in Table 15-2 for a selection of execution speeds and PTPER values. The PWM frequencies in Table 15-2 are for edge-aligned (Free Running PTMR) PWM mode. For center aligned modes (Up/Down PTMR mode), the PWM frequencies will be 1/2 the values as indicated in Table 15-3. Table 15-2: TCY (FCY) 33 ns (30 MHz) 33 ns (30 MHz) 50 ns (20 MHz) 50 ns (20 MHz) 100 ns (10 MHz) 100 ns (10 MHz) 200 ns (5 MHz) 200 ns (5 MHz) Example PWM Frequencies and Resolutions, 1:1 Prescaler, Edge Aligned PWM PTPER Value 0x7FFF 0x3FF 0x7FFF 0x1FF 0x7FFF 0xFF 0x7FFF 0x7F PDCx Value for 100% 0xFFFF 0x7FF 0xFFFF 0x3FF 0xFFFF 0x1FF 0xFFFF 0xFF PWM Resolution 16 bits 11 bits 16 bits 10 bits 16 bits 9 bits 16 bits 8 bits PWM Frequency 915 Hz 29.3 kHz 610 Hz 39.1 kHz 305 Hz 39.1 kHz 153 Hz 39.1 kHz
DS70062D-page 15-20
Table 15-3: TCY (FCY)
Example PWM Frequencies and Resolutions, 1:1 Prescaler, Center Aligned PWM PTPER Value 0x7FFF 0x3FFF 0x7FFF 0x1FFF 0x7FFF 0xFFF 0x7FFF 0x7FF PDCx Value for 100% 0xFFFF 0x7FFF 0xFFFF 0x3FFF 0xFFFF 0x1FFF 0xFFFF 0xFFF PWM Resolution 16 bits 15 bits 16 bits 14 bits 16 bits 13 bits 16 bits 12 bits PWM Frequency 458 Hz 916 Hz 305 Hz 1.22 kHz 153 Hz 1.22 kHz 76.3 Hz 1.22 kHz
33 ns (30 MHz) 33 ns (30 MHz) 50 ns (20 MHz) 50 ns (20 MHz) 100 ns (10 MHz) 100 ns (10 MHz) 200 ns (5 MHz) 200 ns (5 MHz)
The MCPWM module has the ability to produce PWM signal edges with TCY/2 resolution. PTMR increments every TCY with a 1:1 prescaler. To achieve TCY/2 edge resolution, PDCx<15:1> is compared to PTMR<14:0> to determine a duty cycle match. PDCx<0> determines whether the PWM signal edge will occur at the TCY or the TCY/2 boundary. When a 1:4, 1:16 or a 1:64 prescaler is used with the PWM time base, PDCx<0> is compared to the MSbit of the prescaler counter clock to determine when the PWM edge should occur. PTMR and PDCx resolutions are depicted in Figure 15-5. It is shown that PTMR resolution is TCY and PDCx resolution is TCY/2 for 1:1 prescaler selection. Figure 15-5: PTMR and PDCx Resolution Timing Diagram. Free Running Mode and 1:1 Prescaler Selection
TCY
PTPER = 10
TCY
PTMR
PDCx = 14 PDCx = 15
TCY/2
15
Motor Control PWM
DS70062D-page 15-21

Figure 15-6:
14 PTMR 15 15-bit comparison 15 15 PDCx 1 0 1-Bit Comparison Edge Logic
Duty Cycle Comparison Logic

0 N-bit Prescaler N 2 1 TCY
PWM Edge Event
Note: PDCx<0> is compared to the FOSC/2 signal when the prescaler is 1:1.
15.4.2
Edge Aligned PWM

Edge aligned PWM signals are produced by the module when the PWM time base is operating in the Free Running mode. The output signal for a given PWM channel has a period specified by the value loaded in PTPER and a duty cycle specified by the appropriate PDCx register (see Figure 15-7). Assuming a non-zero duty cycle and no immediate updates are enabled (IUE = 0), the outputs of all enabled PWM generators will be driven active at the beginning of the PWM period (PTMR = 0). Each PWM output will be driven inactive when the value of PTMR matches the duty cycle value of the PWM generator. If the value in the PDCx register is zero, then the output on the corresponding PWM pin will be inactive for the entire PWM period. In addition, the output on the PWM pin will be active for the entire PWM period if the value in the PDCx register is greater than the value held in the PTPER register. If immediate updates are enabled (IUE = 1), the new duty cycle value will be loaded at the time the new value is written to any active PDC register. Figure 15-7: Edge-Aligned PWM
New duty cycle loaded from PDCx PTPER PDC1 PDC2 0 PWM1H Duty Cycle PTMR value
PWM2H Period
DS70062D-page 15-22

15.4.3 Single Event PWM Operation
The PWM module will produce single pulse outputs when the PWM time base is configured for the single event mode (PTMOD<1:0> = 01). This mode of operation is useful for driving certain types of electronically commutated motors. In particular, this mode is useful for high-speed SR motor operation. Only edge-aligned outputs may be produced in the Single Event mode. In Single Event mode, the PWM I/O pin(s) are driven to the active state when the PTEN bit is set. When a match with a duty cycle register occurs, the PWM I/O pin is driven to the inactive state. When a match with the PTPER register occurs, the PTMR register is cleared, all active PWM I/O pins are driven to the inactive state, the PTEN bit is cleared, and an interrupt is generated. Operation of the PWM module will stop until the PTEN is set again in software. Figure 15-8: Single Event PWM Operation
PTEN bit set by software PTEN bit cleared by hardware PTPER PDC1 PDC2
PTEN PWM2H PWM1H
PWMIF
PWMIF cleared in software
15
Motor Control PWM
DS70062D-page 15-23

15.4.4 Center Aligned PWM
Center aligned PWM signals are produced by the module when the PWM time base is configured in one of the two Up/Down Counting modes (PTMOD<1:0> = 1x). The PWM compare output is driven to the active state when the value of the Duty Cycle register matches the value of PTMR and the PWM time base is counting downwards (PTDIR = 1). The PWM compare output will be driven to the inactive state when the PWM time base is counting upwards (PTDIR = 0) and the value in the PTMR register matches the duty cycle value. If the value in a particular Duty Cycle register is zero, then the output on the corresponding PWM pin will be inactive for the entire PWM period. In addition, the output on the PWM pin will be active for the entire PWM period if the value in the Duty Cycle register is greater than the value held in the PTPER register. Figure 15-9: Center Aligned PWM
Period/2 PTPER PDC1 PDC2 PTMR Value
0 PWM1H PWM2H PDCx Value Period
15.4.5
Duty Cycle Register Buffering

The four PWM duty cycle registers, PDC1-PDC4, are buffered to allow glitchless updates of the PWM outputs. For each generator, there is the PDCx register (buffer register) that is accessible by the user and the non-memory mapped Duty Cycle register that holds the actual compare value. The PWM duty cycle is updated with the value in the PDCx register at specific times in the PWM period to avoid glitches in the PWM output signal. When the PWM time base is operating in the Free Running or Single Event modes (PTMOD<1:0> = 0x), the PWM duty cycle is updated whenever a match with the PTPER register occurs and PTMR is reset to 0. Note: Any write to the PDCx registers will immediately update the duty cycle when the PWM time base is disabled (PTEN = 0). This allows a duty cycle change to take effect before PWM signal generation is enabled.
When the PWM time base is operating in the Up/Down Counting mode (PTMOD<1:0> = 10), duty cycles are updated when the value of the PTMR register is zero and the PWM time base begins to count upwards. Figure 15-10 indicates the times when the duty cycle updates occur for this mode of the PWM time base. When the PWM time base is in the Up/Down Counting mode with double updates (PTMOD<1:0> = 11), duty cycles are updated when the value of the PTMR register is zero and when the value of the PTMR register matches the value in the PTPER register. Figure 15-11 indicates the times when the duty cycle updates occur for this mode of the PWM time base.
DS70062D-page 15-24

Figure 15-10: Duty Cycle Update Times in Up/Down Count Mode
Duty cycle value loaded from PDCx register, CPU interrupted
PWM output
PTMR Value
PTIF New value written to PDCx register
Figure 15-11:
Duty Cycle Update Times in Up/Down Count Mode with Double Updates
Duty cycle value loaded from PDCx register, CPU interrupted
PWM output
PTMR Value
New values written to PDCx register
15.4.6
PWM Duty Cycle Immediate Updates

When the Immediate Update Enable bit is set (IUE = 1), any write to the duty cycle registers will update the new duty cycle value immediately. This feature gives the option to the user to allow immediate updates of the active PWM duty cycle registers instead of waiting for the end of the current time base period. System stability is improved in closed loop servo applications by reducing the delay between system observation and the issuance of system corrective commands when immediate updates are enabled (IUE = 1). If the PWM output is active at the time the new duty cycle is written and the new duty cycle is less than the current time base value, the PWM pulse width will be shortened. If the PWM output is active at the time the new duty cycle is written and the new duty cycle is greater than the current time base value, the PWM pulse width will be lengthened. If the PWM output is inactive at the time the new duty cycle is written and the new duty cycle is greater than the current time base value, the PWM output will become active immediately and will remain active for the new written duty cycle value. Figure 15-12 indicates the times when the duty cycle updates occur when immediate updates are enabled (IUE = 1).. Note: The IUE bit is not implemented on the dsPIC30F6010 device.
15
Motor Control PWM
DS70062D-page 15-25

Figure 15-12: Duty Cycle Update Times When Immediate Updates Are Enabled (IUE = 1)
New Values Written to PDCx Register
Latest Duty Cycle Value Written to PDCx
50%
90%
10%
90%
PWM Output
PTMR Value
15.5
Complementary PWM Output Mode

The Complementary Output mode is used to drive inverter loads similar to the one shown in Figure 15-13. This inverter topology is typical for ACIM and BLDC applications. In the Complementary Output mode, a pair of PWM outputs cannot be active simultaneously. Each PWM channel and output pin pair is internally configured as shown in Figure 15-14. A dead time may be optionally inserted during device switching where both outputs are inactive for a short period (Refer to Section 15.6 Dead Time Control). Figure 15-13: Typical Load for Complementary PWM Outputs
+V 3 Phase Load
1H
2H
3H
1L
2L
3L
The Complementary mode is selected for each PWM I/O pin pair by clearing the appropriate PMODx bit in PWMCON1. The PWM I/O pins are set to complementary mode by default upon a device reset. Figure 15-14: PWM Channel Block Diagram, Complementary Mode
PWM Generator Dead Time Generator Override and Fault Logic
PWMxH
PWMxL
DS70062D-page 15-26

15.6 Dead Time Control
Dead time generation is automatically enabled when any of the PWM I/O pin pairs are operating in the Complementary Output mode. Because the power output devices cannot switch instantaneously, some amount of time must be provided between the turn-off event of one PWM output in a complementary pair and the turn-on event of the other transistor. The 6-output PWM module has one programmable dead time. The 8-output PWM module allows two different dead times to be programmed. These two dead times may be used in one of two methods described below to increase user flexibility: The PWM output signals can be optimized for different turn-off times in the high-side and low-side transistors. The first dead time is inserted between the turn-off event of the lower transistor of the complementary pair and the turn-on event of the upper transistor. The second dead time is inserted between the turn-off event of the upper transistor and the turn-on event of the lower transistor. The two dead times can be assigned to individual PWM I/O pin pairs. This operating mode allows the PWM module to drive different transistor/load combinations with each complementary PWM I/O pin pair.
15.6.1
Dead Time Generators

Each complementary output pair for the PWM module has a 6-bit down counter that is used to produce the dead time insertion. As shown in Figure 15-15, each dead time unit has a rising and falling edge detector connected to the duty cycle comparison output. One of the two possible dead times is loaded into the timer on the detected PWM edge event. Depending on whether the edge is rising or falling, one of the transitions on the complementary outputs is delayed until the timer counts down to zero. A timing diagram indicating the dead time insertion for one pair of PWM outputs is shown in Figure 15-16. The use of two different dead times for the rising and falling edge events has been exaggerated in the figure for clarity.
Figure 15-15: Dead Time Unit Block Diagram for One Output Pin Pair
Zero Compare
TCY
Prescaler
Clock Control
6-Bit Down Counter High-side PWM signal to output pin Low-Side PWM signal to output pin Dead Time Select Logic
Dead Time A Dead Time B
PWM Generator Input
Note: Logic in dashed lines not present on 6-output PWM module.
15
Motor Control PWM
DS70062D-page 15-27

Figure 15-16: Dead Time Insertion Diagram
PWM Generator
PWMxH Dead time = 0
PWMxL
PWMxH Non-zero dead time
PWMxL
Time selected by DTSxA bit (A or B)
Time selected by DTSxI bit (A or B)
DS70062D-page 15-28

15.6.2 Dead Time Assignment
Note: The dead time assignment logic is only applicable to dsPIC variants that contain the 8-output PWM module. The 6-output PWM module uses dead time A only.
The DTCON2 register contains control bits that allow the two programmable dead times to be assigned to each of the complementary outputs. There are two dead time assignment control bits for each of the complementary outputs. For example, the DTS1A and DTS1I control bits select the dead times to be used for the PWM1H/PWM1L complementary output pair. The pair of dead time selection control bits are referred to as the dead-time-select-active and dead-time-select-inactive control bits, respectively. The function of each bit in a pair is as follows: The DTSxA control bit selects the dead time that is to be inserted before the high-side output is driven active. The DTSxI control bit selects the dead time that is to be inserted before the low-side PWM active is driven active. Table 15-4 summarizes the function of each dead time selection control bit. Table 15-4: Bit DTS1A DTS1I DTS2A DTS2I DTS3A DTS3I DTS4A DTS4I Dead Time Selection Bits Function Selects PWM1H/PWM1L dead time inserted before PWM1H is driven active. Selects PWM1H/PWM1L dead time inserted before PWM1L is driven active. Selects PWM1H/PWM1L dead time inserted before PWM2H is driven active. Selects PWM1H/PWM1L dead time inserted before PWM2L is driven active. Selects PWM1H/PWM1L dead time inserted before PWM3H is driven active. Selects PWM1H/PWM1L dead time inserted before PWM3L is driven active. Selects PWM1H/PWM1L dead time inserted before PWM4H is driven active. Selects PWM1H/PWM1L dead time inserted before PWM4L is driven active.
15.6.3
Dead Time Ranges

Dead time A and dead time B are set by selecting an input clock prescaler value and a 6-bit unsigned dead time count value. Four input clock prescaler selections have been provided to allow a suitable range of dead times based on the device operating frequency. The clock prescaler option may be selected independently for each of the two dead time values. The dead time clock prescaler values are selected using the DTAPS<1:0> and DTBPS<1:0> control bits in the DTCON1 SFR. The following clock prescaler options may be selected for each of the dead time values: TCY 2 TCY 4 TCY 8 TCY Dead Time Calculation Dead Time Prescale Value TCY
Equation 15-4:
DT =
15
Motor Control PWM
Note: DT (Dead Time) is the DTA<5:0> or DTB<5:0> register value.
DS70062D-page 15-29

Table 15-5 shows example dead time ranges as a function of the input clock prescaler selected and the device operating frequency. Table 15-5: TCY (FCY) 33 ns (30 MHz) 50 ns (20 MHz) 100 ns (10 MHz) 100 ns (10 MHz) Example Dead Time Ranges Prescaler Selection 4 TCY 4 TCY 2 TCY 1 TCY Resolution 130 ns 200 ns 200 ns 100 ns Dead Time Range 130 ns -9 s 200 ns -12 s 200 ns -12 s 100 ns - 6 s
15.6.4
Dead Time Distortion

For small PWM duty cycles, the ratio of dead time to the active PWM time may become large. At the extreme case, when the duty cycle is less than or equal to the programmed duty cycle, no PWM pulse will be generated. In these cases, the inserted dead time will introduce distortion into waveforms produced by the PWM module. The user can ensure that dead time distortion is minimized by keeping the PWM duty cycle at least three times larger than the dead time. Dead time distortion can also be corrected by other techniques, such as closed loop current control. A similar effect occurs for duty cycles near 100%. The maximum duty cycle used in the application should be chosen such that the minimum inactive time of the PWM signal is at least three times larger than the dead time.
15.7
Independent PWM Output Mode

An Independent PWM Output mode is useful for driving loads such as the one shown in Figure 15-17. A particular PWM output pair is in the Independent Output mode when the corresponding PMOD bit in the PWMCON1 register is set. The dead time generators are disabled in the Independent mode and there are no restrictions on the state of the pins for a given output pin pair. Figure 15-17: Asymmetric Inverter
+V
1H
1L
Figure 15-18: PWM Block Diagram for One Output Pin Pair, Independent Mode
PWM Generator
Override and Fault Logic
PWMxH
PWMxL
DS70062D-page 15-30

15.8 PWM Output Override
The PWM output override bits allow the user to manually drive the PWM I/O pins to specified logic states independent of the duty cycle comparison units. The PWM override bits are useful when controlling various types of electrically commutated motors. All control bits associated with the PWM output override function are contained in the OVDCON register. The upper half of the OVDCON register contains 8 bits, POVDxx, that determine which PWM I/O pins will be overridden. The lower half of the OVDCON register contains 8 bits, POUTxx, that determine the state of the PWM I/O pin when it is overridden via the POVDxx bit. The POVD bits are active-low control bits. When the POVD bits are set, the corresponding POUTxx bit will have no effect on the PWM output. When one of the POVD bits is cleared, the output on the corresponding PWM I/O pin will be determined by the state of the POUT bit. When a POUT bit is set, the PWM pin will be driven to its active state. When the POUT bit is cleared, the PWM pin will be driven to its inactive state.
15.8.1
Override Control for Complementary Output Mode

The PWM module will not allow certain overrides when a pair of PWM I/O pins are operating in the Complementary mode. (PMODx = 0) The module will not allow both pins in the output pair to become active simultaneously. The high-side pin in each output pair will always take priority. Note: Dead time insertion is still performed when PWM channels are overridden manually.
15.8.2
Override Synchronization
If the OSYNC bit is set (PWMCON2<1>), all output overrides performed via the OVDCON register will be synchronized to the PWM time base. Synchronous output overrides will occur at the following times: Edge aligned mode, when PTMR is zero. Center aligned modes, when PTMR is zero, or When the value of PTMR matches PTPER. The override synchronization function, when enabled, can be used to avoid unwanted narrow pulses on the PWM output pins.
15.8.3
Output Override Examples

Figure 15-19 shows an example of a waveform that might be generated using the PWM output override feature. The Figure shows a six-step commutation sequence for a BLDC motor. The motor is driven through a 3-phase inverter as shown in Figure 15-13. When the appropriate rotor position is detected, the PWM outputs are switched to the next commutation state in the sequence. In this example, the PWM outputs are driven to specific logic states. The OVDCON register values used to generate the signals in Figure 15-19 are given in Table 15-6. The PWM duty cycle registers may be used in conjunction with the OVDCON register. The duty cycle registers controls the current delivered to the load and the OVDCON register controls the commutation. Such an example is shown in Figure 15-20. The OVDCON register values used to generate the signals in Figure 15-20 are given in Table 15-7.
15
Motor Control PWM
DS70062D-page 15-31

Table 15-6: State 1 2 3 4 5 6 PWM Output Override Example #1 OVDCON<15:8> 00000000b 00000000b 00000000b 00000000b 00000000b 00000000b OVDCON<7:0> 00100100b 00100001b 00001001b 00011000b 00010010b 00000110b
Figure 15-19: PWM Output Override Example #1

STATE 1 PWM3H PWM3L PWM2H PWM2L PWM1H PWM1L Note: Switching times between states 1-6 are controlled by user software. The state switch is controlled by writing a new value to OVDCON. 2 3 4 5 6
Table 15-7: State 1 2 3 4
PWM Output Override Example #2 OVDCON<15:8> 11000011b 11110000b 00111100b 00001111b OVDCON<7:0> 00000000b 00000000b 00000000b 00000000b
DS70062D-page 15-32

Figure 15-20: PWM Output Override Example #2
STATE
PWM4H
PWM4L
PWM3H
PWM3L
PWM2H
PWM2L
PWM1H
PWM1L Note: Switching times between states 1-4 are controlled by user software. The state switch is controlled by writing a new value to OVDCON. The PWM outputs are operated in the independent mode for this example.
15
Motor Control PWM
DS70062D-page 15-33

15.9 PWM Output and Polarity Control
The PENxx control bits in PWMCON1 enable each PWM output pin for use by the module. When a pin is enabled for PWM output, the PORT and TRIS registers controlling the pin are disabled. In addition to the PENxx control bits, there are three device configuration bits in the FBORPOR device configuration register that provide PWM output pin control. HPOL configuration bit LPOL configuration bit PWMPIN configuration bit These three configuration bits work in conjunction with the PWM enable bits (PENxx) located in PWMCON1. The configuration bits ensure that the PWM pins are in the correct states after a device reset occurs.
15.9.1
Output Polarity Control

The polarity of the PWM I/O pins is set during device programming via the HPOL and LPOL configuration bits in the FBORPOR Device Configuration register. The HPOL configuration bit sets the output polarity for the high-side PWM outputs PWM1H-PWM4H. The LPOL configuration bit sets the output polarity for the low-side PWM outputs PWM1L-PWM 4L. If the polarity configuration bit is programmed to a 1, the corresponding PWM I/O pins will have active-high output polarity. If the polarity configuration bit is programmed to a 0, then the corresponding PWM pins will have active-low polarity.
15.9.2
PWM Output Pin Reset States

The PWMPIN configuration bit determines the behavior of the PWM output pins on a device reset and can be used to eliminate external pull-up/pull-down resistors connected to the devices controlled by the PWM module. If the PWMPIN configuration bit is programmed to a 1, the PENxx control bits will be cleared on a device reset. Consequently, all PWM outputs will be tri-stated and controlled by the corresponding PORT and TRIS registers. If the PWMPIN configuration bit is programmed to a 0, the PENxx control bits will be set on a device reset. All PWM pins will be enabled for PWM output at the device reset and will be at their inactive states as defined by the HPOL and LPOL configuration bits.
15.10
PWM Fault Pins

There are two Fault pins, FLTA and FLTB, associated with the PWM module. When asserted, these pins can optionally drive each of the PWM I/O pins to a defined state. This action takes place without software intervention so fault events can be managed quickly. The Fault pins may have other multiplexed functions depending on the dsPIC device variant. When used as a fault input, each Fault pin is readable via its corresponding PORT register. The FLTA and FLTB pins function as active low inputs so that it is easy to wire-OR many sources to the same input through an external pull-up resistor. When not used with the PWM module, these pins may be used as general purpose I/O or another multiplexed function. Each Fault pin has its own interrupt vector, Interrupt Flag bit, Interrupt Enable bit and Interrupt Priority bits associated with it. The function of the FLTA pin is controlled by the FLTACON register and the function of the FLTB pin is controlled by the FLTBCON register.
15.10.1
Fault Pin Enable Bits

The FLTACON and FLTBCON registers each have 4 control bits, FxEN1-FxEN4, that determine whether a particular pair of PWM I/O pins is to be controlled by the fault input pin. To enable a specific PWM I/O pin pair for fault overrides, the corresponding bit should be set in the FLTACON or FLTBCON register. If all enable bits are cleared in the FLTACON or FLTBCON registers, then that fault input pin has no effect on the PWM module and no fault interrupts will be produced.
DS70062D-page 15-34

15.10.2 Fault States
The FLTACON and FLTBCON special function registers each have 8 bits that determine the state of each PWM I/O pin when the fault input pin becomes active. When these bits are cleared, the PWM I/O pin will be driven to the inactive state. If the bit is set, the PWM I/O pin will be driven to the active state. The active and inactive states are referenced to the polarity defined for each PWM I/O pin (set by HPOL and LPOL device configuration bits). A special case exists when a PWM module I/O pair is in the Complementary mode and both pins are programmed to be active on a fault condition. The high-side pin will always have priority in the Complementary mode so that both I/O pins cannot be driven active simultaneously.
15.10.3
Fault Input Modes

Each of the fault input pins has two modes of operation: Latched Mode: When the fault pin is driven low, the PWM outputs will go to the states defined in the FLTxCON register. The PWM outputs will remain in this state until the fault pin is driven high AND the corresponding interrupt flag (FLTxIF) has been cleared in software. When both of these actions have occurred, the PWM outputs will return to normal operation at the beginning of the next PWM period or half-period boundary regardless of the Immediate Update Enable bit value (IUE). If the interrupt flag is cleared before the fault condition ends, the PWM module will wait until the fault pin is no longer asserted to restore the outputs. Cycle-by-Cycle Mode: When the fault input pin is driven low, the PWM outputs will remain in the defined fault states for as long as the fault pin is held low. After the fault pin is driven high, the PWM outputs will return to normal operation at the beginning of the following PWM period (or half-period boundary in center aligned modes) even when immediate updates are enabled. The operating mode for each fault input pin is selected using the FLTAM and FLTBM control bits (FLTACON<7> and FLTBCON<7>).
15.10.3.1 Entry Into a Fault Condition When a fault pin is enabled and driven low, the PWM pins are immediately driven to their programmed fault states regardless of the values in the PDCx and OVDCON registers. The fault action has priority over all other PWM control registers. 15.10.3.2 Exit From a Fault Condition A fault condition must be cleared by the external circuitry driving the fault input pin high and clearing the fault interrupt flag (Latched mode only). After the fault pin condition has been cleared, the PWM module will restore the PWM output signals on the next PWM period or half-period boundary. For edge aligned PWM generation, the PWM outputs will be restored when PTMR = 0. For center aligned PWM generation, the PWM outputs will be restored when PTMR = 0 or PTMR = PTPER, whichever event occurs first. An exception to these rules will occur when the PWM time base is disabled (PTEN = 0). If the PWM time base is disabled, the PWM module will restore the PWM output signals immediately after the fault condition has been cleared.
15
Motor Control PWM
DS70062D-page 15-35

15.10.4 Fault Pin Priority
If both fault input pins have been assigned to control a particular pair of PWM pins, the fault states programmed for the FLTA input pin will take priority over the FLTB input pin. One of two actions will take place when the Fault A condition has been cleared. If the FLTB input is still asserted, the PWM outputs will return to the states programmed in the FLTBCON register on the next period or half-period boundary. If the FLTB input is not asserted, the PWM outputs will return to normal operation on the next period or half-period boundary. Note: When the FLTA pin is programmed for Latched mode, the PWM outputs will not return to the Fault B states or normal operation until the Fault A interrupt flag has been cleared and the FLTA pin is de-asserted.
15.10.5
Fault Pin Software Control

Each of the fault pins can be controlled manually in software. Since each fault input is shared with a PORT I/O pin, the PORT pin can be configured as an output by clearing the corresponding TRIS bit. When the PORT bit for the pin is cleared, the fault input will be activated. Note: The user should exercise caution when controlling the fault inputs in software. If the TRIS bit for the fault pin is cleared, then the fault input cannot be driven externally.
15.10.6
Fault Timing Examples
Figure 15-21: Example Fault Timing, Cycle-by-Cycle Mode

PWM Period
PTMR
Case 1: PWM FLTA Case 2: PWM FLTA
duty cycle = 50% Fault state
Case 3: PWM FLTA
Note: Arrows indicate the time when normal PWM operation is restored.
DS70062D-page 15-36

Figure 15-22: Example Fault Timing, Latched Mode
PTMR Return to normal operation Fault state
Duty cycle = 50% PWMxx FLTA FLTAIF Fault condition ends Interrupt flag cleared in software
Figure 15-23: Example Fault Timing, Cycle-by-Cycle Mode, Priority Operation
PTMR Return to normal operation Fault state B Fault state A Fault state B
Duty cycle = 50% PWM FLTA FLTB
Return to fault state B
15.11
PWM Update Lockout

In some applications, it is important that all duty cycle and period registers be written before the new values take effect. The update disable feature allows the user to specify when new duty cycle and period values can be used by the module. The PWM update lockout feature is enabled by setting the UDIS control bit (PWMCON2<0>). The UDIS bit affects all duty cycle registers, PDC1-PDC4, and the PWM time base period buffer, PTPER. To perform an update lockout, the user should perform the following steps: Set the UDIS bit. Write all duty cycle registers and PTPER, if applicable. Clear the UDIS bit to re-enable updates. Note: Immediate updates must be disabled (IUE = 0) in order to use the PWM update lockout feature.
15
Motor Control PWM
DS70062D-page 15-37

15.12 PWM Special Event Trigger
The PWM module has a special event trigger that allows A/D conversions to be synchronized to the PWM time base. The A/D sampling and conversion time may be programmed to occur at any point within the PWM period. The special event trigger allows the user to minimize the delay between the time when A/D conversion results are acquired and the time when the duty cycle value is updated. The PWM special event trigger has one SFR, SEVTCMP, and four postscaler control bits (SEVOPS<3:0>) to control its operation. The PTMR value for which a special event trigger should occur is loaded into the SEVTCMP register. When the PWM time base is in an Up/Down Counting mode, an additional control bit is required to specify the counting phase for the special event trigger. The count phase is selected using the SEVTDIR control bit in the MSb of SEVTCMP. If the SEVTDIR bit is cleared, the special event trigger will occur on the upward counting cycle of the PWM time base. If the SEVTDIR bit is set, the special event trigger will occur on the downward count cycle of the PWM time base. The SEVTDIR control bit has no effect unless the PWM time base is configured for an Up/Down Counting mode.
15.12.1
Special Event Trigger Enable

The PWM module will always produce the special event trigger signal. This signal may optionally be used by the A/D module. Refer to Section Section 17. 10-bit A/D Converter for more information on using the special event trigger.
15.12.2
Special Event Trigger Postscaler

The PWM special event trigger has a postscaler that allows a 1:1 to 1:16 postcale ratio. The postscaler is useful when synchronized A/D conversions do not need to be performed during every PWM cycle. The postscaler is configured by writing the SEVOPS<3:0> control bits in the PWMCON2 SFR. The special event output postscaler is cleared on the following events: Any write to the SEVTCMP register. Any device reset.
15.13 15.13.1
Operation in Device Power Saving Modes PWM Operation in Sleep mode

When the device enters Sleep mode, the system clock is disabled. Since the clock for the PWM time base is derived from the system clock source (TCY), it will also be disabled. All enabled PWM output pins will be frozen in the output states that were in effect prior to entering Sleep. If the PWM module is used to control a load in a power application, it is the users responsibility to put the PWM module outputs into a safe state prior to executing the PWRSAV instruction. Depending on the application, the load may begin to consume excessive current when the PWM outputs are frozen in a particular output state. For example, the OVDCON register can be used to manually turn off the PWM output pins as shown in the code example below. ; This code example drives all PWM pins to the inactive state ; before executing the PWRSAV instruction.
CLR PWRSAV SETM.B
OVDCON #0 OVDCONH
; Force all PWM outputs inactive ; Put the device in SLEEP mode ; Set POVD bits when device wakes.
The Fault A and Fault B input pins, if enabled to control the PWM pins via the FLTxCON registers, will continue to function normally when the device is in Sleep mode. If one of the fault pins is driven low while the device is in Sleep, the PWM outputs will be driven to the programmed fault states in the FLTxCON register.
DS70062D-page 15-38

The fault input pins also have the ability to wake the CPU from Sleep mode. If the fault interrupt enable bit is set (FLTxIE = 1), then the device will wake from Sleep when the fault pin is driven low. If the fault pin interrupt priority is greater than the current CPU priority, then program execution will start at the fault pin interrupt vector location upon wake-up. Otherwise, execution will continue from the next instruction following the PWRSAV instruction.
15.13.2
PWM Operation in Idle Mode

When the device enters Idle mode, the system clock sources remain functional and the CPU stops executing code. The PWM module can optionally continue to operate in Idle mode.The PTSIDL bit (PTCON<13>) selects if the PWM module will stop in Idle mode or continue to operate normally. If PTSIDL = 0, the module will operate normally when the device enters Idle mode. The PWM time base interrupt, if enabled, can be used to wake the device from Idle. If the PWM time base interrupt enable bit is set (PTIE = 1), then the device will wake from Idle when the PWM time base interrupt is generated. If the PWM time base interrupt priority is greater than the current CPU priority, then program execution will start at the PWM interrupt vector location upon wake-up. Otherwise, execution will continue from the next instruction following the PWRSAV instruction. If PTSIDL = 1, the module will stop in Idle mode. If the PWM module is programmed to stop in Idle mode, the operation of the PWM outputs and fault input pins will be the same as the operation in Sleep mode. (See discussion in Section 15.13.1 PWM Operation in Sleep mode.)
15.14
Special Features for Device Emulation

The PWM module has a special feature to support the debugging environment. All enabled PWM pins can be optionally tri-stated when the hardware emulator or debugger device is halted to examine memory contents. The user should install pull-up or pull-down resistors to ensure the PWM outputs are driven to the correct state when device execution is halted. The function of the PWM output pins at a device Reset and the output pin polarity is determined by three device configuration bits (see Section 15.9 PWM Output and Polarity Control). The hardware debugger or emulation tool provides a method to change the values of these configuration bits. Please refer to the tools users manual for more information.
15
Motor Control PWM
DS70062D-page 15-39
Table 15-8:
Bit 13 PWMIP<2:0> FLTAIP<2:0> PTSIDL PWM Time Base register PWM Time Base Period register PWM Special Event Compare register Dead Time B Value register FAOV3H POUT4L FBOV3H POVD3L PWM Duty Cycle #1 register PWM Duty Cycle #2 register PWM Duty Cycle #3 register PWM Duty Cycle #4 register POVD2H POVD2L POVD1H POVD1L POUT4H FBOV3L FBOV2H FBOV2L FBOV1H FBOV1L FLTBM FAOV3L FAOV2H FAOV2L FAOV1H FAOV1L FLTAM FAEN4 FBEN4 DTS4A DTS4I DTS3A DTS3I DTS2A DTAPS<1:0> SEVOPS<3:0> IUE DTS2I FAEN3 FBEN3 POUT2L PMOD4 PMOD3 PMOD2 PMOD1 PEN4H PEN3H PEN2H PEN1H PEN4L PEN3L PEN2L OSYNC DTS1A FAEN2 FBEN2 POUT1H PEN1L UDIS DTS1I FAEN1 FBEN1 POUT1L PTOPS<3:0> PTCKPS<1:0> FLTBIP<2:> PTMOD<1:0> FLTBIE FLTAIE PWMIE FLTBIF FLTAIF PWMIF Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Value on Reset
Registers Associated with the 8-Output PWM Module

0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100 0100 0100 0100 0100 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0111 1111 1111 1111 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 00-0 0000 0000 0000 0000 0000 0000 1111 1111 00-0 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000
Name
ADR
Bit 15
Bit 14
INTCON1
0080
NSTDIS
INTCON2
0082
ALTIVT
IFS2
0088
IEC2
0090
DS70062D-page 15-40
Dead Time A Value register POUT3H POUT3L POUT2H
IPC9
00A6
IPC10
00A8
IPC11
00AA
PTCON
01C0
PTEN
PTMR
01C2
PTDIR
PTPER
01C4
SEVTCMP
01C6
SEVTDIR
PWMCON1
01C8
PWMCON2
01CA
DTCON1
01CC
DTBPS<1:0>
DTCON2
01CE
FLTACON
01D0
FAOV4H
FAOV4L
FLTBCON
01D2
FBOV4H
FBOV4L
OVDCON
01D4
POVD4H
POVD4L POVD3H
PDC1
01D6
PDC2
01D8
PDC3
01DA
PDC4
01DC
Note
1: 2: 3:
Reset state of PENxx control bits depends on the state of the PWMPIN device configuration bit. Shaded register and bit locations not implemented for the 6-output MCPWM module. The IUE bit is not implemented on the dsPIC30F6010 device.
Table 15-9:
Bit 13 PWMIP<2:0> FLTAIP<2:0> PTSIDL PWM Time Base register PWM Time Base Period register PWM Special Event Compare register FAOV3H POVD3H PWM Duty Cycle #1 register PWM Duty Cycle #2 register PWM Duty Cycle #3 register POVD3L POVD2H POVD2L POVD1H POVD1L POUT3H FAOV3L FAOV2H FAOV2L FAOV1H FAOV1L FLTAM FAEN4

Registers Associated with the 6-Output PWM Module

Bit 12 PTOPS<3:0> PTCKPS<1:0> PTMOD<1:0> FLTAIE PWMIE FLATIF PWMIF Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Value on Reset 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100 0100 0100 0100 0100 0000 0000 0000 0000 0000 0000 0000 0000 0111 1111 1111 1111 0000 0000 0000 0000 PEN2H IUE FAEN3 POUT2L PEN1H PEN3L PEN2L OSYNC FAEN2 POUT1H PEN1L UDIS FAEN1 POUT1L 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 00-0 0000 0000 1111 1111 00-0 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000
Name
ADR
Bit 15
Bit 14
INTCON1
0080
NSTDIS
INTCON2
0082
ALTIVT
IFS2
0088
IEC2
0090
IPC9
00A6
IPC10
00A8
PTCON
01C0
PTEN
PTMR
01C2
PTDIR
Motor Control PWM

DTAPS<1:0> PMOD3 PMOD2 PMOD1 PEN3H Dead Time A Value register POUT3L POUT2H
PTPER
01C4
SEVTCMP
01C6
SEVTDIR
PWMCON1
01C8
PWMCON2
01CA
DTCON1
01CC
Reserved
01CE
FLTACON
01D0
Reserved
01D2
OVDCON
01D4
PDC1
01D6
PDC2
01D8
PDC3
01DA
Note
1: 2: 3:
Reset state of PENxx control bits depends on the state of the PWMPIN device configuration bit. Shaded register and bit locations not implemented for the 6-output MCPWM module. The IUE bit is not implemented on the dsPIC30F6010 device.
DS70062D-page 15-41
15

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent, and could be used with modification and possible limitations. The current application notes related to the MCPWM module are: Title PIC18CXXX/PIC16CXXX Servomotor Using the dsPIC30F for Sensorless BLDC Control Using the dsPIC30F for Vector Control of an ACIM Sensored BLDC Motor Control Using dsPIC30F2010 An Introduction to AC Induction Motor Control Using the dsPIC30F MCU Note: Application Note # AN696 AN901 AN908 AN957 AN984
DS70062D-page 15-42

Revision B
This revision provides expanded information for the dsPIC30F MCPWM module.
Revision C
Revision D
This revision includes the Immediate Update Enable Capability (IUE) bit.
15
Motor Control PWM
DS70062D-page 15-43

NOTES:
DS70062D-page 15-44
16
Quadrature Encoder Interface (QEI)
Section 16. Quadrature Encoder Interface (QEI)

HIGHLIGHTS
This section of the manual contains the following major topics: 16.1 Module Introduction ..................................................................................................... 16-2 16.2 Control and Status Registers ....................................................................................... 16-4 16.3 Programmable Digital Noise Filters ............................................................................. 16-9 16.4 Quadrature Decoder .................................................................................................. 16-10 16.5 16-bit Up/Down Position Counter............................................................................... 16-12 16.6 Using QEI as an Alternate 16-bit Timer/Counter........................................................ 16-16 16.7 Quadrature Encoder Interface Interrupts ................................................................... 16-17 16.8 I/O Pin Control ........................................................................................................... 16-18 16.9 QEI Operation During Power Saving Modes ............................................................. 16-19 16.10 Effects of a Reset....................................................................................................... 16-19 16.11 Design Tips ................................................................................................................ 16-21 16.12 Related Application Notes.......................................................................................... 16-22 16.13 Revision History ......................................................................................................... 16-23
DS70063C-page 16-1

16.1 16.1.1 Module Introduction Features Overview
Quadrature encoders (a.k.a. Incremental encoders or Optical encoders) are used in position and speed detection of rotating motion systems. Quadrature encoders enable closed loop control of many motor control applications, such as Switched Reluctance (SR) motor and AC Induction Motor (ACIM). A typical incremental encoder includes a slotted wheel attached to the shaft of the motor and an emitter/detector module sensing the slots in the wheel. Typically, three outputs, termed: Phase A, Phase B and INDEX, provide information that can be decoded to provide information on the movement of the motor shaft including distance and direction. The two channels, Phase A (QEA) and Phase B (QEB), have a unique relationship. If Phase A leads Phase B, then the direction (of the motor) is deemed positive or forward. If Phase A lags Phase B then the direction (of the motor) is deemed negative or reverse. A third channel, termed index pulse, occurs once per revolution and is used as a reference to establish an absolute position. See Figure 16-1 for a relative timing diagram of these three signals. The quadrature signals produced by the encoder can have four unique states. These states are indicated for one count cycle in Figure 16-1. Note that the order of the states are reversed when the direction of travel is changed. A Quadrature Decoder captures the phase signals and index pulse and converts the information into a numeric count of the position pulses. Generally, the count will increment when the shaft is rotating one direction and decrement when the shaft is rotating in the other direction. Figure 16-1: Quadrature Encoder Interface Signals 1 Cycle QEA QEB INDX 01 00 10 11 Reverse Travel QEA QEB INDX 11 10 00 01
Forward Travel
DS70063C-page 16-2
16
The Quadrature Encoder Interface (QEI) module provides an interface to incremental encoders. The QEI consists of quadrature decoder logic to interpret the Phase A and Phase B signals and an up/down counter to accumulate the count. Digital glitch filters on the inputs condition the input signal. Figure 16-2 depicts a simplified block diagram of the QEI Module. The QEI module includes: Figure 16-2: Three input pins for two phase signals and index pulse Programmable digital noise filters on inputs Quadrature decoder providing counter pulses and count direction 16-bit up/down position counter Count direction status X2 and X4 count resolution 2 modes of position counter reset General Purpose16-bit timer/counter mode Interrupts generated by QEI or counter events
Quadrature Encoder Interface Module Simplified Block Diagram

Clock Divider QEA Digital Filter TCY
QEB
Digital Filter
Quadrature Decoder Logic
CLOCK DIR
16-Bit Up/Down Counter (POSCNT)
Reset
INDX
Digital Filter
Comparator/ Zero Detect
EQUAL
Max Count Register (MAXCNT) UPDN
DS70063C-page 16-3

16.2 Control and Status Registers
The QEI module has four user-accessible registers. The registers are accessible in either byte or word mode. The registers are shown in Figure 16-3 and listed below: Control/Status Register (QEICON) This register allows control of the QEI operation and status flags indicating the module state. Digital Filter Control Register (DFLTCON) This register allows control of the digital input filter operation. Position Count Register (POSCNT) This location allows reading and writing of the 16-bit position counter. Maximum Count Register (MAXCNT) The MAXCNT register holds a value that will be compared to the POSCNT counter in some operations. Note: The POSCNT register allows byte accesses, however, reading the register in byte mode may result in partially updated values in subsequent reads. Either use word mode reads/writes or ensure that the counter is not counting during byte operations. QEI Programmers Model
QEICON (16 bits) Bit 15 Bit 7 Bit 0 DFLTCON (8 bits) Bit 0 POSCNT (16 bits) Bit 15 Bit 15 Bit 0 MAXCNT (16 bits) Bit 0
Figure 16-3:
Register 16-1 and Register 16-3 define the QEI module control and digital filter control registers, QEICON and DFLTCON.
DS70063C-page 16-4
16
Register 16-1: Upper Byte: R/W-0 CNTERR bit 15 QEICON: QEI Control Register U-0 R/W-0 QEISIDL R-0 INDEX R/W-0 UPDN R/W-0 R/W-0 QEIM<2:0> R/W-0 bit 8 Lower Byte: R/W-0 R/W-0 SWPAB PCDOUT bit 7 bit 15
R/W-0 TQGATE
R/W-0 R/W-0 TQCKPS<1:0>
R/W-0 POSRES
R/W-0 TQCS
R/W-0 UDSRC bit 0
CNTERR: Count Error Status Flag bit 1 = Position count error has occurred 0 = No position count error has occurred (CNTERR flag only applies when QEIM<2:0> = 110 or 100) Unimplemented: Read as 0 QEISIDL: Stop in Idle Mode bit 1 = Discontinue module operation when device enters Idle mode 0 = Continue module operation in Idle mode INDEX: Index Pin State Status bit (Read Only) 1 = Index pin is High 0 = Index pin is Low UPDN: Position Counter Direction Status bit 1 = Position Counter Direction is positive (+) 0 = Position Counter Direction is negative (-) (Read only bit when QEIM<2:0> = 1XX) (Read/Write bit when QEIM<2:0> = 001) QEIM<2:0>: Quadrature Encoder Interface Mode Select bits 111 = Quadrature Encoder Interface enabled (x4 mode) with position counter reset by match (MAXCNT) 110 = Quadrature Encoder Interface enabled (x4 mode) with Index Pulse reset of position counter 101 = Quadrature Encoder Interface enabled (x2 mode) with position counter reset by match (MAXCNT) 100 = Quadrature Encoder Interface enabled (x2 mode) with Index Pulse reset of position counter 011 = Unused (Module disabled) 010 = Unused (Module disabled) 001 = Starts 16-bit Timer 000 = Quadrature Encoder Interface/Timer off SWPAB: Phase A and Phase B Input Swap Select bit 1 = Phase A and Phase B inputs swapped 0 = Phase A and Phase B inputs not swapped PCDOUT: Position Counter Direction State Output Enable bit 1 = Position Counter Direction Status Output Enable (QEI logic controls state of I/O pin) 0 = Position Counter Direction Status Output Disabled (Normal I/O pin operation) TQGATE: Timer Gated Time Accumulation Enable bit 1 = Timer gated time accumulation enabled 0 = Timer gated time accumulation disabled TQCKPS<1:0>: Timer Input Clock Prescale Select bits 11 = 1:256 prescale value 10 = 1:64 prescale value 01 = 1:8 prescale value 00 = 1:1 prescale value (Prescaler utilized for 16-bit timer mode only)
bit 14 bit 13
bit 12
bit 11
bit 10-8
bit 7
bit 6
bit 5
bit 4-3
DS70063C-page 16-5

Register 16-1: bit 2 QEICON: QEI Control Register (Continued)
POSRES: Position Counter Reset Enable bit 1 = Index Pulse resets Position Counter 0 = Index Pulse does not reset Position Counter (Bit only applies when QEIM<2:0> = 100 or 110) TQCS: Timer Clock Source Select bit 1 = External clock from pin QEA (on the rising edge) 0 = Internal clock (TCY) UDSRC: Position Counter Direction Selection Control bit 1 = QEB pin State Defines Position Counter Direction 0 = Control/Status bit, UPDN (QEICON<11>), Defines Timer Counter (POSCNT) direction Note: When configured for QEI mode, control bit is a dont care Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 1
bit 0
DS70063C-page 16-6
16
Register 16-2: Upper Byte: U-0 bit 15 DFLTCON: Digital Filter Control Register (dsPIC30F6010 Only) U-0 U-0 U-0 U-0 U-0 U-0 R/W-0 CEID bit 8
Lower Byte: R/W-0 QEOUT bit 7 bit 15-9 bit 8
R/W-0
R/W-0 QECK<2:0>
R/W-0
R/W-0 INDOUT
R/W-0
R/W-0 R/W-0 INDCK<2:0> bit 0
Unimplemented: Read as 0 CEID: Count Error Interrupt Disable bit 1 = Interrupts due to position count errors disabled 0 = Interrupts due to position count errors enabled QEOUT: QEA/QEB Digital Filter Output Enable bit 1 = Digital filter outputs enabled 0 = Digital filter outputs disabled (Normal pin operation) QECK<2:0>: QEA/QEB Digital Filter Clock Divide Select bits 111 = 1:256 Clock Divide 110 = 1:128 Clock Divide 101 = 1:64 Clock Divide 100 = 1:32 Clock Divide 011 = 1:16 Clock Divide 010 = 1:4 Clock Divide 001 = 1:2 Clock Divide 000 = 1:1 Clock Divide INDOUT: Index Channel Digital Filter Output Enable bit 1 = Digital filter output is enabled 0 = Digital filter output is disabled (Normal pin operation) INDCK<2:0>: Index Channel Digital Filter Clock Divide Select bits 111 = 1:256 Clock Divide 110 = 1:128 Clock Divide 101 = 1:64 Clock Divide 100 = 1:32 Clock Divide 011 = 1:16 Clock Divide 010 = 1:4 Clock Divide 001 = 1:2 Clock Divide 000 = 1:1 Clock Divide Note: The available control bits in the DFLTCON Register may vary depending on the dsPIC30F device that is used. Refer to Register 16-2 and Register 16-3 for details.
bit 7
bit 6-4
bit 3
bit 2-0
DS70063C-page 16-7

Register 16-3: Upper Half: U-0 bit 15 DFLTCON: Digital Filter Control Register (All dsPIC30F devices except dsPIC30F6010) U-0 U-0 U-0 U-0 R/W-0 R/W-0 IMV<1:0> R/W-0 CEID bit 8
Lower Half: R/W-0 QEOUT bit 7 bit 15-11 Unimplemented: Read as 0 bit 10-9
R/W-0 QECK<2:0>
U-0
U-0
U-0
U-0 bit 0
IMV<1:0>: Index Match Value These bits allow the user to specify the state of the QEA and QEB input pins during an Index pulse when the POSCNT register is to be reset. In 4X Quadrature Count Mode: IMV1= Required State of Phase B input signal for match on index pulse IMV0= Required State of Phase A input signal for match on index pulse In 2X Quadrature Count Mode: IMV1= Selects Phase input signal for Index state match (0 = Phase A, 1 = Phase B) IMV0= Required State of the selected Phase input signal for match on index pulse CEID: Count Error Interrupt Disable 1 = Interrupts due to count errors are disabled 0 = Interrupts due to count errors are enabled QEOUT: QEA/QEB/INDX pin Digital Filter Output Enable 1 = Digital filter outputs enabled 0 = Digital filter outputs disabled (normal pin operation) QECK<2:0>: QEA/QEB/INDX Digital Filter Clock Divide Select Bits 111 = 1:256 Clock Divide 110 = 1:128 Clock Divide 101 = 1:64 Clock Divide 100 = 1:32 Clock Divide 011 = 1:16 Clock Divide 010 = 1:4 Clock Divide 001 = 1:2 Clock Divide 000 = 1:1 Clock Divide Unimplemented: Read as 0 Note: The available control bits in the DFLTCON Register may vary depending on the dsPIC30F device that is used. Refer to Register 16-2 and Register 16-3 for details.
bit 8
bit 7
bit 6-4
bit 3-0
y = Value set from configuration bits on POR or BOR
DS70063C-page 16-8
16
16.3 Programmable Digital Noise Filters
The digital noise filter section is responsible for rejecting noise on the incoming index and quadrature signals. Schmitt trigger inputs and a three-clock cycle delay filter combine to reject low level noise and large, short duration noise spikes that typically occur in noise-prone applications such as a motor system applications. The filter ensures that the filtered output signals are not permitted to change until a stable value has been registered for three consecutive filter cycles. The rate of the filter clocks determines the low passband of the filter. A slower filter clock results in a passband rejecting lower frequencies than a faster filter clock. The filter clock is the device FCY clock divided by a programmable divisor. Setting the QEOUT bit (DFLTCON<7>) enables the filter for channels QEA and QEB. The QECK<2:0> bits (DFLTCON<6:4>) specify the filter clock divisor used for channels QEA and QEB. Setting the INDOUT bit (DFLTCON<3>) enables the filter for the index channel. The INDCK<2:0> bits (DFLTCON<2:0>) specify the filter clock divisor used for the index channel. At reset, the filters for all channels are disabled. Some devices do not have separate control bits for the QEx input digital filters and the INDX input digital filter. For these devices, the QEOUT and QECK<2:0> control bits set the digital filter characteristics for both the QEA/QEB and INDX pins. See Register 16-2 and Register 16-3 for more information. Figure 16-4 depicts a simplified block diagram for the digital noise filter. Figure 16-4: Simplified Digital Noise Filter Block Diagram
Non-filtered Filtered
0 1
QEn Filter Output
J CK QEn pin TCY D Q CK CK D Q CK CK D Q CK CK D Q CK Clock Divider Circuit 3 K
QEOUT
TCY
QECK<2:0> Note: n denotes the phase input, A or B.
Figure 16-5:
Signal Propagation Through Filter, 1:1 Filter Clock Divide
TCY QEn Pin QEn Filter
DS70063C-page 16-9

16.4 Quadrature Decoder
Position measurement modes are selected when QEIM2 = 1 (QEICON<10>). When QEIM1 = 1 (QEICON<9>), the x4 measurement mode is selected and the QEI logic clocks the position counter on both edges of the Phase A and Phase B input signals. The x4 measurement mode provides for finer resolution data (more position counts) to determine the encoder position. Figure 16-6: Quadrature Decoder Signals in 4X Mode
QEA QEB
count_clock POSCNT UPDN

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 -1 -1 -1 -1 -1 -1 -1 -1 -1
When QEIM1 = 0, the x2 measurement mode is selected and the QEI logic only looks at the rising and falling edge of the Phase A input for the position counter increment rate. Every rising and falling edge of the Phase A signal causes the position counter to increment or decrement. The Phase B signal is still utilized for the determination of the counter direction, exactly like the x4 measurement mode. Figure 16-7: Quadrature Decoder Signals in 2X Mode
QEA QEB count_clock POSCNT UPDN

+1 +1 +1 +1 +1 -1 -1 -1 -1
DS70063C-page 16-10
16
16.4.1 Explanation of Lead/Lag Test
The lead/lag test is performed by the quadrature decoder logic to determine the phase relationship of the QEA and QEB signals and hence whether to increment or decrement the POSCNT register. The Table 16-1 clarifies the lead/lag test. Table 16-1: Present Transition QEA Lead/Lag Test Description Previous Transition QEB QEB QEA QEB QEB QEA QEA QEA QEB QEA QEA QEB Condition QEA leads QEB channel QEA lags QEB channel Direction Change QEA lags QEB channel QEA leads QEB channel Direction Change QEA lags QEB channel QEA leads QEB channel Direction Change QEA leads QEB channel QEA lags QEB channel Direction Change Set UPDN Clear UPDN Toggle UPDN Clear UPDN Set UPDN Toggle UPDN Clear UPDN Set UPDN Toggle UPDN Set UPDN Clear UPDN Toggle UPDN Action Increment POSCNT Decrement POSCNT Increment or Decrement POSCNT Decrement POSCNT Increment POSCNT Increment or Decrement POSCNT Decrement POSCNT Increment POSCNT Increment or Decrement POSCNT Increment POSCNT Decrement POSCNT Increment or Decrement POSCNT
QEA
QEB
QEB
16.4.2
Count Direction Status

As mentioned in the previous section, the QEI logic generates an UPDN signal based upon the Phase A and Phase B time relationship. The UPDN signal may be output on an I/O pin. Setting the PCDOUT bit (QEICON<6>) and clearing the appropriate TRIS bit associated with the pin will cause the UPDN signal to drive the output pin. In addition to the output pin, the state of this internal UPDN signal is supplied to a SFR bit QEICON<11> as a Read-Only bit, notated as UPDN.
16.4.3
Encoder Count Direction

The direction of quadrature counting is determined by the SWPAB bit (QEICON<7>). If the SWPAB = 0, the Phase A input is fed to the A input of the quadrature counter and the Phase B input is fed to the B input of the quadrature counter. Therefore as the Phase A signal leads the Phase B signal, the quadrature counter is incremented on each edge. This (A signal leads the B signal) is defined as the forward direction of motion. Setting the SWPAB bit, (QEICON<7>), to a logic 1 causes the Phase A input to be fed to the B input of the quadrature counter and the Phase B signal to be fed to the A input of the quadrature counter. Therefore, if the Phase A signal leads the Phase B signal at the dsPIC30F device pins, the Phase A input to the quadrature counter will now lag the Phase B input. This is recognized as rotation in the reverse direction and the counter will be decremented on each quadrature pulse.
DS70063C-page 16-11

16.4.4 Quadrature Rate
The RPM of the position control system will vary. The RPMs along with the quadrature encoder line count determine the frequency of the QEA and QEB input signals. The quadrature encoder signals can be decoded such that a count pulse is generated for every quadrature signal edge. This allows an angular position measurement resolution of up to 4 times the encoder line count. For example: a 6,000 RPM motor utilizing a 4096 line encoder yields a quadrature count rate of: ((6000/60) * (4096*4)) = 1.6384 MHz. Likewise, a 10,000 RPM motor utilizing a 8,192 line encoder yields a quadrature count rate of: ((10000/60) * (8192*4)) = 5.46 MHz. The QEI allows a quadrature frequency of up to FCY/3. For example, if FCY = 30 MHz, the QEA and QEB signals may have a maximum frequency of 10 MHz. Refer to the Electrical Specifications section of the device data sheet for further details.
16.5
16-bit Up/Down Position Counter

The 16-bit Up/Down Counter counts up or down on every count pulse which is generated by the quadrature decoder logic. The counter then acts as an integrator, whose count value is proportional to position. The direction of the count is determined by the quadrature decoder. The user software may examine the contents of the count by reading the POSCNT register. The user software may also write to the POSCNT register to initialize a count. Changing the QEIM bits does not affect the position counter register contents.
16.5.1
Using the Position Counter

The system may utilize position counter data in one of several methods. In some systems, the position count is accumulated consistently and taken as an absolute value representing the total position of the system. For a typical example, assume that a quadrature encoder is affixed to a motor controlling the print head in a printer. In operation, the system is initialized by moving the print head to the maximum left position and resetting the POSCNT register. As the print head moves to the right, the quadrature encoder will begin to accumulate counts in the POSCNT register. As the print head moves to the left, the accumulated count will decrease. As the print head reaches the right most position, the maximum position count should be reached. If the maximum count is less than 216, the QEI module can encode the entire range of motion. If, however, the maximum count is more than 216, the additional count precision must be captured by the user software. Generally, to accomplish this, the module is set into a mode where it resets the counter at match of a maximum count. QEIM0 = 1 enables modes where the MAXCNT register is used to reset the position counter. When the counter reaches a pre-determined maximum count while incrementing or reaches zero while decrementing, the count is reset and an interrupt is generated to allow the user software to increment or decrement a software counter containing the most significant bits of the position count. The maximum count can be 0xFFFF, to enable a full range of the QEI counter and software counter or some smaller value of significance, such as the number of counts for one encoder revolution. In other systems, the position count may be cyclic. The position count is only used to reference the position of the wheel within number of rotations determined by the index pulse. For example, a tool platform moved by a screw rod uses a quadrature encoder attached to the screw rod. In operation, the screw may require 5.5 rotations to achieve the desired position. The user software will detect 5 index pulses to count the full rotations and use the position count to measure the remaining half rotation. In this method, the index pulse resets the position counter to initialize the counter at each rotation and generates an interrupt for each rotation. QEIM0 = 0 enables these modes.
DS70063C-page 16-12
16
16.5.2 Using MAXCNT to Reset the Position Counter
When the QEIM0 bit is 1, the position counter will reset on a match of the position count with predetermined high and low values. The index pulse reset mechanism is not utilized. For this mode the position counter reset mechanism operates as follows: (See Figure 16-8 for related timing details). - If the encoder is traveling in the forward direction e.g., QEA leads QEB, and the value in the POSCNT register matches the value in the MAXCNT register, POSCNT will reset to zero on the next occurring quadrature pulse edge that increments POSCNT. An interrupt event is generated on this rollover event. - If the encoder is travelling in the reverse direction e.g., QEB leads QEA, and the value in the POSCNT register counts down to 0, the POSCNT is loaded with the value in the MAXCNT register on the next occurring quadrature pulse edge that decrements POSCNT. An interrupt event is generated on this underflow event. When using MAXCNT as a position limit, remember the position counter will count at either 2X or 4X of the encoder counts. For standard rotary encoders, the appropriate value to write to MAXCNT would be 4N 1 for 4x position mode and 2N 1 for 2x position mode, where N is the number of counts per revolution of the encoder. For absolute position information where the range of the system exceeds 216, it is also appropriate to load a value of 0xFFFF into the MAXCNT register. The module will generate an interrupt on rollover or underflow of the position counter. Figure 16-8: Rollover/Rollunder Reset-Up/Down Position Counter
QEA QEB
count_clock POSCNT UPDN MAXCNT

0006
0001 0002 0003 0004 0005 0006 0000 0001 0002 0003 0002 0001 0000 0006 0005 0004 0003 0002 0001
POSCNT = MAXCNT Generate QEI Interrupt POSCNT set to 0000
POSCNT = 0000 Generate QEI Interrupt POSCNT set to MAXCNT
DS70063C-page 16-13

16.5.3 Using Index to Reset Position Counter
When QEIM<0> = 0, the index pulse is utilized for resetting the position counter. For this mode the position counter reset mechanism operates as follows: (See Figure 16-9 for related timing details). The position count is reset each time an index pulse is received on the INDEX pin. If the encoder is travelling in the forward direction e.g., QEA leads QEB, POSCNT is reset to 0. If the encoder is travelling in the reverse direction e.g., QEB leads QEA, the value in the MAXCNT register is loaded into POSCNT. Figure 16-9: Reset by Index Mode-Up/Down Position Counter
Quadrature State QEA QEB INDX
count_clock POSCNT 00E3 00E4 00E5 00E6 0000 0001 0002 0003 0004 UPDN Generate QEI Interrupt POSCNT set to MAXCNT Recognize Index Wheel Reverses 0005 0004 0003 0002 0001 0000 00E6 00E5 00E4 00E3 00E2 00E1 00E0
Recognize Index Generate QEI Interrupt POSCNT set to 0000
16.5.3.1
Index Pulse Detection Criteria Incremental encoders from different manufacturers use differing timing for the index pulse. The index pulse may be aligned to any of the 4 quadrature states and may have a pulse width of either a full cycle (4 quadrature states), a half cycle (2 quadrature states) or a quarter cycle (1 quadrature state). Index pulses of a full cycle width or a half cycle width are normally termed ungated and index pulses of a quarter cycle width are normally termed gated. Regardless of the type of index pulse provided, the QEI maintains symmetry of the count as the wheel reverses direction. This means the index pulse must reset the position counter at the same relative quadrature state transition as the wheel rotates in the forward or reverse direction. For example, in Figure 16-9, the first index pulse is recognized and resets POSCNT as the quadrature state changes from 4 to 1 as highlighted in the diagram. The QEI latches the state of this transition. Any subsequent index pulse detection will use that state transition for the reset. As the wheel reverses, the index pulse again occurs, however the reset of the position counter cannot occur until the quadrature state changes from 1 to 4, again highlighted in the diagram. Note: The QEI index logic ensures that the POSCNT register is always adjusted at the same position relative to the index pulse, regardless of the direction of travel.
DS70063C-page 16-14
16
16.5.3.2 IMV Control Bits The IMV<2:0> control bits are available on some dsPIC devices that have the QEI module. (See Register 16-3). These control bits allow the user to select the state of the QEA and QEB signals for which an index pulse reset will occur. Devices that do not have these control bits will select the QEA and QEB states automatically during the first occurrence of an index pulse. 16.5.3.3 Index Pulse Status The INDEX bit (QEICON<12>) provides status of the logic state on the index pin. This status bit is very useful in position control systems during the homing sequence, where the system searches for a reference position. The index bit indicates the status of the index pin after being processed by the digital filter, if it is enabled. 16.5.3.4 Using the Index Pin and MAXCNT for Error Checking When the counter operates in reset on index pulse mode, the QEI will also detect POSCNT register boundary conditions. This may be used to detect system errors in the incremental encoder system. For example, assume a wheel encoder has 100 lines. When utilized in x4 measurement mode and reset on the index pulse, the counter should count from 0 to 399 (0x018E) and reset. If the POSCNT register ever achieves the values of 0xFFFF or 0x0190, some sort of system error has occurred. The contents of the POSCNT register is compared with MAXCNT + 1, if counting up, and with 0xFFFF, if counting down. If the QEI detects one of these values, a position count error condition is generated by setting the CNTERR bit (QEICON<15>) and optionally generating a QEI interrupt. If the CEID control bit (DFLTCON<8>) is cleared (default), then a QEI interrupt will be generated when a position count error is detected. If the CEID control bit is set, then an interrupt will not occur. The position counter continues to count encoder edges after detecting a position count error. No interrupt is generated for subsequent position count error events until CNTERR is cleared by the user. 16.5.3.5 Position Counter Reset Enable The position counter reset enable bit, POSRES (QEICON<2>) enables reset of the position counter when the index pulse is detected. This bit only applies when the QEI module is configured for modes, QEIM<2:0> = 100 or 110. If the POSRES bit is set to a logic 1 then the position counter is reset when the index pulse is detected as described in this section. If the POSRES bit is set to a logic 0, then the position counter is not reset when the index pulse is detected. The position counter will continue counting up or down and be reset on the rollover or underflow condition. The QEI continues to generate interrupts on the detection of the index pulse.
DS70063C-page 16-15

16.6 Using QEI as an Alternate 16-bit Timer/Counter
When the QEI module is configured QEIM<2:0> = 001, the QEI function is disabled and the QEI module is configured as a 16-bit timer/counter. The setup and control for the auxiliary timer is accomplished through the QEICON register. The QEI timer functions similar to the other dsPIC30F timers. Refer to Section 12. Timers for a general discussion of timers. When configured as a timer, the POSCNT register serves as a timer register similar to the TMRn registers of the GP timers. The MAXCNT register serves as a period register similar to the PRn registers of the GP timers. When a timer/period register match occurs, the QEIF flag asserts. Note: Changing operational modes, i.e., from QEI to Timer or Timer to QEI will not affect the Timer/Position Count Register contents.
Figure 16-10: QEI as Timer/Counter Block Diagram

TQCS TQGATE TQCKPS TCY
Gated TCY
2 00 01 10 11 D Q Prescaler 1, 8, 64, 256 TQGATE TQGATE 1 0 QEIF Event Flag
QEA
Programmable Digital Filter
Synchronize Det
UDSRC UPDN QEB Programmable Digital Filter 0 1
CK Q
16-bit Up/Down Counter (POSCNT) Reset
Comparator/Zero Detect
Equal
Max Count Register (MAXCNT)
DS70063C-page 16-16
16
16.6.1 Up/Down Timer Operation
The QEI timer can increment or decrement. This is a unique feature over most other timers. When the timer is configured to count up, the timer (POSCNT) will increment until the count matches the period register (MAXCNT). The timer resets to zero and restarts incrementing. When the timer is configured to count down, the timer (POSCNT) will decrement until the count matches the period register (MAXCNT). The timer resets to zero and restarts decrementing. When the timer is configured to count down some general operation guidelines must be followed for correct operation. 1. The MAXCNT register will serve as the period match register but because the counter is decrementing, the desired match value is 2 count. For example, to count 0x1000 clocks, the period register must be loaded with 0xF000. On a match condition, the timer resets to zero.
2.
Either an I/O pin or a SFR control bit specify the count direction control. Control bit UDSRC (QEICON<0>) determines what controls the timer count direction state. When UDSRC = 1, the timer count direction is controlled from the QEB pin. If the QEB pin is 1, the count direction will be incrementing. If the QEB pin is 0, the count direction will be decrementing. When UDSRC = 0, the timer count direction is controlled from the UPDN bit (QEICON<11>). When UPDN = 1, the timer increments. When UPDN = 0, the timer decrements.
16.6.2
Timer External Clock

The TQCS bit (QEICON<1>) selects internal or external clock. The QEI timer can use the QEA pin as an external clock input when TQCS is set. The QEI timer does not support the external asynchronous counter mode. If using an external clock source the clock will automatically be synchronized to the internal instruction cycle (TCY).
16.6.3
Timer Gate Operation

The QEA pin functions as a timer gate when the TQGATE bit (QEICON<5>) is set and TQCS is cleared. In the event TQCS and TQGATE are concurrently set, the timer does not increment and does not generate an interrupt.
16.7
Quadrature Encoder Interface Interrupts

Depending on the mode of the QEI, the QEI will generate interrupts for the following events: When operating in reset on match mode, QEIM<2:0> = 111 and 101, an interrupt occurs on position counter rollover/underflow. When operating in reset on index mode, QEIM<2:0> = 110 and 100, an interrupt occurs on detection of index pulse and optionally when CNTERR bit is set. When operating as a Timer/Counter, QEIM<2:0> = 001, an interrupt occurs on a period match event or a timer gate falling edge event when TQGATE = 1. When a QEI interrupt event occurs, the QEIIF bit (IFS2<8>) is asserted and an interrupt will be generated if enabled. The QEIIF bit must be cleared in software. Enabling the QEI interrupt is accomplished via the respective enable bit, QEIIE (IEC2<8>).
DS70063C-page 16-17

16.8 I/O Pin Control
Enabling the QEI module causes the associated I/O pins to come under the control of the QEI and prevents lower priority I/O functions such as Ports from affecting the I/O pin. Depending on the mode specified by QEIM<2:0> and other control bits, the I/O pins may assume differing functions, as shown in Table 16-2 and Table 16-3. Table 16-2: Pin Name QEA Quadrature Encoder Module Pinout I/O Descriptions Pin Type Buffer Type Description I I I I I I O ST ST ST ST ST ST Quadrature Encoder Phase A Input, or Auxiliary Timer External Clock Input, or Auxiliary Timer External gate Input Quadrature Encoder Phase B Input, or Auxiliary Timer Up/Down select input Quadrature Encoder Index Pulse Input Position Up/Down Counter Direction Status, QEI mode
QEB INDX UPDN
Legend: I = Input, O = Output, ST = Schmitt Trigger Table 16-3:

QEIM<2:0>
Module I/O Mode Functions

PCDOUT TQGATE UDSRC TQCS QEA pin QEB pin INDX pin UPDN pin
000,010,011 Module Off 001 Timer Mode
N/A N/A
N/A 0 1 0 1 0 1
N/A 0 0 1 1 N/A N/A
N/A 0 0 0 0 1 1 Input (TQGATE) Port not disabled Input (TQGATE) Port not disabled Input (TQCKI) Port not disabled Input (TQCKI) Port not disabled Input (QEA) Input (UPDN)
Input (UPDN)
Input (UPDN) 101,111 0 N/A N/A N/A Input QEI (QEB) Reset by count 1 N/A N/A N/A Input (QEA) Input Output (QEB) (UPDN) 100,110 0 N/A N/A N/A Input (QEA) Input Input QEI (QEB) (INDX) Reset by Index 1 N/A N/A N/A Input (QEA) Input Input Output (QEB) (INDX) (UPDN) Note: Empty slot indicates pin not used by QEI in this configuration,pin controlled by I/O port logic.
DS70063C-page 16-18
16
16.9 16.9.1 QEI Operation During Power Saving Modes When the Device Enters Sleep
When the device enters Sleep mode, the QEI will cease all operations. POSCNT will stop at the current value. The QEI will not respond to active signals on the QEA, QEB, INDX or UPDN pins. The QEICON register will remain unchanged. If the QEI is configured as a timer/counter, QEIM<2:0> = 001, and the clock is provided externally, TQCS = 1, the module will also cease operation during Sleep mode. When the module wakes up, the quadrature decoder will accept the next transition on the QEA or QEB signals and compare that transition to the last transition before Sleep to determine the next action.
16.9.2
When the Device Enters Idle

The module will enter a power saving state in Idle mode depending on the QEISIDL bit (QEICON<13>). If QEICSIDL = 1, then the module will enter the power saving mode, similar to actions while entering Sleep mode. If QEICSIDL = 0, then the module will not enter a power saving mode. The module will continue to operate normally while the device is in Idle mode.
16.10
Effects of a Reset
Reset forces module registers to their initial reset state. See Register 16-1 for all initialization and reset conditions for QEI module related registers. The quadrature decoder and the POSCNT counter are reset to an initial state.
DS70063C-page 16-19
Table 16-4:
Bit 12 INDX Position Count Register Maximum Count Register PCFG9 OVBTE COVTE DCIIF DCIIE LVDIP<2:0> DCIIP<2:0> QEIIE PWMIE C2IE INT4IE INT3IE OC8IE OC7IE OC6IE QEIIP<2:0> QEIIF PWMIF C2IF INT4IF INT3IF OC8IF OC7IF OC6IF INT4EP INT3EP INT2EP INT1EP MATHERR ADDRERR STKERR OSCFAIL INT0EP OC5IF OC5IE PCFG8 PCFG7 PCFG6 PCFG5 PCFG4 PCFG3 PCFG2 PCFG1 PCFG0 QEOUT QECK2 QECK1 QECK0 INDOUT INDCK2 INDCK1 UPDN QEIM2 QEIM1 QEIM0 SWPAB PCDOUT TQGATE TQCKPS1 TQCKPS0 POSRES TQCS Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Value on ALL Reset UDSRC 0000 0000 0000 0000 INDCK0 ---- ---- ---- ---0000 0000 0000 0000 1111 1111 1111 1111 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100
Special Function Registers Associated with QEI
Name
Bit 15
Bit 14
Bit 13
QEICON
CNTERR
Unused
QEISIDL
DFLTCON
POSCNT
DS70063C-page 16-20
FLTBIE FLTAIE LVDIE FLTBIF FLTAIF LVDIF OVATE
MAXCNT
ADPCFG
PCFG15
PCFG14 PCFG13 PCFG12 PCFG11 PCFG10
INTCON1
NSTDIS
INTCON2
ALTIVT
IFS2
IEC2
IPC10
FLTAIP<2:0>
Note:
The available control bits in the DFLTCON Register may vary depending on the dsPIC30F device that is used. Refer to Register 16-2 and Register 16-3 for details.
Note:
On many devices, the QEI pins are multiplexed with analog input pins. You will need to ensure that the QEI pins are configured as digital pins using the ADPCFG control register.
16
16.11 Design Tips
Question 1: I have initialized the QEI, but the POSCNT Register does not seem to change when quadrature signals are applied to the QEA/QEB pins.
Answer: On many devices, the QEI pins are multiplexed with analog input pins. You will need to ensure that the QEI pins are configured as digital pins using the ADPCFG control register. Question 2: How fast may my quadrature signals be?
Answer: The answer depends on the setting of the filter parameters for the quadrature signals. QEI requires that quadrature signals frequency must be less than FCY/3 when no filter is used and Filter Frequency/6 when a filter is used. Question: 3 My encoder has a 90 Index Pulse and the count does not reset properly.
Answer: Depending on how the count clock is generated and which quadrature state transition is used for the index pulse, a 1/4 cycle index pulse may not be recognized before the required transition. To fix this, use a filter on the quadrature clocks which has a higher filter prescaler than that of the index pulse. This has the effect of delaying the quadrature clocks somewhat, allowing for proper detection of the index pulse. Figure 16-11: Reset by Index Mode (90 Index Pulse) Up/Down Position Counter
Quadrature State QEA Filter QEB Filter INDX
count_clock POSCNT UPDN Generate QEI Interrupt POSCNT set to 0000 Recognize Index Wheel Reverses Generate QEI Interrupt POSCNT set to MAXCNT Recognize Index 00E4 0000 0001 0002 0003 0002 0001 0000 00E4 00E3 00E2
DS70063C-page 16-21

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Quadrature Encoder Interface (QEI) module are: Title Servo Control of a DC-Brush Motor PIC18CXXX/PIC16CXXX DC Servomotor Using the dsPIC30F for Vector Control of an ACIM Note: Application Note # AN532 AN696 AN908
DS70063C-page 16-22
16
Revision B
This revision provides expanded information for the dsPIC30F Quadrature Encoder Interface (QEI) module.
Revision C
DS70063C-page 16-23

NOTES:
DS70063C-page 16-24
Section 17. 10-bit A/D Converter

HIGHLIGHTS
This section of the manual contains the following major topics: 17.1 Introduction .................................................................................................................. 17-2 17.2 Control Registers ......................................................................................................... 17-4 17.3 A/D Result Buffer ......................................................................................................... 17-4 17.4 A/D Terminology and Conversion Sequence ............................................................. 17-11 17.5 A/D Module Configuration.......................................................................................... 17-13 17.6 Selecting the Voltage Reference Source ................................................................... 17-13 17.7 Selecting the A/D Conversion Clock .......................................................................... 17-13 17.8 Selecting Analog Inputs for Sampling ........................................................................ 17-14 17.9 Enabling the Module .................................................................................................. 17-16 17.10 Specifying the Sample/Conversion Sequence........................................................... 17-16 17.11 How to Start Sampling ............................................................................................... 17-17 17.12 How to Stop Sampling and Start Conversions ........................................................... 17-18 17.13 Controlling Sample/Conversion Operation................................................................. 17-29 17.14 Specifying How Conversion Results are Written Into the Buffer ................................ 17-30 17.15 Conversion Sequence Examples............................................................................... 17-31 17.16 A/D Sampling Requirements...................................................................................... 17-45 17.17 Reading the A/D Result Buffer................................................................................... 17-46 17.18 Transfer Function ....................................................................................................... 17-47 17.19 A/D Accuracy/Error .................................................................................................... 17-47 17.20 Connection Considerations........................................................................................ 17-47 17.21 Initialization ................................................................................................................ 17-48 17.22 A/D Conversion Speeds............................................................................................. 17-49 17.23 Operation During Sleep and Idle Modes.................................................................... 17-55 17.24 Effects of a Reset....................................................................................................... 17-55 17.25 Special Function Registers Associated with the 10-bit A/D Converter ...................... 17-56 17.26 Design Tips ................................................................................................................ 17-57 17.27 Related Application Notes.......................................................................................... 17-58 17.28 Revision History ......................................................................................................... 17-59
17
10-bit A/D Converter
DS70064D-page 17-1

17.1 Introduction
The dsPIC30F 10-bit A/D converter has the following key features: Successive Approximation (SAR) conversion Up to 1 Msps conversion speed Up to 16 analog input pins External voltage reference input pins Four unipolar differential S/H amplifiers Simultaneous sampling of up to four analog input pins Automatic Channel Scan mode Selectable conversion trigger source 16 word conversion result buffer Selectable Buffer Fill modes Four result alignment options Operation during CPU Sleep and Idle modes
A block diagram of the 10-bit A/D is shown in Figure 17-1. The 10-bit A/D converter can have up to 16 analog input pins, designated AN0-AN15. In addition, there are two analog input pins for external voltage reference connections. These voltage reference inputs may be shared with other analog input pins. The actual number of analog input pins and external voltage reference input configuration will depend on the specific dsPIC30F device. Refer to the device data sheet for further details. The analog inputs are connected via multiplexers to four S/H amplifiers, designated CH0-CH3. One, two, or four of the S/H amplifiers may be enabled for acquiring input data. The analog input multiplexers can be switched between two sets of analog inputs during conversions. Unipolar differential conversions are possible on all channels using certain input pins (see Figure 17-1). An Analog Input Scan mode may be enabled for the CH0 S/H amplifier. A Control register specifies which analog input channels will be included in the scanning sequence. The 10-bit A/D is connected to a 16-word result buffer. Each 10-bit result is converted to one of four 16-bit output formats when it is read from the buffer.
DS70064D-page 17-2

Figure 17-1: 10-Bit High-Speed A/D Block Diagram
AVDD VREF+ AVSS VREF-
AN0
AN0 AN3 AN6 AN9 VREF-
+ S/H -
CH1
ADC
17
10-bit Result Conversion Logic CH2
AN1
AN1 AN4 AN7 AN10 VREF-
+ S/H
AN2
AN2 AN5 AN8 AN11 VREF0000 0001 0010 0011 0100 0101 0110 0111 1000 1001 1010 1011 1100 1101 1110 1111 VREFAN1
+ S/H -
CH3 CH1,CH2, CH3,CH0 Sample/Sequence Control
Sample
Input Switches
AN3 AN4 AN5 AN6 AN7 AN8 AN9 AN10 AN11 AN12 AN13 AN14 AN15
Input MUX Control
+ S/H -
CH0
Note: VREF+, VREF- inputs may be shared with other analog inputs. See device data sheet for details.
DS70064D-page 17-3
Bus Interface
16-word, 10-bit Dual Port RAM
Data Format

The A/D module has six Control and Status registers. These registers are: ADCON1: A/D Control Register 1 ADCON2: A/D Control Register 2 ADCON3: A/D Control Register 3 ADCHS: A/D Input Channel Select Register ADPCFG: A/D Port Configuration Register ADCSSL: A/D Input Scan Selection Register
The ADCON1, ADCON2 and ADCON3 registers control the operation of the A/D module. The ADCHS register selects the input pins to be connected to the S/H amplifiers. The ADPCFG register configures the analog input pins as analog inputs or as digital I/O. The ADCSSL register selects inputs to be sequentially scanned.
17.3
A/D Result Buffer

The module contains a 16-word dual port RAM, called ADCBUF, to buffer the A/D results. The 16 buffer locations are referred to as ADCBUF0, ADCBUF1, ADCBUF2, ...., ADCBUFE, ADCBUFF. Note: The A/D result buffer is a read only buffer.
DS70064D-page 17-4

Register 17-1: Upper Byte: R/W-0 ADON bit 15 ADCON1: A/D Control Register 1 U-0 R/W-0 ADSIDL U-0 U-0 U-0 R/W-0 R/W-0 FORM<1:0> bit 8
Lower Byte: R/W-0 R/W-0 SSRC<2:0> bit 7 bit 15
R/W-0
U-0
R/W-0 SIMSAM
R/W-0 ASAM
R/W-0 HC, HS SAMP
R/C-0 HC, HS DONE bit 0
17
ADON: A/D Operating Mode bit 1 = A/D converter module is operating 0 = A/D converter is off Unimplemented: Read as 0 ADSIDL: Stop in Idle Mode bit 1 = Discontinue module operation when device enters Idle mode 0 = Continue module operation in Idle mode FORM<1:0>: Data Output Format bits 11 = Signed Fractional (DOUT = sddd dddd dd00 0000) 10 = Fractional (DOUT = dddd dddd dd00 0000) 01 = Signed Integer (DOUT = ssss sssd dddd dddd) 00 = Integer (DOUT = 0000 00dd dddd dddd) SSRC<2:0>: Conversion Trigger Source Select bits 111 = Internal counter ends sampling and starts conversion (auto convert) 110 = Reserved 101 = Reserved 100 = Reserved 011 = Motor Control PWM interval ends sampling and starts conversion 010 = GP Timer3 compare ends sampling and starts conversion 001 = Active transition on INT0 pin ends sampling and starts conversion 000 = Clearing SAMP bit ends sampling and starts conversion Unimplemented: Read as 0 SIMSAM: Simultaneous Sample Select bit (only applicable when CHPS = 01 or 1x) 1 = Samples CH0, CH1, CH2, CH3 simultaneously (when CHPS = 1x) or Samples CH0 and CH1 simultaneously (when CHPS = 01) 0 = Samples multiple channels individually in sequence ASAM: A/D Sample Auto-Start bit 1 = Sampling begins immediately after last conversion completes. SAMP bit is auto set 0 = Sampling begins when SAMP bit set
bit 14 bit 13
bit 7-5
bit 4 bit 3
bit 2
DS70064D-page 17-5

Register 17-1: bit 1 ADCON1: A/D Control Register 1 (Continued)
SAMP: A/D Sample Enable bit 1 = At least one A/D sample/hold amplifier is sampling 0 = A/D sample/hold amplifiers are holding When ASAM = 0, writing 1 to this bit will start sampling When SSRC = 000, writing 0 to this bit will end sampling and start conversion DONE: A/D Conversion Status bit (Rev. B silicon or later) 1 = A/D conversion is done 0 = A/D conversion is NOT done Cleared by software or start of a new conversion Clearing this bit will not effect any operation in progress Legend: R = Readable bit HC = Hardware clear -n = Value at POR W = Writable bit HS = Hardware set 1 = Bit is set U = Unimplemented bit, read as 0 C = Clearable by software 0 = Bit is cleared x = Bit is unknown
bit 0
DS70064D-page 17-6

Register 17-2: ADCON2: A/D Control Register 2 R/W-0 U-0 reserved U-0 R/W-0 CSCNA R/W-0 R/W-0 CHPS<1:0> bit 8
Upper Byte: R/W-0 R/W-0 VCFG<2:0> bit 15
Lower Byte: R-0 BUFS bit 7
U-0
R/W-0
R/W-0 R/W-0 SMPI<3:0>
R/W-0
R/W-0 BUFM
R/W-0 ALTS bit 0
17
bit 15-13 VCFG<2:0>: Voltage Reference Configuration bits

A/D VREFH
000 001 010 011 1XX
A/D VREFL AVSS AVSS External VREF- pin External VREF- pin AVSS
AVDD External VREF+ pin AVDD External VREF+ pin AVDD
bit 12 bit 11 bit 10
Reserved: User should write 0 to this location Unimplemented: Read as 0 CSCNA: Scan Input Selections for CH0+ S/H Input for MUX A Input Multiplexer Setting bit 1 = Scan inputs 0 = Do not scan inputs CHPS<1:0>: Selects Channels Utilized bits 1x = Converts CH0, CH1, CH2 and CH3 01 = Converts CH0 and CH1 00 = Converts CH0 When SIMSAM bit (ADCON1<3>) = 0 multiple channels sampled sequentially When SMSAM bit (ADCON1<3>) = 1 multiple channels sampled as in CHPS<1:0> BUFS: Buffer Fill Status bit Only valid when BUFM = 1 (ADRES split into 2 x 8-word buffers). 1 = A/D is currently filling buffer 0x8-0xF, user should access data in 0x0-0x7 0 = A/D is currently filling buffer 0x0-0x7, user should access data in 0x8-0xF Unimplemented: Read as 0 SMPI<3:0>: Sample/Convert Sequences Per Interrupt Selection bits 1111 = Interrupts at the completion of conversion for each 16th sample/convert sequence 1110 = Interrupts at the completion of conversion for each 15th sample/convert sequence ..... 0001 = Interrupts at the completion of conversion for each 2nd sample/convert sequence 0000 = Interrupts at the completion of conversion for each sample/convert sequence BUFM: Buffer Mode Select bit 1 = Buffer configured as two 8-word buffers ADCBUF(15...8), ADCBUF(7...0) 0 = Buffer configured as one 16-word buffer ADCBUF(15...0.) ALTS: Alternate Input Sample Mode Select bit 1 = Uses MUX A input multiplexer settings for first sample, then alternate between MUX B and MUX A input multiplexer settings for all subsequent samples 0 = Always use MUX A input multiplexer settings Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 9-8
bit 7
bit 6 bit 5-2
bit 1
bit 0
DS70064D-page 17-7

Register 17-3: Upper Byte: U-0 bit 15 ADCON3: A/D Control Register 3 U-0 U-0 R/W-0 R/W-0 R/W-0 SAMC<4:0> R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 ADRC bit 7
U-0
R/W-0
R/W-0
R/W-0 R/W-0 ADCS<5:0>
R/W-0
R/W-0 bit 0
bit 15-13 Unimplemented: Read as 0 bit 12-8 SAMC<4:0>: Auto-Sample Time bits 11111 = 31 TAD 00001 = 1 TAD 00000 = 0 TAD (only allowed if performing sequential conversions using more than one S/H amplifier) ADRC: A/D Conversion Clock Source bit 1 = A/D internal RC clock 0 = Clock derived from system clock Unimplemented: Read as 0 ADCS<5:0>: A/D Conversion Clock Select bits 111111 = TCY/2 (ADCS<5:0> + 1) = 32 TCY 000001 = TCY/2 (ADCS<5:0> + 1) = TCY 000000 = TCY/2 (ADCS<5:0> + 1) = TCY/2 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7
bit 6 bit 5-0
DS70064D-page 17-8

Register 17-4: ADCHS: A/D Input Select Register R/W-0 CH123SB R/W-0 CH0NB R/W-0 R/W-0 R/W-0 CH0SB<3:0> R/W-0 bit 8
Upper Byte: R/W-0 R/W-0 CH123NB<1:0> bit 15
Lower Byte: R/W-0 R/W-0 CH123NA<1:0> bit 7
R/W-0 CH123SA
R/W-0 CH0NA
R/W-0
R/W-0 R/W-0 CH0SA<3:0>
R/W-0 bit 0
17
bit 15-14 CH123NB<1:0>: Channel 1, 2, 3 Negative Input Select for MUX B Multiplexer Setting bits Same definition as bits 6-7 (Note) bit 13 bit 12 bit 11-8 bit 7-6 CH123SB: Channel 1, 2, 3 Positive Input Select for MUX B Multiplexer Setting bit Same definition as bit 5 (Note) CH0NB: Channel 0 Negative Input Select for MUX B Multiplexer Setting bit Same definition as bit 4 (Note) CH0SB<3:0>: Channel 0 Positive Input Select for MUX B Multiplexer Setting bits Same definition as bits 3-0 (Note) CH123NA<1:0>: Channel 1, 2, 3 Negative Input Select for MUX A Multiplexer Setting bits 11 = CH1 negative input is AN9, CH2 negative input is AN10, CH3 negative input is AN11 10 = CH1 negative input is AN6, CH2 negative input is AN7, CH3 negative input is AN8 0x = CH1, CH2, CH3 negative input is VREFCH123SA: Channel 1, 2, 3 Positive Input Select for MUX A Multiplexer Setting bit 1 = CH1 positive input is AN3, CH2 positive input is AN4, CH3 positive input is AN5 0 = CH1 positive input is AN0, CH2 positive input is AN1, CH3 positive input is AN2 CH0NA: Channel 0 Negative Input Select for MUX A Multiplexer Setting bit 1 = Channel 0 negative input is AN1 0 = Channel 0 negative input is VREFCH0SA<3:0>: Channel 0 Positive Input Select for MUX A Multiplexer Setting bits 1111 = Channel 0 positive input is AN15 1110 = Channel 0 positive input is AN14 1101 = Channel 0 positive input is AN13 || || || 0001 = Channel 0 positive input is AN1 0000 = Channel 0 positive input is AN0 Note: The analog input multiplexer supports two input setting configurations, denoted MUX A and MUX B. ADCHS<15:8> determine the settings for MUX B, and ADCHS<7:0> determine the settings for MUX A. Both sets of control bits function identically. The ADCHS register description and functionality will vary depending on the number of A/D inputs available on the selected device. Please refer to the specific device data sheet for additional details on this register.
bit 5
bit 4
bit 3-0
Note:
DS70064D-page 17-9

Register 17-5: ADPCFG: A/D Port Configuration Register R/W-0 PCFG13 R/W-0 PCFG12 R/W-0 PCFG11 R/W-0 PCFG10 R/W-0 PCFG9 R/W-0 PCFG8 bit 8
Upper Byte: R/W-0 R/W-0 PCFG15 PCFG14 bit 15
Lower Byte: R/W-0 R/W-0 PCFG7 PCFG6 bit 7 bit 15-0
R/W-0 PCFG5
R/W-0 PCFG4
R/W-0 PCFG3
R/W-0 PCFG2
R/W-0 PCFG1
R/W-0 PCFG0 bit 0
PCFG<15:0>: Analog Input Pin Configuration Control bits 1 = Analog input pin in Digital mode, port read input enabled, A/D input multiplexer input connected to AVSS 0 = Analog input pin in Analog mode, port read input disabled, A/D samples pin voltage Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
Register 17-6:
ADCSSL: A/D Input Scan Select Register R/W-0 CSSL13 R/W-0 CSSL12 R/W-0 CSSL11 R/W-0 CSSL10 R/W-0 CSSL9 R/W-0 CSSL8 bit 8
Upper Byte: R/W-0 R/W-0 CSSL15 CSSL14 bit 15
Lower Byte: R/W-0 R/W-0 CSSL7 CSSL6 bit 7 bit 15-0
R/W-0 CSSL5
R/W-0 CSSL4
R/W-0 CSSL3
R/W-0 CSSL2
R/W-0 CSSL1
R/W-0 CSSL0 bit 0
CSSL<15:0>: A/D Input Pin Scan Selection bits 1 = Select ANx for input scan 0 = Skip ANx for input scan Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70064D-page 17-10

17.4 A/D Terminology and Conversion Sequence
Figure 17-2 shows a basic conversion sequence and the terms that are used. A sampling of the analog input pin voltage is performed by sample and hold S/H amplifiers. The S/H amplifiers are also called S/H channels. The 10-bit A/D converter has four total S/H channels, designated CH0-CH3. The S/H channels are connected to the analog input pins via the analog input multiplexer. The analog input multiplexer is controlled by the ADCHS register. There are two sets of multiplexer control bits in the ADCHS register that function identically. These two sets of control bits allow two different analog input multiplexer configurations to be programmed, which are called MUX A and MUX B. The A/D converter can optionally switch between the MUX A and MUX B configurations between conversions. The A/D converter can also optionally scan through a series of analog inputs. Sample time is the time that the A/D modules S/H amplifier is connected to the analog input pin. The sample time may be started manually by setting the SAMP bit (ADCON1<1>) or started automatically by the A/D converter hardware. The sample time is ended manually by clearing the SAMP control bit in the user software or automatically by a conversion trigger source. Conversion time is the time required for the A/D converter to convert the voltage held by the S/H amplifier. The A/D is disconnected from the analog input pin at the end of the sample time. The A/D converter requires one A/D clock cycle (TAD) to convert each bit of the result plus one additional clock cycle. A total of 12 TAD cycles are required to perform the complete conversion. When the conversion time is complete, the result is loaded into one of 16 A/D Result registers (ADCBUF0...ADCBUFF), the S/H can be reconnected to the input pin, and a CPU interrupt may be generated. The sum of the sample time and the A/D conversion time provides the total conversion time. There is a minimum sample time to ensure that the S/H amplifier will give the desired accuracy for the A/D conversion (see Section 17.16 A/D Sampling Requirements). Furthermore, there are multiple input clock options for the A/D converter. The user must select an input clock option that does not violate the minimum TAD specification. Figure 17-2: A/D Sample/Conversion Sequence
A/D Total Conversion Time Sample Time A/D Conversion Time
17
A/D conversion complete, result is loaded into A/D result buffer. Optionally generate interrupt.
S/H amplifier is disconnected from input and holds signal lever. A/D conversion is started by the conversion trigger source. S/H amplifier is connected to the analog input pin for sampling.
The 10-bit A/D converter allows many options for specifying the sample/convert sequence. The sample/convert sequence can be very simple, such as the one shown in Figure 17-3. The example in Figure 17-3 uses only one S/H amplifier. A more elaborate sample/convert sequence performs multiple conversions using more than one S/H amplifier. The 10-bit A/D converter can use two S/H amplifiers to perform two conversions in a sample/convert sequence or four S/H amplifiers with four conversions. The number of S/H amplifiers, or channels per sample, used in the sample/convert sequence is determined by the CHPS control bits.
DS70064D-page 17-11

A sample/convert sequence that uses multiple S/H channels can be simultaneously sampled or sequentially sampled, as controlled by the SIMSAM bit (ADCON1<3>). Simultaneously sampling multiple signals ensures that the snapshot of the analog inputs occurs at precisely the same time for all inputs. Sequential sampling takes a snapshot of each analog input just before conversion starts on that input, and the sampling of multiple inputs is not correlated. Figure 17-3: Simultaneous and Sequential Sampling
AN0 AN1 AN2 AN3
Simultaneous Sampling
Sequential Sampling
The start time for sampling can be controlled in software by setting the SAMP control bit. The start of the sampling time can also be controlled automatically by the hardware. When the A/D converter operates in the Auto-Sample mode, the S/H amplifier(s) is reconnected to the analog input pin at the end of the conversion in the sample/convert sequence. The auto-sample function is controlled by the ASAM control bit (ADCON1<2>). The conversion trigger source ends the sampling time and begins an A/D conversion or a sample/convert sequence. The conversion trigger source is selected by the SSRC control bits. The conversion trigger can be taken from a variety of hardware sources, or can be controlled manually in software by clearing the SAMP control bit. One of the conversion trigger sources is an auto-conversion. The time between auto-conversions is set by a counter and the A/D clock. The Auto-Sample mode and auto-conversion trigger can be used together to provide endless automatic conversions without software intervention. An interrupt may be generated at the end of each sample/convert sequence or multiple sample/convert sequences as determined by the value of the SMPI control bits ADCON2<5:2>. The number of sample/convert sequences between interrupts can vary between 1 and 16. The user should note that the A/D conversion buffer holds 16 results when the SMPI value is selected. The total number of conversion results between interrupts is the product of the channels per sample and the SMPI value. The total number of conversions between interrupts should not exceed the buffer length.
DS70064D-page 17-12

17.5 A/D Module Configuration
The following steps should be followed for performing an A/D conversion: 1. Configure the A/D module Select port pins as analog inputs ADPCFG<15:0> Select voltage reference source to match expected range on analog inputs ADCON2<15:13> Select the analog conversion clock to match desired data rate with processor clock ADCON3<5:0> Determine how many S/H channels will be used ADCON2<9:8> and ADPCFG<15:0> Determine how sampling will occur ADCON1<3> and ADCSSL<15:0> Determine how inputs will be allocated to S/H channels ADCHS<15:0> Select the appropriate sample/conversion sequence ADCON1<7:0> and ADCON3<12:8> Select how conversion results are presented in the buffer ADCON1<9:8> Select interrupt rate ADCON2<5:9> Turn on A/D module ADCON1<15> Configure A/D interrupt (if required) Clear ADIF bit Select A/D interrupt priority
17
2.
The options for each configuration step are described in the subsequent sections.
17.6
Selecting the Voltage Reference Source

The voltage references for A/D conversions are selected using the VCFG<2:0> control bits (ADCON2<15:13>). The upper voltage reference (VREFH) and the lower voltage reference (VREFL) may be the internal AVDD and AVSS voltage rails or the VREF+ and VREF- input pins. The external voltage reference pins may be shared with the AN0 and AN1 inputs on low pin count devices. The A/D converter can still perform conversions on these pins when they are shared with the VREF+ and VREF- input pins. The voltages applied to the external reference pins must meet certain specifications. Refer to the Electrical Specifications section of the device data sheet for further details. Note: External VREF+ amd VREF- must be selected for conversion rates above 500 ksps. See Section 17.22 A/D Conversion Speeds for further details.
17.7
Selecting the A/D Conversion Clock

The A/D converter has a maximum rate at which conversions may be completed. An analog module clock, TAD, controls the conversion timing. The A/D conversion requires 12 clock periods (12 TAD). The A/D clock is derived from the device instruction clock or internal RC clock source. The period of the A/D conversion clock is software selected using a 6-bit counter. There are 64 possible options for TAD, specified by the ADCS<5:0> bits (ADCON3<5:0>). Equation 17-1 gives the TAD value as a function of the ADCS control bits and the device instruction cycle clock period, TCY. Equation 17-1: A/D Conversion Clock Period TAD = TCY (ADCS + 1) 2 ADCS = 2TAD TCY For correct A/D conversions, the A/D conversion clock (TAD) must be selected to ensure a minimum TAD time of 83.33 nsec (see Section 17.22 A/D Conversion Speeds for further details). 1
DS70064D-page 17-13

The A/D converter has a dedicated internal RC clock source that can be used to perform conversions. The internal RC clock source should be used when A/D conversions are performed while the dsPIC30F is in Sleep mode. The internal RC oscillator is selected by setting the ADRC bit (ADCON3<7>). When the ADRC bit is set, the ADCS<5:0> bits have no effect on the A/D operation.
17.8
Selecting Analog Inputs for Sampling

All Sample-and-Hold Amplifiers have analog multiplexers (see Figure 17-1) on both their non-inverting and inverting inputs to select which analog input(s) are sampled. Once the sample/convert sequence is specified, the ADCHS bits determine which analog inputs are selected for each sample. Additionally, the selected inputs may vary on an alternating sample basis or may vary on a repeated sequence of samples. The same analog input can be connected to two or more sample and hold channels to improve conversion rates. Note: Different devices will have different numbers of analog inputs. Verify the analog input availability against the device data sheet.
17.8.1
Configuring Analog Port Pins

The ADPCFG register specifies the input condition of device pins used as analog inputs. A pin is configured as analog input when the corresponding PCFGn bit (ADPCFG<n>) is clear. The ADPCFG register is clear at Reset, causing the A/D input pins to be configured for analog input by default at Reset. When configured for analog input, the associated port I/O digital input buffer is disabled so it does not consume current. The ADPCFG register and the TRISB register control the operation of the A/D port pins. The port pins that are desired as analog inputs must have their corresponding TRIS bit set, specifying port input. If the I/O pin associated with an A/D input is configured as an output, TRIS bit is cleared and the ports digital output level (VOH or VOL) will be converted. After a device Reset, all TRIS bits are set. A pin is configured as digital I/O when the corresponding PCFGn bit (ADPCFG<n>) is set. In this configuration, the input to the analog multiplexer is connected to AVSS. Note 1: When reading the A/D Port register, any pin configured as an analog input reads as a 0. 2: Analog levels on any pin that is defined as a digital input (including the AN15:AN0 pins) may cause the input buffer to consume current that is out of the devices specification.
17.8.2
Channel 0 Input Selection

Channel 0 is the most flexible of the 4 S/H channels in terms of selecting analog inputs. The user may select any of the up to 16 analog inputs as the input to the positive input of the channel. The CH0SA<3:0> bits (ADCHS<3:0>) normally select the analog input for the positive input of channel 0. The user may select either VREF- or AN1 as the negative input of the channel. The CH0NA bit (ADCHS<4>) normally selects the analog input for the negative input of channel 0.
DS70064D-page 17-14

17.8.2.1 Specifying Alternating Channel 0 Input Selections The ALTS bit (ADCON2<0>) causes the module to alternate between two sets of inputs that are selected during successive samples. The inputs specified by CH0SA<3:0>, CH0NA, CHXSA and CHXNA<1:0> are collectively called the MUX A inputs. The inputs specified by CH0SB<3:0>, CH0NB, CHXSB and CHXNB<1:0> are collectively called the MUX B inputs. When the ALTS bit is 1 , the module will alternate between the MUX A inputs on one sample and the MUX B inputs on the subsequent sample. For channel 0, if the ALTS bit is 0, only the inputs specified by CH0SA<3:0> and CH0NA are selected for sampling. If the ALTS bit is 1, on the first sample/convert sequence for channel 0, the inputs specified by CH0SA<3:0> and CH0NA are selected for sampling. On the next sample convert sequence for channel 0, the inputs specified by CH0SB<3:0> and CH0NB are selected for sampling. This pattern will repeat for subsequent sample conversion sequences. Note that if multiple channels (CHPS = 01 or 1x) and simultaneous sampling (SIMSAM = 1) are specified, alternating inputs will change every sample because all channels are sampled on every sample time. If multiple channels (CHPS = 01 or 1x) and sequential sampling (SIMSAM = 0) are specified, alternating inputs will change only on each sample of a particular channel. 17.8.2.2 Scanning Through Several Inputs with Channel 0 Channel 0 has the ability to scan through a selected vector of inputs. The CSCNA bit (ADCON2<10>) enables the CH0 channel inputs to be scanned across a selected number of analog inputs. When CSCNA is set, the CH0SA<3:0> bits are ignored. The ADCSSL register specifies the inputs to be scanned. Each bit in the ADCSSL register corresponds to an analog input. Bit 0 corresponds to AN0, bit 1 corresponds to AN1 and so on. If a particular bit in the ADCSSL register is 1, the corresponding input is part of the scan sequence. The inputs are always scanned from lower to higher numbered inputs, starting at the first selected channel after each interrupt occurs. Note: If the number of scanned inputs selected is greater than the number of samples taken per interrupt, the higher numbered inputs will not be sampled.
17
The ADCSSL bits only specify the input of the positive input of the channel. The CH0NA bit still selects the input of the negative input of the channel during scanning. If the ALTS bit is 1, the scanning only applies to the MUX A input selection. The MUX B input selection, as specified by the CH0SB<3:0>, will still select the alternating channel 0 input. When the input selections are programmed in this manner, the channel 0 input will alternate between a set of scanning inputs specified by the ADCSSL register and a fixed input specified by the CH0SB bits.
DS70064D-page 17-15

17.8.3 Channel 1, 2 and 3 Input Selection
Channel 1, 2 and 3 can sample a subset of the analog input pins. Channel 1, 2 and 3 may select one of two groups of 3 inputs. The CHXSA bit (ADCHS<5>) selects the source for the positive inputs of channel 1, 2 and 3. Clearing CHXSA selects AN0, AN1 and AN2 as the analog source to the positive inputs of channel 1, 2 and 3, respectively. Setting CHXSA selects AN3, AN4 and AN5 as the analog source. The CHXNA<1:0> bits (ADCHS<7:6>) select the source for the negative inputs of channel 1, 2 and 3. Programming CHXNA = 0x, selects VREF- as the analog source for the negative inputs of channel 1, 2 and 3. Programming CHXNA = 10 selects AN6, AN7 and AN8 as the analog source to the negative inputs of channel 1, 2 and 3 respectively. Programming CHXNA = 11 selects AN9, AN10 and AN11 as the analog source. 17.8.3.1 Selecting Multiple Channels for a Single Analog Input The analog input multiplexer can be configured so that the same input pin is connected to two or more sample and hold channels. The A/D converts the value held on one S/H channel, while the second S/H channel acquires a new input sample. 17.8.3.2 Specifying Alternating Channel 1, 2 and 3 Input Selections As with the channel 0 inputs, the ALTS bit (ADCON2<0>) causes the module to alternate between two sets of inputs that are selected during successive samples for channel 1,2 and 3. The MUX A inputs specified by CHXSA and CHXNA<1:0> always select the input when ALTS = 0. The MUX A inputs alternate with the MUX B inputs specified by CHXSB and CHXNB<1:0> when ALTS = 1.
17.9
Enabling the Module

When the ADON bit (ADCON1<15>) is 1, the module is in Active mode and is fully powered and functional. When ADON is 0, the module is disabled. The digital and analog portions of the circuit are turned off for maximum current savings. In order to return to the Active mode from the Off mode, the user must wait for the analog stages to stabilize. For the stabilization time, refer to the Electrical Characteristics section of the device data sheet. Note: The SSRC<2:0>, SIMSAM, ASAM, CHPS<1:0>, SMPI<3:0>, BUFM and ALTS bits, as well as the ADCON3 and ADCSSL registers, should not be written to while ADON = 1. This would lead to indeterminate results.
17.10
Specifying the Sample/Conversion Sequence

The 10-bit A/D module has 4 sample/hold amplifiers and one A/D converter. The module may perform 1, 2 or 4 input samples and A/D conversions per sample/convert sequence.
17.10.1
Number of Sample/Hold Channels

The CHPS<1:0> control bits (ADCON2<9:8>) are used to select how many S/H amplifers are used by the A/D module during sample/conversion sequences. The following three options may be selected: CH0 only CH0 and CH1 CH0, CH1, CH2, CH3 The CHPS control bits work in conjunction with the SIMSAM (simultaneous sample) control bit (ADCON1<3>).
DS70064D-page 17-16

17.10.2 Simultaneous Sampling Enable
Some applications may require that multiple signals are sampled at the exact same time instance. The SIMSAM control bit (ADCON1<3>) works in conjunction with the CHPS control bits and controls the sample/convert sequence for multiple channels as shown in Table 17-1. The SIMSAM control bit has no effect on the module operation if CHPS<1:0> = 00. If more than one S/H amplifier is enabled by the CHPS control bits and the SIMSAM bit is 0, the two or four selected channels are sampled and converted sequentially with two or four sampling periods. If the SIMSAM bit is 1, two or four selected channels are sampled simultaneously with one sampling period. The channels are then converted sequentially. Table 17-1: Sample/Conversion Control Options Sample/Conversion Sequence Sample CH0, Convert CH0 # of Sample/ Convert Cycles to Complete 1 Example
17
Figure 17-4, Figure 17-5, Figure 17-6, Figure 17-7, Figure 17-10, Figure 17-11, Figure 17-14, Figure 17-15
CHPS<1:0> SIMSAM 00 x
01 1x
0 0
Sample CH0, Convert CH0 Sample CH1, Convert CH1 Sample CH0, Convert CH0 Sample CH1, Convert CH1 Sample CH2, Convert CH2 Sample CH3, Convert CH3 Sample CH0, CH1 simultaneously Convert CH0 Convert CH1 Sample CH0, CH1, CH2, CH3 simultaneously Convert CH0 Convert CH1 Convert CH2 Convert CH3
2 4 Figure 17-9, Figure 17-13, Figure 17-20 Figure 17-18
01
1x
Figure 17-8 Figure 17-12, Figure 17-16, Figure 17-17, Figure 17-9,
17.11 17.11.1
How to Start Sampling Manual

Setting the SAMP bit (ADCON1<1>) causes the A/D to begin sampling. One of several options can be used to end sampling and complete the conversions. Sampling will not resume until the SAMP bit is once again set. For an example, see Figure 17-4.
17.11.2
Automatic
Setting the ASAM bit (ADCON1<2>) causes the A/D to automatically begin sampling a channel whenever a conversion is not active on that channel. One of several options can be used to end sampling and complete the conversions. If the SIMSAM bit specifies sequential sampling, sampling on a channel resumes after the conversion of that channel completes. If the SIMSAM bit specifies simultaneous sampling, sampling on a channel resumes after the conversion of all channels completes. For an example, see Figure 17-5.
DS70064D-page 17-17

17.12 How to Stop Sampling and Start Conversions
The conversion trigger source will terminate sampling and start a selected sequence of conversions. The SSRC<2:0> bits (ADCON1<7:5>) select the source of the conversion trigger. Note: The available conversion trigger sources may vary depending on the dsPIC30F device variant. Please refer to the specific device data sheet for the available conversion trigger sources.
Note:
The SSRC selection bits should not be changed when the A/D module is enabled. If the user wishes to change the conversion trigger source, the A/D module should be disabled first by clearing the ADON bit (ADCON1<15>).
17.12.1
Manual
When SSRC<2:0> = 000, the conversion trigger is under software control. Clearing the SAMP bit (ADCON1<1>) starts the conversion sequence. Figure 17-4 is an example where setting the SAMP bit initiates sampling and clearing the SAMP bit terminates sampling and starts conversion. The user software must time the setting and clearing of the SAMP bit to ensure adequate sampling time of the input signal. See Example 17-1 for code example.
Figure 17-4:
Converting 1 Channel, Manual Sample Start, Manual Conversion Start
ADCLK TSAMP TCONV
SAMP DONE ADCBUF0 Instruction Execution BSET ADCON1,SAMP BCLR ADCON1,SAMP
Example 17-1:
Converting 1 Channel, Manual Sample Start, Manual Conversion Start Code

// // // // // all PORTB = Digital; RB2 = analog SAMP bit = 0 ends sampling ... and starts converting Connect RB2/AN2 as CH0 input .. in this example RB2/AN2 is the input
ADPCFG = 0xFFFB; ADCON1 = 0x0000; ADCHS = 0x0002;
ADCSSL = 0; ADCON3 = 0x0002; ADCON2 = 0;
// Manual Sample, Tad = internal 2 Tcy
ADCON1bits.ADON = 1; while (1) { ADCON1bits.SAMP = 1; DelayNmSec(100); ADCON1bits.SAMP = 0; while (!ADCON1bits.DONE); ADCValue = ADCBUF0; }
// turn ADC ON // repeat continuously // // // // // // start sampling ... for 100 mS start Converting conversion done? yes then get ADC value repeat
DS70064D-page 17-18

Figure 17-5 is an example where setting the ASAM bit initiates automatic sampling and clearing the SAMP bit terminates sampling and starts conversion. After the conversion completes, the module will automatically return to a sampling state. The SAMP bit is automatically set at the start of the sample interval. The user software must time the clearing of the SAMP bit to ensure adequate sampling time of the input signal, understanding that the time between clearing of the SAMP bit includes the conversion time as well as the sampling time. See Example 17-2 for code example. Figure 17-5: Converting 1 Channel, Automatic Sample Start, Manual Conversion Start
ADCLK TAD0 SAMP DONE ADCBUF0 BSET ADCON1,ASAM BCLR ADCON1,SAMP BCLR ADCON1,SAMP Instruction Execution TSAMP TCONV TAD0 TSAMP TCONV
17
Example 17-2:
Converting 1 Channel, Automatic Sample Start, Manual Conversion Start Code

// // // // // // all PORTB = Digital but RB7 = analog ASAM bit = 1 implies sampling .. starts immediately after last conversion is done Connect RB7/AN7 as CH0 input .. in this example RB7/AN7 is the input
ADPCFG = 0xFF7F; ADCON1 = 0x0004;
ADCHS
= 0x0007;
ADCSSL = 0; ADCON3 = 0x0002; ADCON2 = 0;
// Sample time manual, Tad = internal 2 Tcy
ADCON1bits.ADON = 1; while (1) { DelayNmSec(100); ADCON1bits.SAMP = 0; while (!ADCON1bits.DONE); ADCValue = ADCBUF0; }
// turn ADC ON // repeat continuously // // // // // sample for 100 mS start Converting conversion done? yes then get ADC value repeat
DS70064D-page 17-19

17.12.2 Clocked Conversion Trigger
When SSRC<2:0> = 111, the conversion trigger is under A/D clock control. The SAMC bits (ADCON3<12:8>) select the number of TAD clock cycles between the start of sampling and the start of conversion. This trigger option provides the fastest conversion rates on multiple channels. After the start of sampling, the module will count a number of TAD clocks specified by the SAMC bits. Equation 17-2: Clocked Conversion Trigger Time
TSMP = SAMC<4:0>*TAD
When using only 1 S/H channel or simultaneous sampling, SAMC must always be programmed for at least one clock cycle. When using multiple S/H channels with sequential sampling, programming SAMC for zero clock cycles will result in the fastest possible conversion rate. See Example 17-3 for code example. Figure 17-6: Converting 1 Channel, Manual Sample Start, TAD Based Conversion Start
ADCLK TSAMP = 16 TAD SAMP DONE ADCBUF0 Instruction Execution BSET ADCON1,SAMP TCONV
Example 17-3:
Converting 1 Channel, Manual Sample Start, TAD Based Conversion Start Code
// // // // // // all PORTB = Digital; RB12 = analog SSRC bit = 111 implies internal counter ends sampling and starts converting. Connect RB12/AN12 as CH0 input .. in this example RB12/AN12 is the input
ADPCFG = 0xEFFF; ADCON1 = 0x00E0;
ADCHS
= 0x000C;
ADCSSL = 0; ADCON3 = 0x1F02; ADCON2 = 0;
// Sample time = 31Tad, Tad = internal 2 Tcy
ADCON1bits.ADON = 1; while (1) { ADCON1bits.SAMP = 1; while (!ADCON1bits.DONE); ADCValue = ADCBUF0; }
// turn ADC ON // repeat continuously // // // // // start sampling then ... after 31Tad go to conversion conversion done? yes then get ADC value repeat
// repeat
DS70064D-page 17-20

17.12.2.1 Free Running Sample Conversion Sequence As shown in Figure 17-7, using the Auto-Convert Conversion Trigger mode (SSRC = 111) in combination with the Auto-Sample Start mode (ASAM = 1), allows the A/D module to schedule sample/conversion sequences with no intervention by the user or other device resources. This Clocked mode allows continuous data collection after module initialization. See Example 17-4 for code example. Note: This A/D configuration must be enabled for the conversion rate of 750 ksps (see Section 17.22 A/D Conversion Speeds for details)
Figure 17-7:
Converting 1 Channel, Auto-Sample Start, TAD Based Conversion Start
17
TCONV
ADCLK TSAMP = 16 TAD SAMP DONE ADCBUF0 ADCBUF1 Instruction Execution BSET ADCON1,ASAM TCONV TSAMP = 16 TAD
Example 17-4:
Converting 1 Channel, Auto-Sample Start, TAD Based Conversion Start Code

ADPCFG = 0xFFFB; ADCON1 = 0x00E0;
ADCHS
= 0x0002;
ADCSSL = 0; ADCON3 = 0x0F00; ADCON2 = 0x0004;
// Sample time = 15Tad, Tad = internal Tcy/2 // Interrupt after every 2 samples
ADCON1bits.ADON = 1; while (1) { ADCValue = 0; ADC16Ptr = &ADCBUF0; IFS0bits.ADIF = 0; ADCON1bits.ASAM = 1;
// turn ADC ON // repeat continuously
// clear value // initialize ADCBUF pointer // clear ADC interrupt flag // auto start sampling // for 31Tad then go to conversion while (!IFS0bits.ADIF); // conversion done? ADCON1bits.ASAM = 0; // yes then stop sample/convert for (count = 0; count < 2; count++) // average the 2 ADC value ADCValue = ADCValue + *ADC16Ptr++; ADCValue = ADCValue >> 1; } // repeat
DS70064D-page 17-21

17.12.2.2 Multiple Channels with Simultaneous Sampling As shown in Figure 17-8 when using simultaneous sampling, the SAMC value specifies the sampling time. In the example, SAMC specifies a sample time of 3 TAD. Because automatic sample start is active, sampling will start on all channels after the last conversion ends and will continue for 3 A/D clocks. See Example 17-5 for code example. Figure 17-8: Converting 4 Channels, Auto-Sample Start, TAD Conversion Start, Simultaneous Sampling
ADCLK TCONV TSAMP ch0_samp ch1_samp ch2_samp ch3_samp ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 DONE SAMP TCONV TCONV TCONV TCONV TCONV
Example 17-5:
Converting 4 Channels, Auto-Sample Start, TAD Conversion Start, Simultaneous Sampling Code
// // // // // // RB0,RB1,RB2 & RB7 = analog SIMSAM bit = 1 implies ... simultaneous sampling ASAM = 1 for auto sample after convert SSRC = 111 for 3Tad sample time Connect AN7 as CH0 input
ADPCFG = 0xFF78; ADCON1 = 0x00EC;
ADCHS
= 0x0007;
ADCSSL = 0; ADCON3 = 0x0302; ADCON2 = 0x030C;
// // // //
Auto Sampling 3 Tad, Tad = internal 2 Tcy CHPS = 1x implies simultaneous ... sample CH0 to CH3 SMPI = 0011 for interrupt after 4 converts
ADCON1bits.ADON = 1; while (1) { ADC16Ptr = &ADCBUF0; OutDataPtr = &OutData[0]; IFS0bits.ADIF = 0; while (IFS0bits.ADIF); for (count = 0; count < 4; { ADCValue = *ADC16Ptr++; LoadADC(ADCValue); } }
// turn ADC ON // repeat continuously // initialize ADCBUF pointer // point to first TXbuffer value // clear interrupt // conversion done? count++) // save the ADC values
// repeat
DS70064D-page 17-22

17.12.2.3 Multiple Channels with Sequential Sampling As shown in Figure 17-9 when using sequential sampling, the sample time precedes each conversion time. In the example, 3 TAD clocks are added for sample time for each channel. Note: This A/D configuration must be enabled for the configuration rates of 1 Msps and 600 ksps (see Section 17.22 A/D Conversion Speeds for further details).
Figure 17-9:
ADCLK
Converting 4 Channels, Auto-Sample Start, TAD Conversion Start, Sequential Sampling
TCONV TSAMP ch0_samp ch1_samp ch2_samp ch3_samp ADRES(0) ADRES(1) ADRES(2) ADRES(3) DONE SAMP =0
TCONV
TCONV
TCONV
TCONV TSAMP
17
17.12.2.4 Sample Time Considerations Using Clocked Conversion Trigger and Automatic Sampling Different sample/conversion sequences provide different available sampling times for the S/H channel to acquire the analog signal. The user must ensure the sampling time exceeds the sampling requirements, as outlined in Section 17.16 A/D Sampling Requirements. Assuming that the module is set for automatic sampling and using a clocked conversion trigger, the sampling interval is determined by the sample interval specified by the SAMC bits. If the SIMSAM bit specifies simultaneous sampling or only one channel is active, the sampling time is the period specified by the SAMC bit. Equation 17-3: Available Sampling Time, Simultaneous Sampling
TSMP = SAMC<4:0> * TAD

If the SIMSAM bit specifies sequential sampling, the total interval used to convert all channels is the number of channels times the sampling time and conversion time. The sampling time for an individual channel is the total interval minus the conversion time for that channel. Equation 17-4: TSEQ TSMP = = Available Sampling Time, Simultaneous Sampling Channels per Sample (CH/S) * ((SAMC<4:0> * TAD) + Conversion Time (TCONV)) (TSEQ TCONV)
Note 1: CH/S specified by CHPS<1:0> bits. 2: TSEQ is the total time for the sample/convert sequence.
DS70064D-page 17-23

17.12.3 Event Trigger Conversion Start
It is often desirable to synchronize the end of sampling and the start of conversion with some other time event. The A/D module may use one of three sources as a conversion trigger. 17.12.3.1 External INT Pin Trigger When SSRC<2:0> = 001, the A/D conversion is triggered by an active transition on the INT0 pin. The INT0 pin may be programmed for either a rising edge input or a falling edge input. 17.12.3.2 GP Timer Compare Trigger The A/D is configured in this Trigger mode by setting SSRC<2:0> = 010. When a match occurs between the 32-bit timer TMR3/TMR2 and the 32-bit Combined Period register PR3/PR2, a special ADC trigger event signal is generated by Timer3. This feature does not exist for the TMR5/TMR4 timer pair. Refer to Section 12. Timers for more details. 17.12.3.3 Motor Control PWM Trigger The PWM module has an event trigger that allows A/D conversions to be synchronized to the PWM time base. When SSRC<2:0> = 011, the A/D sampling and conversion times occur at any user programmable point within the PWM period. The special event trigger allows the user to minimize the delay between the time when A/D conversion results are acquired and the time when the duty cycle value is updated. Refer to Section 15. Motor Control PWM for more details. 17.12.3.4 Synchronizing A/D Operations to Internal or External Events Using the modes where an external event trigger pulse ends sampling and starts conversion (SSRC = 001, 10, 011) may be used in combination with auto-sampling (ASAM = 1) to cause the A/D to synchronize the sample conversion events to the trigger pulse source. For example, in Figure 17-11 where SSRC = 010 and ASAM = 1, the A/D will always end sampling and start conversions synchronously with the timer compare trigger event. The A/D will have a sample conversion rate that corresponds to the timer comparison event rate. See Example 17-6 for code example. Figure 17-10: Converting 1 Channel, Manual Sample Start, Conversion Trigger Based Conversion Start
Conversion Trigger ADCLK TSAMP TCONV
SAMP ADCBUF0 Instruction Execution BSET ADCON1,SAMP
DS70064D-page 17-24

Figure 17-11: Converting 1 Channel, Auto-Sample Start, Conversion Trigger Based Conversion Start
Conversion Trigger ADCLK TSAMP TCONV TSAMP TCONV
SAMP DONE ADCBUF0 ADCBUF1 BSET ADCON1,ASAM Instruction Execution
17
Example 17-6:
Converting 1 Channel, Auto-Sample Start, Conversion Trigger Based Conversion Start Code
// // // // // // all PORTB = Digital; RB2 analog SSRC bit = 010 implies GP TMR3 compare ends sampling and starts converting. Connect RB2/AN2 as CH0 input .. in this example RB2/AN2 is the input
ADPCFG = 0xFFFB; ADCON1 = 0x0040;
ADCHS
= 0x0002;
ADCSSL = 0; ADCON3 = 0x0000; ADCON2 = 0x0004;
// Sample time is TMR3, Tad = internal Tcy/2 // Interrupt after 2 conversions
// set TMR3 to time out every 125 mSecs TMR3 = 0x0000; PR3 = 0x3FFF; T3CON = 0x8010; ADCON1bits.ADON = 1; ADCON1bits.ASAM = 1; while (1) { while (!IFS0bits.ADIF); ADCValue = ADCBUF0; IFS0bits.ADIF = 0; } // turn ADC ON // start auto sampling every 125 mSecs // repeat continuously // // // // conversion done? yes then get first ADC value clear ADIF repeat
DS70064D-page 17-25

17.12.3.5 Multiple Channels with Simultaneous Sampling As shown in Figure 17-12 when using simultaneous sampling, the sampling will start on all channels after setting the ASAM bit or when the last conversion ends. Sampling will stop and conversions will start when the conversion trigger occurs. Figure 17-12: Converting 4 Channels, Auto-Sample Start, Trigger Conversion Start, Simultaneous Sampling
TSEQ Conversion Trigger ADCLK TCONV TSAMP ch0_samp ch1_samp ch2_samp ch3_samp ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 DONE SAMP Cleared in software TCONV TCONV TCONV TSAMP
DS70064D-page 17-26

17.12.3.6 Multiple Channels with Sequential Sampling As shown in Figure 17-13 when using sequential sampling, sampling for a particular channel will stop just prior to converting that channel and will resume after the conversion has stopped. Figure 17-13: Converting 4 Channels, Auto-Sample Start, Trigger Conversion Start, Sequential Sampling
TSEQ Conversion Trigger ADCLK TCONV TCONV TCONV TCONV
17
TSAMP
ch0_samp TSAMP
ch1_samp TSAMP
ch2_samp TSAMP
ch3_samp TSAMP
ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 DONE SAMP Cleared in software
DS70064D-page 17-27

17.12.3.7 Sample Time Considerations for Automatic Sampling/Conversion Sequences Different sample/conversion sequences provide different available sampling times for the S/H channel to acquire the analog signal. The user must ensure the sampling time exceeds the sampling requirements, as outlined in Section 17.16 A/D Sampling Requirements. Assuming that the module is set for automatic sampling and an external trigger pulse is used as the conversion trigger, the sampling interval is a portion of the trigger pulse interval. If the SIMSAM bit specifies simultaneous sampling, the sampling time is the trigger pulse period less the time required to complete the specified conversions. Equation 17-5: TSMP TSMP = = Available Sampling Time, Simultaneous Sampling
Trigger Pulse Interval (TSEQ) Channels per Sample (CH/S) * Conversion Time (TCONV) TSEQ (CH/S * TCONV)
Note 1: CH/S specified by CHPS<1:0> bits. 2: TSEQ is the trigger pulse interval time. If the SIMSAM bit specifies sequential sampling, the sampling time is the trigger pulse period less the time required to complete only one conversion. Equation 17-6: TSMP TSMP = = Available Sampling Time, Sequential Sampling
Trigger Pulse Interval (TSEQ) Conversion Time (TCONV) TSEQ TCONV TSEQ is the trigger pulse interval time.
Note:
DS70064D-page 17-28

17.13 Controlling Sample/Conversion Operation
The application software may poll the SAMP and DONE bits to keep track of the A/D operations or the module can interrupt the CPU when conversions are complete. The application software may also abort A/D operations if necessary.
17.13.1
Monitoring Sample/Conversion Status

The SAMP (ADCON1<1>) and DONE (ADCON1<0>) bits indicate the sampling state and the conversion state of the A/D, respectively. Generally, when the SAMP bit clears, indicating end of sampling, the DONE bit is automatically set, indicating end of conversion. If both SAMP and DONE are 0, the A/D is in an inactive state. In some Operational modes, the SAMP bit may also invoke and terminate sampling.
17
17.13.2
Generating an A/D Interrupt

The SMPI<3:0> bits control the generation of interrupts. The interrupt will occur some number of sample/conversion sequences after starting sampling and re-occur on each equivalent number of samples. Note that the interrupts are specified in terms of samples and not in terms of conversions or data samples in the buffer memory. When the SIMSAM bit specifies sequential sampling, regardless of the number of channels specified by the CHPS bits, the module samples once for each conversion and data sample in the buffer. Therefore, the value specified by the SMPI bits will correspond to the number of data samples in the buffer, up to the maximum of 16. When the SIMSAM bit specifies simultaneous sampling, the number of data samples in the buffer is related to the CHPS bits. Algorithmically, the channels/sample times the number of samples will result in the number of data sample entries in the buffer. To avoid loss of data in the buffer due to overruns, the SMPI bits must be set to the desired buffer size divided by the channels per sample. Disabling the A/D interrupt is not done with the SMPI bits. To disable the interrupt, clear the ADIE analog module interrupt enable bit.
17.13.3
Aborting Sampling
Clearing the SAMP bit while in Manual Sampling mode will terminate sampling, but may also start a conversion if SSRC = 000. Clearing the ASAM bit while in Automatic Sampling mode will not terminate an on going sample/convert sequence, however, sampling will not automatically resume after subsequent conversions.
17.13.4
Aborting a Conversion
Clearing the ADON bit during a conversion will abort the current conversion. The A/D Result register pair will NOT be updated with the partially completed A/D conversion sample. That is, the corresponding ADCBUF buffer location will continue to contain the value of the last completed conversion (or the last value written to the buffer).
DS70064D-page 17-29

17.14 Specifying How Conversion Results are Written Into the Buffer
As conversions are completed, the module writes the results of the conversions into the A/D result buffer. This buffer is a RAM array of sixteen 10-bit words. The buffer is accessed through 16 address locations within the SFR space named ADCBUF0...ADCBUFF. User software may attempt to read each A/D conversion result as it is generated, however, this would consume too much CPU time. Generally, to simplify the code, the module will fill the buffer with results and then generate an interrupt when the buffer is filled.
17.14.1
Number of Conversions per Interrupt

The SMPI<3:0> bits (ADCON2<5:2>) will select how many A/D conversions will take place before the CPU is interrupted. This can vary from 1 sample per interrupt to 16 samples per interrupt. The A/D converter module always starts writing its conversion results at the beginning of the buffer, after each interrupt. For example, if SMPI<3:0> = 0000, the conversion results will always be written to ADCBUF0. In this example, no other buffer locations would be used.
17.14.2
Restrictions Due to Buffer Size

The user cannot program a combination of CHPS and SMPI bits that specifies more than 16 conversions per interrupt when the BUFM bit (ADCON2<1>) is 0, or 8 conversions per interrupt when the BUFM bit (ADCON2<1>) is 0. The BUFM bit function is described below.
17.14.3
Buffer Fill Mode

When the BUFM bit (ADCON2<1>) is 1, the 16-word results buffer (ADRES) will be split into two 8-word groups. The 8-word buffers will alternately receive the conversion results after each interrupt event. The initial 8-word buffer used after BUFM is set will be located at the lower addresses of ADCBUF. When BUFM is 0, the complete 16-word buffer is used for all conversion sequences. The decision to use the BUFM feature will depend upon how much time is available to move the buffer contents after the interrupt, as determined by the application. If the processor can quickly unload a full buffer within the time it takes to sample and convert one channel, the BUFM bit can be 0 and up to 16 conversions may be done per interrupt. The processor will have one sample and conversion time before the first buffer location is overwritten. If the processor cannot unload the buffer within the sample and conversion time, the BUFM bit should be 1. For example, if SMPI<3:0> = 0111, then eight conversions will be loaded into 1/2 of the buffer, following which an interrupt will occur. The next eight conversions will be loaded into the other 1/2 of the buffer. The processor will therefore have the entire time between interrupts to move the eight conversions out of the buffer.
17.14.4
Buffer Fill Status

When the conversion result buffer is split using the BUFM control bit, the BUFS status bit (ADCON2<7>) indicates the half of the buffer that the A/D converter is currently filling. If BUFS = 0, then the A/D converter is filling ADCBUF0-ADCBUF7 and the user software should read conversion values from ADCBUF8-ADCBUFF. If BUFS = 1, the situation is reversed and the user software should read conversion values from ADCBUF0-ADCBUF7.
DS70064D-page 17-30

17.15 Conversion Sequence Examples
The following configuration examples show the A/D operation in different sampling and buffering configurations. In each example, setting the ASAM bit starts automatic sampling. A conversion trigger ends sampling and starts conversion.
17.15.1
Example: Sampling and Converting a Single Channel Multiple Times

Figure 17-11 and Table 17-2 illustrate a basic configuration of the A/D. In this case, one A/D input, AN0, will be sampled by one sample and hold channel, CH0, and converted. The results are stored in the ADCBUF buffer. This process repeats 16 times until the buffer is full and then the module generates an interrupt. The entire process will then repeat. The CHPS bits specify that only sample/hold CH0 is active. With ALTS clear, only the MUX A inputs are active. The CH0SA bits and CH0NA bit are specified (AN0-VREF-) as the input to the sample/hold channel. All other input selection bits are not used.
17
Figure 17-14: Converting One Channel 16 Times/Interrupt
Conversion Trigger
ADCLK
TSAMP TCONV
TSAMP TCONV
TSAMP TCONV
TSAMP TCONV
Input to CH0 ASAM SAMP DONE ADCBUF0 ADCBUF1 ADCBUFE ADCBUFF ADIF BSET ADCON1,ASAM
AN0
AN0
AN0
AN0
Instruction Execution
DS70064D-page 17-31

Table 17-2: Converting One Channel 16 Times/Interrupt OPERATION SEQUENCE Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x0 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x1 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x2 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x3 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x4 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x5 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x6 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x7 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x8 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x9 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xA Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xB Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xC Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xD Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xE Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xF Interrupt Repeat
CONTROL BITS Sequence Select SMPI<2:0> = 1111 Interrupt on 16th sample CHPS<1:0> = 00 Sample Channel CH0 SIMSAM = n/a Not applicable for single channel sample BUFM = 0 Single 16-word result buffer ALTS = 0 Always use MUX A input select MUX A Input Select CH0SA<3:0> = 0000 Select AN0 for CH0+ input CH0NA = 0 Select VREF- for CH0- input CSCNA = 0 No input scan CSSL<15:0> = n/a Scan input select unused CH123SA = n/a Channel CH1, CH2, CH3 + input unused CH123NA<1:0> = n/a Channel CH1, CH2, CH3 input unused MUX B Input Select CH0SB<3:0> = n/a Channel CH0+ input unused CH0NB = n/a Channel CH0- input unused CH123SB = n/a Channel CH1, CH2, CH3 + input unused CH123NB<1:0> = n/a Channel CH1, CH2, CH3 input unused
Buffer Address ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 ADCBUF4 ADCBUF5 ADCBUF6 ADCBUF7 ADCBUF8 ADCBUF9 ADCBUFA ADCBUFB ADCBUFC ADCBUFD ADCBUFE ADCBUFF
Buffer @ 1st Interrupt AN0 sample 1 AN0 sample 2 AN0 sample 3 AN0 sample 4 AN0 sample 5 AN0 sample 6 AN0 sample 7 AN0 sample 8 AN0 sample 9 AN0 sample 10 AN0 sample 11 AN0 sample 12 AN0 sample 13 AN0 sample 14 AN0 sample 15 AN0 sample 16
Buffer @ 2nd Interrupt AN0 sample 17 AN0 sample 18 AN0 sample 19 AN0 sample 20 AN0 sample 21 AN0 sample 22 AN0 sample 23 AN0 sample 24 AN0 sample 25 AN0 sample 26 AN0 sample 27 AN0 sample 28 AN0 sample 29 AN0 sample 30 AN0 sample 31 AN0 sample 32
DS70064D-page 17-32

17.15.2 Example: A/D Conversions While Scanning Through All Analog Inputs
Figure 17-15 and Table 17-3 illustrate a very typical setup where all available analog input channels are sampled by one sample and hold channel, CH0, and converted. The set CSCNA bit specifies scanning of the A/D inputs to the CH0 positive input. Other conditions are similar to Subsection 17.15.1. Initially, the AN0 input is sampled by CH0 and converted. The result is stored in the ADCBUF buffer. Then the AN1 input is sampled and converted. This process of scanning the inputs repeats 16 times until the buffer is full and then the module generates an interrupt. The entire process will then repeat. Figure 17-15: Scanning Through 16 Inputs/Interrupt
17
TSAMP TCONV TCONV TSAMP TCONV
Conversion Trigger ADCLK
TSAMP TCONV
TSAMP
Input to CH0 ASAM SAMP DONE ADCBUF0 ADCBUF1 ADCBUFE ADCBUFF ADIF BSET ADCON1,#ASAM
AN0
AN1
AN14
AN15
DS70064D-page 17-33

Table 17-3: Scanning Through 16 Inputs/Interrupt OPERATION SEQUENCE Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x0 Sample MUX A Inputs: AN1 -> CH0 Convert CH0, Write Buffer 0x1 Sample MUX A Inputs: AN2 -> CH0 Convert CH0, Write Buffer 0x2 Sample MUX A Inputs: AN3 -> CH0 Convert CH0, Write Buffer 0x3 Sample MUX A Inputs: AN4 -> CH0 Convert CH0, Write Buffer 0x4 Sample MUX A Inputs: AN5 -> CH0 Convert CH0, Write Buffer 0x5 Sample MUX A Inputs: AN6 -> CH0 Convert CH0, Write Buffer 0x6 Sample MUX A Inputs: AN7 -> CH0 Convert CH0, Write Buffer 0x7 Sample MUX A Inputs: AN8 -> CH0 Convert CH0, Write Buffer 0x8 Sample MUX A Inputs: AN9 -> CH0 Convert CH0, Write Buffer 0x9 Sample MUX A Inputs: AN10 -> CH0 Convert CH0, Write Buffer 0xA Sample MUX A Inputs: AN11 -> CH0 Convert CH0, Write Buffer 0xB Sample MUX A Inputs: AN12 -> CH0 Convert CH0, Write Buffer 0xC Sample MUX A Inputs: AN13 -> CH0 Convert CH0, Write Buffer 0xD Sample MUX A Inputs: AN14 -> CH0 Convert CH0, Write Buffer 0xE Sample MUX A Inputs: AN15 -> CH0 Convert CH0, Write Buffer 0xF Interrupt Repeat
CONTROL BITS Sequence Select SMPI<2:0> = 1111 Interrupt on 16th sample CHPS<1:0> = 00 Sample Channel CH0 SIMSAM = n/a Not applicable for single channel sample BUFM = 0 Single 16-word result buffer ALTS = 0 Always use MUX A input select MUX A Input Select CH0SA<3:0> = n/a Override by CSCNA CH0NA = 0 Select VREF- for CH0- input CSCNA = 1 Scan CH0+ Inputs CSSL<15:0> = 1111 1111 1111 1111 Scan input select unused CH123SA = n/a Channel CH1, CH2, CH3 + input unused CH123NA<1:0> = n/a Channel CH1, CH2, CH3 input unused MUX B Input Select CH0SB<3:0> = n/a Channel CH0+ input unused CH0NB = n/a Channel CH0- input unused CH123SB = n/a Channel CH1, CH2, CH3 + input unused CH123NB<1:0> = n/a Channel CH1, CH2, CH3 input unused
DS70064D-page 17-34

17.15.3 Example: Sampling Three Inputs Frequently While Scanning Four Other Inputs
Figure 17-16 and Table 17-4 shows how the A/D converter could be configured to sample three inputs frequently using sample/hold channels CH1, CH2 and CH3; while four other inputs are sampled less frequently by scanning them using sample/hold channel CH0. In this case, only MUX A inputs are used, and all 4 channels are sampled simultaneously. Four different inputs (AN4, AN5, AN6, AN7) are scanned in CH0, whereas AN0, AN1 and AN2 are the fixed inputs for CH1, CH2 and CH3, respectively. Thus, in every set of 16 samples, AN0, AN1 and AN2 would be sampled 4 times, while AN4, AN5, AN6 and AN7 would be sampled only once each. Figure 17-16: Converting Three Inputs, Four Times and Four Inputs, One Time/Interrupt
17
Conversion Trigger ADCLK TCONV CONV CONV CONV T T T Input to CH0 Input to CH1 Input to CH2 Input to CH3 ASAM SAMP DONE ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 ADCBUFC ADCBUFD ADCBUFE ADCBUFF ADIF AN4 AN0 AN1 AN2 AN5 AN0 AN1 AN2 TCONV CONV CONV CONV T T T AN6 AN0 AN1 AN2 AN7 AN0 AN1 AN2 TCONV CONV CONV CONV T T T AN4 AN0 AN1 AN2 TSAMP TSAMP TSAMP
DS70064D-page 17-35

Table 17-4: Converting Three Inputs, Four Times and Four Inputs, One Time/Interrupt OPERATION SEQUENCE Sample MUX A Inputs: AN4 -> CH0, AN0 -> CH1, AN1 -> CH2, AN2 -> CH3 Convert CH0, Write Buffer 0x0 Convert CH1, Write Buffer 0x1 Convert CH2, Write Buffer 0x2 Convert CH3, Write Buffer 0x3 Sample MUX A Inputs: AN5 -> CH0, AN0 -> CH1, AN1 -> CH2, AN2 -> CH3 Convert CH0, Write Buffer 0x4 Convert CH1, Write Buffer 0x5 Convert CH2, Write Buffer 0x6 Convert CH3, Write Buffer 0x7 Sample MUX A Inputs: AN6 -> CH0, AN0 -> CH1, AN1 -> CH2, AN2 -> CH3 Convert CH0, Write Buffer 0x8 Convert CH1, Write Buffer 0x9 Convert CH2, Write Buffer 0xA Convert CH3, Write Buffer 0xB Sample MUX A Inputs: AN7 -> CH0, AN0 -> CH1, AN1 -> CH2, AN2 -> CH3 Convert CH0, Write Buffer 0xC Convert CH1, Write Buffer 0xD Convert CH2, Write Buffer 0xE Convert CH3, Write Buffer 0xF Interrupt Repeat
CONTROL BITS Sequence Select SMPI<3:0> = 0011 Interrupt on 16th sample CHPS<1:0> = 1x Sample Channels CH0, CH1, CH2, CH3 SIMSAM = 1 Sample all channels simultaneously BUFM = 0 Single 16-word result buffer ALTS = 0 Always use MUX A input select MUX A Input Select CH0SA<3:0> = n/a Override by CSCNA CH0NA = 0 Select VREF- for CH0- input CSCNA = 1 Scan CH0+ Inputs CSSL<15:0> = 0000 0000 1111 0000 Scan AN4, AN5, AN6, AN7 CH123SA = 0 CH1+ = AN0, CH2+ = AN1, CH3+ = AN2 CH123NA<1:0> = 0x CH1-, CH2-, CH3- = VREFMUX B Input Select CH0SB<3:0> = n/a Channel CH0+ input unused CH0NB = n/a Channel CH0- input unused CH123SB = n/a Channel CH1, CH2, CH3 + input unused CH123NB<1:0> = n/a Channel CH1, CH2, CH3 input unused
DS70064D-page 17-36

17.15.4 Example: Using Dual 8-Word Buffers
Figure 17-17 and Table 17-5 demonstrate using dual 8-word buffers and alternating the buffer fill. Setting the BUFM bit enables dual 8-word buffers. The BUFM setting does not affect other operational parameters. First, the conversion sequence starts filling the buffer at ADCBUF0 (buffer location 0x0). After the first interrupt occurs, the buffer begins to fill at ADCBUF8 (buffer location 0x8). The BUFS status bit is set and cleared alternately after each interrupt. In this example, all four channels are sampled simultaneously, and an interrupt occurs after every sample. Figure 17-17: Converting Four Inputs, One Time/Interrupt Using Dual 8-Word Buffers
17
Conversion Trigger ADCLK TCONVTCONV CONV CONV T T Input to CH0 Input to CH1 Input to CH2 Input to CH3 SAMP BUFS ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 ADCBUF8 ADCBUF9 ADCBUFA ADCBUFB ADIF BSET ADCON1,#ASAM BCLR IFS0,#ADIF BCLR IFS0,#ADIF Instruction Execution AN3 AN0 AN1 AN2 AN3 AN0 AN1 AN2 TCONV CONV CONV CONV T T T AN3 AN0 AN1 AN2 TCONV CONV CONV CONV T T T TSAMP TSAMP TSAMP
DS70064D-page 17-37

Table 17-5: Converting Four Inputs, One Time/Interrupt Using Dual 8-Word Buffers OPERATION SEQUENCE Sample MUX A Inputs: AN3 -> CH0, AN0 -> CH1, AN1 -> CH2, AN2 -> CH3 Convert CH0, Write Buffer 0x0 Convert CH1, Write Buffer 0x1 Convert CH2, Write Buffer 0x2 Convert CH3, Write Buffer 0x3 Interrupt; Change Buffer Sample MUX A Inputs: AN3 -> CH0, AN0 -> CH1, AN1 -> CH2, AN2 -> CH3 Convert CH0, Write Buffer 0x8 Convert CH1, Write Buffer 0x9 Convert CH2, Write Buffer 0xA Convert CH3, Write Buffer 0xB Interrupt; Change Buffer Repeat
CONTROL BITS Sequence Select SMPI<2:0> = 0000 Interrupt on each sample CHPS<1:0> = 1x Sample Channels CH1, CH2, CH3, CH0 SIMSAM = 1 Sample all channels simultaneously BUFM = 1 Dual 8-word result buffers ALTS = 0 Always use MUX A input select MUX A Input Select CH0SA<3:0> = 0011 Select AN3 for CH0+ input CH0NA = 0 Select VREF- for CH0- input CSCNA = 0 No input scan CSSL<15:0> = n/a Scan input select unused CH123SA = 0 CH1+ = AN0, CH2+ = AN1, CH3+ = AN2 CH123NA<1:0> = 0x CH1-, CH2-, CH3- = VREFMUX B Input Select CH0SB<3:0> = n/a Channel CH0+ input unused CH0NB = n/a Channel CH0- input unused CH123SB = n/a Channel CH1, CH2, CH3 + input unused CH123NB<1:0> = n/a Channel CH1, CH2, CH3 input unused
Buffer @ 1st Interrupt AN3 sample 1 AN0 sample 1 AN1 sample 1 AN2 sample 1
Buffer @ 2nd Interrupt
AN3 sample 2 AN0 sample 2 AN1 sample 2 AN2 sample 2
DS70064D-page 17-38

17.15.5 Example: Using Alternating MUX A, MUX B Input Selections
Figure 17-18 and Table 17-6 demonstrate alternate sampling of the inputs assigned to MUX A and MUX B. In this example, 2 channels are enabled to sample simultaneously. Setting the ALTS bit enables alternating input selections. The first sample uses the MUX A inputs specified by the CH0SA, CH0NA, CHXSA and CHXNA bits. The next sample uses the MUX B inputs specified by the CH0SB, CH0NB, CHXSB and CHXNB bits. In this example, one of the MUX B input specifications uses 2 analog inputs as a differential source to the sample/hold, sampling (AN3-AN9). This example also demonstrates use of the dual 8-word buffers. An interrupt occurs after every 4th sample, resulting in filling 8-words into the buffer on each interrupt. Note that using 4 sample/hold channels without alternating input selections results in the same number of conversions as this example, using 2 channels with alternating input selections. However, because the CH1, CH2 and CH3 channels are more limited in the selectivity of the analog inputs, this example method provides more flexibility of input selection than using 4 channels. Figure 17-18: Converting Two Sets of Two Inputs Using Alternating Input Selections
17
TSAMP TCONV CONV T
TSAMP TCONV CONV T AN15 AN3-AN9
TSAMP TCONV CONV T AN1 AN0
Input to CH0 Input to CH1 ASAM SAMP DONE BUFS ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 ADCBUF4 ADCBUF5 ADCBUF6 ADCBUF7 ADCBUF8 ADCBUF9 ADCBUFA ADCBUFB ADIF
AN1 AN0
Cleared in software
Cleared by Software
DS70064D-page 17-39

Table 17-6: Converting Two Sets of Two Inputs Using Alternating Input Selections OPERATION SEQUENCE Sample MUX A Inputs: AN1 -> CH0, AN0 -> CH1 Convert CH0, Write Buffer 0x0 Convert CH1, Write Buffer 0x1 Sample MUX B Inputs: AN15 -> CH0, (AN3-AN9) -> CH1 Convert CH0, Write Buffer 0x2 Convert CH1, Write Buffer 0x3 Sample MUX A Inputs: AN1 -> CH0, AN0 -> CH1 Convert CH0, Write Buffer 0x4 Convert CH1, Write Buffer 0x5 Sample MUX B Inputs: AN15 -> CH0, (AN3-AN9) -> CH1 Convert CH0, Write Buffer 0x6 Convert CH1, Write Buffer 0x7 Interrupt; Change Buffer Sample MUX A Inputs: AN1 -> CH0, AN0 -> CH1 Convert CH0, Write Buffer 0x8 Convert CH1, Write Buffer 0x9 Sample MUX B Inputs: AN15 -> CH0, (AN3-AN9) -> CH1 Convert CH0, Write Buffer 0xA Convert CH1, Write Buffer 0xB Sample MUX A Inputs: AN1 -> CH0, AN0 -> CH1 Convert CH0, Write Buffer 0xC Convert CH1, Write Buffer 0xD Sample MUX B Inputs: AN15 -> CH0, (AN3-AN9) -> CH1 Convert CH0, Write Buffer 0xE Convert CH1, Write Buffer 0xF Interrupt; Change Buffer Repeat
CONTROL BITS Sequence Select SMPI<2:0> = 0011 Interrupt on 4th sample CHPS<1:0> = 01 Sample Channels CH0, CH1 SIMSAM = 1 Sample all channels simultaneously BUFM = 1 Dual 8-word result buffers ALTS = 1 Alternate MUX A/B input select MUX A Input Select CH0SA<3:0> = 0001 Select AN1 for CH0+ input CH0NA = 0 Select VREF- for CH0- input CSCNA = 0 No input scan CSSL<15:0> = n/a Scan input select unused CH123SA = 0 CH1+ = AN0, CH2+ = AN1, CH3+ = AN2 CH123NA<1:0> = 0x CH1-, CH2-, CH3- = VREFMUX B Input Select CH0SB<3:0> = 1111 Select AN15 for CH0+ input CH0NB = 0 Select VREF- for CH0- input CH123SB = 1 CH1+ = AN3, CH2+ = AN4, CH3+ = AN5 CH123NB<1:0> = 11 CH1- = AN9, CH2- = AN10, CH3- = AN11
Buffer @ 1st Interrupt AN1 sample 1 AN0 sample 1 AN15 sample 2 (AN3-AN9) sample 2 AN1 sample 3 AN0 sample 3 AN15 sample 4 (AN3-AN9) sample 4
Buffer @ 2nd Interrupt
AN1 sample 5 AN0 sample 5 AN15 sample 6 (AN3-AN9) sample 6 AN1 sample 7 AN0 sample 7 AN15 sample 8 (AN3-AN9) sample 8
DS70064D-page 17-40

17.15.6 Example: Sampling Eight Inputs Using Simultaneous Sampling
Subsection 17.15.6 and Subsection 17.15.7 demonstrate identical setups with the exception that Subsection 17.15.6 uses simultaneous sampling with SIMSAM = 1 and Subsection 17.15.7 uses sequential sampling with SIMSAM = 0. Both examples use alternating inputs and specify differential inputs to the sample/hold. Figure 17-19 and Table 17-7 demonstrate simultaneous sampling. When converting more than one channel and selecting simultaneous sampling, the module will sample all channels, then perform the required conversions in sequence. In this example, with ASAM set, sampling will begin after the conversions complete. Figure 17-19: Sampling Eight Inputs Using Simultaneous Sampling
Conversion Trigger ADCLK TCONVTCONV CONV CONV T T Input to CH0 Input to CH1 Input to CH2 Input to CH3 ASAM SAMP DONE ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 ADCBUFC ADCBUFD ADCBUFE ADCBUFF ADIF AN13-AN1 AN0 AN1 AN2 AN14 AN3-AN6 AN4-AN7 AN5-AN8 TCONVTCONV CONV CONV T T AN14 AN3-AN6 AN4-AN7 AN5-AN8 TCONV CONV CONV CONV T T T AN13-AN1 AN0 AN1 AN2
17
TSAMP
TSAMP
TSAMP
DS70064D-page 17-41

Table 17-7: Sampling Eight Inputs Using Simultaneous Sampling OPERATION SEQUENCE Sample MUX A Inputs: (AN13-AN1) -> CH0, AN0 -> CH1, AN1 -> CH2, AN2 -> CH3 Convert CH0, Write Buffer 0x0 Convert CH1, Write Buffer 0x1 Convert CH2, Write Buffer 0x2 Convert CH3, Write Buffer 0x3 Sample MUX B Inputs: AN14 -> CH0, (AN3-AN6) -> CH1, (AN4-AN7) -> CH2, (AN5-AN8) -> CH3 Convert CH0, Write Buffer 0x4 Convert CH1, Write Buffer 0x5 Convert CH2, Write Buffer 0x6 Convert CH3, Write Buffer 0x7 Sample MUX A Inputs: (AN13-AN1) -> CH0, AN0 -> CH1, AN1 -> CH2, AN2 -> CH3 Convert CH0, Write Buffer 0x8 Convert CH1, Write Buffer 0x9 Convert CH2, Write Buffer 0xA Convert CH3, Write Buffer 0xB Sample MUX B Inputs: AN14 -> CH0, (AN3-AN6) -> CH1, (AN4-AN7) -> CH2, (AN5-AN8) -> CH3 Convert CH0, Write Buffer 0xC Convert CH1, Write Buffer 0xD Convert CH2, Write Buffer 0xE Convert CH3, Write Buffer 0xF Interrupt Repeat
CONTROL BITS Sequence Select SMPI<2:0> = 0011 Interrupt on 4th sample CHPS<1:0> = 1x Sample Channels CH0, CH1, CH2, CH3 SIMSAM = 1 Sample all channels simultaneously BUFM = 0 Single 16-word result buffer ALTS = 1 Alternate MUX A/MUX B input select MUX A Input Select CH0SA<3:0> = 1101 Select AN13 for CH0+ input CH0NA = 1 Select AN1 for CH0- input CSCNA = 0 No input scan CSSL<15:0> = n/a Scan input select unused CH123SA = 0 CH1+ = AN0, CH2+ = AN1, CH3+ = AN2 CH123NA<1:0> = 0x CH1-, CH2-, CH3- = VREFMUX B Input Select CH0SB<3:0> = 1110 Select AN14 for CH0+ input CH0NB = 0 Select VREF- for CH0- input CH123SB = 1 CH1+ = AN3, CH2+ = AN4, CH3+ = AN5 CH123NB<1:0> = 10 CH1- = AN6, CH2- = AN7, CH3- = AN8
Buffer @ 1st Interrupt (AN13-AN1) sample 1 AN0 sample 1 AN1 sample 1 AN2 sample 1 AN14 sample 2 (AN3-AN6) sample 2 (AN4-AN7) sample 2 (AN5-AN8) sample 2 (AN13-AN1) sample 3 AN0 sample 3 AN1 sample 3 AN2 sample 3 AN14 sample 4 (AN3-AN6) sample 4 (AN4-AN7) sample 4 (AN5-AN8) sample 4
Buffer @ 2nd Interrupt (AN13-AN1) sample 5 AN0 sample 5 AN1 sample 5 AN2 sample 5 AN14 sample 6 (AN3-AN6) sample 6 (AN4-AN7) sample 6 (AN5-AN8) sample 6 (AN13-AN1) sample 7 AN0 sample 7 AN1 sample 7 AN2 sample 7 AN14 sample 8 (AN3-AN6) sample 8 (AN4-AN7) sample 8 (AN5-AN8) sample 8
DS70064D-page 17-42

17.15.7 Example: Sampling Eight Inputs Using Sequential Sampling
Figure 17-20 and Table 17-8 demonstrate sequential sampling. When converting more than one channel and selecting sequential sampling, the module will start sampling a channel at the earliest opportunity, then perform the required conversions in sequence. In this example, with ASAM set, sampling of a channel will begin after the conversion of that channel completes. When ASAM is clear, sampling will not resume after conversion completion but will occur when setting the SAMP bit. When utilizing more than one channel, sequential sampling provides more sampling time since a channel may be sampled while conversion occurs on another. Figure 17-20: Sampling Eight Inputs Using Sequential Sampling
17
TSAMP
TSAMP TCONV CONV CONV CONV T T T
TSAMP TCONV CONV CONV CONV T T T AN13-AN1 AN0 AN1 AN2
TCONV CONV CONV CONV T T T AN14 AN3-AN6 AN4-AN7 AN5-AN8 AN13-AN1 AN0 AN1 AN2
Input to CH0 Input to CH1 Input to CH2 Input to CH3 ASAM SAMP DONE ADCBUF0 ADCBUF1 ADCBUF2 ADCBUF3 ADCBUFC ADCBUFD ADCBUFE ADCBUFF ADIF
AN13-AN1 AN0 AN1 AN2
AN14 AN3-AN6 AN4-AN7 AN5-AN8
DS70064D-page 17-43

Table 17-8: Sampling Eight Inputs Using Sequential Sampling OPERATION SEQUENCE Sample: (AN13-AN1) -> CH0 Convert CH0, Write Buffer 0x0 Sample: AN0 -> CH1 Convert CH1, Write Buffer 0x1 Sample: AN1 -> CH2 Convert CH2, Write Buffer 0x2 Sample: AN2 -> CH3 Convert CH3, Write Buffer 0x3 Sample: AN14 -> CH0 Convert CH0, Write Buffer 0x4 Sample: (AN3-AN6) -> CH1 Convert CH1, Write Buffer 0x5 Sample: (AN4-AN7) -> CH2 Convert CH2, Write Buffer 0x6 Sample: (AN5-AN8) -> CH3 Convert CH3, Write Buffer 0x7 Sample: (AN13-AN1) -> CH0 Convert CH0, Write Buffer 0x8 Sample: AN0 -> CH1 Convert CH1, Write Buffer 0x9 Sample: AN1 -> CH2 Convert CH2, Write Buffer 0xA Sample: AN2 -> CH3 Convert CH3, Write Buffer 0xB Sample: AN14 -> CH0 Convert CH0, Write Buffer 0xC Sample: (AN3-AN6) -> CH1 Convert CH1, Write Buffer 0xD Sample: (AN4-AN7) -> CH2 Convert CH2, Write Buffer 0xE Sample: (AN5-AN8) -> CH3 Convert CH3, Write Buffer 0xF Interrupt Repeat
CONTROL BITS Sequence Select SMPI<2:0> = 1111 Interrupt on 16th sample CHPS<1:0> = 1x Sample Channels CH0, CH1, CH2, CH3 SIMSAM = 0 Sample all channels sequentially BUFM = 0 Single 16-word result buffer ALTS = 1 Alternate MUX A/B input select MUX A Input Select CH0SA<3:0> = 0110 Select AN6 for CH0+ input CH0NA = 0 Select VREF- for CH0- input CSCNA = 0 No input scan CSSL<15:0> = n/a Scan input select unused CH123SA = 0 CH1+ = AN0, CH2+ = AN1, CH3+ = AN2 CH123NA<1:0> = 0x CH1-, CH2-, CH3- = VREFMUX B Input Select CH0SB<3:0> = 0111 Select AN7 for CH0+ input CH0NB = 0 Select VREF- for CH0- input CH123SB = 1 CH1+ = AN3, CH2+ = AN4, CH3+ = AN5 CH123NB<1:0> = 0x CH1-, CH2-, CH3- = VREF-
Buffer @ 1st Interrupt (AN13-AN1) sample 1 AN0 sample 2 AN1 sample 3 AN2 sample 4 AN14 sample 5 (AN3-AN6) sample 6 (AN4-AN7) sample 7 (AN5-AN8) sample 8 (AN13-AN1) sample 9 AN0 sample 10 AN1 sample 11 AN2 sample 12 AN14 sample 13 (AN3-AN6) sample 14 (AN4-AN7) sample 15 (AN5-AN8) sample 16
Buffer @ 2nd Interrupt (AN13-AN1) sample 17 AN0 sample 18 AN1 sample 19 AN2 sample 20 AN14 sample 21 (AN3-AN6) sample 22 (AN4-AN7) sample 23 (AN5-AN8) sample 24 (AN13-AN1) sample 25 AN0 sample 26 AN1 sample 27 AN2 sample 28 AN14 sample 29 (AN3-AN6) sample 30 (AN4-AN7) sample 31 (AN5-AN8) sample 32
DS70064D-page 17-44

17.16 A/D Sampling Requirements
The analog input model of the 10-bit A/D converter is shown in Figure 17-21. The total sampling time for the A/D is a function of the internal amplifier settling time and the holding capacitor charge time. For the A/D converter to meet its specified accuracy, the charge holding capacitor (CHOLD) must be allowed to fully charge to the voltage level on the analog input pin. The analog output source impedance (RS), the interconnect impedance (RIC), and the internal sampling switch (RSS) impedance combine to directly affect the time required to charge the capacitor CHOLD. The combined impedance must therefore be small enough to fully charge the holding capacitor within the chosen sample time. To minimize the effects of pin leakage currents on the accuracy of the A/D converter, the maximum recommended source impedance, RS, is 5 k for the conversion rates of up to 500 ksps and a maximum of 500 for conversion rates of up to 1 Msps. After the analog input channel is selected (changed), this sampling function must be completed prior to starting the conversion. The internal holding capacitor will be in a discharged state prior to each sample operation. At least 1 TAD time period should be allowed between conversions for the sample time. For more details, see the device electrical specifications. Figure 17-21: 10-bit A/D Converter Analog Input Model
VDD VT = 0.6V RIC 250 Sampling Switch RSS CHOLD = DAC capacitance = 4.4 pF VSS Legend: CPIN = input capacitance VT = threshold voltage I leakage = leakage current at the pin due to various junctions RIC = interconnect resistance RSS = sampling switch resistance CHOLD = sample/hold capacitance (from DAC) RSS 3 k
17
Rs
ANx
VA
CPIN VT = 0.6V
I leakage 500 nA
Note: CPIN value depends on device package and is not tested. Effect of CPIN negligible if Rs 5 k.
DS70064D-page 17-45

17.17 Reading the A/D Result Buffer
The RAM is 10-bits wide, but the data is automatically formatted to one of four selectable formats when a read from the buffer is performed. The FORM<1:0> bits (ADCON1<9:8>) select the format. The formatting hardware provides a 16-bit result on the data bus for all of the data formats. Figure 17-22 shows the data output formats that can be selected using the FORM<1:0> control bits. Figure 17-22: A/D Output Data Formats RAM Contents: Read to Bus: Integer 0 0 0 0 0 0 d09 d08 d07 d06 d05 d04 d03 d02 d01 d00 d09 d08 d07 d06 d05 d04 d03 d02 d01 d00
Signed Integer
d09 d09 d09 d09 d09 d09 d09 d08 d07 d06 d05 d04 d03 d02 d01 d00
Fractional (1.15)
d09 d08 d07 d06 d05 d04 d03 d02 d01 d00
Signed Fractional (1.15)
d09 d08 d07 d06 d05 d04 d03 d02 d01 d00
Figure 17-23: Numerical Equivalents of Various Result Codes

VIN/VREF 1023/1024 1022/1024 10-bit Output Code 11 1111 1111 11 1111 1110 16-bit Integer Format 0000 0011 1111 1111 = 1023 0000 0011 1111 1110 = 1022 0000 0010 0000 0001 = 513 0000 0010 0000 0000 = 512 0000 0001 1111 1111 = 511 0000 0000 0000 0001 = 1 0000 0000 0000 0000 = 0 16-bit Signed Integer Format 0000 0001 1111 1111 = 511 0000 0001 1111 1110 = 5 10 513/1024 512/1024 511/1024 10 0000 0001 10 0000 0000 01 1111 1111 0000 0000 0000 0001 = 1 0000 0000 0000 0000 = 0 1111 1111 1111 1111 = -1 1/1024 0/1024 00 0000 0001 00 0000 0000 1111 1110 0000 0001 = -511 1111 1110 0000 0000 = -512 0000 0000 0100 0000 = 0.001 0000 0000 0000 0000 = 0.000 1000 0000 0100 0000 = -0.499 1000 0000 0000 0000 = -0.500 1000 0000 0100 0000 = 0.501 1000 0000 0000 0000 = 0.500 0111 1111 1100 0000 = .499 0 000 0000 0100 0000 = 0.001 0000 0000 0000 0000 = 0.000 1111 1111 1100 0000 = -0.001 16-bit Fractional Format 1111 1111 1100 0000 = 0.999 1111 1111 1000 0000 = 0.998 16-bit Signed Fractional Format 0111 1111 1100 0000 = 0.499 0111 1111 1000 0000 = 0.498
DS70064D-page 17-46

17.18 Transfer Function
The ideal transfer function of the A/D converter is shown in Figure 17-24. The difference of the input voltages, (VINH VINL), is compared to the reference, (VREFH VREFL). The first code transition occurs when the input voltage is (VREFH VREFL/2048) or 0.5 LSb. The 00 0000 0001 code is centered at (VREFH VREFL/1024) or 1.0 LSb. The 10 0000 0000 code is centered at (512*(VREFH VREFL)/1024). An input voltage less than (1*(VREFH VREFL)/2048) converts as 00 0000 0000. An input greater than (2045*(VREFH VREFL)/2048) converts as 11 1111 1111.
Figure 17-24: A/D Transfer Function

Output Code
17
11 1111 1111 (= 1023) 11 1111 1110 (= 1022)
10 0000 0011 (= 515) 10 0000 0010 (= 514) 10 0000 0001 (= 513) 10 0000 0000 (= 512) 01 1111 1111 (= 511) 01 1111 1110 (= 510) 01 1111 1101 (= 509)
00 0000 0001 (= 1) 00 0000 0000 (= 0) VREFL VREFL + VREFH VREFL 1024 VREFL + 512*(VREFH VREFL) 1024 VREFL + 1023*(VREFH VREFL) 1024 (VINH VINL) VREFH
17.19
A/D Accuracy/Error
Refer to Section 17.27 Related Application Notesfor a list of documents that discuss A/D accuracy.
17.20
Connection Considerations
Since the analog inputs employ ESD protection, they have diodes to VDD and VSS. This requires that the analog input must be between VDD and VSS. If the input voltage exceeds this range by greater than 0.3V (either direction), one of the diodes becomes forward biased and it may damage the device if the input current specification is exceeded. An external RC filter is sometimes added for anti-aliasing of the input signal. The R component should be selected to ensure that the sampling time requirements are satisfied. Any external components connected (via high-impedance) to an analog input pin (capacitor, zener diode, etc.) should have very little leakage current at the pin.
DS70064D-page 17-47

17.21 Initialization
Example 17-7 shows a simple initialization code example for the A/D module. In this particular configuration, all 16 analog input pins, AN0-AN15, are set up as analog inputs. Operation in Idle mode is disabled output data is in unsigned fractional format, and AVDD and AVSS are used for VREFH and VREFL. The start of sampling, as well as start of conversion (conversion trigger), are performed manually in software. The CH0 S/H amplifier is used for conversions. Scanning of inputs is disabled, and an interrupt occurs after every sample/convert sequence (1 conversion result). The A/D conversion clock is TCY/2. Since sampling is started manually by setting the SAMP bit (ADCON1<1>) after each conversion is complete, the auto-sample time bits, SAMC<4:0> (ADCON3<12:8>), are ignored. Moreover, since the start of conversion (i.e., end of sampling) is also triggered manually, the SAMP bit needs to be cleared each time a new sample needs to be converted. Example 17-7: A/D Initialization Code Example CLR ADPCFG ; Configure A/D port, ; all input pins are analog MOV #0x2208,W0 MOV W0,ADCON1 ; Configure sample clock source ; and conversion trigger mode. ; Unsigned Fractional format, ; Manual conversion trigger, ; Manual start of sampling, ; Simultaneous sampling, ; No operation in IDLE mode. CLR ADCON2 ; Configure A/D voltage reference ; and buffer fill modes. ; VREF from AVDD and AVSS, ; Inputs are not scanned, ; 1 S/H channel used, ; Interrupt every sample CLR CLR ADCON3 ADCHS ; Configure A/D conversion clock ; Configure input channels, ; CH0+ input is AN0. ; CHO- input is VREFL (AVss) ; No inputs are scanned. ; Clear A/D conversion interrupt
CLR BCLR
ADCSSL IFS0,#ADIF
; Configure A/D interrupt priority bits (ADIP<2:0>) here, if ; required. (default priority level is 4) BSET BSET BSET CALL BCLR : : : IEC0,#ADIE ADCON1,#ADON ADCON1,#SAMP DELAY ADCON1,#SAMP ; Enable A/D conversion interrupt ; ; ; ; ; ; ; ; Turn on A/D Start sampling the input Ensure the correct sampling time has elapsed before starting conversion. End A/D Sampling and start Conversion The DONE bit is set by hardware when the convert sequence is finished The ADIF bit will be set.
DS70064D-page 17-48

17.22 A/D Conversion Speeds
The dsPIC30F 10-bit A/D converter specifications permit a maximum 1 Msps sampling rate. The table below summarizes the conversion speeds for the dsPIC30F 10-bit A/D converter and the required operating conditions. Table 17-9: 10-bit Conversion Rate Parameters dsPIC30F 10-bit A/D Converter Conversion Rates A/D Speed Up to 1 MSps(1) Sampling TAD Minimum Time Min 83.33 ns 12 TAD Rs Max 500 VDD 4.5V to 5.5V Temperature -40C to +85C
ANx S/H CH0 S/H ADC
A/D Channels Configuration

VREF- VREF+
17
CH1, CH2 or CH3
Up to 750 ksps(1)
95.24 ns
2 TAD
500
4.5V to 5.5V
-40C to +85C
VREF- VREF+
ANx S/H
CHX ADC
Up to 600 ksps(1)
138.89 ns
12 TAD
500
3.0V to 5.5V
-40C to +125C
ANx S/H CH0 S/H
VREF- VREF+
CH1, CH2 or CH3
ADC
Up to 500 ksps
153.85 ns
1 TAD
5.0 k
4.5V to 5.5V
-40C to +125C
VREF- VREF+ or or AVSS AVDD
ANx S/H ANx or VREF-
CHX ADC
Up to 300 ksps
256.41 ns
1 TAD
5.0 k
3.0V to 5.5V
-40C to +125C
CHX ADC
Note 1: External VREF- and VREF+ pins must be used for correct operation. See Figure 17-25 for recommended circuit.
DS70064D-page 17-49

The following figure depicts the recommended circuit for the conversion rates above 500 ksps. The dsPIC30F6010 is shown as an example. Figure 17-25: A/D Converter Voltage Reference Schematic
VDD
80 79 78 77 76 75 74 73 72
VSS 69 68 67 66
65 64
VDD
1 2 3 4 5 6 7 8 9 10 VDD VSS VDD 13 14 15 16 17 18 19 VREF+ VREFAVDD AVSS 27 VDD R2 10 C2 0.1 F C1 0.01 F 20 21 22 VDD VSS
63 62 61 60 59 58 57 56 55 54 53 52 VSS 50 49 VDD 47 46 45 44 43 42 41 VDD C5 1 F VDD C4 0.1 F VDD C3 0.01 F VDD C8 1 F C7 0.1 F C6 0.01 F VDD VDD VDD 38 39 40
dsPIC30F6010
28
29
30
33
34
35
36
R1 10
VDD VDD
The configuration procedures below give the required setup values for the conversion speeds above 500 ksps.
17.22.1
1 Msps Configuration Guideline

The configuration for 1 Msps operation is dependent on whether a single input pin is to be sampled or whether multiple pins will be sampled.
17.22.1.1 Single Analog Input For conversions at 1 Msps for a single analog input, at least two sample and hold channels must be enabled. The analog input multiplexer must be configured so that the same input pin is connected to both sample and hold channels. The A/D converts the value held on one S/H channel, while the second S/H channel acquires a new input sample.
DS70064D-page 17-50
37

17.22.1.2 Multiple Analog Inputs The A/D converter can also be used to sample multiple analog inputs using multiple sample and hold channels. In this case, the total 1 Msps conversion rate is divided among the different input signals. For example, four inputs can be sampled at a rate of 250 ksps for each signal or two inputs could be sampled at a rate of 500 ksps for each signal. Sequential sampling must be used in this configuration to allow adequate sampling time on each input. 17.22.1.3 1 Msps Configuration Procedure The following configuration items are required to achieve a 1 Msps conversion rate. Comply with conditions provided in Table 17-9. Connect external VREF+ and VREF- pins following the recommended circuit shown in Figure 17-26. Set SSRC<2:0> = 111 in the ADCON1 register to enable the auto convert option. Enable automatic sampling by setting the ASAM control bit in the ADCON1 register. Enable sequential sampling by clearing the SIMSAM bit in the ADCON1 register. Enable at least two sample and hold channels by writing the CHPS<1:0> control bits in the ADCON2 register. Configure at least 2 conversions between interrupts, since at least two sample and hold channels, by writing the SMPI<3:0> control bits in the ADCON2 register. Configure the A/D clock period to be: 1 12 x 1,000,000 = 83.33 ns
17
by writing to the ADCS<5:0> control bits in the ADCON3 register. Configure the sampling time to be 2 TAD by writing: SAMC<4:0> = 00010. Select at least two channels per analog input pin by writing to the ADCHS register. The following figure shows the timing diagram of the A/D converting one input pin using two sample and holds. The TAD selection, in conjunction with the guidelines described above, allows a conversion speed of 1 Msps. See Example 17-8 for code example. Figure 17-26: Converting 1 Input Pin Using Two Channels at 1Msps, Auto-Sample Start, 12 TAD Sampling Time
ADCLK TSAMP TCONV TCONV TCONV TCONV
ch0_samp ch1_samp
ADRES(0) DONE = 0 SAMP
DS70064D-page 17-51

Example 17-8:
ADPCFG = ADCON1 =
Converting 2 Channels, Auto-Sample Start, TAD Conversion Start, Sequential Sampling Code
// // // // // // // all PORTB = Digital; RB2 = analog SSRC bit = 111 implies internal counter ends sampling and starts converting. Connect RB2/AN2 as CH0 input and also connect to positive CH1 input. in this example RB2/AN2 is the input to two
0xFFFB; 0x00E0;
ADCHS = RB2/AN2
0x0002;
channels. ADCSSL = ADCON3 = ADCON2 = and
0; 0x0C04; 0x6104;
// // //
Sample time = 12Tad = 83.33 ns @ MIPS which will give 1 / (12 * 83.33 ns) = 1 Msps Select external VREF+ and VREF- pins, convert CH0 CH1, Interrupt after every 2 samples turn ADC ON repeat continuously // clear interrupt // conversion done? // save the ADC values repeat
// ADCON1bits.ADON = 1; // while (1) // { IFS0bits.ADIF = 0; while (IFS0bits.ADIF); ADCValue = ADCBUF0; } //
17.22.2
750 ksps Configuration Guideline

The following configuration items are required to achieve a 750 ksps conversion rate. This configuration assumes that a single analog input is to be sampled. Comply with conditions provided in Table 17-9. Connect external VREF+ and VREF- pins following the recommended circuit shown in Figure 17-27. Set SSRC<2:0> = 111 in the ADCON1 register to enable the auto convert option. Enable automatic sampling by setting the ASAM control bit in the ADCON1 register. Enable one sample and hold channel by setting CHPS<1:0> = 00 in the ADCON2 register. Write the SMPI<3:0> control bits in the ADCON2 register for the desired number of conversions between interrupts. Configure the A/D clock period to be: 1 (12 + 2) x 750,000 = 95.24 ns
by writing to the ADCS<5:0> control bits in the ADCON3 register. Configure the sampling time to be 2 TAD by writing: SAMC<4:0> = 00010. The following figure shows the timing diagram of the A/D running at 750 ksps. The TAD selection, in conjunctin with the guidelines described above, allows a conversion speed of 750 ksps. See Example 17-9 for code example.
DS70064D-page 17-52

Figure 17-27: Converting 1 Channel at 750 ksps, Auto-Sample Start, 2 TAD Sampling Time
TSAMP = 2 TAD ADCLK TCONV = 12 TAD TCONV = 12 TAD TSAMP = 2 TAD
SAMP DONE
17
ADCBUF0 ADCBUF1 Instruction Execution BSET ADCON1, ASM
Example 17-9:
ADPCFG = ADCON1 =
Converting 1 Channel at 750 ksps, Auto-Sample Start, 2 TAD Sampling Time Code Example
0xFFFB; 0x00E0; // // // // // // // // // // // // all PORTB = Digital; RB2 = analog SSRC bit = 111 implies internal counter ends sampling and starts converting. Connect RB2/AN2 as CH0 input in this example RB2/AN2 is the input Sample time = 2Tad, Tad = 95.24 ns @ 21 MIPS which will give 1 / (14 * 95.24 ns) = 750 ksps Select external VREF+ and VREF- pins Interrupt after every 2 samples turn ADC ON repeat continuously
ADCHS
0x0002; 0; 0x0203; 0x6004;
ADCSSL = ADCON3 = ADCON2 =
// clear value // initialize ADCBUF pointer // clear ADC interrupt flag // auto start sampling // for 31Tad then go to conversion while (!IFS0bits.ADIF); // conversion done? ADCON1bits.ASAM = 0; // yes then stop sample/convert for (count = 0; count <2; count++) // average the 2 ADC value ADCValue = ADCValue + *ADC16Ptr++; ADCValue = ADCValue >> 1; // repeat
17.22.3

The configuration for 600 ksps operation is dependent on whether a single input pin is to be sampled or whether multiple pins will be sampled.
17.22.3.1 Single Analog Input When performing conversions at 600 ksps for a single analog input, at least two sample and hold channels must be enabled. The analog input multiplexer must be configured so that the same input pin is connected to both sample and hold channels. The A/D converts the value held on one S/H channel, while the second S/H channel acquires a new input sample.
DS70064D-page 17-53

17.22.3.2 Multiple Analog Inputs The A/D converter can also be used to sample multiple analog inputs using multiple sample and hold channels. In this case, the total 600 ksps conversion rate is divided among the different input signals. For example, four inputs can be sampled at a rate of 150 ksps for each signal or two inputs could be sampled at a rate of 300 ksps for each signal. Sequential sampling must be used in this configuration to allow adequate sampling time on each input. 17.22.3.3 600 ksps Configuration Items The following configuration items are required to achieve a 600 ksps conversion rate. Comply with conditions provided in Table 17-9. Connect external VREF+ and VREF- pins following the recommended circuit shown in Figure 17-10. Set SSRC<2:0> = 111 in the ADCON1 register to enable the auto convert option. Enable automatic sampling by setting the ASAM control bit in the ADCON1 register. Enable sequential sampling by clearing the SIMSAM bit in the ADCON1 register. Enable at least two sample and hold channels by writing the CHPS<1:0> control bits in the ADCON2 register. Configure at least 2 conversions between interrupts, since at least two sample and hold channels, by writing the SMPI<3:0> control bits in the ADCON2 register. Configure the A/D clock period to be: 1 12 x 600,000 = 138.89 ns
by writing to the ADCS<5:0> control bits in the ADCON3 register. Configure the sampling time to be 2 TAD by writing: SAMC<4:0> = 00010. Select at least two channels per analog input pin by writing to the ADCHS register. The timing diagram for the 600 ksps extended rate is the same as for the 1 Msps shown in Figure 17-10. See Example 17-10 for code example for 600 ksps A/D operation. Example 17-10: Converting 2 Channels, Auto-Sample Start, TAD Conversion Start, Sequential Samplling Code
ADPCFG = ADCON1 = 0xFFFB; 0x00E0; // // // // // // // channels. ADCSSL = ADCON3 = ADCON2 = and 0; 0x0C04; 0x6104; all PORTB = Digital; RB2 = analog SSRC bit = 111 implies internal counter ends sampling and starts converting. Connect RB2/AN2 as CH0 input and also connect to positive CH1 input. in this example RB2/AN2 is the input to two
ADCHS = RB2/AN2
0x0002;
// // //
Sample time = 12Tad = 138.89 ns @ 18 MIPS which will give 1 / (12 * 138.89 ns) = 600 ksps Select external VREF+ and VREF- pins, convert CH0 CH1, Interrupt after every 2 samples turn ADC ON repeat continuously // clear interrupt // conversion done? // save the ADC values repeat
// ADCON1bits.ADON = 1; // while (1) // { IFS0bits.ADIF = 0; while (IFS0bits.ADIF); ADCValue = ADCBUF0; } //
DS70064D-page 17-54

17.23 Operation During Sleep and Idle Modes
Sleep and Idle modes are useful for minimizing conversion noise because the digital activity of the CPU, buses and other peripherals is minimized.
17.23.1
CPU Sleep Mode without RC A/D Clock

When the device enters Sleep mode, all clock sources to the module are shutdown and stay at logic 0. If Sleep occurs in the middle of a conversion, the conversion is aborted unless the A/D is clocked from its internal RC clock generator. The converter will not resume a partially completed conversion on exiting from Sleep mode. Register contents are not affected by the device entering or leaving Sleep mode.
17
17.23.2
CPU Sleep Mode with RC A/D Clock

The A/D module can operate during Sleep mode if the A/D clock source is set to the internal A/D RC oscillator (ADRC = 1). This eliminates digital switching noise from the conversion. When the conversion is completed, the DONE bit will be set and the result loaded into the A/D result buffer, ADCBUF. If the A/D interrupt is enabled (ADIE = 1), the device will wake-up from Sleep when the A/D interrupt occurs. Program execution will resume at the A/D Interrupt Service Routine if the A/D interrupt is greater than the current CPU priority. Otherwise, execution will continue from the instruction after the PWRSAV instruction that placed the device in Sleep mode. If the A/D interrupt is not enabled, the A/D module will then be turned off, although the ADON bit will remain set. To minimize the effects of digital noise on the A/D module operation, the user should select a conversion trigger source that ensures the A/D conversion will take place in Sleep mode. The automatic conversion trigger option can be used for sampling and conversion in Sleep (SSRC<2:0> = 111). To use the automatic conversion option, the ADON bit should be set in the instruction prior to the PWRSAV instruction. Note: For the A/D module to operate in Sleep, the A/D clock source must be set to RC (ADRC = 1).
17.23.3
A/D Operation During CPU Idle Mode

For the A/D, the ADSIDL bit (ADCON1<13>) selects if the module will stop on Idle or continue on Idle. If ADSIDL = 0, the module will continue normal operation when the device enters Idle mode. If the A/D interrupt is enabled (ADIE = 1), the device will wake up from Idle mode when the A/D interrupt occurs. Program execution will resume at the A/D Interrupt Service Routine if the A/D interrupt is greater than the current CPU priority. Otherwise, execution will continue from the instruction after the PWRSAV instruction that placed the device in Idle mode. If ADSIDL = 1, the module will stop in Idle. If the device enters Idle mode in the middle of a conversion, the conversion is aborted. The converter will not resume a partially completed conversion on exiting from Idle mode.
17.24
Effects of a Reset
A device Reset forces all registers to their Reset state. This forces the A/D module to be turned off, and any conversion in progress is aborted. All pins that are multiplexed with analog inputs will be configured as analog inputs. The corresponding TRIS bits will be set. The values in the ADCBUF registers are not initialized during a Power-on Reset. ADCBUF0...ADCBUFF will contain unknown data.
DS70064D-page 17-55
17.25
Special Function Registers Associated with the 10-bit A/D Converter
The following table lists dsPIC30F 10-bit A/D Converter Special Function registers, including their addresses and formats. All unimplemented registers and/or bits within a register read as zeros.
Table 17-10:
Bit 13 SI2CIF SI2CIE ADIP<2:0> ADC Data Buffer 0 ADC Data Buffer 1 ADC Data Buffer 2 ADC Data Buffer 3 ADC Data Buffer 4 ADC Data Buffer 5 ADC Data Buffer 6 ADC Data Buffer 7 ADC Data Buffer 8 ADC Data Buffer 9 ADC Data Buffer 10 ADC Data Buffer 11 ADC Data Buffer 12 ADC Data Buffer 13 ADC Data Buffer 14 ADC Data Buffer 15 ADSIDL OFFCAL SAMC[4:0] CH0NB CH0SB[3:0] CSCNA CHPS[1:0] CHXSB FORM[1:0] BUFS ADRC SSRC[2:0] CHXNA[1:0] CHXSA CH0NA PCFG4 PCFG3 SIMSAM SMPI[3:0] ADCS[5:0] CH0SA[3:0] PCFG2 PCFG1 ASAM SAMP BUFM CONV ALTS U1TXIP<2:0> U1RXIP<2:0> NVMIE ADIE U1TXIE U1RXIE SPI1IE T3IE T2IE OC2IE IC2IE T1IE OC1IE IC1IE SPI1IP<2:0> NVMIF ADIF U1TXIF U1RXIF SPI1IF T3IF T2IF OC2IF IC2IF T1IF OC1IF IC1IF INT4EP INT3EP INT2EP INT1EP INT0 OVATE OVBTE COVTE MATHERR ADDRERR STKERR OSCFAIL Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
ADC Register Map

Reset States 0000 0000 0000 0000
DS70064D-page 17-56
INT0EP 0000 0000 0000 0000 0000 0000 0000 0000 INT0IE 0000 0000 0000 0000 0100 0100 0100 0100 uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 PCFG0 0000 0000 0000 0000 0000 0000 0000 0000 ADC Input Scan Select Register
File Name
ADR
Bit 15
Bit 14
INTCON1
0080
NSTDIS
INTCON2
0082
ALTIVT
IFS0
0084
CNIF
MI2CIF
IEC0
008C
CNIE
MI2CIE
IPC2
0098
ADCBUF0
0280
ADCBUF1
0282
ADCBUF2
0284
ADCBUF3
0286
ADCBUF4
0288
ADCBUF5
028A
ADCBUF6
028C
ADCBUF7
028E
ADCBUF8
0290
ADCBUF9
0292
ADCBUFA
0294
ADCBUFB
0296
ADCBUFC
0298
ADCBUFD
029A
ADCBUFE
029C
ADCBUFF
029E
ADCON1
02A0
ADON
ADFRZ
ADCON2
02A2
VCFG[2:0]
ADCON3
02A4
ADCHS
02A6
CHXNB[1:0]
ADPCFG
02A8
PCFG15 PCFG14 PCFG13 PCFG12 PCFG11 PCFG10 PCFG9 PCFG8 PCFG7 PCFG6 PCFG5
ADCSSL Legend:
02AA u = unknown
Note:

17.26 Design Tips
Question 1: Answer: 1. Make sure you are meeting all of the timing specifications. If you are turning the module off and on, there is a minimum delay you must wait before taking a sample. If you are changing input channels, there is a minimum delay you must wait for this as well and finally, there is TAD, which is the time selected for each bit conversion. This is selected in ADCON3 and should be within a certain range as specified in the Electrical Characteristics. If TAD is too short, the result may not be fully converted before the conversion is terminated, and if TAD is made too long, the voltage on the sampling capacitor can decay before the conversion is complete. These timing specifications are provided in the Electrical Specifications section of the device data sheets. Often the source impedance of the analog signal is high (greater than 10 k), so the current drawn from the source to charge the sample capacitor can affect accuracy. If the input signal does not change too quickly, try putting a 0.1 F capacitor on the analog input. This capacitor will charge to the analog voltage being sampled and supply the instantaneous current needed to charge the 4.4 pF internal holding capacitor. Put the device into Sleep mode before the start of the A/D conversion. The RC clock source selection is required for conversions in Sleep mode. This technique increases accuracy because digital noise from the CPU and other peripherals is minimized. Do you know of a good reference on A/Ds? How can I optimize the system performance of the A/D converter?
17
2.
3.
Question 2:
Answer: A good reference for understanding A/D conversions is the Analog-Digital Conversion Handbook third edition, published by Prentice Hall (ISBN 0-13-03-2848-0). Question 3: My combination of channels/sample and samples/interrupt is greater than the size of the buffer. What will happen to the buffer?
Answer: This configuration is not recommended. The buffer will contain unknown results.
DS70064D-page 17-57

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the 10-bit A/D Converter module are: Title Using the Analog-to-Digital (A/D) Converter Four Channel Digital Voltmeter with Display and Keyboard Understanding A/D Converter Performance Specifications Using the dsPIC30F for Sensorless BLDC Control Using the dsPIC30F for Vector Control of an ACIM Sensored BLDC Motor Control Using the dsPIC30F2010 An Introduction to AC Induction Motor Control Using the dsPIC30F MCU Note: Application Note # AN546 AN557 AN693 AN901 AN908 AN957 AN984
DS70064D-page 17-58

Revision B
To reflect editorial and technical content revisions for the dsPIC30F 10-bit A/D Converter module.
Revision C
Revision D
This revision includes the extended conversion rate guidelines.
17
DS70064D-page 17-59

NOTES:
DS70064D-page 17-60

HIGHLIGHTS
This section of the manual contains the following major topics: 18.1 Introduction .................................................................................................................. 18-2 18.2 Control Registers ......................................................................................................... 18-4 18.3 A/D Result Buffer ......................................................................................................... 18-4 18.4 A/D Terminology and Conversion Sequence ............................................................. 18-10 18.5 A/D Module Configuration.......................................................................................... 18-11 18.6 Selecting the Voltage Reference Source ................................................................... 18-11 18.7 Selecting the A/D Conversion Clock .......................................................................... 18-12 18.8 Selecting Analog Inputs for Sampling ........................................................................ 18-12 18.9 Enabling the Module .................................................................................................. 18-14 18.10 How to Start Sampling ............................................................................................... 18-14 18.11 How to Stop Sampling and Start Conversions ........................................................... 18-14 18.12 Controlling Sample/Conversion Operation................................................................. 18-19 18.13 Specifying How Conversion Results are Written into the Buffer ................................ 18-19 18.14 Conversion Sequence Examples............................................................................... 18-21 18.15 A/D Sampling Requirements...................................................................................... 18-26 18.16 Reading the A/D Result Buffer................................................................................... 18-27 18.17 Transfer Function ....................................................................................................... 18-28 18.18 A/D Accuracy/Error .................................................................................................... 18-28 18.19 Connection Considerations........................................................................................ 18-28 18.20 Initialization ................................................................................................................ 18-29 18.21 A/D Conversion Speeds............................................................................................. 18-30 18.22 Operation During Sleep and Idle Modes.................................................................... 18-33 18.23 Effects of a Reset....................................................................................................... 18-33 18.24 Special Function Registers Associated with the 12-bit A/D Converter ...................... 18-34 18.25 Design Tips ................................................................................................................ 18-35 18.26 Related Application Notes.......................................................................................... 18-36 18.27 Revision History ......................................................................................................... 18-37
18
DS70065D-page 18-1

18.1 Introduction
The dsPIC30F 12-bit A/D converter has the following key features: Successive Approximation Register (SAR) conversion Up to 200 ksps conversion speed Up to 16 analog input pins External voltage reference input pins Unipolar differential S/H amplifier Automatic Channel Scan mode Selectable conversion trigger source 16-word conversion result buffer Selectable Buffer Fill modes Four result alignment options Operation during CPU Sleep and Idle modes
A block diagram of the 12-bit A/D is shown in Figure 18-1. The 12-bit A/D converter can have up to 16 analog input pins, designated AN0-AN15. In addition, there are two analog input pins for external voltage reference connections. These voltage reference inputs may be shared with other analog input pins. The actual number of analog input pins and external voltage reference input configuration will depend on the specific dsPIC30F device. Refer to the dsPIC30F device data sheets (DS70082 and DS70083) for further details. The analog inputs are connected via multiplexers to the S/H amplifier, designated CH0. The analog input multiplexer can be switched between two sets of analog inputs during conversions. Unipolar differential conversions are possible using certain input pins (see Figure 18-1). An Analog Input Scan mode may be enabled for the CH0 S/H amplifier. A Control register specifies which analog input channels will be included in the scanning sequence. The 12-bit A/D is connected to a 16-word result buffer. Each 12-bit result is converted to one of four 16-bit output formats when it is read from the buffer.
DS70065D-page 18-2

Figure 18-1: 12-bit High Speed A/D Block Diagram
AVDD AVSS VREF+ VREF0000 0001 0010 0011 0100 0101 0110 0111 1000 1001 1010 1011 1100 1101 1110 1111 VREFAN1 S/H CH0 Input Switches Sample Sample/Sequence Control 16-word, 12-bit Dual Port RAM 12-bit SAR Conversion Logic DAC Comparator
AN0 AN1 AN2 AN3 AN4 AN5 AN6 AN7 AN8 AN9 AN10 AN11 AN12 AN13 AN14 AN15
Data Format
Bus Interface
18
Input MUX Control
DS70065D-page 18-3

The A/D module has six Control and Status registers. These registers are: ADCON1: A/D Control Register 1 ADCON2: A/D Control Register 2 ADCON3: A/D Control Register 3 ADCHS: A/D Input Channel Select Register ADPCFG: A/D Port Configuration Register ADCSSL: A/D Input Scan Selection Register
The ADCON1, ADCON2 and ADCON3 registers control the operation of the A/D module. The ADCHS register selects the input pins to be connected to the S/H amplifiers. The ADPCFG register configures the analog input pins as analog inputs or as digital I/O. The ADCSSL register selects inputs to be sequentially scanned.
18.3
A/D Result Buffer

The module contains a 16-word dual port RAM, called ADCBUF, to buffer the A/D results. The 16 buffer locations are referred to as ADCBUF0, ADCBUF1, ADCBUF2, ..., ADCBUFE, ADCBUFF. Note: The A/D result buffer is a read only buffer.
DS70065D-page 18-4

Register 18-1: Upper Byte: R/W-0 ADON bit 15 ADCON1: A/D Control Register 1 U-0 R/W-0 ADSIDL U-0 U-0 U-0 R/W-0 R/W-0 FORM<1:0> bit 8
Lower Byte: R/W-0 R/W-0 SSRC<2:0> bit 7 bit 15
R/W-0
U-0
U-0
R/W-0 ASAM
R/W-0 HC, HS SAMP
R/C-0 HC, HS DONE bit 0
ADON: A/D Operating Mode bit 1 = A/D converter module is operating 0 = A/D converter is off Unimplemented: Read as 0 ADSIDL: Stop in Idle Mode bit 1 = Discontinue module operation when device enters Idle mode 0 = Continue module operation in Idle mode FORM<1:0>: Data Output Format bits 11 = Signed fractional (DOUT = sddd dddd dddd 0000) 10 = Fractional (DOUT = dddd dddd dddd 0000) 01 = Signed integer (DOUT = ssss sddd dddd dddd) 00 = Integer (DOUT = 0000 dddd dddd dddd) SSRC<2:0>: Conversion Trigger Source Select bits 111 = Internal counter ends sampling and starts conversion (auto convert) 110 = Reserved 101 = Reserved 100 = Reserved 011 = Motor Control PWM interval ends sampling and starts conversion 010 = General purpose Timer3 compare ends sampling and starts conversion 001 = Active transition on INT0 pin ends sampling and starts conversion 000 = Clearing SAMP bit ends sampling and starts conversion Unimplemented: Read as 0 ASAM: A/D Sample Auto-Start bit 1 = Sampling begins immediately after last conversion completes. SAMP bit is auto set. 0 = Sampling begins when SAMP bit set SAMP: A/D Sample Enable bit 1 = At least one A/D sample/hold amplifier is sampling 0 = A/D sample/hold amplifiers are holding When ASAM = 0, writing 1 to this bit will start sampling. When SSRC = 000, writing 0 to this bit will end sampling and start conversion. DONE: A/D Conversion Status bit 1 = A/D conversion is done 0 = A/D conversion is not done Clearing this bit will not effect any operation in progress. Cleared by software or start of a new conversion. Legend: R = Readable bit HC = Hardware clear -n = Value at POR W = Writable bit HS = Hardware set 1 = Bit is set C = Clearable by software U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 14 bit 13
18
bit 7-5
bit 4-3 bit 2
bit 1
bit 0
DS70065D-page 18-5

Register 18-2: ADCON2: A/D Control Register 2 R/W-0 U-0 U-0 R/W-0 CSCNA U-0 U-0 bit 8
Upper Byte: R/W-0 R/W-0 VCFG<2:0> bit 15
Lower Byte: R-0 BUFS bit 7
U-0
R/W-0
R/W-0 R/W-0 SMPI<3:0>
R/W-0
R/W-0 BUFM
R/W-0 ALTS bit 0
bit 15-13 VCFG<2:0>: Voltage Reference Configuration bits

A/D VREFH
000 001 010 011 1xx
A/D VREFL AVSS AVSS External VREF- pin External VREF- pin AVSS
AVDD External VREF+ pin AVDD External VREF+ pin AVDD
Reserved: User should write 0 to this location Unimplemented: Read as 0 CSCNA: Scan Input Selections for CH0+ S/H Input for MUX A Input Multiplexer Setting bit 1 = Scan inputs 0 = Do not scan inputs Unimplemented: Read as 0 BUFS: Buffer Fill Status bit Only valid when BUFM = 1 (ADRES split into 2 x 8-word buffers) 1 = A/D is currently filling buffer 0x8-0xF, user should access data in 0x0-0x7 0 = A/D is currently filling buffer 0x0-0x7, user should access data in 0x8-0xF Unimplemented: Read as 0 SMPI<3:0>: Sample/Convert Sequences Per Interrupt Selection bits 1111 = Interrupts at the completion of conversion for each 16th sample/convert sequence 1110 = Interrupts at the completion of conversion for each 15th sample/convert sequence ..... 0001 = Interrupts at the completion of conversion for each 2nd sample/convert sequence 0000 = Interrupts at the completion of conversion for each sample/convert sequence BUFM: Buffer Mode Select bit 1 = Buffer configured as two 8-word buffers ADCBUF(15...8), ADCBUF(7...0) 0 = Buffer configured as one 16-word buffer ADCBUF(15...0) ALTS: Alternate Input Sample Mode Select bit 1 = Uses MUX A input multiplexer settings for first sample, then alternate between MUX B and MUX A input multiplexer settings for all subsequent samples 0 = Always use MUX A input multiplexer settings Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 9-8 bit 7
bit 6 bit 5-2
bit 1
bit 0
DS70065D-page 18-6

Register 18-3: Upper Byte: U-0 bit 15 ADCON3: A/D Control Register 3 U-0 U-0 R/W-0 R/W-0 R/W-0 SAMC<4:0> R/W-0 R/W-0 bit 8 Lower Byte: R/W-0 ADRC bit 7
U-0
R/W-0
R/W-0
R/W-0 R/W-0 ADCS<5:0>
R/W-0
R/W-0 bit 0
bit 15-13 Unimplemented: Read as 0 bit 12-8 SAMC<4:0>: Auto Sample Time bits 11111 = 31 TAD 00001 = 1 TAD 00000 = 0 TAD ADRC: A/D Conversion Clock Source bit 1 = A/D internal RC clock 0 = Clock derived from system clock Unimplemented: Read as 0 ADCS<5:0>: A/D Conversion Clock Select bits 111111 = TCY/2 (ADCS<5:0> + 1) = 32 TCY 000001 = TCY/2 (ADCS<5:0> + 1) = TCY 000000 = TCY/2 (ADCS<5:0> + 1) = TCY/2 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 7
bit 6 bit 5-0
18
DS70065D-page 18-7

Register 18-4: Upper Byte: U-0 bit 15 ADCHS: A/D Input Select Register U-0 U-0 R/W-0 CH0NB R/W-0 R/W-0 R/W-0 CH0SB<3:0> R/W-0 bit 8 Lower Byte: U-0 bit 7
U-0
U-0
R/W-0 CH0NA
R/W-0
R/W-0 R/W-0 CH0SA<3:0>
R/W-0 bit 0
bit 15-13 Unimplemented: Read as 0 bit 12 bit 11-8 bit 7-5 bit 4 CH0NB: Channel 0 Negative Input Select for MUX B Multiplexer Setting bit Same definition as bit <4> (see Note). CH0SB<3:0>: Channel 0 Positive Input Select for MUX B Multiplexer Setting bit Same definition as bits <3:0> (see Note). Unimplemented: Read as 0 CH0NA: Channel 0 Negative Input Select for MUX A Multiplexer Setting bit 1 = Channel 0 negative input is AN1 0 = Channel 0 negative input is VREFCH0SA<3:0>: Channel 0 Positive Input Select for MUX A Multiplexer Setting bit 1111 = Channel 0 positive input is AN15 1110 = Channel 0 positive input is AN14 1101 = Channel 0 positive input is AN13 0001 = Channel 0 positive input is AN1 0000 = Channel 0 positive input is AN0 Note: The analog input multiplexer supports two input setting configurations, denoted MUX A and MUX B. ADCHS<15:8> determines the settings for MUX B, and ADCHS<7:0> determines the settings for MUX A. Both sets of control bits function identically.
bit 3-0
DS70065D-page 18-8

Register 18-5: ADPCFG: A/D Port Configuration Register R/W-0 PCFG13 R/W-0 PCFG12 R/W-0 PCFG11 R/W-0 PCFG10 R/W-0 PCFG9 R/W-0 PCFG8 bit 8
Upper Byte: R/W-0 R/W-0 PCFG15 PCFG14 bit 15
Lower Byte: R/W-0 R/W-0 PCFG7 PCFG6 bit 7 bit 15-0
R/W-0 PCFG5
R/W-0 PCFG4
R/W-0 PCFG3
R/W-0 PCFG2
R/W-0 PCFG1
R/W-0 PCFG0 bit 0
PCFG<15:0>: Analog Input Pin Configuration Control bits 1 = Analog input pin in Digital mode, port read input enabled, A/D input multiplexer input connected to AVSS 0 = Analog input pin in Analog mode, port read input disabled, A/D samples pin voltage Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
Register 18-6:
ADCSSL: A/D Input Scan Select Register R/W-0 CSSL13 R/W-0 CSSL12 R/W-0 CSSL11 R/W-0 CSSL10 R/W-0 CSSL9 R/W-0 CSSL8 bit 8
18
Upper Byte: R/W-0 R/W-0 CSSL15 CSSL14 bit 15
Lower Byte: R/W-0 R/W-0 CSSL7 CSSL6 bit 7 bit 15-0
R/W-0 CSSL5
R/W-0 CSSL4
R/W-0 CSSL3
R/W-0 CSSL2
R/W-0 CSSL1
R/W-0 CSSL0 bit 0
CSSL<15:0>: A/D Input Pin Scan Selection bits 1 = Select ANx for input scan 0 = Skip ANx for input scan Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70065D-page 18-9

18.4 A/D Terminology and Conversion Sequence
Figure 18-2 shows a basic conversion sequence and the terms that are used. A sampling of the analog input pin voltage is performed by sample and hold S/H amplifiers. The S/H amplifiers are also called S/H channels. The 12-bit A/D converter has one S/H channel, designated CH0. The S/H channel is connected to the analog input pins via the analog input multiplexer. The analog input multiplexer is controlled by the ADCHS register. There are two sets of multiplexer control bits in the ADCHS register that function identically. These two sets of control bits allow two different analog input multiplexer configurations to be programmed, which are called MUX A and MUX B. The A/D converter can optionally switch between the MUX A and MUX B configurations between conversions. The A/D converter can also optionally scan through a series of analog inputs. Sample time is the time that the A/D modules S/H amplifier is connected to the analog input pin. The sample time may be started manually by setting the SAMP bit (ADCON1<1>) or started automatically by the A/D converter hardware. The sample time is ended manually by clearing the SAMP control bit in the user software or automatically by a conversion trigger source. Conversion time is the time required for the A/D converter to convert the voltage held by the S/H amplifier. The A/D is disconnected from the analog input pin at the end of the sample time. The A/D converter requires one A/D clock cycle (TAD) to convert each bit of the result plus one additional clock cycle. A total of 14 TAD cycles are required to perform the complete conversion. When the conversion time is complete, the result is loaded into one of 16 A/D result registers (ADCBUF0...ADCBUFF), the S/H can be reconnected to the input pin, and a CPU interrupt may be generated. The sum of the sample time and the A/D conversion time provides the total conversion time. There is a minimum sample time to ensure that the S/H amplifier will give the desired accuracy for the A/D conversion (see Section 18.15 A/D Sampling Requirements). Furthermore, there are multiple input clock options for the A/D converter. The user must select an input clock option that does not violate the minimum TAD specification. Figure 18-2: A/D Sample/Conversion Sequence
A/D Total Conversion Time Sample Time A/D Conversion Time
A/D conversion complete, result is loaded into A/D result buffer. Optionally generate interrupt.
S/H amplifier is disconnected from input and holds signal lever. A/D conversion is started by the conversion trigger source. S/H amplifier is connected to the analog input pin for sampling.
DS70065D-page 18-10

The start time for sampling can be controlled in software by setting the SAMP control bit. The start of the sampling time can also be controlled automatically by the hardware. When the A/D converter operates in the Auto Sample mode, the S/H amplifier(s) is reconnected to the analog input pin at the end of the conversion in the sample/convert sequence. The auto sample function is controlled by the ASAM control bit. The conversion trigger source ends the sampling time and begins an A/D conversion or a sample/convert sequence. The conversion trigger source is selected by the SSRC control bits. The conversion trigger can be taken from a variety of hardware sources or can be controlled manually in software by clearing the SAMP control bit. One of the conversion trigger sources is an auto conversion. The time between auto conversions is set by a counter and the A/D clock. The Auto Sample mode and auto conversion trigger can be used together to provide endless automatic conversions without software intervention. An interrupt may be generated at the end of each sample/convert sequence or multiple sample/convert sequences, as determined by the value of the SMPI control bits. The number of sample/convert sequences between interrupts can vary between 1 and 16.
18.5
A/D Module Configuration

The following steps should be followed for performing an A/D conversion: 1. Configure the A/D module Select voltage reference source to match expected range on analog inputs Select the analog conversion clock to match desired data rate with processor clock Determine how sampling will occur Determine how inputs will be allocated to the S/H channel Select how conversion results are presented in the buffer Select interrupt rate Turn on A/D module Configure A/D interrupt (if required) Clear ADIF bit Select A/D interrupt priority
18
2.
The options for each configuration step are described in the subsequent sections. Note: The SSRC<2:0>, SIMSAM, ASAM, CHPS<1:0>, SMPI<3:0>, BUFM and ALTS bits, as well as the ADCON3 and ADCSSL registers, should not be written to while ADON = 1. This would lead to indeterminate results.
18.6
Selecting the Voltage Reference Source

The voltage references for A/D conversions are selected using the VCFG<2:0> control bits (ADCON2<15:13>). The upper voltage reference (VREFH) and the lower voltage reference (VREFL) may be the internal AVDD and AVSS voltage rails or the VREF+ and VREF- input pins. The external voltage reference pins may be shared with the AN0 and AN1 inputs on low pin count devices. The A/D converter can still perform conversions on these pins when they are shared with the VREF+ and VREF- input pins. The voltages applied to the external reference pins must meet certain specifications. Refer to the Electrical Specifications section of the device data sheet for further details.. Note: External VREF+ and VREF- pins must be selected for the conversion rates above 100 ksps. See Section 18.21 A/D Conversion Speeds for further details.
DS70065D-page 18-11

18.7 Selecting the A/D Conversion Clock
The A/D converter has a maximum rate at which conversions may be completed. An analog module clock, TAD, controls the conversion timing. The A/D conversion requires 14 clock periods (14 TAD). The A/D clock is derived from the device instruction clock. The period of the A/D conversion clock is software selected using a six-bit counter. There are 64 possible options for TAD, specified by the ADCS<5:0> bits (ADCON3<5:0>). Equation 18-1 gives the TAD value as a function of the ADCS control bits and the device instruction cycle clock period, TCY. Equation 18-1: A/D Conversion Clock Period TAD = TCY (ADCS + 1) 2 ADCS = 2TAD TCY For correct A/D conversions, the A/D conversion clock (TAD) must be selected to ensure a minimum TAD time of 333.33 nsec (see Section 18.21 A/D Conversion Speeds for further details). The A/D converter has a dedicated internal RC clock source that can be used to perform conversions. The internal RC clock source should be used when A/D conversions are performed while the dsPIC30F is in Sleep mode. The internal RC oscillator is selected by setting the ADRC bit (ADCON3<7>). When the ADRC bit is set, the ADCS<5:0> bits have no effect on the A/D operation. 1
18.8
Selecting Analog Inputs for Sampling

The Sample-and-Hold Amplifier has analog multiplexers (see Figure 18-1) on both its non-inverting and inverting inputs, to select which analog input(s) are sampled. Once the sample/convert sequence is specified, the ADCHS bits determine which analog inputs are selected for each sample. Additionally, the selected inputs may vary on an alternating sample basis, or may vary on a repeated sequence of samples. Note: Different devices will have different numbers of analog inputs. Verify the analog input availability against the device data sheet.
18.8.1
Configuring Analog Port Pins

The ADPCFG register specifies the input condition of device pins used as analog inputs. A pin is configured as analog input when the corresponding PCFGn bit (ADPCFG<n>) is clear. The ADPCFG register is clear at Reset, causing the A/D input pins to be configured for analog input by default at Reset. When configured for analog input, the associated port I/O digital input buffer is disabled so it does not consume current. The ADPCFG register and the TRISB register control the operation of the A/D port pins. The port pins that are desired as analog inputs must have their corresponding TRIS bit set, specifying port input. If the I/O pin associated with an A/D input is configured as an output, TRIS bit is cleared, the pin is in Analog mode (ADPCFG<n> = 0) and the port digital output level (VOH or VOL) will be converted. After a device Reset, all TRIS bits are set. A pin is configured as digital I/O when the corresponding PCFGn bit (ADPCFG<n>) is set. In this configuration, the input to the analog multiplexer is connected to AVSS. Note 1: When reading a port register, any pin configured as an analog input reads as a 0. 2: Analog levels on any pin that is defined as a digital input (including the AN15:AN0 pins) may cause the input buffer to consume current that is out of the devices specification.
DS70065D-page 18-12

18.8.2 Channel 0 Input Selection
The user may select any one of the up to 16 analog inputs to connect to the positive input of the channel. The CH0SA<3:0> bits (ADCHS<3:0>) normally select the analog input for the positive input of channel 0. The user may select either VREF- or AN1 as the negative input of the channel. The CH0NA bit (ADCHS<4>) normally selects the analog input for the negative input of channel 0. 18.8.2.1 Specifying Alternating Channel 0 Input Selections The ALTS bit (ADCON2<0>) causes the module to alternate between two sets of inputs that are selected during successive samples. The inputs specified by CH0SA<3:0>, CH0NA, CHXSA and CHXNA<1:0> are collectively called the MUX A inputs. The inputs specified by CH0SB<3:0>, CH0NB, CHXSB and CHXNB<1:0> are collectively called the MUX B inputs. When the ALTS bit is 1, the module will alternate between the MUX A inputs on one sample and the MUX B inputs on the subsequent sample. For channel 0, if the ALTS bit is 0, only the inputs specified by CH0SA<3:0> and CH0NA are selected for sampling. If the ALTS bit is 1 on the first sample/convert sequence for channel 0, the inputs specified by CH0SA<3:0> and CH0NA are selected for sampling. On the next sample convert sequence for channel 0, the inputs specified by CH0SB<3:0> and CH0NB are selected for sampling. This pattern will repeat for subsequent sample conversion sequences. 18.8.2.2 Scanning Through Several Inputs Channel 0 has the ability to scan through a selected vector of inputs. The CSCNA bit (ADCON2<10>) enables the CH0 channel inputs to be scanned across a selected number of analog inputs. When CSCNA is set, the CH0SA<3:0> bits are ignored. The ADCSSL register specifies the inputs to be scanned. Each bit in the ADCSSL register corresponds to an analog input. Bit 0 corresponds to AN0, bit 1 corresponds to AN1 and so on. If a particular bit in the ADCSSL register is 1, the corresponding input is part of the scan sequence. The inputs are always scanned from lower to higher numbered inputs, starting at the first selected channel after each interrupt occurs. Note: If the number of scanned inputs selected is greater than the number of samples taken per interrupt, the higher numbered inputs will not be sampled.
18
The ADCSSL bits only specify the input of the positive input of the channel. The CH0NA bit still selects the input of the negative input of the channel during scanning. If the ALTS bit is 1, the scanning only applies to the MUX A input selection. The MUX B input selection, as specified by the CH0SB<3:0>, will still select the alternating input. When the input selections are programmed in this manner, the input will alternate between a set of scanning inputs specified by the ADCSSL register and a fixed input specified by the CH0SB bits.
DS70065D-page 18-13

18.9 Enabling the Module
When the ADON bit (ADCON1<15>) is 1, the module is in Active mode and is fully powered and functional. When ADON is 0, the module is disabled. The digital and analog portions of the circuit are turned off for maximum current savings. In order to return to the Active mode from the Off mode, the user must wait for the analog stages to stabilize. For the stabilization time, refer to the Electrical Characteristics section of the device data sheet.
18.10 18.10.1
How to Start Sampling Manual

Setting the SAMP bit (ADCON1<1>) causes the A/D to begin sampling. One of several options can be used to end sampling and complete the conversions. Sampling will not resume until the SAMP bit is once again set. For an example, see Figure 18-3.
18.10.2
Automatic
Setting the ASAM bit (ADCON1<2>) causes the A/D to automatically begin sampling a channel whenever a conversion is not active on that channel. One of several options can be used to end sampling and complete the conversions. Sampling on a channel resumes after the conversion of that channel completes. For an example, see Figure 18-4.
18.11
How to Stop Sampling and Start Conversions

The conversion trigger source will terminate sampling and start a selected sequence of conversions. The SSRC<2:0> bits (ADCON1<7:5>) select the source of the conversion trigger. Note: The available conversion trigger sources may vary depending on the dsPIC30F device variant. Please refer to the specific device data sheet for the available conversion trigger sources. The SSRC selection bits should not be changed when the A/D module is enabled. If the user wishes to change the conversion trigger source, the A/D module should be disabled first by clearing the ADON bit (ADCON1<15>).
Note:
18.11.1
Manual
When SSRC<2:0> = 000, the conversion trigger is under software control. Clearing the SAMP bit (ADCON1<1>) starts the conversion sequence. Figure 18-3 is an example where setting the SAMP bit initiates sampling and clearing the SAMP bit, terminates sampling and starts conversion. The user software must time the setting and clearing of the SAMP bit to ensure adequate sampling time of the input signal.
Figure 18-3:
Converting 1 Channel, Manual Sample Start, Manual Conversion Start
ADCLK TSAMP TCONV
SAMP DONE ADCBUF0 Instruction Execution BSET ADCON1,SAMP BCLR ADCON1,SAMP
DS70065D-page 18-14
Example 18-1:
Converting 1 Channel, Manual Sample Start, Manual Conversion Start Code Example
// // // // // all PORTB = Digital; RB2 = analog SAMP bit = 0 ends sampling ... and starts converting Connect RB2/AN2 as CH0 input .. in this example RB2/AN2 is the input
ADPCFG = 0xFFFB; ADCON1 = 0x0000; ADCHS = 0x0002;
ADCSSL = 0; ADCON3 = 0x0002; ADCON2 = 0; ADCON1bits.ADON = 1; while (1) { ADCON1bits.SAMP = 1; DelayNmSec(100); ADCON1bits.SAMP = 0; while (!ADCON1bits.DONE); ADCValue = ADCBUF0; }
// Manual Sample, Tad = internal 2 Tcy
// turn ADC ON // repeat continuously // // // // // // start sampling ... for 100 mS start Converting conversion done? yes then get ADC value repeat
Figure 18-4 is an example where setting the ASAM bit initiates automatic sampling and clearing the SAMP bit, terminates sampling and starts conversion. After the conversion completes, the module will automatically return to a sampling state. The SAMP bit is automatically set at the start of the sample interval. The user software must time the clearing of the SAMP bit to ensure adequate sampling time of the input signal, understanding that the time between clearing of the SAMP bit includes the conversion time, as well as the sampling time. Figure 18-4:
ADCLK TSAMP TAD0 SAMP ADCBUF0 BSET ADCON1,ASAM BCF ADCON1,SAMP BCLR ADCON1,SAMP Instruction Execution TCONV TAD0 TSAMP TCONV
18
Converting 1 Channel, Automatic Sample Start, Manual Conversion Start
DS70065D-page 18-15

18.11.2 Clocked Conversion Trigger
When SSRC<2:0> = 111, the conversion trigger is under A/D clock control. The SAMC bits (ADCON3<12:8>) select the number of TAD clock cycles between the start of sampling and the start of conversion. After the start of sampling, the module will count a number of TAD clocks specified by the SAMC bits. Equation 18-2: Clocked Conversion Trigger Time
TSMP = SAMC<4:0>*TAD
SAMC must always be programmed for at least 1 clock cycle to ensure sampling requirements are met. Figure 18-5 shows how to use the clocked conversion trigger with the sampling started by the user software. Figure 18-5: Converting 1 Channel, Manual Sample Start, TAD Based Conversion Start
ADCLK TSAMP TCONV
SAMP DONE ADCBUF0 Instruction Execution BSET ADCON1,SAMP
Example 18-2:
Converting 1 Channel, Manual Sample Start, TAD Based Conversion Start Code Example // // // // // // all PORTB = Digital; RB12 = analog SSRC bit = 111 implies internal counter ends sampling and starts converting. Connect RB12/AN12 as CH0 input .. in this example RB12/AN12 is the input
ADPCFG = 0xEFFF; ADCON1 = 0x00E0;
ADCHS = 0x000C; ADCSSL = 0; ADCON3 = 0x1F02; Tcy ADCON2 = 0; ADCON1bits.ADON = 1; while (1) { ADCON1bits.SAMP = 1;
// Sample time = 31Tad, Tad = internal 2
// turn ADC ON // repeat continuously start sampling then ... after 31Tad go to conversion conversion done? yes then get ADC value repeat// repeat
// // while (!ADCON1bits.DONE);// ADCValue = ADCBUF0; // } //
DS70065D-page 18-16

18.11.2.1 Free Running Sample Conversion Sequence As shown in Figure 18-6, using the Auto-Convert Conversion Trigger mode (SSRC = 111) in combination with the Auto-Sample Start mode (ASAM = 1) allows the A/D module to schedule sample/conversion sequences with no intervention by the user or other device resources. This Clocked mode allows continuous data collection after module initialization.. Note: This A/D configuration must be enabled for the conversion rate of 200 ksps (see Section 18.21 A/D Conversion Speeds for details).
Figure 18-6:
Converting 1 Channel, Auto-Sample Start, TAD Based Conversion Start
ADCLK TSAMP TCONV TSAMP TCONV
SAMP DONE ADCBUF0 ADCBUF1 Instruction Execution BSET ADCON1,ASAM Reset by software
18
18.11.2.2 Sample Time Considerations Using Clocked Conversion Trigger and Automatic Sampling The user must ensure the sampling time exceeds the sampling requirements as outlined in Section 18.15 A/D Sampling Requirements. Assuming that the module is set for automatic sampling and using a clocked conversion trigger, the sampling interval is specified by the SAMC bits.
18.11.3
Event Trigger Conversion Start

It is often desirable to synchronize the end of sampling and the start of conversion with some other time event. The A/D module may use one of three sources as a conversion trigger event.
18.11.3.1 External INT Pin Trigger When SSRC<2:0> = 001, the A/D conversion is triggered by an active transition on the INT0 pin. The INT0 pin may be programmed for either a rising edge input or a falling edge input. 18.11.3.2 General Purpose Timer Compare Trigger The A/D is configured in this Trigger mode by setting SSRC<2:0> = 010. When a match occurs between the 32-bit timer TMR3/TMR2 and the 32-bit Combined Period register PR3/PR2, a special ADC trigger event signal is generated by Timer3. This feature does not exist for the TMR5/TMR4 timer pair. Refer to Section 12. Timers for more details. 18.11.3.3 Motor Control PWM Trigger The PWM module has an event trigger that allows A/D conversions to be synchronized to the PWM time base. When SSRC<2:0> = 011, the A/D sampling and conversion times occur at any user programmable point within the PWM period. The special event trigger allows the user to minimize the delay between the time when A/D conversion results are acquired and the time when the duty cycle value is updated. Refer to Section 15. Motor Control PWM for more details.
DS70065D-page 18-17

18.11.3.4 Synchronizing A/D Operations to Internal or External Events The modes where an external event trigger pulse ends sampling and starts conversion (SSRC = 001, 010, 011) may be used in combination with auto sampling (ASAM = 1) to cause the A/D to synchronize the sample conversion events to the trigger pulse source. For example, in Figure 18-8 where SSRC = 010 and ASAM = 1, the A/D will always end sampling and start conversions synchronously with the timer compare trigger event. The A/D will have a sample conversion rate that corresponds to the timer comparison event rate. Figure 18-7: Manual Sample Start, Conversion Trigger Based Conversion Start
Conversion Trigger ADCLK TSAMP TCONV
SAMP ADCBUF0 Instruction Execution BSET ADCON1,SAMP
Figure 18-8:
Auto-Sample Start, Conversion Trigger Based Conversion Start
Conversion Trigger ADCLK TSAMP TCONV TSAMP TCONV
SAMP DONE ADCBUF0 ADCBUF1 BSET ADCON1,ASAM Instruction Execution
Reset by software
18.11.3.5 Sample Time Considerations for Automatic Sampling/Conversion Sequences Different sample/conversion sequences provide different available sampling times for the S/H channel to acquire the analog signal. The user must ensure the sampling time exceeds the sampling requirements, as outlined in Section 18.15 A/D Sampling Requirements. Assuming that the module is set for automatic sampling and an external trigger pulse is used as the conversion trigger, the sampling interval is a portion of the trigger pulse interval. The sampling time is the trigger pulse period, less the time required to complete the conversion. Equation 18-3: TSMP TSMP = = Available Sampling Time, Sequential Sampling
Trigger Pulse Interval (TSEQ) Conversion Time (TCONV) TSEQ TCONV
Note: TSEQ is the trigger pulse interval time.
DS70065D-page 18-18

18.12 Controlling Sample/Conversion Operation
The application software may poll the SAMP and CONV bits to keep track of the A/D operations, or the module can interrupt the CPU when conversions are complete. The application software may also abort A/D operations if necessary.
18.12.1
Monitoring Sample/Conversion Status

The SAMP (ADCON1<1>) and CONV (ADCON1<0>) bits indicate the sampling state and the conversion state of the A/D, respectively. Generally, when the SAMP bit clears indicating end of sampling, the CONV bit is automatically set indicating start of conversion. If both SAMP and CONV are 0, the A/D is in an inactive state. In some operational modes, the SAMP bit may also invoke and terminate sampling and the CONV bit may terminate conversion.
18.12.2
Generating an A/D Interrupt

The SMPI<3:0> bits control the generation of interrupts. The interrupt will occur some number of sample/conversion sequences after starting sampling and re-occur on each equivalent number of samples. The value specified by the SMPI bits will correspond to the number of data samples in the buffer, up to the maximum of 16. Disabling the A/D interrupt is not done with the SMPI bits. To disable the interrupt, clear the ADIE analog module interrupt enable bit.
18.12.3
Aborting Sampling
Clearing the SAMP bit while in Manual Sampling mode will terminate sampling, but may also start a conversion if SSRC = 000. Clearing the ASAM bit while in Automatic Sampling mode will not terminate an on going sample/convert sequence, however, sampling will not automatically resume after a subsequent conversion.
18
18.12.4
Aborting a Conversion
Clearing the ADON bit during a conversion will abort the current conversion. The A/D Result register pair will NOT be updated with the partially completed A/D conversion sample. That is, the corresponding ADCBUF buffer location will continue to contain the value of the last completed conversion (or the last value written to the buffer).
18.13
Specifying How Conversion Results are Written into the Buffer

As conversions are completed, the module writes the results of the conversions into the A/D result buffer. This buffer is a RAM array of sixteen 12-bit words. The buffer is accessed through 16 address locations within the SFR space, named ADCBUF0...ADCBUFF. User software may attempt to read each A/D conversion result as it is generated, however, this might consume too much CPU time. Generally, to simplify the code, the module will fill the buffer with results and then generate an interrupt when the buffer is filled.
DS70065D-page 18-19

18.13.1 Number of Conversions per Interrupt
The SMPI<3:0> bits (ADCON2<5:2>) will select how many A/D conversions will take place before the CPU is interrupted. This can vary from 1 sample per interrupt to 16 samples per interrupt. The A/D converter module always starts writing its conversion results at the beginning of the buffer, after each interrupt. For example, if SMPI<3:0> = 0000, the conversion results will always be written to ADCBUF0. In this example, no other buffer locations would be used.
18.13.2
Restrictions Due to Buffer Size

The user cannot program the SMPI bits to a value that specifies more than 8 conversions per interrupt when the BUFM bit (ADCON2<1>) is 1. The BUFM bit function is described below.
18.13.3
Buffer Fill Mode

When the BUFM bit (ADCON2<1>) is 1, the 16-word results buffer (ADRES) will be split into two 8-word groups. The 8-word buffers will alternately receive the conversion results after each interrupt event. The initial 8-word buffer used after BUFM is set will be located at the lower addresses of ADCBUF. When BUFM is 0, the complete 16-word buffer is used for all conversion sequences. The decision to use the BUFM feature will depend upon how much time is available to move the buffer contents after the interrupt, as determined by the application. If the processor can quickly unload a full buffer within the time it takes to sample and convert one channel, the BUFM bit can be 0 and up to 16 conversions may be done per interrupt. The processor will have one sample and conversion time before the first buffer location is overwritten. If the processor cannot unload the buffer within the sample and conversion time, the BUFM bit should be 1. For example, if SMPI<3:0> = 0111, then eight conversions will be loaded into 1/2 of the buffer, following which an interrupt will occur. The next eight conversions will be loaded into the other 1/2 of the buffer. The processor will, therefore, have the entire time between interrupts to move the eight conversions out of the buffer.
18.13.4
Buffer Fill Status

When the conversion result buffer is split using the BUFM control bit, the BUFS status bit (ADCON2<7>) indicates the half of the buffer that the A/D converter is currently filling. If BUFS = 0, then the A/D converter is filling ADCBUF0-ADCBUF7 and the user software should read conversion values from ADCBUF8-ADCBUFF. If BUFS = 1, the situation is reversed, and the user software should read conversion values from ADCBUF0-ADCBUF7.
DS70065D-page 18-20

18.14 Conversion Sequence Examples
The following configuration examples show the A/D operation in different sampling and buffering configurations. In each example, setting the ASAM bit starts automatic sampling. A conversion trigger ends sampling and starts conversion.
18.14.1
Example: Sampling and Converting a Single Channel Multiple Times

Figure 18-9 and Table 18-1 illustrate a basic configuration of the A/D. In this case, one A/D input, AN0, will be sampled and converted. The results are stored in the ADCBUF buffer. This process repeats 16 times until the buffer is full and then the module generates an interrupt. The entire process will then repeat. With ALTS clear, only the MUX A inputs are active. The CH0SA bits and CH0NA bit are specified (AN0-VREF-) as the input to the sample/hold channel. All other input selection bits are not used.
Figure 18-9:
Conversion Trigger
Converting One Channel 16 Times/Interrupt
TSAMP ADCLK TCONV
TSAMP TCONV
TSAMP TCONV
TSAMP TCONV
Input to CH0 ASAM SAMP DONE
AN0
AN0
AN0
AN0
18
ADCBUF0 ADCBUF1 ADCBUFE ADCBUFF ADIF BSET ADCON1,ASAM Instruction Execution
DS70065D-page 18-21

Example 18-3: Sampling and Converting a Single Channel Multiple Times Code Example
ADPCFG = 0xFFFB; ADCON1 = 0x00E0;
ADCHS = 0x0002; ADCSSL = 0; ADCON3 = 0x0F00; ADCON2 = 0x003C; ADCON1bits.ADON = 1; while (1) { ADCValue = 0; ADC16Ptr = &ADCBUF0; IFS0bits.ADIF = 0; ADCON1bits.ASAM = 1;
// Sample time = 15Tad, Tad = internal Tcy/2 // Interrupt after every 16 samples // turn ADC ON // repeat continuously
// clear value // initialize ADCBUF pointer // clear ADC interrupt flag // auto start sampling // for 31Tad then go to conversion while (!IFS0bits.ADIF); // conversion done? ADCON1bits.ASAM = 0; // yes then stop sample/convert for (count = 0; count < 16; count++) // average the 16 ADC value ADCValue = ADCValue + *ADC16Ptr++; ADCValue = ADCValue >> 4; } // repeat
DS70065D-page 18-22

Table 18-1: Converting One Channel 16 Times/Interrupt OPERATION SEQUENCE Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x0 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x1 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x2 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x3 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x4 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x5 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x6 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x7 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x8 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x9 Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xA Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xB Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xC Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xD Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xE Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0xF Interrupt Repeat
CONTROL BITS Sequence Select SMPI<2:0> = 1111 Interrupt on 16th sample BUFM = 0 Single 16-word result buffer ALTS = 0 Always use MUX A input select MUX A Input Select CH0SA<3:0> = 0000 Select AN0 for CH0+ input CH0NA = 0 Select VREF- for CH0- input CSCNA = 0 No input scan CSSL<15:0> = n/a Scan input select unused MUX B Input Select CH0SB<3:0> = n/a Channel CH0+ input unused CH0NB = n/a Channel CH0- input unused
18
DS70065D-page 18-23

18.14.2 Example: A/D Conversions While Scanning Through All Analog Inputs
Figure 18-10 and Table 18-2 illustrate a typical setup, where all available analog input channels are sampled and converted. The set CSCNA bit specifies scanning of the A/D inputs to the CH0 positive input. Other conditions are similar to Subsection 18.14.1. Initially, the AN0 input is sampled by CH0 and converted. The result is stored in the ADCBUF buffer. Then the AN1 input is sampled and converted. This process of scanning the inputs repeats 16 times until the buffer is full and then the module generates an interrupt. The entire process will then repeat. Figure 18-10: Scanning Through 16 Inputs/Interrupt
TSAMP TCONV
TSAMP TCONV
TSAMP TCONV
TSAMP TCONV
Input to CH0 ASAM SAMP DONE ADCBUF0 ADCBUF1 ADCBUFE ADCBUFF ADIF BSET ADCON1,#ASAM
AN0
AN1
AN14
AN15
DS70065D-page 18-24

Table 18-2: Scanning Through 16 Inputs/Interrupt OPERATION SEQUENCE Sample MUX A Inputs: AN0 -> CH0 Convert CH0, Write Buffer 0x0 Sample MUX A Inputs: AN1 -> CH0 Convert CH0, Write Buffer 0x1 Sample MUX A Inputs: AN2 -> CH0 Convert CH0, Write Buffer 0x2 Sample MUX A Inputs: AN3 -> CH0 Convert CH0, Write Buffer 0x3 Sample MUX A Inputs: AN4 -> CH0 Convert CH0, Write Buffer 0x4 Sample MUX A Inputs: AN5 -> CH0 Convert CH0, Write Buffer 0x5 Sample MUX A Inputs: AN6 -> CH0 Convert CH0, Write Buffer 0x6 Sample MUX A Inputs: AN7 -> CH0 Convert CH0, Write Buffer 0x7 Sample MUX A Inputs: AN8 -> CH0 Convert CH0, Write Buffer 0x8 Sample MUX A Inputs: AN9 -> CH0 Convert CH0, Write Buffer 0x9 Sample MUX A Inputs: AN10 -> CH0 Convert CH0, Write Buffer 0xA Sample MUX A Inputs: AN11 -> CH0 Convert CH0, Write Buffer 0xB Sample MUX A Inputs: AN12 -> CH0 Convert CH0, Write Buffer 0xC Sample MUX A Inputs: AN13 -> CH0 Convert CH0, Write Buffer 0xD Sample MUX A Inputs: AN14 -> CH0 Convert CH0, Write Buffer 0xE Sample MUX A Inputs: AN15 -> CH0 Convert CH0, Write Buffer 0xF Interrupt Repeat
CONTROL BITS Sequence Select SMPI<2:0> = 1111 Interrupt on 16th sample BUFM = 0 Single 16-word result buffer ALTS = 0 Always use MUX A input select MUX A Input Select CH0SA<3:0> = n/a Override by CSCNA CH0NA = 0 Select VREF- for CH0- input CSCNA = 1 Scan CH0+ Inputs CSSL<15:0> = 1111 1111 1111 1111 Scan all inputs MUX B Input Select CH0SB<3:0> = n/a Channel CH0+ input unused CH0NB = n/a Channel CH0- input unused
18
DS70065D-page 18-25

18.14.3 Example: Using Dual 8-Word Buffers
Refer to Subsection 17.15.4 in Section 17. 10-bit A/D Converter for an example that uses dual buffers.
18.14.4
Example: Using Alternating MUX A, MUX B Input Selections

See Subsection 17.15.5 in Section 17. 10-bit A/D Converter for an example that uses the MUX A and MUX B input selections.
18.15
A/D Sampling Requirements

The analog input model of the 12-bit A/D converter is shown in Figure 18-11. The total sampling time for the A/D is a function of the internal amplifier settling time and the holding capacitor charge time. For the A/D converter to meet its specified accuracy, the charge holding capacitor (CHOLD) must be allowed to fully charge to the voltage level on the analog input pin. The source impedance (RS), the interconnect impedance (RIC), and the internal sampling switch (RSS) impedance combine to directly affect the time required to charge the capacitor CHOLD. The combined impedance of the analog sources must therefore be small enough to fully charge the holding capacitor within the chosen sample time. To minimize the effects of pin leakage currents on the accuracy of the A/D converter, the maximum recommended source impedance, RS, is 2.5 k. After the analog input channel is selected (changed), this sampling function must be completed prior to starting the conversion. The internal holding capacitor will be in a discharged state prior to each sample operation. At least 1 TAD time period should be allowed between conversions for the sample time. For more details, see the device electrical specifications. Figure 18-11: 12-bit A/D Converter Analog Input Model
VDD VT = 0.6V RIC 250 Sampling Switch RSS CHOLD = DAC capacitance = 18 pF VSS Legend: CPIN = input capacitance VT = threshold voltage I leakage = leakage current at the pin due to various junctions RIC = interconnect resistance RSS = sampling switch resistance CHOLD = sample/hold capacitance (from DAC) RSS 3 k
Rs
ANx
VA
CPIN VT = 0.6V
I leakage 500 nA
Note: CPIN value depends on device package and is not tested. Effect of CPIN negligible if Rs 2.5 k.
DS70065D-page 18-26

18.16 Reading the A/D Result Buffer
The RAM is 12-bits wide, but the data is automatically formatted to one of four selectable formats when a read from the buffer is performed. The FORM<1:0> bits (ADCON1<9:8>) select the format. The formatting hardware provides a 16-bit result on the data bus for all of the data formats. Figure 18-12 shows the data output formats that can be selected using the FORM<1:0> control bits. Figure 18-12: A/D Output Data Formats RAM Contents: Read to Bus: Integer 0 0 0 0 d11 d10 d09 d08 d07 d06 d05 d04 d03 d02 d01 d00 d11 d10 d09 d08 d07 d06 d05 d04 d03 d02 d01 d00
Signed Integer
d11 d11 d11 d11 d11 d10 d09 d08 d07 d06 d05 d04 d03 d02 d01 d00
Fractional
d11 d10 d09 d08 d07 d06 d05 d04 d03 d02 d01 d00
18
Signed Fractional (1.15) d11 d10 d09 d08 d07 d04 d03 d02 d01 d00 d01 d00 0 0 0 0
Table 18-3:
VIN/VREF 4095/4096 4094/4096
Numerical Equivalents of Various Result Codes

12-bit Output Code 16-bit Unsigned Integer Format 0000 1111 1111 1111 = 4095 0000 1111 1111 1110 = 4094 16-bit Signed Integer Format 0000 0111 1111 1111 = 2047 0000 0111 1111 1110 = 2046 16-bit Unsigned Fractional Format 1111 1111 1111 0000 = 0.9998 1111 1111 1110 0000 = 0.9995 16-bit Signed Fractional Format 0111 1111 1111 0000 = 0.9995 0111 1111 1110 0000 = 0.9990
1111 1111 1111 1111 1111 1110
2049/4096 2048/4096 2047/4096
1000 0000 0001 1000 0000 0000 0111 1111 1111
0000 1000 0000 0001 = 2049 0000 1000 0000 0000 = 2048 0000 0111 1111 1111 = 2047
0000 0000 0000 0001 = 1 0000 0000 0000 0000 = 0 1111 1111 1111 1111 = -1
1000 0000 0001 0000 = 0.5002 1000 0000 0000 0000 = 0.500 0111 1111 1111 0000 = 0.4998
0000 0000 0001 0000 = 0.0005 0000 0000 0000 0000 = 0.000 1111 1111 1111 0000 = -0.0005
1/4096 0/4096
0000 0000 0001 0000 0000 0000
0000 0000 0000 0001 = 1 0000 0000 0000 0000 = 0
1111 1000 0000 0001 = -2047 1111 1000 0000 0000 = -2048
0000 0000 0001 0000 = 0.0002 0000 0000 0000 0000 = 0.000
1000 0000 0001 0000 = -0.9995 1000 0000 0000 0000 = -1.000
DS70065D-page 18-27

18.17 Transfer Function
The ideal transfer function of the A/D converter is shown in Figure 18-13. The difference of the input voltages (VINH VINL) is compared to the reference (VREFH VREFL). The first code transition occurs when the input voltage is (VREFH VREFL/8192) or 0.5 LSb. The 00 0000 0001 code is centered at (VREFH VREFL/4096) or 1.0 LSb. The 10 0000 0000 code is centered at (2048*(VREFH VREFL)/4096). An input voltage less than (1*(VREFH VREFL)/8192) converts as 00 0000 0000. An input greater than (8192*(VREFH VREFL)/8192) converts as 11 1111 1111.
Figure 18-13: A/D Transfer Function

Output Code 1111 1111 1111 (= 4095) 1111 1111 1110 (= 4094)
1000 0000 0011 (= 2051) 1000 0000 0010 (= 2050) 1000 0000 0001 (= 2049) 1000 0000 0000 (= 2048) 0111 1111 1111 (= 2047) 0111 1111 1110 (= 2046) 0111 1111 1101 (= 2045)
0000 0000 0001 (= 1) 0000 0000 0000 (= 0) VREFL VREFH VREFL VREFL + 4096 VREFL + 2048*(VREFH VREFL) 4096 (VINH VINL) VREFH
18.18
A/D Accuracy/Error
Refer to Section 18.26 Related Application Notes for a list of documents that discuss A/D accuracy.
18.19
Connection Considerations
Since the analog inputs employ ESD protection, they have diodes to VDD and VSS. This requires that the analog input must be between VDD and VSS. If the input voltage exceeds this range by greater than 0.3V (either direction), one of the diodes becomes forward biased and it may damage the device if the input current specification is exceeded. An external RC filter is sometimes added for anti-aliasing of the input signal. The R component should be selected to ensure that the sampling time requirements are satisfied. Any external components connected (via high-impedance) to an analog input pin (capacitor, zener diode, etc.) should have very little leakage current at the pin.
DS70065D-page 18-28

18.20 Initialization
Example 18-4 shows a simple initialization code example for the A/D module. In this particular configuration, all 16 analog input pins, AN0-AN15, are set up as analog inputs. Operation in Idle mode is disabled, output data is in unsigned fractional format, and AVDD and AVSS are used for VREFH and VREFL. The start of sampling, as well as the start of conversion (conversion trigger), are performed manually in software. Scanning of inputs is disabled and an interrupt occurs after every sample/convert sequence (1 conversion result). The A/D conversion clock is TCY/2; AN0 is converted. Since sampling is started manually by setting the SAMP bit (ADCON1<1>) after each conversion is complete, the auto-sample time bits, SAMC<4:0> (ADCON3<12:8>), are ignored. Moreover, since the start of conversion (i.e., end of sampling) is also triggered manually, the SAMP bit needs to be cleared each time a new sample needs to be converted. Example 18-4: CLR MOV MOV A/D Initialization Code Example ADPCFG #0x2200,W0 W0,ADCON1 ; Configure A/D port, ; all input pins are analog ; ; ; ; ; ; ; ; ; ; ; Configure sample clock source and conversion trigger mode. Unsigned Fractional format, Manual conversion trigger, Manual start of sampling, No operation in IDLE mode. Configure A/D voltage reference and buffer fill modes. VREF from AVDD and AVSS, Inputs are not scanned, Interrupt every sample
CLR
ADCON2
18
CLR CLR
ADCON3 ADCHS
; Configure A/D conversion clock ; Configure input channels, ; CH0+ input is AN0. ; CHO- input is VREFL (AVss) ; No inputs are scanned. ; Clear A/D conversion interrupt flag
CLR BCLR
ADCSSL IFS0,#ADIF
; Configure A/D interrupt priority bits (ADIP<2:0>) here, if ; required. (default priority level is 4) BSET BSET BSET CALL BCLR : : IEC0,#ADIE ADCON1,#ADON ADCON1,#SAMP DELAY ADCON1,#SAMP ; Enable A/D conversion interrupt ; ; ; ; ; ; Turn on A/D Start sampling the input Ensure the correct sampling time has elapsed before starting conversion. End A/D Sampling and start Conversion The DONE bit is set by hardware when conversion sequence is complete. ; The ADIF bit will be set.
DS70065D-page 18-29

18.21 A/D Conversion Speeds
The dsPIC30F 12-bit A/D converter specifications permit a maximum of 200 ksps sampling rate. The table below summarizes the conversion speeds for the dsPIC30F 12-bit A/D converter and the required operating conditions.
TABLE 18-4:
12-BIT A/D CONVERTER EXTENDED CONVERSION RATES

dsPIC30F 10-bit A/D Converter Conversion Rates
A/D Speed Up to 200 ksps(1)
TAD Sampling Minimum Time Min 333.33 ns 1 TAD
Rs Max 2.5 k
VDD 4.5V to 5.5V
Temperature -40C to +85C
A/D Channels Configuration

VREF- VREF+
ANx S/H
CHX ADC
Up to 100 ksps
666.67 ns
1 TAD
2.5 k
3.0V to 5.5V
-40C to +125C
CHX ADC
Note 1: External VREF- and VREF+ pins must be used for correct operation. See Figure 18-14 for recommended circuit.
DS70065D-page 18-30

The following figure depicts the recommended circuit for the conversion rates above 100 ksps. The dsPIC30F6014 is shown as an example. Figure 18-14: A/D Converter Voltage Reference Schematic
VDD
80 79 78 77 76 75 74 73 72
VSS 69 68 67 66
65 64
VDD
1 2 3 4 5 6 7 8 9 10 VDD VSS VDD 13 14 15 16 17 18 19 VREF+ VREFAVDD AVSS 27 VDD R2 10 C2 0.1 F C1 0.01 F 20 21 22 VDD VSS
63 62 61 60 59 58 57 56 55 54 53 52 VSS 50 49 VDD 47 46 45 44 43 42 41 VDD C5 1 F VDD C4 0.1 F VDD C3 0.01 F VDD C8 1 F C7 0.1 F C6 0.01 F VDD VDD VDD
dsPIC30F6014
18
28
29
30
33
34
35
36
37
38
39
R1 10
VDD VDD
The configuration procedures below give the required setup values for the conversion speeds above 100 ksps.
18.21.1

The following configuration items are required to achieve a 200 ksps conversion rate. Comply with conditions provided in Table 18-4. Connect external VREF+ and VREF- pins following the recommended circuit shown in Figure 18-14. Set SSRC<2.0> = 111 in the ADCON1 register to enable the auto convert option. Enable automatic sampling by setting the ASAM control bit in the ADCON1 register. Write the SMPI<3.0> control bits in the ADCON2 register for the desired number of conversions between interrupts. Configure the A/D clock period to be: 1 (14 + 1) x 200,000 = 333.33 ns
by writing to the ADCS<5:0> control bits in the ADCON3 register. Configure the sampling time to be 1 TAD by writing: SAMC<4:0> = 00001.
40
DS70065D-page 18-31

The following figure shows the timing diagram of the A/D running at 200 ksps. The TAD selection in conjunction with the guidelines described above allows a conversion speed of 200 ksps. See Example 18-1 for code example. Figure 18-15: Converting 1 Channel at 200 ksps, Auto-Sample Start, 1 TAD Sampling Time
TSAMP = 1 TAD ADCLK TCONV = 14 TAD TCONV = 14 TAD TSAMP = 1 TAD
SAMP DONE
ADCBUF0 ADCBUF1 Instruction Execution BSET ADCON1, ASAM
Example 18-1:
ADPCFG = ADCON1 =
Converting at 200 ksps, Auto-Sample Start, 1 TAD Sampling Time Code Example
// // // // // // // // // // // // all PORTB = Digital; RB2 = analog SSRC bit = 111 implies internal counter ends sampling and starts converting. Connect RB2/AN2 as CH0 input in this example RB2/AN2 is the input Sample time = 1Tad, Tad = 333.33 ns @ 30 MIPS which will give 1 / (15 * 333.33 ns) = 200 ksps Select external VREF+ and VREF- pins Interrupt after every 2 samples turn ADC ON repeat continuously clear value initialize ADCBUF pointer clear ADC interrupt flag auto start sampling for 31Tad then go to conversion conversion done? yes then stop sample/convert average the 2 ADC value
0xFFFB; 0x00E0;
ADCHS
0x0002; 0; 0x0113; 0x6004;
ADCSSL = ADCON3 = ADCON2 =
// // // // // while (!IFS0bits.ADIF); // ADCON1bits.ASAM = 0; // for (count = 0; count <2; count++)// ADCValue = ADCValue + *ADC16Ptr++; ADCValue = ADCValue >> 1; } //
repeat
DS70065D-page 18-32

18.22 Operation During Sleep and Idle Modes
Sleep and Idle modes are useful for minimizing conversion noise because the digital activity of the CPU, buses and other peripherals is minimized.
18.22.1
CPU Sleep Mode Without RC A/D Clock

When the device enters Sleep mode, all clock sources to the module are shutdown and stay at logic 0. If Sleep occurs in the middle of a conversion, the conversion is aborted unless the A/D is clocked from its internal RC clock generator. The converter will not resume a partially completed conversion on exiting from Sleep mode. Register contents are not affected by the device entering or leaving Sleep mode.
18.22.2
CPU Sleep Mode With RC A/D Clock

The A/D module can operate during Sleep mode if the A/D clock source is set to the internal A/D RC oscillator (ADRC = 1). This eliminates digital switching noise from the conversion. When the conversion is completed, the CONV bit will be cleared and the result loaded into the A/D result buffer, ADCBUF. If the A/D interrupt is enabled (ADIE = 1), the device will wake-up from Sleep when the A/D interrupt occurs. Program execution will resume at the A/D Interrupt Service Routine if the A/D interrupt is greater than the current CPU priority. Otherwise, execution will continue from the instruction after the PWRSAV instruction, that placed the device in Sleep mode. If the A/D interrupt is not enabled, the A/D module will then be turned off, although the ADON bit will remain set. To minimize the effects of digital noise on the A/D module operation, the user should select a conversion trigger source that ensures the A/D conversion will take place in Sleep mode. The automatic conversion trigger option can be used for sampling and conversion in Sleep (SSRC<2:0> = 111). To use the automatic conversion option, the ADON bit should be set in the instruction prior to the PWRSAV instruction. Note: For the A/D module to operate in Sleep, the A/D clock source must be set to RC (ADRC = 1).
18
18.22.3
A/D Operation During CPU Idle Mode

For the A/D, the ADSIDL bit (ADCON1<13>) selects if the module will stop on Idle or continue on Idle. If ADSIDL = 0, the module will continue normal operation when the device enters Idle mode. If the A/D interrupt is enabled (ADIE = 1), the device will wake-up from Idle mode when the A/D interrupt occurs. Program execution will resume at the A/D Interrupt Service Routine if the A/D interrupt is greater than the current CPU priority. Otherwise, execution will continue from the instruction after the PWRSAV instruction that placed the device in Idle mode. If ADSIDL = 1, the module will stop in Idle. If the device enters Idle mode in the middle of a conversion, the conversion is aborted. The converter will not resume a partially completed conversion on exiting from Idle mode.
18.23
Effects of a Reset
A device Reset forces all registers to their Reset state. This forces the A/D module to be turned off, and any conversion in progress is aborted. All pins that are multiplexed with analog inputs will be configured as analog inputs. The corresponding TRIS bits will be set. The values in the ADCBUF registers are not initialized during a Power-on Reset. ADCBUF0...ADCBUFF will contain unknown data.
DS70065D-page 18-33
18.24
Special Function Registers Associated with the 12-bit A/D Converter
The following table lists dsPIC30F 12-bit A/D Converter Special Function Registers, including their addresses and formats. All unimplemented registers and/or bits within a register read as zeros.
Table 18-4:
Bit 13 SI2CIF SI2CIE ADIP<2:0> ADC Data Buffer 0 ADC Data Buffer 1 ADC Data Buffer 2 ADC Data Buffer 3 ADC Data Buffer 4 ADC Data Buffer 5 ADC Data Buffer 6 ADC Data Buffer 7 ADC Data Buffer 8 ADC Data Buffer 9 ADC Data Buffer 10 ADC Data Buffer 11 ADC Data Buffer 12 ADC Data Buffer 13 ADC Data Buffer 14 ADC Data Buffer 15 ADSIDL CH0NB CH0SB[3:0] SAMC[4:0] CSCNA FORM[1:0] BUFS ADRC SSRC[2:0] CH0NA PCFG4 ADC Input Scan Select Register PCFG3 SMPI[3:0] ADCS[5:0] CH0SA[3:0] PCFG2 PCFG1 ASAM SAMP BUFM CONV ALTS U1TXIP<2:0> U1RXIP<2:0> NVMIE ADIE U1TXIE U1RXIE SPI1IE T3IE T2IE OC2IE IC2IE T1IE OC1IE IC1IE SPI1IP<2:0> NVMIF ADIF U1TXIF U1RXIF SPI1IF T3IF T2IF OC2IF IC2IF T1IF OC1IF IC1IF INT4EP INT3EP INT2EP INT1EP OVATE OVBTE COVTE MATHERR ADDRERR STKERR OSCFAIL Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
ADC Register Map

Reset States 0000 0000 0000 0000
DS70065D-page 18-34
INT0
File Name
ADR
Bit 15
Bit 14
INTCON1
0080
NSTDIS
INTCON2
0082
ALTIVT
INT0EP 0000 0000 0000 0000 0000 0000 0000 0000 INT0IE 0000 0000 0000 0000 0100 0100 0100 0100 uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 PCFG0 0000 0000 0000 0000 0000 0000 0000 0000
IFS0
0084
CNIF
MI2CIF
IEC0
008C
CNIE
MI2CIE
IPC2
0098
ADCBUF0
0280
ADCBUF1
0282
ADCBUF2
0284
ADCBUF3
0286
ADCBUF4
0288
ADCBUF5
028A
ADCBUF6
028C
ADCBUF7
028E
ADCBUF8
0290
ADCBUF9
0292
ADCBUFA
0294
ADCBUFB
0296
ADCBUFC
0298
ADCBUFD
029A
ADCBUFE
029C
ADCBUFF
029E
ADCON1
02A0
ADON
ADCON2
02A2
VCFG[2:0]
ADCON3
02A4
ADCHS
02A6
ADPCFG
02A8
PCFG15 PCFG14 PCFG13 PCFG12 PCFG11 PCFG10 PCFG9 PCFG8 PCFG7 PCFG6 PCFG5
ADCSSL
02AA
Legend:
u = unknown
Note:

18.25 Design Tips
Question 1: Answer: 1. Make sure you are meeting all of the timing specifications. If you are turning the module off and on, there is a minimum delay you must wait before taking a sample. If you are changing input channels, there is a minimum delay you must wait for this as well, and finally, there is TAD, which is the time selected for each bit conversion. This is selected in ADCON3 and should be within a certain range, as specified in the Electrical Characteristics. If TAD is too short, the result may not be fully converted before the conversion is terminated, and if TAD is made too long, the voltage on the sampling capacitor can decay before the conversion is complete. These timing specifications are provided in the Electrical Specifications section of the device data sheets. Often, the source impedance of the analog signal is high (greater than 10 k), so the current drawn from the source by leakage, and to charge the sample capacitor, can affect accuracy. If the input signal does not change too quickly, try putting a 0.1 F capacitor on the analog input. This capacitor will charge to the analog voltage being sampled and supply the instantaneous current needed to charge the 18 pF internal holding capacitor. Put the device into Sleep mode before the start of the A/D conversion. The RC clock source selection is required for conversions in Sleep mode. This technique increases accuracy, because digital noise from the CPU and other peripherals is minimized. Do you know of a good reference on A/Ds? How can I optimize the system performance of the A/D converter?
2.
3.
Question 2:
Answer: A good reference for understanding A/D conversions is the Analog-Digital Conversion Handbook third edition, published by Prentice Hall (ISBN 0-13-03-2848-0). Question 3: My combination of channels/sample and samples/interrupt is greater than the size of the buffer. What will happen to the buffer?
18
Answer: This configuration is not recommended. The buffer will contain unknown results.
DS70065D-page 18-35

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the 12-bit A/D Converter module are: Title Using the Analog-to-Digital (A/D) Converter Four Channel Digital Voltmeter with Display and Keyboard Understanding A/D Converter Performance Specifications Application Note # AN546 AN557 AN693
Note:
DS70065D-page 18-36

Revision B
To reflect editorial and technical content revisions for the dsPIC30F 12-bit A/D Converter module.
Revision C
Revision D
This revision includes the extended conversion rate guidelines.
18
DS70065D-page 18-37

NOTES:
DS70065D-page 18-38
Section 19. UART

HIGHLIGHTS
This section of the manual contains the following major topics: 19.1 Introduction .................................................................................................................. 19-2 19.2 Control Registers ......................................................................................................... 19-3 19.3 UART Baud Rate Generator (BRG) ............................................................................. 19-8 19.4 UART Configuration................................................................................................... 19-10 19.5 UART Transmitter ...................................................................................................... 19-11 19.6 UART Receiver .......................................................................................................... 19-14 19.7 Using the UART for 9-bit Communication.................................................................. 19-18 19.8 Receiving Break Characters ...................................................................................... 19-19 19.9 Initialization ................................................................................................................ 19-20 19.10 Other Features of the UART ...................................................................................... 19-21 19.11 UART Operation During CPU Sleep and Idle Modes ................................................ 19-21 19.12 Registers Associated with UART Module .................................................................. 19-22 19.13 Design Tips ................................................................................................................ 19-23 19.14 Related Application Notes.......................................................................................... 19-24 19.15 Revision History ......................................................................................................... 19-25
19
UART
DS70066C-page 19-1

19.1 Introduction
The Universal Asynchronous Receiver Transmitter (UART) module is one of the serial I/O modules available in the dsPIC30F device family. The UART is a full-duplex asynchronous system that can communicate with peripheral devices, such as personal computers, RS-232 and RS-485 interfaces. The primary features of the UART module are: Full-duplex 8- or 9-bit data transmission through the UxTX and UxRX pins Even, Odd or No Parity options (for 8-bit data) One or two Stop bits Fully integrated Baud Rate Generator with 16-bit prescaler Baud rates ranging from 29 bps to 1.875 Mbps at FCY = 30 MHz 4-deep First-In-First-Out (FIFO) transmit data buffer 4-deep FIFO receive data buffer Parity, Framing and Buffer Overrun error detection Support for 9-bit mode with Address Detect (9th bit = 1) Transmit and Receive Interrupts Loopback mode for diagnostic support Note: Each dsPIC30F device variant may have one or more UART modules. An x used in the names of pins, control/status bits and registers denotes the particular module. Refer to the specific device data sheets for more details.
A simplified block diagram of the UART is shown in Figure 19-1. The UART module consists of the key important hardware elements: Baud Rate Generator Asynchronous Transmitter Asynchronous Receiver Figure 19-1: UART Simplified Block Diagram
Baud Rate Generator
UART Receiver
UxRX
UART Transmitter
UxTX
DS70066C-page 19-2
Section 19. UART

UXMODE: UARTX Mode Register U-0 R/W-0 USIDL U-0 U-0 reserved R/W-0 ALTIO U-0 reserved U-0 reserved bit 8
Register 19-1: Upper Byte: R/W-0 UARTEN bit 15
Lower Byte: R/W-0 R/W-0 WAKE LPBACK bit 7 bit 15
R/W-0 ABAUD
U-0
U-0
R/W-0 R/W-0 PDSEL<1:0>
R/W-0 STSEL bit 0
UARTEN: UART Enable bit 1 = UART is enabled. UART pins are controlled by UART as defined by UEN<1:0> and UTXEN control bits. 0 = UART is disabled. UART pins are controlled by corresponding PORT, LAT, and TRIS bits. Unimplemented: Read as 0 USIDL: Stop in Idle Mode bit 1 = Discontinue operation when device enters Idle mode 0 = Continue operation in Idle mode Unimplemented: Read as 0 Reserved: Write 0 to this location ALTIO: UART Alternate I/O Selection bit 1 = UART communicates using UxATX and UxARX I/O pins 0 = UART communicates using UxTX and UxRX I/O pins Note: The alternate UART I/O pins are not available on all devices. See device data sheet for details. Reserved: Write 0 to these locations WAKE: Enable Wake-up on Start bit Detect During Sleep Mode bit 1 = Wake-up enabled 0 = Wake-up disabled LPBACK: UART Loopback Mode Select bit 1 = Enable Loopback mode 0 = Loopback mode is disabled ABAUD: Auto Baud Enable bit 1 = Input to Capture module from UxRX pin 0 = Input to Capture module from ICx pin Unimplemented: Read as 0 PDSEL<1:0>: Parity and Data Selection bits 11 = 9-bit data, no parity 10 = 8-bit data, odd parity 01 = 8-bit data, even parity 00 = 8-bit data, no parity STSEL: Stop Selection bit 1 = 2 Stop bits 0 = 1 Stop bit Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 14 bit 13
bit 9-8 bit 7
bit 6
bit 5
19
UART
bit 4-3 bit 2-1
bit 0
DS70066C-page 19-3

Register 19-2: Upper Byte: R/W-0 UTXISEL bit 15 UXSTA: UARTX Status and Control Register U-0 U-0 U-0 R/W-0 UTXBRK R/W-0 UTXEN R-0 UTXBF R-1 TRMT bit 8
Lower Byte: R/W-0 R/W-0 URXISEL<1:0> bit 7 bit 15
R/W-0 ADDEN
R-1 RIDLE
R-0 PERR
R-0 FERR
R/C-0 OERR
R-0 URXDA bit 0
UTXISEL: Transmission Interrupt Mode Selection bit 1 = Interrupt when a character is transferred to the Transmit Shift register and as result, the transmit buffer becomes empty 0 = Interrupt when a character is transferred to the Transmit Shift register (this implies that there is at least one character open in the transmit buffer) UTXBRK: Transmit Break bit 1 = UxTX pin is driven low, regardless of transmitter state 0 = UxTX pin operates normally UTXEN: Transmit Enable bit 1 = UART transmitter enabled, UxTX pin controlled by UART (if UARTEN = 1) 0 = UART transmitter disabled, any pending transmission is aborted and buffer is reset. UxTX pin controlled by PORT. UTXBF: Transmit Buffer Full Status bit (Read Only) 1 = Transmit buffer is full 0 = Transmit buffer is not full, at least one more data word can be written TRMT: Transmit Shift Register is Empty bit (Read Only) 1 = Transmit shift register is empty and transmit buffer is empty (the last transmission has completed) 0 = Transmit shift register is not empty, a transmission is in progress or queued in the transmit buffer URXISEL<1:0>: Receive Interrupt Mode Selection bit 11 =Interrupt flag bit is set when Receive Buffer is full (i.e., has 4 data characters) 10 =Interrupt flag bit is set when Receive Buffer is 3/4 full (i.e., has 3 data characters) 0x =Interrupt flag bit is set when a character is received ADDEN: Address Character Detect (bit 8 of received data = 1) 1 = Address Detect mode enabled. If 9-bit mode is not selected, this control bit has no effect. 0 = Address Detect mode disabled RIDLE: Receiver Idle bit (Read Only) 1 = Receiver is Idle 0 = Data is being received PERR: Parity Error Status bit (Read Only) 1 = Parity error has been detected for the current character 0 = Parity error has not been detected FERR: Framing Error Status bit (Read Only) 1 = Framing Error has been detected for the current character 0 = Framing Error has not been detected
bit 10
bit 9
bit 8
bit 7-6
bit 5
bit 4
bit 3
bit 2
DS70066C-page 19-4
Section 19. UART

Register 19-2: bit 1 UXSTA: UARTX Status and Control Register (Continued)
OERR: Receive Buffer Overrun Error Status bit (Read/Clear Only) 1 = Receive buffer has overflowed 0 = Receive buffer has not overflowed URXDA: Receive Buffer Data Available bit (Read Only) 1 = Receive buffer has data, at least one more character can be read 0 = Receive buffer is empty Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared C = Bit can be cleared x = Bit is unknown
bit 0
19
UART
DS70066C-page 19-5

Register 19-3: Upper Byte: U-0 bit 15 UXRXREG: UARTX Receive Register U-0 U-0 U-0 U-0 U-0 U-0 R-0 URX8 bit 8
Lower Byte: R-0 bit 7 bit 15-9 bit 8 bit 7-0
R-0
R-0
R-0 R-0 URX<7:0>
R-0
R-0
R-0 bit 0
Unimplemented: Read as 0 URX8: Data bit 8 of the Received Character (in 9-bit mode) URX<7:0>: Data bits 7-0 of the Received Character Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
UXTXREG: UARTX Transmit Register (Write Only) U-0 U-0 U-0 U-0 U-0 U-0 W-x UTX8 bit 8
Lower Byte: W-x bit 7 bit 15-9 bit 8 bit 7-0
W-x
W-x
W-x W-x UTX<7:0>
W-x
W-x
W-x bit 0
Unimplemented: Read as 0 UTX8: Data bit 8 of the Character to be Transmitted (in 9-bit mode) UTX<7:0>: Data bits 7-0 of the Character to be Transmitted Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70066C-page 19-6
Section 19. UART

Register 19-5: Upper Byte: R/W-0 bit 15 Lower Byte: R/W-0 bit 7 bit 15-0 BRG<15:0>: Baud Rate Divisor bits Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown UXBRG: UARTX Baud Rate Register R/W-0 R/W-0 R/W-0 R/W-0 BRG<15:8> R/W-0 R/W-0 R/W-0 bit 8
R/W-0
R/W-0
R/W-0 R/W-0 BRG<7:0>
R/W-0
R/W-0
R/W-0 bit 0
19
UART
DS70066C-page 19-7

19.3 UART Baud Rate Generator (BRG)
The UART module includes a dedicated 16-bit baud rate generator. The UxBRG register controls the period of a free running 16-bit timer. Equation 19-1 shows the formula for computation of the baud rate. Equation 19-1: UART Baud Rate
Baud Rate =
FCY 16 (UxBRG + 1)
UxBRG =
FCY 1 16 Baud Rate
Note:
FCY denotes the instruction cycle clock frequency.
Example 19-1 shows the calculation of the baud rate error for the following conditions: FCY = 4 MHz Desired Baud Rate = 9600 Example 19-1: Desired Baud Rate Baud Rate Error Calculation = FCY/(16 (UxBRG + 1))
Solving for UxBRG value: UxBRG UxBRG UxBRG Calculated Baud Rate Error = = = = = = = = ( (FCY/Desired Baud Rate)/16) 1 ((4000000/9600)/16) 1 [25.042] = 25 4000000/(16 (25 + 1)) 9615 (Calculated Baud Rate Desired Baud Rate) Desired Baud Rate (9615 9600)/9600 0.16%
The maximum baud rate possible is FCY / 16 (for UxBRG = 0), and the minimum baud rate possible is FCY / (16 * 65536). Writing a new value to the UxBRG register causes the BRG timer to be reset (cleared). This ensures the BRG does not wait for a timer overflow before generating the new baud rate.
DS70066C-page 19-8
Section 19. UART

19.3.1 Baud Rate Tables
UART baud rates are provided in Table 19-1 for common device instruction cycle frequencies (FCY). The minimum and maximum baud rates for each frequency are also shown. Table 19-1:
BAUD RATE (Kbps) 0.3 1.2 2.4 9.6 19.2 38.4 56 115 250 500 MIN. MAX. BAUD RATE (Kbps) 0.3 1.2 2.4 9.6 19.2 38.4 56 115 250 500 MIN. MAX. BAUD RATE (Kbps) 0.3 1.2 2.4 9.6 19.2 38.4 56 115 250 500 MIN. MAX. 0.005 312.5 0.0 0.0 65535 0 0.004 250 0.0 0.0 65535 0 0.003 192 0.0 0.0 65535 0 0.002 115.2 0.0 0.0 65535 0 0.011 750 0.0 0.0 65535 0 BRG value (decimal) 1041 259 129 32 15 7 0.010 625 4 MHz KBAUD 0.3001 1.2019 2.4038 9.6154 19.2308 % ERROR 0.0 +0.2 +0.2 +0.2 +0.2 0.0 0.0 65535 0 BRG value (decimal) 832 207 103 25 12 250 0.0 0.0286 1875 0.0 0.0 65535 0 BRG value (decimal) 2499 624 312 77 38 19 12 6 2 250 500 0.008 500 0.0 0.0 0.0 0.0 1 0 65535 0 BRG value (decimal) 639 159 79 19 9 4 0.007 480 0.0 0.0 65535 0 BRG value (decimal) 0.0238 1562.5 10 MHz KBAUD 0.3 1.1996 2.4038 9.6154 18.9394 39.0625 56.8182 % ERROR 0.0 0.0 +0.2 +0.2 -1.4 +1.7 +1.5 0.0 0.0 65535 0 BRG value (decimal) 2082 520 259 64 32 15 10 0.019 1250 8 MHz KBAUD 0.2999 1.199 2.4038 9.6154 19.2308 38.4615 55.5556 % ERROR -0.02 -0.08 +0.16 +0.16 +0.16 +0.16 -0.79 0.0 0.0 65535 0 BRG value (decimal) 1666 416 207 51 25 12 8
UART Baud Rates

% ERROR 0.0 0.0 0.0 +0.2 -0.4 -0.4 +1.5 +1.9 BRG value (decimal) 6249 1562 780 194 97 48 32 15 25 MHz KBAUD 0.3 1.2001 2.4002 9.5859 19.2901 38.1098 55.8036 111.6071 % ERROR +0.01 +0.01 +0.01 -0.15 0.47 -0.76 -0.35 -2.95 BRG value (decimal) 5207 1301 650 162 80 40 27 13 20 MHz KBAUD 0.3 1.1996 2.3992 9.6154 19.2308 37.8788 56.8182 113.6364 250 % ERROR 0.0 0.0 0.0 +0.2 +0.2 -1.4 +1.5 -1.2 0.0 BRG value (decimal) 4166 1041 520 129 64 32 21 10 4 16 MHz KBAUD 0.3 1.2005 2.3981 9.6154 19.2308 38.4615 55.5556 111.1111 250 500 0.015 1000 % ERROR +0.01 +0.04 -0.08 +0.16 +0.16 +0.16 -0.79 -3.38 0.0 0.0 0.0 0.0 BRG value (decimal) 3332 832 416 103 51 25 17 8 3 1 65535 0 BRG value (decimal) 1599 399 199 49 24
FCY = 30 MHz KBAUD 0.3 1.1996 2.4008 9.6154 19.1327 38.2653 56.8182 117.1875
FCY = 12 MHz KBAUD 0.3 1.2 2.3962 9.6154 19.2308 37.5 57.6923 % ERROR 0.0 0.0 -0.2 -0.2 +0.2 +0.2 -2.3
7.68 MHz KBAUD 0.3 1.2 2.4 9.6 19.2 % ERROR 0.0 0.0 0.0 0.0 0.0
19
UART
FCY = 5 MHz KBAUD 0.2999 1.2019 2.4038 9.4697 19.5313 39.0625 % ERROR 0.0 +0.2 +0.2 -1.4 +1.7 +1.7
3.072 MHz KBAUD 0.3 1.2 2.4 9.6 19.2 38.4 % ERROR 0.0 0.0 0.0 0.0 0.0 0.0
1.8432 MHz KBAUD 0.3 1.2 2.4 9.6 19.2 38.4 % ERROR 0.0 0.0 0.0 0.0 0.0 0.0
383 95 47 11 5 2
DS70066C-page 19-9

19.4 UART Configuration
The UART uses standard non-return-to-zero (NRZ) format (one Start bit, eight or nine data bits, and one or two Stop bits). Parity is supported by the hardware, and may be configured by the user as even, odd or no parity. The most common data format is 8 bits, no parity and one Stop bit (denoted as 8, N, 1), which is the default (POR) setting. The number of data bits and Stop bits, and the parity, are specified in the PDSEL<1:0> (UxMODE<2:1>) and STSEL (UxMODE<0>) bits. An on-chip dedicated 16-bit baud rate generator can be used to derive standard baud rate frequencies from the oscillator. The UART transmits and receives the LSb first. The UARTs transmitter and receiver are functionally independent, but use the same data format and baud rate.
19.4.1
Enabling the UART

The UART module is enabled by setting the UARTEN (UxMODE<15>) bit and UTXEN (UxSTA<10>) bit. Once enabled, the UxTX and UxRX pins are configured as an output and an input, respectively, overriding the TRIS and PORT register bit settings for the corresponding I/O port pins. The UxTX pin is at logic 1 when no transmission is taking place. Note: The UTXEN bit should not be set until the UARTEN bit has been set. Otherwise, UART transmissions will not be enabled.
19.4.2
Disabling the UART

The UART module is disabled by clearing the UARTEN (UxMODE<15>) bit. This is the default state after any Reset. If the UART is disabled, all UART pins operate as port pins under the control of their corresponding PORT and TRIS bits. Disabling the UART module resets the buffers to empty states. Any data characters in the buffers are lost, and the baud rate counter is reset. All error and status flags associated with the UART module are reset when the module is disabled. The URXDA, OERR, FERR, PERR, UTXEN, UTXBRK and UTXBF bits are cleared, whereas RIDLE and TRMT are set. Other control bits, including ADDEN, URXISEL<1:0>, UTXISEL, as well as the UxMODE and UxBRG registers, are not affected. Clearing the UARTEN bit while the UART is active will abort all pending transmissions and receptions and reset the module as defined above. Re-enabling the UART will restart the UART in the same configuration.
19.4.3
Alternate UART I/O Pins

Some dsPIC30F devices have an alternate set of UART transmit and receive pins that can be used for communications. The alternate UART pins are useful when the primary UART pins are shared by other peripherals. The alternate I/O pins are enabled by setting the ALTIO bit (UxMODE<10>). If ALTIO = 1, the UxATX and UxARX pins (alternate transmit and alternate receive pins, respectively) are used by the UART module, instead of the UxTX and UxRX pins. If ALTIO = 0, the UxTX and UxRX pins are used by the UART module.
DS70066C-page 19-10
Section 19. UART

19.5 UART Transmitter
The UART transmitter block diagram is shown in Figure 19-2. The heart of the transmitter is the Transmit Shift register (UxTSR). The Shift register obtains its data from the transmit FIFO buffer, UxTXREG. The UxTXREG register is loaded with data in software. The UxTSR register is not loaded until the Stop bit has been transmitted from the previous load. As soon as the Stop bit is transmitted, the UxTSR is loaded with new data from the UxTXREG register (if available). Note: Figure 19-2: The UxTSR register is not mapped in data memory, so it is not available to the user.
UART Transmitter Block Diagram

16 Internal Data Bus Word Write only Word or Byte Write UxMODE 15 9 8 UTX8 7 UxTXREG Low Byte 0 Transmit Control Control TSR Control Buffer Generate Flags Generate Interrupt Load TSR UxTXIF UTXBRK UxTX Data 0 (Start) 1 (Stop) Parity Parity Generator 16 Divider 16X Baud Clock from Baud Rate Generator UxSTA
Transmit FIFO
Transmit Shift Register (UxTSR)
UxTX
Control Signals
19
UART
Note: x denotes the UART number.
Transmission is enabled by setting the UTXEN enable bit (UxSTA<10>). The actual transmission will not occur until the UxTXREG register has been loaded with data and the Baud Rate Generator (UxBRG) has produced a shift clock (Figure 19-2). The transmission can also be started by first loading the UxTXREG register and then setting the UTXEN enable bit. Normally when transmission is first started, the UxTSR register is empty, so a transfer to the UxTXREG register will result in an immediate transfer to UxTSR. Clearing the UTXEN bit during a transmission will cause the transmission to be aborted and will reset the transmitter. As a result, the UxTX pin will revert to a high-impedance state. In order to select 9-bit transmission, the PDSEL<1:0> bits (UxMODE<2:1>) should be set to 11 and the ninth bit should be written to the UTX9 bit (UxTXREG<8>). A word write should be performed to UxTXREG so that all nine bits are written at the same time. Note: There is no parity in the case of 9-bit data transmission.
DS70066C-page 19-11

19.5.1 Transmit Buffer (UxTXB)
Each UART has a 4-deep, 9-bit wide FIFO transmit data buffer. The UxTXREG register provides user access to the next available buffer location. The user may write up to 4 words in the buffer. Once the UxTXREG contents are transferred to the UxTSR register, the current buffer location becomes available for new data to be written and the next buffer location is sourced to the UxTSR register. The UTXBF (UxSTA<9>) status bit is set whenever the buffer is full. If a user attempts to write to a full buffer, the new data will not be accepted into the FIFO. The FIFO is reset during any device Reset, but is not affected when the device enters a Power Saving mode or wakes up from a Power Saving mode.
19.5.2
Transmit Interrupt
The transmit interrupt flag (UxTXIF) is located in the corresponding interrupt flag status (IFS) register. The UTXISEL control bit (UxSTA<15>) determines when the UART will generate a transmit interrupt. 1. If UTXISEL = 0, an interrupt is generated when a word is transferred from the transmit buffer to the Transmit Shift register (UxTSR). This implies that the transmit buffer has at least one empty word. Since an interrupt is generated after the transfer of each individual word, this mode is useful if interrupts can be handled frequently (i.e., the ISR is completed before the transmission of the next word). If UTXISEL = 1, an interrupt is generated when a word is transferred from the transmit buffer to the Transmit Shift register (UxTSR) and the transmit buffer is empty. Since an interrupt is generated only after all 4 words have been transmitted, this Block Transmit mode is useful if the users code cannot handle interrupts quickly enough (i.e., the ISR is completed before the transmission of the next word).
2.
The UxTXIF bit will be set when the module is first enabled. The user should clear the UxTXIF bit in the ISR. Switching between the two Interrupt modes during operation is possible. Note: When the UTXEN bit is set, the UxTXIF flag bit will also be set if UTXISEL = 0, since the transmit buffer is not yet full (can move transmit data to the UxTXREG register).
While the UxTXIF flag bit indicates the status of the UxTXREG register, the TRMT bit (UxSTA<8>) shows the status of the UxTSR register. The TRMT status bit is a read only bit, which is set when the UxTSR register is empty. No interrupt logic is tied to this bit, so the user has to poll this bit in order to determine if the UxTSR register is empty.
DS70066C-page 19-12
Section 19. UART

19.5.3 Setup for UART Transmit
Steps to follow when setting up a transmission: 1. 2. 3. Initialize the UxBRG register for the appropriate baud rate (Section 19.3 UART Baud Rate Generator (BRG)). Set the number of data bits, number of Stop bits, and parity selection by writing to the PDSEL<1:0> (UxMODE<2:1>) and STSEL (UxMODE<0>) bits. If transmit interrupts are desired, set the UxTXIE control bit in the corresponding Interrupt Enable Control register (IEC). Specify the interrupt priority for the transmit interrupt using the UxTXIP<2:0> control bits in the corresponding Interrupt Priority Control register (IPC). Also, select the Transmit Interrupt mode by writing the UTXISEL (UxSTA<15>) bit. Enable the UART module by setting the UARTEN (UxMODE<15>) bit. Enable the transmission by setting the UTXEN (UxSTA<10>) bit, which will also set the UxTXIF bit. The UxTXIF bit should be cleared in the software routine that services the UART transmit interrupt. The operation of the UxTXIF bit is controlled by the UTXISEL control bit. Load data to the UxTXREG register (starts transmission). If 9-bit transmission has been selected, load a word. If 8-bit transmission is used, load a byte. Data can be loaded into the buffer until the UxTXBF status bit (UxSTA<9>) is set. Note: The UTXEN bit should not be set until the UARTEN bit has been set. Otherwise, UART transmissions will not be enabled.
4. 5.
6.
Figure 19-3:
Transmission (8-bit or 9-bit Data)
Write to UxTXREG Character 1 BCLK/16 (Shift Clock) UxTX Start Bit Bit 0 Bit 1 Character 1 UxTXIF UxTXIF Cleared by User Bit 7/8 Stop Bit
Character 1 to Transmit Shift Reg. TRMT bit
19
UART
Figure 19-4:
Transmission (Back to Back)
Write to UxTXREG BCLK/16 (Shift Clock) UxTX Character 1 Character 2
Start Bit
Bit 0
Bit 1 Character 1
Bit 7/8
Stop Bit
Start Bit
Bit 0 Character 2
UxTXIF (UTXISEL = 0) UxTXIF (UTXISEL = 1) UxTXIF Cleared by User in Software
Character 1 to Transmit Shift Reg. TRMT bit
Character 2 to Transmit Shift Reg.
Note: This timing diagram shows two consecutive transmissions.
DS70066C-page 19-13

19.5.4 Transmission of Break Characters
Setting the UTXBRK bit (UxSTA<11>) will force the UxTX line to 0. UTXBRK overrides any other transmitter activity. The user should wait for the transmitter to be Idle (TRMT = 1) before setting UTXBRK. To send a break character, the UTXBRK bit must be set by software and remain set for a minimum of 13 baud clocks. The baud clock periods are timed in software. The UTXBRK bit is then cleared by software to generate the Stop bit. The user must wait at least one or two baud clocks to ensure a valid Stop bit(s) before loading the UTXBUF again or renewing transmitter activity. Note: Sending a break character does not generate a transmitter interrupt.
19.6
UART Receiver
The receiver block diagram is shown in Figure 19-5. The heart of the receiver is the Receive (Serial) Shift register (UxRSR). The data is received on the UxRX pin and is sent to the data recovery block. The data recovery block operates at 16 times the baud rate, whereas the main receive serial shifter operates at the baud rate. After sampling the UxRX pin for the Stop bit, the received data in UxRSR is transferred to the receive FIFO (if it is empty). Note: The UxRSR register is not mapped in data memory, so it is not available to the user.
The data on the UxRX pin is sampled three times by a majority detect circuit to determine if a high or a low level is present at the UxRX pin. Figure 19-5 shows the sampling scheme.
19.6.1
Receive Buffer (UxRXB)

The UART receiver has a 4-deep, 9-bit wide FIFO receive data buffer. UxRXREG is a memory mapped register that provides access to the output of the FIFO. It is possible for 4 words of data to be received and transferred to the FIFO and a fifth word to begin shifting to the UxRSR register before a buffer overrun occurs.
19.6.2
Receiver Error Handling

If the FIFO is full (four characters) and a fifth character is fully received into the UxRSR register, the overrun error bit, OERR (UxSTA<1>), will be set. The word in UxRSR will be kept, but further transfers to the receive FIFO are inhibited as long as the OERR bit is set. The user must clear the OERR bit in software to allow further data to be received. If it is desired to keep the data received prior to the overrun, the user should first read all five characters, then clear the OERR bit. If the five characters can be discarded, the user can simply clear the OERR bit. This effectively resets the receive FIFO and all prior received data is lost. Note: The data in the receive FIFO should be read prior to clearing the OERR bit. The FIFO is reset when OERR is cleared, which causes all data in the buffer to be lost.
The framing error bit, FERR (UxSTA<2>), is set if a Stop bit is detected as a logic low level. The parity error bit, PERR (UxSTA<3>), is set if a parity error has been detected in the data word at the top of the buffer (i.e., the current word). For example, a parity error would occur if the parity is set to be even, but the total number of ones in the data has been detected to be odd. The PERR bit is irrelevant in the 9-bit mode. The FERR and PERR bits are buffered along with the corresponding word and should be read before reading the data word.
DS70066C-page 19-14
Section 19. UART

19.6.3 Receive Interrupt
The UART receive interrupt flag (UxRXIF) is located in the corresponding Interrupt Flag Status (IFS) register. The URXISEL<1:0> (UxSTA<7:6>) control bits determine when the UART receiver generates an interrupt. a) If URXISEL<1:0> = 00 or 01, an interrupt is generated each time a data word is transferred from the Receive Shift register (UxRSR) to the receive buffer. There may be one or more characters in the receive buffer. If URXISEL<1:0> = 10, an interrupt is generated when a word is transferred from the Receive Shift register (UxRSR) to the receive buffer and as a result, the receive buffer contains 3 or 4 characters. If URXISEL<1:0> = 11, an interrupt is generated when a word is transferred from the Receive Shift register (UxRSR) to the receive buffer and as a result, the receive buffer contains 4 characters (i.e., becomes full).
b)
c)
Switching between the three Interrupt modes during operation is possible. While the URXDA and UxRXIF flag bits indicate the status of the UxRXREG register, the RIDLE bit (UxSTA<4>) shows the status of the UxRSR register. The RIDLE status bit is a read only bit, which is set when the receiver is Idle (i.e., the UxRSR register is empty). No interrupt logic is tied to this bit, so the user has to poll this bit in order to determine if the UxRSR is Idle. The URXDA bit (UxSTA<0>) indicates whether the receive buffer has data or whether the buffer is empty. This bit is set as long as there is at least one character to be read from the receive buffer. URXDA is a read only bit. Figure 19-5 shows a block diagram of the UART receiver.
19
UART
DS70066C-page 19-15

Figure 19-5: UART Receiver Block Diagram
Internal Data Bus 16
Word Read Only
Word or Byte Read
UxMODE
UxSTA
15
8 URX8
7 UxRXREG Low Byte
Receive Buffer Control Generate Flags Generate Interrupt Shift Data Characters
9 Load RSR to Buffer
UxRXIF
LPBACK From UxTX

1
UxRX
Receive Shift Register (UxRSR) START bit Detect Parity Check Stop bit Detect Shift Clock Generation Wake Logic
16 Divider 16X Baud Clock from Baud Rate Generator
DS70066C-page 19-16
PERR
FERR
Control Signals
Section 19. UART

19.6.4 Setup for UART Reception
Steps to follow when setting up a Reception: 1. 2. 3. Initialize the UxBRG register for the appropriate baud rate (Section 19.3 UART Baud Rate Generator (BRG)). Set the number of data bits, number of Stop bits and parity selection by writing to the PDSEL<1:0> (UxMODE<2:1>) and STSEL (UxMODE<0>) bits. If interrupts are desired, then set the UxRXIE bit in the corresponding Interrupt Enable Control register (IEC). Specify the interrupt priority for the interrupt using the UxRXIP<2:0> control bits in the corresponding Interrupt Priority Control register (IPC). Also, select the Receive Interrupt mode by writing to the URXISEL<1:0> (UxSTA<7:6>) bits. Enable the UART module by setting the UARTEN (UxMODE<15>) bit. Receive interrupts will depend on the URXISEL<1:0> control bit settings. If receive interrupts are not enabled, the user can poll the URXDA bit. The UxRXIF bit should be cleared in the software routine that services the UART receive interrupt. Read data from the receive buffer. If 9-bit transmission has been selected, read a word. Otherwise, read a byte. The URXDA status bit (UxSTA<0>) will be set whenever data is available in the buffer.
4. 5.
6.
Figure 19-6:
UART Reception
UxRX
Start bit
bit0
bit1
bit7
Stop bit
Start bit
bit0
bit7
Stop bit
UxRXIF (URXISEL = 0x)
Character 1 to UxRXREG RIDLE bit
Character 2 to UxRXREG
Note: This timing diagram shows 2 characters received on the UxRX input.
Figure 19-7:
UART Reception with Receive Overrun

Character 1 Characters 2,3,4,5 Start bit bit7/8 Stop bit Character 6 Start bit bit7/8 Stop bit
19
bit0
UxRX
Start bit
bit0
bit1
bit7/8 Stop bit
UART
Character 1,2,3,4 stored in Receive FIFO OERR bit
Character 5 held in UxRSR OERR Cleared by User
RIDLE bit
Note: This diagram shows 6 characters received without the user reading the input buffer. The 5th character received is held in the receive shift register. An overrun error occurs at the start of the 6th character.
DS70066C-page 19-17

19.7 Using the UART for 9-bit Communication
A typical multi-processor communication protocol will differentiate between data bytes and address/control bytes. A common scheme is to use a 9th data bit to identify whether a data byte is address or data information. If the 9th bit is set, the data is processed as address or control information. If the 9th bit is cleared, the received data word is processed as data associated with the previous address/control byte. The protocol operates as follows: The master device transmits a data word with the 9th bit set. The data word contains the address of a slave device. All slave devices in the communication chain receive the address word and check the slave address value. The slave device that was addressed will receive and process subsequent data bytes sent by the master device. All other slave devices will discard subsequent data bytes until a new address word (9th bit set) is received.
19.7.1
ADDEN Control Bit

The UART receiver has an Address Detect mode which allows it to ignore data words with the 9th bit cleared. This reduces the interrupt overhead, since data words with the 9th bit cleared are not buffered. This feature is enabled by setting the ADDEN bit (UxSTA<5>). The UART must be configured for 9-bit data to use the Address Detect mode. The ADDEN bit has no effect when the receiver is configured in 8-bit Data mode.
19.7.2
Setup for 9-bit Transmit

The setup procedure for 9-bit transmission is identical to the 8-bit Transmit modes, except that PDSEL<1:0> (UxMODE<2:1) should be set to 11 (see Section 19.5.3 Setup for UART Transmit). Word writes should be performed to the UxTXREG register (starts transmission).
DS70066C-page 19-18
Section 19. UART

19.7.3 Setup for 9-bit Reception Using Address Detect Mode
The setup procedure for 9-bit reception is similar to the 8-bit Receive modes, except that PDSEL<1:0> (UxMODE<2:1) should be set to 11 (see Section 19.6.4 Setup for UART Reception). The Receive Interrupt mode should be configured by writing to the URXISEL<1:0> (UxSTA<7:6>) bits. Note: If the Address Detect mode is enabled (ADDEN = 1), the URXISEL<1:0> control bits should be configured so that an interrupt will be generated after every received word. Each received data word must be checked in software for an address match immediately after reception.
The procedure for using the Address Detect mode is as follows: 1. 2. 3. 4. Set the ADDEN (UxSTA<5>) bit to enable address detect. Ensure that the URXISEL control bits are configured to generate an interrupt after each received word. Check each 8-bit address by reading the UxRXREG register, to determine if the device is being addressed. If this device has not been addressed, then discard the received word. If this device has been addressed, clear the ADDEN bit to allow subsequent data bytes to be read into the receive buffer and interrupt the CPU. If a long data packet is expected, then the Receive Interrupt mode could be changed to buffer more than one data byte between interrupts. When the last data byte has been received, set the ADDEN bit so that only address bytes will be received. Also, ensure that the URXISEL control bits are configured to generate an interrupt after each received word.
5.
Figure 19-8:
Reception with Address Detect (ADDEN = 1)

Bit 8 = 0, Data Byte Bit 8 = 1, Address Byte
UxRX (pin) Transfer to Receive FIFO Read Rcv Buffer Reg UxRXREG UxRXIF (Interrupt Flag)
Start bit
bit0
bit1
bit8
Stop bit
Start bit
bit0
bit8
Stop bit
Word 1 UxRXREG
19
UART
Note: This timing diagram shows a data byte followed by an address byte. The data byte is not read into the UxRXREG (receive buffer) because ADDEN = 1 and bit 8 = 0.
19.8
Receiving Break Characters

The receiver will count and expect a certain number of bit times based on the values programmed in the PDSEL (UxMODE<2:1>) and STSEL (UxMODE<0>) bits. If the break is longer than 13 bit times, the reception is considered complete after the number of bit times specified by PDSEL and STSEL. The URXDA bit is set, FERR is set, zeros are loaded into the receive FIFO, interrupts are generated if appropriate and the RIDLE bit is set. When the module receives a break signal and the receiver has detected the Start bit, the data bits and the invalid Stop bit (which sets the FERR), the receiver must wait for a valid Stop bit before looking for the next Start bit. It cannot assume that the break condition on the line is the next Start bit. Break is regarded as a character containing all 0s with the FERR bit set. The break character is loaded into the buffer. No further reception can occur until a Stop bit is received. Note that RIDLE goes high when the Stop bit has been received.
DS70066C-page 19-19

19.9 Initialization
Example 19-2 is an initialization routine for the Transmitter/Receiver in 8-bit mode. Example 19-3 shows an initialization of the Addressable UART in 9-bit Address Detect mode. In both examples, the value to load into the UxBRG register is dependent on the desired baud rate and the device frequency. Note: The UTXEN bit should not be set until the UARTEN bit has been set. Otherwise, UART transmissions will not be enabled. 8-bit Transmit/Receive (UART1)
; Set Baudrate
Example 19-2:
MOV MOV BSET BCLR BCLR BSET BCLR BCLR CLR MOV
#baudrate,W0 W0,U1BRG IPC2,#U1TXIP2 IPC2,#U1TXIP1 IPC2,#U1TXIP0 IPC2,#U1RXIP2 IPC2,#U1RXIP1 IPC2,#U1RXIP0 U1STA #0x8800,W0
; Set UART TX interrupt priority ; ; ; Set UART RX interrupt priority ; ;
; Enable UART for 8-bit data, ; no parity, 1 STOP bit, ; no wakeup
MOV BSET BSET BSET
W0,U1MODE U1STA,#UTXEN IEC0,#U1TXIE IEC0,#U1RXIE ; Enable transmit ; Enable transmit interrupts ; Enable receive interrupts
Example 19-3:
MOV MOV BSET BCLR BCLR BSET BCLR BCLR BSET MOV
8-bit Transmit/Receive (UART1), Address Detect Enabled

; Set Baudrate
#baudrate,W0 W0,U1BRG IPC2,#U1TXIP2 IPC2,#U1TXIP1 IPC2,#U1TXIP0 IPC2,#U1RXIP2 IPC2,#U1RXIP1 IPC2,#U1RXIP0 U1STA,#ADDEN #0x8883,W0
; Set UART TX interrupt priority ; ; ; Set UART RX interrupt priority ; ; ; Enable address detect ; UART1 enabled for 9-bit data, ; no parity, 1 STOP bit, ; wakeup enabled
MOV BSET BSET BSET
W0,U1MODE U1STA,#UTXEN IEC0,#U1TXIE IEC0,#U1RXIE ; Enable transmit ; Enable transmit interrupts ; Enable receive interrupts
DS70066C-page 19-20
Section 19. UART

19.10 19.10.1 Other Features of the UART UART in Loopback Mode
Setting the LPBACK bit enables this special mode, in which the UxTX output is internally connected to the UxRX input. When configured for the Loopback mode, the UxRX pin is disconnected from the internal UART receive logic. However, the UxTX pin still functions normally. To select this mode: 1. 2. 3. Configure UART for the desired mode of operation. Set LPBACK = 1 to enable Loopback mode. Enable transmission as defined in Section 19.5 UART Transmitter.
The Loopback mode is dependent on the UEN<1:0> bits, as shown in Table 19-2. Table 19-2: 00 01 10 11 Loopback Mode Pin Function UxRX input connected to UxTX; UxTX pin functions; UxRX pin ignored; UxCTS/UxRTS unused UxRX input connected to UxTX; UxTX pin functions; UxRX pin ignored; UxRTS pin functions, UxCTS unused UxRX input connected to UxTX; UxTX pin functions; UxRX pin ignored; UxRTS pin functions, UxCTS input connected to UxRTS; UxCTS pin ignored UxRX input connected to UxTX; UxTX pin functions; UxRX pin ignored; BCLK pin functions; UxCTS/UxRTS unused
UEN<1:0> Pin Function, LPBACK = 1
19.10.2
Auto Baud Support

To allow the system to determine baud rates of the received characters, the UxRX input can be internally connected to a selected input capture channel. When the ABAUD bit (UxMODE<5>) is set, the UxRX pin is internally connected to the input capture channel. The ICx pin is disconnected from the input capture channel. The input capture channel used for auto baud support is device specific. Please refer to the device data sheet for further details. This mode is only valid when the UART is enabled (UARTEN = 1) and the Loopback mode is disabled (LPBACK = 0). Also, the user must program the capture module to detect the falling and rising edges of the Start bit.
19
UART
19.11
UART Operation During CPU Sleep and Idle Modes

The UART does not function in Sleep mode. If entry into Sleep mode occurs while a transmission is in progress, then the transmission is aborted and the UxTX pin is driven to logic 1. Similarly, if entry into Sleep mode occurs while a reception is in progress, then the reception is aborted. The UART can be used to optionally wake the dsPIC device from Sleep mode on the detection of a Start bit. If the WAKE bit (UxSTA<7>) is set, the device is in Sleep mode, and the UART receive interrupt is enabled (UxRXIE = 1), then a falling edge on the UxRX pin will generate a receive interrupt. The Receive Interrupt Select mode bit (URXISEL) has no effect for this function. The UARTEN bit must be set in order to generate a wake-up interrupt. The USIDL bit (UxMODE<13>) selects if the module will stop operation when the device enters Idle mode, or whether the module will continue normal operation in Idle mode. If USIDL = 0, the module will continue normal operation during Idle mode. If USIDL = 1, the module will stop in Idle mode. Any transmission or reception in progress will be aborted.
DS70066C-page 19-21
19.12
Registers Associated with UART Module
Table 19-3:
Bit 13 USIDL UTXBRK UTX8 URX8 Receive Register Transmit Register UTXEN UTXBF TRMT URXISEL<1:0> ADDEN RIDLE PERR FERR OERR URXDA reserved ALTIO reserved reserved WAKE LPBACK ABAUD Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 PDSEL<1:0> STSEL Bit 1 Bit 0 Reset State
Registers Associated with UART1

0000 0000 0000 0000 0000 0001 0001 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 T2IF T2IE U1RXIP<2:0> OC2IE IC2IE T1IE OC1IE IC1IE OC2IF IC2IF T1IF OC1IF SPI1IP<2:0> IC1IF INT0 INT0IE 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100
SFR Name
Bit 15
Bit 14
U1MODE
UARTEN
DS70066C-page 19-22
U1STA
UTXISEL
U1TXREG
U1RXREG Baud Rate Generator Prescaler SI2CIF SI2CIE NVMIE ADIE U1TXIE U1TXIP<2:0> U1RXIE SPI1IE T3IE NVMIF ADIF U1TXIF U1RXIF SPI1IF T3IF
U1BRG
IFS0
CNIF
MI2CIF
IEC0
CNIE
MI2CIE
IPC2
ADIP<2:0>
Note:
The registers associated with UART1 are shown for reference. See the device data sheet for the registers associated with other UART modules.
Section 19. UART

19.13 Design Tips
Question 1: The data I transmit with the UART does not get received correctly. What could cause this?
Answer: The most common reason for reception errors is that an incorrect value has been calculated for the UART baud rate generator. Ensure the value written to the UxBRG register is correct. Question 2: I am getting framing errors even though the signal on the UART receive pin looks correct. What are the possible causes?
Answer: Ensure the following control bits have been setup correctly: UxBRG: UART Baud Rate register PDSEL<1:0>: Parity and Data Size Selection bits STSEL: Stop bit Selection
19
UART
DS70066C-page 19-23

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the UART module are: Title Serial Port Utilities Implementing Table Read and Table Write Application Note # AN547
Note:
DS70066C-page 19-24
Section 19. UART

Revision B
Revision B has been expanded to include a full description of the dsPIC30F UART module.
Revision C
19
UART
DS70066C-page 19-25

NOTES:
DS70066C-page 19-26
Section 20. Serial Peripheral Interface (SPI)

HIGHLIGHTS
This section of the manual contains the following major topics: 20.1 20.2 20.3 20.4 20.5 20.6 20.7 20.8 Introduction .................................................................................................................. 20-2 Status and Control Registers ....................................................................................... 20-4 Modes of Operation ..................................................................................................... 20-7 SPI Master Mode Clock Frequency ........................................................................... 20-19 Operation in Power Save Modes ............................................................................... 20-20 Special Function Registers Associated with SPI Modules......................................... 20-22 Related Application Notes.......................................................................................... 20-23 Revision History ......................................................................................................... 20-24
20
Serial Peripheral Interface (SPI)
DS70067C-page 20-1

20.1 Introduction
The Serial Peripheral Interface (SPI) module is a synchronous serial interface useful for communicating with other peripheral or microcontroller devices. These peripheral devices may be Serial EEPROMs, shift registers, display drivers, A/D converters, etc. The SPI module is compatible with Motorola's SPI and SIOP interfaces. Depending on the variant, the dsPIC30F family offers one or two SPI modules on a single device. SPI1 and SPI2 are functionally identical. The SPI2 module is available in many of the higher pin count packages (64-pin and higher), while the SPI1 module is available on all devices. Note: In this section, the SPI modules are referred together as SPIx or separately as SPI1 and SPI2. Special Function registers will follow a similar notation. For example, SPIxCON refers to the control register for the SPI1 or SPI2 module.
The SPI serial port consists of the following Special Function Registers (SFR): SPIxBUF: Address in SFR space that is used to buffer data to be transmitted and data that is received. This address is shared by the SPIxTXB and SPIxRXB registers. SPIxCON: A control register that configures the module for various modes of operation. SPIxSTAT: A status register that indicates various status conditions.
In addition, there is a 16-bit shift register, SPIxSR, that is not memory mapped. It is used for shifting data in and out of the SPI port. The memory mapped SFR, SPIxBUF, is the SPI Data Receive/Transmit register. Internally, the SPIxBUF register actually comprises of two separate registers - SPIxTXB and SPIxRXB. The Receive Buffer register, SPIxRXB, and the Transmit Buffer register, SPIxTXB, are two unidirectional 16-bit registers. These registers share the SFR address named SPIxBUF. If a user writes data to be transmitted to the SPIxBUF address, internally the data gets written to the SPIxTXB register. Similarly, when the user reads the received data from SPIxBUF, internally the data is read from the SPIxRXB register. This double-buffering of transmit and receive operations allows continuous data transfers in the background. Transmission and reception occur simultaneously. Note: The user cannot write to the SPIxTXB register or read from the SPIxRXB register directly. All reads and writes are performed on the SPIxBUF register.
The SPI serial interface consists of the following four pins: SDIx: serial data input SDOx: serial data output SCKx: shift clock input or output SSx: active low slave select or frame synchronization I/O pulse Note: The SPI module can be configured to operate using 3 or 4 pins. In the 3-pin mode, the SSx pin is not used.
DS70067C-page 20-2

Figure 20-1:
SPI Module Block Diagram

Internal Data Bus
SPIxBUF
Registers share address SPIxBUF SPIxRXB SPIxTXB Transmit
Receive SPIxSR SDIx bit0
SDOx Slave Select and Frame Sync Control
Shift Control Clock Control Edge Select Secondary Prescaler 1:1 1:8 Primary Prescaler 1, 4, 16, 64
SSx
FCY
SCKx
Enable Master Clock
Note: The SPIxTXB and SPIxRXB registers are memory mapped to the SPIxBUF register.
20
DS70067C-page 20-3

20.2 Status and Control Registers
SPIxSTAT: SPI Status and Control Register
Register 20-2: Upper Byte: R/W-0 SPIEN bit 15
U-0
R/W-0 SPISIDL
U-0
U-0
U-0
U-0
U-0 bit 8
R/W-0 HS SPIROV
U-0
U-0
U-0
U-0
R-0 SPITBF
R-0 SPIRBF bit 0
bit 14 bit 13
bit 12-7 bit 6
bit 5-2 bit 1
bit 0
SPIEN: SPI Enable bit 1 = Enables module and configures SCKx, SDOx, SDIx and SSx as serial port pins 0 = Disables module Unimplemented: Read as 0 SPISIDL: Stop in Idle Mode bit 1 = Discontinue module operation when device enters Idle mode 0 = Continue module operation in Idle mode Unimplemented: Read as 0 SPIROV: Receive Overflow Flag bit 1 = A new byte/word is completely received and discarded. The user software has not read the previous data in the SPIxBUF register. 0 = No overflow has occurred Unimplemented: Read as 0 SPITBF: SPI Transmit Buffer Full Status bit 1 = Transmit not yet started, SPIxTXB is full 0 = Transmit started, SPIxTXB is empty Automatically set in hardware when CPU writes SPIxBUF location, loading SPIxTXB. Automatically cleared in hardware when SPIx module transfers data from SPIxTXB to SPIxSR. SPIRBF: SPI Receive Buffer Full Status bit 1 = Receive complete, SPIxRXB is full 0 = Receive is not complete, SPIxRXB is empty Automatically set in hardware when SPIx transfers data from SPIxSR to SPIxRXB. Automatically cleared in hardware when core reads SPIxBUF location, reading SPIxRXB. Legend: R = Readable bit HC = Cleared by Hardware -n = Value at Reset W = Writable bit HS = Set by Hardware 1 = Bit is set 0 = Bit is cleared x = Bit is unknown U = Unimplemented bit, read as 0
DS70067C-page 20-4

Register 20-2: SPIXCON: SPIx Control Register
Upper Byte: U-0 R/W-0 FRMEN bit 15
R/W-0 SPIFSD
U-0
R/W-0 DISSDO
R/W-0 MODE16
R/W-0 SMP
R/W-0 CKE bit 8
Lower Byte: R/W-0 SSEN bit 7 bit 15 bit 14
R/W-0 CKP
R/W-0 MSTEN
R/W-0
R/W-0 SPRE<2:0>
R/W-0
R/W-0 R/W-0 PPRE<1:0> bit 0
bit 13
bit 12 bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
Unimplemented: Read as 0 FRMEN: Framed SPI Support bit 1 = Framed SPI support enabled 0 = Framed SPI support disabled SPIFSD: Frame Sync Pulse Direction Control on SSx pin bit 1 = Frame sync pulse input (slave) 0 = Frame sync pulse output (master) Unimplemented: Read as 0 DISSDO: Disable SDOx pin bit 1 = SDOx pin is not used by module. Pin is controlled by associated port register. 0 = SDOx pin is controlled by the module MODE16: Word/Byte Communication Select bit 1 = Communication is word-wide (16 bits) 0 = Communication is byte-wide (8 bits) SMP: SPI Data Input Sample Phase bit Master mode: 1 = Input data sampled at end of data output time 0 = Input data sampled at middle of data output time Slave mode: SMP must be cleared when SPI is used in Slave mode. CKE: SPI Clock Edge Select bit 1 = Serial output data changes on transition from active clock state to Idle clock state (see bit 6) 0 = Serial output data changes on transition from Idle clock state to active clock state (see bit 6) Note: The CKE bit is not used in the Framed SPI modes. The user should program this bit to 0 for the Framed SPI modes (FRMEN = 1). SSEN: Slave Select Enable (Slave mode) bit 1 = SS pin used for Slave mode 0 = SS pin not used by module. Pin controlled by port function. CKP: Clock Polarity Select bit 1 = Idle state for clock is a high level; active state is a low level 0 = Idle state for clock is a low level; active state is a high level MSTEN: Master Mode Enable bit 1 = Master mode 0 = Slave mode
20
DS70067C-page 20-5

Register 20-2: bit 4-2 SPIXCON: SPIx Control Register (Continued)
bit 1-0
SPRE<2:0>: Secondary Prescale (Master Mode) bits (Supported settings: 1:1, 2:1 through 8:1, all inclusive) 111 = Secondary prescale 1:1 110 = Secondary prescale 2:1 ... 000 = Secondary prescale 8:1 PPRE<1:0>: Primary Prescale (Master Mode) bits 11 = Primary prescale 1:1 10 = Primary prescale 4:1 01 = Primary prescale 16:1 00 = Primary prescale 64:1 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70067C-page 20-6

The SPI module has flexible Operating modes which are discussed in the following subsections: 8-bit and 16-bit Data Transmission/Reception Master and Slave Modes Framed SPI Modes
20.3.1
8-bit vs. 16-bit Operation

A control bit, MODE16 (SPIxCON<10>), allows the module to communicate in either 8-bit or 16-bit modes. The functionality will be the same for each mode except the number of bits that are received and transmitted. Additionally, the following should be noted in this context: The module is reset when the value of the MODE16 (SPIxCON<10>) bit is changed. Consequently, the bit should not be changed during normal operation. Data is transmitted out of bit 7 of the SPIxSR for 8-bit operation while it is transmitted and out of bit 15 of the SPIxSR for 16-bit operation. In both modes, data is shifted into bit 0 of the SPIxSR. 8 clock pulses at the SCKx pin are required to shift in/out data in 8-bit mode, while 16 clock pulses are required in the 16-bit mode.
20.3.2
Master and Slave Modes

SPI Master/Slave Connection
Figure 20-2:
PROCESSOR 1 [SPI Master]
PROCESSOR 2 [SPI Slave]
SDOx
SDIx
Serial Receive Buffer (SPIxRXB)
Shift Register (SPIxSR) MSbit LSbit
SDIx
SDOx
Serial Transmit Buffer (SPIxTXB)
Serial Clock
SPI Buffer (SPIxBUF) SCKx
SCKx
SPI Buffer (SPIxBUF)
SSx
SSx
(MSTEN(SPIxCON<5> = 1))
(SSEN(SPIxCON<7>) = 1 and MSTEN(SPIxCON<5> = 0))
20
Note 1: Using the SSx pin in Slave mode of operation is optional. 2: User must write transmit data to/read received data from SPIxBUF. The SPIxTXB and SPIxRXB registers are memory mapped to SPIxBUF.
DS70067C-page 20-7

20.3.2.1 Master Mode The following steps should be taken to set up the SPI module for the Master mode of operation: 1. If using interrupts: Clear the SPIxIF bit in the respective IFSn register. Set the SPIxIE bit in the respective IECn register. Write the SPIxIP bits in the respective IPCn register. Write the desired settings to the SPIxCON register with MSTEN (SPIxCON<5>) = 1. Clear the SPIROV bit (SPIxSTAT<6>). Enable SPI operation by setting the SPIEN bit (SPIxSTAT<15>). Write the data to be transmitted to the SPIxBUF register. Transmission (and Reception) will start as soon as data is written to the SPIxBUF register.
2. 3. 4. 5.
In Master mode, the system clock is prescaled and then used as the serial clock. The prescaling is based on the settings in the PPRE<1:0> (SPIxCON<1:0>) and SPRE<1:0> (SPIxCON<4:2>) bits. The serial clock is output via the SCKx pin to slave devices. Clock pulses are only generated when there is data to be transmitted. For further information, refer to Section 20.4 SPI Master Mode Clock Frequency. The CKP and CKE bits determine on which edge of the clock, data transmission occurs. Both data to be transmitted and data that is received are respectively written into or read from the SPIxBUF register. The following describes the SPI module operation in Master mode: 1. 2. 3. 4. Once the module is set up for Master mode of operation and enabled, data to be transmitted is written to the SPIxBUF register. The SPITBF (SPIxSTAT<1>) bit is set. The contents of SPIxTXB are moved to the shift register, SPIxSR, and the SPITBF bit is cleared by the module. A series of 8/16 clock pulses shifts out 8/16 bits of transmit data from the SPIxSR to the SDOx pin and simultaneously shifts in the data at the SDIx pin into the SPIxSR. When the transfer is complete, the following events will occur: The interrupt flag bit, SPIxIF, is set. SPI interrupts can be enabled by setting the interrupt enable bit SPIxIE. The SPIxIF flag is not cleared automatically by the hardware. Also, when the ongoing transmit and receive operation is completed, the contents of the SPIxSR are moved to the SPIxRXB register. The SPIRBF (SPIxSTAT<0>) bit is set by the module, indicating that the receive buffer is full. Once the SPIxBUF register is read by the user code, the hardware clears the SPIRBF bit. If the SPIRBF bit is set (receive buffer is full) when the SPI module needs to transfer data from SPIxSR to SPIxRXB, the module will set the SPIROV (SPIxSTAT<6>) bit, indicating an overflow condition. Data to be transmitted can be written to SPIxBUF by the user software at any time as long as the SPITBF (SPIxSTAT<1>) bit is clear. The write can occur while SPIxSR is shifting out the previously written data, allowing continuous transmission. Note: The SPIxSR register cannot be written into directly by the user. All writes to the SPIxSR register are performed through the SPIxBUF register.
5.
6.
DS70067C-page 20-8

Figure 20-3:
User writes to SPIxBUF SPIxTXB to SPIxSR SPITBF SCKx (CKP = 0 CKE = 0) SCKx (CKP = 1 CKE = 0) SCKx (CKP = 0 CKE = 1) SCKx (CKP = 1 CKE = 0) 4 Clock modes (clock output at the SCKx pin in Master mode)
SPI Master Mode Operation
User writes new data during transmission
SDOx (CKE = 0)
bit7
bit6
bit5
bit4
bit3
bit2
bit1
bit0
SDOx (CKE = 1)
bit7
bit6
bit5
bit4
bit3
bit2
bit1
bit0
SDIx (SMP = 0) Input Sample (SMP = 0) SDIx (SMP = 1) Input Sample (SMP = 1)
bit7
bit0 Two modes available for SMP control bit (see Note 4)
bit7
bit0
SPIxIF 1 instruction cycle latency to set SPIxIF flag bit SPIxSR moved into SPIxRXB SPIRBF (SPIxSTAT<0>) User reads SPIxBUF
20
Note 1: Four SPI Clock modes shown to demonstrate CKP (SPIxCON<6>) and CKE (SPIxCON<8>) bit functionality only. Only one of the four modes can be chosen for operation. 2: SDI and input sample shown for two different values of the SMP (SPIxCON<9>) bit, for demonstration purposes only. Only one of the two configurations of the SMP bit can be chosen during operation. 3: If there are no pending transmissions, SPIxTXB is transferred to SPIxSR as soon as the user writes to SPIxBUF. 4: Operation for 8-bit mode shown. The 16-bit mode is similar.
DS70067C-page 20-9

20.3.2.2 Slave Mode The following steps should be taken to set up the SPI module for the Slave mode of operation: 1. 2. Clear the SPIxBUF register. If using interrupts: Clear the SPIxIF bit in the respective IFSn register. Set the SPIxIE bit in the respective IECn register. Write the SPIxIP bits in the respective IPCn register. Write the desired settings to the SPIxCON register with MSTEN (SPIxCON<5>) = 0. Clear the SMP bit. If the CKE bit is set, then the SSEN bit must be set, thus enabling the SSx pin. Clear the SPIROV bit (SPIxSTAT<6>) and, Enable SPI operation by setting the SPIEN bit (SPIxSTAT<15>).
3. 4. 5. 6. 7.
In Slave mode, data is transmitted and received as the external clock pulses appear on the SCKx pin. The CKP (SPIxCON<6>) and CKE (SPIxCON<8>) bits determine on which edge of the clock data transmission occurs. Both data to be transmitted and data that is received are respectively written into or read from the SPIxBUF register. The rest of the operation of the module is identical to that in the Master mode. A few additional features provided in the Slave mode are: Slave Select Synchronization: The SSx pin allows a Synchronous Slave mode. If the SSEN (SPIxCON<7>) bit is set, transmission and reception is enabled in Slave mode only if the SSx pin is driven to a low state. The port output or other peripheral outputs must not be driven in order to allow the SSx pin to function as an input. If the SSEN bit is set and the SSx pin is driven high, the SDOx pin is no longer driven and will tri-state even if the module is in the middle of a transmission. An aborted transmission will be retried the next time the SSx pin is driven low using the data held in the SPIxTXB register. If the SSEN bit is not set, the SSx pin does not affect the module operation in Slave mode. SPITBF Status Flag Operation: The function of the SPITBF (SPIxSTAT<1>) bit is different in the Slave mode of operation. The following describes the function of the SPITBF for various settings of the Slave mode of operation: 1. If SSEN (SPIxCON<7>) is cleared, the SPITBF is set when the SPIxBUF is loaded by the user code. It is cleared when the module transfers SPIxTXB to SPIxSR. This is similar to the SPITBF bit function in Master mode. If SSEN (SPIxCON<7>) is set, the SPITBF is set when the SPIxBUF is loaded by the user code. However, it is cleared only when the SPIx module completes data transmission. A transmission will be aborted when the SSx pin goes high and may be retried at a later time. Each data word is held in SPIxTXB until all bits are transmitted to the receiver. Note: To meet module timing requirements, the SSx pin must be enabled in Slave mode when CKE = 1. (Refer to Figure 20-6 for details.)
2.
DS70067C-page 20-10

Figure 20-4: SPI Slave Mode Operation: Slave Select Pin Disabled
SCKx Input (CKP = 0 CKE = 0) SCKx Input (CKP = 1 CKE = 0) SDOx Output SDIx Input (SMP = 0) Input Sample (SMP = 0) bit7 bit6 bit5 bit4 bit3 bit2 bit1 bit0
bit7
bit0
User writes to SPIxBUF
SPITBF SPISR to SPIxRXB
SPIRBF
1 instruction cycle latency to set SPIxIF flag bit SPIxIF
Note 1: Two SPI Clock modes shown only to demonstrate CKP (SPIxCON<6>) and CKE (SPIxCON<8>) bit functionality. Any combination of CKP and CKE bits can be chosen for module operation. 2: If there are no pending transmissions or a transmission in progress, SPIxBUF is transferred to SPIxSR as soon as the user writes to SPIxBUF. 3: Operation for 8-bit mode shown. The 16-bit mode is similar.
20
DS70067C-page 20-11

Figure 20-5: SPI Slave Mode Operation with Slave Select Pin Enabled
SSx
SCKx (CKP = 0 CKE = 0) SCKx (CKP = 1 CKE = 0) User writes SPIxBUF to to SPIxBUF SPIxSR
SPITBF bit6 bit5 bit4 bit2 bit1 bit0
SDOx
bit7
bit3
SDIx (SMP = 0) Input Sample (SMP = 0)
bit7
bit0
SPIxIF 1 instruction cycle latency SPIxSR to SPIxBUF
SPIRBF User reads SPIxBUF Note 1: When the SSEN (SPIxCON<7>) bit is set to 1, the SSx pin must be driven low so as to enable transmission and reception in Slave mode. 2: Transmit data is held in SPIxTXB and SPITBF remains set until all bits are transmitted. 3: Operation for 8-bit mode shown. The 16-bit mode is similar.
DS70067C-page 20-12

Figure 20-6: SPI Mode Timing (Slave Mode w/CKE = 1)
SSx (see Note 1) SCK Input (CKP = 0 CKE = 1) SCK Input (CKP = 1 CKE = 1)
SDO Output SDI Input (SMP = 0) Input Sample (SMP = 0) SPIxIF
bit7
bit6
bit5
bit4
bit3
bit2
bit1
bit0
bit7
bit0
Write to SPIxBUF
SPISR to SPIRXB
SPITBF
SPIxRBF
Note 1: The SSx pin must be used for Slave mode operation when CKE = 1. 2: When the SSEN (SPIxCON<7>) bit is set to 1, the SSx pin must be driven low so as to enable transmission and reception in Slave mode. 3: Transmit data is held in SPIxTXB and SPITBF remains set until all bits are transmitted. 4: Operation for 8-bit mode shown. The 16-bit mode is similar.
20
DS70067C-page 20-13

20.3.3 SPI Error Handling
When a new data word has been shifted into SPIxSR and the previous contents of SPIxRXB have not been read by the user software, the SPIROV bit (SPIxSTAT<6>) will be set. The module will not transfer the received data from SPIxSR to SPIxRXB. Further data reception is disabled until the SPIROV bit is cleared. The SPIROV bit is not cleared automatically by the module and must be cleared by the user software.
20.3.4
SPI Receive Only Operation

Setting the control bit, DISSDO (SPIxCON<11>), disables transmission at the SDOx pin. This allows the SPIx module to be configured for a Receive Only mode of operation. The SDOx pin will be controlled by the respective port function if the DISSDO bit is set. The DISSDO function is applicable to all SPI Operating modes.
20.3.5
Framed SPI Modes

The module supports a very basic framed SPI protocol while operating in either Master or Slave modes. The following features are provided in the SPI module to support Framed SPI modes: The control bit, FRMEN (SPIxCON<14>), enables Framed SPI modes and causes the SSx pin to be used as a frame synchronization pulse input or output pin. The state of the SSEN (SPIxCON<7>) is ignored. The control bit, SPIFSD (SPIxCON<13>), determines whether the SSx pin is an input or an output (i.e., whether the module receives or generates the frame synchronization pulse). The frame synchronization pulse is an active high pulse for a single SPI clock cycle. The following two framed SPI modes are supported by the SPI module: Frame Master Mode: The SPI module generates the frame synchronization pulse and provides this pulse to other devices at the SSx pin. Frame Slave Mode: The SPI module uses a frame synchronization pulse received at the SSx pin. The Framed SPI modes are supported in conjunction with the Master and Slave modes. Thus, the following four framed SPI configurations are available to the user: SPI Master Mode and Frame Master Mode SPI Master Mode and Frame Slave Mode SPI Slave Mode and Frame Master Mode SPI Slave Mode and Frame Slave Mode
These four modes determine whether or not the SPIx module generates the serial clock and the frame synchronization pulse.
DS70067C-page 20-14

Figure 20-7: SPI Master, Frame Master Connection Diagram
dsPIC30F [SPI Master, Frame Master]
PROCESSOR 2
SDOx
SDIx
SDIx
SDOx
SCKx
Serial Clock
SCKx
SSx
Frame Sync. Pulse
SSx
Note 1: In Framed SPI modes, the SSx pin is used to transmit/receive the frame synchronization pulse. 2: Framed SPI modes require the use of all four pins (i.e., using the SSx pin is not optional). 3: The SPIxTXB and SPIxRXB registers are memory mapped to the SPIxBUF register.
20.3.5.1
SCKx in Framed SPI Modes When FRMEN (SPIxCON<14>) = 1 and MSTEN (SPIxCON<5>) = 1, the SCKx pin becomes an output and the SPI clock at SCKx becomes a free running clock. When FRMEN = 1 and MSTEN = 0, the SCKx pin becomes an input. The source clock provided to the SCKx pin is assumed to be a free running clock. The polarity of the clock is selected by the CKP (SPIxCON<6>) bit. The CKE (SPIxCON<8>) bit is not used for the Framed SPI modes and should be programmed to 0 by the user software. When CKP = 0, the frame sync pulse output and the SDOx data output change on the rising edge of the clock pulses at the SCKx pin. Input data is sampled at the SDIx input pin on the falling edge of the serial clock. When CKP = 1, the frame sync pulse output and the SDOx data output change on the falling edge of the clock pulses at the SCKx pin. Input data is sampled at the SDIx input pin on the rising edge of the serial clock.
20
DS70067C-page 20-15

20.3.5.2 SPIx Buffers in Framed SPI Modes When SPIFSD (SPIxCON<13>) = 0, the SPIx module is in the Frame Master mode of operation. In this mode, the frame sync pulse is initiated by the module when the user software writes the transmit data to SPIxBUF location (thus writing the SPIxTXB register with transmit data). At the end of the frame sync pulse, the SPIxTXB is transferred to the SPIxSR and data transmission/reception begins. When SPIFSD (SPIxCON<13>) = 1, the module is in Frame Slave mode. In this mode, the frame sync pulse is generated by an external source. When the module samples the frame sync pulse, it will transfer the contents of the SPIxTXB register to the SPIxSR and data transmission/ reception begins. The user must make sure that the correct data is loaded into the SPIxBUF for transmission before the frame sync pulse is received. Note: Receiving a frame sync pulse will start a transmission, regardless of whether data was written to SPIxBUF. If no write was performed, the old contents of SPIxTXB will be transmitted.
20.3.5.3
SPI Master Mode and Frame Master Mode This Framed SPI mode is enabled by setting the MSTEN (SPIxCON<5>) and FRMEN (SPIxCON<14>) bits to 1 and the SPIFSD (SPIxCON<13>) bit to 0. In this mode, the serial clock will be output continuously at the SCKx pin, regardless of whether the module is transmitting. When the SPIxBUF is written, the SSx pin will be driven high on the next transmit edge of the SCKx clock. The SSx pin will be high for one SCKx clock cycle. The module will start transmitting data on the next transmit edge of the SCKx, as shown in Figure 20-8. A connection diagram indicating signal directions for this Operating mode is shown in Figure 20-7. Figure 20-8:
SCKx (CKP = 1) SCKx (CKP = 0) SSx SDOx SDIx Bit 15 Bit 15 Bit 14 Bit 14 Bit 13 Bit 13 Bit 12 Bit 12
SPI Master, Frame Master
Write to SPIxBUF Pulse generated at SSx
Receive Samples at SDIx
DS70067C-page 20-16

20.3.5.4 SPI Master Mode and Frame Slave Mode This Framed SPI mode is enabled by setting the MSTEN, FRMEN and the SPIFSD bits to 1. The SSx pin is an input, and it is sampled on the sample edge of the SPI clock. When it is sampled high, data will be transmitted on the subsequent transmit edge of the SPI clock, as shown in Figure 20-9. The interrupt flag, SPIxIF, is set when the transmission is complete. The user must make sure that the correct data is loaded into the SPIxBUF for transmission before the signal is received at the SSx pin. A connection diagram indicating signal directions for this Operating mode is shown in Figure 20-10. Figure 20-9: SPI Master, Frame Slave
SCKx (CKP = 1) SCK (CKP = 0) FSYNC SDO SDI Bit 15 Bit 15 Bit 14 Bit 14 Bit 13 Bit 13 Bit 12 Bit 12
Write to SPIxBUF
Sample SSx pin for frame sync. pulse
Receive Samples at SDIx
Figure 20-10: SPI Master, Frame Slave Connection Diagram
dsPIC30F [SPI Master, Frame Slave] SDOx SDIx SCKx SSx Frame Sync. Pulse Serial Clock SDIx SDOx SCKx SSx
PROCESSOR 2
Note 1: In Framed SPI modes, the SSx pin is used to transmit/receive the frame synchronization pulse. 2: Framed SPI modes require the use of all four pins (i.e., Using the SSx pin is not optional).
20
DS70067C-page 20-17

20.3.5.5 SPI Slave Mode and Frame Master Mode This framed SPI mode is enabled by setting the MSTEN (SPIxCON<5>) bit to 0, the FRMEN (SPIxCON<14>) bit to 1 and the SPIFSD (SPIxCON<13>) bit to 0. The input SPI clock will be continuous in Slave mode. The SSx pin will be an output when the SPIFSD bit is low. Therefore, when the SPIBUF is written, the module will drive the SSx pin high on the next transmit edge of the SPI clock. The SSx pin will be driven high for one SPI clock cycle. Data will start transmitting on the next SPI clock transmit edge. A connection diagram indicating signal directions for this Operating mode is shown in Figure 20-11. Figure 20-11: SPI Slave, Frame Master Connection Diagram
PROCESSOR 2
dsPIC30F [SPI Slave, Frame Slave] SDOx SDIx
SDIx Serial Clock SCKx SSx Frame Sync. Pulse
SDOx SCKx SSx
20.3.5.6
SPI Slave Mode and Frame Slave Mode This Framed SPI mode is enabled by setting the MSTEN (SPIxCON<5>) bit to 0, the FRMEN bit (SPIxCON<14>) to 1 and the SPIFSD (SPIxCON<13>) bit to 1. Therefore, both the SCKx and SSx pins will be inputs. The SSx pin will be sampled on the sample edge of the SPI clock. When SSx is sampled high, data will be transmitted on the next transmit edge of SCKx. A connection diagram indicating signal directions for this Operating mode is shown in Figure 20-12. Figure 20-12: SPI Slave, Frame Slave Connection Diagram
dsPIC30F [SPI Master, Frame Slave] SDOx SDIx
PROCESSOR 2
SDIx Serial Clock SCKx SSx Frame Sync. Pulse
SDOx SCKx SSx
DS70067C-page 20-18

20.4 SPI Master Mode Clock Frequency
In the Master mode, the clock provided to the SPI module is the instruction cycle (TCY). This clock will then be prescaled by the primary prescaler (specified by PPRE<1:0> (SPIxCON<1:0>)), and the secondary prescaler (specified by SPRE<2:0> (SPIxCON<4:2>)). The prescaled instruction clock becomes the serial clock and is provided to external devices via the SCKx pin. Note: Note that the SCKx signal clock is not free running for normal SPI modes. It will only run for 8 or 16 pulses when the SPIxBUF is loaded with data. It will however, be continuous for Framed modes.
Equation 20-1 can be used to calculate the SCKx clock frequency as a function of the primary and secondary prescaler settings. Equation 20-1: FCY FSCK = Primary Prescaler * Secondary Prescaler
Some sample SPI clock frequencies (in kHz) are shown in the table below: Table 20-1: FCY = 30 MHz Primary Prescaler Settings 1:1 4:1 16:1 64:1 FCY = 5 MHz Primary Prescaler Settings 1:1 4:1 16:1 64:1 Note: 5000 1250 313 78 2500 625 156 39 1250 313 78 20 833 208 52 13 625 156 39 10 Sample SCKx Frequencies Secondary Prescaler Settings 1:1 30000 7500 1875 469 2:1 15000 3750 938 234 4:1 7500 1875 469 117 6:1 5000 1250 313 78 8:1 3750 938 234 59
SCKx frequencies shown in kHz.
Note:
Not all clock rates are supported. For further information, refer to the SPI timing specifications in the specific device data sheet.
20
DS70067C-page 20-19

20.5 Operation in Power Save Modes
The dsPIC30FXXXX family of devices has three Power modes: Operational mode: The core and peripherals are running. Power Save modes: These are invoked by the execution of the PWRSAV instruction. There are two Power Save modes supported in the dsPIC30F family of devices. These are specified in the PWRSAV instruction via a parameter. The two modes are: - Sleep mode: Device clock source and entire device is shut down. This is achieved by the following instruction. ;include device p30fxxxx.inc file PWRSAV #SLEEP_MODE - Idle mode: Device clock is operational, CPU and selected peripherals are shut down. ;include device p30fxxxx.inc file PWRSAV #IDLE_MODE
20.5.1
Sleep Mode
When the device enters Sleep mode, the system clock is disabled.
20.5.1.1
Master Mode Operation The following are a consequence of entering Sleep mode when the SPIx module is configured for master operation: The baud rate generator in the SPIx module stops and is reset. If the SPIx module enters Sleep mode in the middle of a transmission/reception, then the transmission/reception is aborted. Since there is no automatic way to prevent an entry into Sleep mode if a transmission or reception is pending, the user software must synchronize entry into Sleep with SPI module operation to avoid aborted transmissions. The transmitter and receiver will stop in Sleep. The transmitter or receiver does not continue with a partially completed transmission at wake-up.
20.5.1.2
Slave Mode Operation Since the clock pulses at SCKx are externally provided for Slave mode, the module will continue to function in Sleep mode. It will complete any transactions during the transition into Sleep. On completion of a transaction, the SPIRBF flag is set. Consequently, the SPIxIF bit will be set. If SPI interrupts are enabled (SPIxIE = 1), the device will wake from Sleep. If the SPI interrupt priority level is greater than the present CPU priority level, code execution will resume at the SPIx interrupt vector location. Otherwise, code execution will continue with the instruction following the PWRSAV instruction that previously invoked Sleep mode. The module is not reset on entering Sleep mode if it is operating as a slave device. Register contents are not affected when the SPIx module is going into or coming out of Sleep mode.
20.5.2
Idle Mode
When the device enters Idle mode, the system clock sources remain functional. The SPISIDL bit (SPIxSTAT<13>) selects whether the module will stop or continue functioning on Idle. If SPISIDL = 1, the SPI module will stop communication on entering Idle mode. It will operate in the same manner as it does in Sleep mode. If SPISID = 0 (default selection), the module will continue operation in Idle mode.
DS70067C-page 20-20

Table 20-2: Pin Name SCK1 SCK2 SDI1 SDI2 SDO1 SDO2 SS1 Pins Associated with the SPI Modules Pin Type I/O I/O I I O O I/O Buffer Type CMOS CMOS CMOS CMOS CMOS CMOS CMOS Description
SPI1 module Clock Input or Output SPI2 module Clock Input or Output SPI1 module Data Receive pin SPI2 module Data Receive pin SPI1 module Data Transmit pin SPI2 module Data Transmit pin SPI1 module Slave Select Control pin 1) Used to enable transmit/receive in Slave mode, if SSEN (SPI1CON<7>) has been set to 1 2) Used as Frame Sync I/O Pulse when FRMEN and SPIFSD (SPI1CON<14:13>) are set to 11 or 10. I/O CMOS SPI2 module Slave Select Control pin SS2 1) Used to enable transmit/receive in Slave mode, if SSEN (SPI2CON<7>) has been set to 1 2) Used as Frame Sync I/O Pulse when FRMEN and SPIFSD (SPI2CON<14:13>) are set to 11 or 10. Legend: CMOS = CMOS compatible input or output, ST = Schmitt Trigger input with CMOS levels, I = Input, O = Output
20
DS70067C-page 20-21
20.6
Special Function Registers Associated with SPI Modules
Table 20-3:
Bit 13 SPISIDL SPIFSD Transmit and Receive Buffer Address shared by SPI1TXB and SPI1RXB registers DISSDO MODE16 SMP CKE SSEN CKP MSTEN SPRE2 SPRE1 SPRE0 PPRE1 PPRE0 SPIROV SPITBF Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
SPI1 Register Map

Reset State
SFR Name
Addr.
Bit 15
Bit 14
SPI1STAT
0220
SPIEN
SPIRBF 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000
DS70067C-page 20-22
Bit 13 SPISIDL SPIFSD Transmit and Receive Buffer Address shared by SPI2TXB and SPI2RXB registers DISSDO MODE16 SMP CKE SSEN CKP MSTEN SPRE2 SPRE1 SPRE0 SPIROV Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Reset State SPITBF SPIRBF 0000 0000 0000 0000 PPRE1 PPRE0 0000 0000 0000 0000 0000 0000 0000 0000 Bit 13 SI2CIF IC4IF SI2CIE IC4IE SPI2IP<2:0> U1TXIP<2:0> C1IP<2:0> IC3IE C1IE SPI2IE U2TXIE U2RXIE INT2IE T5IE T4IE U1RXIP<2:0> U2TXIP<2:0> NVMIE ADIE U1TXIE U1RXIE SPI1IE T3IE T2IE OC2IE IC3IF C1IF SPI2IF U2TXIF U2RXIF INT2IF T5IF T4IF OC4IF IC2IE OC4IE NVMIF ADIF U1TXIF U1RXIF SPI1IF T3IF T2IF OC2IF IC2IF LEV8F INT4EP OVATE OVBTE COVTE SWTRAP OVRFLOW INT3EP T1IF OC3IF T1IE OC3IE Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 ADDRERR STKERR INT2EP OC1IF IC8IF OC1IE IC8IE INT1EP IC1IF IC7IF IC1IE IC7IE SPI1IP<2:0> U2RXIP<2:0> Bit 0 INT0 INT1IF INT0IE INT1IE Reset State 0000 0000 0000 0000
SPI1CON
0222
FRMEN
SPI1BUF
0224
Table 20-4:
SPI2 Register Map
SFR Name
Addr.
Bit 15
Bit 14
SPI2STAT
0226
SPIEN
SPI2CON
0228
FRMEN
SPI2BUF
022A
Table 20-5:
SPI Module Related Interrupt Registers
SFR Name
Addr.
Bit 15
Bit 14
INTCON1 0080 NSTDIS
INTCON2 0082
ALTIVT
DISI
INT0EP 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100 0100 0100 0100 0100
IFS0
0084
CNIF
MI2CIF
IFS1
0086
IC6IF
IC5IF
IEC0
008C
CNIE
MI2CIE
IEC1
008E
IC6IE
IC5IE
IPC2
0098
ADIP<2:0>
IPC6
00A0

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Serial Peripheral Interface (SPI) module are: Title Interfacing Microchips MCP41XXX/MCP42XXX Digital Potentiometers to a PICmicro Microcontroller Interfacing Microchips MCP3201 Analog-to-Digital Converter to the PICmicro Microcontroller Application Note # AN746 AN719
Note:
20
DS70067C-page 20-23

Revision B
This revision reflects editorial and technical content changes for the dsPIC30F Serial Peripheral Interface (SPI) module.
Revision C
DS70067C-page 20-24
21
Inter-Integrated Circuit (I2C)
Section 21. Inter-Integrated Circuit (I2C)

HIGHLIGHTS
This section of the manual contains the following major topics: 21.1 Overview ...................................................................................................................... 21-2 21.2 I2C Bus Characteristics................................................................................................ 21-4 21.3 Control and Status Registers ....................................................................................... 21-7 21.4 Enabling I2C Operation .............................................................................................. 21-13 21.5 Communicating as a Master in a Single Master Environment ................................... 21-15 21.6 Communicating as a Master in a Multi-Master Environment ..................................... 21-29 21.7 Communicating as a Slave ........................................................................................ 21-32 21.8 Connection Considerations for I2C Bus ..................................................................... 21-47 21.9 Module Operation During PWRSAV Instruction......................................................... 21-49 21.10 Effects of a Reset....................................................................................................... 21-49 21.11 Design Tips ................................................................................................................ 21-50 21.12 Related Application Notes.......................................................................................... 21-51 21.13 Revision History ......................................................................................................... 21-52
DS70068D-page 21-1

21.1 Overview
The Inter-Integrated Circuit (I 2C) module is a serial interface useful for communicating with other peripheral or microcontroller devices. These peripheral devices may be serial EEPROMs, display drivers, A/D converters, etc. The I 2C module can operate in any of the following I 2C systems: Where the dsPIC30F acts as a Slave Device Where the dsPIC30F acts as a Master Device in a Single Master System (Slave may also be active) Where the dsPIC30F acts as a Master/Slave Device in a Multi-Master System (Bus collision detection and arbitration available)
The I 2C module contains independent I 2C master logic and I 2C slave logic, each generating interrupts based on their events. In multi-master systems, the software is simply partitioned into master controller and slave controller. When the I 2C master logic is active, the slave logic remains active also, detecting the state of the bus and potentially receiving messages from itself in a single master system or from other masters in a multi-master system. No messages are lost during multi-master bus arbitration. In a multi-master system, bus collision conflicts with other masters in the system are detected and the module provides a method to terminate then restart the message. The I 2C module contains a baud rate generator. The I 2C baud rate generator does not consume other timer resources in the device.
21.1.1
Module Features
Independent Master and Slave logic Multi-Master support. No messages lost in arbitration. Detects 7-bit and 10-bit device addresses Detects general call addresses as defined in the I 2C protocol Bus Repeater mode. Accept all messages as a slave regardless of the address. Automatic SCL clock stretching provides delays for the processor to respond to a slave data request. Supports 100 kHz and 400 kHz bus specifications. Figure 21-1 shows the I2C module block diagram.
DS70068D-page 21-2
21
Figure 21-1: I2C Block Diagram
Internal Data Bus
I2CRCV Read SCL Shift Clock I2CRSR LSB SDA Match Detect Address_Match Write I2CADD Read Start and Stop bit Detect Write Start and Stop bit Generate Control Logic I2CSTAT
Read
Collision Detect
Write I2CCON
Acknowledge Generation Clock Stretching I2CTRN Shift Clock Reload Control I2CBRG BRG Down Counter TCY/2 LSB
Read
Write
Read
Write
Read
DS70068D-page 21-3

21.2 I2C Bus Characteristics
The I 2C bus is a two-wire serial interface. Figure 21-2 is a schematic of a typical I 2C connection between the dsPIC30F device and a 24LC256 I 2C serial EEPROM. The I 2C interface employs a comprehensive protocol to ensure reliable transmission and reception of data. When communicating, one device is the master which initiates transfer on the bus and generates the clock signals to permit that transfer, while the other device(s) acts as the slave responding to the transfer. The clock line, SCL, is output from the master and input to the slave, although occasionally the slave drives the SCL line. The data line, SDA, may be output and input from both the master and slave. Because the SDA and SCL lines are bidirectional, the output stages of the devices driving the SDA and SCL lines must have an open drain in order to perform the wired-AND function of the bus. External pull-up resistors are used to ensure a high level when no device is pulling the line down. In the I 2C interface protocol, each device has an address. When a master wishes to initiate a data transfer, it first transmits the address of the device that it wishes to talk to. All devices listen to see if this is their address. Within this address, bit 0 specifies if the master wishes to read from or write to the slave device. The master and slave are always in opposite modes (transmitter/receiver) of operation during a data transfer. That is, they can be thought of as operating in either of these two relations: Master-transmitter and Slave-receiver Slave-transmitter and Master-receiver In both cases, the master originates the SCL clock signal. Figure 21-2: Typical I2C Interconnection Block Diagram
VDD VDD 5 k SCL MCLR SDA VDD 4.7 F 0.1 F VSS SDA WP A0 A1 A2 OSC1 OSC2 VSS VDD
dsPIC30F VDD
24LC256 SCL VDD
XTAL
DS70068D-page 21-4
21
21.2.1 Bus Protocol
The following I2C bus protocol has been defined: Data transfer may be initiated only when the bus is not busy. During data transfer, the data line must remain stable whenever the SCL clock line is HIGH. Changes in the data line while the SCL clock line is HIGH will be interpreted as a Start or Stop condition. Accordingly, the following bus conditions have been defined (Figure 21-3). 21.2.1.1 Start Data Transfer (S) After a bus Idle state, a HIGH-to-LOW transition of the SDA line while the clock (SCL) is HIGH determines a Start condition. All data transfers must be preceded by a Start condition. 21.2.1.2 Stop Data Transfer (P) A LOW-to-HIGH transition of the SDA line while the clock (SCL) is HIGH determines a Stop condition. All data transfers must end with a Stop condition. 21.2.1.3 Repeated Start (R) After a WAIT state, a HIGH-to-LOW transition of the SDA line while the clock (SCL) is HIGH determines a Repeated Start condition. Repeated Starts allow a master to change bus direction without relinquishing control of the bus. 21.2.1.4 Data Valid (D) The state of the SDA line represents valid data when, after a Start condition, the SDA line is stable for the duration of the HIGH period of the clock signal. There is one bit of data per SCL clock. 21.2.1.5 Acknowledge (A) or Not-Acknowledge (N) All data byte transmissions must be Acknowledged (ACK) or Not Acknowledged (NACK) by the receiver. The receiver will pull the SDA line low for an ACK or release the SDA line for a NACK. The Acknowledge is a one-bit period, using one SCL clock. 21.2.1.6 WAIT/Data Invalid (Q) The data on the line must be changed during the LOW period of the clock signal. Devices may also stretch the clock low time, by asserting a low on SCL line, causing a WAIT on the bus. 21.2.1.7 Bus Idle (I) Both data and clock lines remain HIGH at those times after a Stop condition and before a Start condition. Figure 21-3:
(I) SCL
I2C Bus Protocol States

(S) (D) (Q) (A) or (N) (P) (I)
SDA
NACK ACK Start Condition Data or Address Valid Data Allowed to Change
ACK/NACK Valid
Stop Condition
DS70068D-page 21-5

21.2.2 Message Protocol
A typical I 2C message is shown in Figure 21-4. In this example, the message will read a specified byte from a 24LC256 I 2C serial EEPROM. The dsPIC30F device will act as the master and the 24LC256 device will act as the slave. Figure 21-4 indicates the data as driven by the master device and the data as driven by the slave device, remembering that the combined SDA line is a wired-AND of the master and slave data. The master device controls and sequences the protocol. The slave device will only drive the bus at specifically determined times. Figure 21-4: A Typical I2C Message: Read of Serial EEPROM (Random Address Mode)
R E S T A A CR K T
Bus Activity Master SDA Output Slave SDA Output
S I T D A L R E T
Address Byte
RA / C WK
EE ADDR High Byte
A C K
EE ADDR Low Byte
Address Byte
RA / C WK
Data Byte
N A C K
S T O P
I D L E
S1 01 0 AAA0 2 1 0
R 1 0 1 0 A A A1 2 1 0
NP
21.2.2.1
Start Message Each message is initiated with a Start condition and terminated with a Stop condition. The number of the data bytes transferred between the Start and Stop conditions is determined by the master device. As defined by the system protocol, the bytes of the message may have special meaning such as device address byte or data byte.
21.2.2.2
Address Slave In the figure, the first byte is the device address byte that must be the first part of any I 2C message. It contains a device address and a R/W bit. Refer to Section 26. Appendix for additional information on Address Byte formats. Note that R/W = 0 for this first address byte, indicating that the master will be a transmitter and the slave will be a receiver.
21.2.2.3
Slave Acknowledge The receiving device is obliged to generate an Acknowledge signal, ACK, after the reception of each byte. The master device must generate an extra SCL clock, which is associated with this Acknowledge bit.
21.2.2.4
Master Transmit The next 2 bytes, sent by the master to the slave, are data bytes containing the location of the requested EEPROM data byte. The slave must Acknowledge each of the data bytes.
21.2.2.5
Repeated Start At this point, the slave EEPROM has the address information necessary to return the requested data byte to the master. However, the R/W bit from the first device address byte specified master transmission and slave reception. The bus must be turned in the other direction for the slave to send data to the master. To do this function without ending the message, the master sends a Repeated Start. The Repeated Start is followed with a device address byte containing the same device address as before and with the R/W = 1 to indicate slave transmission and master reception.
DS70068D-page 21-6
21
21.2.2.6 Slave Reply Now the slave transmits the data byte driving the SDA line, while the master continues to originate clocks but releases its SDA drive. 21.2.2.7 Master Acknowledge During reads, a master must terminate data requests to the slave by NOT Acknowledging (generate a NACK) on the last byte of the message. 21.2.2.8 Stop Message The master sends Stop to terminate the message and return the bus to an Idle state.
21.3
Control and Status Registers

The I2C module has six user-accessible registers for I2C operation. The registers are accessible in either Byte or Word mode. The registers are shown in Figure 21-5 and listed below: Control Register (I2CCON): This register allows control of the I2C operation. Status Register (I2CSTAT): This register contains status flags indicating the module state during I2C operation. Receive Buffer Register (I2CRCV): This is the buffer register from which data bytes can be read. The I2CRCV register is a read only register. Transmit Register (I2CTRN): This is the transmit register; bytes are written to this register during a transmit operation. The I2CTRN register is a read/write register. Address Register (I2CADD): The I2CADD register holds the slave device address. Baud Rate Generator Reload Register (I2CBRG): Holds the baud rate generator reload value for the I2C module baud rate generator. Figure 21-5: I2C Programmers Model I2CRCV (8 bits) Bit 7 Bit 7 Bit 8 Bit 15 Bit 15 Bit 9 Bit 0 I2CTRN (8 bits) Bit 0 I2CBRG (9 bits) Bit 0 I2CCON (16 bits) Bit 0 I2CSTAT (16 bits) Bit 0 I2CADD (10 bits) Bit 0
DS70068D-page 21-7

Register 21-1 and Register 21-2 define the I2C module Control and Status registers, I2CCON and I2CSTAT. The I2CTRN is the register to which transmit data is written. This register is used when the module operates as a master transmitting data to the slave or as a slave sending reply data to the master. As the message progresses, the I2CTRN register shifts out the individual bits. Because of this, the I2CTRN may not be written to unless the bus is Idle. The I2CTRN may be reloaded while the current data is transmitting. Data being received by either the master or the slave is shifted into a non-accessible Shift register called I2CRSR. When a complete byte is received, the byte transfers to the I2CRCV register. In receive operations, the I2CRSR and I2CRCV create a double-buffered receiver. This allows reception of the next byte to begin before reading the current byte of received data. If the module receives another complete byte before the software reads the previous byte from the I2CRCV register, a receiver overflow occurs and sets the I2COV (I2CCON<6>). The byte in the I2CRSR is lost. The I2CADD register holds the slave device address. In 10-bit mode, all bits are relevant. In 7-bit addressing mode, only I2CADD<6:0> are relevant. The A10M (I2CCON<10>) specifies the expected mode of the slave address.
DS70068D-page 21-8
21
Register 21-1: Upper Byte: R/W-0 I2CEN bit 15 I2CCON: I2C Control Register U-0 R/W-0 I2CSIDL R/W-1 HC SCLREL R/W-0 IPMIEN R/W-0 A10M R/W-0 DISSLW R/W-0 SMEN bit 8
Lower Byte: R/W-0 GCEN bit 7 bit 15
R/W-0 STREN
R/W-0 ACKDT
R/W-0 HC ACKEN
R/W-0 HC RCEN
R/W-0 HC PEN
R/W-0 HC RSEN
R/W-0 HC SEN bit 0
I2CEN: I2C Enable bit 1 = Enables the I2C module and configures the SDA and SCL pins as serial port pins 0 = Disables I2C module. All I2C pins are controlled by port functions. Unimplemented: Read as 0 I2CSIDL: Stop in Idle Mode bit 1 = Discontinue module operation when device enters an Idle mode 0 = Continue module operation in Idle mode SCLREL: SCL Release Control bit (when operating as I2C Slave) 1 = Release SCL clock 0 = Hold SCL clock low (clock stretch) If STREN = 1: Bit is R/W (i.e., software may write 0 to initiate stretch and write 1 to release clock) Hardware clear at beginning of slave transmission. Hardware clear at end of slave reception. If STREN = 0: Bit is R/S (i.e., software may only write 1 to release clock) Hardware clear at beginning of slave transmission. IPMIEN: Intelligent Peripheral Management Interface (IPMI) Enable bit 1 = Enable IPMI Support mode. All addresses Acknowledged. 0 = IPMI mode not enabled A10M: 10-bit Slave Address bit 1 = I2CADD is a 10-bit slave address 0 = I2CADD is a 7-bit slave address DISSLW: Disable Slew Rate Control bit 1 = Slew rate control disabled 0 = Slew rate control enabled SMEN: SMBus Input Levels bit 1 = Enable I/O pin thresholds compliant with SMBus specification 0 = Disable SMBus input thresholds GCEN: General Call Enable bit (when operating as I2C slave) 1 = Enable interrupt when a general call address is received in the I2CRSR (module is enabled for reception) 0 = General call address disabled STREN: SCL Clock Stretch Enable bit (when operating as I2C slave) Used in conjunction with SCLREL bit. 1 = Enable software or receive clock stretching 0 = Disable software or receive clock stretching ACKDT: Acknowledge Data bit (When operating as I2C Master. Applicable during master receive.) Value that will be transmitted when the software initiates an Acknowledge sequence. 1 = Send NACK during acknowledge 0 = Send ACK during acknowledge
bit 14 bit 13
bit 12
bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
DS70068D-page 21-9

Register 21-1: bit 4 I2CCON: I2C Control Register (Continued)
ACKEN: Acknowledge Sequence Enable bit (When operating as I2C master. Applicable during master receive.) 1 = Initiate Acknowledge sequence on SDA and SCL pins, and transmit ACKDT data bit Hardware clear at end of master Acknowledge sequence. 0 = Acknowledge sequence not in progress RCEN: Receive Enable bit (when operating as I2C master) 1 = Enables Receive mode for I2C Hardware clear at end eighth bit of master receive data byte. 0 = Receive sequence not in progress PEN: Stop Condition Enable bit (when operating as I2C master) 1 = Initiate Stop condition on SDA and SCL pins Hardware clear at end of master Stop sequence. 0 = Stop condition not in progress RSEN: Repeated Start Condition Enabled bit (when operating as I2C master) 1 = Initiate Repeated Start condition on SDA and SCL pins Hardware clear at end of master Repeated Start sequence. 0 = Repeated Start condition not in progress SEN: Start Condition Enabled bit (when operating as I2C master) 1 = Initiate Start condition on SDA and SCL pins Hardware clear at end of master Start sequence. 0 = Start condition not in progress
bit 3
bit 2
bit 1
bit 0
Legend: R = Readable W = Writable HC = Cleared by Hardware 1 = Bit is set at POR C = Clearable bit HS = Set by Hardware 0 = Bit cleared at POR U = Unimplemented bit, read as 0 S = Settable bit x = Bit is unknown at POR
DS70068D-page 21-10
21
Register 21-2: I2CSTAT: I2C Status Register U-0 U-0 U-0 R/C-0 HS BCL R-0 HS, HC GCSTAT R-0 HS, HC ADD10 bit 8
Upper Byte: R-0 R-0 HS, HC HS, HC ACKSTAT TRSTAT bit 15
Lower Byte: R/C-0 R/W-0 HS HS IWCOL bit 7 bit 15 I2COV
R-0 HS, HC D_A
R/C-0 HS, HC P
R/C-0 HS, HC S
R-0 HS, HC R_W
R-0 HS, HC RBF
R-0 HS, HC TBF bit 0
ACKSTAT: Acknowledge Status bit (When operating as I2C master. Applicable to master transmit operation.) 1 = NACK received from slave 0 = ACK received from slave Hardware set or clear at end of slave Acknowledge. TRSTAT: Transmit Status bit (When operating as I2C master. Applicable to master transmit operation.) 1 = Master transmit is in progress (8 bits + ACK) 0 = Master transmit is not in progress Hardware set at beginning of master transmission. Hardware clear at end of slave Acknowledge. BCL: Master Bus Collision Detect bit 1 = A bus collision has been detected during a master operation 0 = No collision Hardware set at detection of bus collision. GCSTAT: General Call Status bit 1 = General call address was received 0 = General call address was not received Hardware set when address matches general call address. Hardware clear at Stop detection. ADD10: 10-bit Address Status bit 1 = 10-bit address was matched 0 = 10-bit address was not matched Hardware set at match of 2nd byte of matched 10-bit address. Hardware clear at Stop detection. IWCOL: Write Collision Detect bit 1 = An attempt to write the I2CTRN register failed because the I2C module is busy 0 = No collision Hardware set at occurrence of write to I2CTRN while busy (cleared by software). I2COV: Receive Overflow Flag bit 1 = A byte was received while the I2CRCV register is still holding the previous byte 0 = No overflow Hardware set at attempt to transfer I2CRSR to I2CRCV (cleared by software). D_A: Data/Address bit (when operating as I2C slave) 1 = Indicates that the last byte received was data 0 = Indicates that the last byte received was device address Hardware clear at device address match. Hardware set by write to I2CTRN or by reception of slave byte.
bit 14
bit 9
bit 8
bit 7
bit 6
bit 5
DS70068D-page 21-11

Register 21-2: bit 4 I2CSTAT: I2C Status Register (Continued)
P: Stop bit 1 = Indicates that a Stop bit has been detected last 0 = Stop bit was not detected last Hardware set or clear when Start, Repeated Start or Stop detected. S: Start bit 1 = Indicates that a Start (or Repeated Start) bit has been detected last 0 = Start bit was not detected last Hardware set or clear when Start, Repeated Start or Stop detected. R_W: Read/Write bit Information (when operating as I2C slave) 1 = Read - indicates data transfer is output from slave 0 = Write - indicates data transfer is input to slave Hardware set or clear after reception of I 2C device address byte. RBF: Receive Buffer Full Status bit 1 = Receive complete, I2CRCV is full 0 = Receive not complete, I2CRCV is empty Hardware set when I2CRCV written with received byte. Hardware clear when software reads I2CRCV. TBF: Transmit Buffer Full Status bit 1 = Transmit in progress, I2CTRN is full 0 = Transmit complete, I2CTRN is empty Hardware set when software writes I2CTRN. Hardware clear at completion of data transmission.
bit 3
bit 2
bit 1
bit 0
Legend: R = Readable HC = Cleared by Hardware 1 = Bit is set at POR W = Writable HS = Set by Hardware 0 = Bit cleared at POR C = Clearable bit U = Unimplemented bit, read as 0 x = Bit is unknown at POR
DS70068D-page 21-12
21
21.4 Enabling I 2C Operation
The module is enabled by setting the I2CEN (I2CCON<15>) bit. The I 2C module fully implements all master and slave functions. When the module is enabled, the master and slave functions are active simultaneously and will respond according to the software or the bus events. When initially enabled, the module will release SDA and SCL pins, putting the bus into the Idle state. The master functions will remain in the Idle state unless software sets a control bit to initiate a master event. The slave functions will begin to monitor the bus. If the slave logic detects a Start event and a valid address on the bus, the slave logic will begin a slave transaction.
21.4.1
Enabling I 2C I/O
Two pins are used for bus operation. These are the SCL pin, which is the clock, and the SDA pin, which is the data. When the module is enabled, assuming no other module with higher priority has control, the module will assume control of the SDA and SCL pins. The module software need not be concerned with the state of the port I/O of the pins, the module overrides the port state and direction. At initialization, the pins are tri-state (released).
21.4.2
I 2C Interrupts
The I 2C module generates two interrupts. One interrupt is assigned to master events and the other interrupt is assigned to slave events. These interrupts will set a corresponding interrupt flag bit and will interrupt the software process if the corresponding interrupt enable bit is set and the corresponding interrupt priority is high enough. The master interrupt is called MI2CIF and is activated on completion of a master message event. The following events generate the MI2CIF interrupt. Start condition Stop condition Data transfer byte transmitted/received Acknowledge transmit Repeated Start Detection of a bus collision event
The slave interrupt is called SI2CIF and is activated on detection of a message directed to the slave. Detection of a valid device address (including general call) Request to transmit data Reception of data
DS70068D-page 21-13

21.4.3 Setting Baud Rate when Operating as a Bus Master
When operating as an I2C master, the module must generate the system SCL clock. Generally, I 2C system clocks are specified to be either 100 kHz, 400 kHz or 1 MHz. The system clock rate is specified as the minimum SCL low time plus the minimum SCL high time. In most cases, that is defined by 2 TBRG intervals. The reload value for the baud rate generator is the I2CBRG register, as shown in Figure 21-6. When the baud rate generator is loaded with this value, the generator counts down to 0 and stops until another reload has taken place. The generator count is decremented twice per instruction cycle (TCY). The baud rate generator is reloaded automatically on baud rate restart. For example, if clock synchronization is taking place, the baud rate generator will be reloaded when the SCL pin is sampled high. Note: I2CBRG value of 0x0 is not supported. To compute the baud rate generator reload value, use the following equation. Equation 21-1: I2CBRG = FCY ( FSCL FCY 1,111,111
Table 21-1: Required System FSCL 100 kHz 100 kHz 100 kHz 400 kHz 400 kHz 400 kHz 1 MHz* 1 MHz
I2C Clock Rates FCY 30 MHz 20 MHz 1 MHz 10 MHz 5 MHz 1 MHz 11 MHz 1 MHz I2CBRG Decimal 272 181 8 15 7 1 1 0 I2CBRG HEX 0x110 0x0B5 0x008 0x00F 0x007 0x001 0x001 0x000 (invalid) Actual FSCL 100 kHz 100 kHz 101 kHz 400 kHz 400 kHz 345 kHz** 1 MHz* 1 MHz
*FCY = 11 MHz is the minimum input clock frequency to have FSCL = 1 MHz. ** This is closest value to 400 kHz for this value of FCY. Figure 21-6: Baud Rate Generator Block Diagram
I2CBRG<8:0>
SCL
Reload Control CLK
Reload
Down Counter
2 TCY
DS70068D-page 21-14
21
21.5 Communicating as a Master in a Single Master Environment
Typical operation of the I 2C module in a system is using the I 2C to communicate with an I 2C peripheral, such as an I 2C serial memory. In an I 2C system, the master controls the sequence of all data communication on the bus. In this example, the dsPIC30F and its I 2C module have the role of the single master in the system. As the single master, it is responsible for generating the SCL clock and controlling the message protocol. In the I 2C module, the module controls individual portions of the I 2C message protocol, however, sequencing of the components of the protocol to construct a complete message is a software task. For example, a typical operation in a single master environment may be to read a byte from an I 2C serial EEPROM. This example message is depicted in Figure 21-7. To accomplish this message, the software will sequence through the following steps. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. Figure 21-7: Assert a Start condition on SDA and SCL. Send the I 2C device address byte to the slave with a write indication. Wait for and verify an Acknowledge from the slave. Send the serial memory address high byte to the slave. Wait for and verify an Acknowledge from the slave. Send the serial memory address low byte to the slave. Wait for and verify an Acknowledge from the slave. Assert a Repeated Start condition on SDA and SCL. Send the device address byte to the slave with a read indication. Wait for and verify an Acknowledge from the slave. Enable master reception to receive serial memory data. Generate an ACK or NACK condition at the end of a received byte of data. Generate a Stop condition on SDA and SCL.
A Typical I2C Message: Read Of Serial EEPROM (Random Address Mode)

R E S T A A CR K T
I D L E
S T A R T
Address Byte
RA / C WK
EE ADDR High Byte
A C K
EE ADDR Low Byte
Address Byte
RA / C WK
Data Byte
N A C K
S T O P
I D L E
S10 1 0AAA0 2 1 0
R 1 0 1 0 A A A1 2 1 0
NP
The I 2C module supports Master mode communication with the inclusion of Start and Stop generators, data byte transmission, data byte reception, Acknowledge generator and a baud rate generator. Generally, the software will write to a control register to start a particular step, then wait for an interrupt or poll status to wait for completion. Subsequent sub-sections detail each of these operations
DS70068D-page 21-15

21.5.1 Generating Start Bus Event
To initiate a Start event, the software sets the Start enable bit, SEN (I2CCON<0>). Prior to setting the Start bit, the software can check the P (I2CSTAT<4>) status bit to ensure that the bus is in an Idle state. Figure 21-8 shows the timing of the Start condition. Slave logic detects the Start condition, sets the S bit (I2CSTAT<3>) and clears the P bit (I2CSTAT<4>). SEN bit is automatically cleared at completion of the Start condition. MI2CIF interrupt generated at completion of the Start condition. After Start condition, SDA line and SCL line are left low (Q state). 21.5.1.1 IWCOL Status Flag If the software writes the I2CTRN when a Start sequence is in progress, then IWCOL is set and the contents of the transmit buffer are ignored. Note: Because queueing of events is not allowed, writing to the lower 5 bits of I2CCON is disabled until the Start condition is complete.
Figure 21-8:
Master Start Timing Diagram
SEN I2C Bus State (I) TBRG SCL (Master) SDA (Master) S P
MI2CIF Interrupt
(S) TBRG
(Q)
1 - Writing SEN = 1 initiates a master Start event. Baud generator starts. 2 - Baud generator times out. Master module drives SDA low. Baud generator restarts. 3 - Slave module detects Start, sets S = 1, P = 0. 4 - Baud generator times out. Master module drives SCL low, generates interrupt and clears SEN.
2 3
DS70068D-page 21-16
21
21.5.2 Sending Data to a Slave Device
Transmission of a data byte, a 7-bit device address byte or the second byte of a 10-bit address, is accomplished by simply writing the appropriate value to the I2CTRN register. Loading this register will start the following process: The software loads the I2CTRN with the data byte to transmit. Writing I2CTRN sets the buffer full flag bit, TBF (I2CSTAT<0>). The data byte is shifted out the SDA pin until all 8 bits are transmitted. Each bit of address/data will be shifted out onto the SDA pin after the falling edge of SCL. On the ninth SCL clock, the module shifts in the ACK bit from the slave device and writes its value into the ACKSTAT bit (I2CCON<15>). The module generates the MI2CIF interrupt at the end of the ninth SCL clock cycle. Note that the module does not generate or validate the data bytes. The contents and usage of the byte is dependant on the state of the message protocol maintained by the software. 21.5.2.1 Sending a 7-bit Address to the Slave Sending a 7-bit device address involves sending 1 byte to the slave. A 7-bit address byte must contain the 7 bits of I 2C device address and a R/W bit that defines if the message will be a write to the slave (master transmission and slave receiver) or a read from the slave (slave transmission and master receiver). 21.5.2.2 Sending a 10-bit Address to the Slave Sending a 10-bit device address involves sending 2 bytes to the slave. The first byte contains 5 bits of I 2C device address reserved for 10-bit Addressing modes and 2 bits of the 10-bit address. Because the next byte, which contains the remaining 8 bits of the 10-bit address must be received by the slave, the R/W bit in the first byte must be 0, indicating master transmission and slave reception. If the message data is also directed toward the slave, the master can continue sending the data. However, if the master expects a reply from the slave, a Repeated Start sequence with the R/W bit at 1 will change the R/W state of the message to a read of the slave. 21.5.2.3 Receiving Acknowledge from the Slave On the falling edge of the eighth SCL clock, the TBF bit is cleared and the master will de-assert the SDA pin allowing the slave to respond with an Acknowledge. The master will then generate a ninth SCL clock. This allows the slave device being addressed to respond with an ACK bit during the ninth bit time if an address match occurs, or if data was received properly. A slave sends an Acknowledge when it has recognized its device address (including a general call), or when the slave has properly received its data. The status of ACK is written into the Acknowledge status bit, ACKSTAT (I2CSTAT<15>), on the falling edge of the ninth SCL clock. After the ninth SCL clock, the module generates the MI2CIF interrupt and enters an Idle state until the next data byte is loaded into I2CTRN. 21.5.2.4 ACKSTAT Status Flag The ACKSTAT bit (I2CCON<15>) is cleared when the slave has sent an Acknowledge (ACK = 0), and is set when the slave does not Acknowledge (ACK = 1).
DS70068D-page 21-17

21.5.2.5 TBF Status Flag When transmitting, the TBF bit (I2CSTAT<0>) is set when the CPU writes to I2CTRN and is cleared when all 8 bits are shifted out. 21.5.2.6 IWCOL Status Flag If the software writes the I2CTRN when a transmit is already in progress (i.e., the module is still shifting out a data byte), then IWCOL is set and the contents of the buffer are ignored. IWCOL must be cleared in software. Note: Because queueing of events is not allowed, writing to the lower 5 bits of I2CCON is disabled until the transmit condition is complete.
Figure 21-9:
Master Transmission Timing Diagram
I2CTRN I2C Bus State (Q) (D) (Q) (D) (Q) (A) (Q)
TBRG TBRG
SCL (Master) SCL (Slave) SDA (Master) SDA (Slave) TRSTAT TBF MI2CIF Interrupt ACKSTAT D7 D6 D5 D4 D3 D2 D1 D0
1 2
1 - Writing the I2CTRN register will start a master transmission event. TBF bit is set. 2 - Baud generator starts. The MSB of the I2CTRN drives SDA. SCL remains low. TRSTAT bit is set. 3 - Baud generator times out. SCL released. Baud generator restarts. 4 - Baud generator times out. SCL driven low. After SCL detected low, next bit of I2CTRN drives SDA. 5 - While SCL is low, the slave can also pull SCL low to initiate a WAIT (clock stretch). 6 - Master has already released SCL, and slave can release to end WAIT. Baud generator restarts. 7 - At falling edge of 8th SCL clock, master releases SDA. TBF bit is cleared. Slave drives ACK/NACK. 8 - At falling edge of 9th SCL clock, master generates interrupt. SCL remains low until next event. Slave releases SDA. TRSTAT bit is clear.
DS70068D-page 21-18
21
21.5.3 Receiving Data from a Slave Device
Setting the receive enable bit, RCEN (I2CCON<3>), enables the master to receive data from a slave device. Note: The lower 5 bits of I2CCON must be 0 before attempting to set the RCEN bit. This ensures the master logic is inactive.
The master logic begins to generate clocks and before each falling edge of the SCL, SDA line is sampled and data is shifted into the I2CRSR. After the falling edge of the eighth SCL clock: The RCEN bit is automatically cleared. The contents of the I2CRSR transfer into the I2CRCV. The RBF flag bit is set. The module generates the MI2CIF interrupt.
When the CPU reads the buffer, the RBF flag bit is automatically cleared. The software can process the data and then do an Acknowledge sequence. 21.5.3.1 RBF Status Flag When receiving data, the RBF bit is set when an device address or data byte is loaded into I2CRCV from I2CRSR. It is cleared when software reads the I2CRCV register. 21.5.3.2 I2COV Status Flag If another byte is received in the I2CRSR while the RBF bit remains set and the previous byte remains in the I2CRCV register, the I2COV bit is set and the data in the I2CRSR is lost. Leaving I2COV set does not inhibit further reception. If RBF is cleared by reading the I2CRCV, and the I2CRSR receives another byte, that byte will be transferred to the I2CRCV. 21.5.3.3 IWCOL Status Flag If the software writes the I2CTRN when a receive is already in progress (i.e., I2CRSR is still shifting in a data byte), then the IWCOL bit is set and the contents of the buffer are ignored. Note: Since queueing of events is not allowed, writing to the lower 5 bits of I2CCON is disabled until the data reception condition is complete.
DS70068D-page 21-19

Figure 21-10: Master Reception Timing Diagram
RCEN I2C Bus State (Q) (Q)

TBRG
(D) (Q)
TBRG
(Q) (D)
(Q)
SCL (Master) SCL (Slave) SDA (Master) SDA (Slave) I2CRCV RBF MI2CIF Interrupt D7 D6 D5 D4 D3 D2 D1 D0
1 - Typically, the slave can pull SCL low (clock stretch) to request a wait to prepare data response. The slave will drive MSB of data response on SDA when ready. 2 - Writing the RCEN bit will start a master reception event. The baud generator starts. SCL remains low. 3 - Baud generator times out. Master attempts to release SCL. 4 - When slave releases SCL, baud generator restarts. 5 - Baud generator times out. MSB of response shifted to I2CRSR. SCL driven low for next baud interval. 6 - At falling edge of 8th SCL clock, I2CRSR transferred to I2CRCV. Module clears RCEN bit. RBF bit is set. Master generates interrupt.
DS70068D-page 21-20
21
21.5.4 Acknowledge Generation
Setting the Acknowledge sequence enable bit, ACKEN (I2CCON<4>), enables generation of a master Acknowledge sequence. Note: The lower 5 bits of I2CCON must be 0 (master logic inactive) before attempting to set the ACKEN bit.
Figure 21-11 shows an ACK sequence and Figure 21-12 shows a NACK sequence. The Acknowledge data bit, ACKDT (I2CCON<5>), specifies ACK or NACK. After two baud periods: The ACKEN bit is automatically cleared. The module generates the MI2CIF interrupt. 21.5.4.1 IWCOL Status Flag If the software writes the I2CTRN when an Acknowledge sequence is in progress, then IWCOL is set and the contents of the buffer are ignored. Note: Because queueing of events is not allowed, writing to the lower 5 bits of I2CCON is disabled until the Acknowledge condition is complete.
Figure 21-11:
Master Acknowledge (ACK) Timing Diagram
ACKDT = 0 ACKEN I2C Bus State (Q) (Q) TBRG SCL (Master) SDA (Master) MI2CIF Interrupt (A) TBRG (Q) 1 - Writing ACKDT = 0 specifies sending an ACK. Writing ACKEN = 1 initiates a master Acknowledge event. Baud generator starts. SCL remains low. 2 - When SCL detected low, module drives SDA low. 3 - Baud generator times out. Module releases SCL. Baud generator restarts. 4 - Baud generator times out. Module drives SCL low then releases SDA. Module clears ACKEN. Master generates interrupt. 1 2 3 4
Figure 21-12: Master Not Acknowledge (NACK) Timing Diagram
ACKDT = 1 ACKEN I2C Bus State (Q) TBRG SCL (Master) SDA (Master) MI2CIF Interrupt (A) TBRG (I) 1 - Writing ACKDT = 1 specifies sending an NACK. Writing ACKEN = 1 initiates a master Acknowledge event. Baud generator starts. 2 - When SCL detected low, module releases SDA. 3 - Baud generator times out. Module releases SCL. Baud generator restarts. 4 - Baud generator times out. Module drives SCL low then releases SDA. Module clears ACKEN. Master generates interrupt. 1 2 3 4
DS70068D-page 21-21

21.5.5 Generating Stop Bus Event
Setting the Stop sequence enable bit, PEN (I2CCON<2>), enables generation of a master Stop sequence. Note: The lower 5 bits of I2CCON must be 0 (master logic inactive) before attempting to set the PEN bit.
When the PEN bit is set, the master generates the Stop sequence as shown in Figure 21-13. The slave detects the Stop condition, sets the P bit (I2CSTAT<4>) and clears the S bit (I2CSTAT<3>). The PEN bit is automatically cleared. The module generates the MI2CIF interrupt. 21.5.5.1 IWCOL Status Flag If the software writes the I2CTRN when a Stop sequence is in progress, then the IWCOL bit is set and the contents of the buffer are ignored. Note: Because queueing of events is not allowed, writing to the lower 5 bits of I2CCON is disabled until the Stop condition is complete.
Figure 21-13: Master Stop Timing Diagram
PEN I2C Bus State (Q) (Q) TBRG SCL (Master) SDA (Master) S P MI2CIF Interrupt TBRG (P) TBRG (I)
1 - Writing PEN = 1 initiates a master Stop event. Baud generator starts. Module drives SDA low. 2 - Baud generator times out. Module releases SCL. Baud generator restarts. 3 - Baud generator times out. Module releases SDA. Baud generator restarts. 4 - Slave logic detects Stop. Module sets P = 1, S = 0. 5 - The baud generator times out. Module clears PEN. Master generates interrupt.
3 4
DS70068D-page 21-22
21
21.5.6 Generating Repeated Start Bus Event
Setting the Repeated Start sequence enable bit, RSEN (I2CCON<1>), enables generation of a master Repeated Start sequence (see Figure 21-14). Note: The lower 5 bits of I2CCON must be 0 (master logic inactive) before attempting to set the RSEN bit.
To generate a Repeated Start condition, software sets the RSEN bit (I2CCON<1>). The module asserts the SCL pin low. When the module samples the SCL pin low, the module releases the SDA pin for one baud rate generator count (TBRG). When the baud rate generator times out, if the module samples SDA high, the module de-asserts the SCL pin. When the module samples SCL pin high, the baud rate generator reloads and begins counting. SDA and SCL must be sampled high for one TBRG. This action is then followed by assertion of the SDA pin low for one TBRG while SCL is high. The following is the Repeated Start sequence: The slave detects the Start condition, sets the S bit (I2CSTAT<3>) and clears the P bit (I2CSTAT<4>). The RSEN bit is automatically cleared. The module generates the MI2CIF interrupt. 21.5.6.1 IWCOL Status Flag If the software writes the I2CTRN when a Repeated Start sequence is in progress, then IWCOL is set and the contents of the buffer are ignored. Note: Because queueing of events is not allowed, writing of the lower 5 bits of I2CCON is disabled until the Repeated Start condition is complete.
Figure 21-14: Master Repeated Start Timing Diagram
RSEN I C Bus State (Q)

2
(Q) TBRG TBRG
(S) TBRG
(Q)
1 - Writing RSEN = 1 initiates a master Repeated Start event. Baud generator starts. Module drives SCL low and releases SDA. 2 - Baud generator times out. Module releases SCL. Baud generator restarts. 3 - Baud generator times out. Module drives SDA low. Baud generator restarts. 4 - Slave logic detects Start. Module sets S = 1, P = 0. 5 - The baud generator times out. Module drives SCL low. Module clears RSEN. Master generates interrupt.
SCL (Master) SDA (Master) S P MI2CIF Interrupt
3 4
DS70068D-page 21-23

21.5.7 Building Complete Master Messages
As described at the beginning of Section 21.5, the software is responsible for constructing messages with the correct message protocol. The module controls individual portions of the I 2C message protocol, however, sequencing of the components of the protocol to construct a complete message is a software task. The software can use polling or interrupt methods while using the module. The examples shown use interrupts. The software can use the SEN, RSEN, PEN, RCEN and ACKEN bits (Least Significant 5 bits of the I2CCON register) and the TRSTAT bit as a state flag when progressing through a message. For example, Table 21-2 shows some example state numbers associated with bus states. Table 21-2: Master Message Protocol States I2CCON<4:0> TRSTAT (I2CSTAT<14>) State
Example State Number
0 00000 0 Bus Idle or WAIT 1 00001 n/a Sending Start Event 2 00000 1 Master Transmitting 3 00010 n/a Sending Repeated Start Event 4 00100 n/a Sending Stop Event 5 01000 n/a Master Reception 6 10000 n/a Master Acknowledgement Note: Example state numbers for reference only. User software may assign as desired. The software will begin a message by issuing a Start command. The software will record the state number corresponding to Start. As each event completes and generates an interrupt, the interrupt handler may check the state number. So, for a Start state, the interrupt handler will confirm execution of the Start sequence and then start a master transmission event to send the I2C device address, changing the state number to correspond to master transmission. On the next interrupt, the interrupt handler will again check the state, determining that a master transmission just completed. The interrupt handler will confirm successful transmission of the data, then move on to the next event, depending on the contents of the message. In this manner, on each interrupt, the interrupt handler will progress through the message protocol until the complete message is sent. Figure 21-15 provides a more detailed examination of the same message sequence of Figure 21-7. Figure 21-16 shows some simple examples of messages using 7-bit addressing format. Figure 21-17 shows an example of a 10-bit address format message sending data to a slave. Figure 21-18 shows an example of a 10-bit address format message receiving data from a slave.
DS70068D-page 21-24
Figure 21-15: Master Message (Typical I2C Message: Read of Serial EEPROM)
SEN
RSEN
RCEN
ACKEN
SDA (Master) SCL (Slave) SDA (Slave) A A11 A10 A9 A8 0 0 0 0 A7A6A5A4 A3A2A1A0 1 0 1 0 A2 A1A0 R A A A

1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 N D7D6D5D4D3D2D1D0
ACKDT
PEN
SCL (Master)
1 2 3 4 5 6 7 8 9
1 0 1 0 A2 A1A0 W
I2CTRN
TBF
I2CRCV
RBF
ACKSTAT
MI2CIF cleared by user software.
MI2CIF
1 - Setting the SEN bit starts a Start event.
2 - Writing the I2CTRN register starts a master transmission. The data is the serial EE device address byte, with R/W clear indicating a write.
6 - Writing the I2CTRN register starts a master transmission. The data is a resend of the serial EE device address byte, but with R/W bit set indicating a read. 7 - Setting the RCEN bit starts a master reception. On interrupt, the software reads the I2CRCV register, which clears the RBF flag. 8 - Setting the ACKEN bit starts an Acknowledge event. ACKDT = 1 to send NACK. 9 - Setting the PEN bit starts a master Stop event.
3 - Writing the I2CTRN register starts a master transmission. The data is the first byte of the EE data address.
4 - Writing the I2CTRN register starts a master transmission. The data is the second byte of the EE data address.
DS70068D-page 21-25
5 - Setting the RSEN bit starts a Repeated Start event.
21
Figure 21-16: Master Message (7-bit Address: Transmission And Reception)
SEN
RSEN
DS70068D-page 21-26
1 2 3 4 5 6 7 8 9 D7D6D5D4D3D2D1D0 A6A5A4A3A2 A1A0 R 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 N A A A D7D6D5D4D3D2D1D0
RCEN
ACKEN
ACKDT
PEN
1 2 3 4 5 6 7 8 9
SCL (Master) SDA (Master) SCL (Slave)
A6A5A4A3A2 A1A0 W
SDA (Slave)
I2CTRN
TBF
I2CRCV
RBF
ACKSTAT
MI2CIF
2 - Writing the I2CTRN register starts a master transmission. The data is the address byte with R/W bit clear.
6 - Writing the I2CTRN register starts a master transmission. The data is the address byte with R/W bit set. 7 - Setting the RCEN bit starts a master reception. 8 - Setting the ACKEN bit starts an Acknowledge event. ACKDT = 1 to send NACK. 9 - Setting the PEN bit starts a master Stop event.
3 - Writing the I2CTRN register starts a master transmission. The data is the message byte.
4 - Setting the PEN bit starts a master Stop event.
Figure 21-17: Master Message (10-bit Transmission)
SEN
RSEN
RCEN
ACKEN

1 2 3 4 5 6 7 8 9 A7A6A5A4A3A2A1A0 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 D7D6D5D4D3D2D1D0 A A A A A
ACKDT
PEN
SCL (Master)
1 2 3 4 5 6 7 8 9
SDA (Master)
1 1 1 1 0 A9A8 W
SCL (Slave)
SDA (Slave)
I2CTRN
TBF
I2CRCV
RBF
ACKSTAT
MI2CIF
5 - Writing the I2CTRN register starts a master transmission. The data is the second byte of the message data. 6 - Writing the I2CTRN register starts a master transmission. The data is the third byte of the message data. 7 - Setting the PEN bit starts a master Stop event.
2 - Writing the I2CTRN register starts a master transmission. The data is the first byte of the address.
3 - Writing the I2CTRN register starts a master transmission. The data is the second byte of the address.
DS70068D-page 21-27
4 - Writing the I2CTRN register starts a master transmission. The data is the first byte of the message data.
21
Figure 21-18: Master Message (10-bit Reception)
SEN
RSEN
DS70068D-page 21-28
1 2 3 4 5 6 7 8 9 A7A6A5A4A3A2A1A0 1 1 1 1 0 A9A8 R A 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 N A D7D6D5D4D3D2D1D0 A A D7D6D5D4D3D2D1D0
RCEN
ACKEN
ACKDT
PEN SCL (Master)
1 2 3 4 5 6 7 8 9
SDA (Master)
1 1 1 1 0 A9A8 W
SCL (Slave) SDA (Slave)
I2CTRN
TBF
I2CRCV
RBF
ACKSTAT
MI2CIF cleared in user software.
MI2CIF
10
6 - Setting the RCEN bit starts a master reception. On interrupt, the software reads the I2CRCV register, which clears the RBF flag. 7 - Setting the ACKEN bit starts an Acknowledge event. ACKDT = 0 to send ACK. 8 - Setting the RCEN bit starts a master reception. 9 - Setting the ACKEN bit starts an Acknowledge event. ACKDT = 1 to send NACK. 10 - Setting the PEN bit starts a master Stop event.
2 - Writing the I2CTRN register starts a master transmission. The data is the first byte of the address with the R/W bit cleared.
3 - Writing the I2CTRN register starts a master transmission. The data is the second byte of the address.
4 - Setting the RSEN bit starts a master REStart event.
5 - Writing the I2CTRN register starts a master transmission. The data is a resend of the first byte with the R/W bit set.
21
21.6 Communicating as a Master in a Multi-Master Environment
The I2C protocol allows for more than one master to be attached to a system bus. Remembering that a master can initiate message transactions and generate clocks for the bus, the protocol has methods to account for situations where more than one master is attempting to control the bus. Clock synchronization ensures that multiple nodes can synchronize their SCL clocks to result in one common clock on the SCL line. Bus arbitration ensures that if more than one node attempts a message transaction, one and only one node will be successful in completing the message. The other nodes will lose bus arbitration and be left with a bus collision.
21.6.1
Multi-Master Operation
The master module has no special settings to enable multi-master operation. The module performs clock synchronization and bus arbitration at all times. If the module is used in a single master environment, clock synchronization will only occur between the master and slaves and bus arbitration will not occur.
21.6.2
Master Clock Synchronization

In a multi-master system, different masters may have different baud rates. Clock synchronization will ensure that when these masters are attempting to arbitrate the bus, their clocks will be coordinated. Clock synchronization occurs when the master de-asserts the SCL pin (SCL intended to float high). When the SCL pin is released, the baud rate generator (BRG) is suspended from counting until the SCL pin is actually sampled high. When the SCL pin is sampled high, the baud rate generator is reloaded with the contents of I2CBRG<8:0> and begins counting. This ensures that the SCL high time will always be at least one BRG rollover count in the event that the clock is held low by an external device, as shown in Figure 21-19.
Figure 21-19: Baud Rate Generator Timing with Clock Synchronization

TBRG SDA (Master) SCL (Master) SCL (Slave) Baud Counter 000 003 002 TCY 1 2 3 4 5 6 001 000 003 002 001 000 003 TBRG
1 - The baud counter decrements twice per TCY. On rollover, the master SCL will transition. 2 - The slave has pulled SCL low to initiate a wait. 3 - At what would be the master baud counter rollover, detecting SCL low holds counter. 4 - Logic samples SCL once per TCY. Logic detects SCL high. 5 - The baud counter rollover occurs on next cycle. 6 - On next rollover, the master SCL will transition.
DS70068D-page 21-29

21.6.3 Bus Arbitration and Bus Collision
Bus arbitration supports multi-master system operation. The wired-and nature of the SDA line permits arbitration. Arbitration takes place when the first master outputs a 1 on SDA by letting SDA float high and, simultaneously, the second master outputs a 0 on SDA by pulling SDA low. The SDA signal will go low. In this case, the second master has won bus arbitration. The first master has lost bus arbitration and thus has a bus collision. For the first master, the expected data on SDA is a 1 yet the data sampled on SDA is a 0. This is the definition of a bus collision. The first master will set the bus collision bit, BCL (I2CSTAT<10>), and generate a master interrupt. The master module will reset the I2C port to its Idle state. In multi-master operation, the SDA line must be monitored for arbitration to see if the signal level is the expected output level. This check is performed by the master module, with the result placed in the BCL bit. The states where arbitration can be lost are: A Start condition A Repeated Start condition Address, Data or Acknowledge bit A Stop condition
21.6.4
Detecting Bus Collisions and Resending Messages

When a bus collision occurs, the module sets the BCL bit and generates a master interrupt. If bus collision occurs during a byte transmission, the transmission is halted, the TBF flag is cleared and the SDA and SCL pins are de-asserted. If bus collision occurs during a Start, Repeated Start, Stop or Acknowledge condition, the condition is aborted, the respective control bits in the I2CCON register are cleared and the SDA and SCL lines are de-asserted. The software is expecting an interrupt at the completion of the master event. The software can check the BCL bit to determine if the master event completed successfully or if a collision occurred. If a collision occurs, the software must abort sending the rest of the pending message and prepare to resend the entire message sequence beginning with Start condition, after the bus returns to an Idle state. The software can monitor the S and P bits to wait for an Idle bus. When the software services the master Interrupt Service Routine and the I2C bus is free, the software can resume communication by asserting a Start condition.
DS70068D-page 21-30
21
21.6.5 Bus Collision During a Start Condition
Before issuing a Start command, the software should verify an Idle state of the bus using the S and P status bits. Two masters may attempt to initiate a message at a similar point in time. Typically, the masters will synchronize clocks and continue arbitration into the message until one loses arbitration. However, certain conditions can cause a bus collision to occur during a Start. In this case, the master that loses arbitration during the Start bit generates a bus collision interrupt.
21.6.6
Bus Collision During a Repeated Start Condition

Should two masters not collide throughout an address byte, a bus collision may occur when one master attempts to assert a Repeated Start while another transmits data. In this case, the master generating the Repeated Start will lose arbitration and generate a bus collision interrupt.
21.6.7
Bus Collision During Message Bit Transmission

The most typical case of data collision occurs while the master is attempting to transmit the device address byte, a data byte or an Acknowledge bit. If the software is properly checking the bus state, it is unlikely that a bus collision will occur on a Start condition. However, because another master can at a very similar time, check the bus and initiate its own Start condition, it is likely that SDA arbitration will occur and synchronize the starts of two masters. In this condition, both masters will begin and continue to transmit their messages until one master loses arbitration on a message bit. Remember that SCL clock synchronization will keep the two masters synchronized until one loses arbitration. Figure 21-20 shows an example of message bit arbitration.
Figure 21-20: Bus Collision During Message Bit Transmission
I2C Bus State (Q)
(D) TBRG
(Q) TBRG
(D)
(Q)
SCL (Master) SDA (Master) SCL (Bus) SDA (Bus) BCL TBF MI2CIF Interrupt
1 - Master transmits bit value of 1 in next SCL clock. Module releases SDA. 2 - Another master on bus transmits bit value of 0 in next SCL clock. Another master pulls SDA low. 3 - Baud generator times out. Module attempts to verify SDA high. Bus collision detected. Module releases SDA, SCL. Module sets BCL bit and clears TBF bit. Master generates interrupt.
21.6.8
Bus Collision During a Stop Condition

If the master software loses track of the state of the I 2C bus, there are conditions which cause a bus collision during a Stop condition. In this case, the master generating the Stop condition will lose arbitration and generate a bus collision interrupt.
DS70068D-page 21-31

21.7 Communicating as a Slave
In some systems, particularly where multiple processors communicate with each other, the dsPIC30F device may communicate as a slave (see Figure 21-21). When the module is enabled, the slave module is active. The slave may not initiate a message, it can only respond to a message sequence initiated by a master. The master requests a response from a particular slave as defined by the device address byte in the I 2C protocol. The slave module replies to the master at the appropriate times as defined by the protocol. As with the master module, sequencing the components of the protocol for the reply is a software task. However, the slave module detects when the device address matches the address specified by the software for that slave. Figure 21-21: A Typical Slave I2C Message: Multiprocessor Command/Status
10-bit Address S T A R T First Address Byte Second Address Byte Command Data Byte
RA / C WK
A C K
R E S T A AR CR K T
Address Byte
RA / C WK
Status Data Byte
N A C K
S T O P
S111 1 0AA0 9 8
AAA AA A A A 7 65 43 2 1 0
R 1 1 1 1 0 A A1 9 8
NP
After a Start condition, the slave module will receive and check the device address. The slave may specify either a 7-bit address or a 10-bit address. When a device address is matched, the module will generate an interrupt to notify the software that its device is selected. Based on the R/W bit sent by the master, the slave will either receive or transmit data. If the slave is to receive data, the slave module automatically generates the Acknowledge (ACK), loads the I2CRCV register with the received value currently in the I2CRSR register and notifies the software through an interrupt. If the slave is to transmit data, the software must load the I2CTRN register.
21.7.1
Sampling Receive Data

All incoming bits are sampled with the rising edge of the clock (SCL) line.
21.7.2
Detecting Start and Stop Conditions

The slave module will detect Start and Stop conditions on the bus and indicate that status on the S bit (I2CSTAT<3>) and P bit (I2CSTAT<4>). The Start (S) and Stop (P) bits are cleared when a Reset occurs or when the module is disabled. After detection of a Start or Repeated Start event, the S bit is set and the P bit is cleared. After detection of a Stop event, the P bit is set and the S bit is clear.
21.7.3
Detecting the Address

Once the module has been enabled, the slave module waits for a Start condition to occur. After a Start, depending on the A10M bit (I2CCON<10>), the slave will attempt to detect a 7-bit or 10-bit address. The slave module will compare 1 received byte for a 7-bit address or 2 received bytes for a 10-bit address. A 7-bit address also contains a R/W bit that specifies the direction of data transfer after the address. If R/W = 0, a write is specified and the slave will receive data from the master. If R/W = 1, a read is specified and the slave will send data to the master. The 10-bit address contains a R/W bit, however by definition, it is always R/W = 0 because the slave must receive the second byte of the 10-bit address.
DS70068D-page 21-32
21
Table 21-3: 0x00 0x01-0x03 0x04-0x77 0x78-0x7b 0x7c-0x7f Slave Addresses Suppported by the I2C Module: General call address or start byte Reserved Valid 7-bit addresses Valid 10-bit addresses (lower 7 bits) Reserved
Refer to dsPIC30F Family Reference Manual (DS70046) for descriptions of register bit fields. 21.7.3.1 7-bit Address and Slave Write Following the Start condition, the module shifts 8 bits into the I2CRSR register (see Figure 21-22). The value of register I2CRSR<7:1> is compared to the value of the I2CADD<6:0> register. The device address is compared on the falling edge of the eighth clock (SCL). If the addresses match, the following events occur: 1. 2. 3. 4. An ACK is generated. The D_A and R_W bits are cleared. The module generates the SI2CIF interrupt on the falling edge of the ninth SCL clock. The module will wait for the master to send data.
Figure 21-22: Slave Write 7-bit Address Detection Timing Diagram
I2C Bus State SCL (Master) SDA (Master) SDA (Slave) SI2CIF Interrupt R_W D_A ADD10 SCLREL
(S)
(D) (D)
(D) (A)
(Q)
1 - Detecting Start bit enables address detection. 2 - R/W = 0 bit indicates that slave receives data bytes. 3 - Address match of first byte clears D_A bit. Slave generates ACK. 4 - R_W bit cleared. Slave generates interrupt. 5 - Bus waiting. Slave ready to receive data.
A6
A5
A4
A3
A2
A1
A0 R/W =0
DS70068D-page 21-33

21.7.3.2 7-bit Address and Slave Read When a slave read is specified by having R/W = 1 in a 7-bit address byte, the process of detecting the device address is similar to that for a slave write (see Figure 21-23). If the addresses match, the following events occur: 1. 2. 3. An ACK is generated. The D_A bit is cleared and the R_W bit is set. The module generates the SI2CIF interrupt on the falling edge of the ninth SCL clock.
Since the slave module is expected to reply with data at this point, it is necessary to suspend the operation of the I 2C bus to allow the software to prepare a response. This is done automatically when the module clears the SCLREL bit. With SCLREL low, the slave module will pull down the SCL clock line, causing a wait on the I 2C bus. The slave module and the I 2C bus will remain in this state until the software writes the I2CTRN register with the response data and sets the SCLREL bit. Note: SCLREL will automatically clear after detection of a slave read address regardless of the state of the STREN bit. Figure 21-23: Slave Read 7-bit Address Detection Timing Diagram
I2C Bus State SCL (Master) SCL (Slave) SDA (Master) SDA (Slave) SI2CIF Interrupt R_W D_A ADD10 SCLREL
(S)
(D) (D)
(D) (A)
(Q)
1 - Detecting Start bit enables address detection. 2 - R/W = 1 bit indicates that slave sends data bytes.
A6
A5
A4
A3
A2
A1
A0 R/W
=1
3 - Address match of first byte clears D_A bit. Slave generates ACK. 4 - R_W bit set. Slave generates interrupt. SCLREL cleared. Slave pulls SCL low while SCLREL = 0. 5 - Bus waiting. Slave prepares to send data.
DS70068D-page 21-34
21
21.7.3.3 10-bit Address In 10-bit Address mode, the slave must receive two device address bytes (see Figure 21-24). The five Most Significant bits (MSbs) of the first address byte specify a 10-bit address. The R/W bit of the address must specify a write, causing the slave device to receive the second address byte. For a 10-bit address the first byte would equal 11110 A9 A8 0, where A9 and A8 are the two MSbs of the address. Following the Start condition, the module shifts 8 bits into the I2CRSR register. The value of register I2CRSR<2:1> is compared to the value of the I2CADD<9:8> register. The value of I2CRSR<7:3> is compared to 11110. The device address is compared on the falling edge of the eighth clock (SCL). If the addresses match, the following events occur: 1. 2. 3. An ACK is generated. The D_A and R_W bits are cleared. The module generates the SI2CIF interrupt on the falling edge of the ninth SCL clock.
The module does generate an interrupt after the reception of the first byte of a 10-bit address, however this interrupt is of little use. The module will continue to receive the second byte into I2CRSR. This time, I2CRSR<7:0> is compared to I2CADD<7:0>. If the addresses match, the following events occur: 1. 2. 3. 4. An ACK is generated. The ADD10 bit is set. The module generates the SI2CIF interrupt on the falling edge of the ninth SCL clock. The module will wait for the master to send data or initiate a Repeated Start condition. Note: Following a Repeated Start condition in 10-bit mode, the slave module only matches the first 7-bit address, 11110 A9 A8 0.
Figure 21-24: 10-bit Address Detection Timing Diagram

I2C Bus State SCL (Master) SDA (Master) SDA (Slave) SI2CIF Interrupt R_W D_A ADD10 SCLREL 1 1 - Detecting Start bit enables address detection. 2 - Address match of first byte clears D_A bit and causes slave logic to generate ACK. 3 - Reception of first byte clears R_W bit. Slave logic generates interrupt. 4 - Address match of first and second byte sets ADD10 and causes slave logic to generate ACK. 5 - Reception of second byte completes 10-bit address. Slave logic generates interrupt. 5 - Bus waiting. Slave ready to receive data. 2 3 4 5 6 1 1 1 1 0 A9 A8
R/W =0
(S)
(D) (D)
(D) (A)
(D) (D)
(D) (A)
(Q)
A7
A6
A5
A4
A3
A2
A1
A0
DS70068D-page 21-35

21.7.3.4 General Call Operation The addressing procedure for the I2C bus is such that the first byte after a Start condition usually determines which slave device the master is addressing. The exception is the general call address, which can address all devices. When this address is used, all enabled devices should respond with an Acknowledge. The general call address is one of eight addresses reserved for specific purposes by the I2C protocol. It consists of all 0s with R/W = 0. The general call is always a slave write operation. The general call address is recognized when the general call enable bit, GCEN (I2CCON<7>), is set (see Figure 21-25). Following a Start bit detect, 8 bits are shifted into the I2CRSR and the address is compared against the I2CADD, and is also compared to the general call address. If the general call address matches, the following events occur: 1. 2. 3. 4. 5. 6. An ACK is generated. Slave module will set the GCSTAT bit (I2CSTAT<9>). The D_A and R_W bits are cleared. The module generates the SI2CIF interrupt on the falling edge of the ninth SCL clock. The I2CRSR is transferred to the I2CRCV and the RBF flag bit is set (during the eighth bit). The module will wait for the master to send data.
When the interrupt is serviced, the cause for the interrupt can be checked by reading the contents of the GCSTAT bit to determine if the device address was device specific or a general call address. Note that general call addresses are 7-bit addresses. If A10M bit is set, configuring the slave module for 10-bit addresses and GCEN is set, the slave module continues to detect the 7-bit general call address. Figure 21-25: General Call Address Detection Timing Diagram (GCEN = 1)
I2C Bus State SCL (Master) SDA (Master) SDA (Slave) SI2CIF Interrupt R_W D_A CGSTAT I2CRCV RBF
(S)
(D) (D)
(D) (A)
(Q) 1 - Detecting Start bit enables address detection. 2 - All 0s and R/W = 0 bit indicates general call. 3 - Address match clears D_A bit and sets GCSTAT. Slave generates ACK. Address loaded into I2CRCV. 4 - R_W bit cleared. Slave generates interrupt. 5 - Bus waiting. Slave ready to receive data.
R/W =0
DS70068D-page 21-36
21
21.7.3.5 Receiving All Addresses (IPMI Operation) Some I 2C system protocols require a slave to act upon all messages on the bus. For example, the IPMI (Intelligent Peripheral Management Interface) bus uses I 2C nodes as message repeaters in a distributed network. To allow a node to repeat all messages, the slave module must accept all messages, regardless of the device address. Setting the IPMIEN bit (I2CCON<11>) enables this mode (see Figure 21-26). Regardless of the state of the I2CADD register and the A10M and GCEN bits, all addresses will be accepted. Figure 21-26: IPMI Address Detection Timing Diagram (IPMIEN = 1)
I2C Bus State SCL (Master) SDA (Master) SDA (Slave) SI2CIF Interrupt R_W D_A I2CRCV RBF
R/W
(S)
(D) (D)
(D) (A)
(Q)
1 - Detecting Start bit enables address detection. 2 - Regardless of contents of byte address is matched. Address match clears D_A bit. Slave generates ACK. Address loaded into I2CRCV. 3 - R_W bit set/clear. Slave generates interrupt. 4 - Bus waiting.
21.7.3.6
When an Address is Invalid If a 7-bit address does not match the contents of I2CADD<6:0>, the slave module will return to an Idle state and ignore all bus activity until after the Stop condition. If the first byte of a 10-bit address does not match the contents of I2CADD<9:8>, the slave module will return to an Idle state and ignore all bus activity until after the Stop condition. If the first byte of a 10-bit address matches the contents of I2CADD<9:8>, however, the second byte of the 10-bit address does not match I2CADD<7:0>, the slave module will return to an Idle state and ignore all bus activity until after the Stop condition.
21.7.4
Receiving Data from a Master Device

When the R/W bit of the device address byte is zero and an address match occurs, the R_W bit (I2CSTAT<2>) is cleared. The slave module enters a state waiting for data sent by the master. After the device address byte, the contents of the data byte are defined by the system protocol and are only received by the slave module. The slave module shifts 8 bits into the I2CRSR register. On the falling edge of the eighth clock (SCL), the following events occur: 1. 2. 3. 4. 5. 6. The module begins to generate an ACK or NACK. The RBF bit is set to indicate received data. The I2CRSR byte is transferred to the I2CRCV register for access by the software. The D_A bit is set. A slave interrupt is generated. Software may check the status of the I2CSTAT register to determine the cause of the event and then clear the SI2CIF flag. The module will wait for the next data byte.
DS70068D-page 21-37

21.7.4.1 Acknowledge Generation Normally, the slave module will Acknowledge all received bytes by sending an ACK on the ninth SCL clock. If the receive buffer is overrun, the slave module does not generate this ACK. Overrun is indicated if either (or both): 1. 2. The buffer full bit, RBF (I2CSTAT<1>), was set before the transfer was received. The overflow bit, I2COV (I2CSTAT<6>), was set before the transfer was received.
Table 21-4 shows what happens when a data transfer byte is received, given the status of the RBF and I2COV bits. If the RBF bit is already set when the slave module attempts to transfer to the I2CRCV, the transfer does not occur but the interrupt is generated and the I2COV bit is set. If both the RBF and I2COV bits are set, the slave module acts similarly. The shaded cells show the condition where software did not properly clear the overflow condition. Reading the I2CRCV clears the RBF bit. The I2COV is cleared by writing to a 0 through software. Table 21-4: Data Transfer Received Byte Actions Transfer I2CRSR I2CRCV Generate ACK Generate SI2CIF Interrupt (Interrupt occurs if enabled) Set RBF Set I2COV No change Yes Yes No change
Status Bits as Data Byte Received RBF 0 1 1 0 Note: 21.7.4.2 I2COV
0 Yes Yes Yes Yes 0 No No Yes No change 1 No No Yes No change 1 Yes No Yes Yes Shaded cells show state where the software did not properly clear the overflow condition. WAIT States During Slave Receptions
When the slave module receives a data byte, the master can potentially begin sending the next byte immediately. This allows the software controlling the slave module 9 SCL clock periods to process the previously received byte. If this is not enough time, the slave software may want to generate a bus WAIT period. The STREN bit (I2CCON<6>) enables a bus WAIT to occur on slave receptions. When STREN = 1 at the falling edge of the 9th SCL clock of a received byte, the slave module clears the SCLREL bit. Clearing the SCLREL bit causes the slave module to pull the SCL line low, initiating a WAIT. The SCL clock of the master and slave will synchronize, as shown in Section 21.6.2 Master Clock Synchronization. When the software is ready to resume reception, the software sets SCLREL. This causes the slave module to release the SCL line and the master resumes clocking.
DS70068D-page 21-38
21
21.7.4.3 Example Messages of Slave Reception Receiving a slave message is a rather automatic process. The software handling the slave protocol uses the slave interrupt to synchronize to the events. When the slave detects the valid address, the associated interrupt will notify the software to expect a message. On receive data, as each data byte transfers to the I2CRCV register, an interrupt notifies the software to unload the buffer. Figure 21-27 shows a simple receive message. Being a 7-bit address message, only one interrupt occurs for the address bytes. Then, interrupts occur for each of four data bytes. At an interrupt, the software may monitor the RBF, D_A and R_W bits to determine the condition of the byte received. Figure 21-28 shows a similar message using a 10-bit address. In this case, two bytes are required for the address. Figure 21-29 shows a case where the software does not respond to the received byte and the buffer overruns. On reception of the second byte, the module will automatically NACK the master transmission. Generally, this causes the master to resend the previous byte. The I2COV bit indicates that the buffer has overrun. The I2CRCV buffer retains the contents of the first byte. On reception of the third byte, the buffer is still full and again the module will NACK the master. After this, the software finally reads the buffer. Reading the buffer will clear the RBF bit, however the I2COV bit remains set. The software must clear the I2COV bit. The next received byte will be moved to the I2CRCV buffer and the module will respond with a ACK. Figure 21-30 highlights clock stretching while receiving data. Note in the previous examples, STREN = 0 which disables clock stretching on receive messages. In this example, the software sets STREN to enable clock stretching. When STREN = 1, the module will automatically clock stretch after each received data byte, allowing the software more time to move the data from the buffer. Note that if RBF = 1 at the falling edge of the 9th clock, the module will automatically clear the SCLREL bit and pull the SCL bus line low. As shown with the second received data byte, if the software can read the buffer and clear the RBF before the falling edge of the 9th clock, the clock stretching will not occur. The software can also suspend the bus at any time. By clearing the SCLREL bit, the module will pull the SCL line low after it detects the bus SCL low. The SCL line will remain low, suspending transactions on the bus until the SCLREL bit is set.
DS70068D-page 21-39
Figure 21-27: Slave Message (Write Data to Slave: 7-bit Address; Address Matches; A10M = 0; GCEN = 0; IPMIEN = 0)
SCL (Master) 1 2 3 4 5 6 7 8 9 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
SDA (Master)
A6A5A4A3A2 A1A0 W
SCL (Slave) A A A A A

SI2CIF cleared by user software.
SDA (Slave)
I2CRCV
RBF
I2COV
R_W
D_A
STREN
SCLREL
SI2CIF
1 - Slave recognizes Start event, S and P bits set/clear accordingly.
2 - Slave receives address byte. Address matches. Slave Acknowledges and generates interrupt. Address byte is moved to I2CRCV register and must be read by user software to prevent buffer overflow.
3 - Next received byte is message data. Byte moved to I2CRCV register, sets RBF. Slave generates interrupt. Slave Acknowledges reception.
4 - Software reads I2CRCV register. RBF bit clears.
DS70068D-page 21-40
5 - Slave recognizes Stop event, S and P bits set/clear accordingly.
21
Figure 21-28: Slave Message (Write Data to Slave: 10-bit Address; Address Matches; A10M=1; GCEN=0; IPMIEN=0)
SCL (Master) 1 2 3 4 5 6 7 8 9 A7A6A5A4A3A2A1A0 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
SDA (Master)
1 1 1 1 0 A9A8 W
DS70068D-page 21-41
A A A A A SI2CIF cleared by user software.
SCL (Slave)
SDA (Slave)
I2CRCV
RBF
I2COV
R_W
D_A
STREN
SCLREL
SI2CIF
4 - Next received byte is message data. Byte moved to I2CRCV register, sets RBF. Slave Acknowledges and generates interrupt. 5 - Software reads I2CRCV register. RBF bit clears. 6 - Slave recognizes Stop event, S and P bits set/clear accordingly.
2 - Slave receives address byte. High order address matches. Slave Acknowledges and generates interrupt. Address byte not moved to I2CRCV register.
3 - Slave receives address byte. Low order address matches. Slave Acknowledges and generates interrupt. Address byte not moved to I2CRCV register.
Figure 21-29: Slave Message (Write Data to Slave: 7-bit Address; Buffer Overrun; A10M = 0; GCEN = 0; IPMIEN = 0)
SCL (Master) 1 2 3 4 5 6 7 8 9 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
SDA (Master)
A6A5A4A3A2 A1A0 W
DS70068D-page 21-42
A N N A A SI2CIF cleared by user software.
SCL (Slave)
SDA (Slave)
I2CRCV
RBF
I2COV
R_W
D_A
STREN
SCLREL
SI2CIF
1 - Slave receives address byte. Address matches. Slave generates interrupt. Address byte not moved to I2CRCV register.
4 - Next byte also received before I2CRCV read by software. I2CRCV register unchanged. Slave generates interrupt. Slave sends NACK for reception. 6 - Software reads I2CRCV register. RBF bit clears. 7 - Software clears I2COV bit.
2 - Next received byte is message data. Byte moved to I2CRCV register, sets RBF. Slave generates interrupt. Slave Acknowledges reception.
3 - Next byte received before I2CRCV read by software. I2CRCV register unchanged. I2COV overflow bit set. Slave generates interrupt. Slave sends NACK for reception.
Figure 21-30: Slave Message (Write Data to Slave: 7-bit Address; Clock Stretching Enabled; A10M = 0; GCEN = 0; IPMIEN = 0)
SCL (Master) 1 2 3 4 5 6 7 8 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 D7D6D5D4 D3D2D1D0 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
SDA (Master)
A6A5A4A3A2 A1A0 W
SCL (Slave) A A A A
SDA (Slave)
2005 Microchip Technology Inc. 2 3 4 5 6 3 5 7 8 9 3 5 6 - Software sets SCLREL bit to release clock. 7 - Slave does not clear SCLREL because RBF = 0 at this time. 8 - Software may clear SCLREL to cause a clock hold. Module must detect SCL low before asserting SCL low. 9 - Software may set SCLREL to release a clock hold.
I2CTRN
TBF
I2CRCV
RBF
I2COV
R_W
D_A
STREN
SCLREL
SI2CIF
1 - Software sets the STREN bit to enable clock stretching.
2 - Slave receives address byte.
3 - Next received byte is message data. Byte moved to I2CRCV register, sets RBF.
4 - Because RBF = 1 at 9th clock, automatic clock stretch begins. Slave clears SCLREL bit. Slave pulls SCL line low to stretch clock.
DS70068D-page 21-43
5 - Software reads I2CRCV register. RBF bit clears.
21

21.7.5 Sending Data to a Master Device
When the R/W bit of the incoming device address byte is one and an address match occurs, the R_W bit (I2CSTAT<2>) is set. At this point, the master device is expecting the slave to respond by sending a byte of data. The contents of the byte are defined by the system protocol and are only transmitted by the slave module. When the interrupt from the address detection occurs, the software can write a byte to the I2CTRN register to start the data transmission. The slave module sets the TBF bit. The eight data bits are shifted out on the falling edge of the SCL input. This ensures that the SDA signal is valid during the SCL high time. When all eight bits have been shifted out, the TBF bit will be cleared. The slave module detects the Acknowledge from the master-receiver on the rising edge of the ninth SCL clock. If the SDA line is low indicating an Acknowledge (ACK), the master is expecting more data and the message is not complete. The module generates a slave interrupt to signal more data is requested. A slave interrupt is generated on the falling edge of the ninth SCL clock. Software must check the status of the I2CSTAT register and clear the SI2CIF flag. If the SDA line is high, indicating a Not Acknowledge (NACK), then the data transfer is complete. The slave module resets and does not generate an interrupt. The slave module will wait for detection of the next Start bit. 21.7.5.1 WAIT States During Slave Transmissions During a slave transmission message, the master expects return data immediately after detection of the valid address with R/W = 1. Because of this, the slave module will automatically generate a bus WAIT whenever the slave returns data. The automatic WAIT occurs at the falling edge of the 9th SCL clock of a valid device address byte or transmitted byte Acknowledged by the master, indicating expectation of more transmit data. The slave module clears the SCLREL bit. Clearing the SCLREL bit causes the slave module to pull the SCL line low, initiating a WAIT. The SCL clock of the master and slave will synchronize as shown in Section 21.6.2 Master Clock Synchronization. When the software loads the I2CTRN and is ready to resume transmission, the software sets SCLREL. This causes the slave module to release the SCL line and the master resumes clocking. 21.7.5.2 Example Messages of Slave Transmission Slave transmissions for 7-bit address messages are shown in Figure 21-31. When the address matches and the R/W bit of the address indicates a slave transmission, the module will automatically initiate clock stretching by clearing the SCLREL bit and generate an interrupt to indicate a response byte is required. The software will write the response byte into the I2CTRN register. As the transmission completes, the master will respond with an Acknowledge. If the master replies with an ACK, the master expects more data and the module will again clear the SCLREL bit and generate another interrupt. If the master responds with a NACK, no more data is required and the module will not stretch the clock nor generate an interrupt. Slave transmissions for 10-bit address messages require the slave to first recognize a 10-bit address. Because the master must send two bytes for the address, the R/W bit in the first byte of the address specifies a write. To change the message to a read, the master will send a Repeated Start and repeat the first byte of the address with the R/W bit specifying a read. At this point, the slave transmission begins as shown in Figure 21-32.
DS70068D-page 21-44
Figure 21-31: Slave Message (Read Data from Slave: 7-bit Address)
SCL (Master) 1 2 3 4 5 6 7 8 9 A N A 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
SDA (Master)
A6A5A4A3A2 A1A0 R
SCL (Slave) A D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0 D7D6D5D4D3D2D1D0
SDA (Slave)
2005 Microchip Technology Inc. 2 3 4 5 6 3 4 5 6 3 4 5 7 8 5 - After last bit, module clears TBF bit indicating buffer is available for next byte. 8 - Slave recognizes Stop event, S and P bits set/clear accordingly.
I2CTRN
TBF
I2CRCV
RBF
I2COV
R_W
D_A
STREN
SCLREL
SI2CIF
2 - Slave receives address byte. Address matches. Slave generates interrupt. Address byte not moved to I2CRCV register. R_W = 1 to indicate read from slave. SCLREL = 0 to suspend master clock.
6 - At end of 9th clock, if master sent ACK, module clears SCLREL to suspend clock. Slave generates interrupt. 7 - At end of 9th clock, if master sent NACK, no more data expected. Module does not suspend clock and will generate an interrupt.
3 - Software writes I2CTRN with response data. TBF = 1 indicates that buffer is full. Writing I2CTRN sets D_A, indicating data byte.
DS70068D-page 21-45
4 - Software sets SCLREL to release clock hold. Master resumes clocking and slave transmits data byte.
21
Figure 21-32: Slave Message (Read Data from Slave: 10-bit Address)
SCL (Master) 1 2 3 4 5 6 7 8 9 A7A6A5A4A3A2A1A0 1 1 1 1 0 A9A8 R A N 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
SDA (Master)
1 1 1 1 0 A9A8 W
DS70068D-page 21-46
A D7D6D5D4D3D2D1D0 A A D7D6D5D4D3D2D1D0
SCL (Slave)
SDA (Slave)
I2CTRN
TBF
I2CRCV
RBF
ADD10
R_W
D_A
STREN
SCLREL
SI2CIF
9 7 - Software sets SCLREL to release clock hold. Master resumes clocking and slave transmits data byte.
10
2 - Slave receives first address byte. Write indicated. Slave Acknowledges and generates interrupt.
3 - Slave receives address byte. Address matches. Slave Acknowledges and generates interrupt.
8 - At end of 9th clock, if master sent ACK, module clears SCLREL to suspend clock. Slave generates interrupt. 9 - At end of 9th clock, if master sent NACK, no more data expected. Module does not suspend clock or generate interrupt. 10 - Slave recognizes Stop event, S and P bits set/clear accordingly.
4 - Master sends a Repeated Start to redirect the message.
5 - Slave receives resend of first address byte. Read indicated. Slave suspends clock.
6 - Software writes I2CTRN with response data.
21
21.8 Connection Considerations for I2C Bus
By definition of the I2C bus being a wired AND bus connection, pull-up resistors on the bus are required, shown as RP in Figure 21-33. Series resistors, shown as RS are optional and used to improve ESD susceptibility. The values of resistors RP and RS depend on the following parameters: Supply voltage Bus capacitance Number of connected devices (input current + leakage current)
Because the device must be able to pull the bus low against RP, current drawn by RP must be greater than the I/O pin minimum sink current IOL of 3 mA at VOL(MAX) = 0.4V for the device output stage. For example, with a supply voltage of VDD = 5V +10%: RP(MIN) = (VDD(MAX) VOL(MAX)) / IOL = (5.5-0.4) / 3 mA = 1.7 k In a 400 kHz system, a minimum rise time specification of 300 nsec exists and in a 100 kHz system, the specification is 1000 nsec. Because RP must pull the bus up against the total capacitance CB with a maximum rise time of 300 nsec to 0.7 VDD, the maximum resistance for RP must be less than: RP(MAX) = -tR / CB * ln(1 (VIL(MAX) VDD(MAX)) = -300 nsec / (100pf * ln(1-0.7)) = 2.5 k The maximum value for RS is determined by the desired noise margin for the low level. RS cannot drop enough voltage to make the device VOL plus voltage across RS more than the maximum VIL. Rs(MAX) = (VIL(MAX) VOL(MIN)) / IOL(MAX) = (0.3 VDD-0.4) / 3 mA = 366 The SCL clock input must have a minimum high and low time for proper operation. The high and low times of the I2C specification as well as the requirements of the I2C module, are shown in the Electrical Specifications section in the specific device data sheet. Figure 21-33: Sample Device Configuration for I2C Bus
VDD + 10%
RP
RP
Device
RS
RS
SDA SCL CB = 10 - 400 pF
Note:
I2C devices with input levels related to VDD must have one common supply line to which the pull-up resistor is also connected.
DS70068D-page 21-47

21.8.1 Integrated Signal Conditioning
The SCL and SDA pins have an input glitch filter. The I2C bus requires this filter in both the 100 kHz and 400 kHz systems. When operating on a 400 kHz bus, the I2C specification requires a slew rate control of the device pin output. This slew rate control is integrated into the device. If the DISSLW bit (I2CCON<9>) is cleared, the slew rate control is active. For other bus speeds, the I2C specification does not require slew rate control and DISSLW should be set. Some system implementations of I2C busses require different input levels for VIL(MAX) and VIH(MIN). In a normal I2C system: VIL(MAX) = lesser of 1.5V and 0.3 VDD VIH(MIN) = greater of 3.0V and 0.7 VDD In an SMBus (System Management Bus) system: VIL(MAX) = 0.2 VDD VIH(MIN) = 0.8 VDD The SMEN bit (I2CCON<8>) controls the input levels. SMEN is set to change the input levels to SMBus specifications.
DS70068D-page 21-48
21
21.9 21.9.1 Module Operation During PWRSAV Instruction When the Device Enters Sleep Mode
When the device executes a PWRSAV 0 instruction, the device enters Sleep mode. When the device enters Sleep mode, the master and slave module abort any pending message activity and reset the state of the modules. Any transmission/reception that is in progress will not continue when the device wakes from Sleep. After the device returns to Operational mode, the master module will be in an Idle state waiting for a message command and the slave module will be waiting for a Start condition. During Sleep, the IWCOL, I2COV and BCL bits are cleared. Additionally, because the master functions are aborted, the SEN, RSEN, PEN, RCEN, ACKEN and TRSTAT bits are cleared. TBF and RBF are cleared and the buffers are available at wake-up. There is no automatic method to prevent Sleep entry if a transmission or reception is active or pending. The software must synchronize Sleep entry with I2C operation to avoid aborted messages. During Sleep, the slave module will not monitor the I2C bus. Thus, it is not possible to generate a wake-up event based on the I2C bus using the I2C module. Other interrupt inputs, such as the interrupt-on-change inputs can be used to detect message traffic on a I2C bus and cause a device wake-up.
21.9.2
When the Device Enters Idle Mode

When the device executes a PWRSAV 1 instruction, the device enters Idle mode. The module will enter a power saving state in Idle mode depending on the I2CSIDL bit (I2CCON<13>). If I2CSIDL = 1, the module will enter the Power Saving mode similarly to actions while entering Sleep mode. If I2CSIDL = 0, the module will not enter a Power Saving mode. The module will continue to operate normally.
21.10
Effects of a Reset
A Reset disables the I2C module and terminates any active or pending message activity. See the register definitions of I2CCON and I2CSTAT for the Reset conditions of those registers. Note: In this discussion, Idle refers to the CPU power saving state. The lower-case idle refers to the time when the I2C module is not transferring data on the bus.
DS70068D-page 21-49

21.11 Design Tips
Question 1: Im operating as a bus master and transmitting data, however, slave and receive interrupts are also occurring.
Answer: The master and slave circuits are independent. The slave module will receive events from the bus sent by the master. Question 2: Im operating as a slave and I write data to the I2CTRN register, but the data did not transmit.
Answer: The slave enters an automatic wait when preparing to transmit. Ensure that you set the SCLREL bit to release the I2C clock. Question 3: How do I tell what state the master module is in?
Answer: Looking at the condition of SEN, RSEN, PEN, RCEN, ACKEN and TRSTAT bits will indicate the state of the master module. If all bits are 0, the module is Idle. Question 4: Operating as a slave, I receive a byte while STREN = 0. What should the software do if it cannot process the byte before the next one is received?
Answer: Because STREN was 0, the module did not generate an automatic WAIT on the received byte. However, the software may, at any time during the message, set STREN then clear SCLREL. This will cause a WAIT on the next opportunity to synchronize the SCL clock. Question 5: My I 2C system is a multi-master system. When I attempt to send a message, it is being corrupted.
Answer: In a multi-master system, other masters may cause bus collisions. In the Interrupt Service Routine for the master, check the BCL bit to ensure that the operation completed without a collision. If a collision is detected, the message must be resent from the beginning. Question 6: My I 2C system is a multi-master system. How can I tell when it is OK to begin a message?
Answer: Look at the S and P bits. If S = 0 and P = 0 the bus is Idle. If S = 0 and P = 1, the bus is Idle. Question 7: I tried to send a Start condition on the bus, then transmit a byte by writing to the I2CTRN register. The byte did not get transmitted. Why?
Answer: You must wait for each event on the I2C bus to complete before starting the next one. In this case, you should poll the SEN bit to determine when the Start event completed, or wait for the master I2C interrupt before data is written to I2CTRN.
DS70068D-page 21-50
21
This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Inter-Integrated Circuit (I2C) module are: Title Use of the SSP Module in the I C Multi-Master Environment Using the PICmicro SSP for Slave I2C Communication Using the PICmicro MSSP Module for Master I2C Communications An I2C Network Protocol for Environmental Monitoring
2
Application Note # AN578 AN734 AN735 AN736
Note:
DS70068D-page 21-51

Revision B
This revision has been expanded to contain a full description of the dsPIC30F Inter-Integrated Circuit (I2C) module.
Revision C
DS70068D-page 21-52
Section 22. Data Converter Interface (DCI)

HIGHLIGHTS
This section of the manual contains the following topics: 22.1 Introduction .................................................................................................................. 22-2 22.2 Control Register Descriptions ...................................................................................... 22-2 22.3 Codec Interface Basics and Terminology..................................................................... 22-8 22.4 DCI Operation ............................................................................................................ 22-10 22.5 Using the DCI Module................................................................................................ 22-17 22.6 Operation in Power Saving Modes ............................................................................ 22-28 22.7 Registers Associated with DCI................................................................................... 22-28 22.8 Design Tips ................................................................................................................ 22-30 22.9 Related Application Notes.......................................................................................... 22-31 22.10 Revision History ......................................................................................................... 22-32
22
Data Converter Interface (DCI)
DS70069C-page 22-1

22.1 Introduction
The dsPIC Data Converter Interface (DCI) module allows simple interfacing of devices, such as audio coder/decoders (codecs), A/D converters, and D/A converters. The following interfaces are supported: Framed Synchronous Serial Transfer (Single or Multi-Channel) Inter-IC Sound (I2S) Interface AC-Link Compliant mode Many codecs intended for use in audio applications support sampling rates between 8 kHz and 48 kHz and use one of the interface protocols listed above. The DCI automatically handles the interface timing associated with these codecs. No overhead from the CPU is required until the requested amount of data has been transmitted and/or received by the DCI. Up to four data words may be transferred between CPU interrupts. The data word length for the DCI is programmable up to 16 bits to match the data size of the dsPIC30F CPU. However, many codecs have data word sizes greater than 16 bits. Long data word lengths can be supported by the DCI. The DCI is configured to transmit/receive the long word in multiple 16-bit time slots. This operation is transparent to the user and the long data word is stored in consecutive register locations. The DCI can support up to 16 time slots in a data frame, for a maximum frame size of 256 bits. There are control bits for each time slot in the data frame that determine whether the DCI will transmit/receive during the time slot.
22.2
Control Register Descriptions

The DCI has five Control registers and one Status register, which are listed below: DCICON1: DCI module enable and mode bits. DCICON2: DCI module word length, data frame length, and buffer setup. DCICON3: DCI module bit clock generator setup. DCISTAT: DCI module status information. RSCON: Active frame time slot control for data reception. TSCON: Active frame time slot control for data transmit.
In addition to these Control and Status registers, there are four Transmit registers, TXBUF0....TXBUF3, and four Receive registers, RXBUF0....RXBUF3.
DS70069C-page 22-2

Register 22-1: Upper Byte: R/W-0 DCIEN bit 15 DCICON1 U-0 R/W-0 DCISIDL U-0 R/W-0 DLOOP R/W-0 CSCKD R/W-0 CSCKE R/W-0 COFSD bit 8
Lower Byte: R/W-0 R/W-0 UNFM CSDOM bit 7 bit 15 DCIEN: DCI Module Enable bit 1 = Module is enabled 0 = Module is disabled Reserved: Read as 0
R/W-0 DJST
U-0
U-0
U-0
R/W-0 R/W-0 COFSM<1:0> bit 0
22
bit 14 bit 13
DCISIDL: DCI Stop in Idle Control bit 1 = Module will halt in CPU Idle mode 0 = Module will continue to operate in CPU Idle mode Reserved: Read as 0 DLOOP: Digital Loopback Mode Control bit 1 = Digital Loopback mode is enabled. CSDI and CSDO pins internally connected. 0 = Digital Loopback mode is disabled CSCKD: Sample Clock Direction Control bit 1 = CSCK pin is an input when DCI module is enabled 0 = CSCK pin is an output when DCI module is enabled CSCKE: Sample Clock Edge Control bit 1 = Data changes on serial clock falling edge, sampled on serial clock rising edge 0 = Data changes on serial clock rising edge, sampled on serial clock falling edge COFSD: Frame Synchronization Direction Control bit 1 = COFS pin is an input when DCI module is enabled 0 = COFS pin is an output when DCI module is enabled UNFM: Underflow Mode bit 1 = Transmit last value written to the Transmit registers on a transmit underflow 0 = Transmit 0s on a transmit underflow CSDOM: Serial Data Output Mode bit 1 = CSDO pin will be tri-stated during disabled transmit time slots 0 = CSDO pin drives 0s during disabled transmit time slots DJST: DCI Data Justification Control bit 1 = Data transmission/reception is begun during the same serial clock cycle as the frame synchronization pulse 0 = Data transmission/reception is begun one serial clock cycle after frame synchronization pulse Reserved: Read as 0 COFSM<1:0>: Frame Sync Mode bits 11 = 20-bit AC-Link mode 10 = 16-bit AC-Link mode 01 = I2S Frame Sync mode 00 = Multi-Channel Frame Sync mode Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 12 bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
bit 4-2 bit 1-0
DS70069C-page 22-3

Register 22-2: Upper Byte: U-0 bit 15 DCICON2 U-0 U-0 U-0 R/W-0 R/W-0 BLEN<1:0> U-0 R/W-0 COFSG3 bit 8
Lower Byte: R/W-0 R/W-0 R/W-0 COFSG<2:0> bit 7 bit 15-12 Reserved: Read as 0
U-0
R/W-0
R/W-0 R/W-0 WS<3:0>
R/W-0 bit 0
bit 11-10 BLEN<1:0>: Buffer Length control bits 11 = Four data words will be buffered between interrupts 10 = Three data words will be buffered between interrupts 01 = Two data words will be buffered between interrupts 00 = One data word will be buffered between interrupts bit 9 bit 8-5 Reserved: Read as 0 COFSG<3:0>: Frame Sync Generator control bits 1111 = Data frame has 16 words || 0010 = Data frame has 3 words 0001 = Data frame has 2 words 0000 = Data frame has 1 word Reserved: Read as 0 WS<3:0>: DCI Data Word Size bits 1111 = Data word size is 16 bits || 0100 = Data word size is 5 bits 0011 = Data word size is 4 bits 0010 = Invalid Selection. Do not use. Unexpected results may occur. 0001 = Invalid Selection. Do not use. Unexpected results may occur. 0000 = Invalid Selection. Do not use. Unexpected results may occur. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 4 bit 3-0
DS70069C-page 22-4

Register 22-3: Upper Byte: U-0 bit 15 DCICON3 U-0 U-0 U-0 R/W-0 R/W-0 R/W-0 BCG<11:8> R/W-0 bit 8 Lower Byte: R/W-0 R/W-0 bit 7 bit 15-12 Reserved: Read as 0. bit 11-0 BCG<11:0>: DCI Bit Clock Generator Control bits Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
R/W-0
R/W-0 R/W-0 BCG<7:0>
R/W-0
R/W-0
R/W-0 bit 0
22
DS70069C-page 22-5

Register 22-4: Upper Byte: U-0 bit 15 DCISTAT U-0 U-0 U-0 R-0 R-0 R-0 SLOT<3:0> R-0 bit 8 Lower Byte: U-0 bit 7 bit 15-12 bit 11-8 Reserved: Read as 0 SLOT<3:0>: DCI Slot Status bits 1111 = Slot #15 is currently active || 0010 = Slot #2 is currently active 0001 = Slot #1 is currently active 0000 = Slot #0 is currently active Reserved: Read as 0 ROV: Receive Overflow Status bit 1 = A receive overflow has occurred for at least one receive register 0 = A receive overflow has not occurred RFUL: Receive Buffer Full Status bit 1 = New data is available in the receive registers 0 = The receive registers have old data TUNF: Transmit Buffer Underflow Status bit 1 = A transmit underflow has occurred for at least one transmit register 0 = A transmit underflow has not occurred TMPTY: Transmit Buffer Empty Status bit 1 = The Transmit registers are empty 0 = The Transmit registers are not empty Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
U-0
U-0
U-0
R-0 ROV
R-0 RFUL
R-0 TUNF
R-0 TMPTY bit 0
bit 7-4 bit 3
bit 2
bit 1
bit 0
DS70069C-page 22-6

Register 22-5: RSCON R/W-0 RSE13 R/W-0 RSE12 R/W-0 RSE11 R/W-0 RSE10 R/W-0 RSE9 R/W-0 RSE8 bit 8
Upper Byte: R/W-0 R/W-0 RSE15 RSE14 bit 15
Lower Byte: R/W-0 R/W-0 RSE7 RSE6 bit 7 bit 11
R/W-0 RSE5
R/W-0 RSE4
R/W-0 RSE3
R/W-0 RSE2
R/W-0 RSE1
R/W-0 RSE0 bit 0
22
RSE<15:0>: Receive Slot Enable bits 1 = CSDI data is received during the individual time slot n 0 = CSDI data is ignored during the individual time slot n Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
Register 22-6:
TSCON R/W-0 TSE13 R/W-0 TSE12 R/W-0 TSE11 R/W-0 TSE10 R/W-0 TSE9 R/W-0 TSE8 bit 8
Upper Byte: R/W-0 R/W-0 TSE15 TSE14 bit 15
Lower Byte: R/W-0 R/W-0 TSE7 TSE6 bit 7 bit 11
R/W-0 TSE5
R/W-0 TSE4
R/W-0 TSE3
R/W-0 TSE2
R/W-0 TSE1
R/W-0 TSE0 bit 0
TSE<15:0>: Transmit Slot Enable Control bits 1 = Transmit buffer contents are sent during the individual time slot n 0 = CSDO pin is tri-stated or driven to logic 0 during the individual time slot, depending on the state of the CSDOM bit Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70069C-page 22-7

22.3 Codec Interface Basics and Terminology
The interface protocols supported by the DCI require the use of a Frame Synchronization (FS) signal to initiate a data transfer between two devices. In most cases, the rising edge of FS starts a new data transfer. In any codec application there is, at a minimum, a controller and a codec device. Either device may produce FS. The device that generates FS is the master device. Conceptually, the master device does not have to be the transmitting or receiving device. Various connection examples are shown in Figure 22-1. The frequency of the FS signal is usually the system sampling rate, fs. Note: The details given in this section are not specific to the DCI module. This discussion is intended to provide the user some background and terminology related to the digital serial interface protocols found in most codec devices.
Figure 22-1:
Codec CONNECTION EXAMPLES

CSCK SCK FS SDI SDO Codec is Master Codec dsPIC (Controller) CSCK COFS CSDO CSDI Controller is Master SCK FS SDI SDO Codec
dsPIC COFS (Controller) CSDO CSDI
CSCK dsPIC (Controller) COFS CSDO CSDI
SCK FS Codec SDI SDO (See Note)
Codec Generates SCK
CSCK dsPIC (Controller) COFS CSDI
SCK FS A/D #1 SDO FSO SCK FS
Daisy-Chained Configuration to Other Devices
A/D #2 SDO FSO
Controller
CSCK dsPIC (Controller) COFS CSDO CSDI
SCK FS SDI SDO Codec
External Controller is Master Note: Codec oscillator circuit generates SCK signal.
DS70069C-page 22-8

All interfaces have a serial transfer clock, SCK. The SCK signal may be generated by any of the connected devices or can be provided externally. In some systems, SCK is also referred to as the bit clock. For codecs that offer high signal fidelity, it is common for the SCK signal to be derived from the crystal oscillator on the codec device. The protocol defines the edge of SCK on which data is sampled. The master device generates the FS signal with respect to SCK. The period of the FS signal delineates one data frame. This period is the same as the data sample period. The number of SCK cycles that occur during the data frame will depend on the type of codec that is selected. The ratio of the SCK frequency to the system sample rate is expressed as a ratio of n, where n is the number of SCK periods per data frame. One advantage of using a framed interface protocol is that multiple data words can be transferred during each sample period, or data frame. Each division of the data frame is referred to as a time slot. The time slots can be used for multiple codec data channels and/or control information. Furthermore, multiple devices can be multiplexed on the same serial data pins. Each slave device is programmed to place its data on the serial data connection during the proper time slot. The output of each slave device is tri-stated at all other times to permit other devices to use the serial bus. Some devices allow the FS signal to be daisy-chained via Frame Synchronization Output (FSO) pins. A typical daisy-chained configuration is shown in Figure 22-1. When the transfer from the first slave device has completed, a FS pulse is sent to the second device in the chain via its FSO pin. This process continues until the last device in the chain has sent its data. The controller (master) device should be programmed for a data frame size that accommodates all of the data words that will be transferred. The timing for a typical data transfer is shown in Figure 22-2. Most protocols begin the data transfer one SCK cycle after the FS signal is detected. This example uses a 16 fs clock and transfers four 4-bit data words per frame. Figure 22-2: FRAMED DATA TRANSFER EXAMPLE
Data Frame Period (1/fs)
22
SCK FS SDI or SDO Time Slot 0 Time Slot 1 Time Slot 2 Time Slot 3
The timing for a typical data transfer with daisy-chained devices is shown in Figure 22-3. This example uses a 16 fs SCK frequency and transfers two 8-bit data words per frame. After the FS pulse is detected, the first device in the chain transfers the first 8-bit data word and generates the FSO signal at the end of the transfer. The FSO signal begins the transfer of the second data word from the second device in the chain. Figure 22-3: DAISY-CHAINED DATA TRANSFER EXAMPLE
Data Frame Period (1/fs)
SCK FS FSO SDI or SDO Time Slot 0 Time Slot 1
DS70069C-page 22-9

The FS pulse has a minimum active time of one SCK period so the slave device can detect the start of the data frame. The duty cycle of the FS pulse may vary depending on the specific protocol that is used to mark certain boundaries in the data frame. For example, the I2S protocol uses a FS signal that has a 50% duty cycle. The I2S protocol is optimized for the transfer of two data channels (left and right channel audio information). The edges of the FS signal mark the boundaries of the left and right channel data words. The AC-Link protocol uses a FS signal that is high for 16 SCK periods and low for 240 SCK periods. The edges of the AC-Link FS signal mark the boundaries of control information and data in the frame. Note: Refer to Section 26. Appendix of this manual for additional information on codec communication protocols.
22.4
DCI Operation
A simplified block diagram of the module is shown in Figure 22-4. The module consists of a Transmit/Receive Shift register that is connected to a small range of memory buffers via a buffer control unit. This arrangement allows the DCI to support various codec serial protocols. The DCI Shift register is 16-bits wide. Data is transmitted and received by the DCI MSbit first.
Figure 22-4:
DCI Module Block Diagram
BCG<11:0> CSCKD TCY Clock Generator CSCK
COFSD WS<3:0> COFSG<3:0> COFSM<1:0> 16-bit Data Bus Frame Synchronization Generator COFS
Receive Registers w/ Buffer Buffer Control
15 Transmit Registers w/ Buffer DCI Shift Register
0 CSDI
CSDO
DS70069C-page 22-10

22.4.1 DCI Pins
There are four I/O pins associated with the DCI. The DCI, when enabled, controls the data direction of each of the four pins. 22.4.1.1 CSCK Pin The CSCK pin provides the serial clock connection for the DCI. The CSCK pin may be configured as an input or output using the CSCKD control bit, DCICON1<10>. When the CSCK pin is configured as an output (CSCKD = 0), the serial clock is derived from the dsPIC30F system clock source and supplied to external devices by the DCI. When the CSCK pin is configured as an input (CSCKD = 1), the serial clock must be provided by an external device. 22.4.1.2 CSDO Pin The Serial Data Output (CSDO) pin is configured as an output only pin when the module is enabled. The CSDO pin drives the serial bus whenever data is to be transmitted. The CSDO pin can be tri-stated or driven to 0 during serial clock periods when data is not transmitted, depending on the state of the CSDOM control bit (DCICON1<6>). The tri-state option allows other devices to be multiplexed onto the CSDO connection. 22.4.1.3 CSDI Pin The serial data input (CSDI) pin is configured as an input only pin when the module is enabled. 22.4.1.4 COFS Pin The frame synchronization (COFS) pin is used to synchronize data transfers that occur on the CSDO and CSDI pins. The COFS pin may be configured as an input or an output. The data direction for the COFS pin is determined by the COFSD control bit (DCICON1<8>). When the COFSD bit is cleared, the COFS pin is an output. The DCI module will generate frame synchronization pulses to initiate a data transfer. The DCI is the master device for this configuration. When the COFSD bit is set, the COFS pin becomes an input. Incoming synchronization signals to the module will initiate data transfers. The DCI is a slave device when the COFSD control bit is set.
22
22.4.2
Module Enable
The DCI module is enabled or disabled by setting/clearing the DCIEN control bit (DCICON1<15>). Clearing the DCIEN control bit has the effect of resetting the module. In particular, all counters associated with serial clock generation, frame sync, and the buffer control logic are reset (see Section 22.5.1.1 DCI Start-up and Data Buffering and Section 22.5.1.2 DCI Disable for additional information). When enabled, the DCI controls the data direction for the CSCK, CSDI, CSDO and COFS I/O pins associated with the module. The PORT, LAT, and TRIS register values for these I/O pins are overridden by the DCI module when the DCIEN bit is set. It is also possible to override the CSCK pin separately when the bit clock generator is enabled. This permits the bit clock generator to be operated without enabling the rest of the DCI module.
DS70069C-page 22-11

22.4.3 Bit Clock Generator
The DCI module has a dedicated 12-bit time base that produces the bit clock. The bit clock rate (period) is set by writing a non-zero 12-bit value to the BCG<11:0> control bits (DCICON3<11:0>). When the BCG<11:0> bits are set to zero, the bit clock will be disabled. Note: The CSCK I/O pin will be controlled by the DCI module if the DCIEN bit is set OR the bit clock generator is enabled by writing a non-zero value to BCG<11:0>. This allows the bit clock generator to be operated independently of the DCI module.
When the CSCK pin is controlled by the DCI module, the corresponding PORT, LAT and TRIS Control register values for the CSCK pin will be overridden and the data direction for the CSCK pin will be controlled by the CSCKD control bit (DCICON1<10>). If the serial clock for the DCI is to be provided by an external device, the BCG<11:0> bits should be set to 0 and the CSCKD bit set to 1. If the serial clock is to be generated by the DCI module, the BCG<11:0> control bits should be set to a non-zero value (see Equation 22-1) and the CSCKD control bit should be set to zero. The formula for the bit clock frequency is given in Equation 22-1. Equation 22-1: DCI Bit Clock Generator Value BCG<11:0> = fCY 2 fCSCK 1
The required bit clock frequency will be determined by the system sampling rate and frame size. Typical bit clock frequencies range from 16x to 512x the converter sample rate, depending on the data converter and the communication protocol that is used. Note: The BCG<11:0> bits have no effect on the operation of the DCI module when the CSCK signal is provided externally (CSCKD = 1).
22.4.4
Sample Clock Edge Selection

The CSCKE control bit (DCICON1<9>) determines the sampling edge for the serial clock signal. If the CSCKE bit is cleared (default), data will be sampled on the falling edge of the CSCK signal. The AC-Link protocols and most multi-channel formats require that data be sampled on the falling edge of the CSCK signal. If the CSCKE bit is set, data will be sampled on the rising edge of CSCK. The I2S protocol requires that data be sampled on the rising edge of the serial clock signal.
22.4.5
Frame Sync Mode Control Bits

The type of interface protocol supported by the DCI is selected using the COFSM<1:0> control bits (DCICON1<1:0>). The following Operating modes can be selected: Multi-channel Mode I2S Mode AC-Link Mode (16-bit) AC-Link Mode (20-bit)
Specific information for each of the protocols is provided in subsequent sections.
22.4.6
Word-Size Selection Bits

The WS<3:0> word-size selection bits (DCICON2<3:0>) determine the number of bits in each DCI data word. Any data length from 4 to 16 bits may be selected. Note: The WS control bits are used only in the multi-channel and I2S modes. These bits have no effect in AC-Link mode since the data slot sizes are fixed by the protocol.
DS70069C-page 22-12

22.4.7 Frame Synchronization Generator
The Frame Sync Generator (FSG) is a 4-bit counter that sets the frame length in data words. The period for the FSG is set by writing the COFSG<3:0> control bits (DCICON2<8:5>). The FSG period (in serial clock cycles) is determined by the following formula: Equation 22-2: Frame Length, In CSCK Cycles FrameLength = (WS<3:0> + 1) (COFSG<3:0> + 1)
Frame lengths up to 16 data words may be selected. The frame length in serial clock periods will vary up to a maximum of 256 depending on the word size that is selected. Note: The COFSG control bits will have no effect in AC-Link mode, since the frame length is set to 256 serial clock periods by the protocol.
22
22.4.8
Transmit and Receive Registers

The DCI has four Transmit registers, TXBUF0...TXBUF3, and four Receive registers, RXBUF0..RXBUF3. All of the Transmit and Receive registers are memory mapped.
22.4.8.1
Buffer Data Alignment Data values are always stored left-justified in the DCI registers, since audio PCM data is represented as a signed 2s complement fractional number. If the programmed DCI word size is less than 16 bits, the unused LSbs in the Receive registers are set to 0 by the module. Also, the unused LSbs in the Transmit register are ignored by the module.
22.4.8.2
Transmit and Receive Buffers The Transmit and Receive registers each have a set of buffers that are not accessible by the user. Effectively, each transmit and receive buffer location is double-buffered. The DCI transmits data from the transmit buffers and writes received data to the receive buffers. The buffers allow the user to read and write the RXBUF and TXBUF registers, while the DCI uses data from the buffers.
22.4.9
DCI Buffer Control Unit

The DCI module contains a buffer control unit that transfers data between the buffer memory and the Serial Shift register. The buffer control unit also transfers data between the buffer memory and the TXBUF and RXBUF registers. The buffer control unit allows the DCI to queue the transmission and reception of multiple data words without CPU overhead. The DCI generates an interrupt each time a transfer between the buffer memory and the TXBUF and RXBUF registers takes place. The number of data words buffered between interrupts is determined by the BLEN<1:0> control bits (DCICON2<11:10>). The size of the transmit and receive buffering may be varied from 1 to 4 data words using the BLEN<1:0> bits. Each time a data transfer takes place between the DCI Shift register and the buffer memory, the DCI buffer control unit is incremented to point to the next buffer location. If the number of transmitted or received data words is equal to the BLEN value + 1, the following will occur: The buffer control unit is reset to point to the first buffer location The received data held in the buffer is transferred to the RXBUF registers The data in the TXBUF registers is transferred to the buffer A CPU interrupt is generated
The DCI buffer control unit will also reset the buffer pointer to the first buffer location each time a frame boundary is reached. This action ensures alignment between the buffer locations and the enabled time slots in the data frame. The DCI buffer control unit always accesses the same relative location in the Transmit and Receive buffers. If the DCI is transmitting data from TXBUF3, for example, then any data received during that time slot will be written to RXBUF3.
DS70069C-page 22-13

Figure 22-5: DCI Buffer Control Unit
Transmit Buffer
Transmit Registers TXBUF0 TXBUF1 TXBUF2 TXBUF3 4 4
Transmit Buffer Select
BLEN Buffer Transfer Signal 2 DCI Shift Register
Buffer Control
Receive Registers RXBUF0 RXBUF1 RXBUF2 RXBUF3 4 4
Receive Buffer
Receive Buffer Select
22.4.10
Transmit Slot Enable Bits

The TSCON SFR has control bits that are used to enable up to 16 time slots for transmission. These control bits are the TSE<15:0> bits. The size of each time slot is determined by the WS<3:0> word size selection bits and can vary up to 16 bits. If a transmit time slot is enabled via one of the TSE bits (TSEx = 1), the contents of the current transmit buffer location will be loaded into the CSDO Shift register and the DCI buffer control unit will increment to point to the next buffer location. Not all TSE control bits will have an effect on the module operation if the selected frame size has less than 16 data slots. The Most Significant TSE control bits are not used. For example, if COFSG<3:0> = 0111 (8 data slots per frame), TSE8 through TSE15 will have no effect on the DCI operation.
22.4.10.1 CSDO Mode Control During disabled transmit time slots, the CSDO pin can drive 0s or can be tri-stated, depending on the state of the CSDOM bit (DCICON1<6>). A given transmit time slot is disabled if its corresponding TSEx bit is cleared in the TSCON register. If the CSDOM bit is cleared (default), the CSDO pin will drive 0s onto the CSDO pin during disabled time slot periods. This mode is used when there are only two devices (1 master and 1 slave) attached to the serial bus. If the CSDOM bit is set, the CSDO pin will be tri-stated during unused time slot periods. This mode allows multiple dsPIC30F devices to share the same CSDO line in a multiplexed application. Each device on the CSDO line is configured so that it will only transmit data during specific time slots. No two devices should transmit data during the same time slot.
DS70069C-page 22-14

22.4.11 Receive Slot Enable Bits
The RSCON SFR contains control bits (RSE<15:0>) that are used to enable up to 16 time slots for reception. The size of each receive time slot is determined by the WS<3:0> control bits and can vary from 4 to 16 bits. If a receive time slot is enabled via one of the RSE bits (RSEx = 1), the Shift register contents will be written to the current DCI receive buffer location and the buffer control logic will advance to the next available buffer location. Data is not packed in the receive memory buffer locations if the selected word size is less than 16 bits. Each received slot data word is stored in a seperate 16-bit buffer location. Data is always stored in a left-justified format in the receive memory buffer.
22.4.12
TSCON and RSCON Operation with Buffer Control Unit

The slot enable bits in the TSCON and RSCON registers function independently, with the exception of the buffer control logic. For each time slot in a data frame, the buffer location is advanced if either the TSEx or the RSEx bit is set for the current time slot. That is, the buffer control unit synchronizes the Transmit and Receive buffering so that the Transmit and Receive buffer location will always be the same for each time slot in the data frame. If the TSEx bit and the RSEx bit are both set for every time slot that is used in the data frame, the DCI will Transmit and Receive equal amounts of data . In some applications, the number of data words transmitted during a frame may not equal the number of words received. As an example, assume that the DCI is configured for a 2-word data frame, TSCON = 0x0001 and RSCON = 0x0003. This configuration would allow the DCI to transmit one data word per frame and receive two data words per frame. Since two data words are received for each data word that is transmitted, the user would write every other transmit buffer location. Specifically, only TXBUF0 and TXBUF2 would be used to transmit data. Figure 22-6: DCI Buffer Operation: TSCON = 0x0001, RSCON = 0x0003, BLEN<1:0> = 11b
Transmit Registers TXBUF0 TXBUF1 TXBUF2 Data Word #2 Data Word #1 RXBUF0 RXBUF1 RXBUF2 Receive Registers Data Word #1
22
Data Word #2 Data Word #3 Data Word #4
TXBUF3
RXBUF3
Note: User writes to TXBUF0 and TXBUF2. TXBUF1 and TXBUF3 not used by transmit logic.
22.4.13
Receive Status Bits

There are two receive status bits, RFUL and ROV. The receive status bits only indicate status for register locations that are enabled for use by the module. This is a function of the BLEN<1:0> control bits. If the buffer length is set to less than four words, the unused buffer locations will not affect the receive status bits. The RFUL status bit (DCISTAT<2>) is read only and indicates that new data is available in the Receive registers. The RFUL bit is cleared automatically when all RXBUF registers in use have been read by the user software. The ROV status bit (DCISTAT<3>) is read only and indicates that a receive overflow has occurred for at least one of the Receive register locations. A receive overflow occurs when the RXBUF register location is not read by the user software before new data is transferred from the buffer memory. When a receive overflow occurs, the old contents of the register are overwritten. The ROV status bit is cleared automatically when the register that caused the overflow is read.
DS70069C-page 22-15

22.4.14 Transmit Status Bits
There are two transmit status bits, TMPTY and TUNF. The transmit status bits only indicate status for register locations that are used by the module. If the buffer length is set to less than four words, for example, the unused register locations will not affect the transmit status bits. The TMPTY bit (DCISTAT<0>) is read only and is set when the contents of the active TXBUF registers are transferred to the Transmit Buffer registers. The TMPTY bit may be polled in software to determine when the Transmit registers may be written. The TMPTY bit is cleared automatically by the hardware when a write to any of the TXBUF registers in use occurs. The TUNF bit (DCISTAT<1>) is read only and indicates that a transmit underflow has occurred for at least one of the Transmit registers that is in use. The TUNF bit is set when the TXBUF register contents are transferred to the transmit buffer memory and the user did not write all of the TXBUF registers in use since the last buffer transfer. The TUNF status bit is cleared automatically when the TXBUF register that underflowed is written by the user software.
22.4.15
SLOT Status Bits

The SLOT<3:0> status bits (DCISTAT<11:7>) indicate the current active time slot in the data frame and are useful when more than four words per data frame need to be transferred. The user may poll these status bits in software when a DCI interrupt occurs to determine what time slot data was last received and which time slot data should be loaded into the TXBUF registers.
22.4.16
Digital Loopback Mode

Digital Loopback mode is enabled by setting the DLOOP control bit (DCICON1<11>). When the DLOOP bit is set, the module internally connects the CSDO signal to CSDI. The actual data input on the CSDI pin will be ignored in Digital Loopback mode.
22.4.17
Underflow Mode Control Bit

When a transmit underflow occurs, one of two actions may occur depending on the state of the UNFM control bit (DCICON1<7>). If the UNFM bit is cleared (default), the module will transmit 0s on the CSDO pin during the active time slot for the buffer location. In this Operating mode, the codec device attached to the DCI module will simply be fed digital silence. If the UNFM control bit is set, the module will transmit the last data written to the buffer location. This Operating mode permits the user to send a continuous data value to the codec device without consuming software overhead.
22.4.18
Data Justification Control

In most applications, the data transfer begins one serial clock cycle after the FS signal is sampled active. This is the DCI module default. An alternate data alignment can be selected by setting the DJST control bit (DCICON2<5>). When DJST = 1, data transfers will begin during the same serial clock cycle as the FS signal.
22.4.19
DCI Module Interrupts

The frequency of DCI module interrupts is dependent on the number of active time slots (TSCON and RSCON registers), length of the data frame (WS and COFSG control bits), and the BLEN control bits. An interrupt is generated at the following times: When the buffer length has been reached When a frame boundary is reached A buffer memory transfer takes place each time the above events occur. A buffer memory transfer is defined as the time when the previously written TXBUF values are transferred to the transmit buffer memory and new received values in the receive buffer memory are transferred into the RXBUF registers.
DS70069C-page 22-16

22.5 Using the DCI Module
This section explains how to configure and use the DCI with specific kinds of data converters.
22.5.1
How to Transmit and Receive Data Using the DCI Buffers, Status Bits and Interrupts
The DCI can buffer up to four data words between CPU interrupts depending on the setting of the BLEN control bits. The buffered data can be transmitted and received in a single data frame, or across multiple data frames, depending on the TSCON and RSCON register settings. For example, assume BLEN<1:0> = 00b ( buffer one data word per interrupt) and TSCON = RSCON = 0x0001. This particular configuration represents the most basic setup and would cause the DCI to transmit/receive one data word at the beginning of every data frame. The CPU would be interrupted after every data word transmitted/received since BLEN<1:0> = 00b. For a second configuration example, assume BLEN<1:0> = 11b (buffer four data words per interrupt) and TSCON = RSCON = 0x0001. This configuration would cause the DCI to transmit/receive one data word at the beginning of every data frame, but a CPU interrupt would be generated after four data words were transmitted/received. This configuration would be useful for block processing, where multiple data samples are processed at once. For a third configuration example, assume BLEN<1:0> = 11b (buffer four data words per interrupt) and TSCON = RSCON = 0x000F. This configuration would cause the DCI to transmit/receive four data words at the beginning of every data frame. A CPU interrupt would be generated every data frame in this case because the DCI was setup to buffer four data words in a data frame. This configuration represents a typical multi-channel buffering setup. The DCI can also be configured to buffer more than four data words per frame. For example, assume BLEN<1:0> = 11b (buffer four data words per interrupt) and TSCON = RSCON = 0x00FF. In this configuration, the DCI will transmit/receive 8 data words per data frame. An interrupt will be generated twice per data frame. To determine which portion of the data is in the Transmit/Receive registers at each interrupt, the user will need to check the SLOT status bits (DCISTAT <11:7>) in the Interrupt Service Routine to determine the current data frame position. The Transmit and Receive registers are double-buffered, so the DCI module can work on one set of Transmit and Receive data while the user software is manipulating the other set of data. Because of the double-buffers, it will take three interrupt periods to receive the data, process that data, and transmit the processed data. For each DCI interrupt, the CPU will process a data word that was received during a prior interrupt period and generate a data word that will be transmitted during the next interrupt period. The buffering and data processing time of the dsPIC device will insert a two-interrupt period delay into the processed data. This data delay is negligible, in most cases. The DCI status flags and CPU interrupt indicate that a buffer transfer has taken place and that it is time for the CPU to process more data. In a typical application, the following steps will occur each time the DCI data is processed: 1. The RXBUF registers are read by the user software. The RFUL status bit (DCISTAT<2>) will have been set by the module to indicate the Receive registers contain new data. The RFUL bit is cleared automatically after all the active Receive registers have been read. The user software will process the received data. The processed data is written to the TXBUF registers. The TMPTY status bit (DCISTAT<0>) will have been previously set to indicate that the Transmit registers are ready for more data to be written.
22
2. 3.
For applications that are configured to Transmit and Receive data (TSCON and RSCON are non-zero), the RFUL and TMPTY status bits can be polled in user software to determine when a DCI buffer transfer takes place. If the DCI is only used to transmit data (RSCON = 0), then the TMPTY bit can be polled to indicate a buffer transfer. If the DCI is configured to only receive data (TSCON = 0), then the RFUL bit can be polled to indicate a buffer transfer. The DCIIF status bit (IFS2<9>) is set each time a DCI buffer transfer takes place and generates a CPU interrupt, if enabled. The DCIIF status bit is generated by the logical ORing of the RFUL and TMPTY status bits.
DS70069C-page 22-17

22.5.1.1 DCI Start-up and Data Buffering Data transfers are begun by setting the DCIEN control bit (DCICON1<15>). Prior to this, the DCI Control registers should have been initialized for the desired operating mode. (See Section 22.5.4 Multi-Channel Operation, Section 22.5.5 I2S Operation, and Section 22.5.6 AC-Link Operation) A timing diagram for DCI startup is shown in Figure 22-7. In this example, the DCI is configured for an 8-bit data word (WS<3:0> = 0111b) and an 8-bit data frame (COFSG<3:0> = 0000b). The Multi-Channel mode (COFSM<1:0> = 00b) is used. The steps required to transmit and receive data are described below. 1. The TXBUF registers should be pre-loaded with the first data to be transmitted before the module is enabled. If the transmit data will be based on data received from the codec, then the user can simply clear the TXBUF registers. This will transmit digital silence until data is first received into the RXBUF registers from the codec. Enable the DCI module by setting the DCIEN bit (DCICON1<15>). If the DCI is the master device, the data in the TXBUF registers will be transferred to the transmit buffers and transmission of the first data frame will commence. Otherwise, the TXBUF data will be held in the transmit buffers until a frame sync signal is received from the master device. The TMPTY bit will be set immediately after the module is enabled and a DCI interrupt will be generated, if enabled. At this time, the module is ready for the TXBUF registers to be reloaded with data to be transferred on the second data frame. No data has been received by the module at this time, so the TXBUF registers should be cleared again if the transmitted data is calculated from the received data. The DCIIF status bit should be cleared by the user in software if interrupts are enabled. After the first data frame is transferred, the TMPTY bit will set, the RFUL status bit will be set, and a DCI interrupt will occur, if enabled. This is the first data word received from the device connected to the DCI. The user reads the Receive register(s), automatically clearing the RFUL status bit. The user software processes the received data at this time. The Transmit register(s) is written with data to be transmitted during the next data frame. The TMPTY status bit is cleared automatically when the write occurs. The write data may be calculated from data that was received at the prior interrupt. The next DCI interrupt occurs and the cycle repeats.
2.
3.
4.
5. 6.
7. Figure 22-7:
CSCK Data COFS DCIEN TMPTY TXBUF RFUL RXBUF DCIIF
DCI Start-up and Data Buffering Example
Word 1
Word 2
TX Word 1
TX Word 2
TX Word 3
RX Word 1 Cleared by User
DS70069C-page 22-18

22.5.1.2 DCI Disable The DCI module is disabled by clearing the DCIEN control bit (DCICON1<15>). When the DCIEN bit is cleared, the module will finish the current data frame transfer that is in progress. An interrupt will be generated if the transmit/receive buffers need to be written/read before the end of the frame. The DCIEN bit must be cleared at least 3 CSCK cycles before the end of the frame disables the module at that frame. If not, the module will disable on the next frame. The DCI will not generate any further frame sync pulses after the DCIEN bit is cleared, nor will it respond to an incoming frame sync pulse. When the frame sync generator has reached the final time slot in the data frame, all state machines associated with the DCI will be reset to their Idle state and control of the I/O pins associated with the module will be released. The user may poll the SLOT<3:0> status bits (DCISTAT<11:7>) after the DCIEN bit is cleared to determine when the module is Idle. The DCI is Idle when SLOT<3:0> = 0000b and DCIEN = 0. When the module enters the Idle state, any data in the Receive Shadow registers will be transferred to the RXBUF registers, and the RFUL and ROV status bits will be affected accordingly. Figure 22-8: DCI Timing, Module Disable
FS pulse not generated.
22
WS = 0011b
COFSG = 0011b
CSCK Data COFS DCIEN SLOT RFUL Receive buffer contents transferred to RXBUF. 0011 0000 0001 0010 0011 0000 3 2 1 0 3 2 1 0 3 2 1 0 3 2 1 0 3 2 1 0
DS70069C-page 22-19

22.5.2 Master vs. Slave Operation
The DCI can be configured for master or slave operation. The master device generates the frame sync signal to initiate a data transfer. The Operating mode (master or slave) is selected by the COFSD control bit (DCICON1<8>). When the DCI module is operating as a master device (COFSD = 0), the COFSM mode bits determine the type of frame sync pulse that is generated by the frame sync generator logic. A new frame synchronization signal is generated when the frame sync generator resets and is output on the COFS pin. When the DCI module is operating as a frame sync slave (COFSD = 1), data transfers are controlled by the device attached to the DCI module. The COFSM control bits control how the DCI module responds to incoming FS signals. In the Multi-Channel mode, a new data frame transfer will begin one serial clock cycle after the COFS pin is sampled high. The pulse on the COFS pin resets the frame sync generator logic. In the I2S mode, a new data word will be transferred one serial clock cycle after a low-to-high or a high-to-low transition is sampled on the COFS pin. A rising or falling edge on the COFS pin resets the frame sync generator logic. In the AC-Link mode, the tag slot and subsequent data slots for the next frame will be transferred one serial clock cycle after the COFS pin is sampled high. The COFSG and WS bits must be configured to provide the expected frame length when the module is operating in the Slave mode. Once a valid frame sync pulse has been sampled by the module on the COFS pin, an entire data frame transfer will take place. The module will not respond to further frame sync pulses until the current data frame transfer has fully completed.
22.5.3
Data Packing for Long Data Word Support

Many codecs have data word lengths in excess of 16 bits. The DCI natively supports word lengths up to 16 bits, but longer word lengths can be supported by enabling multiple Transmit and Receive slots and packing data into multiple transmit and receive buffer locations. For example, assume that a particular codec transmits/receives 24-bit data words. This data could be transmitted and received by setting BLEN<1:0> = 01b (two data words per interrupt) and setting TSCON = RSCON = 0x0003. This will enable transmission and reception during the first two time slots of the data frame. The 16 MSbs of the transmit data are written to TXBUF0. The 8 LSbs of the transmit data are written left-justified to TXBUF1 as shown in Figure 22-9. The 8 LSbs of TXBUF1 can be written to 0. The 24-bit data received from the codec will be loaded into RXBUF0 and RXBUF1 with the same format as the transmit data. Any combination of word size and enabled time slots may be used to transmit and receive long data words in multiple Transmit and Receive registers. For example, the 24 bit data word example shown in Figure 22-9 could be transmitted/received in three consecutive registers by setting WS<3:0> = 0111 (word size = 8 bits), BLEN<1:0> = 10 (buffer three words between interrupts), and TSCON = RSCON = 0x0007 (transmit/receive during the first three time slots of the data frame). Each Transmit and Receive register would contain 8 bits of the data word. Figure 22-9: Data Packing Example for Long Data Words
Transmit Registers TXBUF0 TXBUF1 TXBUF2 Data Word bits 24:8 bits 7:0 0 00 0 0 000
TXBUF3
TSCON = RSCON = 0x0003 BLEN<1:0> = 01b
DS70069C-page 22-20

22.5.4 Multi-Channel Operation
The Multi-Channel mode (COFSM<1:0> = 00) is used for codecs that require a frame sync pulse that is driven high for one serial clock period to initiate a data transfer. One or more data words can be transferred in the data frame. The number of clock cycles between successive frame sync pulses will depend on the device connected to the DCI module. A timing diagram for the frame sync signal in Multi-Channel mode is shown in Figure 22-10. A timing example, indicating a four-word data transfer is also shown in Figure 22-2. Figure 22-10: Frame Sync Timing, Multi-Channel Mode
Frame Synch Sampled Here First Data Bit Sampled Here CSCK COFS
22
Data
MSB
LSB
22.5.4.1
Multi-Channel Setup Details The steps required to configure the DCI for a codec using the Multi-Channel mode are provided in this section. This Operating mode can be used for codecs with one or more data channels. The setup is similar regardless of the number of channels. For this setup example, a hypothetical codec will be considered. The single channel codec used for this setup example will use a 256 fs serial clock frequency with a 16-bit data word transmitted at the beginning of each frame. The steps required for setup and operation are described below. 1. 2. Determine the sample rate and data word size required by the codec. An 8 kHz sampling rate is assumed for this example. Determine the serial transfer clock frequency required by the codec. Most codecs require a serial clock signal that is some multiple of the sampling frequency. The example codec requires a frequency that is 256 fs, or 1.024 MHz. Therefore, a frame sync pulse must be generated every 256 serial clock cycles to start a data transfer. The DCI must be configured for the serial transfer clock. If the CSCK signal will be generated by the DCI, clear the CSCKD control bit (DCICON1<10>) and write a value to DCICON3 that will produce the correct clock frequency (See Section 22.4.3 Bit Clock Generator). If the CSCK signal is generated by the codec or other external source, set the CSCKD control bit and clear the DCICON3 register. Clear the COFSM<1:0> control bits (DCICON1<1:0>) to set the frame synchronization signal to Multi-Channel mode. If the DCI will generate the frame sync signal (master), then clear the COFSD control bit (DCICON1<8>). If the DCI will receive the frame sync signal (slave), then set the COFSD control bit. Clear the CSCKE control bit (DCICON1<9>) to sample incoming data on the falling edge of CSCK. This is the typical configuration for most codecs. Refer to the codec data sheet to ensure the correct sampling edge is used. Write the WS<3:0> control bits (DCICON2<3:0>) for the desired data word size. The example codec requires WS<3:0> = 1111b for a 16-bit data word size.
3.
4. 5.
6.
7.
DS70069C-page 22-21

8. Write the COFSG<3:0> control bits (DCICON2<8:5>) for the desired number of data words per frame. The WS and COFSG control bits will determine the length of the data frame in CSCK cycles (see Section 22.4.7 Frame Synchronization Generator) COFSG<3:0> = 1111b is used for this codec to provide the 256-bit data frame required by the example codec. Set the Output mode for the CSDO pin using the CSDOM control bit (DCICON1<6>). If a single device is attached to the DCI, CSDOM can be cleared. This will force the CSDO pin to 0 during unused data time slots. You may need to set CSDOM if multiple devices are attached to the CSDO pin. Write the TSCON and RSCON registers to determine which data time slots in the frame are to be transmitted and received, respectively. For this single channel codec, use TSCON = RSCON = 0x0001 to enable transmission and reception during the first 16-bit time slot of the data frame. Set the BLEN control bits (DCICON2<11:10>) to buffer the desired amount of data words. For the single channel codec, BLEN = 00 will provide an interrupt at each data frame. A higher value of BLEN could be used for this codec to buffer multiple samples between interrupts. If interrupts are to be used, clear the DCIIF status bit (IFS2<9>) and set the DCIIE control bit (IEC2<9>). Begin operation as described in Section 22.5.1.1 DCI Start-up and Data Buffering.
9.
10.
11.
12. 13.
22.5.5
I2S Operation
The I2S Operating mode is used for codecs that require a frame sync signal that has a 50% duty cycle. The period of the I2S frame sync signal in serial clock cycles is determined by the word size of the codec that is connected to the DCI module. The start of a new word boundary is marked by a high-to-low or a low-to-high transition edge on the COFS pin as shown in Figure 22-11. I2S codecs are generally stereo or two-channel devices, with one data word transferred during the low time of the frame sync signal and the other data word transmitted during the high time. Figure 22-11: I2S Interface Frame Sync Timing
Frame Synch Edge Sampled First Data Bit Sampled
CSCK Data MSB LSB MSB LSB
COFS
Note: A 5-bit transfer is shown here for illustration purposes. The I2S protocol does not specify word length, this will be system dependent.
The DCI module is configured for I2S mode by writing a value of 01h to the COFSM<1:0> control bits in the DCICON1 SFR. When operating in the I2S mode, the DCI module will generate frame synchronization signals with a 50% duty cycle. Each edge of the frame synchronization signal marks the boundary of a new data word transfer. Refer to the Appendix of this manual for more information about the I2S protocol. The user must also select the frame length and data word size using the COFSG and WS control bits in the DCICON2 SFR.
DS70069C-page 22-22

22.5.5.1 I2S Setup Details
The steps required to configure the DCI for an I2S codec are provided in this section. For this setup example, a hypothetical I2S codec will be considered. The I2S codec in this setup example will use a 64 fs serial clock frequency, with two 16 bit data words during the data frame. Therefore, the frame length will be 64 CSCK cycles, with the COFS signal high for 32 cycles and low for 32 cycles. The first data word will be transmitted one CSCK cycle after the falling edge of COFS, and the second data word will be transmitted one CSCK cycle after the rising edge of COFS as shown in Figure 22-11. 1. 2. 3. Determine the sample rate used by the codec to determine the CSCK frequency. It is assumed in this example that fs is 48 kHz. Determine the serial transfer clock frequency required by the codec. The example codec requires a frequency that is 64 fs, or 3.072 MHz. The DCI must be configured for the serial transfer clock. If the CSCK signal will be generated by the DCI, clear the CSCKD control bit (DCICON1<10>) and write a value to DCICON3 that will produce the correct clock frequency (see Section 22.4.3 Bit Clock Generator). If the CSCK signal is generated by the codec or other external source, set the CSCKD control bit and clear the DCICON3 register. Next, set COFSM<1:0> = 01b to set the frame synchronization signal to I2S mode. If the DCI will generate the frame sync signal (master), then clear the COFSD control bit (DCICON1<8>). If the DCI will receive the frame sync signal (slave), then set the COFSD control bit. Set the CSCKE control bit (DCICON1<9>) to sample incoming data on the rising edge of CSCK. This is the typical configuration for most I2S codecs. Write the WS<3:0> control bits (DCICON2<3:0>) for the desired data word size. For the example codec, use WS<3:0> = 1111b for a 16-bit data word size. Write the COFSG<3:0> control bits (DCICON2<8:5) for the desired number of data words per frame. The WS and COFSG control bits will determine the length of the data frame in CSCK cycles (see Section 22.4.7 Frame Synchronization Generator). For this example codec, set COFSG<3:0> = 0001b. Note: In the I2S mode, the COFSG bits are set to the length of 1/2 of the data frame. For this example codec, set COFSG<3:0> = 0001b (two data words per frame) to produce a 32-bit frame. This will produce an I2S data frame that is 64 bits in length.
22
4. 5.
6. 7. 8.
9.
10.
11.
12. 13.
Set the Output mode for the CSDO pin using the CSDOM control bit (DCICON1<6>). If a single device is attached to the DCI, CSDOM can be cleared. You may need to set CSDOM if multiple devices are attached to the CSDO pin. Write the TSCON and RSCON registers to determine which data time slots in the frame are to be transmitted and received, respectively. For this codec, set TSCON = 0x0001 and RSCON = 0x0001 to enable transmission and reception during the first 16-bit time slot of the 32-bit data frame. Adjacent time slots can be enabled to buffer data words longer than 16 bits. Set the BLEN<1:0> control bits (DCICON2<11:10>) to buffer the desired amount of data words. For a two-channel I2S codec, BLEN<1:0> = 01b will generate an interrupt after transferring two data words. If interrupts are to be used, clear the DCIIF status bit (IFS2<9>) and set the DCIIE control bit (IEC2<9>). Begin operation as described in Section 22.5.1.1 DCI Start-up and Data Buffering. In the I2S Master mode, the COFS pin will be driven high after the module is enabled and begin transmitting the data loaded in TXBUF0.
DS70069C-page 22-23

22.5.5.2 How to Determine the I2S Channel Alignment Most I2S codecs support two channels of data and the level of the frame sync signal indicates the channel that is transferred during that half of the data frame. The COFS pin can be polled in software using its associated Port register to determine the present level on the pin in the DCI Interrupt Service Routine. This will indicate which data is in the Receive register and which data should be written to the Transmit registers for transfer on the next frame. 22.5.5.3 I2S Data Justification As per the I2S specification, a data word transfer will by default begin one serial clock cycle following a transition of the frame sync signal. An MSb left-justified option can be selected using the DJST control bit (DCICON1<5>). If DJST = 1, the I2S data transfers will be MSb left justified. The MSb of the data word will be presented on the CSDO pin during the same serial clock cycle as the rising or falling edge of the FS signal. After the data word has been transmitted, the state of the CSDO pin is dictated by the CSDOM (DCICON1<6>) bit. The left-justified data option allows two stereo codecs to be connected to the same serial bus. Many I2S compatible devices have configuration options for left-justified or right-justified data. The word size selection bits are set to twice the codec word length and data is read/written to the DCI memory in a packed format. The connection details for a dual I2S codec system are shown in Figure 22-12. Timing diagrams for I2S mode are shown in Figure 22-13. For reference, these diagrams assume an 8-bit word size (WS<3:0> = 0111b). Two data words per frame would be required to achieve a 16-bit sub-frame (COFSG<3:0> = 0001b). The 3rd timing diagram in Figure 22-13 uses packed data to read/write from two codecs. For this example, the DCI module is configured for a 16-bit data word (WS<3:0> = 1111b). Two packed 8-bit words are written to each 16-bit location in the DCI memory buffer. Figure 22-12: Dual I2S Codec Interface
Codec #1 dsPIC Device
SCK FS SDO I2C/SPI
SCK WS SDI
Codec #2 SCK WS SDI I2C/SPI
DS70069C-page 22-24

Figure 22-13: I2S Data Justification Options
1. Standard I2S Data Alignment
CSCK COFS Data Channel #1 Transfer 7 6 5 4 3 2 1 0 Channel #2 Transfer 7 6 5 4 3 2 1 0
2. Left-Justified Data Alignment
22
CSCK COFS Data Channel #1 Transfer 7 6 5 4 3 2 1 0 Channel #2 Transfer 7 6 5 4 3 2 1 0
3. Left-Justified Data Alignment with Packed Data
CSCK COFS Data

Channel #1 Transfer, Word 1 Channel #1 Transfer, Word 2 Channel #2 Transfer, Word 1 Channel #2 Transfer, Word 2
7 6
5 4 3
2 1
0 7 6
5 4 3
2 1
0 7 6
5 4 3
2 1
0 7 6
5 4 3
2 1
22.5.6
AC-Link Operation
This section describes how to use the DCI in the AC-Link modes. The AC-Link modes are used to communicate with AC-97 compliant codec devices.
22.5.6.1
AC-Link Data Frame The AC-Link data frame is 256 bits subdivided into one 16-bit control slot, followed by twelve 20-bit data slots. The AC-97 codec usually provides the serial transfer clock signal which is derived from a crystal oscillator as shown in Figure 22-14. The controller receives the serial clock and generates the frame sync signal. The default data frame rate is 48 kHz. The frame sync signal used for AC-Link systems is high for 16 CSCK periods at the beginning of the data frame and low for 240 CSCK periods. The data transfer begins one CSCK period after the rising edge of the frame sync signal as shown in Figure 22-16. Data is sampled by the receiving device on the falling edge of CSCK. The control and data time slots in the AC-Link have defined uses in the protocol as shown in Figure 22-15. Refer to the Appendix of this manual or the Intel AC 97 Codec Specification, Rev 2.2 for a complete definition of the AC-Link protocol.
DS70069C-page 22-25

Figure 22-14: AC-Link Signal Connections
CSCK COFS dsPIC (AC 97 CSDO Controller) CSDI BIT_CLK SYNC SDATA_OUT SDATA_IN /RESET AC 97 Codec 24.576 MHz
I/O
Figure 22-15: AC-Link Data Frame

256 16 20 20 20 20 20 20 20
SYNC
SDATA_OUT
Tag Frame
Command Address
Command Data
Slot 3 Left PCM Data
Slot 4 Right PCM Data
Slot 10 Line 2 DAC
Slot 11 Handset DAC
Slot 12 Codec I/O Control
SDATA_IN
Tag Frame
Status Address
Status Data
Slot 3 Left PCM Data
Slot 10 Line 2 ADC
Slot 11 Handset ADC
Slot 12 Codec I/O Status
Figure 22-16: Frame Sync Timing, AC-Link Start of Frame

Frame Synch Edge Sampled First bit of Data Frame Sampled
CSCK S12 bit 2 S12 bit 1 S12 LSb Tag Tag Tag MSb bit 14 bit 13
Data
COFS
The DCI module has two Operating modes for the AC-Link protocol to accommodate the 20-bit data time slots. These Operating modes are selected by the COFSM<1:0> control bits (DCICON1<1:0>). The first AC-Link mode is called 16-bit AC-Link mode and is selected by setting COFSM<1:0> = 10b. The second AC-Link mode is called 20-bit AC-Link mode and is selected by setting COFSM<1:0> = 11b.
DS70069C-page 22-26

22.5.6.2 16-bit AC-Link Mode In the 16-bit AC-Link mode, transmit and receive data word lengths are restricted to 16 bits to fit the DCI Transmit and Receive registers. Note that this restriction only affects the 20-bit data time slots of the AC-Link protocol. For received time slots, the incoming data will be truncated to 16 bits. For outgoing time slots, the 4 LSbs of the data word are set to 0 by the module. This Operating mode simplifies the AC-Link data frame by treating every time slot as a 16-bit time slot. The frame sync generator maintains alignment to the time slot boundaries. 22.5.6.3 20-bit AC-Link Mode The 20-bit AC-Link mode allows all bits in the data time slots to be transmitted and received, but does not maintain data alignment to the specific time slot boundaries defined in the AC-Link protocol. The 20-bit AC-Link mode functions similarly to the Multi-Channel mode of the DCI module, except for the duty cycle of the frame synchronization signal that is produced. The AC-Link frame synchronization signal should remain high for 16 clock cycles and should be low for the following 240 cycles. The 20-bit mode treats each 256-bit AC-Link frame as sixteen 16-bit time slots. In the 20-bit AC-Link mode, the module operates as if COFSG<3:0> = 1111b and WS<3:0> = 1111b. The data alignment for 20-bit data slots is not maintained in this Operating mode. For example, an entire 256-bit AC-Link data frame can be transmitted and received in a packed fashion by setting all bits in the TSCON and RSCON registers. Since the total available buffer length is 64 bits, it would take 4 consecutive interrupts to transfer the AC-Link frame. The application software must keep track of the current AC-Link frame segment by monitoring the SLOT<3:0> status bits (DCISTAT<11:7>). 22.5.6.4 AC-Link Setup Details The module is enabled for AC-Link mode by writing 10h or 11h to the COFSM<1:0> control bits in the DCICON1 SFR. The word size selection bits (WS<3:0>) and the frame synchronization generator bits (COFSG<3:0>) have no effect for the 16 and 20-bit AC-Link modes since the frame and word sizes are set by the protocol. Most AC 97 codecs generate the clock signal that controls data transfers. Therefore, the CSCKD control bit is set in software. The COFSD control bit is cleared because the DCI will generate the FS signal from the incoming clock signal. The CSCKE bit is cleared so that data is sampled on the rising edge. The user must decide which time slots in the AC-Link data frame are to be buffered and set the TSE and RSE control bits accordingly. At a minimum, it will be necessary to buffer the transmit and receive TAG slots, so the TSCON<0> and RSCON<1> control bits should be set in software. Note: 1. 2. 3. 4. Only the TSCON<12:0> control bits and the RSCON<12:0> control bits will have an effect in the 16-bit AC-Link mode, since an AC-Link frame has 13 time slots.
22
The DCI must be configured to accept the serial transfer clock from the AC 97 codec. Set the CSCKD control bit and clear the DCICON3 register. Next, set the COFSM<1:0> control bits (DCICON1<1:0>) to 10b or 11b to set the desired AC-Link Frame Synchronization mode. Clear the COFSD control bit (DCICON1<8>), so the DCI will output the frame sync signal. Clear the CSCKE control bit (DCICON1<9>) to sample incoming data on the falling edge of CSCK. Note: The word size selection bits (WS<3:0>) and the frame synchronization generator bits (COFSG<3:0>) have no effect for the 16- and 20-bit AC-Link modes, since the frame and word sizes are set by the protocol.
DS70069C-page 22-27

5. 6. Clear the CSDOM control bit (DCICON1<6>). Write the TSCON and RSCON registers to determine which data time slots in the frame are to be transmitted and received, respectively. This will depend on which data time slots in the AC-Link protocol will be used. At a minimum, communication on slot #0 (Tag Slot) is required. Refer to the discussion in Section 22.5.6.2 16-bit AC-Link Mode, Section 22.5.6.3 20-bit AC-Link Mode and Section 26. Appendix of this manual for additional information. Set the BLEN control bits (DCICON2<11:10>) to buffer the desired amount of data words. For the single channel codec, BLEN = 00 will provide an interrupt at each data frame. A higher value of BLEN could be used for this codec to buffer multiple samples between interrupts. If interrupts are to be used, clear the DCIIF status bit (IFS2<9>) and set the DCIIE control bit (IEC2<9>). Begin operation as described in Section 22.5.1.1 DCI Start-up and Data Buffering.
7.
8. 9.
22.6 22.6.1
Operation in Power Saving Modes CPU Idle Mode

The DCI module may optionally continue to operate while the CPU is in Idle mode. The DCISIDL control bit (DCICON1<13>) determines whether the DCI module will operate when the CPU is in Idle mode. If the DCISIDL control bit is cleared (default), the module will continue to operate normally in Idle mode. If the DCISIDL bit is set, the module will halt when the CPU enters Idle mode.
22.6.2
Sleep Mode
The DCI will not operate while the device is in Sleep mode if the CSCK signal is derived from the device instruction clock, TCY. However, the DCI module has the ability to operate while in Sleep mode and wake the CPU when the CSCK signal is supplied by an external device (CSCKD = 1). The DCI interrupt enable bit, DCIIE, must be set to allow a wake-up event from Sleep mode. When the DCI interrupt flag, DCIIF is set, the device will wake from Sleep mode. If the DCI interrupt priority level is greater than the current CPU priority, program execution will resume from the DCI ISR. Otherwise, execution will resume with the instruction following the PWRSAV instruction that previously entered Sleep mode.
22.7
Registers Associated with DCI

Table 22-1 lists the registers associated with the DCI module.
DS70069C-page 22-28
Table 22-1:
Bit 13 FLTAIP<2:0> TSE13 RSE13 Receive #0 Data Register Receive #1 Data Register Receive #2 Data Register Receive #3 Data Register Transmit #0 Data Register Transmit #1 Data Register Transmit #2 Data Register Transmit #3 Data Register RSE12 RSE11 RSE10 RSE9 RSE8 RSE7 RSE6 RSE5 RSE4 RSE3 RSE2 RSE1 TSE12 TSE11 TSE10 TSE9 TSE8 TSE7 TSE6 TSE5 TSE4 TSE3 TSE2 TSE1 SLOT<3:0> ROV RFUL TUNF TSE0 RSE0 BCG<11:0> TMPTY BLEN<1:0> COFSG<3:0> WS<3:0> DCISIDL DLOOP CSCKD CSCKE COFSD UNFM SDOM DJST COFSM<1:0> LVDIP<2:0> DCIIP<2:0> QEIIP<2:0> FLTBIE FLTAIE LVDIE DCIIE QEIIE PWMIE C2IE INT4IE INT3IE OC8IE OC7IE OC6IE OC5IE FLTBIF FLTAIF LVDIF DCIIF QEIIF PWMIF C2IF INT4IF INT3IF OC8IF OC7IF OC6IF OC5IF Bit 12 Bit 11 Bit 10 Bit 9 Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
DCI Register Map

Value on all Resets 0000 0000 0000 0000 0000 0000 0000 0000 0100 0100 0100 0100 000- -000 000- --00 ---- 00-0 000- 0000 ---- 0000 0000 0000 ---- 0000 ---0 0000 0000 0000 0000 0000 0000 0000 0000 0000 uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu uuuu 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000
Name
Address
Bit 15
Bit 14
IFS2
0088
IEC2
0090
IPC10
00A8
DCICON1
240
DCIEN
DCICON2
242
DCICON3
244
DCISTAT
246
TSCON
248
TSE15
TSE14
RSCON
24C
RSE15
RSE14
RXBUF0
250
RXBUF1
252
RXBUF2
254
RXBUF3
256
TXBUF0
258
TXBUF1
25A
TXBUF2
25C
TXBUF3
25E
Legend: r = Reserved, x = Unknown, u = Unchanged.
Note:
Grayed locations indicate reserved space in SFR map for future module expansion. Read reserved locations as 0s.
DS70069C-page 22-29
22

22.8 Design Tips
Question 1: Can the DCI support data word lengths greater than 16-bits?
Answer: Yes. A long data word can be transmitted and received using multiple Transmit and Receive registers. See Section 22.5.3 Data Packing for Long Data Word Support for details.
DS70069C-page 22-30

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Data Converter Interface (DCI) module are: Title No related application notes at this time. Application Note #
Note:
22
DS70069C-page 22-31

Revision B
This revision incorporates additional technical content and changes for the dsPIC30F Data Converter Interface (DCI) module.
Revision C
DS70069C-page 22-32
Section 23. CAN Module

HIGHLIGHTS
This section of the manual contains the following major topics: 23.1 Introduction .................................................................................................................. 23-2 23.2 Control Registers for the CAN Module......................................................................... 23-2 23.3 CAN Module Features ............................................................................................... 23-28 23.4 CAN Module Implementation ..................................................................................... 23-29 23.5 CAN Module Operation Modes .................................................................................. 23-36 23.6 Message Reception ................................................................................................... 23-39 23.7 Transmission.............................................................................................................. 23-49 23.8 Error Detection........................................................................................................... 23-58 23.9 CAN Baud Rate ......................................................................................................... 23-60 23.10 Interrupts.................................................................................................................... 23-64 23.11 CAN Capture.............................................................................................................. 23-65 23.12 CAN Module I/O......................................................................................................... 23-65 23.13 Operation in CPU Power Saving Modes.................................................................... 23-66 23.14 CAN Protocol Overview ............................................................................................. 23-68 23.15 Related Application Notes.......................................................................................... 23-72 23.16 Revision History ......................................................................................................... 23-73
23
CAN Module
DS70070C-page 23-1

23.1 Introduction
The Controller Area Network (CAN) module is a serial interface useful for communicating with other peripherals or microcontroller devices. This interface/protocol was designed to allow communications within noisy environments. Figure 23-1 shows an example CAN bus network. Figure 23-1: Example CAN Bus Network
dsPIC30F with CAN
CAN bus
MCP2551 Transceiver
MCP2551 Transceiver Microchip MCP2510 SPI Interface PICmicro Microcontroller
MCP2551 Transceiver
MCP2551 Transceiver
MCP2551 Transceiver
dsPIC30F with integrated CAN
dsPIC30F with integrated CAN
PICmicro with integrated CAN
23.2
Control Registers for the CAN Module

There are many registers associated with the CAN module. Descriptions of these registers are grouped into sections. These sections are: Control and Status Registers Transmit Buffer Registers Receive Buffer Registers Baud Rate Control Registers Interrupt Status and Control Registers Note 1: i in the register identifier denotes the specific CAN module (CAN1 or CAN2). 2: n in the register identifier denotes the buffer, filter or mask number. 3: m in the register identifier denotes the word number within a particular CAN data field.
DS70070C-page 23-2
Section 23. CAN

23.2.1 CAN Control and Status Registers
CiCTRL: CAN Module Control and Status Register U-0 R/W-0 CSIDL R/W-0 ABAT R/W-0 CANCKS R/W-1 R/W-0 REQOP<2:0> bit 8 Lower Byte: R-1 bit 7 bit 15 CANCAP: CAN Message Receive Capture Enable bit 1 = Enable CAN capture 0 = Disable CAN capture Note: CANCAP is always writable, regardless of CAN module Operating mode. Unimplemented: Read as 0 CSIDL: Stop in Idle Mode bit 1 = Discontinue CAN module operation when device enters Idle mode 0 = Continue CAN module operation in Idle mode ABAT: Abort All Pending Transmissions bit 1 = Abort pending transmissions in all Transmit Buffers 0 = No effect Note: Module will clear this bit when all transmissions aborted. CANCKS: CAN Master Clock Select bit 1 = FCAN clock is FCY 0 = FCAN clock is 4 FCY REQOP<2:0>: Request Operation Mode bits 111 = Set Listen All Messages mode 110 = Reserved 101 = Reserved 100 = Set Configuration mode 011 = Set Listen Only mode 010 = Set Loopback mode 001 = Set Disable mode 000 = Set Normal Operation mode OPMODE<2:0>: Operation Mode bits Note: These bits indicate the current Operating mode of the CAN module. See description for REQOP bits (CiCTRL<10:8>). Unimplemented: Read as 0 R-0 OPMODE<2:0> R-0 U-0 R-0 R-0 ICODE<2:0> R-0 U-0 bit 0 R/W-0
Register 23-1: Upper Byte: R/W-x CANCAP bit 15
bit 14 bit 13
bit 12
23
CAN Module
bit 11
bit 10-8
bit 7-5
bit 4
DS70070C-page 23-3

Register 23-1: bit 3-1 CiCTRL: CAN Module Control and Status Register (Continued)
ICODE<2:0>: Interrupt Flag Code bits 111 = Wake-up interrupt 110 = RXB0 interrupt 101 = RXB1 interrupt 100 = TXB0 interrupt 011 = TXB1 interrupt 010 = TXB2 interrupt 001 = Error interrupt 000 = No interrupt Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
bit 0
DS70070C-page 23-4
Section 23. CAN

23.2.2 CAN Transmit Buffer Registers
This subsection describes the CAN Transmit Buffer Register and the associated Transmit Buffer Control Registers. Register 23-2: Upper Byte: U-0 bit 15 Lower Byte: U-0 bit 7 bit 15-7 bit 6 Unimplemented: Read as '0' TXABT: Message Aborted bit 1 = Message was aborted 0 = Message has not been aborted Note: This bit is cleared when TXREQ is set. TXLARB: Message Lost Arbitration bit 1 = Message lost arbitration while being sent 0 = Message did not lose arbitration while being sent Note: This bit is cleared when TXREQ is set. TXERR: Error Detected During Transmission bit 1 = A bus error occurred while the message was being sent 0 = A bus error did not occur while the message was being sent Note: This bit is cleared when TXREQ is set. TXREQ: Message Send Request bit 1 = Request message transmission 0 = Abort message transmission if TXREQ already set, otherwise no effect Note: The bit will automatically clear when the message is successfully sent. Unimplemented: Read as 0 TXPRI<1:0>: Message Transmission Priority bits 11 = Highest message priority 10 = High intermediate message priority 01 = Low intermediate message Priority 00 = Lowest message priority Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R-0 TXABT R-0 TXLARB R-0 TXERR R/W-0 TXREQ U-0 R/W-0 R/W-0 bit 0 TXPRI<1:0> U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8 CiTXnCON: Transmit Buffer Status and Control Register
bit 5
23
CAN Module
bit 4
bit 3
bit 2 bit 1-0
DS70070C-page 23-5

Register 23-3: Upper Byte: R/W-x bit 15 Lower Byte: R/W-x bit 7 bit 15-11 SID<10:6>: Standard Identifier bits bit 10-8 bit 7-2 bit 1 Unimplemented: Read as 0 SID<6:0>: Standard Identifier bits SRR: Substitute Remote Request Control bit 1 = Message will request a remote transmission 0 = Normal message. TXIDE: Extended Identifier bit
1 = Message will transmit extended identifier 0 = Message will transmit standard identifier
CiTXnSID: Transmit Buffer n Standard Identifier R/W-x R/W-x SID<10:6> R/W-x R/W-x U-0 U-0 U-0 bit 8
R/W-x
R/W-x SID<5:0>
R/W-x
R/W-x
R/W-x
R/W-x SRR
R/W-x TXIDE bit 0
bit 0
Register 23-4: Upper Byte: R/W-x bit 15
CiTXnEID: Transmit Buffer n Extended Identifier R/W-x R/W-x R/W-x U-0 U-0 U-0 U-0 bit 8 Lower Byte: R/W-x bit 7 R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x bit 0 EID<13:6>
EID<17:14>
bit 15-12 EID<17:14>: Extended Identifier bits 17-14 bit 11-8 bit 7-0 Unimplemented: Read as 0 EID<13:6>: Extended Identifier bits 13-6 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70070C-page 23-6
Section 23. CAN

Register 23-5: Upper Byte: R/W-x bit 15 Lower Byte: R/W-x TXRB0 bit 7 bit 15-10 EID<5:0>: Extended Identifier bits 5-0 bit 9 TXRTR: Remote Transmission Request bit 1 = Message will request a remote transmission 0 = Normal message TXRB<1:0>: Reserved Bits Note: User must set these bits to 0 according to CAN protocol. DLC<3:0>: Data Length Code bits Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = bit is cleared x = Bit is unknown R/W-x R/W-x R/W-x R/W-x U-0 U-0 U-0 bit 0 DLC<3:0> R/W-x R/W-x EID<5:0> R/W-x R/W-x R/W-x R/W-x TXRTR R/W-x TXRB1 bit 8 CiTXnDLC: Transmit Buffer n Data Length Control
bit 8-7 bit 6-3 bit 2-0
23
CAN Module
Register 23-6: Upper Byte: R/W-x bit 15
CiTXnBm: Transmit Buffer n Data Field Word m R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x bit 8 Lower Byte: R/W-x bit 7 R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x bit 0 CTXB<7:0>
CTXB<15:8>
bit 15-0
CTXB<15:0>: Data Field Buffer Word bits (2 bytes) Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = bit is cleared x = Bit is unknown
DS70070C-page 23-7

23.2.3 CAN Receive Buffer Registers
This subsection shows the Receive buffer registers with their associated control registers. Register 23-7: Upper Byte: U-0 bit 15 Lower Byte: R/C-0 RXFUL bit 7 bit 15-8 bit 7 Unimplemented: Read as 0 RXFUL: Receive Full Status bit 1 = Receive buffer contains a valid received message 0 = Receive buffer is open to receive a new message Note: This bit is set by the CAN module and should be cleared by software after the buffer is read. Unimplemented: Read as 0 RXRTRRO: Received Remote Transfer Request bit (read only) 1 = Remote Transfer Request was received 0 = Remote Transfer Request not received Note: This bit reflects the status of the last message loaded into Receive Buffer 0. DBEN: Receive Buffer 0 Double Buffer Enable bit 1 = Receive Buffer 0 overflow will write to Receive Buffer 1 0 = No Receive Buffer 0 overflow to Receive Buffer 1 JTOFF: Jump Table Offset bit (read only copy of DBEN) 1 = Allows Jump Table offset between 6 and 7 0 = Allows Jump Table offset between 0 and 1 FILHIT0: Indicates Which Acceptance Filter Enabled the Message Reception bit 1 = Acceptance Filter 1 (RXF1) 0 = Acceptance Filter 0 (RXF0) Note: This bit reflects the status of the last message loaded into Receive Buffer 0. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set C = Bit can be cleared 0 = Bit is cleared U = Unimplemented bit, read as 0 x = Bit is unknown U-0 U-0 U-0 R-0 RXRTRRO R/W-0 DBEN R/W-0 JTOFF R-0 FILHIT0 bit 0 U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8 CiRX0CON: Receive Buffer 0 Status and Control Register
bit 6-4 bit 3
bit 2
bit 1
bit 0
DS70070C-page 23-8
Section 23. CAN

Register 23-8: Upper Byte: U-0 bit 15 Lower Byte: R/C-0 RXFUL bit 7 bit 15-8 bit 7 Unimplemented: Read as 0 RXFUL: Receive Full Status bit 1 = Receive buffer contains a valid received message 0 = Receive buffer is open to receive a new message Note: This bit is set by the CAN module and should be cleared by software after the buffer is read. Unimplemented: Read as 0 RXRTRRO: Received Remote Transfer Request bit (read only) 1 = Remote transfer request was received 0 = Remote transfer request not received Note: This bit reflects the status of the last message loaded into Receive Buffer 1. FILHIT<2:0>: Indicates Which Acceptance Filter Enabled the Message Reception bits 101 = Acceptance filter 5 (RXF5) 100 = Acceptance filter 4 (RXF4) 011 = Acceptance filter 3 (RXF3) 010 = Acceptance filter 2 (RXF2) 001 = Acceptance filter 1 (RXF1) (Only possible when DBEN bit is set) 000 = Acceptance filter 0 (RXF0) (Only possible when DBEN bit is set) Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown U-0 U-0 U-0 R-0 RXRTRRO R-0 R-0 FILHIT<2:0> bit 0 R-0 U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8 CiRX1CON: Receive Buffer 1 Status and Control Register
bit 6-4 bit 3
bit 2-0
23
CAN Module
DS70070C-page 23-9

Register 23-9: Upper Byte: U-0 bit 15 Lower Byte: R/W-x bit 7 bit 15-13 Unimplemented: Read as 0 bit 12-2 bit 1 SID<10:0>: Standard Identifier bits SRR: Substitute Remote Request bit (Only when RXIDE = 1) 1 = Remote transfer request occured 0 = No remote transfer request occured RXIDE: Extended Identifier Flag bit 1 = Received message is an extended data frame, SID<10:0> are EID<28:18> 0 = Received message is a standard data frame Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-x R/W-x SID<5:0> R/W-x R/W-x R/W-x R/W-x SRR R/W-x RXIDE bit 0 U-0 U-0 R/W-x R/W-x R/W-x SID<10:6> bit 8 R/W-x R/W-x CiRXnSID: Receive Buffer n Standard Identifier
bit 0
Register 23-10: CiRXnEID: Receive Buffer n Extended Identifier Upper Byte: U-0 bit 15 Lower Byte: R/W-x bit 7 bit 15-12 Unimplemented: Read as 0 bit 11-0 EID<17:6>: Extended Identifier bits 17-6 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x bit 0 EID<13:6> U-0 U-0 U-0 R/W-x R/W-x R/W-x R/W-x bit 8 EID<17:14>
DS70070C-page 23-10
Section 23. CAN

Register 23-11: CiRXnBm: Receive Buffer n Data Field Word m Upper Byte: R/W-x bit 15 Lower Byte: R/W-x bit 7 bit 15-0 CRXB<15:0>: Data Field Buffer Word bits (2 bytes) Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x bit 0 CRXB<7:0> R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x bit 8 CRXB<15:8>
Register 23-12: CiRXnDLC: Receive Buffer n Data Length Control Upper Byte: R/W-x bit 15 Lower Byte: U-0 bit 7 bit 15-10 EID<5:0>: Extended Identifier bits bit 9 RXRTR: Receive Remote Transmission Request bit
1 = Remote transfer request 0 = No remote transfer request
23
R/W-x RXRTR R/W-x RB1 bit 8
R/W-x
R/W-x EID<5:0>
R/W-x
R/W-x
R/W-x
CAN Module
U-0
U-0
R/W-x RB0
R/W-x
R/W-x
R/W-x
R/W-x bit 0
DLC<3:0>
Note: bit 8 bit 4 bit 3-0
This bit reflects the status of the RTR bit in the last received message.
RB1: Reserved bit 1 Reserved by CAN Spec and read as 0 RB0: Reserved bit 0 Reserved by CAN Spec and read as 0 DLC<3:0>: Data Length Code bits (Contents of Receive Buffer) Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown
DS70070C-page 23-11

23.2.4 Message Acceptance Filters
This subsection describes the Message Acceptance filters. Register 23-13: CiRXFnSID: Acceptance Filter n Standard Identifier Upper Byte: U-0 bit 15 Lower Byte: R/W-x bit 7 bit 15-13 Unimplemented: Read as 0 bit 12-2 bit 1 bit 0 SID<10:0>: Standard Identifier bits Unimplemented: Read as 0 EXIDE: Extended Identifier Enable bits If MIDE = 1, then 1 = Enable filter for extended identifier 0 = Enable filter for standard identifier If MIDE = 0, then EXIDE is dont care Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-x R/W-x SID<5:0> R/W-x R/W-x R/W-x U-0 R/W-x EXIDE bit 0 U-0 U-0 R/W-x R/W-x R/W-x SID<10:6> bit 8 R/W-x R/W-x
Register 23-14: CiRXFnEIDH: Acceptance Filter n Extended Identifier High Upper Byte: U-0 bit 15 Lower Byte: R/W-x bit 7 bit 15-12 Unimplemented: Read as 0 bit 11-0 EID<17:6>: Extended Identifier bits 17-6 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x bit 0 EID<13:6> U-0 U-0 U-0 R/W-x R/W-x R/W-x R/W-x bit 8 EID<17:14>
DS70070C-page 23-12
Section 23. CAN

Register 23-15: CiRXFnEIDL: Acceptance Filter n Extended Identifier Low Upper Byte: R/W-x bit 15 Lower Byte: U-0 bit 7 bit 15-10 EID<5:0>: Extended Identifier bits bit 9-0 Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 0 R/W-x R/W-x EID<5:0> R/W-x R/W-x R/W-x U-0 U-0 bit 8
23
CAN Module
DS70070C-page 23-13

23.2.5 Acceptance Filter Mask Registers
Register 23-16: CiRXMnSID: Acceptance Filter Mask n Standard Identifier Upper Byte: U-0 bit 15 Lower Byte: R/W-x bit 7 bit 15-13 Unimplemented: Read as 0 bit 12-2 SID<10:0>: Standard Identifier Mask bits 1 = Include bit in the filter comparison 0 = Dont include bit in the filter comparison Unimplemented: Read as 0 MIDE: Identifier Mode Selection bit 1 = Match only message types (standard or extended address) as determined by EXIDE bit in filter 0 = Match either standard or extended address message if the filters match Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-x R/W-x SID<5:0> R/W-x R/W-x R/W-x U-0 R/W-x MIDE bit 0 U-0 U-0 R/W-x R/W-x R/W-x SID<10:6> bit 8 R/W-x R/W-x
bit 1 bit 0
Register 23-17: CiRXMnEIDH: Acceptance Filter Mask n Extended Identifier High Upper Byte: U-0 bit 15 Lower Byte: R/W-x bit 7 bit 15-12 Unimplemented: Read as 0 bit 11-0 EID<17:6>: Extended Identifier Mask bits 17-6 1 = Include bit in the filter comparison 0 = Dont include bit in the filter comparison Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x bit 0 EID<13:6> U-0 U-0 U-0 R/W-x R/W-x R/W-x R/W-x bit 8 EID<17:14>
DS70070C-page 23-14
Section 23. CAN

Register 23-18: CiRXMnEIDL: Acceptance Filter Mask n Extended Identifier Low Upper Byte: R/W-x bit 15 Lower Byte: U-0 bit 7 bit 15-10 EID<5:0>: Extended Identifier bits bit 9-0 Unimplemented: Read as 0 Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 0 R/W-x R/W-x EID<5:0> R/W-x R/W-x R/W-x U-0 U-0 bit 8
23
CAN Module
DS70070C-page 23-15

23.2.6 CAN Baud Rate Registers
This subsection describes the CAN baud rate registers. Register 23-19: CiCFG1: Baud Rate Configuration Register 1 Upper Byte: U-0 bit 15 Lower Byte: R/W-0 bit 7 bit 15-8 bit 7-6 Unimplemented: Read as 0 SJW<1:0>: Synchronized Jump Width bits 11 = Synchronized jump width time is 4 x TQ 10 = Synchronized jump width time is 3 x TQ 01 = Synchronized jump width time is 2 x TQ 00 = Synchronized jump width time is 1 x TQ BRP<5:0>: Baud Rate Prescaler bits 11 1111 = TQ = 2 x (BRP + 1)/FCAN = 128/FCAN 11 1110 = TQ = 2 x (BRP + 1)/FCAN = 126/FCAN . . . 00 0001 = TQ = 2 x (BRP + 1)/FCAN = 4/FCAN 00 0000 = TQ = 2 x (BRP + 1)/FCAN = 2/FCAN Note: FCAN is FCY or 4 FCY, depending on the CANCKS bit setting. Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 bit 0 SJW<1:0> BRP<5:0> U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8
bit 5-0
DS70070C-page 23-16
Section 23. CAN

Register 23-20: CiCFG2: Baud Rate Configuration Register 2 Upper Byte: U-0 bit 15 Lower Byte: R/W-x SEG2PHT S bit 7 bit 15 bit 14 Unimplemented: Read as 0 WAKFIL: Select CAN bus Line Filter for Wake-up bit 1 = Use CAN bus line filter for wake-up 0 = CAN bus line filter is not used for wake-up SEG2PH<2:0>: Phase Buffer Segment 2 bits 111 = length is 8 x TQ . . 000 = length is 1 x TQ SEG2PHTS: Phase Segment 2 Time Select bit 1 = Freely programmable 0 = Maximum of SEG1PH or information processing time (3 TQs), whichever is greater SAM: Sample of the CAN bus Line bit 1 = Bus line is sampled three times at the sample point 0 = Bus line is sampled once at the sample point SEG1PH<2:0>: Phase Buffer Segment 1 bits 111 = length is 8 x TQ . . 000 = length is 1 x TQ PRSEG<2:0>: Propagation Time Segment bits 111 = length is 8 x TQ . . 000 = length is 1 x TQ Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-x SAM R/W-x R/W-x SEG1PH<2:0> R/W-x R/W-x R/W-x PRSEG<2:0> bit 0 R/W-x R/W-x WAKFIL U-0 U-0 U-0 R/W-x R/W-x SEG2PH<2:0> bit 8 R/W-x
23
CAN Module
bit 7
bit 6
bit 5-3
bit 2-0
DS70070C-page 23-17

23.2.7 CAN Module Error Count Register
This subsection describes the CAN Module Transmission/Reception Error Count register. The various error status flags are present in the CAN Interrupt Flag Register. Register 23-21: CiEC: Transmit/Receive Error Count Upper Byte: R-0 bit 15 Lower Byte: R-0 bit 7 bit 15-8 bit 7-0 TERRCNT<7:0>: Transmit Error Count bits RERRCNT<7:0>: Receive Error Count bits Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R-0 R-0 R-0 R-0 R-0 R-0 R-0 bit 0 RERRCNT<7:0> R-0 R-0 R-0 R-0 R-0 R-0 R-0 bit 8 TERRCNT<7:0>
DS70070C-page 23-18
Section 23. CAN

23.2.8 CAN Interrupt Registers
This subsection documents the CAN Registers which are associated with interrupts. Register 23-22: CiINTE: Interrupt Enable Register Upper Byte: U-0 bit 15 Lower Byte: R/W-0 IVRIE bit 7 bit 15-8 bit 7 Unimplemented: Read as 0 IVRIE: Invalid Message Received Interrupt Enable bit 1 = Enabled 0 = Disabled WAKIE: Bus Wake Up Activity Interrupt Enable bit 1 = Enabled 0 = Disabled ERRIE: Error Interrupt Enable bit 1 = Enabled 0 = Disabled TX2IE: Transmit Buffer 2 Interrupt Enable bit 1 = Enabled 0 = Disabled TX1IE: Transmit Buffer 1 Interrupt Enable bit 1 = Enabled 0 = Disabled TX0IE: Transmit Buffer 0 Interrupt Enable bit 1 = Enabled 0 = Disabled RX1IE: Receive Buffer 1 Interrupt Enable bit 1 = Enabled 0 = Disabled RX0IE: Receive Buffer 0 Interrupt Enable bit 1 = Enabled 0 = Disabled Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set U = Unimplemented bit, read as 0 0 = Bit is cleared x = Bit is unknown R/W-0 WAKIE R/W-0 ERRIE R/W-0 TX2IE R/W-0 TX1IE R/W-0 TX0IE R/W-0 RX1IE R/W-0 RX0IE bit 0 U-0 U-0 U-0 U-0 U-0 U-0 U-0 bit 8
bit 6
23
CAN Module
bit 5
bit 4
bit 3
bit 2
bit 1
bit 0
DS70070C-page 23-19

Register 23-23: CiINTF: Interrupt Flag Register Upper Byte: R/C-0 RX0OVR bit 15 Lower Byte: R/W-0 IVRIF bit 7 bit 15 RX0OVR: Receive Buffer 0 Overflowed bit 1 = Receive buffer 0 overflowed 0 = Receive buffer 0 not overflowed RX1OVR: Receive Buffer 1 Overflowed bit 1 = Receive buffer 1 overflowed 0 = Receive buffer 1 not overflowed TXBO: Transmitter in Error State, Bus Off bit 1 = Transmitter in error state, bus off 0 = Transmitter not in error state, bus off TXEP: Transmitter in Error State, Bus Passive bit 1 = Transmitter in error state, bus passive 0 = Transmitter not in error state, bus passive RXEP: Receiver in Error State, Bus Passive bit 1 = Receiver in error state, bus passive 0 = Receiver not in error state, bus passive TXWAR: Transmitter in Error State, Warning bit 1 = Transmitter in error state, warning 0 = Transmitter not in error state, warning RXWAR: Receiver in Error State, Warning bit 1 = Receiver in error state, warning 0 = Receiver not in error state, warning EWARN: Transmitter or Receiver is in Error State, Warning bit 1 = Transmitter or receiver is in error state, warning 0 = Transmitter and receiver are not in error state IVRIF: Invalid Message Received Interrupt Flag bit 1 = Some type of error occurred during reception of the last message 0 = Receive error has not occurred WAKIF: bus Wake-up Activity Interrupt Flag bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred ERRIF: Error Interrupt Flag bit (multiple sources in CiINTF<15:8> register) 1 = Interrupt request has occurred 0 = Interrupt request has not occurred TX2IF: Transmit Buffer 2 Interrupt Flag bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred TX1IF: Transmit Buffer 1 Interrupt Flag bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred R/W-0 WAKIF R/W-0 ERRIF R/W-0 TX2IF R/W-0 TX1IF R/W-0 TX0IF R/W-0 RX1IF R/W-0 RX0IF bit 0 R/C-0 RX1OVR R-0 TXBO R-0 TXEP R-0 RXEP R-0 TXWAR R-0 RXWAR R-0 EWARN bit 8
bit 14
bit 13
bit 12
bit 11
bit 10
bit 9
bit 8
bit 7
bit 6
bit 5
bit 4
bit 3
DS70070C-page 23-20
Section 23. CAN

Register 23-23: bit 2 CiINTF: Interrupt Flag Register (Continued)
TX0IF: Transmit Buffer 0 Interrupt Flag bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred RX1IF: Receive Buffer 1 Interrupt Flag bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred RX0IF: Receive Buffer 0 Interrupt Flag bit 1 = Interrupt request has occurred 0 = Interrupt request has not occurred Legend: R = Readable bit -n = Value at POR W = Writable bit 1 = Bit is set C = Bit can be cleared 0 = Bit is cleared U = Unimplemented bit, read as 0 x = Bit is unknown
bit 1
bit 0
23
CAN Module
DS70070C-page 23-21
Table 23-1:
Bit 14 SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<5:0> EID<13:6> SID<5:0> EID<13:6> MIDE SID<5:0> EID<13:6> MIDE SID<5:0> EID<13:6> EXIDE SID<5:0> EID<13:6> EXIDE EID<13:6> EXIDE SID<5:0> EID<13:6> EXIDE SID<5:0> EID<13:6> EXIDE SID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EXIDE 13 12 11 10 9 8 7 6 5 4 3 2 1 0 xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx Reset
CAN1 Register Map
File Name
ADR
15
C1RXF0SID
300
C1RXF0EIDH
302
DS70070C-page 23-22
C1RXF0EIDL
304
unused
306
C1RXF1SID
308
C1RXF1EIDH
30A
C1RXF1EIDL
30C
unused
30E
C1RXF2SID
310
C1RXF2EIDH
312
C1RXF2EIDL
314
unused
316
C1RXF3SID
318
C1RXF3EIDH
31A
C1RXF3EIDL
31C
unused
31E
C1RXF4SID
320
C1RXF4EIDH
322
C1RXF4EIDL
324
unused
326
C1RXF5SID
328
C1RXF5EIDH
32A
C1RXF5EIDL
32C
unused
32E
C1RXM0SID
330
C1RXM0EIDH
332
C1RXM0EIDL
334
unused
336
C1RXM1SID
338
C1RXM1EIDH
33A
C1RXM1EIDL
33C
unused
33E
Table 23-1:
Bit 14 SID<10:6> SID<5:0> EID<13:6> TX RB0 Transmit Buffer 0 Byte 0 Transmit Buffer 0 Byte 2 Transmit Buffer 0 Byte 4 Transmit Buffer 0 Byte 6 TX RTR TX RB1 TX RB0 SID<5:0> EID<13:6> DLC<3:0> Transmit Buffer 0 Byte 0 Transmit Buffer 0 Byte 2 Transmit Buffer 0 Byte 4 Transmit Buffer 0 Byte 6 TX RTR TX RB1 TX RB0 TX ABT TX LARB TX ERR SID<5:0> EID<13:6> DLC<3:0> Transmit Buffer 0 Byte 0 Transmit Buffer 0 Byte 2 Transmit Buffer 0 Byte 4 Transmit Buffer 0 Byte 6 TX ABT TX LARB TX ERR TX REQ TXPRI[1:0] TX REQ TXPRI[1:0] SRR TX IDE TX ABT TX LARB TX ERR TX REQ TXPRI[1:0] SRR TX IDE DLC<3:0> SRR EID<17:14> EID<5:0> Transmit Buffer 0 Byte 1 Transmit Buffer 0 Byte 3 Transmit Buffer 0 Byte 5 Transmit Buffer 0 Byte 7 SID<10:6> EID<17:14> EID<5:0> Transmit Buffer 0 Byte 1 Transmit Buffer 0 Byte 3 Transmit Buffer 0 Byte 5 Transmit Buffer 0 Byte 7 SID<10:6> EID<17:14> EID<5:0> Transmit Buffer 0 Byte 1 Transmit Buffer 0 Byte 3 Transmit Buffer 0 Byte 5 Transmit Buffer 0 Byte 7 TX RTR TX RB1 TX IDE 13 12 11 10 9 8 7 6 5 4 3 2 1 0
CAN1 Register Map (Continued)

Reset
File Name
ADR
15
C1TX2SID
340
xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000 xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000 xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000
C1TX2EID
342
C1TX2DLC
342
C1TX2B1
346
CAN Module
C1TX2B2
348
C1TX2B3
34A
C1TX2B4
34C
C1TX2CON
34E
C1TX1SID
350
C1TX1EID
352
C1TX1DLC
352
C1TX1B1
356
C1TX1B2
358
C1TX1B3
35A
C1TX1B4
35C
C1TX1CON
35E
C1TX0SID
360
C1TX0EID
362
C1TX0DLC
362
C1TX0B1
366
C1TX0B2
368
C1TX0B3
36A
C1TX0B4
36C
Section 23. CAN
DS70070C-page 23-23
C1TX0CON
36E
23
Table 23-1:
Bit 14 SID<10:6> EID<17:14> RX RTR Receive Buffer 1 Byte 0 Receive Buffer 1 Byte 2 Receive Buffer 1 Byte 4 Receive Buffer 1 Byte 6 RX FUL SID<5:0> EID<13:6> RX RB1 RX RB0 Receive Buffer 0 Byte 0 Receive Buffer 0 Byte 2 Receive Buffer 0 Byte 4 Receive Buffer 0 Byte 6 RX FUL RX ERR OPMODE[2:0] SJW[1:0]S SEG2 PHTS E WARN IVR IF IVR IE SAM WAK IF WAK IE ERR IF ERR IE SEG1PH[2:0] TXB2 IF TXB2 IE TXB1 IF TXB1 IE Receive Error Counter TXB0 IF TXB0 IE RX RTR R0 RXB0 DBEN ICODE[2:0] BRP[5:0] PRSEG[2:0] RXB1 IF RXB1 IE RXB0 IF RXB0 IE JTOFF FIL HIT 0 DLC[3:0] RX ERR RX RTR R0 FILHIT[2:0] RX RB1 RX RB0 DLC[3:0] EID<13:6> SID<5:0> SRR RX IDE EID<0:5> Receive Buffer 1 Byte 1 Receive Buffer 1 Byte 3 Receive Buffer 1 Byte 5 Receive Buffer 1 Byte 7 13 12 11 10 9 8 7 6 5 4 3 2 1 0 xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000 Reset
File Name
ADR
15
C1RX1SID
370
DS70070C-page 23-24
EID<17:14> RX RTR EID<0:5> Receive Buffer 0 Byte 1 Receive Buffer 0 Byte 3 Receive Buffer 0 Byte 5 Receive Buffer 0 Byte 7 SID<10:6> SRR RX IDE xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000 WAK FIL TXBO Transmit Error Counter TXBP RXBP TX WARN RX WARN RXB1 OVR SEG2PH[2:0] C SIDL ABAT CAN CKS REQOP[2:0] 0480 0000 0000 0000 0000 0000 xxxx
C1RX1EID
372
C1RX1DLC
374
C1RX1B1
376
C1RX1B2
378
C1RX1B3
37A
C1RX1B4
37C
C1RX1CON
37E
C1RX1SID
380
C1RX1EID
382
C1RX1DLC
384
C1RX0B1
386
C1RX0B2
388
C1RX0B3
38A
C1RX0B4
38C
C1RX0CON
38E
C1CTRL
390
C1CFG1
392
CAN CAP
C1CFG2
394
C1INTF
396
RXB0 OVR
C1INTE
398
C1EC
39A
Reserved
39C 3FE
Legend: x = Unknown
Table 23-2:
Bit 14 SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<10:6> EID<17:14> SID<5:0> EID<13:6> SID<5:0> EID<13:6> MIDE SID<5:0> EID<13:6> MIDE SID<5:0> EID<13:6> EXIDE SID<5:0> EID<13:6> EXIDE SID<5:0> EID<13:6> EXIDE EID<13:6> EXIDE SID<5:0> EID<13:6> EXIDE SID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> EID<5:0> 13 12 11 10 9 8 7 6 5 4 3 2 1 0 EXIDE
CAN2 Register Map

Reset
File Name
ADR
15
C2RXF0SID
3C0
xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx
C2RXF0EIDH
3C2
C2RXF0EIDL
3C4
unused
3C6
C2RXF1SID
3C8
CAN Module
C2RXF1EIDH
3CA
C2RXF1EIDL
3CC
unused
3CE
C2RXF2SID
3D0
C2RXF2EIDH
3D2
C2RXF2EIDL
3D4
unused
3D6
C2RXF3SIDH
3D8
C2RXF3EID
3DA
C2RXF3EIDL
3DC
unused
3DE
C2RXF4SID
3E0
C2RXF4EIDH
3E2
C2RXF4EIDL
3E4
unused
3E6
C2RXF5SID
3E8
C2RXF5EIDH
3EA
C2RXF5EIDL
3EC
unused
3EE
C2RXM0SID
3F0
C2RXM0EIDH
3F2
C2RXM0EIDL
3F4
unused
3F6
C2RXM1SID
3F8
C2RXM1EIDH
3FA
C2RXM1EIDL
3FC
Section 23. CAN
DS70070C-page 23-25
unused
3FE
23
Table 23-2:
Bit 14 SID<10:6> SID<5:0> EID<13:6> TX RB0 Transmit Buffer 0 Byte 0 Transmit Buffer 0 Byte 2 Transmit Buffer 0 Byte 4 Transmit Buffer 0 Byte 6 TX RTR TX RB1 TX RB0 DLC<3:0> Transmit Buffer 0 Byte 0 Transmit Buffer 0 Byte 2 Transmit Buffer 0 Byte 4 Transmit Buffer 0 Byte 6 TX RTR TX RB1 TX RB0 TX ABT TX LARB TX ERR SID<5:0> EID<13:6> DLC<3:0> Transmit Buffer 0 Byte 0 Transmit Buffer 0 Byte 2 Transmit Buffer 0 Byte 4 Transmit Buffer 0 Byte 6 TX ABT TX LARB TX ERR TX REQ TXPRI[1:0] TX REQ TXPRI[1:0] SRR TX IDE SID<5:0> EID<13:6> TX ABT TX LARB TX ERR TX REQ TXPRI[1:0] SRR TX IDE DLC<3:0> SRR TX IDE EID<17:14> EID<5:0> Transmit Buffer 0 Byte 1 Transmit Buffer 0 Byte 3 Transmit Buffer 0 Byte 5 Transmit Buffer 0 Byte 7 SID<10:6> EID<17:14> EID<5:0> Transmit Buffer 0 Byte 1 Transmit Buffer 0 Byte 3 Transmit Buffer 0 Byte 5 Transmit Buffer 0 Byte 7 SID<10:6> EID<17:14> EID<5:0> Transmit Buffer 0 Byte 1 Transmit Buffer 0 Byte 3 Transmit Buffer 0 Byte 5 Transmit Buffer 0 Byte 7 TX RTR TX RB1 13 12 11 10 9 8 7 6 5 4 3 2 1 0 xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000 xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000 xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000 Reset
File Name
ADR
15
C2TX2SID
400
DS70070C-page 23-26
C2TX2EID
402
C2TX2DLC
404
C2TX2B1
406
C2TX2B2
408
C2TX2B3
40A
C2TX2B4
40C
C2TX2CON
40E
C2TX1SID
410
C2TX1EID
412
C2TX1DLC
414
C2TX1B1
416
C2TX1B2
418
C2TX1B3
41A
C2TX1B4
41C
C2TX1CON
41E
C2TX0SID
420
C2TX0EID
422
C2TX0DLC
424
C2TX0B1
426
C2TX0B2
428
C2TX0B3
42A
C2TX0B4
42C
C2TX0CON
42E
Table 23-2:
Bit 14 SID<10:6> EID<17:14> RX RTR Receive Buffer 1 Byte 0 Receive Buffer 1 Byte 2 Receive Buffer 1 Byte 4 Receive Buffer 1 Byte 6 RX FUL SID<5:0> EID<13:6> RX RB1 RX RB0 Receive Buffer 0 Byte 0 Receive Buffer 0 Byte 2 Receive Buffer 0 Byte 4 Receive Buffer 0 Byte 6 RX FUL RX ERR OPMODE[2:0] SJW[1:0]S SEG2 PHTS E WARN IVR IF IVR IE SAM WAK IF WAK IE ERR IF ERR IE SEG1PH[2:0] TXB2 IF TXB2 IE TXB1 IF TXB1 IE Receive Error Counter TXB0 IF TXB0 IE RX RTR R0 RXB0 DBEN ICODE[2:0] BRP[5:0] PRSEG[2:0] RXB1 IF RXB1 IE RXB0 IF RXB0 IE JTOFF FIL HIT 0 DLC[3:0] RX ERR RX RTR R0 FILHIT[2:0] RX RB1 RX RB0 DLC[3:0] EID<13:6> SID<5:0> SRR EID<0:5> Receive Buffer 1 Byte 1 Receive Buffer 1 Byte 3 Receive Buffer 1 Byte 5 Receive Buffer 1 Byte 7 RX IDE 13 12 11 10 9 8 7 6 5 4 3 2 1 0

Reset
File Name
ADR
15
C2RX1SID
430
xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000
C2RX1EID
432
C2RX1DLC
434
C2RX1B1
436
CAN Module

EID<17:14> RX RTR EID<0:5> Receive Buffer 0 Byte 1 Receive Buffer 0 Byte 3 Receive Buffer 0 Byte 5 Receive Buffer 0 Byte 7 SID<10:6> SRR RX IDE WAK FIL TXBO Transmit Error Counter TXBP RXBP TX WARN RXB1 OVR SEG2PH[2:0] RX WARN C SIDL ABAT CAN CKS REQOP[2:0]
C2RX1B2
438
C2RX1B3
43A
C2RX1B4
43C
C2RX1CON
43E
C2RX1SID
440
xxxx xxxx xxxx xxxx xxxx xxxx xxxx 0000
C2RX1EID
442
C2RX1DLC
444
C2RX0B1
446
C2RX0B2
448
C2RX0B3
44A
C2RX0B4
44C
C2RX0CON
44E
C2CTRL
450
C2CFG1
452
CAN CAP
0480 0000 0000 0000 0000 0000 xxxx
C2CFG2
454
C2INTF
456
RXB0 OVR
C2INTE
458
C2EC
45A
Reserved
45C 4FE
Section 23. CAN
DS70070C-page 23-27
Legend: x = Unknown
23

23.3 CAN Module Features
The CAN module is a communication controller implementing the CAN 2.0A/B protocol as defined in the BOSCH specification. The module will support CAN 1.2, CAN 2.0A, CAN 2.0B Passive and CAN 2.0B Active versions of the protocol. The module implementation is a Full CAN system. The module features are as follows: Implementation of the CAN protocol CAN 1.2, CAN 2.0A and CAN 2.0B Standard and extended data frames Data length from 0-8 bytes Programmable bit rate up to 1 Mbit/sec Support for remote data frames Double buffered receiver with two prioritized received message storage buffers Six full (standard/extended identifier) acceptance filters, 2 associated with the high priority receive buffer, and 4 associated with the low priority receive buffer Two full acceptance filter masks, one each associated with the high and low priority receive buffers Three Transmit Buffers with application specified prioritization and abort capability Programmable wake-up functionality with integrated low-pass filter Programmable Loopback mode supports self-test operation Signaling via interrupt capabilities for all CAN receiver and transmitter error states Programmable clock source Programmable link to input capture module for time-stamping and network synchronization Low power Sleep mode
DS70070C-page 23-28
Section 23. CAN

23.4 CAN Module Implementation
The CAN bus module consists of a Protocol Engine and message buffering and control. The Protocol Engine can best be understood by defining the types of data frames to be transmitted and received by the module. These blocks are shown in Figure 23-2. Figure 23-2: CAN Buffers and Protocol Engine Block Diagram
Acceptance Mask RXM1 Acceptance Filter RXF2 TXB0 MESSAGE MSGREQ TXABT TXLARB TXERR MTXBUFF TXB1 MESSAGE MSGREQ TXABT TXLARB TXERR MTXBUFF TXB2 A c c e p t MESSAGE MSGREQ TXABT TXLARB TXERR MTXBUFF Acceptance Mask RXM0 Acceptance Filter RXF0 Acceptance Filter RXF1 Acceptance Filter RXF3 Acceptance Filter RXF4 Acceptance Filter RXF5 A c c e p t
BUFFERS
Message Queue Control
R X B 0 Transmit Byte Sequencer
Identifier
Message Assembly Buffer
Identifier
R X B 1
Data Field
Data Field
23
CAN Module
RERRCNT TERRCNT ErrPas BusOff
PROTOCOL ENGINE
Receive Error Counter
Transmit Error Counter Transmit Shift Receive Shift
CRC Generator
CRC Check
Protocol Finite State Machine
Transmit Logic
Bit Timing Logic
Bit Timing Generator
CxTX
CxRX
Note: x = 1 or 2
DS70070C-page 23-29

23.4.1 CAN Message Formats
The CAN protocol engine handles all functions for receiving and transmitting messages on the CAN bus. Messages are transmitted by first loading the appropriate data registers. Status and errors can be checked by reading the appropriate registers. Any message detected on the CAN bus is checked for errors and then matched against filters to see if it should be received and stored in one of the two receive registers. The CAN Module supports the following frame types: 23.4.1.1 Standard Data Frame Extended Data Frame Remote Frame Error Frame Interframe Space
Standard Data Frame A standard data frame is generated by a node when the node wishes to transmit data. The standard CAN data frame is shown in Figure 23-3. In common with all other frames, the frame begins with a Start-Of-Frame bit (SOF - dominant state) for hard synchronization of all nodes. The SOF is followed by the Arbitration field consisting of 12 bits, the 11-bit identifier (reflecting the contents and priority of the message) and the RTR bit (Remote Transmission Request bit). The RTR bit is used to distinguish a data frame (RTR - dominant) from a remote frame. The next field is the Control field, consisting of 6 bits. The first bit of this field is called the Identifier Extension (IDE) bit and is at dominant state to specify that the frame is a standard frame. The following bit is reserved by the CAN protocol, RB0, and defined as a dominant bit. The remaining 4 bits of the Control field are the Data Length Code (DLC) and specify the number of bytes of data contained in the message. The data being sent follows in the Data field which is of the length defined by the DLC above (0-8 bytes). The Cyclic Redundancy Check (CRC) field follows and is used to detect possible transmission errors. The CRC field consists of a 15-bit CRC sequence and a delimiter bit. The message is completed by the End-Of-Frame (EOF) field, which consists of seven recessive bits with no bit-stuffing. The final field is the Acknowledge field. During the ACK Slot bit the transmitting node sends out a recessive bit. Any node that has received an error free frame acknowledges the correct reception of the frame by sending back a dominant bit (regardless of whether the node is configured to accept that specific message or not). The recessive Acknowledge Delimiter completes the Acknowledge Slot and may not be overwritten by a dominant bit, except when an error frame occurs.
23.4.1.2
Extended Data Frame In the extended CAN data frame, shown in Figure 23-4, the Start-Of-Frame bit (SOF) is followed by the Arbitration Field consisting of 38 bits. The first 11 bits are the 11 Most Significant bits of the 29-bit identifier (Base-lD). These 11 bits are followed by the Substitute Remote Request bit (SRR), which is transmitted as recessive. The SRR is followed by the lDE bit which is recessive to denote that the frame is an extended CAN frame. It should be noted from this, that if arbitration remains unresolved after transmission of the first 11 bits of the identifier, and one of the nodes involved in arbitration is sending a standard CAN frame (11-bit identifier), then the standard CAN frame will win arbitration due to the assertion of a dominant lDE bit. Also, the SRR bit in an extended CAN frame must be recessive to allow the assertion of a dominant RTR bit by a node that is sending a standard CAN remote frame. The SRR and lDE bits are followed by the remaining 18 bits of the identifier (lD-Extension) and a dominant Remote Transmission Request bit.
DS70070C-page 23-30
Section 23. CAN

To enable standard and extended frames to be sent across a shared network, it is necessary to split the 29-bit extended message identifier into 11-bit (Most Significant) and 18-bit (Least Significant) sections. This split ensures that the Identifier Extension bit (lDE) can remain at the same bit position in both standard and extended frames. The next field is the Control field, consisting of 6 bits. The first 2 bits of this field are reserved and are at dominant state. The remaining 4 bits of the Control field are the Data Length Code (DLC) and specify the number of data bytes. The remaining portion of the frame (Data field, CRC field, Acknowledge field, End-Of-Frame and intermission) is constructed in the same way as for a standard data frame. 23.4.1.3 Remote Frame A data transmission is usually performed on an autonomous basis with the data source node (e.g., a sensor sending out a data frame). It is possible however for a destination node to request the data from the source. For this purpose, the destination node sends a remote frame with an identifier that matches the identifier of the required data frame. The appropriate data source node will then send a data frame as a response to this remote request. There are two differences between a remote frame and a data frame, shown in Figure 23-5. First, the RTR bit is at the recessive state and second there is no Data field. In the very unlikely event of a data frame and a remote frame with the same identifier being transmitted at the same time, the data frame wins arbitration due to the dominant RTR bit following the identifier. In this way, the node that transmitted the remote frame receives the desired data immediately. 23.4.1.4 The Error Frame An error frame is generated by any node that detects a bus error. An error frame, shown in Figure 23-6, consists of 2 fields, an error flag field followed by an Error Delimiter field. The Error Delimiter consists of 8 recessive bits and allows the bus nodes to restart bus communications cleanly after an error. There are two forms of error flag fields. The form of the error flag field depends on the error status of the node that detects the error. If an error-active node detects a bus error then the node interrupts transmission of the current message by generating an active error flag. The active error flag is composed of six consecutive dominant bits. This bit sequence actively violates the bit-stuffing rule. All other stations recognize the resulting bit-stuffing error and in turn generate error frames themselves, called Error Echo Flags. The error flag field therefore consists of between six and twelve consecutive dominant bits (generated by one or more nodes). The Error Delimiter field completes the error frame. After completion of the error frame, bus activity retains to normal and the interrupted node attempts to resend the aborted message. If an error passive node detects a bus error then the node transmits an Error Passive flag followed, again, by the Error Delimiter field. The Error Passive flag consists of six consecutive recessive bits. From this it follows that, unless the bus error is detected by the transmitting node or other error active receiver that is actually transmitting, the transmission of an error frame by an error passive node will not affect any other node on the network. If the bus master node generates an error passive flag then this may cause other nodes to generate error frames due to the resulting bit-stuffing violation. After transmission of an error frame, an error passive node must wait for 6 consecutive recessive bits on the bus before attempting to rejoin bus communications. 23.4.1.5 The Interframe Space Interframe Space separates a proceeding frame (of whatever type) from a following data or remote frame. lnterframe Space is composed of at least 3 recessive bits, called the intermission. This is provided to allow nodes time for internal processing of the message by receiving nodes before the start of the next message frame. After the intermission, the bus line remains in the recessive state (bus idle) until the next transmission starts. If the transmitting node is in the error passive state, an additional 8 recessive bit times will be inserted in the Interframe Space before any other message is transmitted by that node. This time period is called the Suspend Transmit field. The Suspend Transmit field allows additional delay time for other transmitting nodes to take control of the bus.
23
CAN Module
DS70070C-page 23-31
Figure 23-3:
111111111111111111111 Data Frame (number of bits = 44 + 8 N) 12 Arbitration Field 11 ID3 ID0 RTR IDE RB0 DLC3 DLC0 15 CRC 1 6 Control Field 4 16 CRC Field 8 8 8 N ( N 8) Data Field 7 End-OfFrame Start-Of-Frame ID 10 CRC Del Acknowledgment ACK Del
1110
Start-Of-Frame Data Frame or Remote Frame
Reserved Bits
111111111111111111111
1110
Start-Of-Frame
DS70070C-page 23-32
bus Idle 0 Identifier Message Filtering Stored in Buffers Bit-Stuffing Stored in Transmit/Receive Buffers Data Length Code Inter-Frame Space 3 Any Frame INT 8 Suspend Transmit bus Idle 000 00000000 11111111 Data Frame or Remote Frame
Inter-Frame Space
Standard Data Frame
Any Frame INT
Suspend Transmit
Figure 23-4:
Start-Of-Frame
ID3
EID0 RTR RB1 RB0 DLC3
ID0 SRR IDE EID17
DLC0
Start-Of-Frame ID10
0 Extended Identifier Data Length Code Stored in Transmit/Receive Buffers
11
100
000000000000000000000001
Identifier
Stored in Buffers Bit-Stuffing
Reserved bits
Message Filtering
CRC Del Acknowledgment ACK Del
111111111111111111111
1110
Start-Of-Frame
23
CAN Module

Extended Data Frame (number of bits = 64 + 8 N) 32 Arbitration Field 18 8 8 15 CRC 8 N (N 8) Data Field 16 CRC Field 7 End-OfFrame 6 Control Field 4 11111111 Inter-Frame Space 3 Any Frame INT 8 Suspend Transmit bus Idle Data Frame or Remote Frame
Extended Data Format
bus Idle
Data Frame or Remote Frame
11
1110
11
Section 23. CAN
DS70070C-page 23-33
Figure 23-5:
111111111111111111111
1110
Start-Of-Frame
Start-Of-Frame ID 10
0 Identifier Message Filtering Stored in Buffers Bit-Stuffing Data Length Code
ID0 RTR IDE RB0 DLC3
100
DLC0
Reserved Bits
111111111111111111111
1110
Start-Of-Frame
DS70070C-page 23-34
Inter-Frame Space 3 8 Suspend Transmit bus Idle Data Frame or Remote Frame Remote Frame (number of bits = 44) 12 Arbitration Field 11 15 CRC 1 6 Control Field 4 16 CRC Field End-OfFrame 7 11111111 Inter-Frame Space 3 Any Frame INT 8 Suspend Transmit bus Idle Data Frame or Remote Frame
Remote Data Frame
Any Frame INT
Figure 23-6:
Inter-Frame Space 8 Suspend Transmit bus Idle Data Frame or Remote Frame
Error Frame
111111111111111111111 Interrupted Data Frame 12 Arbitration Field 11 ID3 ID0 RTR IDE RB0 DLC3 DLC0 6 Control Field 4 8 8 Start-Of-Frame ID 10 8N ( N 8) Data Field
1110
Start-Of-Frame 0 Identifier Message Filtering Reserved Bits Data Length Code
Section 23. CAN
111111111111111111111
1110
Star-Of-Frame 3 Any Frame INT 8 Suspend Transmit bus Idle
23
CAN Module

000 Error Frame Bit-Stuffing Data Frame or Remote Frame 6 Error Flag 6 Echo Error Flag 0000000 8 Error Delimiter Inter-Frame Space 00111111110 Inter-Frame Space Data Frame or Remote Frame
Any Frame INT
DS70070C-page 23-35

23.5 CAN Module Operation Modes
The CAN Module can operate in one of several Operation modes selected by the user. These modes include: Normal Operation mode Disable mode Loopback mode Listen Only mode Configuration mode Listen to All Messages mode
Modes are requested by setting the REQOP<2:0> bits (CiCTRL<10:8>). Entry into a mode is acknowledged by monitoring the OPMODE<2:0> bits (CiCTRL<7:5>). The module does not change the mode and the OPMODE bits until a change in mode is acceptable, generally during bus idle time which is defined as at least 11 consecutive recessive bits.
23.5.1
Normal Operation Mode

Normal Operating mode is selected when REQOP<2:0> = 000. In this mode, the module is activated, the I/O pins will assume the CAN bus functions. The module will transmit and receive CAN bus messages as described in subsequent sections.
23.5.2
Disable Mode
In Disable mode, the module will not transmit or receive. The module has the ability to set the WAKIF bit due to bus activity, however any pending interrupts will remain and the error counters will retain their value. If the REQOP<2:0> bits (CiCTRL<10:8>) = 001, the module will enter the Module Disable mode. This mode is similar to disabling other peripheral modules by turning off the module enables. This causes the module internal clock to stop unless the module is active (i.e., receiving or transmitting a message). If the module is active, the module will wait for 11 recessive bits on the CAN bus, detect that condition as an idle bus, then accept the module disable command. When the OPMODE<2:0> bits (CiCTRL<7:5>) = 001, this indicates that the module successfully entered Module Disable mode (see Figure 23-7). The WAKIF interrupt is the only module interrupt that is still active in the Module Disable mode. If the WAKIE bit (CiINTE<6>) is set, the processor will receive an interrupt whenever the CAN bus detects a dominant state, as occurs with a Start-Of-Frame (SOF). The I/O pins will revert to normal I/O function when the module is in the Module Disable mode. Note: Typically, if the CAN module is allowed to transmit in a particular mode of operation and a transmission is requested immediately after the CAN module has been placed in that mode of operation, the module waits for 11 consecutive recessive bits on the bus before starting transmission. If the user switches to Disable Mode within this 11-bit period, then this transmission is aborted and the corresponding TXABT bit is set and TXREQ bit is cleared.
DS70070C-page 23-36
Section 23. CAN

Figure 23-7:
OSC1
Entering and Exiting Module Disable Mode
REQOP<2:0>
000
001
000
OPMODE<2:0>
000
001
000
CAN bus
WAKIF
WAKIE
CAN Module Disabled 1 2 3 4 5
1 - Processor writes REQOP<2:0> while module receiving/transmitting message. Module continues with CAN message. 2 - Module detects 11 recessive bits. Module acknowledges Disable mode and sets OPMODE<2:0> bits. Module disables. 3 - CAN bus message will set WAKIF bit. If WAKIE = 1, processor will vector to the interrupt address. CAN message ignored. 4 - Processor writes REQOP<2:0> during CAN bus activity. Module waits for 11 recessive bits before accepting activate. 5 - Module detects 11 recessive bits. Module acknowledges Normal mode and sets OPMODE<2:0> bits. Module activates.
23
CAN Module
23.5.3
Loopback Mode
If the Loopback mode is activated, the module will connect the internal transmit signal to the internal receive signal at the module boundary. The transmit and receive pins revert to their PORT I/O function. The transmitter will receive an acknowledge for its sent messages. Special hardware will generate an acknowledge for the transmitter.
23.5.4
Listen Only Mode

Listen Only mode and Loopback modes are special cases of Normal Operation mode to allow system debug. If the Listen Only mode is activated, the module on the CAN bus is passive. The transmitter buffers revert to the PORT I/O function. The receive pins remain as inputs to the CAN module. For the receiver, no error flags or Acknowledge signals are sent. The error counters are deactivated in this state. The Listen Only mode can be used for detecting the baud rate on the CAN bus. To use this, it is necessary that there are at least two further nodes that communicate with each other. The baud rate can be detected empirically by testing different values. This mode is also useful as a bus monitor without influencing the data traffic.
DS70070C-page 23-37

23.5.5 Configuration Mode
In the Configuration mode, the module will not transmit or receive. The error counters are cleared and the interrupt flags remain unchanged. The programmer will have access to configuration registers that are access restricted in other modes. After a device Reset the CAN module is in the Configuration mode (OPMODE<2:0> = 100). The error counters are cleared and all registers contain the Reset values. It should be ensured that the initialization is performed before REQOP<2> bit is cleared. The CAN module has to be initialized before its activation. This is only possible if the module is in the Configuration mode. The Configuration mode is requested by setting the REQOP<2> bit. Only when the Status bit OPMODE<2> has a high level, the initialization can be performed. Afterwards the configuration registers and the acceptance mask registers and the acceptance filter registers can be written. The module is activated by clearing the control bits REQOP<2:0>. The module will protect the user from accidentally violating the CAN protocol through programming errors. All registers which control the configuration of the module can not be modified while the module is on-line. The CAN module will not be allowed to enter the Configuration mode while a transmission is taking place. The Configuration mode serves as a lock to protect the following registers. All Module Control Registers Baud Rate and Interrupt Configuration Registers Bus Timing Registers Identifier Acceptance Filter Registers Identifier Acceptance Mask Registers
23.5.6
Listen All Messages Mode

Listen All Messages mode is a special case of Normal Operation mode to allow system debug. If the Listen All Messages mode is activated, the module on the CAN bus is passive. The transmitter buffers revert to the PORT I/O function. The receive pins remain inputs. For the receiver, no error flags or Acknowledge signals are sent. The error counters are deactivated in this state. The filters are disabled. Receive Buffer 0 will receive any message transferred on the bus. This mode is useful to record all bus traffic as a bus monitor without influencing the data traffic.
DS70070C-page 23-38
Section 23. CAN

23.6 Message Reception
This subsection describes CAN module message reception.
23.6.1
Receive Buffers
The CAN bus module has 3 receive buffers. However, one of the receive buffers is always committed to monitoring the bus for incoming messages. This buffer is called the message assembly buffer, MAB. So there are 2 receive buffers visible, RXB0 and RXB1, that can essentially instantaneously receive a complete message from the protocol engine. The CPU can be operating on one while the other is available for reception or holding a previously received message. The MAB holds the destuffed bit stream from the bus line to allow parallel access to the whole data or remote frame for the acceptance match test and the parallel transfer of the frame to the receive buffers. The MAB will assemble all messages received. These messages will be transferred to the RXBn buffers only if the acceptance filter criterion are met. When a message is received, the RXnIF flag (CiINTF<0> or CiINRF<1>) will be set. This bit can only be set by the module when a message is received. The bit is cleared by the CPU when it has completed processing the message in the buffer. This bit provides a positive lockout to ensure that the CPU has finished with the message buffer. If the RXnIE bit (CiINTE<0> or CiINTE<1>) is set , an interrupt will be generated when a message is received. There are 2 programmable acceptance filter masks associated with the receive buffers, one for each buffer. When the message is received, the FILHIT bits (CiRX0CON<0> for Receive Buffer 0 and CiRX1CON<2:0> for Receive Buffer 1) indicate the acceptance criterion for the message. The number of the acceptance filter that enabled the reception will be indicated as well as a Status bit that indicates that the received message is a remote transfer request. Note: In the case of Receive Buffer 0, a limited number of Acceptance Filters can be used to enable a reception. A single bit, FILHIT0 (CiRX0CON<0>) determines which of the 2 filters, RXF0 or RXF1, enabled the message reception.
23
CAN Module
DS70070C-page 23-39

23.6.1.1 Receive Buffer Priority To provide flexibility, there are several acceptance filters corresponding to each receive buffer. There is also an implied priority to the receive buffers. RXB0 is the higher priority buffer and has 2 message acceptance filters associated with it. RXB1 is the lower priority buffer and has 4 acceptance filters associated with it. The lower number of possible acceptance filters makes the match on RXB0 more restrictive and implies the higher priority associated with that buffer. Additionally, if the RXB0 contains a valid message, and another valid message is received, the RXB0 can be setup such that it will not overrun and the new message for RXB0 will be placed into RXB1. Figure 23-8 shows a block diagram of the receive buffer, while Figure 23-9 shows a flow chart for a receive operation. Figure 23-8: The Receive Buffers
Acceptance Mask RXM1
Acceptance Filter RXF2
Acceptance Mask RXM0
Acceptance Filter RXF0 A c c e p t
A c c e p t
R X B 0
Identifier
Message Assembly Buffer
Identifier
R X B 1
Data Field
Data Field
DS70070C-page 23-40
Section 23. CAN

Figure 23-9: Receive Flowchart
START
No
Detect Start of Message ? Yes
Begin Loading Message into Message Assembly Buffer (MAB)
Generate Error Frame
No
Valid Message Received ? Yes
Yes, meets criteria Yes, meets criteria Message for RXB1 for RXB0 Identifier meets a filter criteria ? No Go to Start The RXFUL bit determines if the receive register is empty and able to accept a new message. The DBEN bit determines if RXB0 can roll over into RXB1 if it is full.
23
CAN Module
Is RXFUL = 0 ? Yes
No
Is DBEN = 1 ? No
Yes
Move message into RXB0
Generate Overrun Error: Set RX0OVR
Generate Overrun Error: Set RX1OVR
No
Is RXFUL = 0 ? Yes
Set RXFUL = 1 No
Move message into RXB1 Set FILHIT<0> according to which filter criteria was met Does ERRIE=1 ? Yes Go to Start Set FILHIT<2:0> according to which filter criteria was met
Set RXFUL = 1
Is RXnIE = 1 ? No
Yes
Generate Interrupt Set ICODE<3:0> according to which receive buffer the message was loaded into
Yes
Does RXnIE = 1 ? No
DS70070C-page 23-41

23.6.2 Message Acceptance Filters
The message acceptance filters and masks are used to determine if a message in the message assembly buffer should be loaded into either of the receive buffers. Once a valid message has been received into the Message Assembly Buffer (MAB), the identifier fields of the message are compared to the filter values. If there is a match, that message will be loaded into the appropriate receive buffer. The filter masks are used to determine which bits in the identifiers are examined with the filters. A truthtable is shown in Table 23-3 that indicates how each bit in the identifier is compared to the masks and filters to determine if the message should be loaded into a receive buffer. The mask bit essentially determines which bits to apply the filter to. If any mask bit is set to a zero, then that bit will automatically be accepted regardless of the filter bit. Table 23-3: 0 1 1 1 1 Legend: x = dont care 23.6.2.1 Identifier Mode Selection The EXIDE control bits (CiRXFnSID<0>) and the MIDE control bits (CiRXMnSID<0>) enable an acceptance filter for standard or extended identifiers. The acceptance filters look at incoming messages for the RXIDE bit to determine how to compare the identifiers. If the RXIDE bit is clear, the message is a standard frame. If the RXIDE bit is set, the message is an extended frame. If the MIDE control bit for the filter is set, then the identifier type for the filter is determined by the EXIDE control bit for the filter. If the EXIDE control bit is cleared, then the filter will accept standard identifiers. If the EXIDE bit is set, then the filter will accept extended identifiers. Most CAN systems will use only standard identifiers or only extended identifiers. If the MIDE control bit for the filter is cleared, the filter will accept both standard and extended identifiers if a match occurs with the filter bits. This mode can be used in CAN systems that support both standard and extended identifiers on the same bus. 23.6.2.2 FILHIT Status Bits As shown in the Receive Buffers Block Diagram, Figure 23-8, RXF0 and RXF1 filters with the RXM0 mask are associated with RXB0. The filters RXF2, RXF3, RXF4 and RXF5 and the mask RXM1 are associated with RXB1. When a filter matches and a message is loaded into the receive buffer, the number of the filter that enabled the message reception is indicated in the CiRXnCON register via the FILHIT bits. The CiRX0CON register contains one FILHIT Status bit to indicate whether the RXF0 or the RXF1 filter enabled the message reception. The CiRX1CON register contains the FILHIT<2:0> bits. They are coded as shown in Table 23-4. Filter/Mask Truth Table Filter Bit n x 0 0 1 1 Message Identifier bit x 0 1 0 1 Accept or Reject bit n Accept Accept Reject Reject Accept
Mask Bit n
DS70070C-page 23-42
Section 23. CAN

Table 23-4: Acceptance Filter Acceptance Filter RXF0 RXF1 RXF2 RXF3 RXF4 RXF5 Comment Only if DBEN = 1 Only if DBEN = 1
(1)
FILHIT<2:0> 000 001(1) 010 011 100 101
Note 1: Is only valid if the DBEN bit is set. The DBEN bit (CiRX0CON<2>) allows the FILHIT bits to distinguish a hit on filter RXF0 and RXF1 in either RXB0 or overrun into RXB1. 111 = Acceptance Filter 1 (RXF1) 110 = Acceptance Filter 0 (RXF0) 001 = Acceptance Filter 1 (RXF1) 000 = Acceptance Filter 0 (RXF0) If the DBEN bit is clear, there are 6 codes corresponding to the 6 filters. If the DBEN bit is set, there are 6 codes corresponding to the 6 filters plus 2 additional codes corresponding to RXF0 and RXF1 filters overrun to RXB1. If more than 1 acceptance filter matches, the FILHIT bits will encode the lowest binary value of the filters that matched. In other words, if filter 2 and filter 4 match, FILHIT will code the value for 2. This essentially prioritizes the acceptance filters with lower numbers having priority. Figure 23-10 shows a block diagram of the message acceptance filters. Figure 23-10: Message Acceptance Filter
Acceptance Filter Register RXMn0 Acceptance Mask Register
23
CAN Module
RXFn0
RXFn1
RXMn1
RxRqst
RXFnn
RXMnn
Message Assembly Buffer Identifier
DS70070C-page 23-43

23.6.3 Receiver Overrun
An overrun condition occurs when the Message Assembly Buffer (MAB) has assembled a valid received message, the message is accepted through the acceptance filters, and when the receive buffer associated with the filter has not been designated as clear of the previous message. The overrun error flag, RXnOVR (CiINTF<15> or CiINTF<14>) and the ERRIF bit (CiINTF<5>) will be set and the message in the MAB will be discarded. While in the overrun situation, the module will stay synchronized with the CAN bus and is able to transmit messages, but it will discard all incoming messages destined for the overflowed buffer. If the DBEN bit is clear, RXB1 and RXB0 operate independently. When this is the case, a message intended for RXB0 will not be diverted into RXB1 if RXB0 contains an unread message and the RX0OVR bit will be set. If the DBEN bit is set, the overrun for RXB0 is handled differently. If a valid message is received for RXB0 and RXFUL = 1 (CiRX0CON<7>) indicating that RXB0 is full, and RXFUL = 0 (CiRX1CON<7>) indicating that RXB1 is empty, the message for RXB0 will be loaded into RXB1. An overrun error will not be generated for RXB0. If a valid message is received for RXB0 and RXFUL = 1, and RXFUL = 1 indicating that both RXB0 and RXB1 are full, the message will be lost and an overrun will be indicated for RXB1. If the DBEN bit is clear, there are six codes corresponding to the six filters. If the DBEN bit is set, there are six codes corresponding to the six filters plus two additional codes corresponding to RXF0 and RXF1 filters overrun to RXB1. These codes are given in Table 23-5. Table 23-5: Message Matches Filter 0 or 1 0 0 0 1 1 1 1 1 1 0 0 Buffer Reception and Overflow Truth Table Message Matches Filter 2,3,4,5 0 1 1 0 0 0 0 1 1 0 1 RXFUL0 Bit X X X 0 1 1 1 0 1 X X RXFUL1 Bit X 0 1 X X 0 1 X X X 0 DBEN Bit X X X X 0 1 1 X 0 X X None MAB RXB1 MAB discarded RX1OVR = 1 MAB RXB0 MAB discarded RX0OVR = 1 MAB RXB1 MAB discarded RX1OVR = 1 MAB RXB0 MAB discarded RX0OVR = 1 None MAB RXB1 Action Results
No message received Message for RXB1, RXB1 available Message for RXB1, RXB1 full Message for RXB0, RXB0 available Message for RXB0, RXB0 full, DBEN not enabled Message for RXB0, RXB0 full, DBEN enabled, RXB1 available Message for RXB0, RXB0 full, DBEN enabled, RXB1 full Message for RXB0 and RXB1, RXB0 available Message for RXB0 and RXB1, RXB0 full, DBEN not enabled No message received Message for RXB1, RXB1 available
Legend: X = Dont care
DS70070C-page 23-44
Section 23. CAN

23.6.4 Effects of a Reset
Upon any Reset the CAN module has to be initialized. All registers are set according to the Reset values. The content of a received message is lost. The initialization is discussed in Section 23.5.5 Configuration Mode.
23.6.5
Receive Errors
The CAN module will detect the following receive errors: Cyclic Redundancy Check (CRC) Error Bit Stuffing Error Invalid message receive error These receive errors do not generate an interrupt. However, the receive error counter is incremented by one in case one of these errors occur. The RXWAR bit (CiINTF<9>) indicates that the Receive Error Counter has reached the CPU warning limit of 96 and an interrupt is generated.
23.6.5.1
Cyclic Redundancy Check (CRC) Error With the Cyclic Redundancy Check, the transmitter calculates special check bits for the bit sequence from the start of a frame until the end of the data field. This CRC sequence is transmitted in the CRC Field. The receiving node also calculates the CRC sequence using the same formula and performs a comparison to the received sequence. If a mismatch is detected, a CRC error has occurred and an Error Frame is generated. The message is repeated. The receive error interrupt counter is incremented by one. An Interrupt will only be generated if the error counter passes a threshold value.
23
CAN Module
23.6.5.2
Bit Stuffing Error If, between the Start -Of-Frame and the CRC Delimiter, 6 consecutive bits with the same polarity are detected, the bit-stuffing rule has been violated. A bit-stuffing error occurs and an error frame is generated. The message is repeated. No interrupt will be generated upon this event.
23.6.5.3
Invalid Message Received Error If any type of error occurs during reception of a message, an error will be indicated by the IVRIF bit (CiINTF<7>). This bit can be used (optionally with an interrupt) for autobaud detection with the device in Listen Only mode. This error is not an indicator that any action needs to be taken, but it does indicate that an error has occurred on the CAN bus.
23.6.5.4
Rules for Modifying the Receive Error Counter The Receive Error Counter is modified according to the following rules: When the receiver detects an error, the Receive Error Counter is incremented by 1, except when the detected error was a bit error during the transmission of an active error flag. When the receiver detects a dominant bit as the first bit after sending an error flag, the Receive Error Counter will be incremented by 8. If a receiver detects a bit error while sending an active error flag, the Receive Error Counter is incremented by 8. Any node tolerates up to 7 consecutive dominant bits after sending an active error flag or passive error flag. After detecting the 14th consecutive dominant bit (in case of an Active error flag) or after detecting the 8th consecutive dominant bit following a passive error flag, and after each sequence of eight additional consecutive dominant bits, every transmitter increases its Transmission Error Counter and every receiver increases its Receive Error Counter by 8. After a successful reception of a message (reception without error up to the ACK slot and the successful sending of the ACK bit), the Receive Error Counter is decreased by one, if the Receive Error Counter was between 1 and 127. If the Receive Error Counter was 0, it will stay 0. If the Receive Error Counter was greater than 127, it will change to a value between 119 and 127.
DS70070C-page 23-45

23.6.6 Receive Interrupts
Several Interrupts are linked to the message reception. The receive interrupts can be broken up into two separate groups: Receive Error Interrupts Receive interrupts 23.6.6.1 Receive Interrupt A message has been successfully received and loaded into one of the receive buffers. This interrupt is activated immediately after receiving the End-Of-Frame (EOF) field. Reading the RXnIF flag will indicate which receive buffer caused the interrupt. Figure 23-11 depicts when the receive buffer interrupt flag RXnIF will be set. 23.6.6.2 Wake-up Interrupt The Wake-up interrupt sequences are described in Section 23.13.1 Operation in Sleep Mode.
DS70070C-page 23-46
Section 23. CAN

Figure 23-11: Receive Buffer Interrupt Flag
EOF EOF EOF EOF EOF EOF EOF ACK DELIMITER ACK SIST BIT CRCDEL CRC0 CRC1 CRC2 CRC3 CRC4 CRC5 CRC6 CRC7 CRC8 CRC9 CRC10 CRC11 CRC12 CRC13 CRC14 DLC0 DLC1 STUFF DLC2 DLC3 RB0 IDE RTR ID0 ID1 ID2 ID3 ID4 ID5 ID6 ID7 ID8 ID9 ID10 SOF Receive Buffer Interrupt Flag
23
CAN Module
CAN bit Timing
CAN bit Names
Data
DS70070C-page 23-47

23.6.6.3 Receive Error Interrupts A receive error interrupt will be indicated by the ERRIF bit (CiINTF<5>). This bit shows that an error condition occurred. The source of the error can be determined by checking the bits in the CAN Interrupt Status Register CiINTF. The bits in this register are related to receive and transmit errors. The following subsequences will show which flags are linked to the receive errors. 23.6.6.3.1 Invalid Message Received Interrupt If any type of error occurred during reception of the last message, an error will be indicated by the IVRIF bit (CiINTF<7>). The specific error that occurred is unknown. This bit can be used (optionally with an interrupt) for autobaud detection with the device in Listen Only mode. This error is not an indicator that any action needs to be taken, but an indicator that an error has occurred on the CAN bus. 23.6.6.3.2 Receiver Overrun Interrupt The RXnOVR bit (CiINTF<15>, CiINTF<14>) indicates that an overrun condition occurred for the receive buffer. An overrun condition occurs when the Message Assembly Buffer (MAB) has assembled a valid received message, the message is accepted through the acceptance filters, however, the receive buffer associated with the filter is not clear of the previous message. The overflow error interrupt will be set and the message is discarded. While in the overrun situation, the module will stay synchronized with the CAN bus and is able to transmit and receive messages. 23.6.6.4 Receiver Warning Interrupt The RXWAR bit (CiINTF<8>) indicates that the Receive Error Counter has reached the CPU warning limit of 96. When RXWAR transitions from a 0 to a 1, it will cause the Error Interrupt Flag ERRIF to become set. This bit cannot be manually cleared, as it should remain an indicator that the Receive Error Counter has reached the CPU warning limit of 96. The RXWAR bit will become clear automatically if the Receive Error Counter becomes less than or equal to 95. The ERRIF bit can be manually cleared allowing the interrupt service routine to be exited without affecting the RXWAR bit. 23.6.6.5 Receiver Error Passive The RXEP bit (CiINTF<11>) indicates that the Receive Error Counter has exceeded the Error Passive limit of 127 and the module has gone to Error Passive state. When the RXEP bit transitions from a 0 to a 1, it will cause the error interrupt flag to become set. The RXEP bit cannot be manually cleared, as it should remain an indicator that the bus is in Error State Passive. The RXEP bit will become clear automatically if the Receive Error Counter becomes less than or equal to 127. The ERRIF bit can be manually cleared allowing the interrupt service routine to be exited without affecting the RXEP bit.
DS70070C-page 23-48
Section 23. CAN

23.7 Transmission
This subsection describes how the CAN module is used to transmit CAN messages.
23.7.1
Real Time Communication and Transmit Message Buffering

For an application to effectively transmit messages in real-time, the CAN nodes must be able to dominate and hold the bus, assuming that nodes messages are of a high enough priority to win arbitration on the bus. If a node only has 1 transmission buffer, it must transmit a message, then release the bus while the CPU reloads the buffer. If a node has two transmission buffers, one buffer could be transmitting while the second buffer is being reloaded. However, the CPU would need to maintain tight tracking of the bus activity to ensure that the second buffer is reloaded before the first message completes. Typical applications require three transmit message buffers. With three buffers, one buffer can be transmitting, the second buffer can be ready to transmit as soon as the first is complete, and the third can be reloaded by the CPU. This eases the burden of the software to maintain synchronization with the bus (see Figure 23-12). Additionally, the three buffers allow some degree of prioritizing of the outgoing messages. For example, the application software may have a message enqueued in the second buffer while it is working on the third buffer. The application may require that the message going into the third buffer is of higher importance than the one already enqueued. If only 2 buffers are available, the enqueued message would have to be deleted and replaced with the third. The process of deleting the message may mean losing control of the bus. With 3 buffers, both the second and the third message can be enqueued, and the module can be instructed that the third message is higher priority than the second. The third message will be the next one sent followed by the second.
23
CAN Module
23.7.2
Transmit Message Buffers

The CAN module has three Transmit Buffers. Each of the three buffers occupies 14 bytes of data. Eight of the bytes are the maximum 8 bytes of the transmitted message. Five bytes hold the standard and extended identifiers and other message arbitration information. The last byte is a control byte associated with each message. The information in this byte determines the conditions under which the message will be transmitted and indicates status of the transmission of the message. The TXnIF bit (CiINTF<2>, CiINTF<3> or CiINTF<4>) will be set and the TXREQ bit (CiTXnCON<3>) will be clear, indicating that the message buffer has completed a transmission. The CPU will then load the message buffer with the contents of the message to be sent. At a minimum, the standard identifier register CiTXnSID must be loaded. If data bytes are present in the message, the TXBnDm registers are loaded. If the message is to use extended identifiers, the CiTXnEID register and the EID<5:0> bits (CiTXnDLC<15:10>) are loaded and the TXIDE bit is set (CiTXnSID<0>). Prior to sending the message, the user must initialize the TXnIE bit (CiINTE<2>, CiINTE<3> or CiINTE<4>) to enable or disable an interrupt when the message is sent. The user must also initialize the transmit priority. Figure 23-12 shows a block diagram of the Transmit Buffers.
DS70070C-page 23-49

Figure 23-12: Transmit Buffers
MESSAGE MESSAGE MESSAGE TXB0 TXLARB TXREQ TXERR TXABT TXPRI TXB1 TXLARB TXREQ TXERR TXABT TXPRI TXB2 TXLARB TXREQ TXERR TXABT TXPRI
Message Queue Control
Transmit Byte Sequencer
23.7.3
Transmit Message Priority

Transmit priority is a prioritization within each node of the pending transmittable messages. Prior to sending the SOF (Start-Of-Frame), the priorities of all buffers ready for transmission are compared. The Transmit Buffer with the highest priority will be sent first. For example, if Transmit Buffer 0 has a higher priority setting than Transmit Buffer 1, Buffer 0 will be sent first. If two buffers have the same priority setting, the buffer with the highest address will be sent. For example, if Transmit Buffer 1 has the same priority setting as Transmit Buffer 0, Buffer 1 will be sent first. There are 4 levels of transmit priority. If TXPRI<1:0> (CiTXnCON<1:0>) for a particular message buffer is set to 11, that buffer has the highest priority. If TXPRI<1:0> for a particular message buffer is set to 10 or 01, that buffer has an intermediate priority. If TXPRI<1:0> for a particular message buffer is 00, that buffer has the lowest priority.
23.7.4
Message Transmission
To initiate transmitting the message, the TXREQ bit (CiTXnCON<3>) must be set. The CAN bus module resolves any timing conflicts between setting of the TXREQ bit and the SOF time, ensuring that if the priority was changed, it is resolved correctly before SOF. When TXREQ is set the TXABT (CiTXnCON<6>), TXLARB (CiTXnCON<5>) and TXERR (CiTXnCON<4>) flag bits will be cleared by the module. Setting TXREQ bit does not actually start a message transmission, it flags a message buffer as enqueued for transmission. Transmission will start when the module detects an available bus for SOF. The module will then begin transmission on the message which has been determined to have the highest priority. If the transmission completes successfully on the first try, the TXREQ bit will clear and an interrupt will be generated if the TXnIE bit (CiINTE<2>, CiINTE<3>, CiINTE<4>) is set. If the message fails to transmit, other condition flags will be set and the TXREQ bit will remain set indicating that the message is still pending for transmission. If the message tried to transmit but encountered an error condition, the TXERR bit (CiTXnCON<4>) will be set. In this case, the error condition can also cause an interrupt. If the message tried to transmit but lost arbitration, the TXLARB bit (CiTXnCON<5>) will be set. In this case, no interrupt is available to signal the loss of arbitration.
DS70070C-page 23-50
Section 23. CAN

23.7.5 Transmit Message Aborting
The system can abort a message by clearing the TXREQ bit associated with each message buffer. Setting the ABAT bit (CiCTRL<12>) will request an abort of all pending messages (see Figure 23-14). A queued message is aborted by clearing the TXREQ bit. Aborting a queued message is illustrated in Figure 23-13. If the message has not yet started transmission, or if the message started but is interrupted by loss of arbitration or an error; the abort will be processed. The abort is indicated when the module sets the TXABT bit (CiTXnCON<6>), and the TXnIF flag is not set. If the message has started to transmit, it will attempt to transmit the current message fully (see Figure 23-15). If the current message is transmitted fully, and is not lost to arbitration or an error, the TXABT bit will not be set, because the message was transmitted successfully. Likewise, if a message is being transmitted during an abort request, and the message is lost to arbitration (see Figure 23-16) or an error, the message will not be re-transmitted, and the TXABT bit will be set, indicating that the message was successfully aborted. Figure 23-13: Abort Queued Message
CAN bus
CiTX
TXREQ TXnIF
23
CAN Module
TXABT
1 - Processor sets TXREQ while module receiving/transmitting message. Module continues with CAN message. 2 - Processor clears TXREQ while module looking for 11 recessive bits. Module aborts pending transmission, sets TXABT bit in 2 clocks. 3 - Another module takes the available transmit slot.
DS70070C-page 23-51

Figure 23-14: Abort All Messages
CAN bus
CiTX
ABAT
TXREQ TXnIF
TXABT
1 - Processor sets TXREQ while module receiving/transmitting message. Module continues with CAN message. 2 - Processor sets ABAT while module looking for 11 recessive bits. Module clears TXREQ bits. Module aborts pending transmission, sets TXABT bit. 3 - Another module takes the available transmit slot.
Figure 23-15: Failed Abort During Transmission

CAN bus
CiTX
TXREQ TXnIF
TXABT
1 - Processor sets TXREQ while module receiving/transmitting message. Module continues with CAN message. 2 - Module detects 11 recessive bits. Module begins transmission of queued message. 3 - Processor clears TXREQ requesting message abort. Abort cannot be acknowledged. 4 - At successful completion of transmission, TXREQ bit remains clear and TXnIF bit set. TXABT remains clear.
DS70070C-page 23-52
Section 23. CAN

Figure 23-16: Loss of Arbitration During Transmission
CAN bus
CiTX TXREQ TXnIF TXLARB 1 2 3 4 5
1 - Processor sets TXREQ while module inactive. TXLARB bit cleared. 2 - Module in inactive state. Module begins transmission of queued message. 3 - Message loses arbitration. Module releases bus and sets TXLARB bit. 4 - Module waits for 11 recessive bits before re-trying transmission of queued message. 5 - At successful completion of transmission, TXREQ bit cleared and TXnIF bit set.
23
CAN Module
DS70070C-page 23-53

Figure 23-17: Transmit Flowchart
START The message transmission sequence begins when the device determines that the TXREQ for any of the Transmit registers has been set. No Are any TXREQ bits = 1 ? Yes Clear: TXABT, TXLARB and TXERR Clearing the TXREQ bit while it is set, or setting the ABAT bit before the message has started transmission will abort the message.
Is CAN bus available to start transmission ? Yes
No
Does TXREQ = 0 ABAT = 1 ? Yes
No
Examine TXPRI<1:0> to Determine Highest Priority Message
Begin transmission (SOF)
Was message transmitted successfully?
No
Set TXERR = 1
Yes Set TXREQ = 0 Does TXLARB = 1? Yes Generate Interrupt Is TXnIE = 1? Yes Arbitration lost during transmission
No A message can also be aborted if a message error or lost arbitration condition occurred during transmission.
Set TXBUFE = 1 The TXnIE bit determines if an interrupt should be generated when a message is successfully transmitted.
Does TXREQ = 0 or TXABT =1 ? No
Yes
Abort Transmission: Set TXABT = 1
END
DS70070C-page 23-54
Section 23. CAN

23.7.6 Transmit Boundary Conditions
The module handles transmit commands which are not necessarily synchronized to the CAN bus message framing time. 23.7.6.1 Clearing TXREQ bit as a Message Starts The TXREQ bit can be cleared just when a message is starting transmission, with the intent of aborting the message. If the message is not being transmitted, the TXABT bit will be set, indicating that the Abort was successfully processed. When the user clears the TXREQ bit and the TXABT bit is not set two cycles later, the message has already begun transmission. If the message is being transmitted, the abort is not immediately processed, at some point later, the TXnIF interrupt flag or the TXABT bit is set. If transmission has begun the message will only be aborted if either an error or a loss of arbitration occurs. 23.7.6.2 Setting TXABT bit as a Message Starts Setting the ABAT bit will abort all pending Transmit Buffers and has the function of clearing all of the TXREQ bits for all buffers. The boundary conditions are the same as clearing the TXREQ bit. 23.7.6.3 Clearing TXREQ bit as a Message Completes The TXREQ bit can be cleared when a message is just about to successfully complete transmission. Even if the TXREQ bit is cleared by the Data bus a short time before it will be cleared by the successful transmission of the message, the TXnIF flag will still be set due to the successful transmission. 23.7.6.4 Setting TXABT bit as a Message Completes The boundary conditions are the same as clearing the TXREQ bit. 23.7.6.5 Clearing TXREQ bit as a Message Loses Transmission The TXREQ bit can be cleared when a message is just about to be lost to arbitration or an error. If the TXREQ signal falls before the loss of arbitration signal or error signal, the result will be like clearing TXREQ during transmission. When the arbitration is lost or the error is set, the TXABT bit will be set, as it will see that an error has occurred while transmitting, and that the TXREQ bit was not set. If the TXREQ bit falls after the arbitration signal has entered the block, the result will be like clearing TXREQ during an inactive transmit time. The TXABT bit will be set. 23.7.6.6 Setting TXABT bit as a Message Loses Transmission The boundary conditions are the same as clearing the TXREQ bit.
23
CAN Module
23.7.7
Effects of a Reset
Upon any Reset the CAN module has to be initialized. All registers are set according to the reset values. The content of a transmitted message is lost. The initialization is discussed in Section 23.5.5 Configuration Mode.
DS70070C-page 23-55

23.7.8 Transmission Errors
The CAN module will detect the following transmission errors: Acknowledge Error Form Error Bit Error These transmission errors will not necessarily generate an interrupt but are indicated by the transmission error counter. However, each of these errors will cause the transmission error counter to be incremented by one. Once the value of the error counter exceeds the value of 96, the ERRIF (CiINTF<5>) and the TXWAR bit (CiINTF<10>) are set. Once the value of the error counter exceeds the value of 96 an interrupt is generated and the TXWAR bit in the error flag register is set. A transmission error example is illustrated in Figure 23-18. Figure 23-18: Error During Transmission
CAN bus
CiTX TXREQ TXnIF TXERR 1 2 3 4 5
1 - Processor sets TXREQ while module inactive. TXERR bit is cleared. 2 - Module in inactive state. Module begins transmission of queued message. 3 - Module detects error during transmission, releases bus and sets TXERR bit. 4 - Module waits for 11 recessive bits before re-trying transmission of queued message. 5 - At successful completion of transmission, TXREQ bit cleared and TXnIF bit set.
23.7.8.1
Acknowledge Error In the Acknowledge field of a message, the transmitter checks if the Acknowledge Slot (which it has sent out as a recessive bit) contains a dominant bit. If not, no other node has received the frame correctly. An acknowledge error has occurred and the message has to be repeated. No error frame is generated.
23.7.8.2
Form Error lf a transmitter detects a dominant bit in one of the four segments including End-Of-Frame, lnterframe Space, Acknowledge Delimiter or CRC Delimiter; then a form error has occurred and an error frame is generated. The message is repeated.
23.7.8.3
Bit Error A bit error occurs if a transmitter sends a dominant bit and detects a recessive bit. In the case where the transmitter sends a recessive bit and a dominant bit is detected during the Arbitration field and the Acknowledge Slot, no bit error is generated because normal arbitration is occurring.
DS70070C-page 23-56
Section 23. CAN

23.7.8.4 Rules for Modifying the Transmit Error Counter The Transmit Error Counter is modified according to the following rules: When the transmitter sends an error flag the Transmit Error Counter is increased by 8 with the following exceptions. In these two exceptions, the Transmit Error Counter is not changed. - If the transmitter is error passive and detects an acknowledgment error because of not detecting a dominant ACK, and does not detect a dominant bit while sending a passive error flag. - If the transmitter sends an error flag because of a bit-stuffing error occurred during arbitration whereby the Stuffbit is located before the RTR bit, and should have been recessive, and has been sent as recessive but monitored as dominant. If a transmitter detects a bit error while sending an active error flag the Transmit Error Counter is increased by 8. Any Node tolerates up to 7 consecutive dominant bits after sending an active error flag or passive error flag. After detecting the 14th consecutive dominant bit (in case of an active error flag) or after detecting the 8th consecutive dominant following a passive error flag, and after each sequence of eight additional consecutive dominant bits, every transmitter increases its Transmission Error Counter and every receiver increases its Receive Error Counter by 8. After the successful transmission of a message (getting an acknowledge and no error until End-Of-Frame is finished) the Transmit Error Counter is decreased by one unless it was already 0.
23.7.9
Transmission Interrupts
There are several interrupts linked to the message transmission. The transmission interrupts can be broken up into two groups: Transmission interrupts Transmission error interrupts
23
CAN Module
23.7.9.1
Transmit Interrupt At least one of the three transmit buffers is empty (not scheduled) and can be loaded to schedule a message for transmission. Reading the TXnIF flags in the CiINTF register will indicate which transmit buffer is available and caused the interrupt.
DS70070C-page 23-57

23.7.9.2 Transmission Error Interrupts A transmission error interrupt will be indicated by the ERRIF flag. This flag shows that an error condition occurred. The source of the error can be determined by checking the error flags in the CAN Interrupt Status register CiINTF. The flags in this register are related to receive and transmit errors. The TXWAR bit (CiINTF<10>) indicates that the Transmit Error Counter has reached the CPU warning limit of 96. When this bit transitions from a 0 to a 1, it will cause the error interrupt flag to become set. The TXWAR bit cannot be manually cleared, as it should remain as an indicator that the Transmit Error Counter has reached the CPU warning limit of 96. The TXWAR bit will become clear automatically if the Transmit Error Counter becomes less than or equal to 95. The ERRIF flag can be manually cleared allowing the interrupt service routine to be exited without affecting the TXWAR bit. The TXEP bit (CiINTF<12>) indicates that the Transmit Error Counter has exceeded the Error Passive limit of 127 and the module has gone to Error Passive state. When this bit transitions froma 0 to a 1, it will cause the error interrupt flag to become set. The TXEP bit cannot be manually cleared, as it should remain as an indicator that the bus is in Error Passive state. The TXEP bit will become clear automatically if the Transmit Error Counter becomes less than or equal to 127. The ERRIF flag can be manually cleared allowing the interrupt service routine to be exited without affecting the TXEP bit. The TXBO bit (CiINTF<13>) indicates that the Transmit Error Counter has exceeded 255 and the module has gone to bus off state. When this bit transitions from a 0 to a 1, it will cause the error interrupt flag to become set. The TXBO bit cannot be manually cleared, as it should remain as an indicator that the bus is off. The ERRIF flag can be manually cleared allowing the interrupt service routine to be exited without affecting the TXBO bit.
23.8
Error Detection
The CAN protocol provides sophisticated error detection mechanisms. The following errors can be detected. These errors are either receive or transmit errors. Receive errors are: Cyclic Redundancy Check (CRC) Error (see Section 23.6.5.1 Cyclic Redundancy Check (CRC) Error) Bit Stuffing Bit Error (see Section 23.6.5.2 Bit Stuffing Error) lnvalid Message Received Error (see Section 23.6.5.3 Invalid Message Received Error) The transmit errors are: Acknowledge Error (see Section 23.7.8.1 Acknowledge Error) Form Error (see Section 23.7.8.2 Form Error) Bit Error (see Section 23.7.8.3 Bit Error)
23.8.1
Error States
Detected errors are made public to all other nodes via error frames. The transmission of the erroneous message is aborted and the frame is repeated as soon as possible. Furthermore, each CAN node is in one of the three error states error active, error passive or bus off according to the value of the internal error counters. The error active state is the usual state where the bus node can transmit messages and active error frames (made of dominant bits) without any restrictions. In the error passive state, messages and passive error frames (made of recessive bits) may be transmitted. The bus off state makes it temporarily impossible for the station to participate in the bus communication. During this state, messages can neither be received nor transmitted.
DS70070C-page 23-58
Section 23. CAN

23.8.2 Error Modes and Error Counters
The CAN controller contains the two error counters Receive Error Counter (RERRCNT) and Transmit Error Counter (TERRCNT). The values of both counters can be read by the CPU from the Error Count Register CiEC. These counters are incremented or decremented according to the CAN bus specification. The CAN controller is error active if both error counters are below the error passive limit of 128. It is error passive if at least one of the error counters equals or exceeds 128. It goes bus off if the Transmit Error Counter equals or exceeds the bus off limit of 256. The device remains in this state, until the bus off recovery sequence is finished, which is 128 consecutive 11 recessive bit times. Additionally, there is a error state warning flag bit, EWARN (CiINTF<8>), which is set if at least one of the error counters equals or exceeds the error warning limit of 96. EWARN is reset if both error counters are less than the error warning limit. Figure 23-19: Error Modes
Reset
RERRCNT > 127 or TERRCNT > 127
Error Active 128 occurrences of 11 consecutive recessive bits
23
CAN Module
RERRCNT < 127 or TERRCNT < 127 Error Passive TERRCNT > 255 Bus Off
23.8.3
Error Flag Register

The values in the error flag register indicate which error(s) caused the error interrupt flag. The RXnOVR error flags (CiINTF<15> and CiINTF<14>) have a different function than the other error flag bits in this register. The RXnOVR bits must be cleared in order to clear the ERRIF interrupt flag. The other error flag bits in this register will cause the ERRIF interrupt flag to become set as the value of the Transmit and Receive Error Counters crosses a specific threshold. Clearing the ERRIF interrupt flag in these cases will allow the interrupt service routine to be exited without recursive interrupt occurring. It may be desirable to disable specific interrupts after they have occurred once to stop the device from interrupting repeatedly as the Error Counter moves up and down in the vicinity of a threshold value.
DS70070C-page 23-59

23.9 CAN Baud Rate
All nodes on any particular CAN bus must have the same nominal bit rate. The CAN bus uses NRZ coding which does not encode a clock. Therefore the receivers independent clock must be recovered by the receiving nodes and synchronized to the transmitters clock. In order to set the baud rate the following bits have to be initialized: Synchronization Jump Width (see Section Section 23.9.6.2 Re-synchronization) Baud Rate Prescaler (see Section Section 23.9.2 Prescaler Setting) Phase Segments (see Section Section 23.9.4 Phase Segments) Length Determination of Phase Segment 2 (see Section Section 23.9.4 Phase Segments) Sample Point (see Section Section 23.9.5 Sample Point) Propagation Segment Bits (see Section Section 23.9.3 Propagation Segment)
23.9.1
Bit Timing
As oscillators and transmission time may vary from node to node, the receiver must have some type of PLL synchronized to data transmission edges to synchronize and maintain the receiver clock. Since the data is NRZ coded, it is necessary to include bit-stuffing to ensure that an edge occurs at least every 6 bit times, to maintain the Digital Phase Lock Loop (DPLL) synchronization. Bus timing functions executed within the bit time frame, such as synchronization to the local oscillator, network transmission delay compensation, and sample point positioning, are defined by the programmable bit timing logic of the DPLL. All controllers on the CAN bus must have the same baud rate and bit length. However, different controllers are not required to have the same master oscillator clock. At different clock frequencies of the individual controllers, the baud rate has to be adjusted by adjusting the number of time quanta in each segment. The nominal bit time can be thought of as being divided into separate non-overlapping time segments. These segments are shown in Figure 23-20. Synchronization segment (Sync Seg) Propagation time segment (Prop Seg) Phase buffer segment 1 (Phase1 Seg) Phase buffer segment 2 (Phase2 Seg)
The time segments and also the nominal bit time are made up of integer units of time called time quanta or TQ. By definition, the nominal bit time has a minimum of 8 TQ and a maximum of 25 TQ. Also, by definition the minimum nominal bit time is 1 sec, corresponding to a maximum 1 MHz bit rate. Figure 23-20: CAN Bit Timing
Input Signal
Sync
Prop Segment
Phase Segment 1 Sample Point
Phase Segment 2
Sync
TQ
DS70070C-page 23-60
Section 23. CAN

23.9.2 Prescaler Setting
There is a programmable prescaler, with integral values ranging from 1 to 64, in addition to a fixed divide-by-2 for clock generation. The Time Quanta (TQ) is a fixed unit of time derived from the input clock frequency, FCAN. Time quanta is defined as shown in Equation 23-1. Note: FCAN must not exceed 30 MHz. If CANCKS = 0, then FCY must not exceed 7.5 MHz. Time Quanta for Clock Generation TQ = 2 (BRP<5:0> + 1) FCAN
Equation 23-1:
Where BRP is the binary value of BRP <5:0> FCAN is FCY or 4 FCY depending on CANCKS bit Example 23-1: Bit Rate Calculation Example
If 4FCY = 32 MHz, BRP<5:0> = 0x01 and CANCKS = 0, then: TQ = 2 (BRP + 1) TCY = 2 X 2 X (1/32X106) = 125ns 4
If Nominal Bit Time = 8 TQ then: 9 Nominal Bit Rate = 1 ( 8 125 10 ) Mbps Example 23-2: Baud Rate Prescaler Calculation Example
23
CAN Module
CAN Baud Rate = 125 kHz FCY = 5 MHz, CANCKS = 1 1. Select number of TQ clocks per bit time (e.g., K=16). 2. Calculate TQ from baud rate: 1 ( BaudRate ) 1 125 10 TQ = ------------------------------------- = --------------------------- = 500ns K 16 3. Calculate BRP<5:0>: TQ = 2 (BRP + 1) TCAN BRP = ( 2T Q T CY ) 1 2 ( 500 10 ) = ------------------------------ 1 6 1 ( 5 10 ) = 4 The frequencies of the oscillators in the different nodes must be coordinated in order to provide a system-wide specified time quantum. This means that all oscillators must have a TOSC that is a integral divisor of TQ.
9 3
DS70070C-page 23-61

23.9.3 Propagation Segment
This part of the bit time is used to compensate physical delay times within the network. These delay times consist of the signal propagation time on the bus line and the internal delay time of the nodes. The delay is calculated as the round trip from transmitter to receiver as twice the signal's propagation time on the bus line, the input comparator delay, and the output driver delay. The Propagation Segment can be programmed from 1 TQ to 8 TQ by setting the PRSEG<2:0> bits (CiCFG2<2:0>).
23.9.4
Phase Segments
The phase segments are used to optimally locate the sampling of the received bit within the transmitted bit time. The sampling point is between Phase1 Segment and Phase2 Segment. These segments are lengthened or shortened by re-synchronization. The end of the Phase1 Segment determines the sampling point within a bit period. The segment is programmable from 1 TQ to 8 TQ. Phase2 Segment provides delay to the next transmitted data transition. The segment is programmable from 1 TQ to 8 TQ or it may be defined to be equal to the greater of Phase1 Segment or the Information Processing Time (3 TQs). The phase segment 1 is initialized by setting bits SEG1PH<2:0> (CiCFG2<5:3>), and phase segment 2 is initialized by setting SEG2PH<2:0> (CiCFG2<10:8>).
23.9.5
Sample Point
The sample point is the point of time at which the bus level is read and interpreted as the value of that respective bit. The location is at the end of phase segment 1. If the bit timing is slow and contains many TQ, it is possible to specify multiple sampling of the bus line at the sample point. The level determined by the CAN bus then corresponds to the result from the majority decision of three values. The majority samples are taken at the sample point and twice before with a distance of TQ/2. The CAN module allows to chose between sampling three times at the same point or once at the same point. This is done by setting or clearing the SAM bit (CiCFG2<6>).
23.9.6
Synchronization
To compensate for phase shifts between the oscillator frequencies of the different bus stations, each CAN controller must be able to synchronize to the relevant signal edge of the incoming signal. When an edge in the transmitted data is detected, the logic will compare the location of the edge to the expected time (Synchronous Segment). The circuit will then adjust the values of Phase1 Segment and Phase2 Segment. There are 2 mechanisms used to synchronize.
23.9.6.1
Hard Synchronization Hard Synchronization is only done whenever there is a recessive to dominant edge during bus Idle, indicating the start of a message. After hard synchronization, the bit time counters are restarted with Synchronous Segment. Hard synchronization forces the edge which has caused the hard synchronization to lie within the synchronization segment of the restarted bit time. Due to the rules of synchronization, if a hard synchronization is done, there will not be a re-synchronization within that bit time.
DS70070C-page 23-62
Section 23. CAN

23.9.6.2 Re-synchronization As a result of re-synchronization phase segment 1 may be lengthened or phase segment 2 may be shortened. The amount of lengthening or shortening of the phase buffer segment, specified by the SJW<1:0> bits (CiCFG1<7:6>), has an upper bound given by the re-synchronization jump width bits. The value of the synchronization jump width will be added to phase segment 1 or subtracted from phase segment 2. The re-synchronization jump width is programmable between 1 TQ and 4 TQ. Clocking information will only be derived from transitions of recessive to dominant bus states. The property that only a fixed maximum number of successive bits have the same value ensures resynchronizing a bus unit to the bit stream during a frame (e.g., bit-stuffing). The phase error of an edge is given by the position of the edge relative to Synchronous Segment, measured in Time Quanta. The phase error is defined in magnitude of TQ as follows: e = 0 if the edge lies within the Synchronous Segment. e > 0 if the edge lies before the Sample Point. e < 0 if the edge lies after the Sample Point of the previous bit. If the magnitude of the phase error is less than or equal to the programmed value of the re-synchronization jump width, the effect of a re-synchronization is the same as that of a hard synchronization. If the magnitude of the phase error is larger than the re-synchronization jump width, and if the phase error is positive, then phase segment 1 is lengthened by an amount equal to the re-synchronization jump width. If the magnitude of the phase error is larger than the re-synchronization jump width, and if the phase error is negative, then phase segment 2 is shortened by an amount equal to the re-synchronization jump width. Figure 23-21: Lengthening a Bit Period
Input Signal
23
CAN Module
Sync
Propagation Segment
Phase Segment 1
sjw
Phase Segment 2
Sample Point TQ
Nominal Actual Bit Length Bit Length
Figure 23-22: Shortening a Bit Period

Input Signal
Sync
Propagation Segment
Phase Segment 1
Phase Segment 2
sjw
Sample Point TQ
Actual Bit Length
Nominal Bit Length
DS70070C-page 23-63

23.9.7 Programming Time Segments
Some requirements for programming of the time segments are as follows: Propagation Segment + Phase1 Segment > = Phase2 Segment Phase2 Segment > Synchronous Jump Width Typically, the sampling of the bit should take place at about 60-70% through the bit time, depending on the system parameters. Example 23-2 has a 16 TQ bit time. If we choose Synchronous Segment = 1 TQ and Propagation Segment = 2 TQ, setting Phase Segment 1 = 7 TQ would place the sample point at 10 TQ (62% of the bit time) after the initial transition. This would leave 6 TQ for phase segment 2. Since phase segment 2 is 6, according to the rules, the SJWS<1:0> bits could be set to the maximum of 4 TQ. However, normally a large synchronization jump width is only necessary when the clock generation of the different nodes is inaccurate or unstable, such as using ceramic resonators. So a synchronization jump width of 1 is typically enough.
23.10
Interrupts
The module has several sources of interrupts. Each of these interrupts can be individually enabled or disabled. A CiINTF register contains interrupt flags. A CiINTE register controls the enabling of the 8 main interrupts. A special set of read only bits in the CiCTRL register (ICODE<2:0>) can be used in combination with a jump table for efficient handling of interrupts. All interrupts have one source, with the exception of the error interrupt. Any of the error interrupt sources can set the error interrupt flag. The source of the error interrupt can be determined by reading the CiINTF register. The interrupts can be broken up into two categories: receive and transmit interrupts. The receive related interrupts are: Receive interrupt Wake-up interrupt Receiver Overrun interrupt Receiver Warning interrupt Receiver Error Passive interrupt Transmit interrupt Transmitter Warning interrupt Transmitter Error Passive interrupt Bus Off interrupt
The Transmit related interrupts are:
23.10.1
Interrupt Acknowledge
Interrupts are directly associated with one or more status flags in CiINTF register. Interrupts are pending as long as one of the corresponding flags is set. The flags in the registers must be reset within the interrupt handler in order to handshake the interrupt. A flag can not be cleared if the respective condition still prevails, with the exception being interrupts that are caused by a certain value being reached in one of the error counter registers.
DS70070C-page 23-64
Section 23. CAN

23.10.2 The ICODE Bits
The ICODE<2:0> bits (CiCTRL<3:1>) are a set of read only bits designed for efficient handling of interrupts via a jump table. The ICODE<2:0> bits can only display one interrupt at a time because the interrupt bits are multiplexed into this register. Therefore, the pending interrupt with the highest priority and enabled interrupt is reflected in the ICODE<2:0> bits. Once the highest priority interrupt flag has been cleared, the next highest priority interrupt code is reflected in the ICODE<2:0> bits. An interrupt code for a corresponding interrupt can only be displayed if both its interrupt flag and interrupt enable are set. Table 23-6 describes the operation of the ICODE<2:0> bits. Table 23-6: ICODE Bits Decode Table Boolean Expression ERRWAKTX0TX1TX2RX0RX1 ERR ERRTX0 ERRTX0TX1 ERRTX0TX1TX2 ERRTX0TX1TX2RX0 ERRTX0TX1TX2RX0RX1 ERRTX0TX1TX2RX0RX1WAK
ICODE<2:0> 000 001 100 011 010 110 101 111
23
CAN Module
Legend: ERR = ERRIF ERRIE TX0 = TX0IF TX0IE TX1 = TX1IF TX1IE TX2 = TX2IF TX2IE RX0 = RX0IF RX0IE RX1 = RX1IF RX1IE WAK = WAKIF WAKIE
23.11
CAN Capture
The CAN module will generate a signal that can be sent to a timer capture input whenever a valid frame has been accepted. This is useful for time-stamping and network synchronization. Because the CAN specification defines a frame to be valid if no errors occurred before the EOF field has been transmitted successfully, the timer signal will be generated right after the EOF. A pulse of one bit time is generated. Time-stamping is enabled by the TSTAMP control bit (CiCTRL<15>). The IC2 capture input is used for time-stamping. Note: If the CAN capture is enabled, the IC2 pin becomes unusable as a general input capture pin. In this mode, the IC2 channel derives its input signal from the C1RX or C2RX pin instead of the IC2 pin.
23.12
CAN Module I/O

The CAN bus module communicates on up to 2 I/O pins. There is 1 transmit pin and 1 receive pin. These pins are multiplexed with normal digital I/O functions of the device. When the module is in the Configuration mode, Module Disable mode or Loopback mode, the I/O pins revert to the PORT I/O function. When the module is active, the CiTX pin (i = 1 or 2) is always dedicated to the CAN output function. The TRIS bits associated with the transmit pins are overridden by the CAN bus modes. The module receives the CAN input on the CiRX input pin.
DS70070C-page 23-65

23.13 23.13.1 Operation in CPU Power Saving Modes Operation in Sleep Mode
Sleep mode is entered by executing a PWRSAV #0 instruction. This will stop the crystal oscillator and shut down all system clocks. The user should ensure that the module is not active when the CPU goes into Sleep mode. The pins will revert into normal I/O function, dependent on the value in the TRIS register. Because the CAN bus is not disruptable, the user must never execute a PWRSAV #0 instruction while the module is in an Operating mode. The module must first be switched to Disable mode by setting REQOP<2:0> =001 (CiCTRL<10:8>). When OPMODE<2:0> = 001 (CiCTRL<7:5>), indicating that Disable mode is achieved, then the Sleep instruction may be used. Figure 23-23 depicts how the CAN module will behave when the CPU enters Sleep mode and how the module wakes up on bus activity. When the CPU exits Sleep mode due to activity on the CAN bus, the WAKIF flag (CiINTF<6>) is set. The module will monitor the CiRX line for activity while the device is in Sleep mode. If the device is in Sleep mode and the WAKIE wake-up interrupt enable is set, the module will generate an interrupt, waking the CPU. Due to the delays in starting up the oscillator and CPU, the message activity that caused the wake-up will be lost. If the module is in CPU Sleep mode and the WAKIE is not set, no interrupt will be generated and the CPU and the CAN module will continue to sleep. If the CAN module is in Disable mode, the module will wake-up and, depending on the condition of the WAKIE bit, may generate an interrupt. It is expected that the module will correctly receive the message that caused the wake-up from Sleep mode. The module can be programmed to apply a low-pass filter function to the CiRX input line while the module or the CPU is in Sleep mode. This feature can be used to protect the module from wake-up due to short glitches on the CAN bus lines. Such glitches can result from electromagnetic inference within noisy environments. The WAKFIL bit (CiCFG2<14>) enables or disables the filter.
DS70070C-page 23-66
Section 23. CAN

Figure 23-23: Processor Sleep and CAN bus Wake-up Interrupt
OSC1 TOST REQOP<2:0> 000 001 000
OPMODE<2:0>
000
001
000
CAN bus
Sleep WAKIF
WAKIE Processor in SLEEP CAN Module Disabled 3 4 5
23
CAN Module
1 - Processor requests and receives Module Disable mode. Wake-up interrupt enabled. 2 - Processor executes SLEEP (PWRSAV #0) instruction. 3 - SOF of message wakes up processor. Oscillator start time begins. CAN message lost. WAKIF bit set. 4 - Processor completes oscillator start time. Processor resumes program or interrupt, based on GIE bits. Processor requests Normal Operating mode. Module waits for 11 recessive bits before accepting CAN bus activity. CAN message lost. 5 - Module detects 11 recessive bits. Module will begin to receive messages and transmit any pending messages.
23.13.2
CAN Module Operation during CPU Idle Mode

On executing a CPU Idle (PWRSAV #1) instruction, the operation of the CAN module is determined by the state of the CSIDL bit (CiCTRL<13>). If CSIDL = 0, the module will continue operation on assertion of Idle mode. The CAN module can wake the device from Idle mode if the CAN module interrupt is enabled. If CSIDL = 1, the module will discontinue operation in Idle mode. The same rules and conditions for entry to and wake from Sleep mode apply. Refer to Section 23.13.1 Operation in Sleep Mode for further details.
DS70070C-page 23-67

23.14 CAN Protocol Overview
The Controller Area Network (CAN) is a serial communications protocol which efficiently supports distributed real-time control with a very high level of robustness. The CAN Protocol is fully defined by Robert Bosch GmbH, in the CAN Specification V2.0B from 1991. Its domain of application ranges from high speed networks to low cost multiplex wiring. Automotive electronics (i.e., engine control units, sensors, anti-skid-systems, etc.) are connected using CAN with bit rates up to 1 Mbit/sec. The CAN Network allows a cost effective replacement of wiring harnesses in the automobile. The robustness of the bus in noisy environments and the ability to detect and recover from fault conditions makes the bus suitable for industrial control applications such as DeviceNet, SDS and other fieldbus protocols. CAN is an asynchronous serial bus system with one logical bus line. It has an open, linear bus structure with equal bus nodes. A CAN bus consists of two or more nodes. The number of nodes on the bus may be changed dynamically without disturbing the communication of other nodes. This allows easy connection and disconnection of bus nodes (e.g., for addition of system function, error recovery or bus monitoring). The bus logic corresponds to a wired-AND mechanism, recessive bits (mostly, but not necessarily equivalent to the logic level 1) are overwritten by dominant bits (mostly logic level 0). As long as no bus node is sending a dominant bit, the bus line is in the recessive state, but a dominant bit from any bus node generates the dominant bus state. Therefore, for the CAN bus line, a medium must be chosen that is able to transmit the two possible bit states (dominant and recessive). One of the most common and cheapest ways is to use a twisted wire pair. The bus lines are then called CANH and CANL, and may be connected directly to the nodes, or via a connector. There's no standard defined by CAN regarding connector requirements. The twisted wire pair is terminated by terminating resistors at each end of the bus line. The maximum bus speed is 1 Mbit, which can be achieved with a bus length of up to 40 meters. For bus lengths longer than 40 meters, the bus speed must be reduced (a 1000 m bus can be realized with a 40 Kbit bus speed). For bus lengths above 1000 meters, special drivers should be used. At least 20 nodes may be connected without additional equipment. Due to the differential nature of transmission, CAN is not inherently susceptible to radiated electromagnetic energy because both bus lines are affected in the same way, which leaves the differential signal unaffected. The bus lines may also be shielded to reduce the radiated electromagnetic emission from the bus itself, especially at high baud rates. The binary data is coded corresponding to the NRZ code (Non-Return-to-Zero; low level = dominant state; high level = recessive state). To ensure clock synchronization of all bus nodes, bit-stuffing is used. This means that during the transmission of a message, a maximum of five consecutive bits may have the same polarity. Whenever five consecutive bits of the same polarity have been transmitted, the transmitter will insert one additional bit of the opposite polarity into the bit stream before transmitting further bits. The receiver also checks the number of bits with the same polarity and removes the stuff bits from the bit stream (destuffing). In the CAN protocol, it is not bus nodes that are addressed. The address information is contained in the messages that are transmitted. This is done via an identifier (part of each message) which identifies the message content (e.g., engine speed, oil temperature etc.). This identifier also indicates the priority of the message. The lower the binary value of the identifier, the higher the priority of the message. For bus arbitration, Carrier Sense Multiple Access/Collision Detection (CSMA/CD) with Non-Destructive Arbitration (NDA) is used. If bus node A wants to transmit a message across the network, it first checks that the bus is in the Idle state (Carrier Sense), (i.e., no node is currently transmitting). If this is the case (and no other node wishes to start a transmission at the same moment), node A becomes the bus master and sends its message. All other nodes switch to Receive mode during the first transmitted bit (Start-Of-Frame bit). After correct reception of the message (which is Acknowledged by each node), each bus node checks the message identifier and stores the message, if required. Otherwise, the message is discarded.
DS70070C-page 23-68
Section 23. CAN

If two or more bus nodes start their transmission at the same time (Multiple Access), collision of the messages is avoided by bitwise arbitration (Collision Detection/Non-Destructive Arbitration together with the Wired-AND mechanism, dominant bits override recessive bits). Each node sends the bits of its message identifier (MSb first) and monitors the bus level. A node that sends a recessive identifier bit but reads back a dominant one loses bus arbitration and switches to Receive mode. This condition occurs when the message identifier of a competing node has a lower binary value (dominant state = logic 0) and therefore, the competing node is sending a message with a higher priority. In this way, the bus node with the highest priority message wins arbitration without losing time by having to repeat the message. All other nodes automatically try to repeat their transmission once the bus returns to the Idle state. It is not permitted for different nodes to send messages with the same identifier, as arbitration could fail, leading to collisions and errors later in the message. The original CAN specifications (Versions 1.0, 1.2 and 2.0A) defined the message identifier as having a length of 11 bits giving a possible 2048 message identifiers. The specification has since been updated (to version 2.0B) to remove this limitation. CAN specification Version 2.0B allows message identifier lengths of 11 and/or 29 bits to be used (an identifier length of 29 bits allows over 536 million message identifiers). Version 2.0B CAN is also referred to as Extended CAN; and Versions 1.0, 1.2 and 2.0A) are referred to as Standard CAN.
23.14.1
Standard CAN vs. Extended CAN

Those data frames and remote frames, which only contain the 11-bit identifier, are called standard frames according to CAN specification V2.0A. With these frames, 2048 different messages can be identified (identifiers 0-2047). However, the 16 messages with the lowest priority (2032-2047) are reserved. Extended frames according to CAN specification V2.0B have a 29-bit identifier. As already mentioned, this 29-bit identifier is made up of the 11-bit identifier (Standard lD) and the 18-bit Extended identifier (Extended ID). CAN modules specified by CAN V2.0A are only able to transmit and receive standard frames according to the Standard CAN protocol. Messages using the 29-bit identifier cause errors. If a device is specified by CAN V2.0B, there is one more distinction. Modules named Part B Passive can only transmit and receive standard frames but tolerate extended frames without generating error frames. Part B Active devices are able to transmit and receive both standard and extended frames.
23
CAN Module
23.14.2
ISO Model
The lSO/OSl Reference Model is used to define the layers of protocol of a communication system as shown in Figure 23-24. At the highest end, the applications need to communicate between each other. At the lowest end, some physical medium is used to provide electrical signaling. The higher levels of the protocol are run by software. Within the CAN bus specification, there is no definition of the type of message, or the contents, or meaning of the messages transferred. These definitions are made in systems such as Volcano, the Volvo automotive CAN specification J1939, the U.S. heavy truck multiplex wiring specification; and Allen-Bradley DeviceNet and Honeywell SDS, industrial protocols. The CAN bus module definition encompasses two levels of the overall protocol: The Data Link Layer - The Logical Link Control (LLC) sub layer - The Medium Access Control (MAC) sub layer The Physical Layer - The Physical Signaling (PLS) sub layer
DS70070C-page 23-69

The LLC sub layer is concerned with Message Filtering, Overload Notification and Error Recovery Management. The scope of the LLC sub layer is: To provide services for data transfer and for remote data request. To decide which messages received by the LLC sub layer are actually to be accepted. To provide means for error recovery management and overload notifications. The MAC sub layer represents the kernel of the CAN protocol. The MAC sub layer defines the transfer protocol, (i.e., controlling the Framing, Performing Arbitration, Error Checking, Error Signalling and Fault Confinement). It presents messages received from the LLC sub layer and accepts messages to be transmitted to the LLC sub layer. Within the MAC sub layer is where its decided whether the bus is free for starting a new transmission, or whether a reception is just starting. The MAC sub layer is supervised by a management entity called Fault Confinement which is a self-checking mechanism for distinguishing short disturbances from permanent failures. Also, some general features of the bit timing are regarded as part of the MAC sub layer. The physical layer defines the actual transfer of the bits between the different nodes with respect to all electrical properties. The PLS sub layer defines how signals are actually transmitted and therefore deals with the description of Bit Timing, Bit Encoding and Synchronization. The lower levels of the protocol are implemented in driver/receiver chips and the actual interface such as twisted pair wiring or optical fiber, etc. Within one network, the physical layer has to be the same for all nodes. The driver/receiver characteristics of the physical layer are not defined in the CAN specification so as to allow transmission medium and signal level implementations to be optimized for their application. The most common example of the physical transmission medium is defined in Road Vehicles ISO11898, a multiplex wiring specification.
DS70070C-page 23-70
Section 23. CAN

Figure 23-24: CAN Bus in ISO/OSI Reference Model
OSI REFERENCE LAYERS Application Presentation Session Transport Network Data Link Layer Supervisor
LLC (Logical Link Control) Acceptance Filtering Overload Notification Recovery Management MAC (Medium Access Control) Data Encapsulation/Decapsulation Frame Coding (stuffing, destuffing) Medium Access Management Error Detection Error Signalling Acknowledgment Serialization/Deserialization Physical Layer PLS (Physical Signalling) Bit Encoding/Decoding Bit Timing Synchronization
Fault Confinement
23
CAN Module
Bus Failure Management
PMA (Physical Medium Attachment) Driver/Receiver Characteristics
MDI (Medium Dependent Interface) Connectors
Shaded Regions Implemented by the CAN Module
Has to be Implemented in dsPIC30F Firmware
CAN Transceiver Chip
Connector
DS70070C-page 23-71

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the CAN module are: Title An Introduction to the CAN Protocol Application Note # AN713
Note:
DS70070C-page 23-72
Section 23. CAN

Revision B
This revision incorporates additional technical content for the dsPIC30F CAN module.
Revision C
23
CAN Module
DS70070C-page 23-73

NOTES:
DS70070C-page 23-74
Section 24. Device Configuration

HIGHLIGHTS
This section of the manual contains the following topics: 24.1 24.2 24.3 24.4 24.5 24.6 Introduction .................................................................................................................. 24-2 Device Configuration Registers ................................................................................... 24-2 Configuration Bit Descriptions...................................................................................... 24-6 Device Identification Registers..................................................................................... 24-7 Related Application Notes............................................................................................ 24-8 Revision History ........................................................................................................... 24-9
24
Device Configuration
DS70071D-page 24-1

24.1 Introduction
The device Configuration registers allow each user to customize certain aspects of the device to fit the needs of the application. Device Configuration registers are nonvolatile memory locations in the program memory map that hold settings for the dsPIC device during power-down. The Configuration registers hold global setup information for the device, such as the oscillator source, Watchdog Timer mode and code protection settings. The device Configuration registers are mapped in program memory locations, starting at address 0xF80000 and are accessible during normal device operation. This region is also referred to as configuration space. The Configuration bits can be programmed (read as 0), or left unprogrammed (read as 1) to select various device configurations.
24.2
Device Configuration Registers

Each device Configuration register is a 24-bit register, but only the lower 16 bits of each register are used to hold configuration data. There are four device Configuration registers available to the user: FOSC (0xF80000): Oscillator Configuration Register (Note 2) FWDT (0xF80002): Watchdog Timer Configuration Register FBORPOR (0xF80004): BOR and POR Configuration Register FGS (0xF8000A): General Code Segment Configuration Register
The device Configuration registers can be programmed using Run-Time Self-Programming (RTSP), In-Circuit Serial Programming (ICSP), or by a device programmer. Note 1: Not all device Configuration bits shown in the subsequent Configuration register descriptions may be available on a specific device. Refer to the device data sheet for more information. 2: dsPIC30F devices in the General Purpose, Sensor and Motor Control families feature one of three versions of the Oscillator system Version 1, Version 2 and Version 3. For information on the Configuration bits of the FOSC device Configuration register available in each of these versions, please refer to Section 7. "Oscillator.
DS70071D-page 24-2

Register 24-1: Upper Byte: U bit 23 FWDT: Watchdog Timer Configuration Register U U U U U U U bit 16
Middle Byte: R/P

FWDTEN
U bit 8
bit 15 Lower Byte: U bit 7 bit 23-16 Unimplemented: Read as 0 bit 15
R/P R/P FWPSA<1:0>
R/P
R/P R/P FWPSB<3:0>
R/P bit 0
FWDTEN: Watchdog Enable Configuration bit 1 = Watchdog Enabled (LPRC oscillator cannot be disabled. Clearing the SWDTEN bit in the RCON register. Will have no effect.) 0 = Watchdog Disabled (LPRC oscillator can be disabled by clearing the SWDTEN bit in the RCON register.) Unimplemented: Read as 0 FWPSA<1:0>: Prescale Value Selection for Watchdog Timer Prescaler A bits 11 = 1:512 10 = 1:64 01 = 1:8 00 = 1:1 FWPSB<3:0>: Prescale Value Selection for Watchdog Timer Prescaler B bits 1111 = 1:16 1110 = 1:15 0001 = 1:2 0000 = 1:1 Legend: R = Readable bit P = Programmable bit U = Unimplemented bit
bit 14-6 bit 5-4:
bit 3-0
24
2005 Microchip Technology Inc..
DS70071D-page 24-3

Register 24-2: Upper Byte: U bit 23 FBORPOR: BOR and POR Configuration Register U U U U U U U bit 16
Middle Byte: R/P MCLREN bit 15
R/P PWMPIN
R/P HPOL
R/P LPOL bit 8
Lower Byte: R/P BOREN bit 7
R/P R/P BORV<1:0>
R/P R/P FPWRT<1:0> bit 0
bit 23-16 Unimplemented: Read as 0 bit 15 MCLREN: MCLR Pin Function Enable bit 1 = Pin function is MCLR (default case) 0 = Pin is disabled PWMPIN: Motor Control PWM Module Pin Mode bit 1 = PWM module pins controlled by PORT register at device Reset (tri-stated) 0 = PWM module pins controlled by PWM module at device Reset (configured as output pins) HPOL: Motor Control PWM Module High Side Polarity bit 1 = PWM module high-side output pins have active-high output polarity 0 = PWM module high-side output pins have active-low output polarity LPOL: Motor Control PWM Module Low Side Polarity bit 1 = PWM module low-side output pins have active-high output polarity 0 = PWM module low-side output pins have active-low output polarity BOREN: PBOR Enable bit 1 = PBOR Enabled 0 = PBOR Disabled Unimplemented: Read as 0 BORV<1:0>: Brown-out Voltage Select bits 11 = 2.0V 10 = 2.7V 01 = 4.2V 00 = 4.5V Unimplemented: Read as 0 FPWRT<1:0>: Power-on Reset Timer Value Selection bits 11 = PWRT = 64 ms 10 = PWRT = 16 ms 01 = PWRT = 4 ms 00 = Power-up timer disabled Note: Legend: R = Readable bit P = Programmable bit U = Unimplemented bit PWMPIN, HPOL, and LPOL Configuration bits are only available on devices that feature a Motor Control PWM module.
bit 9
bit 8
bit 7
bit 6 bit 5-4
bit 3-2 bit 1-0
DS70071D-page 24-4

Register 24-3: Upper Byte: U bit 23 FGS: General Code Segment Configuration Register U U U U U U U bit 16
Middle Byte: U bit 15
U bit 8
Lower Byte: U bit 7 bit 23-2 bit 1
P GCP
P GWRP bit 0
Unimplemented: Read as 0 GCP: General Code Segment Code-Protect bit 1 = User program memory is not code-protected 0 = User program memory is code-protected GWRP: General Code Segment Write-Protect bit 1 = User program memory is not write-protected 0 = User program memory is write-protected Note: Legend: R = Readable bit P = Programmable bit U = Unimplemented bit The BCP and GWRP Configuration bits can only be programmed to a 0.
bit 0
24
DS70071D-page 24-5

24.3 Configuration Bit Descriptions
This section provides specific functional information on each of the device Configuration bits.
24.3.1
Oscillator Configuration Bits

dsPIC30F devices in the General Purpose, Sensor and Motor Control families feature one of three versions of the Oscillator system Version 1, Version 2 and Version 3. For information on the Configuration bits of the FOSC device Configuration register available in each of these versions, please refer to Section 7. "Oscillator.
24.3.2
BOR and POR Configuration Bits

The BOR and POR Configuration bits found in the FBORPOR Configuration register are used to set the Brown-out Reset voltage for the device, enable the Brown-out Reset circuit, and set the Power-up Timer delay time. For more information on these Configuration bits, please refer to Section 8. "Reset.
24.3.3
Motor Control PWM Module Configuration Bits

The motor control PWM module Configuration bits are located in the FBORPOR Configuration register and are present only on devices that have the PWM module. The Configuration bits associated with the PWM module have two functions: 1. 2. Select the state of the PWM pins at a device Reset (high-Z or output). Select the active signal polarity for the PWM pins. The polarity for the high side and low side PWM pins may be selected independently.
For more information on these Configuration bits, please refer to Section 15. "Motor Control PWM.
24.3.4
General Code Segment Configuration Bits

The general code segment Configuration bits in the FGS Configuration register are used to code-protect or write-protect the user program memory space. The general code segment includes all user program memory with the exception of the interrupt vector table space (0x000000-0x0000FE). If the general code segment is code-protected by programming the GCP Configuration bit (FGS<1>) to a 0, the device program memory cannot be read from the device using In-Circuit Serial Programming (ICSP), or the device programmer. Additionally, further code cannot be programmed into the device without first erasing the entire general code segment. When the general segment is code-protected, user code can still access the program memory data via table read instructions, or Program Space Visibility (PSV) accesses from data space. If the GWRP (FGS<0>) Configuration bit is programmed, all writes to the user program memory space are disabled.
24.3.4.1
General Code Segment Configuration Bit Group The GCP and GWRP Configuration bits in the FGS Configuration register must be programmed/erased as a group. If one or both of the Configuration bits is programmed to a 0, a full chip erase must be performed to change the state of either bit. Note: If the code protection Configuration fuse group (FGS<GCP:GWRP>) bits have been programmed, an erase of the entire code-protected device is only possible at voltages, VDD >= 4.5 volts.
DS70071D-page 24-6

24.4 Device Identification Registers
The dsPIC30F devices have two sets of registers located in configuration space that provide identification information.
24.4.1
Device ID (DEVID) Registers

The configuration memory space locations 0xFF0000 and 0xFF0002 are used to store a read-only Device ID number that is set when the device is manufactured. This number identifies the dsPIC30F device type and the silicon revision. The Device ID registers can be read by the user using table read instructions.
24.4.2
Unit ID Field
The Unit ID field is located at configuration memory space locations 0x800600 through 0x80063E. This field consists of 32 program memory locations and can be programmed at the Microchip factory with unique device information. This field cannot be written or erased by the user, but can be read using table read instructions. Please contact Microchip technical support or your local Microchip representative for further details.
24
DS70071D-page 24-7

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Device Configuration module are: Title Using the dsPIC30F for Sensorless BLDC Control Note: Application Note # AN901
DS70071D-page 24-8

Revision B
This revision incorporates technical content changes for the dsPIC30F Device Configuration module.
Revision C
Revision D
Descriptions of three versions of the Oscillator Control module have been added. The definition of the FOSC Configuration register was moved to DS70054.
24
DS70071D-page 24-9
DS70071D-page 24-10
Section 25. Development Tool Support

HIGHLIGHTS
This section of the manual contains the following topics: 25.1 25.2 25.3 25.4 25.5 25.6 Introduction .................................................................................................................. 25-2 Microchip Hardware and Language Tools.................................................................... 25-2 Third Party Hardware/Software Tools and Application Libraries.................................. 25-6 dsPIC30F Hardware Development Boards................................................................ 25-11 Related Application Notes.......................................................................................... 25-15 Revision History ......................................................................................................... 25-16
25
Development Tool Support
DS70072C-page 25-1

Note: Some development tools described in this section are not available at the time of this writing, however they are currently under development. Some of the product details may change. Please check the Microchip web site or your local Microchip sales office for the most current information and the availability of each product. Introduction
Microchip will offer a comprehensive package of development tools and libraries to support the dsPIC architecture. In addition, the company is partnering with many third party tool manufacturers for additional dsPIC device support.
25.1
25.2
Microchip Hardware and Language Tools

The Microchip tools proposed include: MPLAB Integrated Development Environment (IDE) dsPIC Language Suite, including MPLAB C30 C Compiler, Assembler, Linker and Librarian MPLAB SIM Software Simulator MPLAB ICE 4000 In-Circuit Emulator MPLAB ICD 2 In-Circuit Debugger PRO MATE II Universal Device Programmer PICSTART Plus Development Programmer
25.2.1
MPLAB 6.XX Integrated Development Environment Software

Note: This product is currently available on Microchips web site, www.microchip.com.
The MPLAB Integrated Development Environment (IDE) is available at no cost. MPLAB IDE software is a desktop development environment with tool sets for developing and debugging a microcontroller design application. MPLAB IDE allows quick changes between different development and debugging activities. Designed for use with the Windows operating environment, it is a powerful, affordable, run-time development tool. It is also the common user interface for Microchip's development systems tools, including MPLAB Editor, MPLAB ASM30 Assembler, MPLAB SIM software simulator, MPLAB LIB30 Library, MPLAB LINK30 Linker, MPLAB ICE 4000 In-Circuit Emulators, PRO MATE II programmer and In-Circuit Debugger (ICD 2). The MPLAB IDE gives users the flexibility to edit, compile and emulate, all from a single user interface. Engineers can design and develop code for the dsPIC devices in the same design environment that they have used for PICmicro microcontrollers. The MPLAB IDE is a 32-bit Windows-based application. It provides many advanced features for the engineer in a modern, easy-to-use interface. MPLAB IDE integrates: Full featured, color coded text editor Easy-to-use project manager with visual display Source level debugging Enhanced source level debugging for C - (Structures, automatic variables, etc.) Customizable toolbar and key mapping Dynamic status bar that displays processor condition at a glance Context sensitive, interactive on-line help Integrated MPLAB SIM instruction simulator User interface for PRO MATE II and PICSTART Plus device programmers (sold separately) User interface for MPLAB ICE 4000 In-Circuit Emulator (sold separately) User interface for MPLAB ICD 2 In-Circuit Debugger (sold separately)
DS70072C-page 25-2

The MPLAB IDE allows the engineer to: Edit source files in either assembly or C One-touch compile and download to dsPIC program memory on emulator or simulator. All project information is updated. Debug using: - Source files - Machine code - Mixed mode source and machine code The ability to use the MPLAB IDE with multiple development and debugging targets allows users to easily switch from the cost effective simulator to a full featured emulator with minimal retraining.
25.2.2
dsPIC Language Suite

Note: This product is currently available on Microchips web site, www.microchip.com. The Assembler, Linker and Librarian are included with MPLAB IDE. Contact your local Microchip sales office for availability of the MPLAB C30 C compiler.
The Microchip Technology MPLAB C30 C compiler is a complete, easy-to-use language product. It allows dsPIC applications codes to be written in high level C language and then be fully converted into machine-object code for programming of the microcontroller. It simplifies development of code by removing code obstacles and allowing the designer to focus on program flow and not on program elements. Several options for compiling are available so the user can select those that will maximize the efficiency of the code characteristics. It is a fully ANSI compliant product with standard libraries for the dsPIC family of microcontrollers. It uses the many advanced features of the dsPIC devices to provide very efficient assembly code generation. MPLAB C30 also provides extensions that will allow for excellent support of the hardware, such as interrupts and peripherals. It is fully integrated with the MPLAB IDE for high level, source debugging. Some features include: 16-bit native data types Efficient use of register-based, 3-operand instructions Complex Addressing modes Efficient multi-bit shift operations Efficient signed/unsigned comparisons
MPLAB C30 comes complete with its own assembler, linker and librarian. These allow the user to write mixed mode C and assembly programs and link the resulting object files into a single executable file. The compiler is sold separately. The assembler, linker and librarian is available for free with MPLAB IDE.
25.2.3
MPLAB SIM Software Simulator

Note: This product is included with MPLAB IDE.
The MPLAB SIM software simulator allows code development in a PC-hosted environment by simulating the dsPIC device on an instruction level. On any given instruction, the data areas are able to be examined or modified. The execution is able to be performed in Single Step, Execute Until Break or Trace mode.(1) The MPLAB SIM simulator fully supports symbolic debugging using the MPLAB C30 compiler and assembler. The software simulator offers the flexibility to develop and debug code outside of the laboratory environment, making it an excellent multi-project software development tool. Note 1: Some features, including peripheral support, have not been implemented at the time of this writing. Please check Microchips web site or your local Microchip sales office for the most current information.
25
DS70072C-page 25-3

25.2.4 MPLAB ICE 4000 In-Circuit Emulator
Note: This product is currently under development at the time of this writing. Some of the product details may change. Please refer to the Microchip web site or your local Microchip sales office for the most current information and the availability of this product.
The MPLAB ICE 4000 In-Circuit Emulator will provide the product development engineer with a complete hardware design tool for the dsPIC devices. Software control of the emulator will be provided by MPLAB IDE. The MPLAB ICE 4000 will be a full-featured emulator system with enhanced trace, trigger and data monitoring features. Interchangeable processor modules will allow the system to be easily reconfigured for emulation of different processors. The MPLAB ICE 4000 will support the extended, high-end PICmicro microcontrollers, the PIC18CXXX and PIC18FXXX devices, as well as the dsPIC Family of digital signal controllers. The modular architecture of the MPLAB ICE 4000 in-circuit emulator will allow expansion to support new devices. The MPLAB ICE 4000 in-circuit emulator system has been designed as a real-time emulation system, with advanced features that are generally found on more expensive development tools. Features will include: Full speed emulation, up to 50 MHz bus speed or 200 MHz external clock speed Low voltage emulation down to 1.8 volts Configured with 2 Mb program emulation memory; additional modular memory up to 16 Mb 64K x 136-bit wide Trace Memory Unlimited software breakpoints Complex break, trace and trigger logic Multi-level trigger up to 4 levels Filter trigger functions to trace specific event 16-bit Pass counter for triggering on sequential events 16-bit Delay counter 48-bit time-stamp Stopwatch feature Time between events Statistical performance analysis Code coverage analysis USB and parallel printer port PC connection
DS70072C-page 25-4

25.2.5 MPLAB ICD 2 In-Circuit Debugger
Note: This product is available, but does not provide support for dsPIC30F devices at this time. Please refer to the Microchip web site for information about product upgrades.
Microchip's In-Circuit Debugger, MPLAB ICD, will be a powerful, low cost, run-time development tool. This tool is based on the PICmicro and dsPIC Flash devices. The MPLAB ICD 2 will utilize the in-circuit debugging capability built into the various devices. This feature, along with Microchip's In-Circuit Serial Programming protocol (ICSP), will offer cost effective, in-circuit debugging from the graphical user interface of MPLAB IDE. This will enable a designer to develop and debug source code by watching variables, single-stepping and setting break points. Running at full speed enables testing hardware in real-time. Some of its features will include: Full speed operation to the range of the device Serial or USB PC connector Serial interface externally powered USB powered from PC interface Low noise power (VPP and VDD) for use with analog and other noise sensitive applications Operation down to 2.0V Can be used as an ICD or inexpensive serial programmer Modular application connector as MPLAB ICD Limited number of breakpoints Smart watch variable windows Some chip resources required (RAM, program memory and 2 pins)
25.2.6
PRO MATE II Universal Device Programmer

Note: This product is available, but does not provide support for dsPIC30F devices at this time. Please refer to the Microchip web site for information about product upgrades.
The PRO MATE II universal device programmer will be a full-featured programmer capable of operating in Stand-alone mode, as well as PC-hosted mode. The PRO MATE II device programmer will have programmable VDD and VPP supplies, which will allow it to verify programmed memory at VDDMIN and VDDMAX for maximum reliability when programming requires this capability. It will have an LCD display for instructions and error messages and keys to enter commands. Interchangeable optional socket modules will support all package types. In Stand-alone mode, the PRO MATE II device programmer will be able to read, verify or program PICmicro and dsPIC30F devices. It will also be able to set code protection in this mode. PRO MATE II features will include: Runs under MPLAB IDE Field upgradable firmware DOS Command Line interface for production Host, Safe and Stand-alone operation Automatic downloading of object file SQTPSM serialization adds an unique serial number to each device programmed In-Circuit Serial Programming Kit (sold separately) Interchangeable socket modules supporting all package options (sold separately)
25
DS70072C-page 25-5

25.3 Third Party Hardware/Software Tools and Application Libraries
Note: These products are currently under development at the time of this writing. Some of the product details may change. Please refer to the Microchip web site or your local Microchip sales office for the most current information and the availability of this product.
Microchip is partnering with key third party tool manufacturers for the development of quality hardware and software tools in support of the dsPIC30F Product Family. Microchip plans to offer this initial set of tools and libraries, which will enable customers to rapidly develop their dsPIC30F based application(s). Microchip will expand this current list to provide our customers with additional value added services, (i.e., repository of skilled/certified technical applications contacts, reference designs, hardware and software developers). Please refer to the Microchip web site (www.microchip.com) for the most current information about third party support for the dsPIC30F Device Family. The dsPIC30F software tools and libraries will include: Third Party C compilers Floating Point and Double Precision Math Library DSP Algorithm Library Digital Filter Design Software Utility Peripheral Driver Library CAN Library Real-Time Operating Systems (RTOS) OSEK Operating Systems TCP/IP Protocol Stacks V.22/V.22bis and V.32 ITU Specifications
The dsPIC30F hardware development board tools include: General Purpose Development Board Motor Control Development System Connectivity Development Board
25.3.1
Third Party C Compilers

Note: These products are currently under development at the time of this writing. Some of the product details may change. Please refer to the Microchip web site or your local Microchip sales office for the most current information and the availability of these products.
In addition to the Microchip MPLAB C30 C Compiler, the dsPIC30F will be supported by ANSI C compilers developed by IAR, HI-TECH and Custom Computer Services (CCS). The compilers will allow dsPIC application code to be written in high level C language, and then be fully converted into machine object code for programming of the microcontroller. Each compiler tool will provide several options for compiling, so the user can select those that will maximize the efficiency of the generated code characteristics. The multiple C compiler solutions will have different price targets and features, enabling the customer to select the compiler best suited for their application requirements.
DS70072C-page 25-6

25.3.2 Math Library
The Math Library will support several standard C functions, including, but not limited to: sin(), cos(), tan() asin(), acos(), atan(), log(), log10() sqrt(), power() ceil(), floor() fmod(), frexp()
The math function routines will be developed and optimized in dsPIC30F assembly language and will be callable from both assembly and C language. Floating point and double precision versions of each function shall be provided. The Microchip MPLAB C30 and IAR C compilers will be supported.
25.3.3
DSP Algorithm Library

The DSP library will support multiple filtering, convolution, vector and matrix functions. Some of the functions will include, but will not be limited to: Cascaded Infinite Impulse Response (IIR) Filters Correlation Convolution Finite Impulse Response (FIR) Filters Windowing Functions FFTs LMS Filter Vector Addition and Subtraction Vector Dot Product Vector Power Matrix Addition and Subtraction Matrix Multiplication
25
DS70072C-page 25-7

25.3.4 DSP Filter Design Software Utility
Microchip will offer a digital filter design software tool which will enable the user to develop optimized assembly code for Low-pass, High-pass, Band-pass and Band-stop IIR and FIR filters, including 16-bit fractional data size filter coefficients from a graphical user interface. The application developer will enter the required filter frequency specifications and the software tool develops the filter code and coefficients. Ideal filter frequency response and time domain plots are generated for analysis. FIR filter lengths up to 513 taps and IIR filter lengths up to 10 cascaded sections will be supported. All IIR and FIR routines are generated in assembly language and will be callable from both assembly and C language. The Microchip MPLAB C30 C compiler will be supported.
25.3.5
Peripheral Driver Library

Microchip will offer a peripheral driver library that will support the setup and control of dsPIC30F hardware peripherals, including, but not limited to: Analog-to-Digital Converter Motor Control PWM Quadrature Encoder Interface UART SPI Data Converter Interface I2C General Purpose Timers Input Capture Output Compare/Simple PWM
DS70072C-page 25-8

25.3.6 CAN Library
Microchip will offer a CAN driver library, which will support the dsPIC30F CAN peripheral. Some of the CAN functions which will be supported are: Initialize CAN Module Set CAN Operational Mode Set CAN Baud Rate Set CAN Masks Set CAN Filters Send CAN Message Receive CAN Message Abort CAN Sequence Get CAN TX Error Count Get CAN RX Error Count
25.3.7
Real-Time Operating System (RTOS)

Real-Time Operating System (RTOS) solutions for the dsPIC30F Product Family will be provided. These RTOS solutions will provide the necessary function calls and operating system routines to write efficient C and/or assembly code for multi-tasking applications. In addition, RTOS solutions will be provided that address those applications in which program and more importantly, data memory resources, are limited. Configurable and optimized kernels will be available to support various RTOS application requirements. The RTOS solutions will range from a fully-true, preemptive and multi-tasking scheduler to a cooperative type scheduler, both of which will be designed to optimally run on the dsPIC30F devices. Depending on the RTOS implementation, some of the function calls provided in the system kernel will be: Control Tasks Send And Receive Messages Handle Events Control Resources Control Semaphores Regulate Timing in a Variety of Ways Provide Memory Management Handle Interrupts and Swap Tasks
Most functions will be written in ANSI C, with the exception of time critical functions, which will be optimized in assembly, thereby reducing execution time for maximum code efficiency. The ANSI C and assembly routines will be supported by the Microchip MPLAB C30 C compiler. Electronic documentation will accompany the RTOS, enabling the user to efficiently understand and implement the RTOS in their application.
25
DS70072C-page 25-9

25.3.8 OSEK Operating Systems
Note: This product is currently under development at the time of this writing. Some of the product details may change. Please refer to the Microchip web site or your local Microchip sales office for the most current information and the availability of this product. OSEK/VDX will be developed for support of OSEK, Offene Systeme und deren (Open systems and the corresponding with VDX Vehicle Distributed eXecutive
Operating Systems for the vehicle software standard of the dsPIC30F product family. The functionality Schnittstellen fr die Elektronik im Kraftfahrzeug interfaces for automotive electronics), is harmonized yielding OSEK/VDX.
Structured and modular RTOS software implementations based on standardized interfaces and protocols will be provided. Structured and modular implementations will provide for portability and extendability for distributed control units for vehicles. Various OSEK COM modules will be provided, such as: OSEK/COM Standard API OSEK/COM Communication API OSEK/COM Network API OSEK/COM Standard Protocols OSEK/COM Device Driver Interface
Microchip will also provide Internal and External CAN driver support. The physical layer will be integrated into the communication controllers hardware and will not be covered by the OSEK specifications. Most module functions will be developed in ANSI C, with the exception of time critical functions and peripheral utilization, which will be optimized in assembly, thereby reducing execution time for maximum code efficiency. The Microchip MPLAB C30 C compiler will be supported.
25.3.9
TCP/IP Protocol Stack

Microchip will offer various Transmission Control Protocol/Internet Protocol (TCP/IP) Stack Layer solutions for Internet connectivity solutions implemented on the dsPIC30F product family. Both reduced and full stack implementations will be provided, which will allow the user to select the optimum TCP/IP stack solution for their application. Application protocol layers, such as FTP, TFTP and SMTP, Transport and Internet layers, such as TCP, UDP, ICMP and IP, and Network Access layers, such as PPP, SLIP, ARP and DHCP, will be provided. Various configurations, such as a minimal UDP/IP stack will be available for limited connectivity requirements. Most stack protocol functions will be developed and optimized in Microchips MPLAB C30 C language. Assembly language coding may be developed for specific dsPIC30F hardware peripherals and Ethernet drivers to optimize code size and execution time. These assembly language specific routines will be assembly and C callable. Electronic documentation will accompany the TCP/IP protocol stack, enabling the user to efficiently understand and implement the protocol stack in their application.
DS70072C-page 25-10

25.3.10 V.22/V.22bis and V.32 Specification
Microchip will offer ITU compliant V.22/V.22bis (1200/2400 bps) and V.32 (non-trellis coding at 9600 bps) modem specifications to support a range of connected applications. Applications which will benefit from these modem specifications will be numerous and will fall into many applications, some of which are listed here: Internet enabled home security systems Internet connected power, gas and water meters Internet connected vending machines Smart Appliances Industrial monitoring POS Terminals Set Top Boxes Drop Boxes Fire Panels
Most ITU specification modules will be developed and optimized in Microchips MPLAB C30 C language. Assembly language coding may be developed for specific dsPIC30F hardware peripherals, along with key transmitter and receiver filtering routines to optimize code size and execution time. These assembly language specific routines will be assembly and C callable. Electronic documentation will accompany the modem library, enabling the user to efficiently understand and implement the library functions.
25.4
dsPIC30F Hardware Development Boards

Note: These products are currently under development at the time of this writing. Some of the product details may change. Please refer to the Microchip web site or your local Microchip sales office for the most current information and the availability of these products.
Microchip will initially provide three hardware development boards, which will provide the application developer with a tool in which to quickly prototype and validate key design requirements. Each board will feature key dsPIC30F peripherals and support Microchips MPLAB In-Circuit Debugger (ICD 2) tool for cost effective debugging and programming of the dsPIC30F device. The three initial boards to be provided are: General Purpose Development Board Motor Control Development System Connectivity Development Board
25
DS70072C-page 25-11

25.4.1 General Purpose Development Board
The dsPIC30F general purpose development board will provide the application designer with a low cost development tool in which to become familiar with the dsPIC30F 16-bit architecture, high performance peripherals and powerful instruction set. The development board will serve as an ideal prototyping tool in which to quickly develop and validate key design requirements. Some key features and attributes of the general purpose development board will be: Supports various dsPIC30F packages CAN communication channel RS-232 and RS-485 communication channels Codec interface with line in/out jacks In-Circuit Debugger interface MPLAB ICE 4000 emulation support Microchip temperature sensor Microchip Op Amp circuit, supporting user input signals Microchip Digital-to-Analog Converter 2x16 LCD General purpose prototyping area Various LEDS, switches and potentiometers
The general purpose development board will be shipped with a 9V power supply, RS-232 I/O cable, preprogrammed dsPIC30F device, example software and appropriate documentation to enable the user to exercise the development board demonstration programs.
DS70072C-page 25-12

25.4.2 Motor Control Development System
The dsPIC30F motor control development system will initially provide the application developer with three main components for quick prototyping and validation of BLDC, PMAC and ACIM applications. The three main components will be: dsPIC30F Motor Control Main Board 3-phase Low Voltage Power Module 3-phase High Voltage Power Module The main control board will support the dsPIC30F6010 device, various peripheral interfaces, and a custom interface header system that will allow different motor power modules to be connected. The control board also will have connectors for mechanical position sensors, such as incremental rotary encoders and hall effect sensors, and a breadboard area for custom circuits. The main control board will receive its power from a standard plug-in transformer. The low voltage power module will be optimized for 3-phase motor applications that will require a DC bus voltage less than 60 volts and will deliver up to 400W power output. The 3-phase low voltage power module is intended to power BLDC and PMAC motors. The high voltage power module will be optimized for 3-phase motor applications that require DC bus voltages up to 400 volts and up to 1 kW power output. The high voltage module will have an active power factor correction circuit that will be controlled by the dsPIC30F device. This power module is intended for AC induction motor and power inverter applications. Both power modules will have automatic Fault protection and electrical isolation from the control interface. Both power module boards will provide preconditioned voltage and current signals to the main control board. All position feedback devices that will be isolated from the motor control circuitry, such as incremental encoders, hall-effect sensors or tachometer sensors, will be directly connected to the main control board. Both modules will be equipped with motor braking circuits.
25
DS70072C-page 25-13

25.4.3 Connectivity Development Board
The dsPIC30F connectivity development board will provide the application developer a basic platform for developing and evaluating various connectivity solutions, implementing TCP/IP protocol layers combined with V.22/V.22bis and V.32 (non-trellis coding) ITU specifications, across PSTN or Ethernet communication channels. Some key features and attributes of the connectivity development board will be: Supports the dsPIC30F6014 device Media Access Control (MAC) and PHY interface PSTN interface with DAA/AFE RS-232 and RS-485 communication channels In-Circuit Debugger interface MPLAB ICE 4000 emulation support Microchip temperature sensor Microchip Digital-to-Analog Converter 2x16 LCD General purpose prototyping area Various LEDs, switches and potentiometers
The connectivity development board will be shipped with a 9V power supply, RS-232 I/O cable and preprogrammed dsPIC30F devices with example connectivity software and appropriate documentation to enable the user to exercise the development board connectivity demo program.
DS70072C-page 25-14

This section lists application notes that are related to this section of the manual. These application notes may not be written specifically for the dsPIC30F Product Family, but the concepts are pertinent and could be used with modification and possible limitations. The current application notes related to the Development Tool Support are: Title No related application notes at this time. Application Note #
Note:
25
DS70072C-page 25-15

This is the initial released revision of the dsPIC30F Development Tool Support description.
Revision B
Revision C
DS70072C-page 25-16
26
Appendix
Section 26. Appendix

HIGHLIGHTS
This section of the manual contains the following topics: Appendix A: I2C Overview .................................................................................................. 26-2 Appendix B: CAN Overview ................................................................................................. 26-12 Appendix C: Codec Protocol Overview ................................................................................ 26-25
I2C is a trademark of Philips Corporation.
DS70074C-page 26-1

APPENDIX A: I 2C OVERVIEW
This appendix provides an overview of the Inter-Integrated Circuit (I 2C) bus, with Subsection A.2 Addressing I2C Devices discussing the operation of the SSP modules in I 2C mode. The I 2C bus is a two-wire serial interface. The original specification, or standard mode, is for data transfers of up to 100 Kbps. An enhanced specification, or fast mode (400 Kbps), is supported. Standard and Fast mode devices will operate when attached to the same bus, if the bus operates at the speed of the slower device. The I 2C interface employs a comprehensive protocol to ensure reliable transmission and reception of data. When transmitting data, one device is the master, which initiates transfer on the bus and generates the clock signals to permit that transfer, while the other device(s) acts as the slave. All portions of the slave protocol are implemented in the SSP modules hardware, except general call support, while portions of the master protocol need to be addressed in the PIC16CXX software. The MSSP module supports the full implementation of the I 2C master protocol, the general call address and data transfers up to 1 Mbps. The 1 Mbps data transfers are supported by some of Microchips Serial EEPROMs. Table A-1 defines some of the I 2C bus terminology. In the I 2C interface protocol, each device has an address. When a master wishes to initiate a data transfer, it first transmits the address of the device that it wishes to talk to. All devices listen to see if this is their address. Within this address, a bit specifies if the master wishes to read-from/write-to the slave device. The master and slave are always in opposite modes (transmitter/receiver) of operation during a data transfer. That is, they can be thought of as operating in either of these two relations: Master-transmitter and Slave-receiver Slave-transmitter and Master-receiver In both cases, the master generates the clock signal. The output stages of the clock (SCL) and data (SDA) lines must have an open-drain or open-collector in order to perform the wired-AND function of the bus. External pull-up resistors are used to ensure a high level when no device is pulling the line down. The number of devices that may be attached to the I 2C bus is limited only by the maximum bus loading specification of 400 pF and addressing capability.
DS70074C-page 26-2
Section 26. Appendix 26

A.1 Initiating and Terminating Data Transfer During times of no data transfer (idle time), both the clock line (SCL) and the data line (SDA) are pulled high through the external pull-up resistors. The Start and Stop conditions determine the start and stop of data transmission. The Start condition is defined as a high-to-low transition of the SDA when the SCL is high. The Stop condition is defined as a low-to-high transition of the SDA when the SCL is high. Figure A-1 shows the Start and Stop conditions. The master generates these conditions for starting and terminating data transfer. Due to the definition of the Start and Stop conditions, when data is being transmitted, the SDA line can only change state when the SCL line is low. Figure A-1: Start and Stop Conditions
Appendix
SDA SCL S P Change Stop of Data Condition Allowed
Change Start Condition of Data Allowed
Table A-1: Term Transmitter Receiver Master
I2C Bus Terminology Description
The device that sends the data to the bus. The device that receives the data from the bus. The device which initiates the transfer, generates the clock and terminates the transfer. Slave The device addressed by a master. Multi-master More than one master device in a system. These masters can attempt to control the bus at the same time without corrupting the message. Arbitration Procedure that ensures that only one of the master devices will control the bus. This ensures that the transfer data does not get corrupted. Synchronization Procedure where the clock signals of two or more devices are synchronized.
DS70074C-page 26-3

A.2 Addressing I 2C Devices There are two address formats. The simplest is the 7-bit address format with a R/W bit (Figure A-2). The more complex is the 10-bit address with a R/W bit (Figure A-3). For 10-bit address format, two bytes must be transmitted. The first five bits specify this to be a 10-bit address format. The 1st transmitted byte has 5 bits which specify a 10-bit address, the two MSbs of the address, and the R/W bit. The second byte is the remaining 8 bits of the address. Figure A-2: 7-bit Address Format
MSb S Slave Address LSb R/W ACK
Sent by Slave
S R/W ACK
Start Condition Read/Write pulse Acknowledge
Figure A-3:
I2C 10-bit Address Format
1 1 1 1 0 A9 A8R/W ACK A7 A6 A5 A4 A3 A2 A1 A0 Sent By Slave = 0 for write
ACK
S R/W ACK
- Start Condition - Read/Write Pulse - Acknowledge
DS70074C-page 26-4

A.3 Transfer Acknowledge All data must be transmitted per byte, with no limit to the number of bytes transmitted per data transfer. After each byte, the slave-receiver generates an Acknowledge bit (ACK) (Figure A-4). When a slave-receiver doesnt acknowledge the slave address or received data, the master must abort the transfer. The slave must leave SDA high so that the master can generate the Stop condition (Figure A-1). Figure A-4: Slave-Receiver Acknowledge
Data Output by Transmitter Data Output by Receiver SCL from Master S Start Condition 1
Appendix
not acknowledge acknowledge 2 8 9
Clock Pulse for Acknowledgment
If the master is receiving the data (master-receiver), it generates an Acknowledge signal for each received byte of data, except for the last byte. To signal the end of data to the slave-transmitter, the master does not generate an acknowledge (not acknowledge). The slave then releases the SDA line so the master can generate the Stop condition. The master can also generate the Stop condition during the Acknowledge pulse for valid termination of data transfer. If the slave needs to delay the transmission of the next byte, holding the SCL line low will force the master into a wait state. Data transfer continues when the slave releases the SCL line. This allows the slave to move the received data or fetch the data it needs to transfer before allowing the clock to start. This wait state technique can also be implemented at the bit level, Figure A-5. Figure A-5: Data Transfer Wait State
SDA MSb Acknowledgment Byte Complete Signal from Receiver Interrupt with Receiver Acknowledgment Signal from Receiver
Clock Line Held Low while Interrupts are Serviced SCL S Start Condition 1 2 Address 7 8 R/W 9 ACK Wait State 1 2 Data 38 9 ACK P Stop Condition
DS70074C-page 26-5

Figure A-6 and Figure A-7 illustrate master-transmitter and master-receiver data transfer sequences. Figure A-6: Master-Transmitter Sequence
For 7-bit address: S Slave Address '0' (write) R/W A Data A Data A/A P
data transferred (n bytes - acknowledge)
A master transmitter addresses a slave receiver with a 7-bit address. The transfer direction is not changed.
For 10-bit address: S Slave Address R/W A1 (Code + A9:A8) (write) A master transmitter addresses a slave receiver with a 10-bit address. A = acknowledge (SDA low) A = not acknowledge (SDA high) S = Start Condition P = Stop Condition Slave Address (A7:A0) A2 Data A Data A/A P
From master to slave From slave to master
Figure A-7:
Master-Receiver Sequence
For 7-bit address: S Slave Address '1' (read) R/W A Data A Data A P
data transferred (n bytes - acknowledge)
A master reads a slave immediately after the first byte.
For 10-bit address: S Slave Address (Code + A9:A8) (write) A master transmitter addresses a slave receiver with a 10-bit address. A = acknowledge (SDA low) A = not acknowledge (SDA high) S = Start Condition P = Stop Condition R/W A1 Slave Address (A7:A0) A2 Sr Slave Address (Code + A9:A8) (read) R/W A3 Data A Data A P
DS70074C-page 26-6

When a master does not wish to relinquish the bus (which occurs by generating a Stop condition), a repeated Start condition (Sr) must be generated. This condition is identical to the Start condition (SDA goes high-to-low, while SCL is high), but occurs after a data transfer Acknowledge pulse (not the bus-free state). This allows a master to send commands to the slave and then receive the requested information or to address a different slave device. This sequence is illustrated in Figure A-8. Figure A-8: Combined Format
(read or write) (n bytes + acknowledge) S Slave Address R/W A Data A/A Sr Slave Address (read) Sr = repeated Start Condition R/W A Data A/A P
Appendix
(write)
Direction of transfer may change at this point
Transfer direction of data and acknowledgment bits depends on R/W bits. Combined format: Sr Slave Address R/W A (Code + A9:A8) (write) Slave Address A Data A (A7:A0) Data A/A Sr Slave Address R/W A Data A (Code + A9:A8) (read) Data A P
Combined format - A master addresses a slave with a 10-bit address, then transmits data to this slave and reads data from this slave. A = acknowledge (SDA low) A = not acknowledge (SDA high) S = Start Condition P = Stop Condition
DS70074C-page 26-7

A.4 Multi-master The I2C protocol allows a system to have more than one master. This is called a multi-master system. When two or more masters try to transfer data at the same time, arbitration and synchronization occur. A.4.1 Arbitration Arbitration takes place on the SDA line, while the SCL line is high. The master which transmits a high when the other master transmits a low, loses arbitration (Figure A-9) and turns off its data output stage. A master which lost arbitration can generate clock pulses until the end of the data byte where it lost arbitration. When the master devices are addressing the same device, arbitration continues into the data. Figure A-9: Multi-Master Arbitration (Two Masters)
Transmitter 1 Loses Arbitration DATA 1 SDA DATA 1 DATA 2 SDA SCL
Masters that also incorporate the slave function, and have lost arbitration must immediately switch over to Slave-receiver mode. This is because the winning master-transmitter may be addressing it. Arbitration is not allowed between: A repeated Start condition A Stop condition and a data bit A repeated Start condition and a Stop condition Care needs to be taken to ensure that these conditions do not occur.
DS70074C-page 26-8

A.4.2 Clock Synchronization Clock synchronization occurs after the devices have started arbitration. This is performed using a wired-AND connection to the SCL line. A high-to-low transition on the SCL line causes the concerned devices to start counting off their low period. Once a device clock has gone low, it will hold the SCL line low until its SCL high state is reached. The low-to-high transition of this clock may not change the state of the SCL line if another device clock is still within its low period. The SCL line is held low by the device with the longest low period. Devices with shorter low periods enter a high wait-state until the SCL line comes high. When the SCL line comes high, all devices start counting off their high periods. The first device to complete its high period will pull the SCL line low. The SCL line high time is determined by the device with the shortest high period, Figure A-10. Figure A-10: Clock Synchronization
Wait State Start Counting High Period
Appendix
CLK 1 Counter Reset
CLK 2
SCL
DS70074C-page 26-9

Table A-2 and Table A-3 show the specifications of a compliant I2C bus. The column titled, Parameter No., is provided to ease the users correlation to the corresponding parameter in the device data sheet. Figure A-11 and Figure A-12 show these times on the appropriate waveforms. Figure A-11: I2C Bus Start/Stop Bits Timing Specification
SCL 90 SDA
91 92
93
Start Condition
Stop Condition
Table A-2: Parameter No. 90 91 92 93
I2C Bus Start/Stop Bits Timing Specification Sym Characteristic 100 kHz mode 400 kHz mode 100 kHz mode 400 kHz mode 100 kHz mode 400 kHz mode 100 kHz mode 400 kHz mode Min 4700 600 4000 600 4700 600 4000 600 Typ Max Units ns ns ns ns Conditions Only relevant for repeated Start condition After this period the first clock pulse is generated
TSU:STA Start condition Setup time THD:STA Start condition Hold time TSU:STO Stop condition Setup time THD:STO Stop condition Hold time
Figure A-12:
I2C Bus Data Timing Specification
103 SCL
100 101
102
90 91
106
107 92
SDA In 110 109 SDA Out MSb 109
DS70074C-page 26-10

Table A-3: Parameter No. 100 101 102 I2C Bus Data Timing Specification Sym THIGH TLOW TR Characteristic Clock high time Clock low time SDA and SCL rise time SDA and SCL fall time 100 kHz mode 400 kHz mode 100 kHz mode 400 kHz mode 100 kHz mode 400 kHz mode 100 kHz mode 400 kHz mode Min 4.0 0.6 4.7 1.3 20 + 0.1Cb 20 + 0.1Cb 4.7 0.6 4.0 0.6 0 0 250 100 4.7 0.6 4.7 1.3 Max 1000 300 300 300 0.9 3500 1000 Units s s s s ns ns ns ns s s s s ns s ns ns s s ns ns s s Conditions
Appendix
Cb is specified to be from 10 to 400 pF Cb is specified to be from 10 to 400 pF Only relevant for repeated Start condition After this period the first clock pulse is generated
103
TF
90 91 106 107 92 109 110
TSU:STA Start condition setup time THD:STA THD:DAT TSU:DAT TSU:STO TAA TBUF
100 kHz mode 400 kHz mode Start condition hold 100 kHz mode time 400 kHz mode Data input hold time 100 kHz mode 400 kHz mode Data input setup time 100 kHz mode 400 kHz mode Stop condition setup 100 kHz mode time 400 kHz mode Output valid from 100 kHz mode clock 400 kHz mode Bus free time 100 kHz mode 400 kHz mode
Note 2
Note 1 Time the bus must be free before a new transmission can start
D102 Cb Bus capacitive loading 400 pF Note 1: As a transmitter, the device must provide this internal minimum delay time to bridge the undefined region (min. 300 ns) of the falling edge of SCL to avoid unintended generation of Start or Stop conditions. 2: A Fast mode I2C-bus device can be used in a Standard mode I2C-bus system, but the requirement TSU;DAT 250 ns must then be met. This will automatically be the case if the device does not stretch the LOW period of the SCL signal. If such a device does stretch the LOW period of the SCL signal, it must output the next data bit to the SDA line, TR max.+TSU; DAT = 1000 + 250 = 1250 ns (according to the Standard mode I2C bus specification) before the SCL line is released.
DS70074C-page 26-11

APPENDIX B: CAN OVERVIEW
This appendix provides an overview of the Controller Area Network (CAN) bus. The CAN Section of this reference manual discusses the implementation of the CAN protocol for that hardware module. B.1 CAN Bus Background The Controller Area Network (CAN) is a serial communications protocol which efficiently supports distributed real-time control with a very high level of security. Its domain of application ranges from high speed networks to low cost multiplex wiring. In automotive electronics, engine control units, sensors, anti-skid-systems, etc., are connected using CAN with bit rates up to 1 Mbit/sec. The silicon cost is also low enough to be cost effective at replacing wiring harnesses in the automobile. The robustness of the bus in noisy environments and the ability to detect and recover from fault conditions makes the bus suitable for industrial control applications such as DeviceNet, SDS and other field bus protocols. CAN is an asynchronous serial bus system with one logical bus line. It has an open, linear bus structure with equal bus nodes. A CAN bus consists of two or more nodes. The number of nodes on the bus may be changed dynamically without disturbing the communication of other nodes. This allows easy connection and disconnection of bus nodes (e.g., for addition of system function, error recovery or bus monitoring). The bus logic corresponds to a wired-AND mechanism, recessive bits (mostly, but not necessarily equivalent to the logic level 1) are overwritten by dominant bits (mostly logic level 0). As long as no bus node is sending a dominant bit, the bus line is in the recessive state, but a dominant bit from any bus node generates the dominant bus state. Therefore, for the CAN bus line, a medium must be chosen that is able to transmit the two possible bit states (dominant and recessive). One of the most common and cheapest ways is to use a twisted wire pair. The bus lines are then called CANH and CANL, and may be connected directly to the nodes or via a connector. There's no standard defined by CAN regarding the connector to be used. The twisted wire pair is terminated by terminating resistors at each end of the bus line. The maximum bus speed is 1 Mbit, which can be achieved with a bus length of up to 40 meters. For bus lengths longer than 40 meters, the bus speed must be reduced (a 1000 m bus can be realized with a 40 Kbit bus speed). For a bus length above 1000 meters, special drivers should be used. At least 20 nodes may be connected without additional equipment. Due to the differential nature of transmission, CAN is insensitive to EMI because both bus lines are affected in the same way which leaves the differential signal unaffected. The bus lines can also be shielded to reduce the electromagnetic emission of the bus itself, especially at high baud rates. The binary data is coded corresponding to the NRZ code (Non-Return-to-Zero; low level = dominant state; high level = recessive state). To ensure exact synchronization of all bus nodes, bit stuffing is used. This means that during the transmission of a message a maximum of five consecutive bits may have the same polarity. Whenever five consecutive bits of the same polarity have been transmitted, the transmitter will insert one additional bit of the opposite polarity into the bit stream before transmitting further bits. The receiver also checks the number of bits with the same polarity and removes the stuff bits from the bit stream (-destuffing). In the CAN protocol it is not bus nodes that are addressed, but the address information is contained in the messages that are transmitted. This is done via an identifier (part of each message) which identifies the message content (e.g., engine speed, oil temperature etc.). The identifier additionally indicates the priority of the message. The lower the binary value of the identifier, the higher the priority of the message.
DS70074C-page 26-12

For bus arbitration, CSMA/CD with NDA is used (Carrier Sense Multiple Access/Collision Detection with Non-Destructive Arbitration). If bus node A wants to transmit a message across the network, it first checks that the bus is in the Idle state (Carrier Sense) (i.e., no node is currently transmitting). If this is the case (and no other node wishes to start a transmission at the same moment), node A becomes the bus master and sends its message. All other nodes switch to Receive mode during the first transmitted bit (Start-Of-Frame bit). After correct reception of the message (which is acknowledged by each node), each bus node checks the message identifier and stores the message, if required. Otherwise, the message is discarded. If two or more bus nodes start their transmission at the same time (Multiple Access), collision of the messages is avoided by bitwise arbitration (Collision Detection/Non-Destructive Arbitration together with the Wired-AND mechanism, dominant bits override recessive bits). Each node sends the bits of its message identifier (MSB first) and monitors the bus level. A node that sends a recessive identifier bit but reads back a dominant one loses bus arbitration and switches to Receive mode. This condition occurs when the message identifier of a competing node has a lower binary value (dominant state = logic 0) and therefore, the competing node is sending a message with a higher priority. In this way, the bus node with the highest priority message wins arbitration without losing time by having to repeat the message. All other nodes automatically try to repeat their transmission once the bus returns to the Idle state. It is not permitted for different nodes to send messages with the same identifier as arbitration could fail leading to collisions and errors. The original CAN specifications (versions 1.0, 1.2 and 2.0A) defined the message identifier as having a length of 11 bits giving a possible 2048 message identifiers. The specification has since been updated (to version 2.0B) to remove this possible limitation. CAN specification, version 2.0B, allows message identifier lengths of 11 and/or 29 bits to be used (an identifier length of 29 bits allows over 536 Million message identifiers). Version 2.0B CAN is also referred to as Extended CAN and versions 1.0, 1.2 and 2.0A are referred to as Standard CAN. B.2 B.2.1 Different CAN Implementations Standard CAN, Extended CAN Those data frames and remote frames, which only contain the 11-bit identifier, are called standard frames according to CAN specification V2.0A. With these frames, 2048 different messages can be identified (identifiers 0-2047). However, the 16 messages with the lowest priority (2032-2047) are reserved. Extended frames, according to CAN specification V2.0B, have a 29-bit identifier. As already mentioned, this 29-bit identifier is made up of the 11-bit identifier (Base lD) and the 18-bit Extended identifier (ID Extension). CAN modules specified by CAN V2.0A are only able to transmit and receive standard frames according to the Standard CAN protocol. Messages using the 29-bit identifier cause errors. If a device is specified by CAN V2.0B, there is one more distinction. Modules named Part B Passive can only transmit and receive standard frames, but tolerate extended frames without generating error frames. Part B Active devices are able to transmit and receive both standard and extended frames. B.3 Basic CAN, Full CAN There is one more CAN characteristic concerning the interface between the CAN module and the host CPU, dividing CAN chips into Basic CAN and Full CAN devices. This has nothing to do with the used protocol though (Standard or Extended CAN), which makes it possible to use both Basic and Full CAN devices in the same network. In the Basic CAN devices, only basic functions of the protocol are implemented in hardware, (e.g., the generation and the check of the bit steam). The decision, if a received message has to be stored or not (acceptance filtering), and the whole message management, has to be done by software (i.e., by the host CPU). Mostly the CAN chip only provides one transmit buffer and one or two receive buffers. So the host CPU load is quite high using Basic CAN modules, therefore these devices should only be used at low baud rates and low bus loads with only a few different messages. The advantages of Basic CAN are the small chip size leading to low costs of these devices.
Appendix
DS70074C-page 26-13

Full CAN devices do the whole bus protocol in hardware, including the acceptance filtering and the message management. They contain several so called message objects which handle the identifier, the data, the direction (receive or transmit) and the information Standard CAN/Extended CAN. During the initialization of the device, the host CPU defines which messages are to be sent and which are to be received. The host CPU is informed by interrupt if the identifier of a received message matches with one of the programmed (receive-) message objects. In this way. the CPU load is reduced. Using Full CAN devices, high baud rates and high bus loads with many messages can be handled. These chips are more expensive than the Basic CAN devices, though. Many Full CAN chips provide a Basic-CAN Feature. One of their messages objects can be programmed in a way that every message is stored there that does not match with one of the other message objects. This can be very helpful in a number of applications. B.4 ISO Model The lSO/OSl Reference Model is used to define the layers of protocol of a communication system, as shown in Figure B-1. At the highest end, the applications need to communicate between each other. At the lowest end, some physical medium is used to provide electrical signaling. The higher levels of the protocol are run by software. Typically, only the application layer is implemented. Within the CAN bus specification, there is no definition of the type of message or the contents or meaning of the messages transferred. These definitions are made in systems such as Volcano, the Volvo automotive CAN specification; J1939, the U.S. heavy truck multiplex wiring spec; and Allen-Bradly DeviceNet and Honeywell SDS, examples of industrial protocols. The CAN bus module definition encompasses two levels of the overall protocol. The Data Link Layer - The Logical Link Control (LLC) sub-layer - The Medium Access Control (MAC) sub-layer - The Physical Layer - The Physical Signaling (PLS) sub-layer The LLC sub layer is concerned with Message Filtering, Overload Notification and Error Recovery Management. The scope of the LLC sub-layer is: To provide services for data transfer and for remote data request To decide which messages received by the LLC sub layer are actually to be accepted To provide means for error recovery management and overload notifications
The MAC sub-layer represents the kernel of the CAN protocol. The MAC sub-layer defines the transfer protocol (i.e., controlling the Framing, Performing Arbitration, Error Checking, Error Signalling and Fault Confinement). It presents messages received from the LLC sub-layer and accepts messages to be transmitted to the LLC sub-layer. Within the MAC sub-layer, it is decided whether the bus is free for starting a new transmission or whether a reception is just starting. The MAC sub-layer is supervised by a management entity called Fault Confinement, which is a self-checking mechanism for distinguishing short disturbances from permanent failures. Also, some general features of the bit timing are regarded as part of the MAC sub-layer. The physical layer defines the actual transfer of the bits between the different nodes with respect to all electrical properties. The PLS sub-layer defines how signals are actually transmitted and therefore deals with the description of Bit Timing, Bit Encoding and Synchronization. The lower levels of the protocol are implemented in driver/receiver chips and the actual interface, such as twisted pair wiring or optical fiber etc. Within one network, the physical layer has to be the same for all nodes. The Driver/Receiver Characteristics of the Physical Layer are not defined so as to allow transmission medium and signal level implementations to be optimized for their application. The most common example is defined in the ISO11898 Road Vehicles Multiplex Wiring specification.
DS70074C-page 26-14

Figure B-1: CAN Bus in ISO/OSI Reference Model OSI REFERENCE LAYERS Application Presentation Session Transport Network Data Link Layer Supervisor LLC (Logical Link Control) Acceptance Filtering Overload Notification Recovery Management MAC (Medium Access Control) Data Encapsulation/Decapsulation Frame Coding (stuffing, destuffing) Medium Access Management Error Detection Error Signalling Acknowledgment Serialization/Deserialization Physical Layer PLS (Physical Signalling) Bit Encoding/Decoding Bit Timing Synchronization PMA (Physical Medium Attachment) Driver/Receiver Characteristics MDI (Medium Dependent Interface) Connectors Bus Failure management (PLS-LME) SHADED REGIONS DEFINED BY CAN BUS SPECIFICATION
Appendix
Fault confinement (MAC-LME)
DS70074C-page 26-15

B.5 CAN Bus Features CAN has the following properties: Prioritization of messages Latency times ensured Configuration flexibility Multi-cast reception with time synchronization System wide data consistency Multi-master Error detection and signaling Automatic retransmission of corrupted messages Distinction between temporary errors and permanent failures of nodes and autonomous switching off of defect nodes 1. Messages information on the bus is sent in fixed format messages of different but limited length. When the bus is free any connected unit may start to transmit a new message. 2. Information Routing In CAN systems, a CAN node does not make use of any information about the system configuration (e.g., station addresses). 3. System Flexibility Nodes can be added to the CAN network without requiring any change in the software or hardware of any node and application layer. 4. Message Routing The content of a message is named by an ldentifier. The ldentifier does not indicate the destination of the message, but describes the meaning of the data, so that all nodes in the network are able to decide by Message Filtering whether the data is to be acted upon by them or not. 5. Multicast As a consequence of the concept of Message Filtering, any number of nodes can receive and simultaneously act upon the same message. 6. Data Consistency Within a CAN network, it is ensured that a message is simultaneously accepted either by all nodes or by no node. Thus, data consistency of a system is achieved by the concepts of multicast and by error handling. 7. Bit Rate The speed of CAN may be different in different systems. However, in a given system, the bit-rate is uniform and fixed. 8. Prioritization The ldentifier defines a static message priority during bus access. 9. Remote Data Request By sending a remote frame, a node requiring data may request another node to send the corresponding data frame. The date frame and the corresponding remote frame are named by the same ldentifier. 10. Multimaster When the bus is free, any unit may start to transmit a message. The unit with the message of higher priority to be transmitted gains bus access. 11. Arbitration Whenever the bus is free, any unit may start to transmit a message. If 2 or more units start transmitting messages at the same time, the bus access conflict is resolved by bitwise arbitration using the ldentifier. The mechanism of arbitration ensures that neither information nor time is lost. If a data frame and a remote frame with the same ldentifier are initiated at the same time, the data frame prevails over the remote frame. During arbitration, every transmitter compares the level of the bit transmitted with the level that is monitored on the bus. If these levels are equal, the unit may continue to send. When a recessive level is sent and a dominant level is monitored, the unit has lost arbitration and must withdraw without sending one more bit. 12. Safety In order to achieve the highest safety of data transfer, powerful measures for error detection, signaling and self-checking are implemented in every CAN node. 13. Error Detection For detecting errors the following measures have been taken: Monitoring (transmitters compare the bit levels to be transmitted with the bit levels detected on the bus) Cyclic Redundancy Check Bit Stuffing Message Frame Check
DS70074C-page 26-16

The error detection mechanisms have the following properties: all global errors are detected all local errors at transmitters are detected up to 5 randomly distributed errors in a message are detected burst errors of length less than 15 in a message are detected errors of any odd number in a message are detected Error Signalling and Recovery Time Corrupted messages are flagged by any node detecting an error. Such messages are aborted and will be retransmitted automatically. The recovery time from detecting an error until the start of the next message is at most 31 bit times, if there is no further error. Fault Confinement CAN nodes are able to distinguish short disturbances from permanent failures. Defective nodes are switched off. Connections The CAN serial communication link is a bus to which a number of units may be connected. This number has no theoretical limit. Practically the total number of units will be limited by delay times and/or electrical loads on the bus line. Single Channel The bus consists of a single channel that carries bits. From this data resynchronization, information can be derived. The way in which this channel is implemented is not fixed in this specification (i.e., single wire (plus ground), two differential wires, optical fires, etc). Bus values The bus can have one of two complementary logical values; dominant or recessive. During simultaneous transmission of dominant and recessive bits, the resulting bus value will be dominant. For example, in case of a wired-AND implementation of the bus, the dominant level would be represented by a logical 0 and the recessive level by a logical 1. Physical states (e.g., electrical voltage, light) that represent the logical levels are not given in the specification. Acknowledgment All receivers check the consistency of the message being received and will acknowledge a consistent message and flag an inconsistent message. Sleep Mode; Wake-up To reduce the system's power consumption, a CAN device may be set into Sleep mode without any internal activity and with disconnected bus drivers. The Sleep mode is finished with a wake-up by any bus activity or by internal conditions of the system. On wake-up, the internal activity is restarted, although the MAC sub-layer will be waiting for the system's oscillator to stabilize and it will then wait until it has synchronized itself to the bus activity (by checking for eleven consecutive recessive bits), before the bus drivers are set to on-bus again.
Appendix
14.
15. 16.
17.
18.
19. 20.
DS70074C-page 26-17

B.6 B.6.1 Frame Types Standard Data Frame A data frame is generated by a node when the node wishes to transmit data. The Standard CAN Data Frame is shown in Figure B-2. In common with all other frames, the frame begins with a Start-Of-Frame bit (SOF dominant state) for hard synchronization of all nodes. The SOF is followed by the Arbitration field consisting of 12 bits, the 11-bit ldentifier (reflecting the contents and priority of the message) and the RTR bit (Remote Transmission Request bit). The RTR bit is used to distinguish a data frame (RTR dominant) from a remote frame. The next field is the Control field, consisting of 6 bits. The first bit of this field is called the lDE bit (Identifier Extension) and is at dominant state to specify that the frame is a standard frame. The following bit is reserved, RB0, and defined as a dominant bit. The remaining 4 bits of the Control field are the Data Length Code (DLC) and specify the number of bytes of data contained in the message. The data being sent follows in the Data field, which is of the length defined by the DLC above (1-8 bytes). The Cyclic Redundancy field (CRC) follows and is used to detect possible transmission errors. The CRC field consists of a 15-bit CRC sequence, completed by the recessive CRC Delimiter bit. The final field is the Acknowledge field. During the ACK Slot bit, the transmitting node sends out a recessive bit. Any node that has received an error free frame acknowledges the correct reception of the frame by sending back a dominant bit (regardless of whether the node is configured to accept that specific message or not). From this, it can be seen that CAN belongs to the in-bit-response group of protocols. The recessive Acknowledge Delimiter completes the Acknowledge slot and may not be overwritten by a dominant bit. B.7 Extended Data Frame In the Extended CAN Data frame, shown in Figure B-3, the Start-Of-Frame bit (SOF) is followed by the Arbitration field consisting of 38 bits. The first 11 bits are the 11 most significant bits of the 29-bit identifier (Base-lD). These 11 bits are followed by the Substitute Remote Request bit, SRR, which is transmitted as recessive. The SRR is followed by the lDE bit, which is recessive to denote that the frame is an extended CAN frame. It should be noted from this, that if arbitration remains unresolved after transmission of the first 11 bits of the identifier, and one of the nodes involved in arbitration is sending a standard CAN frame (11-bit identifier), then the standard CAN frame will win arbitration due to the assertion of a dominant lDE bit. Also, the SRR bit in an extended CAN frame must be recessive to allow the assertion of a dominant RTR bit by a node that is sending a standard CAN remote frame. The SRR and lDE bits are followed by the remaining 18 bits of the identifier (lD-Extension) and the Remote Transmission Request bit. To enable standard and extended frames to be sent across a shared network, it is necessary to split the 29-bit extended message identifier into 11-bit (most significant) and 18-bit (least significant) sections. This split ensures that the Identifier Extension bit (lDE) can remain at the same bit position in both standard and extended frames. The next field is the Control field, consisting of 6 bits. The first 2 bits of this field are reserved and are at dominant state. The remaining 4 bits of the Control field are the Data Length Code (DLC) and specify the number of data bytes. The remaining portion of the frame (Data field, CRC field, Acknowledge field, End-Of-Frame and lntermission) is constructed in the same way as for a standard data frame.
DS70074C-page 26-18

B.8 Remote Frame Normally data transmission is performed on an autonomous basis with the data source node (e.g., a sensor sending out a data frame). It is possible, however, for a destination node to request the data from the source. For this purpose, the destination node sends a remote frame with an identifier that matches the identifier of the required data frame. The appropriate data source node will then send a data frame as a response to this remote request. There are 2 differences between a remote frame and a data frame, shown in Figure B-4. First, the RTR bit is at the recessive state and second, there is no data field. In the very unlikely event of a data frame and a remote frame with the same identifier being transmitted at the same time, the data frame wins arbitration due to the dominant RTR bit following the identifier. In this way, the node that transmitted the remote frame receives the desired data immediately. B.9 Error Frame An error frame is generated by any node that detects a bus error. An error frame, shown in Figure B-5, consists of 2 fields, an error flag field followed by an error delimiter field. The error delimiter consists of 8 recessive bits and allows the bus nodes to restart bus communications cleanly after an error. There are two forms of error flag fields. The form of the error flag field depends an the error status of the node that detects the error. If an error-active node detects a bus error then the node interrupts transmission of the current message by generating an active error flag. The active error flag is composed of six consecutive dominant bits. This bit sequence actively violates the bit stuffing rule. All other stations recognize the resulting bit stuffing error and in turn generate error frames themselves, called error echo flags. The error flag field therefore consists of between six and twelve consecutive dominant bits (generated by one or more nodes). The error delimiter field completes the error frame. After completion of the error frame, bus activity returns to normal and the interrupted node attempts to resend the aborted message. If an error passive node detects a bus error, then the node transmits an error passive flag followed again by the error delimiter field. The error passive flag consists of six consecutive recessive bits, and therefore the error frame for an error passive node consists of 14 recessive bits. From this, it follows that, unless the bus error is detected by the bus master node that is actually transmitting, the transmission of an error frame by an error passive node will not affect any other node on the network. If the bus master node generates an error passive flag, then this may cause other nodes to generate error frames due to the resulting bit stuffing violation. After transmission of an error frame, an error passive node must wait for 6 consecutive recessive bits on the bus before attempting to rejoin bus communications. B.10 Interframe Space lnterframe space separates a proceeding frame (of whatever type) from a following data or remote frame. lnterframe space is composed of at least 3 recessive bits, called the intermission. This is provided to allow nodes time for internal processing before the start of the next message frame. After the intermission, the bus line remains in the recessive state (Bus idle) until the next transmission starts.
Appendix
DS70074C-page 26-19
Start-Of-Frame
12 Arbitration Field
Stored in Buffers
ID3
0 0 0
Reserved Bits Data Length Code
ID0 RTR IDE RB0 DLC3 DLC0
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90
DS70074C-page 26-20
Any Frame 3 INT 8 Suspend Transmit 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Inter-Frame Space bus Idle 1 1 1 0 0 Data Frame or Remote Frame Message Filtering 11 Identifier 6 Control Field 4 8 Stored in Transmit/Receive Buffers Data Frame (number of bits = 44 + 8N) Bit Stuffing 8N ( N 8) Data Field 8 0 0 0 0 0 0 0 0 15 CRC 16 CRC Field 1 7 End-OfFrame 1 1 1 1 1 1 1 1 Any Frame 3 INT 8 Suspend Transmit
Figure B-2:
Standard Data Frame

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 Inter-Frame Space bus Idle
Start-Of-Frame ID10
ID3
ID0 SRR IDE EID17
0 0 0
Bit Stuffing
Reserved bits Data Length Code
EID0 RTR RB1 RB0 DLC3 4 DLC0
Appendix
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90

1 1 bus Idle 1 1 1 0 0 Start-Of-Frame 11 Data Frame or Remote Frame Identifier Message Filtering 0 1 32 Stored in Buffers Arbitration Field 18 Extended Identifier 6 Control Field 8 8N ( N 8) Data Field Data Frame (number of bits = 64 + 8N) Stored in Transmit/Receive Buffers 8 15 16 CRC Field CRC 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 7 1 1 1 1 1 1 1 1 Any Frame End-OfFrame 3 INT 8 Suspend Transmit
Figure B-3:
Extended Data Format
DS70074C-page 26-21
26
Start-Of-Frame
1 0 0
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90
DS70074C-page 26-22
Any Frame 3 INT 8 Suspend Transmit 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Inter-Frame Space bus Idle 1 1 1 0 0 Data Frame or Remote Frame Message Filtering 11 12 Arbitration Field Identifier Stored in Buffers 6 Control Field 4 Bit Stuffing Remote Frame (number of bits = 44) 15 CRC 16 CRC Field 1 7 End-OfFrame 1 1 1 1 1 1 1 1 Any Frame 3 INT 8 Suspend Transmit
Figure B-4:
Remote Data Frame

12 Arbitration Field
ID3
0 0 0
Appendix
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90
90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90

Any Frame 3 INT 8 Suspend Transmit 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Inter-Frame Space bus Idle 1 1 1 0 0 Start-Of-Frame Data Frame or Remote Frame 11 Identifier Message Filtering 6 Control Field 4 Interrupted Data Frame Bit Stuffing 8 8N ( N 8) Data Field Data Frame or Remote Frame 8 6 Error Flag 0 0 0 0 0 0 0 6 Echo Error Flag Error Frame 8 Error Delimiter 0 0 1 1 1 1 1 1 1 1 0 Any Frame 3 INT 8 Inter-Frame Space Suspend Transmit
Figure B-5:
Error Frame
DS70074C-page 26-23
26

B.11 Referenced Documents Title Document Road Vehicles; Interchange of Digital Information, Controller Area Network Bosch CAN Specification Version 2.0 ISO11898
DS70074C-page 26-24

APPENDIX C: CODEC PROTOCOL OVERVIEW
This appendix summarizes audio coder/decoder (codec) protocols for Inter-IC Sound (I2S) and AC-Link Compliant mode interfaces. Many codecs intended for use in audio applications support sampling rates between 8 kHz and 48 kHz and typically use one of the interface protocols previously mentioned. The Data Converter Interface (DCI) module automatically handles the interface timing associated with these codecs. No overhead from the CPU is required until the requested amount of data has been transmitted and/or received by the DCI. Up to four data words may be transferred between CPU interrupts. C.1 I2S Protocol Description Inter-IC Sound (I2S) is a simple, three-wire bus interface used for the transfer of digital audio data between the following devices: DSP processors A/D and D/A converters Digital input/output interfaces This Appendix information is intended to supplement the I2S Protocol Specification, which is published by Philips, Inc. The I2S bus is a time division multiplexed and transfers two channels of data. These two data channels are typically the left and right channels of a digital audio stream. The I2S bus has the following connection pins: SCK: The I2S serial clock line SDx: The I2S serial data line (input or output) WS: The I2S word select line A timing diagram for a data transfer is shown in Figure C-2. Serial data is transmitted on the I2S bus in twos complement format with the MSb transmitted first. The MSb must be transferred first because the protocol allows different transmitter and receiver data word lengths. If a receiver is sent more bits than in can accept for a data word, the LSbits are ignored. If a receiver is sent fewer bits than its native word length, it must set the remaining LSbits to zero internally. The WS line indicates the data channel that is being transmitted. The following standard is used: WS = 0: Channel 1 or left audio channel WS = 1: Channel 2 or right audio channel The WS line is sampled by the receiver on the rising edge of SCK and the MSb of the next data word is transmitted one SCK period after WS changes. The one period delay after WS changes provides the receiver time to store the previously transmitted word and prepare for the next word. Serial data sent by the transmitter is placed on the bus on the falling edge of SCK and is latched by the receiver on the rising edge of SCK. Any device may act as the system master in an I2S system. The system master generates the SCK and WS signals. Typically, the transmitter is the system master, but the receiver or a third device may perform this function. Figure C-1 shows possible I2S bus configurations. Although it is not indicated in Figure C-1, the two connected devices may have both a data transmit and a data receive connection.
Appendix
DS70074C-page 26-25

Figure C-1: I2S Bus Connections
SCK I2S Transmitter WS SD I2S Receiver Transmitter master
SCK I S Transmitter
2
WS SD
I2S Receiver
Receiver master
I2S Controller
SCK I2S Transmitter WS SD I2S Receiver Separate controller as master
Figure C-2:
I2S Interface Timing Diagram
SCK
SD
MSB
LSB
MSB
LSB
WS
Note:
A 5 bit transfer is shown here for illustration purposes. The I2S protocol does not specify word length this is system dependent.
C.2
AC 97 Protocol The Audio Codec 97 (AC 97) specification defines a standard architecture and digital interface protocol for audio codecs used in PC platforms. The digital interface protocol for an AC 97 compliant codec is called AC-Link and is the focus of this discussion. The specific requirements and features of the AC 97 controller device are not described here. This Appendix information is intended to supplement the AC 97 Component Specification document, which is published by Intel, Corp.
C.3
AC-Link Signal Descriptions All AC-Link signals are derived from the AC 97 master clock source. The recommended clock source is a 24.576 MHz crystal connected to the AC 97 codec to minimize clock jitter. The 24.576 MHz clock may also be provided by the AC 97 controller or by an external source. All AC-Link signal names are referenced to the AC 97 controller, not the AC 97 codec. The controller is the device that generates the SYNC signal to initiate data transfers. Each signal is described in subsequent sections.
DS70074C-page 26-26

C.3.1 Bit Clock (BIT_CLK) A 12.288 MHz BIT_CLK signal is provided by the master AC 97 codec in a system. The BIT_CLK signal is an input to the AC 97 controller and up to three slave AC 97 codec devices in the system. All data on the AC-Link transitions on the rising edge of BIT_CLK and is sampled by the receiving device on the falling edge of BIT_CLK. C.3.2 Serial Data Output (SDO) SDO is a time division multiplexed data stream sent to the AC 97 codec C.3.3 Serial Data Input (SDI) SDI is the time division multiplexed data stream from the AC 97 codec. C.3.4 SYNC SYNC is a 48 kHz fixed rate sample synchronization signal that is supplied from the AC 97 controller to the AC 97 codec. The SYNC signal is derived by dividing the BIT_CLK signal by 256. The SYNC signal is high for 16 BIT_CLK periods and is low for 240 BIT_CLK periods. The SYNC signal only changes on the rising edge of BIT_CLK and its period defines the boundaries of one audio data frame. C.3.5 Reset The RESET signal is an input to each AC 97 codec in the system and resets the codec hardware. C.4 C.4.1 AC-Link Protocol AC-Link Serial Interface Protocol The AC-Link serial data stream uses a time division multiplexed (TDM) scheme with a 256-bit data frame. Each data frame is subdivided into 13 time slots, numbered Slot #0 Slot #12. Slot #0 is a special time slot that contains 16 bits. The remaining 12 slots are 20-bits wide. An example of an AC-Link frame is shown in Figure C-4. The frame begins with a rising edge of the SYNC signal which is coincident with the rising edge of BIT_CLK. The AC 97 codec samples the assertion of SYNC on the falling edge of BIT_CLK that immediately follows. This falling edge marks the time when both the codec and controller are aware of the start of a new frame. On the next rising edge of BIT_CLK, the codec asserts the MSb of SDATA_IN and the codec asserts the first edge of SDATA_OUT. This sequence ensures that data transitions and subsequent sample points for both incoming and outgoing data streams are time aligned. Slot #0, Slot #1 and Slot #2 have special use for status and control in the AC-Link protocol. The remaining time slots are assigned to certain types of digital audio data. The data assignment for Slot #3 Slot #12 is dependent on the AC 97 codec that is selected, so the slot usage is summarized briefly here. For more details on slot usage, refer to the AC 97 Component Specification.
Appendix
DS70074C-page 26-27

C.4.2 Slot #0, TAG Frame Slot #0 is commonly called the tag frame. The tag frame has a bit location for each data time slot in the AC-Link protocol. These bits are used to specify which time slots in a frame are valid for use by the controller. A 1 in a given bit position of Slot #0 indicates that the corresponding time slot within the current audio frame has been assigned to a data stream, and contains valid data. If a slot is tagged invalid, it is the responsibility of the source of the data, (AC 97 codec for the input stream, AC 97 controller for the output stream), to stuff all bit positions with 0s during the slots active time. There are also special bits in the tag frame. The MSb of the tag frame for SDATA_OUT is a Frame Valid Status bit. The Frame Valid bit serves as a global indicator to the codec that at least one time slot in the frame has valid data. If the entire frame is tagged invalid, the codec can ignore all subsequent slots in the frame. This feature is used to implement sample rates other than 48 kHz. The two LSbs of the SDATA_OUT tag frame indicate the codec address. Up to four AC 97 codecs may be connected in a system. If only one codec is used in a system, these bits remain 0s. The MSb of the SDATA_IN is used as a Codec Ready Status bit. If this bit location is a 0, then the codec is powered down and/or not ready for normal operation. If the Codec Ready bit is set, it is the responsibility of the controller to query the status registers in the codec to see which subsections are operable. C.4.3 Slot #1 (Command Address) and Slot #2 (Command Data) Slot #1 and Slot #2 also have special uses in the AC-Link protocol. These time slots are used for address and data values when reading or writing the AC 97 codec control registers. These time slots must be tagged as valid in Slot #0 in order to read and write the control registers. The AC 97 Component Specification allows for sixty-four (64) 16-bit control registers in the codec. Seven address bits are provided in the AC-Link protocol, but only even-numbered addresses are used. The odd numbered address values are reserved. Slot #1 and Slot #2 for the SDATA_OUT line are called the Command Address and Command Data, respectively. The Command Address slot on the SDATA_OUT line is used to specify the codec register address and to specify whether the register access will be a read or a write. The Command Data slot on SDATA_OUT contains the 16-bit value that will be written to one of the codec control registers. If a read of the codec registers is being performed, the Command Data bits are set to 0s. Slot #1 and Slot #2 for the SDATA_IN line are called the Status Address and Status Data slots, respectively. The Status Address time slot echos the register address that was previously sent to the codec. If this value is 0, an invalid address was previously sent to the codec. The Status Address time slot also has ten Slot Request bits. The Slot Request bits can be manipulated by the codec for applications with variable sample rates. The Status Data time slot returns 16-bit data read from the codec control/status registers. C.4.4 Slot #3 (PCM Left Channel) Slot #3 in the SDATA_OUT signal is used for the composite digital audio left playback stream. For soundcard applications, this is typically the combined .WAV audio and MIDI synthesizer output. Slot #3 in the SDATA_IN signal is the left channel record data taken from the AC 97 codec input mixer. C.4.5 Slot #4 (PCM Right Channel) Slot #4 in the SDATA_OUT signal is used for the composite digital audio right playback stream. For soundcard applications, this is typically the combined .WAV audio and MIDI synthesizer output. Slot #4 in the SDATA_IN signal is the right channel record data taken from the AC 97 codec input mixer.
DS70074C-page 26-28

C.4.6 Slot #5 (Modem Line 1) Slot #5 in the SDATA_OUT signal is used for modem DAC data. The default resolution for modem compatible AC 97 codecs is 16 bits. As in all time slots, the unused bits in the slot are set to 0. Slot #5 in the SDATA_IN signal is used for the modem ADC data. C.4.7 Slot #6 Slot #6 in the SDATA_OUT signal is used for PCM Center Channel DAC data in 4 or 6-channel sound configurations. Slot #6 in the SDATA_IN signal is used for dedicated microphone record data. The data in this slot allows echo cancellation algorithms to be used for speakerphone applications. C.4.8 Slot #7 Slot #7 in the SDATA_OUT signal is used for PCM Left Channel DAC data in 4 or 6-channel sound configurations. Slot #7 in the SDATA_IN signal is reserved for future use in the AC 97 Component Specification. C.4.9 Slot #8 Slot #8 in the SDATA_OUT signal is used for PCM Right Channel DAC data in 4 or 6-channel sound configurations. Slot #8 in the SDATA_IN signal is reserved for future use in the AC 97 Component Specification. C.4.10 Slot #9 Slot #9 in the SDATA_OUT signal is used for PCM LFE DAC data in 6-channel sound configurations. Slot #9 in the SDATA_IN signal is reserved for future use in the AC 97 Component Specification. C.4.11 Slot #10 (Modem Line 2) Slot #10 is used for the modem line 2 ADC and DAC data in modem compatible devices. C.4.12 Slot #11 (Modem Handset) Slot #11 is used for the modem handset ADC and DAC data in modem compatible devices. C.4.13 Slot #12 (GPIO Control/Status) The bits in Slot #12 are used for reading and writing GPIO pins in the AC 97 codec. The GPIO pins are provided for modem control functions on modem compatible devices. Figure C-3: AC-Link Signal Connections
BIT_CLK SYNC 24.576 MHz
Appendix
AC 97 Controller
SDATA_OUT SDATA_IN
AC 97 Codec
/RESET
DS70074C-page 26-29

Figure C-4: AC-Link Data Frame
256 16 20 20 20 20 20 20 20
SYNC
SDATA_OUT
Tag Frame
Command Address
Command Data
Slot3 Left PCM Data
Slot 10 Line 2 DAC
Slot 11 Handset DAC
Slot 12 Codec I/O Control
SDATA_IN
Tag Frame
Status Address
Status Data
Slot3 Left PCM Data
Slot 10 Line 2 ADC
Slot 11 Handset ADC
Slot 12 Codec I/O Status
Figure C-5:
SDATA_IN Bit Locations for SLOT#0, SLOT#1, SLOT#2
SLOT#0: TAG Slot

bit 15 Codec Slot 1 Slot 2 Slot 3 Slot 4 Slot 5 Slot 6 Slot 7 Slot 8 Slot 9 Slot 10 Slot 11 Slot 12 ___ Ready Valid Valid Valid Valid Valid Valid Valid Valid Valid Valid Valid Valid ___ bit 0 ___
SLOT#1: Control Address Echo And SLOTREQ Bits

bit 19 bit 18 Resvd 0 Control Register Index Echo (Set to 0siftaggedinvalidbyAC'97codec.) bit 12 bit 11 bit 2 bit 0 Reserved (set to 0)
Slot 3 Slot 4 Slot 5 Slot 6 Slot 7 Slot 8 Slot 9 Slot 10 Slot 11 Slot 12
bits 11-2: On demand data request flags 0 = send data, 1 = do NOTsend data
SLOT#2: Codec Control Register Read Data

bit 19 Control Register 16-bit Read Data (Set to 0siftaggedinvalidbyAC'97codec.) bit 4 Reserved (set to 0) bit 0
DS70074C-page 26-30

Figure C-6: SDATA_OUT Bit Locations for SLOT#0, SLOT#1, SLOT#2
SLOT#0: TAG Slot

bit 15 Valid Slot 1 Slot 2 Slot 3 Slot 4 Slot 5 Slot 6 Slot 7 Slot 8 Slot 9 Slot 10 Slot 11 Slot 12 ___ Frame Valid Valid Valid Valid Valid Valid Valid Valid Valid Valid Valid Valid bit 0 Codec Codec ID ID
Appendix
SLOT#1: Command Address

bit 19 bit 18 R/W bit Control Register Index bit 12 bit 11 Reserved (set to 0) bit 0
SLOT#2: Command Data

bit 19 Control Register 16-bit Write Data (Set to 0sifcurrentlyperformingareadoperation.)' bit 4 Reserved (set to 0) bit 0
DS70074C-page 26-31

NOTES:
DS70074C-page 26-32

A
A/D Accuracy/Error ............................................17-47, 18-28 ADCON0 Register...........................................17-4, 18-4 Configuring Analog Port Pins......................17-14, 18-12 Edge Detection Mode .............................................. 13-8 Effects of a Reset........................................16-19, 17-55 Effects of a Reset (12-bit) ...................................... 18-33 Enabling the Module .............................................. 17-16 Enabling the Module (12-bit).................................. 18-14 How to Start Sampling ........................................... 17-17 How to Stop Sampling and Start Conversions....... 17-18 Reading the A/D Result Buffer............................... 17-46 Sampling Requirements..............................17-45, 18-26 Transfer Function........................................17-47, 18-28 Transfer Function (12-bit) ...................................... 18-28 A/D Accuracy/Error ........................................................ 17-47 A/D Accuracy/Error (12-bit) ............................................ 18-28 A/D Conversion Speeds......................................17-49, 18-30 A/D Converter External Conversion Request .................. 6-10 A/D Module Configuration .............................................. 17-13 A/D Module Configuration (12-bit) ................................. 18-11 A/D Sampling Requirements (12-bit) ............................. 18-26 A/D Special Event Trigger.............................................. 12-22 A/D Terminology and Conversion Sequence ................. 17-11 A/D Terminology and Conversion Sequence (12-bit) .... 18-10 Accumulator Write Back ................................................. 2-25 ACK................................................................................ 21-38 Acknowledge Pulse........................................................ 21-38 Address Generator Units and DSP Class Instructions....... 3-6 Address Register Dependencies ..................................... 2-35 AKS ................................................................................ 21-17 Alternate Vector Table ....................................................... 6-2 Arithmetic Logic Unit (ALU).............................................. 2-17 UART Transmitter.................................................. 19-11 WDT ........................................................................ 10-6 Brown-out Reset (BOR)..................................................... 8-8 Configuration ............................................................. 8-9 Current Consumption for Operation .......................... 8-9 Byte to Word Conversion................................................. 2-17
Index
C
CAN Buffer Reception and Overflow Truth Table .......... 23-44 Message Acceptance Filters ................................. 23-12 CAN Library ..................................................................... 25-9 Capture Buffer Operation................................................. 13-8 Clock Switching Aborting ................................................................... 7-36 Enable ..................................................................... 7-34 Entering Sleep Mode During ................................... 7-36 Operation................................................................. 7-34 Recommended Code Sequence ............................. 7-36 Tips.......................................................................... 7-36 CN Change Notification Pins ......................................... 11-7 Configuration and Operation ................................... 11-8 Control Registers..................................................... 11-7 Operation in Sleep and Idle Modes ......................... 11-8 Code Examples Clock Switching ....................................................... 7-37 Compare Mode Toggle Mode Pin State Setup........ 14-8 Compare Mode Toggle Setup and Interrupt Servicing.......................................................... 14-8 Configuration Register Write ................................... 5-15 Continuous Output Pulse Setup and Interrupt Servicing........................................................ 14-16 Initialization Code for 16-bit Asynchronous Counter Mode Using an External Clock Input ............. 12-11 Initialization Code for 16-bit Gated Time Accumulation Mode ............................................................. 12-13 Initialization Code for 16-bit Synchronous Counter Mode Using an External Clock Input ....................... 12-10 Initialization Code for 16-bit Timer Using System Clock................................................... 12-9 Initialization Code for 32-bit Gated Time Accumulation Mode ....................................... 12-20 Initialization Code for 32-bit Synchronous Counter Mode Using an External Clock Input ....................... 12-19 Initialization Code for 32-bit Timer Using Instruction Cycle as Input Clock...................................... 12-18 Prescaled Capture................................................... 13-7 PWM Mode Pulse Setup and Interrupt Servicing........................................................ 14-22 Reading from a 32-bit Timer.................................. 12-21 Single Output Pulse Setup and Interrupt Servicing........................................................ 14-12 Single Row Programming........................................ 5-14 8-bit Transmit/Receive (UART1) ........................... 19-20 9-bit Transmit/Receive (UART1), Address Detect Enabled ......................................................... 19-20 CODEC Interface Basics and Terminology ..................... 22-8 Complementary PWM Output Mode.............................. 15-26 Configuration Bit Descriptions ......................................... 24-6 BOR and POR ......................................................... 24-6 General Code Segment ........................................... 24-6 Motor Control PWM Module .................................... 24-6 Oscillator.................................................................. 24-6 Connection Considerations............................................ 17-47 Connectivity Development Board .................................. 25-14 Control Register Descriptions .......................................... 3-18
B
Barrel Shifter .................................................................... 2-26 Baud Rate Generator (BRG)...................................................... 19-8 Tables ...................................................................... 19-9 BF .......................................................................21-18, 21-19 Bit-Reversed Addressing ................................................. 3-14 and Modulo Addressing ........................................... 3-15 Code Example ......................................................... 3-18 Intro.......................................................................... 3-14 Modifier Value .......................................................... 3-16 Operation ................................................................. 3-15 Block Diagrams Dedicated Port Structure.......................................... 11-2 DSP Engine ............................................................. 2-19 dsPIC30F CPU Core.................................................. 2-3 External Power-on Reset Circuit (For Slow VDD Rise Time) ......................................................... 8-7 Input Capture ........................................................... 13-2 Input Change Notification......................................... 11-7 Low Voltage Detect (LVD) ......................................... 9-3 Oscillator System ....................................................... 7-4 Output Compare Module.......................................... 14-2 Reset System............................................................. 8-2 Shared Port Structure .............................................. 11-4 Type A Timer ........................................................... 12-3 Type B - Type C Timer Pair (32-bit Timer)............. 12-17 Type B Timer ........................................................... 12-4 Type C Timer ........................................................... 12-5 UART ....................................................................... 19-2 UART Receiver ...................................................... 19-16
DS70046E-page 1
Index
Control Registers .......................................... 12-6, 17-4, 18-4 Assignment of Interrupts .......................................... 6-14 Controlling Sample/Conversion Operation..................... 17-29 Controlling Sample/Conversion Operation (12-bit) ........ 18-19 Conversion Sequence Examples ................................... 17-31 Conversion Sequence Examples (12-bit).......................18-21 CPU Register Maps .......................................................... 2-38 Related Application Notes........................................ 2-40 Revision History ....................................................... 2-41 CPU Clocking Scheme....................................................... 7-5 CPU Priority Status ............................................................ 6-5 CPU Register Descriptions ..............................................2-11 Crystal Oscillators, Ceramic Resonators .........................7-22
E
Equations Calculating the PWM Period.................................. 14-19 Calculation for Maximum PWM Resolution ........... 14-20 Modulo End Address for Incrementing Buffer............ 3-9 Modulo Start Address for Decrementing Buffer ......... 3-9 WDT Time-out Period .............................................. 10-7 External Clock Input......................................................... 7-25 External Interrupt Support................................................ 6-10 External RC Oscillator ..................................................... 7-26 Operating Frequency ............................................... 7-27 Start-up .................................................................... 7-27 with I/O Enabled ...................................................... 7-27 External Reset (EXTR) ...................................................... 8-7
D
Data Accumulator Status Bits ................................................................ 2-23 Data Accumulator Adder/Subtractor ................................ 2-23 Data Accumulators........................................................... 2-20 Data Alignment................................................................... 3-7 Data EEPROM Programming .......................................... 5-15 Erasing One Row ..................................................... 5-19 Erasing One Word ................................................... 5-17 Reading Memory...................................................... 5-21 Row Algorithm.......................................................... 5-16 Single Word Algorithm ............................................. 5-16 Write One Row......................................................... 5-20 Writing One Word .................................................... 5-18 Data Memory Map ............................................................................ 3-3 Near ........................................................................... 3-4 Data Space Address Generator Units (AGUs)............................................. 3-5 X Address Generator Unit .......................................... 3-5 Y Address Generator Unit .......................................... 3-5 Data Space Write Saturation............................................ 2-24 DCI Buffer Control Unit ................................................. 22-13 Control Register Descriptions .................................. 22-2 Operation ............................................................... 22-10 Using the Module ................................................... 22-17 Dead Time Control ......................................................... 15-27 Determining Best Values for Crystals, Clock Mode, C1, C2, and Rs ........................................................ 7-24 Device Configuration Register Descriptions ............................................... 24-2 Device Identification Registers......................................... 24-7 Device ID (DEVID) ................................................... 24-7 Unit ID Field ............................................................. 24-7 Device Reset Times ......................................................... 8-11 Device Start-up Time Lines..............................................8-13 Device Wake-up on Sleep/Idle....................................... 13-10 Disable Interrupts Instruction ............................................. 6-8 Divide Support.................................................................. 2-27 DSP Algorithm Library...................................................... 25-7 DSP Engine...................................................................... 2-18 DSP Engine Mode Selection............................................ 2-26 DSP Engine Trap Events ................................................. 2-26 DSP Filter Design Software Utility.................................... 25-8 dsPIC Language Suite ..................................................... 25-3 dsPIC30F Hardware Development Boards .................... 25-11
F
Fail-Safe Clock Monitor (FSCM)...................................... 7-32 and Slow Oscillator Start-up .................................... 7-33 and WDT ................................................................. 7-33 Delay ....................................................................... 7-32 Flash and Data EEPROM Programming Control Registers ....................................................... 5-5 NVMADR ........................................................... 5-6 NVMCON........................................................... 5-5 NVMKEY ........................................................... 5-6 Flash Program Memory Erasing a Row ......................................................... 5-12 Loading Write Latches ............................................. 5-13 Programming Algorithm ........................................... 5-11 FSCM and Device Resets................................................... 8-12 Delay for Crystal and PLL Clock Sources................ 8-12
G
General Purpose Development Board........................... 25-12
H
Hard Traps......................................................................... 6-7 Address Error (Level 13)............................................ 6-8 Oscillator Failure (Level 14)....................................... 6-8 Priority and Conflicts.................................................. 6-7 Stack Error (Level 12)................................................ 6-6 How to Start Sampling (12-bit)....................................... 18-14 How to Start Sampling and Start Conversions (12-bit) .............................................. 18-14
I
I/O Multiplexing with Multiple Peripherals ........................ 11-4 I/O Pin Control ..............................12-22, 13-10, 14-23, 16-18 I/O Port Control Registers................................................ 11-3 I/O Ports Related Application Notes ..................................... 11-11 Revision History..................................................... 11-12 Idle Mode ......................................................................... 10-4 Time Delays on Wake-up from ................................ 10-5 Wake-up from on Interrupt....................................... 10-5 Wake-up from on Reset........................................... 10-5 Wake-up from on WDT Time-out............................. 10-5 Independent PWM Output Mode ................................... 15-30 Initialization .................................................................... 17-48 Initialization (12-bit)........................................................ 18-29
DS70046E-page 2

Input Capture Associated Special Function Registers.................. 13-11 Buffer Not Empty (ICBNE) ....................................... 13-9 Design Tips ............................................................ 13-12 Interrupts.................................................................. 13-9 Control Bits ...................................................... 13-9 Operation in Power Saving States ......................... 13-10 Overflow (ICOV)....................................................... 13-9 Related Application Notes...................................... 13-13 Input Capture Event Modes ............................................. 13-4 Input Capture Registers ................................................... 13-3 Instruction Flow Types ..................................................... 2-27 Instruction Stall Cycles..................................................... 2-36 Internal Fast RC Oscillator (FRC) .................................... 7-31 Internal Low Power RC (LPRC) Oscillator ....................... 7-32 Enabling ................................................................... 7-32 Internal Voltage Reference ................................................ 9-3 Interrupt Control and Status Registers............................. 6-14 CORCON ................................................................. 6-14 IECx ......................................................................... 6-14 IFSx.......................................................................... 6-14 INTCON1, INTCON2 ............................................... 6-14 IPCx ......................................................................... 6-14 SR ............................................................................ 6-14 Interrupt Controller Associated Special Function Registers.................... 6-43 Interrupt Latency One-Cycle Instructions............................................. 6-11 Two-Cycle Instructions............................................. 6-12 Interrupt Operation ............................................................. 6-9 Nesting..................................................................... 6-10 Return From Interrupt ................................................ 6-9 Interrupt Priority ................................................................. 6-5 Interrupt Processing Timing ............................................. 6-11 Interrupt Setup Procedures .............................................. 6-42 Initialization .............................................................. 6-42 Interrupt Disable....................................................... 6-42 Interrupt Service Routine ......................................... 6-42 Trap Service Routine ............................................... 6-42 Interrupt Vector Table ........................................................ 6-2 Interrupts Design Tips .............................................................. 6-44 Related Application Notes........................................ 6-45 Revision History ....................................................... 6-46 Interrupts Coincident with Power Save Instructions......... 10-5 Introduction Revision History ......................................................... 1-7 IWCOL ........................................................................... 21-22 I2C Acknowledge Generation....................................... 21-21 Building Complete Master Messages .................... 21-24 Bus Arbitration and Bus Collision........................... 21-30 Bus Collision During a Repeated Start Condition........................................................ 21-31 Bus Collision During a Start Condition................... 21-31 Bus Collision During a Stop Condition ................... 21-31 Bus Collision During Message Bit Transmission ................................................. 21-31 Bus Connection Considerations............................. 21-47 Communicating as a Master in a Multi-Master Environment................................................... 21-29 Communicating as a Slave .................................... 21-32 Detecting Bus Collisions and Resending Messages ...................................................... 21-30 Detecting Start and Stop Conditions...................... 21-32 Detecting the Address............................................ 21-32 Enabling I/O ........................................................... 21-13 Enabling Operation................................................ 21-13 Generating Repeated Start Bus Event .................. 21-23 Generating Start Bus Event ................................... 21-16 Generating Stop Bus Event ................................... 21-22 Initiating and Terminating Data Transfer ................. 26-3 Interrupts ............................................................... 21-13 Master Message Protocol States........................... 21-24 Module Operation during PWRSAV Instruction..... 21-49 Receiving Data from a Master Device ................... 21-37 Receiving Data from a Slave Device ..................... 21-19 Sending Data to a Master Device.......................... 21-44 Sending Data to a Slave Device............................ 21-17 Start ......................................................................... 26-3 Stop ......................................................................... 26-3 I2C Module Multi-master Mode................................................. 21-29 Register Map ......................................................... 21-33 10-bit Address Mode ............................................. 21-35
Index
L
LAT (I/O Latch) Registers ................................................ 11-3 Loop Constructs .............................................................. 2-30 DO ........................................................................... 2-32 REPEAT .................................................................. 2-30 Low Power 32 kHz Crystal Oscillator............................... 7-31 Low Power 32 kHz Crystal Oscillator Input.................... 12-15 LP Oscillator Continuous Operation.............................................. 7-31 Enable ..................................................................... 7-31 Intermittent Operation.............................................. 7-31 Operation with Timer1 ............................................. 7-31 LVD Control Bits ................................................................ 9-3 Current Consumption for Operation .......................... 9-5 Design Tips................................................................ 9-6 Initialization Steps...................................................... 9-5 Operation................................................................... 9-5 Operation During Sleep and Idle Mode ..................... 9-5 Related Application Notes ......................................... 9-7 Trip Point Selection ................................................... 9-3
M
Math Library..................................................................... 25-7 Microchip Hardware and Language Tools ....................... 25-2 Modes of Operation ......................................................... 14-4 Compare Mode Output Driven High ........................ 14-5 Compare Mode Output Driven Low ......................... 14-6 Compare Mode Toggle Output ................................ 14-7 Dual Compare Match............................................... 14-9 Dual Compare, Continuous Output Pulses............ 14-14 Dual Compare, Generating Continuous Output Pulses Special Cases (table) .................................... 14-17 Dual Compare, Single Output Pulse........................ 14-9 Special Cases (table) .................................... 14-13 Single Compare Match ............................................ 14-4 Modulo Addressing ............................................................ 3-7 Applicability.............................................................. 3-11 Calculation................................................................. 3-9 Initialization for Decrementing Buffer....................... 3-13 Initialization for Incrementing Modulo Buffer ........... 3-12 Start and End Address Selection............................... 3-8 W Address Register Selection................................... 3-9
DS70046E-page 3
Index
Modulo Start and End Address Selection XMODEND Register ..................................................3-8 XMODSRT Register................................................... 3-8 YMODEND Register ..................................................3-8 YMODSRT Register................................................... 3-8 Motor Control Development Board................................. 25-13 MPLAB ICD 2 In-Circuit Debugger................................... 25-5 MPLAB ICE 4000 In-Circuit Emulator .............................. 25-4 MPLAB SIM Software Simulator ...................................... 25-3 MPLAB 6.XX Integrated Development Environment Software ...................................................................25-2 Multi-Master Mode ......................................................... 21-29 Multiplier........................................................................... 2-20 Multiply Instructions.......................................................... 2-22 Loss of Lock During Normal Device Operation........ 7-30 POR and Long Oscillator Start-up Times ........................ 8-12 Port (I/O Port) Registers .................................................. 11-3 Power Saving Modes....................................................... 10-2 Power-on Reset (POR)...................................................... 8-5 Using ......................................................................... 8-7 Power-up Timer (PWRT) ................................................... 8-7 Prescaler Capture Events................................................ 13-6 Primary Oscillator ............................................................ 7-20 PRO MATE II Universal Device Programmer .................. 25-5 Program Memory Address Map ............................................................. 4-2 Counter ...................................................................... 4-4 Data Access From ..................................................... 4-4 Data Storage ............................................................. 4-7 High Word Access ..................................................... 4-7 Low Word Access ...................................................... 4-6 Program Space Visibility from Data Space................ 4-8 Related Application Notes ....................................... 4-12 Table Address Generation......................................... 4-6 Table Instruction Summary........................................ 4-5 Writes ............................................................. 4-10, 4-11 Programmable Digital Noise Filters ................................. 16-9 Programmable Oscillator Postscaler ............................... 7-33 Programmers Model .................................................. 2-3, 2-4 Register Description .................................................. 2-4 Protection Against Accidental Writes to OSCCON .......... 7-13 PSV Configuration ............................................................. 4-8 PSV Mapping with X and Y Data Spaces .................. 4-8 Timing ...................................................................... 4-10 Instruction Stalls .............................................. 4-10 Using PSV in a Repeat Loop........................... 4-10 Pulse Width Modulation Mode ....................................... 14-18 Duty Cycle ............................................................. 14-20 Period .................................................................... 14-19 With Fault Protection Input Pin .............................. 14-19 PWM Duty Cycle Comparison Units .............................. 15-20 PWM Fault Pins ............................................................. 15-34 PWM Output and Polarity Control.................................. 15-34 PWM Output Override ................................................... 15-31 PWM Special Event Trigger........................................... 15-38 PWM Time Base............................................................ 15-16 PWM Update Lockout .................................................... 15-37
N
Non-Maskable Traps.......................................................... 6-6 Address Error ............................................................. 6-6 Arithmetic Error .......................................................... 6-6 Oscillator Failure ........................................................ 6-6 Stack Error ................................................................. 6-6
O
Oscillator Configuration.............................................................. 7-6 Clock Switching Mode Configuration Bits ..........7-6 Design Tips .............................................................. 7-38 Related Application Notes........................................ 7-39 Resonator Start-up................................................... 7-22 Revision History ....................................................... 7-40 System Features Summary ....................................... 7-2 Oscillator Control Register (OSCCON) ............................ 7-13 Oscillator Mode Selection Guidelines .............................. 7-21 Oscillator Start-up From Sleep Mode............................... 7-23 Oscillator Start-up Timer (OST) ....................................... 7-31 Oscillator Switching Sequence......................................... 7-35 OSEK Operating Systems..............................................25-10 Other dsPIC30F CPU Control Registers.......................... 2-16 DISICNT...................................................................2-16 MODCON................................................................. 2-16 PSVPAG .................................................................. 2-16 TBLPAG ...................................................................2-16 XBREV .....................................................................2-16 XMODSRT, XMODEND........................................... 2-16 YMODSRT, YMODEND........................................... 2-16 Output Compare Associated Register Map ....................................... 14-24 Design Tips ............................................................ 14-26 Related Application Notes...................................... 14-27 Revision History ..................................................... 14-28 Output Compare Operating in Power Saving States......14-23 Output Compare Operation in Power Saving States Idle Mode ............................................................... 14-23 Sleep Mode ............................................................ 14-23 Output Compare Registers ..............................................14-3
Q
QEI Operation During Power Saving Modes ................. 16-19 Quadrature Decoder ...................................................... 16-10 Quadrature Encoder Interface Interrupts ....................... 16-17
R
R/W Bit.................................................................. 21-35, 26-4 Read-After-Write Dependency Rules .............................. 2-36 Reading A/D Result Buffer (12-bit) ................................ 18-27 Reading and Writing into 32-bit Timers ......................... 12-21 Reading and Writing 16-bit Timer Module Registers ............................................................... 12-15 Real-Time Operating System (RTOS) ............................. 25-9
P
Peripheral Driver Library ..................................................25-8 Peripheral Module Disable (PMD) Registers ................... 10-9 Peripheral Multiplexing..................................................... 11-4 Peripherals Using Timer Modules .................................. 12-22 Phase Locked Loop (PLL)................................................7-30 Frequency Range .................................................... 7-30 Lock Status .............................................................. 7-30 Loss of Lock During a Power-on Reset ................... 7-30 Loss of Lock During Clock Switching ....................... 7-30
DS70046E-page 4

Registers A/D Input Scan Select (ADCSSL - 12-bit)................ 18-9 A/D Input Scan Select (ADCSSL) .......................... 17-10 ADCHS (A/D Input Select) Register................17-9, 18-8 ADCHS A/D Input Select ......................................... 17-9 ADCHS A/D Input Select (12-bit) ............................. 18-8 ADCON1 (A/D Control) Register1........17-5, 18-5, 21-11 ADCON1 A/D Control 1 ..................................17-5, 17-6 ADCON1 A/D Control 1 (12-bit) ............................... 18-5 ADCON2 (A/D Control) Register2...................17-7, 18-6 ADCON2 A/D Control 2 ........................................... 17-7 ADCON2 A/D Control 2 (12-bit) ............................... 18-6 ADCON3 (A/D Control) Register3...................17-8, 18-7 ADCON3 A/D Control 3 ........................................... 17-8 ADCON3 A/D Control 3 (12-bit) ............................... 18-7 ADPCFG (A/D Port Configuration) Register ................................................17-10, 18-9 ADPCFG A/D Port Configuration ........................... 17-10 ADPCFG A/D Port Configuration (12-bit)................. 18-9 CiCFG1 (Baud Rate Configuration Register)......... 23-16 CiCFG2 (Baud Rate Configuration Register 2)...... 23-17 CiCTRL (CAN Module Control and Status Register) .......................................................... 23-3 CiEC (Transmit/Receive Error Count).................... 23-18 CiINTE (Interrupt Enable Register) ........................ 23-19 CiINTF (Interrupt Flag Register) ............................ 23-20 CiRXFnEIDH (Acceptance Filter n Extended Identifier High) ............................................... 23-12 CiRXFnEIDL (Acceptance Filter n Extended Identifier Low) ................................................ 23-13 CiRXMnEIDH (Acceptance Filter Mask n Extended Identifier High) ............................................... 23-14 CiRXMnEIDL (Acceptance Filter Mask n Extended Identifier Low) ................................................ 23-15 CiRXMnSID (Acceptance Filter Mask n Standard Identifier) ........................................................ 23-14 CiRXnBm (Receive Buffer n Data Field Word m) ......................................................... 23-11 CiRXnDLC (Receive Buffer n Data Length Control) .......................................................... 23-11 CiRXnEID (Receive Buffer n Extended Identifier) ........................................................ 23-10 CiRXnSID (Receive Buffer n Standard Identifier) ........................................................ 23-10 CiRX0CON (Receive Buffer 0 Status and Control Register) .......................................................... 23-8 CiRX1CON (Receive Buffer 1 Status and Control Register) .......................................................... 23-9 CiRX1FnSID (Acceptance Filter n Standard Identifier) ........................................................ 23-12 CiTXnBm (Transmit Buffer n Data Field Word m) ........................................................... 23-7 CiTXnCON (Transmit Buffer Status and Control Register) .......................................................... 23-5 CiTXnDLC (Transmit Buffer n Data Length Control) ............................................................ 23-7 CiTXnEID (Transmit Buffer n Extended Identifier) ............ 23-6 CiTXnSID (Transmit Buffer n Standard Identifier) .......................................................... 23-6 CNEN1 (Input Change Notification Interrupt Enable 1) ......................................................... 11-9 CNEN2 (Input Change Notification Interrupt Enable 2) ......................................................... 11-9 CNPU1 (Input Change Notification Pull-up Enable 1) ....................................................... 11-10 CNPU2 (Input Change Notification Pull-up Enable 2) ....................................................... 11-10 Control and Status................................................... 16-4 Control Registers..................................................... 15-4 CORCON (Core Control) ................................ 2-14, 6-15 DCICON1 ................................................................ 22-3 DCICON2 ................................................................ 22-4 DCICON3 ................................................................ 22-5 DCISTAT ................................................................. 22-6 DFLTCON Digital Filter Control ...................... 16-7, 16-8 DTCON1 Dead Time Control 1................................ 15-9 DTCON2 Dead Time Control 2.............................. 15-10 FBORPOR (BOR and POR Configuration Register) .......................................................... 24-4 FBORPOR BOR and POR Device Configuration ................................................. 15-15 FGS (General Code Segment Configuration Register) .......................................................... 24-5 FLTACON Fault A Control..................................... 15-11 FLTBCON Fault B Control..................................... 15-12 FOSC (Oscillator Configuration Register) ............................................ 7-7, 7-9, 7-11 FWDT (Watchdog Timer Configuration Register) .......................................................... 24-3 ICxCON (Input Capture x Control)........................... 13-3 IEC0 (Interrupt Enable Control 0) ............................ 6-24 IEC1 (Interrupt Enable Control 1) ............................ 6-26 IEC2 (Interrupt Enable Control 2) ................... 6-28, 6-29 IFS0 (Interrupt Flag Status 0) .................................. 6-18 IFS1 (Interrupt Flag Status 1) .................................. 6-20 IFS2 (Interrupt Flag Status 2) ......................... 6-22, 6-23 INTCON1 (Interrupt Control 1) ................................ 6-16 INTCON2 (Interrupt Control 2) ................................ 6-17 IPC0 (Interrupt Priority Control 0) ............................ 6-30 IPC1 (Interrupt Priority Control 1) ............................ 6-31 IPC10 (Interrupt Priority Control 10) ........................ 6-40 IPC11 (Interrupt Priority Control 11) ........................ 6-41 IPC2 (Interrupt Priority Control 2) ............................ 6-32 IPC3 (Interrupt Priority Control 3) ............................ 6-33 IPC4 (Interrupt Priority Control 4) ............................ 6-34 IPC5 (Interrupt Priority Control 5) ............................ 6-35 IPC6 (Interrupt Priority Control 6) ............................ 6-36 IPC7 (Interrupt Priority Control 7) ............................ 6-37 IPC8 (Interrupt Priority Control 8) ............................ 6-38 IPC9 (Interrupt Priority Control 9) ............................ 6-39 I2CSTAT (I2C Status) Register.......... 21-9, 21-10, 21-11 MODCON (Modulo and Bit-Reversed Addressing Control)............................................................ 3-19 NVMADR (Non-Volatile Memory Address) ................ 5-8 NVMADRU (Non-Volatile Memory Address) ............. 5-9 NVMCON (Non-Volatile Memory Control) ................. 5-7 NVMKEY (Non-Volatile Memory Key) ....................... 5-9 OCxCON (Output Compare x Control) .................... 14-3 OVDCON Override Control ................................... 15-13 PDC1 PWM Duty Cycle 1...................................... 15-13 PDC2 PWM Duty Cycle 2...................................... 15-14 PDC3 PWM Duty Cycle 3...................................... 15-14 PDC4 PWM Duty Cycle 4...................................... 15-15 PTCON PWM Time Base Control ........................... 15-5 PTMR PWM Time Base .......................................... 15-6 PTPER PWM Time Base Period ............................. 15-6 PWMCON1 PWM Control 1 .................................... 15-7 PWMCON2 PWM Control 2 .................................... 15-8 QEI Special Function............................................. 16-20 QEICON QEI Control...................................... 16-5, 16-6 RCON (Reset Control)........................................ 8-3, 9-4 RSCON.................................................................... 22-7
Index
DS70046E-page 5
Index
SEVTCMP Special Event Compare .........................15-7 SR (CPU Status) ...................................................... 2-12 SR (Status in CPU) ..................................................6-15 TSCON .................................................................... 22-7 TxCON (Timer Control for Type A Time Base) ........12-6 TxCON (Timer Control for Type B Time Base) ........12-7 TxCON (Timer Control for Type C Time Base) ........12-8 UxBRG (UARTx Baud Rate) .................................... 19-7 UxMODE (UARTx Mode) ......................................... 19-3 UxRXREG (UARTx Receive) ................................... 19-6 UxSTA (UARTx Status and Control) ........................ 19-4 UxTXREG (UARTx Transmit - Write Only) .............. 19-6 XBREV (X Write AGU Bit-Reversal Addressing Control) ............................................................ 3-22 XMODEND (X AGU Modulo Addressing End) ......... 3-20 XMODSRT (X AGU Modulo Addressing Start) ........3-20 YMODEND (Y AGU Modulo Addressing End) ......... 3-21 YMODSRT (Y AGU Modulo Addressing Start) ........3-21 10-bit A/D Converter Special Function................... 17-56 12-bit A/D Converter Special Function................... 18-34 6-Output PWM Module .......................................... 15-41 8-Output PWM Module .......................................... 15-40 Reset Design Tips .............................................................. 8-17 Illegal Opcode ............................................................ 8-9 Trap Conflict............................................................... 8-9 Uninitialized W Register ............................................. 8-9 Reset Sequence................................................................. 6-2 Returning From Interrupt..................................................6-13 Round Logic .....................................................................2-25 Run-Time Self Programming (RTSP)............................... 5-10 FLASH Operations ................................................... 5-10 Operation ................................................................. 5-10 W14 Stack Frame Pointer........................................ 2-10 Special Conditions for Interrupt Latency.......................... 6-13 Special Features for Device Emulation.......................... 15-39 Special Function Register Reset States .......................... 8-16 Specifying How Conversion Results are Written Into Buffer ..................................................................... 17-30 Specifying How Conversion Results are Written into Buffer (12-bit)......................................................... 18-19 SSPOV .......................................................................... 21-19
T
Table Instruction Operation ............................................... 5-2 TCP/IP Protocol Stack ................................................... 25-10 Third Party C Compilers .................................................. 25-6 Third Party Hardware/Software Tools and Application Libraries ................................................................... 25-6 Time-base for Input Capture/Output Compare .............. 12-22 Timer as an External Interrupt Pin ................................. 12-22 Timer Interrupts ............................................................. 12-14 Timer Modes of Operation ............................................... 12-9 Synchronous Counter Using External Clock Input............................................................... 12-10 Timer Mode.............................................................. 12-9 Type A Timer Asynchronous Counter Mode Using External Clock Input ...................................... 12-11 32-bit Timer............................................................ 12-18 Timer Modules Associated Special Function Registers ................. 12-23 Timer Operation in Power Saving States....................... 12-21 Timer Operation Modes Gated Time Accumulation ..................................... 12-12 with Fast External Clock Source............................ 12-12 Timer Prescalers............................................................ 12-14 Timer Selection................................................................ 13-4 Timer Variants ................................................................. 12-3 Timers Design Tips............................................................ 12-24 Related Application Notes ..................................... 12-25 Revision History..................................................... 12-26 Timing Diagrams Brown-out Situations.................................................. 8-8 Clock Transition ....................................................... 7-35 Clock/Instruction Cycle .............................................. 7-5 Data Space Access .......................................... 2-35, 3-6 Dead Time ............................................................. 15-28 Device Reset Delay, Crystal + PLL Clock Source, PWRT Disabled ............................................... 8-13 Device Reset Delay, Crystal + PLL Clock Source, PWRT Enabled................................................ 8-14 Device Reset Delay, EC + PLL Clock, PWRT Enabled ........................................................... 8-15 Device Reset Delay, EC or RC Clock, PWRT Disabled........................................................... 8-16 Dual Compare Mode.............................................. 14-10 Dual Compare Mode (Continuous Output Pulse, PR2 = OCxRS) ................................... 14-14, 14-15 Dual Compare Mode (Single Output Pulse, OCxRS > PR2) .............................................. 14-10 Edge Detection Mode .............................................. 13-8 Gated Timer Mode Operation ................................ 12-13 Interrupt Timing During a Two-Cycle Instruction ..... 6-12 Interrupt Timing for Timer Period Match ................ 12-14 Interrupt Timing, Interrupt Occurs During 1st Cycle of a Two-Cycle Instruction ............................... 6-12 POR Module for Rising VDD ...................................... 8-6 Postscaler Update ................................................... 7-34 PWM Output ............................................... 14-18, 14-21
S
Saturation and Overflow Modes....................................... 2-24 Selecting A/D Conversion Clock .................................... 17-13 Selecting Analog Inputs for Sampling ............................ 17-14 Selecting Analog Inputs for Sampling (12-bit)................ 18-12 Selecting the A/D Conversion Clock (12-bit).................. 18-12 Selecting the Voltage Reference Source .......................17-13 Selecting the Voltage Reference Source (12-bit)........... 18-11 Setup for Continuous Output Pulse Generation............. 14-15 Shadow Registers .............................................................. 2-6 DO Loop.....................................................................2-7 PUSH.S and POP.S................................................... 2-7 Simple Capture Events .................................................... 13-4 Sleep and Idle Modes Operation.................................... 17-55 Sleep and Idle Modes Operation (12-bit) .......................18-33 Sleep Mode ...................................................................... 10-2 and FSCM Delay...................................................... 10-3 Clock Selection on Wake-up from............................ 10-2 Delay on Wake-up from ........................................... 10-3 Delay Times for Exit ................................................. 10-3 Wake-up from on Interrupt ....................................... 10-4 Wake-up from on Reset ........................................... 10-4 Wake-up from on Watchdog Time-out ..................... 10-4 Wake-up from with Crystal Oscillator or PLL ........... 10-3 Slow Oscillator Start-up.................................................... 10-3 Soft Traps........................................................................... 6-6 Arithmetic Error (Level 11) ......................................... 6-7 Software Reset Instruction (SWR) ..................................... 8-7 Software Stack Examples ................................................................... 2-9 Pointer........................................................................ 2-8 Pointer Overflow ...................................................... 2-10 Pointer Underflow .................................................... 2-10
DS70046E-page 6

Reception with Address Detect (ADDEN = 1)........ 19-19 Return From Interrupt .............................................. 6-13 Simple Capture Event, Time-base Prescaler = 1:1................................................. 13-5 Simple Capture Event, Time-base Prescaler = 1:4................................................. 13-5 Single Compare Mode (Force OCx Low on Compare Match Event).................................................... 14-6 Single Compare Mode (Set OCx High on Compare Match Event).................................................... 14-5 Single Compare Mode (Toggle Output on Compare Match Event, PR2 = OCxR)............................. 14-7 Single Compare Mode (Toggle Output on Compare Match Event, PR2 > OCxR)............................. 14-7 SPI Mode Timing (No SS Control) ..............20-12, 20-13 Transmission (Back to Back) ................................. 19-13 Transmission (8-bit or 9-bit Data) .......................... 19-13 UART Reception .................................................... 19-17 UART Reception with Receive Overrun................. 19-17 TRIS (Data Direction) Registers ...................................... 11-3 Tuning the Oscillator Circuit ............................................. 7-23 Type A Timer ................................................................... 12-3 Type B Timer ................................................................... 12-4 Type C Timer ................................................................... 12-5 Using Table Write Instructions........................................... 5-4 Byte Mode ................................................................. 5-5 Holding Latches......................................................... 5-4 Word Mode ................................................................ 5-4 Using the RCON Status Bits............................................ 8-10
Index
V
V.22/V.22bis and V.32 Specification.............................. 25-11
W
Wake-up from Sleep and Idle .......................................... 6-10 Watchdog Time-out Reset (WDTR)................................... 8-7 Watchdog Timer .............................................................. 10-6 Enabling and Disabling............................................ 10-6 Operation................................................................. 10-7 Operation in Sleep and Idle Modes ......................... 10-8 Period Selection ...................................................... 10-7 Prescalers................................................................ 10-7 Resetting ................................................................. 10-8 Software Controlled ................................................. 10-6 WCOL ...................................................... 21-18, 21-19, 21-21 WDT and Power Saving Modes Design Tips............................................................ 10-10 Related Application Notes ..................................... 10-11 Revision History..................................................... 10-12 Working Register Array...................................................... 2-6 W Register Memory Mapping .................................... 2-6 W Registers and Byte Mode Instructions .................. 2-6 W0 and File Register Instructions.............................. 2-6 Writing to Device Configuration Registers ....................... 5-14 Write Algorithm ........................................................ 5-14
U
UART ADDEN Control Bit................................................. 19-18 Alternate I/O Pins................................................... 19-10 Associated Registers ............................................. 19-22 Baud Rate Generator............................................... 19-8 Configuration.......................................................... 19-10 Control Registers ..................................................... 19-3 Design Tips ............................................................ 19-23 Disabling ................................................................ 19-10 Enabling ................................................................. 19-10 Other Features....................................................... 19-21 Auto Baud Support ........................................ 19-21 Loopback Mode ............................................. 19-21 Operation During CPU Sleep and Idle Modes .............................................. 19-21 Receiver................................................................. 19-14 Buffer (UxRXB) .............................................. 19-14 Error Handling................................................ 19-14 Interrupt ......................................................... 19-15 Setup for Reception ....................................... 19-17 Related Application Notes...................................... 19-24 Setup for 9-bit Transmit ......................................... 19-18 Transmitter............................................................. 19-11 Buffer (UxTXB) .............................................. 19-12 Interrupt ......................................................... 19-12 Setup ............................................................. 19-13 Transmission of Break Characters ................ 19-14 Using for 9-bit Communication............................... 19-18 UART Autobaud Support ................................................. 13-9 Uninitialized W Register Reset .......................................... 2-7 USART Initialization ............................................................ 19-20 Introduction .............................................................. 19-2 Receiving Break Characters .................................. 19-19 Revision History ..................................................... 19-25 Setup for 9-bit Reception Using Address Detect Mode .............................................................. 19-19 Using QEI as Alternate 16-bit Timer/Counter ................ 16-16 Using Table Read Instructions ........................................... 5-3 Byte Mode.................................................................. 5-3 Word Mode ................................................................ 5-3
Z
10-bit Address Mode...................................................... 21-35 12-Bit A/D ADCHS ........................................................... 17-4, 18-4 ADPCFG......................................................... 17-4, 18-4 12-bit A/D Operation During CPU Idle Mode............... 17-55, 18-33 Operation During CPU Sleep Mode ........... 17-55, 18-33 16-bit Up/Down Position Counter .................................. 16-12 32-bit Timer Configuration ............................................. 12-16
DS70046E-page 7
WORLDWIDE SALES AND SERVICE

AMERICAS
Corporate Office 2355 West Chandler Blvd. Chandler, AZ 85224-6199 Tel: 480-792-7200 Fax: 480-792-7277 Technical Support: http://support.microchip.com Web Address: www.microchip.com Atlanta Alpharetta, GA Tel: 770-640-0034 Fax: 770-640-0307 Boston Westborough, MA Tel: 774-760-0087 Fax: 774-760-0088 Chicago Itasca, IL Tel: 630-285-0071 Fax: 630-285-0075 Dallas Addison, TX Tel: 972-818-7423 Fax: 972-818-2924 Detroit Farmington Hills, MI Tel: 248-538-2250 Fax: 248-538-2260 Kokomo Kokomo, IN Tel: 765-864-8360 Fax: 765-864-8387 Los Angeles Mission Viejo, CA Tel: 949-462-9523 Fax: 949-462-9608 San Jose Mountain View, CA Tel: 650-215-1444 Fax: 650-961-0286 Toronto Mississauga, Ontario, Canada Tel: 905-673-0699 Fax: 905-673-6509
ASIA/PACIFIC
Australia - Sydney Tel: 61-2-9868-6733 Fax: 61-2-9868-6755 China - Beijing Tel: 86-10-8528-2100 Fax: 86-10-8528-2104 China - Chengdu Tel: 86-28-8676-6200 Fax: 86-28-8676-6599 China - Fuzhou Tel: 86-591-8750-3506 Fax: 86-591-8750-3521 China - Hong Kong SAR Tel: 852-2401-1200 Fax: 852-2401-3431 China - Qingdao Tel: 86-532-8502-7355 Fax: 86-532-8502-7205 China - Shanghai Tel: 86-21-5407-5533 Fax: 86-21-5407-5066 China - Shenyang Tel: 86-24-2334-2829 Fax: 86-24-2334-2393 China - Shenzhen Tel: 86-755-8203-2660 Fax: 86-755-8203-1760 China - Shunde Tel: 86-757-2839-5507 Fax: 86-757-2839-5571 China - Wuhan Tel: 86-27-5980-5300 Fax: 86-27-5980-5118 China - Xian Tel: 86-29-8833-7250 Fax: 86-29-8833-7256
ASIA/PACIFIC
India - Bangalore Tel: 91-80-4182-8400 Fax: 91-80-4182-8422 India - New Delhi Tel: 91-11-5160-8631 Fax: 91-11-5160-8632 India - Pune Tel: 91-20-2566-1512 Fax: 91-20-2566-1513 Japan - Yokohama Tel: 81-45-471- 6166 Fax: 81-45-471-6122 Korea - Gumi Tel: 82-54-473-4301 Fax: 82-54-473-4302 Korea - Seoul Tel: 82-2-554-7200 Fax: 82-2-558-5932 or 82-2-558-5934 Malaysia - Penang Tel: 60-4-646-8870 Fax: 60-4-646-5086 Philippines - Manila Tel: 63-2-634-9065 Fax: 63-2-634-9069 Singapore Tel: 65-6334-8870 Fax: 65-6334-8850 Taiwan - Hsin Chu Tel: 886-3-572-9526 Fax: 886-3-572-6459 Taiwan - Kaohsiung Tel: 886-7-536-4818 Fax: 886-7-536-4803 Taiwan - Taipei Tel: 886-2-2500-6610 Fax: 886-2-2508-0102 Thailand - Bangkok Tel: 66-2-694-1351 Fax: 66-2-694-1350
EUROPE
Austria - Wels Tel: 43-7242-2244-399 Fax: 43-7242-2244-393 Denmark - Copenhagen Tel: 45-4450-2828 Fax: 45-4485-2829 France - Paris Tel: 33-1-69-53-63-20 Fax: 33-1-69-30-90-79 Germany - Munich Tel: 49-89-627-144-0 Fax: 49-89-627-144-44 Italy - Milan Tel: 39-0331-742611 Fax: 39-0331-466781 Netherlands - Drunen Tel: 31-416-690399 Fax: 31-416-690340 Spain - Madrid Tel: 34-91-708-08-90 Fax: 34-91-708-08-91 UK - Wokingham Tel: 44-118-921-5869 Fax: 44-118-921-5820
02/16/06
DS70046E

Dspic30F Family Reference Manual: High-Performance Digital Signal Controllers

Uploaded by

Copyright:

Available Formats

Dspic30F Family Reference Manual: High-Performance Digital Signal Controllers

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

Dspic30F Family Reference Manual: High-Performance Digital Signal Controllers

Uploaded by

Copyright:

Available Formats

dsPIC30F Family Reference Manual

High-Performance Digital Signal Controllers

2006 Microchip Technology Inc.

2006 Microchip Technology Inc.

SECTION 3. DATA MEMORY

SECTION 4. PROGRAM MEMORY

2006 Microchip Technology Inc.

SECTION 6. RESET INTERRUPTS

2006 Microchip Technology Inc.

SECTION 9. LOW VOLTAGE DETECT (LVD)

SECTION 10. WATCHDOG TIMER AND POWER SAVING MODES

SECTION 11. I/O PORTS

2006 Microchip Technology Inc.

SECTION 13. INPUT CAPTURE

SECTION 14. OUTPUT COMPARE

2006 Microchip Technology Inc.

SECTION 16. QUADRATURE ENCODER INTERFACE (QEI)

2006 Microchip Technology Inc.

2006 Microchip Technology Inc.

SECTION 19. UART

2006 Microchip Technology Inc.

SECTION 21. INTER-INTEGRATED CIRCUIT (I2C)

SECTION 22. DATA CONVERTER INTERFACE (DCI)

2006 Microchip Technology Inc.

SECTION 24. DEVICE CONFIGURATION

SECTION 25. DEVELOPMENT TOOL SUPPORT

SECTION 26. APPENDIX

2006 Microchip Technology Inc.

2006 Microchip Technology Inc.

2004 Microchip Technology Inc.