Pic18f66k90t I PTRSL

PIC18F87K90 Family
Data Sheet
64/80-Pin, High-Performance
Microcontrollers with LCD Driver and
nanoWatt XLP Technology
 2009-2011 Microchip Technology Inc. DS39957D

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 Microchip’s 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 Microchip’s 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 Trademarks

applications and the like is provided only for your convenience The Microchip name and logo, the Microchip logo, dsPIC,
and may be superseded by updates. It is your responsibility to KEELOQ, KEELOQ logo, MPLAB, PIC, PICmicro, PICSTART,
ensure that your application meets with your specifications.
PIC32 logo, rfPIC and UNI/O are registered trademarks of
MICROCHIP MAKES NO REPRESENTATIONS OR
Microchip Technology Incorporated in the U.S.A. and other
WARRANTIES OF ANY KIND WHETHER EXPRESS OR
countries.
IMPLIED, WRITTEN OR ORAL, STATUTORY OR
OTHERWISE, RELATED TO THE INFORMATION, FilterLab, Hampshire, HI-TECH C, Linear Active Thermistor,
INCLUDING BUT NOT LIMITED TO ITS CONDITION, MXDEV, MXLAB, SEEVAL and The Embedded Control
QUALITY, PERFORMANCE, MERCHANTABILITY OR Solutions Company are registered trademarks of Microchip
FITNESS FOR PURPOSE. Microchip disclaims all liability Technology Incorporated in the U.S.A.
arising from this information and its use. Use of Microchip Analog-for-the-Digital Age, Application Maestro, chipKIT,
devices in life support and/or safety applications is entirely at chipKIT logo, CodeGuard, dsPICDEM, dsPICDEM.net,
the buyer’s risk, and the buyer agrees to defend, indemnify and dsPICworks, dsSPEAK, ECAN, ECONOMONITOR,
hold harmless Microchip from any and all damages, claims, FanSense, HI-TIDE, In-Circuit Serial Programming, ICSP,
suits, or expenses resulting from such use. No licenses are Mindi, MiWi, MPASM, MPLAB Certified logo, MPLIB,
conveyed, implicitly or otherwise, under any Microchip MPLINK, mTouch, Omniscient Code Generation, PICC,
intellectual property rights. PICC-18, PICDEM, PICDEM.net, PICkit, PICtail, REAL ICE,
rfLAB, Select Mode, Total Endurance, TSHARC,
UniWinDriver, 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.
© 2009-2011, Microchip Technology Incorporated, Printed in
the U.S.A., All Rights Reserved.
Printed on recycled paper.
ISBN: 978-1-61341-351-7
Microchip received ISO/TS-16949:2009 certification for its worldwide

headquarters, design and wafer fabrication facilities in Chandler and
Tempe, Arizona; Gresham, Oregon and design centers in California
and India. The Company’s quality system processes and procedures
are for its PIC® MCUs and dsPIC® DSCs, KEELOQ® code hopping
devices, Serial EEPROMs, microperipherals, nonvolatile memory and
analog products. In addition, Microchip’s quality system for the design
and manufacture of development systems is ISO 9001:2000 certified.
DS39957D-page 2  2009-2011 Microchip Technology Inc.

PIC18F87K90 FAMILY
64/80-Pin, High-Performance Microcontrollers with
LCD Driver and nanoWatt XLP Technology
Low-Power Features: Peripheral Highlights:

• Power-Managed modes: • Ten or Eight CCP/ECCP modules:
- Run: CPU on, peripherals on - Seven Capture/Compare/PWM (CCP) modules
- Idle: CPU off, peripherals on - Three Enhanced Capture/Compare/PWM
- Sleep: CPU off, peripherals off (ECCP) modules
• Two-Speed Oscillator Start-up • Eleven 8/16-Bit Timer/Counter modules:
• Fail-Safe Clock Monitor - Timer0 – 8/16-bit timer/counter with 8-bit
• Power-Saving Peripheral Module Disable (PMD) programmable prescaler
• Ultra Low-Power Wake-up - Timer1, 3, 5, 7 – 16-bit timer/counter
• Fast Wake-up, 1 s Typical - Timer2, 4, 6, 8, 10, 12 – 8-bit timer/counter
• Low-Power WDT, 300 nA Typical • Three Analog Comparators
• Ultra Low 50 nA Input Leakage • Configurable Reference Clock Output
• Run mode Currents Down to 5.5 A, Typical • Hardware Real-Time Clock and Calendar (RTCC)
• Idle mode Currents Down to 1.7 A, Typical module with Clock, Calendar and Alarm Functions
• Sleep mode Current Down to Very Low 20 nA, Typical - Time-out from 0.5s to 1 year
• Charge Time Measurement Unit (CTMU):
• RTCC Current Down to Very Low 700 nA, Typical
- Capacitance measurement for mTouch™
• LCD Current Down to Very Low 300 nA, Typical
Sensing
- Time measurement with 1 ns typical resolution
LCD Driver and Keypad Features:
• High-Current Sink/Source 25 mA/25 mA (PORTB
• Direct LCD Panel Drive Capability: and PORTC)
- Can drive LCD panel while in Sleep mode • Up to Four External Interrupts
• Up to 48 Segments and 192 Pixels, • Two Master Synchronous Serial Port (MSSP)
Software-Selectable modules:
• Programmable LCD Timing module: - 3/4-wire SPI (supports all four SPI modes)
- Multiple LCD timing sources available - I2C™ Master and Slave mode
- Up to four commons: static, 1/2, 1/3 or
1/4 multiplex
- Bias configuration: Static, 1/2 or 1/3
• Low-Power Resistor Bias Network for LCD
Comparators
(Channels)
12-Bit A/D
Flash SRAM
EUSART
8/16-Bit
Timers
CTMU
RTCC
Program Data EEPROM LCD CCP/

Device I/O SPI I2C™
Memory Memory (Bytes) Pixels ECCP
(Bytes) (Bytes)
PIC18F65K90 32K 2K 1K 53 132 4/4 5/3 Yes Yes 2 16 3 Y Y

 2009-2011 Microchip Technology Inc. DS39957D-page 3

PIC18F87K90 FAMILY
Special Microcontroller Features: • Priority Levels for Interrupts
• Operating Voltage Range: 1.8V to 5.5V • 8 x 8 Single-Cycle Hardware Multiplier
• On-Chip 3.3V Regulator • Extended Watchdog Timer (WDT):
• Operating Speed up to 64 MHz - Programmable period from 4 ms to 4,194s
• Up to 128 Kbytes On-Chip Flash Program (about 70 minutes)
Memory • In-Circuit Serial Programming™ (ICSP™) via
• Data EEPROM of 1,024 Bytes Two Pins
• 4K x 8 General Purpose Registers (SRAM) • In-Circuit Debug via Two Pins
• 10,000 Erase/Write Cycle Flash Program • Programmable:
Memory, Minimum - BOR
• 1,000,000 Erase/write Cycle Data EEPROM - LVD
Memory, Typical • Two Enhanced Addressable USART modules:
• Flash Retention 40 Years, Minimum - LIN/J2602 support
• Three Internal Oscillators: LF-INTRC (31 kHz), - Auto-Baud Detect (ABD)
MF-INTOSC (500 kHz) and HF-INTOSC • 12-Bit A/D Converter with up to 24 Channels:
(16 MHz) - Auto-acquisition and Sleep operation
• Self-Programmable under Software Control - Differential Input mode of operation

PIC18F87K90 FAMILY
Pin Diagrams – PIC18F6XK90
64-Pin QFN(3), TQFP
RE3/COM0/P3C/CCP9(2)/REFO
RE2/LCDBIAS3/P2B/CCP10(2)
RE7/ECCP2(1)/SEG31/P2A
RD6/SEG6/SCK2/SCL2
RD5/SEG5/SDI2/SDA2
RE5/COM2/P1C/CCP7
RD1/SEG1/T5CKI/T7G
RE4/COM1/P3B/CCP8
RE6/COM3/P1B/CCP6
RD0/SEG0/CTPLS
RD4/SEG4/SDO2
RD7/SEG7/SS2
RD2/SEG2
RD3/SEG3
VDD
VSS
64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49
RE1/LCDBIAS2/P2C 1 48 RB0/INT0/SEG30/FLTO
RE0/LCDBIAS1/P2D 2 47 RB1/INT1/SEG8
RG0/ECCP3/P3A 3 46 RB2/INT2/SEG9/CTED1
RG1/TX2/CK2/AN19/C3OUT 4 45 RB3/INT3/SEG10/CTED2/P2A
RG2/RX2/DT2/AN18/C3INA 5 44 RB4/KBI0/SEG11
RG3/CCP4/AN17/P3D/C3INB 6 PIC18F65K90 43 RB5/KBI1/SEG29/T3CKI/T1G
MCLR/RG5 7 42 RB6/KBI2/PGC
PIC18F66K90
RG4/SEG26/RTCC/T7CKI(2)/T5G/CCP5/AN16/P1D/C3INC 8 41 VSS
VSS 9 PIC18F67K90 40 OSC2/CLKO/RA6
VDDCORE/VCAP 10 39 OSC1/CLKI/RA7
RF7/AN5/SS1/SEG25 11 38 VDD
RF6/AN11/SEG24/C1INA 12 37 RB7/KBI3/PGD
RF5/AN10/CVREF/SEG23/C1INB 13 36 RC5/SDO1/SEG12
RF4/AN9/SEG22/C2INA 14 35 RC4/SDI1/SDA1/SEG16
RF3/AN8/SEG21/C2INB/CTMUI 15 34 RC3/SCK1/SCL1/SEG17
RF2/AN7/C1OUT/SEG20 16 33 RC2/ECCP1/P1A/SEG13
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
ENVREG
RF1/AN6/C2OUT/SEG19/CTDIN
RA2/AN2/VREF-
RC7/RX1/DT1/SEG28
RC1/SOSCI/ECCP2(1)/P2A/SEG32
RC0/SOSCO/SCLKI
RC6/TX1/CK1/SEG27
RA3/AN3/VREF+
RA5/AN4/T1CKI/SEG15/T3G/HLVDIN
AVDD
RA1/AN1/SEG18
VDD
RA4/T0CKI/SEG14
AVSS
RA0/AN0/ULPWU
VSS
Note 1: The ECCP2 pin placement depends on the CCP2MX Configuration bit setting.
2: Not available on the PIC18F65K90 and PIC18F85K90.
3: For the QFN package, it is recommended that the bottom pad be connected to VSS.

PIC18F87K90 FAMILY
Pin Diagrams – PIC18F8XK90
80-Pin TQFP
RE3/COM0/P3C/CCP9(2)(3)/REFO
RE2/LCDBIAS3/P2B/CCP10(2)
RD1/SEG1/T5CKI/T7G/PSP1
RE7/ECCP2(1)/P2A/SEG31
RD0/PSP0/SEG0/CTPLS
RE5/COM2/P1C/CCP7(3)
RE4/COM1/P3B/CCP8(3)
RE6/COM3/P1B/CCP6(3)
RD6/SEG6/SCK2/SCL2
RD5/SEG5/SDI2/SDA2
RH1/SEG46/AN22
RH0/SEG47/AN23
RD4/SEG4/SDO2
RD7/SEG7/SS2
RJ1/SEG33
RD2/SEG2
RD3/SEG3
VDD
RJ0
VSS
80 79 78 77 76 75 74 73 72 71 70 69 68 6766 65 64 63 62 61
RH2/SEG45/AN21 1 60 RJ2/SEG34
RH3/SEG44/AN20 2 59 RJ3/SEG35
RE1/LCDBIAS2/P2C 3 58 RB0/INT0/SEG30/FLT0
RE0/LCDBIAS1/P2D 4 57 RB1/INT1/SEG8
RG0/ECCP3/P3A 5 56 RB2/INT2/SEG9/CTED1
RG1/TX2/CK2/AN19/C3OUT 6 55 RB3/INT3/SEG10/CTED2/P2A
RG2/RX2/DT2/AN18/C3INA 7 54 RB4/KBI0/SEG11
RG3/CCP4/AN17/P3D/C3INB 8 53 RB5/KBI1/SEG29/T3CKI/T1G
MCLR/RG5 9 PIC18F85K90 52 RB6/KBI2/PGC
RG4/SEG26/RTCC/T7CKI(2)/T5G/CCP5/AN16/P1D/C3INC 10 PIC18F86K90 51 VSS
VSS 11 50 OSC2/CLKO/RA6
VDDCORE/VCAP 12 PIC18F87K90 49 OSC1/CLKI/RA7
RF7/AN5/SS1/SEG25 13 48 VDD
RF6/AN11/SEG24/C1INA 14 47 RB7/KBI3/PGD
RF5/AN10/CVREF/SEG23/C1INB 15 46 RC5/SDO1/SEG12
RF4/AN9/SEG22/C2INA 16 45 RC4/SDI1/SDA1/SEG16
RF3/AN8/SEG21/C2INB/CTMUI 17 44 RC3/SCK1/SCL1/SEG17
RF2/AN7/C1OUT/SEG20 18 43 RC2/ECCP1/P1A/SEG13
RH7/SEG43/CCP6(3)/P1B/AN15 19 42 RJ7/SEG36
RH6/SEG42/CCP7(3)/P1C/AN14/C1INC 20 41 RJ6/SEG37
21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
RA3/AN3/VREF+
RC1/SOSCI/ECCP2(1)I/SEG32/P2A
RA2/AN2/VREF-
RA1/AN1/SEG18
RJ4/SEG39
RJ5/SEG38
RA4/T0CKI/SEG14
ENVREG
AVDD
VDD
RC7/RX1/DT1/SEG28
RF1/AN6/C2OUT/SEG19/CTDIN
RC0/SOSCO/SCKLI
RC6/TX1/CK1/SEG27
RA0/AN0/ULPWU
RA5/AN4/T1CKI/SEG15/T3G/HLVDIN
RH5/SEG41/CCP8(3)/P3B/AN13/C2IND
RH4/SEG40/CCP9(2)(3)/P3C/AN12/C2INC
AVSS
VSS
Note 1: The ECCP2 pin placement depends on the CCP2MX Configuration bit setting.
2: Not available on the PIC18F65K90 and PIC18F85K90.
3: The CCP6, CCP7, CCP8 and CCP9 pin placement depends on the ECCPMX Configuration bit setting

PIC18F87K90 FAMILY
Table of Contents
1.0 Device Overview .......................................................................................................................................................................... 9
2.0 Guidelines for Getting Started with PIC18FXXKXX Microcontrollers ......................................................................................... 35
3.0 Oscillator Configurations ............................................................................................................................................................ 41
4.0 Power-Managed Modes ............................................................................................................................................................. 53
5.0 Reset .......................................................................................................................................................................................... 69
6.0 Memory Organization ................................................................................................................................................................. 85
7.0 Flash Program Memory............................................................................................................................................................ 111
8.0 Data EEPROM Memory ........................................................................................................................................................... 121
9.0 8 x 8 Hardware Multiplier.......................................................................................................................................................... 127
10.0 Interrupts .................................................................................................................................................................................. 129
11.0 I/O Ports ................................................................................................................................................................................... 153
12.0 Timer0 Module ......................................................................................................................................................................... 183
13.0 Timer1 Module ......................................................................................................................................................................... 187
14.0 Timer2 Module ......................................................................................................................................................................... 199
15.0 Timer3/5/7 Modules.................................................................................................................................................................. 201
16.0 Timer4/6/8/10/12 Modules........................................................................................................................................................ 213
17.0 Real-Time Clock and Calendar (RTCC)................................................................................................................................... 217
18.0 Capture/Compare/PWM (CCP) Modules ................................................................................................................................. 237
19.0 Enhanced Capture/Compare/PWM (ECCP) Module................................................................................................................ 251
20.0 Liquid Crystal Display (LCD) Driver Module............................................................................................................................. 273
21.0 Master Synchronous Serial Port (MSSP) Module .................................................................................................................... 303
22.0 Enhanced Universal Synchronous Asynchronous Receiver Transmitter (EUSART) ............................................................... 349
23.0 12-Bit Analog-to-Digital Converter (A/D) Module ..................................................................................................................... 373
24.0 Comparator Module.................................................................................................................................................................. 389
25.0 Comparator Voltage Reference Module................................................................................................................................... 397
26.0 High/Low-Voltage Detect (HLVD)............................................................................................................................................. 401
27.0 Charge Time Measurement Unit (CTMU) ................................................................................................................................ 407
28.0 Special Features of the CPU.................................................................................................................................................... 425
29.0 Instruction Set Summary .......................................................................................................................................................... 451
30.0 Development Support............................................................................................................................................................... 501
31.0 Electrical Characteristics .......................................................................................................................................................... 505
32.0 Packaging Information.............................................................................................................................................................. 545
Appendix A: Revision History............................................................................................................................................................. 553
Appendix B: Migration From PIC18F85J90 and PIC18F87J90 to PIC18F87K90 .............................................................................. 553
Index ................................................................................................................................................................................................. 555
The Microchip Web Site ..................................................................................................................................................................... 567
Customer Change Notification Service .............................................................................................................................................. 567
Customer Support .............................................................................................................................................................................. 567
Reader Response .............................................................................................................................................................................. 568
Product Identification System ............................................................................................................................................................ 569

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
1.0 DEVICE OVERVIEW • An internal oscillator block that provides a 16 MHz
clock (±2% accuracy) and an INTRC source
This document contains device-specific information for (approximately 31 kHz, stable over temperature
the following devices: and VDD)
• PIC18F65K90 • PIC18F85K90 - Operates as HF-INTOSC or MF-INTOSC
• PIC18F66K90 • PIC18F86K90 when block selected for 16 MHz or 500 kHz
• PIC18F67K90 • PIC18F87K90 - Frees the two oscillator pins for use as
additional general purpose I/O
This family combines the traditional advantages of all
PIC18 microcontrollers – namely, high computational The internal oscillator block provides a stable reference
performance and a rich feature set – with a versatile source that gives the family additional features for
on-chip LCD driver, while maintaining an extremely robust operation:
competitive price point. These features make the • Fail-Safe Clock Monitor: This option constantly
PIC18F87K90 family a logical choice for many monitors the main clock source against a reference
high-performance applications where price is a primary signal provided by the internal oscillator. If a clock
consideration. failure occurs, the controller is switched to the inter-
nal oscillator, allowing for continued low-speed
1.1 Core Features operation or a safe application shutdown.
• Two-Speed Start-up: This option allows the
1.1.1 nanoWatt TECHNOLOGY internal oscillator to serve as the clock source
All of the devices in the PIC18F87K90 family incorpo- from Power-on Reset, or wake-up from Sleep
rate a range of features that can significantly reduce mode, until the primary clock source is available.
power consumption during operation. Key items include:
1.1.3 MEMORY OPTIONS
• Alternate Run Modes: By clocking the controller
from the Timer1 source or the internal RC The PIC18F87K90 family provides ample room for
oscillator, power consumption during code application code, from 32 Kbytes to 128 Kbytes of code
execution can be reduced. space. The Flash cells for program memory are rated
• Multiple Idle Modes: The controller can also run to last up to 10,000 erase/write cycles. Data retention
with its CPU core disabled but the peripherals still without refresh is conservatively estimated to be
active. In these states, power consumption can be greater than 40 years.
reduced even further. The Flash program memory is readable and writable.
• On-the-Fly Mode Switching: The power-managed During normal operation, the PIC18F87K90 family also
modes are invoked by user code during operation, provides plenty of room for dynamic application data
allowing the user to incorporate power-saving ideas with up to 3,828 bytes of data RAM.
into their application’s software design.
1.1.4 EXTENDED INSTRUCTION SET
• nanoWatt XLP: An extra low-power BOR, RTCC
and low-power Watchdog Timer. Also, an ultra The PIC18F87K90 family implements the optional
low-power regulator for Sleep mode is provided in extension to the PIC18 instruction set, adding 8 new
regulator-enabled modes. instructions and an Indexed Addressing mode.
Enabled as a device configuration option, the extension
1.1.2 OSCILLATOR OPTIONS AND has been specifically designed to optimize re-entrant
FEATURES application code originally developed in high-level
languages, such as ‘C’.
All of the devices in the PIC18F87K90 family offer
different oscillator options, allowing users a range of 1.1.5 EASY MIGRATION
choices in developing application hardware. These
include: Regardless of the memory size, all devices share the
same rich set of peripherals (except the 32-Kbyte parts,
• External Resistor/Capacitor (RC); RA6 available which have two less CCPs and three less Timers),
• External Resistor/Capacitor with Clock Out (RCIO) allowing for a smooth migration path as applications
• Three External Clock modes: grow and evolve.
- External Clock (EC); RA6 available The consistent pinout scheme, used throughout the
- External Clock with Clock Out (ECIO) entire family, also aids in migrating to the next larger
- External Crystal (XT, HS, LP) device. This is true when moving between the 64-pin
members, between the 80-pin members, or even
• A Phase Lock Loop (PLL) frequency multiplier,
jumping from 64-pin to 80-pin devices.
available to the External Oscillator modes which
allows clock speeds of up to 64 MHz. PLL can
also be used with the internal oscillator.

PIC18F87K90 FAMILY
The PIC18F87K90 family is also largely • LP Watchdog Timer (WDT): This enhanced
pin-compatible with other PIC18 families, such as the version incorporates a 22-bit prescaler, allowing
PIC18F8720, PIC18F8722, PIC18F85J11, PIC18F8490, an extended time-out range that is stable across
PIC18F85J90, PIC18F87J90 and PIC18F87J93 families operating voltage and temperature. See
of microcontrollers with LCD drivers. This allows a new Section 31.0 “Electrical Characteristics” for
dimension to the evolution of applications, allowing time-out periods.
developers to select different price points within • Real-Time Clock and Calendar Module (RTCC):
Microchip’s PIC18 portfolio, while maintaining a similar The RTCC module is intended for applications
feature set. requiring that accurate time be maintained for
extended periods of time with minimum to no
1.2 LCD Driver intervention from the CPU.
The on-chip LCD driver includes many features that The module is a 100-year clock and calendar with
ease the integration of displays in low-power automatic leap year detection. The range of the
applications. These include an integrated internal clock is from 00:00:00 (midnight) on January 1, 2000
resistor ladder, so bias voltages can be generated to 23:59:59 on December 31, 2099.
internally. This enables software-controlled contrast
control and eliminates the need for external bias 1.4 Details on Individual Family
voltage resistors. Members
Devices in the PIC18F87K90 family are available in
1.3 Other Special Features 64-pin and 80-pin packages. Block diagrams for the
two groups are shown in Figure 1-1 and Figure 1-2.
• Communications: The PIC18F87K90 family
incorporates a range of serial communication The devices are differentiated from each other in these ways:
peripherals including two Enhanced USART, that • Flash Program Memory:
support LIN/J2602, and two Master SSP modules - PIC18FX5K90 (PIC18F65K90 and
capable of both SPI and I2C™ (Master and Slave) PIC18F85K90) – 32 Kbytes
modes of operation.
- PIC18FX6K90 (PIC18F66K90 and
• CCP Modules: PIC18F87K90 family devices PIC18F86K90) – 64 Kbytes
incorporate up to seven or five Capture/
- PIC18FX7K90 (PIC18F67K90 and
Compare/PWM (CCP) modules. Up to six differ-
PIC18F87K90) – 128 Kbytes
ent time bases can be used to perform several
different operations at once. • Data RAM:
• ECCP Modules: The PIC18F87K90 family has - All devices except PIC18FX5K90 – 4 Kbytes
three Enhanced CCP (ECCP) modules to - PIC18FX5K90 – 2 Kbytes
maximize flexibility in control applications: • I/O Ports:
- Up to eight different time bases for performing - PIC18F6XK90 (64-pin devices) –
several different operations at once 7 bidirectional ports
- Up to four PWM outputs for each module, for - PIC18F8XK90 (80-pin devices) –
a total of 12 PWMs 9 bidirectional ports
- Other beneficial features, such as polarity • LCD Pixels:
selection, programmable dead time, - PIC18F6XK90 – 132 pixels (33 SEGs x 4 COMs)
auto-shutdown and restart, and Half-Bridge
- PIC18F8XK90 – 192 pixels (48 SEGs x 4 COMs)
and Full-Bridge Output modes
• CCP Module:
• 12-Bit A/D Converter: The PIC18F87K90 family
has differential ADC. It incorporates program- - All devices except PIC18FX5K90 have seven CCP
mable acquisition time, allowing for a channel to modules, PIC18FX5K90 has only five CCP modules
be selected and a conversion to be initiated with- • Timers:
out waiting for a sampling period, and thus, - All devices except 18FX5K90 have six 8-bit timers
reducing code overhead. and five 16-bit timers, PIC18FX5K90 has only four
• Charge Time Measurement Unit (CTMU): The 8-bit timers and four 16-bit timers.
CTMU is a flexible analog module that provides accu- • A/D Channels:
rate differential time measurement between pulse - All PIC18F8XK90 devices have 24 A/D
sources, as well as asynchronous pulse generation. channels, all PIC18F6XK90 devices have
Together with other on-chip analog modules, the 16 A/D channels
CTMU can precisely measure time, measure All other features for devices in this family are identical.
capacitance or relative changes in capacitance, or These are summarized in Table 1-1 and Table 1-2.
generate output pulses that are independent of the The pinouts for all devices are listed in Table 1-3 and
system clock. Table 1-4.

PIC18F87K90 FAMILY
TABLE 1-1: DEVICE FEATURES FOR THE PIC18F6XK90 (64-PIN DEVICES)
Features PIC18F65K90 PIC18F66K90 PIC18F67K90
Operating Frequency DC – 64 MHz
Program Memory (Bytes) 32K 64K 128K
Program Memory (Instructions) 16,384 32,768 65,536
Data Memory (Bytes) 2K 4K 4K
Interrupt Sources 42 48
I/O Ports Ports A, B, C, D, E, F, G
LCD Driver (available pixels to drive) 132 (33 SEGs x 4 COMs)
Timers 8 11
Comparators 3
CTMU Yes
RTCC Yes
Capture/Compare/PWM (CCP) Modules 5 7 7
Enhanced CCP (ECCP) Modules 3
Serial Communications Two MSSP and two Enhanced USART (EUSART)
12-Bit Analog-to-Digital Module 16 Input Channels
Resets (and Delays) POR, BOR, RESET Instruction, Stack Full, Stack Underflow, MCLR, WDT
(PWRT, OST)
Instruction Set 75 Instructions, 83 with Extended Instruction Set Enabled
Packages 64-Pin QFN, 64-Pin TQFP
TABLE 1-2: DEVICE FEATURES FOR THE PIC18F8XK90 (80-PIN DEVICES)

Features PIC18F85K90 PIC18F86K90 PIC18F87K90
Operating Frequency DC – 64 MHz
Program Memory (Bytes) 32K 64K 128K
Program Memory (Instructions) 16,384 32,768 65,536
Data Memory (Bytes) 2K 4K 4K
Interrupt Sources 42 48
I/O Ports Ports A, B, C, D, E, F, G, H, J
LCD Driver (available pixels to drive) 192 (48 SEGs x 4 COMs)
Timers 8 11
Comparators 3
CTMU Yes
RTCC Yes
Capture/Compare/PWM (CCP) Modules 5 7 7
Enhanced CCP (ECCP) Modules 3
Serial Communications Two MSSP and two Enhanced USART (EUSART)
12-Bit Analog-to-Digital Module 24 Input Channels
Resets (and Delays) POR, BOR, RESET Instruction, Stack Full, Stack Underflow, MCLR, WDT
(PWRT, OST)
Instruction Set 75 Instructions, 83 with Extended Instruction Set Enabled
Packages 80-Pin TQFP

PIC18F87K90 FAMILY
FIGURE 1-1: PIC18F6XK90 (64-PIN) BLOCK DIAGRAM
Data Bus<8>
Table Pointer<21>
PORTA
Data Latch
inc/dec logic 8 8 RA0:RA7(1,2)
Data Memory
(2/4 Kbytes)
21 PCLATU PCLATH
20 Address Latch
PCU PCH PCL
Program Counter 12 PORTB
Data Address<12>
RB0:RB7(1)
31-Level Stack
Address Latch 4 12 4
BSR FSR0 Access
Program Memory STKPTR Bank
FSR1
FSR2 12
Data Latch PORTC
inc/dec RC0:RC7(1)
8 logic
Table Latch
ROM Latch Address

Instruction Bus <16> Decode
PORTD
IR RD0:RD7(1)
8
Instruction State Machine
Decode and Control Signals
Control
PRODH PRODL PORTE
RE0:RE7(1)
8 x 8 Multiply
Timing 3
Power-up 8
OSC2/CLKO Generation
OSC1/CLKI Timer
BITOP W
INTRC Oscillator 8 8
Oscillator 8
Start-up Timer
16 MHz PORTF
Oscillator Power-on 8 8
Reset RF1:RF7(1)
Precision ALU<8>
Band Gap Watchdog
Reference Timer
8
ENVREG BOR and
Voltage LVD
Regulator
PORTG
RG0:RG5(1)
VDDCORE/VCAP VDD, VSS MCLR
Timer Timer ADC Comparator

Timer0 Timer1 2/4/6/8/10(3)/12(3) 3/5/7(3) CTMU 12-Bit 1/2/3
CCP ECCP LCD

4/5/6/7/8/9(3)/10(3) 1/2/3 EUSART1 EUSART2 RTCC MSSP1/2
Driver
Note 1: See Table 1-3 for I/O port pin descriptions.

2: RA6 and RA7 are only available as digital I/O in select oscillator modes. For more information, see Section 3.0 “Oscillator
Configurations”.
3: Unimplemented in the PIC18F65K90.

PIC18F87K90 FAMILY
FIGURE 1-2: PIC18F8XK90 (80-PIN) BLOCK DIAGRAM
Data Bus<8>
Table Pointer<21>
PORTA
Data Latch
inc/dec logic 8 8
RA0:RA7(1,2)
Data Memory
(2/4 Kbytes)
21 PCLATU PCLATH
20 Address Latch
PCU PCH PCL PORTB
Program Counter 12 RB0:RB7(1)
Data Address<12>
31-Level Stack
Address Latch 4 12 4
BSR Access PORTC
Program Memory STKPTR FSR0
Bank
FSR1 RC0:RC7(1)
Data Latch FSR2 12
inc/dec
8 logic PORTD
Table Latch
RD0:RD7(1)
ROM Latch Address

Instruction Bus <16> Decode
PORTE
IR
RE0:RE7
8
Instruction State Machine
Decode and Control Signals
Control PORTF
PRODH PRODL
RF1:RF7(1)
8 x 8 Multiply
Timing 3
Power-up 8
OSC2/CLKO Generation
OSC1/CLKI Timer
BITOP W PORTG
INTRC Oscillator 8 8
Oscillator 8
Start-up Timer RG0:RG5(1)
16 MHz
Oscillator Power-on 8 8
Reset
Precision ALU<8> PORTH
Band Gap Watchdog
Reference Timer
8 RH0:RH7(1)
ENVREG BOR and
Voltage
Regulator LVD
PORTJ
VDDCORE/VCAP VDD,VSS MCLR RJ0:RJ7(1)
Timer Timer ADC Comparator

Timer0 Timer1 2/4/6/8/10(3)/12(3) 3/5/7(3) CTMU 12-Bit 1/2/3
CCP ECCP LCD

4/5/6/7/8/9(3)/10(3) 1/2/3 EUSART1 EUSART2 RTCC MSSP1/2
Driver
Note 1: See Table 1-3 for I/O port pin descriptions.

2: RA6 and RA7 are only available as digital I/O in select oscillator modes. See Section 3.0 “Oscillator Configurations” for
more information.
3: Unimplemented in the PIC18F85K90.

PIC18F87K90 FAMILY
TABLE 1-3: PIC18F6XK90 PINOUT I/O DESCRIPTIONS
Pin Number Pin Buffer
Pin Name Description
QFN/TQFP Type Type
MCLR/RG5 7 Master Clear (input) or programming voltage (input).

MCLR I ST This pin is an active-low Reset to the device.
RG5 I ST General purpose, input only pin.
OSC1/CLKI/RA7 39 Oscillator crystal or external clock input.
OSC1 I CMOS Oscillator crystal input.
CLKI I CMOS External clock source input. Always associated
with pin function, OSC1. (See related OSC1/CLKI,
OSC2/CLKO pins.)
RA7 I/O TTL General purpose I/O pin.
OSC2/CLKO/RA6 40 Oscillator crystal or clock output.
OSC2 O — Oscillator crystal output. Connects to crystal or
resonator in Crystal Oscillator mode.
CLKO O — In certain oscillator modes, OSC2 pin outputs CLKO,
which has 1/4 the frequency of OSC1 and denotes the
instruction cycle rate.
Legend: TTL = TTL compatible input CMOS = CMOS compatible input or output
ST = Schmitt Trigger input with CMOS levels Analog = Analog input
I = Input O = Output
P = Power OD = Open-Drain (no P diode to VDD)
I2C™ = I2C/SMBus
Note 1: Default assignment for ECCP2 when the CCP2MX Configuration bit is set.
2: Alternate assignment for ECCP2 when the CCP2MX Configuration bit is cleared.
3: Not available on PIC18F65K90 and PIC18F85K90 devices.

PIC18F87K90 FAMILY
TABLE 1-3: PIC18F6XK90 PINOUT I/O DESCRIPTIONS (CONTINUED)
QFN/TQFP Type Type
PORTA is a bidirectional I/O port.

RA0/AN0/ULPWU 24
RA0 I/O TTL Digital I/O.
AN0 I Analog Analog Input 0.
ULPWU I Analog Ultra Low-Power Wake-up (ULPW) input.
RA1/AN1/SEG18 23
SEG18 O Analog SEG18 output for LCD.
RA2/AN2/VREF- 22
VREF- I Analog A/D reference voltage (low) input.
RA3/AN3/VREF+ 21
VREF+ I Analog A/D reference voltage (high) input.
RA4/T0CKI/SEG14 28
RA4 I/O ST Digital I/O.
T0CKI I ST Timer0 external clock input.
RA5/AN4/SEG15/T1CKI/ 27
T3G/HLVDIN
T1CKI I ST Timer1 clock input.
T3G I ST Timer3 external clock gate input.
HLVDIN I Analog High/Low-Voltage Detect (HLVD) input.
RA6 See the OSC2/CLKO/RA6 pin.
RA7 See the OSC1/CLKI/RA7 pin.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
QFN/TQFP Type Type
PORTB is a bidirectional I/O port. PORTB can be software

programmed for internal weak pull-ups on all inputs.
RB0/INT0/SEG30/FLTO 48
RB0 I/O TTL Digital I/O.
INT0 I ST External Interrupt 0.
FLTO I ST Enhanced PWM Fault input for ECCP1/2/3.
RB1/INT1/SEG8 47
RB2/INT2/SEG9/CTED1 46
CTED1 I ST CTMU Edge 1 input.
RB3/INT3/SEG10/CTED2/ 45
ECCP2/P2A
ECCP2 I/O ST Capture 2 input/Compare 2 output/PWM2.
P2A O — Enhanced PWM2 Output A.
RB4/KBI0/SEG11 44
KBI0 I TTL Interrupt-on-change pin.
RB5/KBI1/SEG29/T3CKI/ 43
T1G
RB6/KBI2/PGC 42
PGC I/O ST In-Circuit Debugger and ICSP™ programming clock pin.
RB7/KBI3/PGD 37
PGD I/O ST In-Circuit Debugger and ICSP programming data pin.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
QFN/TQFP Type Type
PORTC is a bidirectional I/O port.

RC0/SOSCO/SCLKI 30
RC0 I/O ST Digital I/O.
SOSCO O — SOSC oscillator output.
SCLKI I ST Digital SOSC input.
RC1/SOSCI/ECCP2/P2A/ 29
SEG32
SOSCI I CMOS SOSC oscillator input.
ECCP2(1) I/O ST Capture 2 input/Compare 2 output/PWM2 output.
RC2/ECCP1/P1A/SEG13 33
ECCP1 I/O ST Capture 1 input/Compare 1 output/PWM1 output.
RC3/SCK1/SCL1/SEG17 34
SCK1 I/O ST Synchronous serial clock input/output for SPI mode.
SCL1 I/O I2C Synchronous serial clock input/output for I2C™ mode.
RC4/SDI1/SDA1/SEG16 35
SDI1 I ST SPI data in.
SDA1 I/O I2C I2C data I/O.
RC5/SDO1/SEG12 36
SDO1 O — SPI data out.
RC6/TX1/CK1/SEG27 31
TX1 O — EUSART asynchronous transmit.
CK1 I/O ST EUSART synchronous clock (see related RX1/DT1).
RC7/RX1/DT1/SEG28 32
RX1 I ST EUSART asynchronous receive.
DT1 I/O ST EUSART synchronous data (see related TX1/CK1).
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
QFN/TQFP Type Type
PORTD is a bidirectional I/O port.

RD0/SEG0/CTPLS 58
RD0 I/O ST Digital I/O.
CTPLS O — CTMU pulse generator output.
RD1/SEG1/T5CKI/T7G 55
RD2/SEG2 54
RD3/SEG3 53
RD4/SEG4/SDO2 52
RD5/SEG5/SDI2/SDA2 51
SDA2 I/O I2C I2C™ data I/O.
RD6/SEG6/SCK2/SCL2 50
SCK2 I/O ST Synchronous serial clock.
SCL2 I/O I2C Synchronous serial clock for I2C mode.
RD7/SEG7/SS2 49
SS2 I TTL SPI slave select input.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
QFN/TQFP Type Type
PORTE is a bidirectional I/O port.

RE0/LCDBIAS1/P2D 2
RE0 I/O ST Digital I/O.
LCDBIAS1 I Analog BIAS1 input for LCD.
P2D O — EECP2 PWM Output D.
RE1/LCDBIAS2/P2C 1
P2C O — ECCP2 PWM Output C.
RE2/LCDBIAS3/P2B/ 64
CCP10
P2B O — ECCP2 PWM Output B.
CCP10(3) I/O S/T Capture 10 input/Compare 10 output/PWM10 output.
RE3/COM0/P3C/CCP9/ 63
REFO
COM0 O Analog COM0 output for LCD.
CCP9(3) I/O S/T Capture 9 input/Compare 9 output/PWM9 output.
REFO O — Reference clock out.
RE4/COM1/P3B/CCP8 62
CCP8 I/O S/T Capture 8 input/Compare 8 output/PWM8 output.
RE5/COM2/P1C/CCP7 61
RE7/ECCP2/SEG31/P2A 59
SEG31 O Analog SEG31 Output for LCD.
P2A O — ECCP2 PWM Output A.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
QFN/TQFP Type Type
PORTF is a bidirectional I/O port.

RF1/AN6/C2OUT/SEG19/ 17
CTDIN
RF1 I/O ST Digital I/O.
C2OUT O — Comparator 2 output.
CTDIN I ST CTMU pulse delay input.
RF2/AN7/C1OUT/SEG20 16
RF3/AN8/SEG21/C2INB/ 15
CTMUI
C2INB I Analog Comparator 2 Input B.
CTMUI O — CTMU pulse generator charger for the C2INB
comparator input.
RF4/AN9/SEG22/C2INA 14
SEG22 O Analog SEG22 output for LCD
C2INA I Analog Comparator 2 Input A.
RF5/AN10/CVREF/ 13
SEG23/C1INB
CVREF O Analog Comparator reference voltage output.
SEG24 O Analog SEG24 output for LCD
RF7/AN5/SS1/SEG25 11
AN5 O AnalogT Analog Input 5.
SS1 I TL SPI1 slave select input.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
QFN/TQFP Type Type
PORTG is a bidirectional I/O port.

RG0/ECCP3/P3A 3
RG0 I/O ST Digital I/O.
RG1/TX2/CK2/AN19/ 4
C3OUT
TX2 O — USART asynchronous transmit.
CK2 I/O ST USART synchronous clock (see related RX2/DT2).
RG2/RX2/DT2/AN18/ 5
C3INA
RG3/CCP4/AN17/P3D/ 6
C3INB
P3D O — ECCP3 PWM Output D.
RG4/SEG26/RTCC/ 8
T7CKI/T5G/CCP5/AN16/
P1D/C3INC
RTCC O — RTCC output
T7CKI(3) I ST Timer7 clock input.
CCP5 I/O ST Capture 5 input/Compare 5 output/PWM5 output.
C3INC I Analog Comparator 3 Input C.
RG5 7 See the MCLR/RG5 pin.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
QFN/TQFP Type Type
VSS 9, 25, 41, 56 P — Ground reference for logic and I/O pins.
VDD 26, 38, 57 P — Positive supply for logic and I/O pins.
AVSS 20 P — Ground reference for analog modules.
AVDD 19 P — Positive supply for analog modules.
ENVREG 18 I ST Enable for on-chip voltage regulator.
VDDCORE/VCAP 10 Core logic power or external filter capacitor connection.
VDDCORE
VCAP P — External filter capacitor connection (regulator
enabled/disabled).
I2C™ = I2C/SMBus
TABLE 1-4: PIC18F8XK90 PINOUT I/O DESCRIPTIONS

TQFP Type Type
MCLR/RG5 9 Master Clear (input) or programming voltage (input).

I ST This pin is an active-low Reset to the device.
RG5 I ST General purpose, input only pin.
OSC1/CLKI/RA7 49 Oscillator crystal or external clock input.
OSC1 I CMOS Oscillator crystal input.
CLKI I CMOS External clock source input. Always associated
with pin function, OSC1. (See related OSC1/CLKI,
OSC2/CLKO pins.)
OSC2/CLKO/RA6 50 Oscillator crystal or clock output.
OSC2 O — Oscillator crystal output. Connects to crystal or
resonator in Crystal Oscillator mode.
CLKO O — In certain oscillator modes, OSC2 pin outputs CLKO,
which has 1/4 the frequency of OSC1 and denotes the
instruction cycle rate.
I2C™ = I2C/SMBus
4: The CCP6, CCP7, CCP8 and CCP9 pin placement depends on the ECCPMX Configuration bit setting.

PIC18F87K90 FAMILY
TQFP Type Type
PORTA is a bidirectional I/O port.

RA0/AN0/ULPWU 30
ULPWU I Analog Ultra low-power wake-up input.
RA1/AN1/SEG18 29
RA2/AN2/VREF- 28
VREF- I Analog A/D reference voltage (low) input.
RA3/AN3/VREF+ 27
VREF+ I Analog A/D reference voltage (high) input.
RA4/T0CKI/SEG14 34
RA4 I/O ST Digital I/O.
T0CKI I ST Timer0 external clock input.
RA5/AN4/SEG15/T1CKI/ 33
T3G/HLVDIN
HLVDIN I Analog High/Low-Voltage Detect (HLVD) input.
RA6 See the OSC2/CLKO/RA6 pin.
RA7 See the OSC1/CLKI/RA7 pin.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
PORTB is a bidirectional I/O port. PORTB can be software

programmed for internal weak pull-ups on all inputs.
RB0/INT0/SEG30/FLT0 58
FLT0 I ST Enhanced PWM Fault input for ECCP1/2/3.
RB1/INT1/SEG8 57
RB2/INT2/SEG9/CTED1 56
RB3/INT3/SEG10/ 55
CTED2/ECCP2/P2A
P2A O ST Enhanced PWM2 Output A.
RB4/KBI0/SEG11 54
RB5/KBI1/SEG29/T3CKI/ 53
T1G
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
RB6/KBI2/PGC 52
PGC I/O ST In-Circuit Debugger and ICSP™ programming clock pin.
RB7/KBI3/PGD 47
PGD I/O ST In-Circuit Debugger and ICSP programming data pin.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
PORTC is a bidirectional I/O port.

RC0/SOSCO/SCKLI 36
SOSCO O — SOSC oscillator output.
SCKLI I ST Digital SOSC input.
RC1/SOSCI/ECCP2/ 35
SEG32/P2A
SOSCI I CMOS SOSC oscillator input.
RC2/ECCP1/P1A/SEG13 43
RC3/SCK1/SCL1/SEG17 44
SCL1 I/O ST Synchronous serial clock input/output for I2C™ mode.
RC4/SDI1/SDA1/SEG16 45
SDA1 I/O ST I2C data I/O.
RC5/SDO1/SEG12 46
RC6/TX1/CK1/SEG27 37
RC7/RX1/DT1/SEG28 38
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
PORTD is a bidirectional I/O port.

RD0/SEG0/CTPLS 72
CTPLS O ST CTMU pulse generator output.
RD1/SEG1/T5CKI/T7G 69
RD2/SEG2 68
RD3/SEG3 67
RD4/SEG4/SDO2 66
RD5/SEG5/SDI2/SDA2 65
SDA2 I/O I2C I2C™ data in.
RD6/SEG6/SCK2/SCL2 64
SCL2 I/O I2C Synchronous serial clock input/output for I2C™ mode.
RD7/SEG7/SS2 63
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
PORTE is a bidirectional I/O port.

RE0/LCDBIAS1/P2D 4
RE1/LCDBIAS2/P2C 3
RE2/LCDBIAS3/P2B/ 78
CCP10
P2B O ST ECCP2 PWM Output B.
CCP10(3) I/O ST Capture 10 input/Compare 10 output/PWM10 output.
RE3/COM0/P3C/CCP9/ 77
REFO
CCP9(3,4) I/O S/T Capture 9 input/Compare 9 output/PWM9 output.
REFO O — Reference clock out.
RE5/COM2/P1C/CCP7 75
RE7/ECCP2/P2A/SEG31 73
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
PORTF is a bidirectional I/O port.

RF1/AN6/C2OUT/SEG19/ 23
CTDIN
CTDIN I ST CTMU pulse delay input.
RF2/AN7/C1OUT/ 18
SEG20/CTMUI
CTMUI O — CTMU pulse generator charger for the C2INB
comparator input.
RF3/AN8/SEG21/C2INB 17
RF5/AN10/CVREF/ 15
SEG23/C1INB
CVREF O Analog Comparator reference voltage output.
RF7/AN5/SS1/SEG25 13
AN5 O Analog Analog Input 5.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
PORTG is a bidirectional I/O port.

RG0/ECCP3/P3A 5
RG1/TX2/CK2/AN19/ 6
C3OUT
RG2/RX2/DT2/AN18/ 7
C3INA
RG3/CCP4/AN17/P3D/ 8
C3INB
RG4/SEG26/RTCC/ 10
T7CKI/T5G/CCP5/AN16/
P1D/C3INC
RTCC O — RTCC output.
T7CKI(3) I ST Timer7 clock input.
RG5 9 See the MCLR/RG5 pin.
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
PORTH is a bidirectional I/O port.

RH0/SEG47/AN23 79
RH0 I/O ST Digital I/O.
RH1/SEG46/AN22 80
RH2/SEG45/AN21 1
RH3/SEG44/AN20 2
RH4/SEG40/CCP9/P3C/ 22
AN12/C2INC
CCP9(3,4) I/O ST Capture 9 input/Compare 9 output/PWM9 output.
RH5/SEG41/CCP8/P3B/ 21
AN13/C2IND
C2IND I Analog Comparator 1 Input D.
RH6/SEG42/CCP7/P1C/ 20
AN14/C1INC
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
RH7/SEG43/CCP6/P1B/ 19
AN15
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
TQFP Type Type
PORTJ is a bidirectional I/O port.

RJ0 62 I/O ST Digital I/O.
RJ1/SEG33 61
RJ1 I/O ST Digital I/O.
RJ2/SEG34 60
RJ3/SEG35 59
RJ4/SEG39 39
RJ5/SEG38 40
RJ5 I/O ST Digital I/O
RJ6/SEG37 41
RJ7/SEG36 42
VSS 11, 31, 51, 70 P — Ground reference for logic and I/O pins.
VDD 32, 48, 71 P — Positive supply for logic and I/O pins.
AVSS 26 P — Ground reference for analog modules.
AVDD 25 P — Positive supply for analog modules.
ENVREG 24 I ST Enable for on-chip voltage regulator.
VDDCORE/VCAP 12 Core logic power or external filter capacitor connection.
VDDCORE
VCAP P — External filter capacitor connection (regulator
enabled/disabled).
I2C™ = I2C/SMBus

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
2.0 GUIDELINES FOR GETTING FIGURE 2-1: RECOMMENDED
STARTED WITH PIC18FXXKXX MINIMUM CONNECTIONS
MICROCONTROLLERS
C2(2)
2.1 Basic Connection Requirements VDD
Getting started with the PIC18F87K90 family family of
VDD
VSS
R1 (1) (1)
8-bit microcontrollers requires attention to a minimal R2
set of device pin connections before proceeding with ENVREG
MCLR
development. VCAP/VDDCORE
C1
The following pins must always be connected: C7(2)
PIC18FXXKXX
• All VDD and VSS pins
VDD
(see Section 2.2 “Power Supply Pins”) VSS
C6(2) C3(2)
• All AVDD and AVSS pins, regardless of whether or VSS
VDD
not the analog device features are used
AVDD
AVSS
VDD
VSS
(see Section 2.2 “Power Supply Pins”)
• MCLR pin
(see Section 2.3 “Master Clear (MCLR) Pin”)
C5(2) C4(2)
• ENVREG (if implemented) and VCAP/VDDCORE pins
(see Section 2.4 “Voltage Regulator Pins
(ENVREG and VCAP/VDDCORE)”)
Key (all values are recommendations):
These pins must also be connected if they are being C1 through C6: 0.1 F, 20V ceramic
used in the end application:
R1: 10 kΩ
• PGC/PGD pins used for In-Circuit Serial R2: 100Ω to 470Ω
Programming™ (ICSP™) and debugging purposes
Note 1: See Section 2.4 “Voltage Regulator Pins
(see Section 2.5 “ICSP Pins”) (ENVREG and VCAP/VDDCORE)” for
• OSCI and OSCO pins when an external oscillator explanation of ENVREG pin connections.
source is used 2: The example shown is for a PIC18F device
(see Section 2.6 “External Oscillator Pins”) with five VDD/VSS and AVDD/AVSS pairs.
Other devices may have more or less pairs;
Additionally, the following pins may be required:
adjust the number of decoupling capacitors
• VREF+/VREF- pins are used when external voltage appropriately.
reference for analog modules is implemented
Note: The AVDD and AVSS pins must always be
connected, regardless of whether any of
the analog modules are being used.
The minimum mandatory connections are shown in
Figure 2-1.

PIC18F87K90 FAMILY
2.2 Power Supply Pins 2.3 Master Clear (MCLR) Pin
2.2.1 DECOUPLING CAPACITORS The MCLR pin provides two specific device
functions: Device Reset, and Device Programming
The use of decoupling capacitors on every pair of and Debugging. If programming and debugging are
power supply pins, such as VDD, VSS, AVDD and not required in the end application, a direct
AVSS, is required. connection to VDD may be all that is required. The
Consider the following criteria when using decoupling addition of other components, to help increase the
capacitors: application’s resistance to spurious Resets from
voltage sags, may be beneficial. A typical
• Value and type of capacitor: A 0.1 F (100 nF),
configuration is shown in Figure 2-1. Other circuit
10-20V capacitor is recommended. The capacitor
designs may be implemented, depending on the
should be a low-ESR device, with a resonance
application’s requirements.
frequency in the range of 200 MHz and higher.
Ceramic capacitors are recommended. During programming and debugging, the resistance
• Placement on the printed circuit board: The and capacitance that can be added to the pin must
decoupling capacitors should be placed as close be considered. Device programmers and debuggers
to the pins as possible. It is recommended to drive the MCLR pin. Consequently, specific voltage
place the capacitors on the same side of the levels (VIH and VIL) and fast signal transitions must
board as the device. If space is constricted, the not be adversely affected. Therefore, specific values
capacitor can be placed on another layer on the of R1 and C1 will need to be adjusted based on the
PCB using a via; however, ensure that the trace application and PCB requirements. For example, it is
length from the pin to the capacitor is no greater recommended that the capacitor, C1, be isolated
than 0.25 inch (6 mm). from the MCLR pin during programming and
debugging operations by using a jumper (Figure 2-2).
• Handling high-frequency noise: If the board is
The jumper is replaced for normal run-time
experiencing high-frequency noise (upward of
operations.
tens of MHz), add a second ceramic type capaci-
tor in parallel to the above described decoupling Any components associated with the MCLR pin
capacitor. The value of the second capacitor can should be placed within 0.25 inch (6 mm) of the pin.
be in the range of 0.01 F to 0.001 F. Place this
second capacitor next to each primary decoupling FIGURE 2-2: EXAMPLE OF MCLR PIN
capacitor. In high-speed circuit designs, consider CONNECTIONS
implementing a decade pair of capacitances as
close to the power and ground pins as possible VDD
(e.g., 0.1 F in parallel with 0.001 F).
• Maximizing performance: On the board layout R1
from the power supply circuit, run the power and R2
return traces to the decoupling capacitors first, MCLR
and then to the device pins. This ensures that the PIC18FXXKXX
JP
decoupling capacitors are first in the power chain.
Equally important is to keep the trace length C1
between the capacitor and the power pins to a
minimum, thereby reducing PCB trace
inductance.
Note 1: R1  10 k is recommended. A suggested
2.2.2 TANK CAPACITORS starting value is 10 k. Ensure that the
On boards with power traces running longer than MCLR pin VIH and VIL specifications are met.
six inches in length, it is suggested to use a tank capac- 2: R2  470 will limit any current flowing into
itor for integrated circuits, including microcontrollers, to MCLR from the external capacitor, C, in the
supply a local power source. The value of the tank event of MCLR pin breakdown, due to
Electrostatic Discharge (ESD) or Electrical
capacitor should be determined based on the trace
Overstress (EOS). Ensure that the MCLR pin
resistance that connects the power supply source to
VIH and VIL specifications are met.
the device, and the maximum current drawn by the
device in the application. In other words, select the tank
capacitor so that it meets the acceptable voltage sag at
the device. Typical values range from 4.7 F to 47 F.

PIC18F87K90 FAMILY
2.4 Voltage Regulator Pins (ENVREG Some PIC18FXXKXX families, or some devices within
and VCAP/VDDCORE) a family, do not provide the option of enabling or
disabling the on-chip voltage regulator:
The on-chip voltage regulator enable pin, ENVREG,
• Some devices (with the name, PIC18LFXXKXX)
must always be connected directly to either a supply
permanently disable the voltage regulator.
voltage or to ground. Tying ENVREG to VDD enables
These devices do not have the ENVREG pin and
the regulator, while tying it to ground disables the
require a 0.1 F capacitor on the VCAP/VDDCORE
regulator. Refer to Section 28.3 “On-Chip Voltage
pin. The VDD level of these devices must comply
Regulator” for details on connecting and using the
with the “voltage regulator disabled” specification
on-chip regulator.
for Parameter D001, in Section 31.0 “Electrical
When the regulator is enabled, a low-ESR (< 5Ω) Characteristics”.
capacitor is required on the VCAP/VDDCORE pin to • Some devices permanently enable the voltage
stabilize the voltage regulator output voltage. The regulator.
VCAP/VDDCORE pin must not be connected to VDD and These devices also do not have the ENVREG pin.
must use a capacitor of 10 µF connected to ground. The The 10 F capacitor is still required on the
type can be ceramic or tantalum. Suitable examples of VCAP/VDDCORE pin.
capacitors are shown in Table 2-1. Capacitors with
equivalent specifications can be used.
FIGURE 2-3: FREQUENCY vs. ESR
Designers may use Figure 2-3 to evaluate ESR PERFORMANCE FOR
equivalence of candidate devices. SUGGESTED VCAP
It is recommended that the trace length not exceed 10
0.25 inch (6 mm). Refer to Section 31.0 “Electrical
Characteristics” for additional information.
1
When the regulator is disabled, the VCAP/VDDCORE pin
must be tied to a voltage supply at the VDDCORE level.
ESR ()
Refer to Section 31.0 “Electrical Characteristics” for 0.1

information on VDD and VDDCORE.
0.01
0.001
0.01 0.1 1 10 100 1000 10,000
Frequency (MHz)
Note: Typical data measurement at 25°C, 0V DC bias.
.
TABLE 2-1: SUITABLE CAPACITOR EQUIVALENTS
Nominal
Make Part # Base Tolerance Rated Voltage Temp. Range
Capacitance
TDK C3216X7R1C106K 10 µF ±10% 16V -55 to 125ºC
TDK C3216X5R1C106K 10 µF ±10% 16V -55 to 85ºC
Panasonic ECJ-3YX1C106K 10 µF ±10% 16V -55 to 125ºC
Panasonic ECJ-4YB1C106K 10 µF ±10% 16V -55 to 85ºC
Murata GRM32DR71C106KA01L 10 µF ±10% 16V -55 to 125ºC
Murata GRM31CR61C106KC31L 10 µF ±10% 16V -55 to 85ºC

PIC18F87K90 FAMILY
2.4.1 CONSIDERATIONS FOR CERAMIC FIGURE 2-4: DC BIAS VOLTAGE vs.
CAPACITORS CAPACITANCE
In recent years, large value, low-voltage, surface-mount CHARACTERISTICS
ceramic capacitors have become very cost effective in
sizes up to a few tens of microfarad. The low-ESR, small
Capacitance Change (%)

10
physical size and other properties make ceramic 0
capacitors very attractive in many types of applications. -10

16V Capacitor
-20
Ceramic capacitors are suitable for use with the inter- -30
-40
nal voltage regulator of this microcontroller. However, -50
10V Capacitor
some care is needed in selecting the capacitor to -60
-70
6.3V Capacitor
ensure that it maintains sufficient capacitance over the -80
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
intended operating range of the application.
DC Bias Voltage (VDC)
Typical low-cost, 10 F ceramic capacitors are available
in X5R, X7R and Y5V dielectric ratings (other types are
also available, but are less common). The initial toler- When selecting a ceramic capacitor to be used with the
ance specifications for these types of capacitors are internal voltage regulator, it is suggested to select a
often specified as ±10% to ±20% (X5R and X7R), or high-voltage rating, so that the operating voltage is a
-20%/+80% (Y5V). However, the effective capacitance small percentage of the maximum rated capacitor volt-
that these capacitors provide in an application circuit will age. For example, choose a ceramic capacitor rated at
also vary based on additional factors, such as the 16V for the 2.5V core voltage. Suggested capacitors
applied DC bias voltage and the temperature. The total are shown in Table 2-1.
in-circuit tolerance is, therefore, much wider than the
initial tolerance specification. 2.5 ICSP Pins
The X5R and X7R capacitors typically exhibit satisfac- The PGC and PGD pins are used for In-Circuit Serial
tory temperature stability (ex: ±15% over a wide Programming™ (ICSP™) and debugging purposes. It
temperature range, but consult the manufacturer’s data is recommended to keep the trace length between the
sheets for exact specifications). However, Y5V capaci- ICSP connector and the ICSP pins on the device as
tors typically have extreme temperature tolerance short as possible. If the ICSP connector is expected to
specifications of +22%/-82%. Due to the extreme experience an ESD event, a series resistor is recom-
temperature tolerance, a 10 F nominal rated Y5V type mended, with the value in the range of a few tens of
capacitor may not deliver enough total capacitance to ohms, not to exceed 100Ω.
meet minimum internal voltage regulator stability and
transient response requirements. Therefore, Y5V Pull-up resistors, series diodes and capacitors on the
capacitors are not recommended for use with the PGC and PGD pins are not recommended as they will
internal regulator if the application must operate over a interfere with the programmer/debugger communica-
wide temperature range. tions to the device. If such discrete components are an
application requirement, they should be removed from
In addition to temperature tolerance, the effective the circuit during programming and debugging. Alter-
capacitance of large value ceramic capacitors can vary natively, refer to the AC/DC characteristics and timing
substantially, based on the amount of DC voltage requirements information in the respective device
applied to the capacitor. This effect can be very signifi- Flash programming specification for information on
cant, but is often overlooked or is not always capacitive loading limits, and pin input voltage high
documented. (VIH) and input low (VIL) requirements.
A typical DC bias voltage vs. capacitance graph for For device emulation, ensure that the “Communication
X7R type and Y5V type capacitors is shown in Channel Select” (i.e., PGCx/PGDx pins), programmed
Figure 2-4. into the device, matches the physical connections for
the ICSP to the Microchip debugger/emulator tool.
For more information on available Microchip
development tools connection requirements, refer to
Section 30.0 “Development Support”.

PIC18F87K90 FAMILY
2.6 External Oscillator Pins FIGURE 2-5: SUGGESTED
PLACEMENT OF THE
Many microcontrollers have options for at least two
OSCILLATOR CIRCUIT
oscillators: a high-frequency primary oscillator and a
low-frequency secondary oscillator (refer to Single-Sided and In-Line Layouts:
Section 3.0 “Oscillator Configurations” for details).
Copper Pour Primary Oscillator
The oscillator circuit should be placed on the same (tied to ground) Crystal
side of the board as the device. Place the oscillator
DEVICE PINS
circuit close to the respective oscillator pins with no
more than 0.5 inch (12 mm) between the circuit
components and the pins. The load capacitors should
be placed next to the oscillator itself, on the same side Primary OSC1
Oscillator
of the board.
C1 ` OSC2
Use a grounded copper pour around the oscillator cir-
cuit to isolate it from surrounding circuits. The C2 GND
grounded copper pour should be routed directly to the `
MCU ground. Do not run any signal traces or power T1OSO
traces inside the ground pour. Also, if using a two-sided
T1OS I
board, avoid any traces on the other side of the board Timer1 Oscillator
where the crystal is placed. Crystal
`
Layout suggestions are shown in Figure 2-4. In-line
packages may be handled with a single-sided layout
that completely encompasses the oscillator pins. With T1 Oscillator: C1 T1 Oscillator: C2
fine-pitch packages, it is not always possible to com-
pletely surround the pins and components. A suitable
solution is to tie the broken guard sections to a mirrored Fine-Pitch (Dual-Sided) Layouts:
ground layer. In all cases, the guard trace(s) must be
returned to ground. Top Layer Copper Pour
(tied to ground)
In planning the application’s routing and I/O assign-
ments, ensure that adjacent port pins, and other Bottom Layer
signals in close proximity to the oscillator, are benign Copper Pour
(i.e., free of high frequencies, short rise and fall times, (tied to ground)
and other similar noise).

OSCO
For additional information and design guidance on
oscillator circuits, please refer to these Microchip
C2
Application Notes, available at the corporate web site
Oscillator
(www.microchip.com): GND Crystal
• AN826, “Crystal Oscillator Basics and Crystal C1
Selection for rfPIC™ and PICmicro® Devices”
• AN849, “Basic PICmicro® Oscillator Design” OSCI
• AN943, “Practical PICmicro® Oscillator Analysis
and Design”
• AN949, “Making Your Oscillator Work”
2.7 Unused I/Os DEVICE PINS
Unused I/O pins should be configured as outputs and

driven to a logic low state. Alternatively, connect a 1 kΩ
to 10 kΩ resistor to VSS on unused pins and drive the
output to logic low.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
3.0 OSCILLATOR To optimize power consumption when using EC/HS/
XT/LP/RC as the primary oscillator, the frequency input
CONFIGURATIONS
range can be configured to yield an optimized power
bias:
3.1 Oscillator Types
• Low-Power Bias – External frequency less than
The PIC18F87K90 family of devices can be operated in 160 kHz
the following oscillator modes:
• Medium Power Bias – External frequency
• EC External Clock, RA6 available between 160 kHz and 16 MHz
• ECIO External Clock, Clock Out RA6 • High-Power Bias – External frequency greater
(FOSC/4 on RA6) than 16 MHz
• HS High-Speed Crystal/Resonator All of these modes are selected by the user by
• XT Crystal/Resonator programming the OSC<3:0> Configuration bits
• LP Low-Power Crystal (CONFIG1H<3:0>). In addition, PIC18F87K90 family
• RC External Resistor/Capacitor, RA6 devices can switch between different clock sources,
available either under software control or under certain condi-
tions, automatically. This allows for additional power
• RCIO External Resistor/Capacitor, Clock
savings by managing device clock speed in real time
Out RA6 (FOSC/4 on RA6)
without resetting the application. The clock sources for
• INTIO2 Internal Oscillator with I/O on RA6 the PIC18F87K90 family of devices are shown in
and RA7 Figure 3-1.
• INTIO1 Internal Oscillator with FOSC/4 Output
For the HS and EC mode, there are additional power
on RA6 and I/O on RA7
modes of operation – depending on the frequency of
There is also an option for running the 4xPLL on any of operation.
the clock sources in the input frequency range of 4 to
HS1 is the Medium Power mode with a frequency
16 MHz.
range of 4 MHz to 16 MHz. HS2 is the High-Power
The PLL is enabled by setting the PLLCFG bit mode where the oscillator frequency can go from
(CONFIG1H<4>) or the PLLEN bit (OSCTUNE<6>). 16 MHz to 25 MHz. HS1 and HS2 are achieved by
For the EC and HS mode, the PLLEN (software) or setting the CONFIG1H<3:0> correctly. (For details, see
PLLCFG (CONFIG) bit can be used to enable the PLL. Register 28-2 on page 428.)
For the INTIOx modes (HF-INTOSC): EC mode has these modes of operation:
• Only the PLLEN can enable the PLL (PLLCFG is • EC1 – For low power with a frequency range up to
ignored). 160 kHz
• When the oscillator is configured for the internal • EC2 – Medium power with a frequency range of
oscillator (OSC<3:0> = 100x), the PLL can be 160 kHz to 16 MHz
enabled only when the HF-INTOSC frequency is • EC3 – High power with a frequency range of
8 or 16 MHz. 16 MHz to 64 MHz
When the RA6 and RA7 pins are not used for an oscil- EC1, EC2 and EC3 are achieved by setting the
lator function or CLKOUT function, they are available CONFIG1H<3:0> correctly. (For details, see
as general purpose I/Os. Register 28-2 on page 428.)
Table 3-1 shows the HS and EC modes’ frequency
range and OSC<3:0> settings.

PIC18F87K90 FAMILY
TABLE 3-1: HS, EC, XT, LP AND RC MODES: RANGES AND SETTINGS
Mode Frequency Range OSC<3:0> Setting
EC1 (low power) 1101
DC-160 kHz
(EC1 & EC1IO) 1100
EC2 (medium power) 1011
160 kHz-16 MHz
(EC2 & EC2IO) 1010
EC3 (high power) 0101
16 MHz-64 MHz
(EC3 & EC3IO) 0100
HS1 (medium power) 4 MHz-16 MHz 0011
HS2 (high power) 16 MHz-25 MHz 0010
XT 100 kHz-4 MHz 0001
LP 31.25 kHz 0000
RC (External) 0-4 MHz 011x
100x
INTIO 32 kHz-16 MHz
(and OSCCON, OSCCON2)
FIGURE 3-1: PIC18F87K90 FAMILY CLOCK DIAGRAM
SOSCO
SOSCI
4x PLL Peripherals
MUX
MUX
OSC2
MUX
CPU
OSC1 PLLEN
FOSC<3:0> and PLLCFG
IDLEN
16 MHz 16 MHz 111

8 MHz 8 MHz Clock Control SCS<1:0>
110
4 MHz 4 MHz
Postscaler
101
HF INTOSC 2 MHz 2 MHz
MUX
100 FOSC<3:0>
16 MHz to 1 MHz 1 MHz
011
31 kHz 500 kHz
500 kHz
010
250 kHz 250 kHz
001
31 kHz 31 kHz
000
MUX
500 kHz
Postscaler
IRCF<2:0>
MF INTOSC 250 kHz
MUX
500 kHz to
31 kHz
31 kHz
INTSRC
MFIOSEL
LF INTOSC 31 kHz
31 kHz

PIC18F87K90 FAMILY
3.2 Control Registers The OSCTUNE register (Register 3-3) controls the
tuning and operation of the internal oscillator block. It also
The OSCCON register (Register 3-1) controls the main implements the PLLEN bit which controls the operation of
aspects of the device clock’s operation. It selects the the Phase Locked Loop (PLL) (see Section 3.5.2 “PLL
oscillator type to be used, which of the power-managed Frequency Multiplier”).
modes to invoke and the output frequency of the
INTOSC source. It also provides status on the oscillators.
REGISTER 3-1: OSCCON: OSCILLATOR CONTROL REGISTER

R/W-0 R/W-1 R/W-1 R/W-0 R(1) R-0 R/W-0 R/W-0
(2) (2) (2) (4)
IDLEN IRCF2 IRCF1 IRCF0 OSTS HFIOFS SCS1 SCS0(4)
bit 7 bit 0
Legend:
R = Readable bit W = Writable bit U = Unimplemented bit, read as ‘0’
-n = Value at POR ‘1’ = Bit is set ‘0’ = Bit is cleared x = Bit is unknown
bit 7 IDLEN: Idle Enable bit

1 = Device enters an Idle mode when a SLEEP instruction is executed
0 = Device enters Sleep mode when a SLEEP instruction is executed
bit 6-4 IRCF<2:0>: Internal Oscillator Frequency Select bits(2)
111 = HF-INTOSC output frequency is used (16 MHz)
110 = HF-INTOSC/2 output frequency is used (8 MHz, default)
101 = HF-INTOSC/4 output frequency is used (4 MHz)
If INTSRC = 0 and MFIOSEL = 0:(3,5)
010 = HF-INTOSC/32 output frequency is used (500 kHz)
000 = LF-INTOSC output frequency is used (31.25 kHz)
000 = HF-INTOSC/512 output frequency is used (31.25 kHz)
010 = MF-INTOSC output frequency is used (500 kHz)
001 = MF-INTOSC/2 output frequency is used (250 kHz)
000 = LF-INTOSC output frequency is used (31.25 kHz)
010 = MF-INTOSC output frequency is used (500 kHz)
001 = MF-INTOSC/2 output frequency is used (250 kHz)
000 = MF-INTOSC/16 output frequency is used (31.25 kHz)
bit 3 OSTS: Oscillator Start-up Timer Time-out Status bit(1)
1 = Oscillator Start-up Timer (OST) time-out has expired: primary oscillator is running as defined by
OSC<3:0>
0 = Oscillator Start-up Timer (OST) time-out is running: primary oscillator is not ready; device is
running from an internal oscillator (HF-INTOSC, MF-INTOSC or LF-INTOSC)
Note 1: Reset state depends on the state of the IESO Configuration bit (CONFIG1H<7>).
2: Modifying these bits will cause an immediate clock frequency switch if the internal oscillator is providing
the device clocks.
3: Source selected by the INTSRC bit (OSCTUNE<7>).
4: Modifying these bits will cause an immediate clock source switch.
5: INTSRC = OSCTUNE<7> and MFIOSEL = OSCCON2<0>.

PIC18F87K90 FAMILY
REGISTER 3-1: OSCCON: OSCILLATOR CONTROL REGISTER (CONTINUED)
bit 2 HFIOFS: INTOSC Frequency Stable bit
1 = HF-INTOSC oscillator frequency is stable
0 = HF-INTOSC oscillator frequency is not stable
bit 1-0 SCS<1:0>: System Clock Select bits(4)
1x = Internal oscillator block (LF-INTOSC, MF-INTOSC or HF-INTOSC)
01 = SOSC oscillator
00 = Default primary oscillator (OSC1/OSC2 or HF-INTOSC with or without PLL; defined by the
OSC<3:0> Configuration bits, CONFIG1H<3:0>.)
Note 1: Reset state depends on the state of the IESO Configuration bit (CONFIG1H<7>).
2: Modifying these bits will cause an immediate clock frequency switch if the internal oscillator is providing
the device clocks.
3: Source selected by the INTSRC bit (OSCTUNE<7>).
4: Modifying these bits will cause an immediate clock source switch.
5: INTSRC = OSCTUNE<7> and MFIOSEL = OSCCON2<0>.
REGISTER 3-2: OSCCON2: OSCILLATOR CONTROL REGISTER 2

U-0 R-0 U-0 U-0 R/W-0 U-0 R-x R/W-0
— SOSCRUN — — SOSCGO — MFIOFS MFIOSEL
bit 7 bit 0
Legend:
bit 7 Unimplemented: Read as ‘0’

bit 6 SOSCRUN: SOSC Run Status bit
1 = System clock comes from a secondary SOSC
0 = System clock comes from an oscillator other than SOSC
bit 5-4 Unimplemented: Read as ‘0’
bit 3 SOSCGO: Oscillator Start Control bit
1 = Oscillator is running, even if no other sources are requesting it
0 = Oscillator is shut off if no other sources are requesting it (When the SOSC is selected to run from
a digital clock input, rather than an external crystal, this bit has no effect.)
bit 1 MFIOFS: MF-INTOSC Frequency Stable bit
1 = MF-INTOSC is stable
0 = MF-INTOSC is not stable
bit 0 MFIOSEL: MF-INTOSC Select bit
1 = MF-INTOSC is used in place of HF-INTOSC frequencies of 500 kHz, 250 kHz and 31.25 kHz
0 = MF-INTOSC is not used

PIC18F87K90 FAMILY
REGISTER 3-3: OSCTUNE: OSCILLATOR TUNING REGISTER
R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0
INTSRC PLLEN TUN5 TUN4 TUN3 TUN2 TUN1 TUN0
bit 7 bit 0
Legend:
bit 7 INTSRC: Internal Oscillator Low-Frequency Source Select bit

1 = 31.25 kHz device clock is derived from 16 MHz INTOSC source (divide-by-512 enabled, HF-INTOSC)
0 = 31 kHz device clock is derived from INTRC 31 kHz oscillator (LF-INTOSC)
bit 6 PLLEN: Frequency Multiplier PLL Enable bit
1 = PLL is enabled
0 = PLL is disabled
bit 5-0 TUN<5:0>: Fast RC Oscillator (INTOSC) Frequency Tuning bits
011111 = Maximum frequency
•
•
000001
000000 = Center frequency. Fast RC oscillator is running at the calibrated frequency.
111111
•
•
100000 = Minimum frequency
3.3 Clock Sources and mode. PIC18F87K90 family devices offer the SOSC
Oscillator Switching (Timer1/3/5/7) oscillator as a secondary oscillator
source. This oscillator, in all power-managed modes, is
Essentially, PIC18F87K90 family devices have these often the time base for functions, such as a Real-Time
independent clock sources: Clock (RTC).
• Primary oscillators The SOSCEN bit in the corresponding timer should be
• Secondary oscillators set correctly for the enabled SOSC. The
• Internal oscillator SOSCEL<1:0> bits (CONFIG1L<4:3>) decide the
The primary oscillators can be thought of as the main SOSC mode of operation:
device oscillators. These are any external oscillators • 11 = High-power SOSC circuit
connected to the OSC1 and OSC2 pins, and include • 10 = Digital (SCLKI) mode
the External Crystal and Resonator modes and the
• 01 = Low-power SOSC circuit
External Clock modes. If selected by the OSC<3:0>
Configuration bits (CONFIG1H<3:0>), the internal In addition to being a primary clock source in some
oscillator block may be considered a primary oscillator. circumstances, the internal oscillator is available as a
The internal oscillator block can be one of the following: power-managed mode clock source. The LF-INTOSC
source is also used as the clock source for several
• 31 kHz LF-INTRC source
special features, such as the WDT and Fail-Safe Clock
• 31 kHz to 500 kHz MF-INTOSC source Monitor. The internal oscillator block is discussed in
• 31 kHz to 16 MHz HF-INTOSC source more detail in Section 3.6 “Internal Oscillator
The particular mode is defined by the OSC Block”.
Configuration bits. The details of these modes are The PIC18F87K90 family includes features that allow
covered in Section 3.4 “External Oscillator Modes”. the device clock source to be switched from the main
The secondary oscillators are external clock oscillator, chosen by device configuration, to one of the
sources that are not connected to the OSC1 or OSC2 alternate clock sources. When an alternate clock
pin. These sources may continue to operate, even source is enabled, various power-managed operating
after the controller is placed in a power-managed modes are available.

PIC18F87K90 FAMILY
3.3.1 OSC1/OSC2 OSCILLATOR 3.3.2.1 System Clock Selection and Device
The OSC1/OSC2 oscillator block is used to provide the Resets
oscillator modes and frequency ranges: Since the SCS bits are cleared on all forms of Reset,
this means the primary oscillator, defined by the
Mode Design Operating Frequency OSC<3:0> Configuration bits, is used as the primary
LP 31.25-100 kHz clock source on device Resets. This could either be the
internal oscillator block by itself, or one of the other
XT 100 kHz to 4 MHz
primary clock source (HS, EC, XT, LP, External RC and
HS 4 MHz to 25 MHz PLL-enabled modes).
EC 0 to 64 MHz (external clock) In those cases when the internal oscillator block, with-
EXTRC 0 to 4 MHz (external RC) out PLL, is the default clock on Reset, the Fast RC
The crystal-based oscillators (XT, HS and LP) have a oscillator (INTOSC) will be used as the device clock
built-in start-up time. The operation of the EC and source. It will initially start at 8 MHz; the postscaler
EXTRC clocks is immediate. selection that corresponds to the Reset value of the
IRCF<2:0> bits (‘110’).
3.3.2 CLOCK SOURCE SELECTION Regardless of which primary oscillator is selected,
The System Clock Select bits, SCS<1:>0 INTRC will always be enabled on device power-up. It
(OSCCON2<1:0>), select the clock source. The avail- serves as the clock source until the device has loaded
able clock sources are the primary clock defined by the its configuration values from memory. It is at this point
OSC<3:0> Configuration bits, the secondary clock that the OSC Configuration bits are read and the
(SOSC oscillator) and the internal oscillator. The clock oscillator selection of the operational mode is made.
source changes after one or more of the bits is written Note that either the primary clock source or the internal
to, following a brief clock transition interval. oscillator will have two bit setting options for the possible
The OSTS (OSCCON<3>) and SOSCRUN values of the SCS<1:0> bits, at any given time.
(OSCCON2<6>) bits indicate which clock source is
currently providing the device clock. The OSTS bit 3.3.3 OSCILLATOR TRANSITIONS
indicates that the Oscillator Start-up Timer (OST) has PIC18F87K90 family devices contain circuitry to
timed out and the primary clock is providing the device prevent clock “glitches” when switching between clock
clock in primary clock modes. The SOSCRUN bit sources. A short pause in the device clock occurs dur-
indicates when the SOSC oscillator (from Timer1/3/5/7) ing the clock switch. The length of this pause is the sum
is providing the device clock in secondary clock modes. of two cycles of the old clock source and three to four
In power-managed modes, only one of these bits will cycles of the new clock source. This formula assumes
be set at any time. If neither of these bits is set, the that the new clock source is stable.
INTRC is providing the clock, or the internal oscillator Clock transitions are discussed in greater detail in
has just started and is not yet stable. Section 4.1.2 “Entering Power-Managed Modes”.
The IDLEN bit (OSCCON<7>) determines if the device
goes into Sleep mode or one of the Idle modes when 3.4 External Oscillator Modes
the SLEEP instruction is executed.
The use of the flag and control bits in the OSCCON 3.4.1 CRYSTAL OSCILLATOR/CERAMIC
register is discussed in more detail in Section 4.0 RESONATORS (HS MODES)
“Power-Managed Modes”. In HS or HSPLL Oscillator modes, a crystal or ceramic
resonator is connected to the OSC1 and OSC2 pins to
Note 1: The secondary oscillator must be enabled
establish oscillation. Figure 3-2 shows the pin
to select the secondary clock source. The
connections.
SOSC oscillator is enabled by setting the
SOSCGO bit in the OSCCON2 register The oscillator design requires the use of a crystal rated
(OSCCON<3>). If the SOSC oscillator is for parallel resonant operation.
not enabled, then any attempt to select a
Note: Use of a crystal rated for series resonant
secondary clock source when executing a
operation may give a frequency out of the
SLEEP instruction will be ignored.
crystal manufacturer’s specifications.
2: It is recommended that the secondary
oscillator be operating and stable before
executing the SLEEP instruction or a very
long delay may occur while the SOSC
oscillator starts.

PIC18F87K90 FAMILY
TABLE 3-2: CAPACITOR SELECTION FOR Note 1: Higher capacitance increases the stability
CERAMIC RESONATORS of the oscillator but also increases the
Typical Capacitor Values Used: start-up time.
2: Since each resonator/crystal has its
Mode Freq. OSC1 OSC2
own characteristics, the user should
HS 8.0 MHz 27 pF 27 pF consult the resonator/crystal manufac-
16.0 MHz 22 pF 22 pF turer for appropriate values of external
Capacitor values are for design guidance only. components.
Different capacitor values may be required to produce 3: Rs may be required to avoid overdriving
acceptable oscillator operation. The user should test crystals with low drive level specification.
the performance of the oscillator over the expected 4: Always verify oscillator performance over
VDD and temperature range for the application. Refer the VDD and temperature range that is
to the following application notes for oscillator-specific expected for the application.
information:
• AN588, “PIC® Microcontroller Oscillator Design FIGURE 3-2: CRYSTAL/CERAMIC
Guide” RESONATOR OPERATION
• AN826, “Crystal Oscillator Basics and Crystal (HS OR HSPLL
Selection for rfPIC® and PIC® Devices” CONFIGURATION)
• AN849, “Basic PIC® Oscillator Design” C1(1) OSC1
• AN943, “Practical PIC® Oscillator Analysis and
Design” To
Internal
• AN949, “Making Your Oscillator Work” XTAL Logic
RF(3)
See the notes following Table 3-3 for additional
OSC2 Sleep
information.
C2(1) RS(2) PIC18F87K90
TABLE 3-3: CAPACITOR SELECTION FOR Note 1: See Table 3-2 and Table 3-3 for initial values of
CRYSTAL OSCILLATOR C1 and C2.
2: A series resistor (RS) may be required for AT
Typical Capacitor Values
Crystal strip cut crystals.
Osc Type Tested:
Freq. 3: RF varies with the oscillator mode chosen.
C1 C2
HS 4 MHz 27 pF 27 pF 3.5 RC Oscillator
8 MHz 22 pF 22 pF
For timing-insensitive applications, the RC and RCIO
20 MHz 15 pF 15 pF Oscillator modes offer additional cost savings. The
Capacitor values are for design guidance only. actual oscillator frequency is a function of several
factors:
Different capacitor values may be required to produce
acceptable oscillator operation. The user should test • Supply voltage
the performance of the oscillator over the expected • Values of the external resistor (REXT) and
VDD and temperature range for the application. capacitor (CEXT)
Refer to the Microchip application notes cited in • Operating temperature – Given the same device,
Table 3-2 for oscillator-specific information. Also see operating voltage and temperature and
the notes following this table for additional component values, there will also be unit-to-unit
information. frequency variations. These are due to factors,
such as:
- Normal manufacturing variation
- Difference in lead frame capacitance
between package types (especially for low
CEXT values)
- Variations within the tolerance of limits of
REXT and CEXT

PIC18F87K90 FAMILY
In the RC Oscillator mode, the oscillator frequency FIGURE 3-5: EXTERNAL CLOCK
divided by 4 is available on the OSC2 pin. This signal INPUT OPERATION
may be used for test purposes or to synchronize other (EC CONFIGURATION)
logic. Figure 3-3 shows how the R/C combination is
connected.
Clock from OSC1/CLKI
FIGURE 3-3: RC OSCILLATOR MODE Ext. System PIC18F87K90
VDD OSC2/CLKO
FOSC/4
REXT
Internal
OSC1
Clock An external clock source may also be connected to the
OSC1 pin in the HS mode, as shown in Figure 3-6. In
CEXT
PIC18FXXXX this configuration, the divide-by-4 output on OSC2 is
VSS not available. Current consumption in this configuration
OSC2/CLKO will be somewhat higher than EC mode, as the internal
FOSC/4
oscillator’s feedback circuitry will be enabled (in EC
Recommended values: 3 k  REXT  100 k mode, the feedback circuit is disabled).
20 pF CEXT  300 pF
FIGURE 3-6: EXTERNAL CLOCK INPUT
The RCIO Oscillator mode (Figure 3-4) functions like
OPERATION (HS OSC
the RC mode, except that the OSC2 pin becomes an
CONFIGURATION)
additional general purpose I/O pin. The I/O pin
becomes bit 6 of PORTA (RA6).
Clock from OSC1
FIGURE 3-4: RCIO OSCILLATOR MODE Ext. System PIC18F87K90
VDD (HS Mode)
Open OSC2
REXT
OSC1 Internal
Clock
3.5.2 PLL FREQUENCY MULTIPLIER
CEXT
A Phase Lock Loop (PLL) circuit is provided as an
PIC18FXXXX
VSS option for users who want to use a lower frequency
RA6 I/O (OSC2) oscillator circuit, or to clock the device up to its highest
rated frequency from a crystal oscillator. This may be
Recommended values: 3 k  REXT  100 k useful for customers who are concerned with EMI due
20 pF CEXT  300 pF to high-frequency crystals, or users who require higher
clock speeds from an internal oscillator.
3.5.1 EXTERNAL CLOCK INPUT
(EC MODES) 3.5.2.1 HSPLL and ECPLL Modes
The EC and ECPLL Oscillator modes require an The HSPLL and ECPLL modes provide the ability to
external clock source to be connected to the OSC1 pin. selectively run the device at four times the external
There is no oscillator start-up time required after a oscillating source to produce frequencies up to 64 MHz.
Power-on Reset or after an exit from Sleep mode. The PLL is enabled by setting the PLLEN bit
In the EC Oscillator mode, the oscillator frequency, (OSCTUNE<6>) or the PLLCFG bit (CONFIG1H<4>).
divided by 4, is available on the OSC2 pin. This signal The PLLEN bit provides software control for the PLL,
may be used for test purposes or to synchronize other even if PLLCFG is set to ‘0’. The PLL is enabled only
logic. Figure 3-5 shows the pin connections for the EC when the HS or EC oscillator frequency is within the
Oscillator mode. 4 MHz to 16 MHz input range.
This enables additional flexibility for controlling the
application’s clock speed in software. The PLLEN
should be enabled in HS or EC Oscillator mode only if
the input frequency is in the range of 4 MHz-16 MHz.

PIC18F87K90 FAMILY
FIGURE 3-7: PLL BLOCK DIAGRAM For MF-INTOSC mode to provide a frequency range of
500 kHz to 31 kHz, INTSRC = 1 and MFIOSEL = 1.
PLLCFG (CONFIG1H<4>)
PLL Enable (OSCTUNE)
The postscaler (IRCF<2:0>), in this mode, provides the
frequency range of 31 kHz to 500 kHz.
The LF-INTRC can provide only 31 kHz if INTSRC = 0.
OSC2 The LF-INTRC provides 31 kHz and is enabled if it is
Phase selected as the device clock source. The mode is
HS or EC FIN Comparator
Mode
enabled automatically when any of the following is
OSC1 FOUT
enabled:
• Power-up Timer
Loop • Fail-Safe Clock Monitor
Filter
• Watchdog Timer
• Two-Speed Start-up
4 VCO These features are discussed in greater detail in
SYSCLK Section 28.0 “Special Features of the CPU”.
MUX The clock source frequency (HF-INTOSC, MF-INTOSC
or LF-INTRC direct) is selected by configuring the IRCF
bits of the OSCCON register, as well the INTSRC and
MFIOSEL bits. The default frequency on device Resets
3.5.2.2 PLL and HF-INTOSC is 8 MHz.
The PLL is available to the internal oscillator block
3.6.1 INTIO MODES
when the internal oscillator block is configured as the
primary clock source. In this configuration, the PLL is Using the internal oscillator as the clock source elimi-
enabled in software and generates a clock output of up nates the need for up to two external oscillator pins,
to 64 MHz. which can then be used for digital I/O. Two distinct
oscillator configurations, which are determined by the
The operation of INTOSC with the PLL is described in OSC Configuration bits, are available:
Section 3.6.2 “INTPLL Modes”. Care should be taken • In INTIO1 mode, the OSC2 pin (RA6) outputs
that the PLL is enabled only if the HF-INTOSC FOSC/4, while OSC1 functions as RA7 (see
postscaler is configured for 8 MHz or 16 MHz. Figure 3-8) for digital input and output.
• In INTIO2 mode, OSC1 functions as RA7 and
3.6 Internal Oscillator Block OSC2 functions as RA6 (see Figure 3-9). Both
The PIC18F87K90 family of devices includes an are available as digital input and output ports.
internal oscillator block which generates two different
clock signals. Either clock can be used as the micro- FIGURE 3-8: INTIO1 OSCILLATOR MODE
controller’s clock source, which may eliminate the need
for an external oscillator circuit on the OSC1 and/or
OSC2 pins. RA7 I/O (OSC1)
The internal oscillator consists of three blocks, depend- PIC18F87K90

ing on the frequency of operation. They are FOSC/4 OSC2
HF-INTOSC, MF-INTOSC and LF-INTRC.
In HF-INTOSC mode, the internal oscillator can provide
a frequency, ranging from 31 kHz to 16 MHz, with the
postscaler deciding the selected frequency FIGURE 3-9: INTIO2 OSCILLATOR MODE
(IRCF<2:0>).
The INTSRC bit (OSCTUNE<7>) and MFIOSEL bit RA7 I/O (OSC1)
(OSCCON2<0>) also decide which INTOSC provides
PIC18F87K90
the lower frequency (500 kHz to 31 kHz). For the
HF-INTOSC to provide these frequencies, INTSRC = 1 RA6 I/O (OSC2)
and MFI0SEL = 0.
In HF-INTOSC, the postscaler (IRCF<2:0>) provides the
frequency range of 31 kHz to 16 MHz. If HF-INTOSC is
used with the PLL, the input frequency to the PLL should
be 8 MHz or 16 MHz (IRCF<2:0> = 111, 110).

PIC18F87K90 FAMILY
3.6.2 INTPLL MODES Tuning INTOSC requires knowing when to make the
adjustment, in which direction it should be made, and in
The 4x Phase Locked Loop (PLL) can be used with the
some cases, how large a change is needed. Three
HF-INTOSC to produce faster device clock speeds
compensation techniques are shown here.
than are normally possible with the internal oscillator
sources. When enabled, the PLL produces a clock
3.6.4.1 Compensating with the EUSART
speed of 32 MHz or 64 MHz.
An adjustment may be required when the EUSART
PLL operation is controlled through software. The con-
begins to generate framing errors or receives data with
trol bit, PLLEN (OSCTUNE<6>) is used to enable or
errors while in Asynchronous mode. Framing errors
disable its operation. Additionally, the PLL will only func-
indicate that the device clock frequency is too high. To
tion when the selected HF-INTOSC frequency is either
adjust for this, decrement the value in OSCTUNE to
8 MHz or 16 MHz (OSCCON<6:4> = 111 or 110).
reduce the clock frequency. On the other hand, errors
Like the INTIO modes, there are two distinct INTPLL in data may suggest that the clock speed is too low. To
modes available: compensate, increment OSCTUNE to increase the
• In INTPLL1 mode, the OSC2 pin outputs FOSC/4, clock frequency.
while OSC1 functions as RA7 for digital input and
output. Externally, this is identical in appearance 3.6.4.2 Compensating with the Timers
to INTIO1 (Figure 3-8). This technique compares device clock speed to some
• In INTPLL2 mode, OSC1 functions as RA7 and reference clock. Two timers may be used; one timer is
OSC2 functions as RA6, both for digital input and clocked by the peripheral clock, while the other is
output. Externally, this is identical to INTIO2 clocked by a fixed reference source, such as the SOSC
(Figure 3-9). oscillator.
Both timers are cleared, but the timer clocked by the
3.6.3 INTERNAL OSCILLATOR OUTPUT reference source generates interrupts. When an inter-
FREQUENCY AND TUNING rupt occurs, the internally clocked timer is read and
The internal oscillator block is calibrated at the factory both timers are cleared. If the internally clocked timer
to produce an INTOSC output frequency of 16 MHz. It value is much greater than expected, then the internal
can be adjusted in the user’s application by writing to oscillator block is running too fast. To adjust for this,
TUN<5:0> (OSCTUNE<5:0>) in the OSCTUNE decrement the OSCTUNE register.
register (Register 3-3).
3.6.4.3 Compensating with the CCP Module
When the OSCTUNE register is modified, the INTOSC
in Capture Mode
(HF-INTOSC and MF-INTOSC) frequency will begin
shifting to the new frequency. The oscillator will require A CCP module can use free-running Timer1 (or
some time to stabilize. Code execution continues Timer3), clocked by the internal oscillator block and an
during this shift and there is no indication that the shift external event with a known period (i.e., AC power
has occurred. frequency). The time of the first event is captured in the
CCPRxH:CCPRxL registers and is recorded for use
The LF-INTOSC oscillator operates independently of
later. When the second event causes a capture, the
the HF-INTOSC or the MF-INTOSC source. Any
time of the first event is subtracted from the time of the
changes in the HF-INTOSC or the MF-INTOSC source,
second event. Since the period of the external event is
across voltage and temperature, are not necessarily
known, the time difference between events can be
reflected by changes in LF-INTOSC or vice versa. The
calculated.
frequency of LF-INTOSC is not affected by OSCTUNE.
If the measured time is much greater than the
3.6.4 INTOSC FREQUENCY DRIFT calculated time, the internal oscillator block is running
too fast. To compensate, decrement the OSCTUNE
The INTOSC frequency may drift as VDD or tempera-
register. If the measured time is much less than the
ture changes and can affect the controller operation in
calculated time, the internal oscillator block is running
a variety of ways. It is possible to adjust the INTOSC
too slow. To compensate, increment the OSCTUNE
frequency by modifying the value in the OSCTUNE
register.
register. Depending on the device, this may have no
effect on the LF-INTOSC clock source frequency.

PIC18F87K90 FAMILY
3.7 Reference Clock Output The ROSSLP and ROSEL bits (REFOCON<5:4>) con-
trol the availability of the reference output during Sleep
In addition to the FOSC/4 clock output in certain oscilla- mode. The ROSEL bit determines if the oscillator on
tor modes, the device clock in the PIC18F87K90 family OSC1 and OSC2, or the current system clock source,
can also be configured to provide a reference clock out- is used for the reference clock output. The ROSSLP bit
put signal to a port pin. This feature is available in all determines if the reference source is available on RE3
oscillator configurations and allows the user to select a when the device is in Sleep mode.
greater range of clock submultiples to drive external
devices in the application. To use the reference clock output in Sleep mode, both
the ROSSLP and ROSEL bits must be set. The device
This reference clock output is controlled by the clock must also be configured for an EC or HS mode;
REFOCON register (Register 3-4). Setting the ROON otherwise, the oscillator on OSC1 and OSC2 will be
bit (REFOCON<7>) makes the clock signal available powered down when the device enters Sleep mode.
on the REFO (RE3) pin. The RODIV<3:0> bits enable
the selection of 16 different clock divider options. Clearing the ROSEL bit allows the reference output
frequency to change as the system clock changes
during any clock switches.
REGISTER 3-4: REFOCON: REFERENCE OSCILLATOR CONTROL REGISTER

R/W-0 U-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0
ROON — ROSSLP ROSEL(1) RODIV3 RODIV2 RODIV1 RODIV0
bit 7 bit 0
Legend:
bit 7 ROON: Reference Oscillator Output Enable bit

1 = Reference oscillator output is available on REFO pin
0 = Reference oscillator output is disabled
bit 5 ROSSLP: Reference Oscillator Output Stop in Sleep bit
1 = Reference oscillator continues to run in Sleep
0 = Reference oscillator is disabled in Sleep
bit 4 ROSEL: Reference Oscillator Source Select bit(1)
1 = Primary oscillator (EC or HS) is used as the base clock
0 = System clock is used as the base clock; base clock reflects any clock switching of the device
bit 3-0 RODIV<3:0>: Reference Oscillator Divisor Select bits
1111 = Base clock value divided by 32,768
1001 = Base clock value divided by 512
0000 = Base clock value
Note 1: For ROSEL (REFOCON<4>), the primary oscillator is only available when configured as a default via the
FOSC settings (regardless of whether the device is in Sleep mode).

PIC18F87K90 FAMILY
3.8 Effects of Power-Managed Modes Real-Time Clock (RTC). Other features may be operat-
on the Various Clock Sources ing that do not require a device clock source (i.e.,
MSSP slave, INTx pins and others). Peripherals that
When PRI_IDLE mode is selected, the designated pri- may add significant current consumption are listed in
mary oscillator continues to run without interruption. Section 31.2 “DC Characteristics: Power-Down and
For all other power-managed modes, the oscillator Supply Current PIC18F87K90 Family (Industrial/
using the OSC1 pin is disabled. The OSC1 pin (and Extended)”.
OSC2 pin if used by the oscillator) will stop oscillating.
In secondary clock modes (SEC_RUN and 3.9 Power-up Delays
SEC_IDLE), the SOSC oscillator is operating and
Power-up delays are controlled by two timers, so that
providing the device clock. The SOSC oscillator may
no external Reset circuitry is required for most applica-
also run in all power-managed modes if required to
tions. The delays ensure that the device is kept in
clock SOSC.
Reset until the device power supply is stable under nor-
In RC_RUN and RC_IDLE modes, the internal mal circumstances and the primary clock is operating
oscillator provides the device clock source. The 31 kHz and stable. For additional information on power-up
LF-INTOSC output can be used directly to provide the delays, see Section 5.6 “Power-up Timer (PWRT)”.
clock and may be enabled to support various special
The first timer is the Power-up Timer (PWRT), which
features, regardless of the power-managed mode (see
provides a fixed delay on a power-up time of about
Section 28.2 “Watchdog Timer (WDT)” through
64 ms (Parameter 33, Table 31-10); it is always
Section 28.5 “Fail-Safe Clock Monitor” for more
enabled.
information on WDT, Fail-Safe Clock Monitor and
Two-Speed Start-up). The second timer is the Oscillator Start-up Timer
(OST), intended to keep the chip in Reset until the
If the Sleep mode is selected, all clock sources are
crystal oscillator is stable (HS, XT or LP modes). The
stopped. Since all the transistor switching currents
OST does this by counting 1,024 oscillator cycles
have been stopped, Sleep mode achieves the lowest
before allowing the oscillator to clock the device.
current consumption of the device (only leakage
currents). There is a delay of interval, TCSD (Parameter 38,
Table 31-10), following POR, while the controller
Enabling any on-chip feature that will operate during
becomes ready to execute instructions.
Sleep will increase the current consumed during Sleep.
The INTOSC is required to support WDT operation.
The SOSC oscillator may be operating to support a
TABLE 3-4: OSC1 AND OSC2 PIN STATES IN SLEEP MODE

Oscillator Mode OSC1 Pin OSC2 Pin
EC, ECPLL Floating, pulled by external clock At logic low (clock/4 output)
HS, HSPLL Feedback inverter is disabled at quiescent Feedback inverter is disabled at quiescent
voltage level voltage level
INTOSC, INTPLL1/2 I/O pin, RA6, direction is controlled by I/O pin, RA6, direction is controlled by
TRISA<6> TRISA<7>
Note: See Section 5.0 “Reset” for time-outs due to Sleep and MCLR Reset.

PIC18F87K90 FAMILY
4.0 POWER-MANAGED MODES The IDLEN bit (OSCCON<7>) controls CPU clocking,
while the SCS<1:0> bits (OSCCON<1:0>) select the
The PIC18F87K90 family of devices offers a total of clock source. The individual modes, bit settings, clock
seven operating modes for more efficient power man- sources and affected modules are summarized in
agement. These modes provide a variety of options for Table 4-1.
selective power conservation in applications where
resources may be limited (such as battery-powered 4.1.1 CLOCK SOURCES
devices).
The SCS<1:0> bits select one of three clock sources
There are three categories of power-managed modes: for power-managed modes. Those sources are:
• Run modes • The primary clock, as defined by the OSC<3:0>
• Idle modes Configuration bits
• Sleep mode • The secondary clock (the SOSC oscillator)
There is an Ultra Low-Power Wake-up (ULPWU) for • The internal oscillator block (for LF-INTOSC
waking from the Sleep mode. modes)
These categories define which portions of the device 4.1.2 ENTERING POWER-MANAGED
are clocked, and sometimes, at what speed. The Run MODES
and Idle modes may use any of the three available
clock sources (primary, secondary or internal oscillator Switching from one power-managed mode to another
block). The Sleep mode does not use a clock source. begins by loading the OSCCON register. The
SCS<1:0> bits select the clock source and determine
The ULPWU mode, on the RA0 pin, enables a slow fall- which Run or Idle mode is used. Changing these bits
ing voltage to generate a wake-up, even from Sleep, causes an immediate switch to the new clock source,
without excess current consumption. (See Section 4.7 assuming that it is running. The switch may also be
“Ultra Low-Power Wake-up”.) subject to clock transition delays. These considerations
The power-managed modes include several power- are discussed in Section 4.1.3 “Clock Transitions
saving features offered on previous PIC® devices. One and Status Indicators” and subsequent sections.
is the clock switching feature, offered in other PIC18 Entering the power-managed Idle or Sleep modes is
devices. This feature allows the controller to use the triggered by the execution of a SLEEP instruction. The
SOSC oscillator instead of the primary one. Another actual mode that results depends on the status of the
power-saving feature is Sleep mode, offered by all PIC IDLEN bit.
devices, where all device clocks are stopped.
Depending on the current and impending mode, a
change to a power-managed mode does not always
4.1 Selecting Power-Managed Modes
require setting all of the previously discussed bits. Many
Selecting a power-managed mode requires two transitions can be done by changing the oscillator select
decisions: bits, or changing the IDLEN bit, prior to issuing a SLEEP
instruction. If the IDLEN bit is already configured as
• Will the CPU be clocked or not
desired, it may only be necessary to perform a SLEEP
• What will be the clock source instruction to switch to the desired mode.
TABLE 4-1: POWER-MANAGED MODES

OSCCON Bits Module Clocking
Mode Available Clock and Oscillator Source
IDLEN<7>(1) SCS<1:0> CPU Peripherals
Sleep 0 N/A Off Off None – All clocks are disabled
Primary – XT, LP, HS, EC, RC and PLL modes.
PRI_RUN N/A 00 Clocked Clocked
This is the normal, Full-Power Execution mode.
SEC_RUN N/A 01 Clocked Clocked Secondary – SOSC Oscillator
RC_RUN N/A 1x Clocked Clocked Internal oscillator block(2)
PRI_IDLE 1 00 Off Clocked Primary – LP, XT, HS, RC, EC
SEC_IDLE 1 01 Off Clocked Secondary – SOSC oscillator
RC_IDLE 1 1x Off Clocked Internal oscillator block(2)
Note 1: IDLEN reflects its value when the SLEEP instruction is executed.
2: Includes INTOSC (HF-INTOSC and MG-INTOSC) and INTOSC postscaler, as well as the LF-INTISC source.

PIC18F87K90 FAMILY
4.1.3 CLOCK TRANSITIONS AND STATUS 4.1.4 MULTIPLE SLEEP COMMANDS
INDICATORS The power-managed mode that is invoked with the
The length of the transition between clock sources is SLEEP instruction is determined by the setting of the
the sum of two cycles of the old clock source and IDLEN bit at the time the instruction is executed. If
three to four cycles of the new clock source. This for- another SLEEP instruction is executed, the device will
mula assumes that the new clock source is stable. enter the power-managed mode specified by IDLEN at
The HF-INTOSC and MF-INTOSC are termed as that time. If IDLEN has changed, the device will enter
INTOSC in this chapter. the new power-managed mode specified by the new
Three bits indicate the current clock source and its setting.
status, as shown in Table 4-2. The three bits are:
4.2 Run Modes
• OSTS (OSCCON<3>)
In the Run modes, clocks to both the core and
• HFIOFS (OSCCON<2>)
peripherals are active. The difference between these
• SOSCRUN (OSCCON2<6>)
modes is the clock source.
TABLE 4-2: SYSTEM CLOCK INDICATOR
4.2.1 PRI_RUN MODE
HFIOFS or The PRI_RUN mode is the normal, Full-Power Execu-
Main Clock Source OSTS SOSCRUN
MFIOFS tion mode of the microcontroller. This is also the default
Primary Oscillator 1 0 0 mode upon a device Reset, unless Two-Speed Start-up
is enabled. (For details, see Section 28.4 “Two-Speed
INTOSC (HF-INTOSC or
0 1 0 Start-up”.) In this mode, the OSTS bit is set. The
MF-INTOSC)
HFIOFS or MFIOFS bit may be set if the internal
Secondary Oscillator 0 0 1 oscillator block is the primary clock source. (See
MF-INTOSC or Section 3.2 “Control Registers”.)
HF-INTOSC as Primary 1 1 0
Clock Source 4.2.2 SEC_RUN MODE
LF-INTOSC is Running or The SEC_RUN mode is the compatible mode to the
0 0 0 “clock-switching” feature offered in other PIC18
INTOSC is Not Yet Stable
devices. In this mode, the CPU and peripherals are
When the OSTS bit is set, the primary clock is providing
clocked from the SOSC oscillator. This enables lower
the device clock. When the HFIOFS or MFIOFS bit is
power consumption while retaining a high-accuracy
set, the INTOSC output is providing a stable clock clock source.
source to a divider that actually drives the device clock.
When the SOSCRUN bit is set, the SOSC oscillator is SEC_RUN mode is entered by setting the SCS<1:0>
providing the clock. If none of these bits are set, either bits to ‘01’. The device clock source is switched to the
the LF-INTOSC clock source is clocking the device or SOSC oscillator (see Figure 4-1), the primary oscillator
the INTOSC source is not yet stable. is shut down, the SOSCRUN bit (OSCCON2<6>) is set
and the OSTS bit is cleared.
If the internal oscillator block is configured as the
primary clock source by the OSC<3:0> Configuration Note: The SOSC oscillator can be enabled by
bits (CONFIG1H<3:0>), then the OSTS and HFIOFS or setting the SOSCGO bit (OSCCON2<3>).
MFIOFS bits can be set when in PRI_RUN or If this bit is set, the clock switch to the
PRI_IDLE modes. This indicates that the primary clock SEC_RUN mode can switch immediately
(INTOSC output) is generating a stable output. once SCS<1:0> are set to ‘01’.
Entering another INTOSC power-managed mode at
the same frequency would clear the OSTS bit. On transitions from SEC_RUN mode to PRI_RUN
mode, the peripherals and CPU continue to be clocked
Note 1: Caution should be used when modifying from the SOSC oscillator while the primary clock is
a single IRCF bit. At a lower VDD, it is started. When the primary clock becomes ready, a
possible to select a higher clock speed clock switch back to the primary clock occurs (see
than is supportable by that VDD. Improper Figure 4-2). When the clock switch is complete, the
device operation may result if the VDD/ SOSCRUN bit is cleared, the OSTS bit is set and the
FOSC specifications are violated. primary clock is providing the clock. The IDLEN and
2: Executing a SLEEP instruction does not SCS bits are not affected by the wake-up and the
necessarily place the device into Sleep SOSC oscillator continues to run.
mode. It acts as the trigger to place the
controller into either the Sleep mode or
one of the Idle modes, depending on the
setting of the IDLEN bit.

PIC18F87K90 FAMILY
FIGURE 4-1: TRANSITION TIMING FOR ENTRY TO SEC_RUN MODE
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3
SOSCI 1 2 3 n-1 n
OSC1 Clock Transition(1)
CPU
Clock
Peripheral
Clock
Program
Counter PC PC + 2 PC + 4
Note 1: Clock transition typically occurs within 2-4 TOSC.
FIGURE 4-2: TRANSITION TIMING FROM SEC_RUN MODE TO PRI_RUN MODE (HSPLL)
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3
SOSC
OSC1
TOST(1) TPLL(1)
1 2 n-1 n
PLL Clock
Output
Clock
Transition(2)
CPU Clock
Peripheral
Clock
Program PC PC + 2 PC + 4
Counter
SCS<1:0> bits Changed OSTS bit Set
Note 1: TOST = 1024 TOSC; TPLL = 2 ms (approx). These intervals are not shown to scale.
2: Clock transition typically occurs within 2-4 TOSC.
4.2.3 RC_RUN MODE This mode is entered by setting the SCS1 bit to ‘1’. To
maintain software compatibility with future devices, it is
In RC_RUN mode, the CPU and peripherals are
recommended that the SCS0 bit also be cleared, even
clocked from the internal oscillator block using the
though the bit is ignored. When the clock source is
INTOSC multiplexer. In this mode, the primary clock is
switched to the INTOSC multiplexer (see Figure 4-3),
shut down. When using the LF-INTOSC source, this
the primary oscillator is shut down and the OSTS bit is
mode provides the best power conservation of all the
cleared. The IRCF bits may be modified at any time to
Run modes, while still executing code. It works well for
immediately change the clock speed.
user applications which are not highly timing-sensitive
or do not require high-speed clocks at all times. Note: Caution should be used when modifying a
If the primary clock source is the internal oscillator single IRCF bit. At a lower VDD, it is
block – either LF-INTOSC or INTOSC (MF-INTOSC or possible to select a higher clock speed
HF-INTOSC) – there are no distinguishable differences than is supportable by that VDD. Improper
between the PRI_RUN and RC_RUN modes during device operation may result if the VDD/
execution. Entering or exiting RC_RUN mode, FOSC specifications are violated.
however, causes a clock switch delay. Therefore, if the
primary clock source is the internal oscillator block,
using RC_RUN mode is not recommended.

PIC18F87K90 FAMILY
If the IRCF bits and the INTSRC bit are all clear, the If the IRCF bits are changed from all clear (thus,
INTOSC output (HF-INTOSC/MF-INTOSC) is not enabling the INTOSC output) or if INTSRC or
enabled, and the HFIOFS and MFIOFS bits will remain MFIOSEL is set, the HFIOFS or MFIOFS bit is set after
clear. There will be no indication of the current clock the INTOSC output becomes stable. For details, see
source. The LF-INTOSC source is providing the device Table 4-3.
clocks.
TABLE 4-3: INTERNAL OSCILLATOR FREQUENCY STABILITY BITS

IRCF<2:0> INTSRC MFIOSEL Status of MFIOFS or HFIOFS when INTOSC is Stable
000 0 x MFIOFS = 0, HFIOFS = 0 and clock source is LF-INTOSC
000 1 0 MFIOFS = 0, HFIOFS = 1 and clock source is HF-INTOSC
000 1 1 MFIOFS = 1, HFIOFS = 0 and clock source is MF-INTOSC
Non-Zero x 0 MFIOFS = 0, HFIOFS = 1 and clock source is HF-INTOSC
Non-Zero x 1 MFIOFS = 1, HFIOFS = 0 and clock source is MF-INTOSC
Clocks to the device continue while the INTOSC source On transitions from RC_RUN mode to PRI_RUN mode,
stabilizes after an interval of TIOBST (Parameter 39, the device continues to be clocked from the INTOSC
Table 31-10). multiplexer while the primary clock is started. When the
If the IRCF bits were previously at a non-zero value, or primary clock becomes ready, a clock switch to the
if INTSRC was set before setting SCS1, and the primary clock occurs (see Figure 4-4). When the clock
INTOSC source was already stable, the HFIOFS or switch is complete, the HFIOFS or MFIOFS bit is
MFIOFS bit will remain set. cleared, the OSTS bit is set and the primary clock is
providing the device clock. The IDLEN and SCS bits
are not affected by the switch. The LF-INTOSC source
will continue to run if either the WDT or the Fail-Safe
Clock Monitor is enabled.

PIC18F87K90 FAMILY
FIGURE 4-3: TRANSITION TIMING TO RC_RUN MODE
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3
LF-INTOSC 1 2 3 n-1 n
OSC1 Clock Transition(1)
CPU
Clock
Peripheral
Clock
Program
Counter PC PC + 2 PC + 4
Note 1: Clock transition typically occurs within 2-4 TOSC.
FIGURE 4-4: TRANSITION TIMING FROM RC_RUN MODE TO PRI_RUN MODE
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3
INTOSC
Multiplexer
OSC1
TOST(1) TPLL(1)
1 2 n-1 n
PLL Clock
Output
Clock
Transition(2)
CPU Clock
Peripheral
Clock
Program PC PC + 2 PC + 4
Counter
SCS<1:0> bits Changed OSTS bit Set
Note1: TOST = 1024 TOSC; TPLL = 2 ms (approx). These intervals are not shown to scale.

PIC18F87K90 FAMILY
4.3 Sleep Mode 4.4 Idle Modes
The power-managed Sleep mode in the PIC18F87K90 The Idle modes allow the controller’s CPU to be
family of devices is identical to the legacy Sleep mode selectively shut down while the peripherals continue to
offered in all other PIC devices. It is entered by clearing operate. Selecting a particular Idle mode allows users
the IDLEN bit (the default state on device Reset) and to further manage power consumption.
executing the SLEEP instruction. This shuts down the If the IDLEN bit is set to a ‘1’ when a SLEEP instruction is
selected oscillator (Figure 4-5). All clock source status executed, the peripherals will be clocked from the clock
bits are cleared. source selected using the SCS<1:0> bits. The CPU,
Entering Sleep mode from any other mode does not however, will not be clocked. The clock source status bits
require a clock switch. This is because no clocks are are not affected. This approach is a quick method to
needed once the controller has entered Sleep. If the switch from a given Run mode to its corresponding Idle
WDT is selected, the LF-INTOSC source will continue mode.
to operate. If the SOSC oscillator is enabled, it will also If the WDT is selected, the LF-INTOSC source will
continue to run. continue to operate. If the SOSC oscillator is enabled,
When a wake event occurs in Sleep mode (by interrupt, it will also continue to run.
Reset or WDT time-out), the device will not be clocked Since the CPU is not executing instructions, the only
until the clock source selected by the SCS<1:0> bits exits from any of the Idle modes are by interrupt, WDT
becomes ready (see Figure 4-6). Alternately, the device time-out or a Reset. When a wake event occurs, CPU
will be clocked from the internal oscillator block if either execution is delayed by an interval of TCSD
the Two-Speed Start-up or the Fail-Safe Clock Monitor is (Parameter 38, Table 31-10) while it becomes ready to
enabled (see Section 28.0 “Special Features of the execute code. When the CPU begins executing code,
CPU”). In either case, the OSTS bit is set when the it resumes with the same clock source for the current
primary clock is providing the device clocks. The IDLEN Idle mode. For example, when waking from RC_IDLE
and SCS bits are not affected by the wake-up. mode, the internal oscillator block will clock the CPU
and peripherals (in other words, RC_RUN mode). The
IDLEN and SCS bits are not affected by the wake-up.
While in any Idle mode or Sleep mode, a WDT time-
out will result in a WDT wake-up to the Run mode
currently specified by the SCS<1:0> bits.
FIGURE 4-5: TRANSITION TIMING FOR ENTRY TO SLEEP MODE

Q1 Q2 Q3 Q4 Q1
OSC1
CPU
Clock
Peripheral
Clock
Sleep
Program
Counter PC PC + 2
FIGURE 4-6: TRANSITION TIMING FOR WAKE FROM SLEEP (HSPLL)

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
OSC1
TOST(1) TPLL(1)
PLL Clock
Output
CPU Clock
Peripheral
Clock
Program PC PC + 2 PC + 4 PC + 6
Counter
Wake Event OSTS bit Set
Note1: TOST = 1024 TOSC; TPLL = 2 ms (approx). These intervals are not shown to scale.

PIC18F87K90 FAMILY
4.4.1 PRI_IDLE MODE 4.4.2 SEC_IDLE MODE
This mode is unique among the three low-power Idle In SEC_IDLE mode, the CPU is disabled but the
modes, in that it does not disable the primary device peripherals continue to be clocked from the SOSC
clock. For timing-sensitive applications, this allows for oscillator. This mode is entered from SEC_RUN by set-
the fastest resumption of device operation with its more ting the IDLEN bit and executing a SLEEP instruction. If
accurate, primary clock source, since the clock source the device is in another Run mode, set the IDLEN bit
does not have to “warm-up” or transition from another first, then set the SCS<1:0> bits to ‘01’ and execute
oscillator. SLEEP. When the clock source is switched to the SOSC
PRI_IDLE mode is entered from PRI_RUN mode by oscillator, the primary oscillator is shut down, the OSTS
setting the IDLEN bit and executing a SLEEP instruc- bit is cleared and the SOSCRUN bit is set.
tion. If the device is in another Run mode, set IDLEN When a wake event occurs, the peripherals continue to
first, then clear the SCS bits and execute SLEEP. be clocked from the SOSC oscillator. After an interval of
Although the CPU is disabled, the peripherals continue TCSD following the wake event, the CPU begins execut-
to be clocked from the primary clock source specified ing code being clocked by the SOSC oscillator. The
by the OSC<3:0> Configuration bits. The OSTS bit IDLEN and SCS bits are not affected by the wake-up and
remains set (see Figure 4-7). the SOSC oscillator continues to run (see Figure 4-8).
When a wake event occurs, the CPU is clocked from the
primary clock source. A delay of interval, TCSD
(Parameter 39, Table 31-10), is required between the
wake event and the start of code execution. This is
required to allow the CPU to become ready to execute
instructions. After the wake-up, the OSTS bit remains
set. The IDLEN and SCS bits are not affected by the
wake-up (see Figure 4-8).
FIGURE 4-7: TRANSITION TIMING FOR ENTRY TO IDLE MODE
Q1 Q2 Q3 Q4 Q1
OSC1
CPU Clock
Peripheral
Clock
Program PC PC + 2
Counter
FIGURE 4-8: TRANSITION TIMING FOR WAKE FROM IDLE TO RUN MODE
Q1 Q2 Q3 Q4
OSC1
TCSD
CPU Clock
Peripheral
Clock
Program PC
Counter
Wake Event

PIC18F87K90 FAMILY
4.4.3 RC_IDLE MODE 4.5 Selective Peripheral Module
In RC_IDLE mode, the CPU is disabled but the periph- Control
erals continue to be clocked from the internal oscillator
Idle mode allows users to substantially reduce power
block using the INTOSC multiplexer. This mode
consumption by stopping the CPU clock. Even so,
provides controllable power conservation during Idle
peripheral modules still remain clocked, and thus,
periods.
consume power. There may be cases where the
From RC_RUN, this mode is entered by setting the application needs what this mode does not provide: the
IDLEN bit and executing a SLEEP instruction. If the allocation of power resources to the CPU, processing
device is in another Run mode, first set IDLEN, then set with minimal power consumption from the peripherals.
the SCS1 bit and execute SLEEP. To maintain software
PIC18F87K90 family devices address this requirement
compatibility with future devices, it is recommended
by allowing peripheral modules to be selectively
that SCS0 also be cleared, though its value is ignored.
disabled, reducing or eliminating their power
The INTOSC multiplexer may be used to select a
consumption. This can be done with two control bits:
higher clock frequency by modifying the IRCF bits
before executing the SLEEP instruction. When the • Peripheral Enable bit, generically named XXXEN –
clock source is switched to the INTOSC multiplexer, the Located in the respective module’s main control
primary oscillator is shut down and the OSTS bit is register
cleared. • Peripheral Module Disable (PMD) bit, generically
If the IRCF bits are set to any non-zero value, or the named XXXMD – Located in one of the PMDx
INTSRC/MFIOSEL bit is set, the INTOSC output is Control registers (PMD0, PMD1, PMD2 or PMD3)
enabled. The HFIOFS/MFIOFS bits become set, after Disabling a module by clearing its XXXEN bit disables
the INTOSC output becomes stable after an interval of the module’s functionality, but leaves its registers
TIOBST (Parameter 38, Table 31-10). (For information available to be read and written to. This reduces power
on the HFIOFS/MFIOFS bits, see Table 4-3.) consumption, but not by as much as the second
Clocks to the peripherals continue while the INTOSC approach.
source stabilizes. The HFIOFS/MFIOFS bits will Most peripheral modules have an enable bit.
remain set if the IRCF bits were previously at a non- In contrast, setting the PMD bit for a module disables all
zero value or if INTSRC was set before the SLEEP clock sources to that module, reducing its power
instruction was executed and the INTOSC source was consumption to an absolute minimum. In this state, the
already stable. If the IRCF bits and INTSRC are all control and status registers associated with the
clear, the INTOSC output will not be enabled, the peripheral are also disabled, so writes to those registers
HFIOFS/MFIOFS bits will remain clear and there will be have no effect and read values are invalid. Many
no indication of the current clock source. peripheral modules have a corresponding PMD bit.
When a wake event occurs, the peripherals continue to There are four PMD registers in the PIC18F87K90 family
be clocked from the INTOSC multiplexer. After a delay devices: PMD0, PMD1, PMD2 and PMD3. These
of TCSD (Parameter 38, Table 31-10), following the registers have bits associated with each module for
wake event, the CPU begins executing code clocked disabling or enabling a particular peripheral.
by the INTOSC multiplexer. The IDLEN and SCS bits
are not affected by the wake-up. The INTRC source will
continue to run if either the WDT or the Fail-Safe Clock
Monitor is enabled.

PIC18F87K90 FAMILY
REGISTER 4-1: PMD3: PERIPHERAL MODULE DISABLE REGISTER 3
(1)
CCP10MD CCP9MD(1) CCP8MD CCP7MD CCP6MD CCP5MD CCP4MD TMR12MD(1)
bit 7 bit 0
Legend:
bit 7 CCP10MD: PMD CCP10 Enable/Disable bit(1)

1 = Peripheral Module Disable (PMD) is enabled for CCP10, disabling all of its clock sources
0 = PMD is disabled for CCP10
bit 6 CCP9MD: PMD CCP9 Enable/Disable bit(1)
bit 5 CCP8MD: PMD CCP8 Enable/Disable bit
bit 0 TMR12MD: TMR12MD Disable bit(1)
1 = PMD is enabled and all TMR12MD clock sources are disabled
0 = PMD is disabled and TMR12MD is enabled
Note 1: Unimplemented in devices with a program memory of 32 Kbytes (PIC18FX5K90).

PIC18F87K90 FAMILY
(1)
TMR10MD TMR8MD TMR7MD(1) TMR6MD TMR5MD CMP3MD CMP2MD CMP1MD
bit 7 bit 0
Legend:

1 = Peripheral Module Disable (PMD) is enabled and all TMR10MD clock sources are disabled
bit 6 TMR8MD: TMR8MD Disable bit
bit 2 CMP3MD: PMD Comparator 3 Enable/Disable bit
1 = PMD is enabled for Comparator 3, disabling all of its clock sources
0 = PMD is disabled for Comparator 3

PIC18F87K90 FAMILY
— CTMUMD RTCCMD(1) TMR4MD TMR3MD TMR2MD TMR1MD —
bit 7 bit 0
Legend:

bit 6 CTMUMD: PMD CTMU Enable/Disable bit
1 = Peripheral Module Disable (PMD) is enabled for CMTU, disabling all of its clock sources
0 = PMD is disabled for CMTU
bit 5 RTCCMD: PMD RTCC Enable/Disable bit(1)
1 = PMD is enabled for RTCC, disabling all of its clock sources
0 = PMD is disabled for RTCC
Note 1: RTCCMD can only be set to ‘1’ after an EECON2 unlock sequence. Refer to Section 17.0 “Real-Time
Clock and Calendar (RTCC)” for the unlock sequence (see Example 17-1).

PIC18F87K90 FAMILY
CCP3MD CCP2MD CCP1MD UART2MD UART1MD SSP2MD SSP1MD ADCMD
bit 7 bit 0
Legend:
bit 7 CCP3MD: PMD ECCP3 Enable/Disable bit

1 = Peripheral Module Disable (PMD) is enabled for ECCP3, disabling all of its clock sources
0 = PMD is disabled for ECCP3
1 = PMD is enabled for ECCP2, disabling all of its clock sources
1 = PMD is enabled for ECCP1, disabling all of its clock sources
bit 4 UART2MD: PMD UART2 Enable/Disable bit
1 = PMD is enabled for UART2, disabling all of its clock sources
0 = PMD is disabled for UART2
bit 3 UART1MD: PMD UART1 Enable/Disable bit
1 = PMD is enabled for UART1, disabling all of its clock sources
0 = PMD is disabled for UART1
bit 2 SSP2MD: PMD MSSP2 Enable/Disable bit
1 = PMD is enabled for MSSP2, disabling all of its clock sources
0 = PMD is disabled for MSSP2
bit 1 SSP1MD: PMD MSSP1 Enable/Disable bit
1 = PMD is enabled for MSSP1, disabling all of its clock sources
0 = PMD is disabled for MSSP1
bit 0 ADCMD: PMD Analog/Digital Converter PMD Enable/Disable bit
1 = PMD is enabled for Analog/Digital Converter, disabling all of its clock sources
0 = PMD is disabled for Analog/Digital Converter

PIC18F87K90 FAMILY
4.6 Exiting Idle and Sleep Modes 4.6.3 EXIT BY RESET
An exit from Sleep mode or any of the Idle modes is Normally, the device is held in Reset by the Oscillator
triggered by an interrupt, a Reset or a WDT time-out. Start-up Timer (OST) until the primary clock becomes
This section discusses the triggers that cause exits ready. At that time, the OSTS bit is set and the device
from power-managed modes. The clocking subsystem begins executing code. If the internal oscillator block is
actions are discussed in each of the power-managed the new clock source, the HFIOFS/MFIOFS bits are set
modes (see Section 4.2 “Run Modes”, Section 4.3 instead.
“Sleep Mode” and Section 4.4 “Idle Modes”). The exit delay time from Reset to the start of code
execution depends on both the clock sources before
4.6.1 EXIT BY INTERRUPT and after the wake-up, and the type of oscillator if the
Any of the available interrupt sources can cause the new clock source is the primary clock. Exit delays are
device to exit from an Idle or Sleep mode to a Run summarized in Table 4-4.
mode. To enable this functionality, an interrupt source Code execution can begin before the primary clock
must be enabled by setting its enable bit in one of the becomes ready. If either the Two-Speed Start-up (see
INTCONx or PIEx registers. The exit sequence is Section 28.4 “Two-Speed Start-up”) or Fail-Safe
initiated when the corresponding interrupt flag bit is set. Clock Monitor (see Section 28.5 “Fail-Safe Clock
On all exits from Idle or Sleep modes by interrupt, code Monitor”) is enabled, the device may begin execution
execution branches to the interrupt vector if the GIE/ as soon as the Reset source has cleared. Execution is
GIEH bit (INTCON<7>) is set. Otherwise, code execu- clocked by the INTOSC multiplexer, driven by the inter-
tion continues or resumes without branching (see nal oscillator block. Execution is clocked by the internal
Section 10.0 “Interrupts”). oscillator block until either the primary clock becomes
ready or a power-managed mode is entered before the
4.6.2 EXIT BY WDT TIME-OUT primary clock becomes ready; the primary clock is then
shut down.
A WDT time-out will cause different actions depending
on which power-managed mode the device is in when
4.6.4 EXIT WITHOUT AN OSCILLATOR
the time-out occurs.
START-UP DELAY
If the device is not executing code (all Idle modes and
Certain exits from power-managed modes do not
Sleep mode), the time-out will result in an exit from the
invoke the OST at all. The two cases are:
power-managed mode (see Section 4.2 “Run
Modes” and Section 4.3 “Sleep Mode”). If the device • When in PRI_IDLE mode, where the primary
is executing code (all Run modes), the time-out will clock source is not stopped
result in a WDT Reset (see Section 28.2 “Watchdog • When the primary clock source is not any of the
Timer (WDT)”). LP, XT, HS or HSPLL modes
Executing a SLEEP or CLRWDT instruction clears the In these instances, the primary clock source either
WDT timer and postscaler, loses the currently selected does not require an oscillator start-up delay, since it is
clock source (if the Fail-Safe Clock Monitor is enabled) already running (PRI_IDLE), or normally does not
and modifies the IRCF bits in the OSCCON register (if require an oscillator start-up delay (RC, EC and INTIO
the internal oscillator block is the device clock source). Oscillator modes). However, a fixed delay of interval,
TCSD, following the wake event, is still required when
leaving Sleep and Idle modes to allow the CPU to
prepare for execution. Instruction execution resumes
on the first clock cycle following this delay.

PIC18F87K90 FAMILY
4.7 Ultra Low-Power Wake-up A series resistor, between RA0 and the external capac-
itor, provides overcurrent protection for the RA0/AN0/
The Ultra Low-Power Wake-up (ULPWU) on pin, RA0, ULPWU pin and enables software calibration of the
allows a slow falling voltage to generate an interrupt time-out (see Figure 4-9).
without excess current consumption.
To use this feature: FIGURE 4-9: ULTRA LOW-POWER
1. Charge the capacitor on RA0 by configuring the WAKE-UP INITIALIZATION
RA0 pin to an output and setting it to ‘1’.
RA0/AN0/ULPWU
2. Stop charging the capacitor by configuring RA0
as an input.
3. Discharge the capacitor by setting the ULPEN
and ULPSINK bits in the WDTCON register.
4. Configure Sleep mode.
5. Enter Sleep mode.
When the voltage on RA0 drops below VIL, the device
wakes up and executes the next instruction.
This feature provides a low-power technique for
periodically waking up the device from Sleep mode.
The time-out is dependent on the discharge time of the
A timer can be used to measure the charge time and
RC circuit on RA0.
discharge time of the capacitor. The charge time can
When the ULPWU module wakes the device from then be adjusted to provide the desired delay in Sleep.
Sleep mode, the ULPLVL bit (WDTCON<5>) is set. This technique compensates for the affects of
Software can check this bit upon wake-up to determine temperature, voltage and component accuracy. The
the wake-up source. peripheral can also be configured as a simple
See Example 4-1 for initializing the ULPWU module. programmable Low-Voltage Detect (LVD) or
temperature sensor.
EXAMPLE 4-1: ULTRA LOW-POWER Note: For more information, see AN 879, “Using
WAKE-UP INITIALIZATION the Microchip Ultra Low-Power Wake-up
Module” (DS00879).
//***************************
//Charge the capacitor on RA0
//***************************
TRISAbits.TRISA0 = 0;
PORTAbits.RA0 = 1;
for(i = 0; i < 10000; i++) Nop();
//*****************************
//Stop Charging the capacitor
//on RA0
//*****************************
TRISAbits.TRISA0 = 1;
//*****************************
//Enable the Ultra Low Power
//Wakeup module and allow
//capacitor discharge
//*****************************
WDTCONbits.ULPEN = 1;
WDTCONbits.ULPSINK = 1;
//For Sleep
OSCCONbits.IDLEN = 0;
//Enter Sleep Mode
//
Sleep();
//for sleep, execution will
//resume here

PIC18F87K90 FAMILY
TABLE 4-4: EXIT DELAY ON WAKE-UP BY RESET FROM SLEEP MODE OR ANY IDLE MODE
(BY CLOCK SOURCES)
Power-Managed Clock Ready
Clock Source(5) Exit Delay
Mode Status Bits
LP, XT, HS
HSPLL OSTS
EC, RC
PRI_IDLE mode TCSD(1)
HF-INTOSC(2) HFIOFS
MF-INTOSC(2) MFIOFS
LF-INTOSC None
SEC_IDLE mode SOSC TCSD(1) SOSCRUN
HF-INTOSC(2) HFIOFS
RC_IDLE mode MF-INTOSC(2) TCSD(1) MFIOFS
LF-INTOSC None
LP, XT, HS TOST(3)
HSPLL TOST + trc(3) OSTS
EC, RC TCSD(1)
Sleep mode
HF-INTOSC(2) HFIOFS
MF-INTOSC(2) TIOBST(4) MFIOFS
LF-INTOSC None
Note 1: TCSD (Parameter 38, Table 31-10) is a required delay when waking from Sleep and all Idle modes, and
runs concurrently with any other required delays (see Section 4.4 “Idle Modes”).
2: Includes postscaler derived frequencies. On Reset, INTOSC defaults to HF-INTOSC at 8 MHz.
3: TOST is the Oscillator Start-up Timer (Parameter 32, Table 31-10). TRC is the PLL Lock-out Timer
(Parameter F12, Table 31-7); it is also designated as TPLL.
4: Execution continues during TIOBST (Parameter 39, Table 31-10), the INTOSC stabilization period.
5: The clock source is dependent upon the settings of the SCS (OSCCON<1:0>), IRCF (OSCCON<6:4>)
and FOSC (CONFIG1H<3:0>) bits.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
5.0 RESET A simplified block diagram of the on-chip Reset circuit
is shown in Figure 5-1.
The PIC18F87K90 family of devices differentiates
between various kinds of Reset: 5.1 RCON Register
a) Power-on Reset (POR)
Device Reset events are tracked through the RCON
b) MCLR Reset during normal operation
register (Register 5-1). The lower five bits of the
c) MCLR Reset during power-managed modes register indicate that a specific Reset event has
d) Watchdog Timer (WDT) Reset (during occurred. In most cases, these bits can only be set by
execution) the event and must be cleared by the application after
e) Configuration Mismatch (CM) Reset the event.
f) Brown-out Reset (BOR) The state of these flag bits, taken together, can be read
g) RESET Instruction to indicate the type of Reset that just occurred. This is
h) Stack Full Reset described in more detail in Section 5.7 “Reset State
of Registers”.
i) Stack Underflow Reset
The RCON register also has a control bit for setting
This section discusses Resets generated by MCLR,
interrupt priority (IPEN). Interrupt priority is discussed
POR and BOR, and covers the operation of the various
in Section 10.0 “Interrupts”.
start-up timers. Stack Reset events are covered in
Section 6.1.3.4 “Stack Full and Underflow Resets”.
WDT Resets are covered in Section 28.2 “Watchdog
Timer (WDT)”.
FIGURE 5-1: SIMPLIFIED BLOCK DIAGRAM OF ON-CHIP RESET CIRCUIT
RESET Instruction
Configuration Word Mismatch
Stack Stack Full/Underflow Reset

Pointer
External Reset
MCLR
( )_IDLE
Sleep
WDT
Time-out
VDD Rise POR Pulse

Detect
VDD
Brown-out
Reset
S
PWRT
32 s PWRT 66 ms Chip_Reset
R Q
LF-INTOSC 11-Bit Ripple Counter

PIC18F87K90 FAMILY
REGISTER 5-1: RCON: RESET CONTROL REGISTER
R/W-0 R/W-1 R/W-1 R/W-1 R-1 R-1 R/W-0 R/W-0
IPEN SBOREN CM RI TO PD POR BOR
bit 7 bit 0
Legend:
bit 7 IPEN: Interrupt Priority Enable bit

1 = Enable priority levels on interrupts
0 = Disable priority levels on interrupts (PIC16CXXX Compatibility mode)
bit 6 SBOREN: BOR Software Enable bit
If BOREN<1:0> = 01:
1 = BOR is enabled
0 = BOR is disabled
If BOREN<1:0> = 00, 10 or 11:
Bit is disabled and read as ‘0’.
bit 5 CM: Configuration Mismatch Flag bit
1 = A Configuration Mismatch Reset has not occurred
0 = A Configuration Mismatch Reset has occurred (must be set in software after a Configuration
Mismatch Reset occurs)
bit 4 RI: RESET Instruction Flag bit
1 = The RESET instruction was not executed (set by firmware only)
0 = The RESET instruction was executed, causing a device Reset (must be set in software after a
Brown-out Reset occurs)
bit 3 TO: Watchdog Time-out Flag bit
1 = Set by power-up, CLRWDT instruction or SLEEP instruction
0 = A WDT time-out has occurred
bit 2 PD: Power-Down Detection Flag bit
1 = Set by power-up or by the CLRWDT instruction
0 = Set by execution of the SLEEP instruction
bit 1 POR: Power-on Reset Status bit
1 = A Power-on Reset has not occurred (set by firmware only)
0 = A Power-on Reset has occurred (must be set in software after a Power-on Reset occurs)
bit 0 BOR: Brown-out Reset Status bit
1 = A Brown-out Reset has not occurred (set by firmware only)
0 = A Brown-out Reset has occurred (must be set in software after a Brown-out Reset occurs)
Note 1: It is recommended that the POR bit be set after a Power-on Reset has been detected, so that subsequent
Power-on Resets may be detected.
2: Brown-out Reset is said to have occurred when BOR is ‘0’ and POR is ‘1’ (assuming that POR was set to
‘1’ by software immediately after a Power-on Reset).

PIC18F87K90 FAMILY
5.2 Master Clear (MCLR) In Zero-Power BOR (ZPBORMV), the module monitors
the VDD voltage and re-arms the POR at about 2V.
The MCLR pin provides a method for triggering a hard ZPBORMV does not cause a Reset, but re-arms the
external Reset of the device. A Reset is generated by POR.
holding the pin low. PIC18 extended microcontroller
devices have a noise filter in the MCLR Reset path The BOR accuracy varies with its power level. The
which detects and ignores small pulses. lower the power setting, the less accurate the BOR trip
levels are. So, the high-power BOR has the highest
The MCLR pin is not driven low by any internal Resets, accuracy and the low-power BOR has the lowest accu-
including the WDT. racy. The trip levels (BVDD, Parameter D005), current
consumption (Section 31.2 “DC Characteristics:
5.3 Power-on Reset (POR) Power-Down and Supply Current PIC18F87K90
Family (Industrial/Extended)”) and time required
A Power-on Reset condition is generated on-chip
below BVDD (TBOR, Parameter 35) can all be found in
whenever VDD rises above a certain threshold. This
Section 31.0 “Electrical Characteristics”
allows the device to start in the initialized state when
VDD is adequate for operation.
FIGURE 5-2: EXTERNAL POWER-ON
To take advantage of the POR circuitry, tie the MCLR RESET CIRCUIT (FOR
pin through a resistor (1 k to 10 k) to VDD. This will
SLOW VDD POWER-UP)
eliminate external RC components usually needed to
create a Power-on Reset delay. A minimum rise rate for
VDD VDD
VDD is specified (Parameter D004). For a slow rise
time, see Figure 5-2.
D R
When the device starts normal operation (exiting the
R1
Reset condition), device operating parameters (such MCLR
as voltage, frequency and temperature) must be met to PIC18F87K90
ensure operation. If these conditions are not met, the C
device must be held in Reset until the operating
conditions are met.
Note 1: External Power-on Reset circuit is required
Power-on Reset events are captured by the POR bit only if the VDD power-up slope is too slow.
(RCON<1>). The state of the bit is set to ‘0’ whenever The diode, D, helps discharge the capacitor
a Power-on Reset occurs and does not change for any quickly when VDD powers down.
other Reset event. POR is not reset to ‘1’ by any 2: R < 40 k is recommended to make sure that
hardware event. To capture multiple events, the user the voltage drop across R does not violate
manually resets the bit to ‘1’ in software following any the device’s electrical specification.
Power-on Reset. 3: R1  1 k will limit any current flowing into
MCLR from external capacitor, C, in the event
5.4 Brown-out Reset (BOR) of MCLR/VPP pin breakdown, due to
Electrostatic Discharge (ESD) or Electrical
The PIC18F87K90 family has four BOR modes: Overstress (EOS).
• High-Power BOR
• Medium Power BOR 5.4.1 DETECTING BOR
• Low-Power BOR The BOR bit always resets to ‘0’ on any Brown-out
• Zero-Power BOR Reset or Power-on Reset event. This makes it difficult
Each power mode is selected by the BORPWR<1:0> to determine if a Brown-out Reset event has occurred
bits setting (CONFIG2L<6:5>). For low, medium and just by reading the state of BOR alone. A more reliable
high-power BOR, the module monitors the VDD depend- method is to simultaneously check the state of both
ing on the BORV<1:0> setting (CONFIG1L<3:2>). A POR and BOR. This assumes that the POR bit is reset
BOR event re-arms the Power-on Reset. It also causes to ‘1’ in software immediately after any Power-on Reset
a Reset depending on which of the trip levels has been event. If BOR is ‘0’ while POR is ‘1’, it can be reliably
set: 1.8V, 2V, 2.7V or 3V. The typical (IBOR) trip level for assumed that a Brown-out Reset event has occurred.
the Low and Medium Power BOR will be 0.75 A and LP-BOR cannot be detected with the BOR bit in the
3 A. RCON register. LP-BOR can rearm the POR and can
cause a Power-on Reset.

PIC18F87K90 FAMILY
5.5 Configuration Mismatch (CM) 5.6 Power-up Timer (PWRT)
The Configuration Mismatch (CM) Reset is designed to PIC18F87K90 family devices incorporate an on-chip
detect, and attempt to recover from, random, memory Power-up Timer (PWRT) to help regulate the Power-on
corrupting events. These include Electrostatic Discharge Reset process. The PWRT is enabled by setting the
(ESD) events that can cause widespread, single bit PWRTEN bit (CONFIG2L<0>). The main function is to
changes throughout the device and result in catastrophic ensure that the device voltage is stable before code is
failure. executed.
In PIC18F87K90 family Flash devices, the device The Power-up Timer (PWRT) of the PIC18F87K90
Configuration registers (located in the configuration family devices is a 13-bit counter that uses the
memory space) are continuously monitored during LF-INTOSC source as the clock input. This yields an
operation by comparing their values to complimentary approximate time interval of 2,048 x 32 s = 66 ms.
shadow registers. If a mismatch is detected between While the PWRT is counting, the device is held in
the two sets of registers, a CM Reset automatically Reset.
occurs. These events are captured by the CM bit The power-up time delay depends on the LF-INTOSC
(RCON<5>). The state of the bit is set to ‘0’ whenever clock and will vary from chip-to-chip due to temperature
a CM event occurs and does not change for any other and process variation. See DC Parameter 33 for
Reset event. details.
A CM Reset behaves similarly to a Master Clear Reset,
RESET instruction, WDT time-out or Stack Event Reset. 5.6.1 TIME-OUT SEQUENCE
As with all hard and power Reset events, the device If enabled, the PWRT time-out is invoked after the POR
Configuration Words are reloaded from the Flash pulse has cleared. The total time-out will vary based on
Configuration Words, in program memory, as the the status of the PWRT. Figure 5-3, Figure 5-4,
device restarts. Figure 5-5 and Figure 5-6 all depict time-out
sequences on power-up with the Power-up Timer
enabled.
Since the time-outs occur from the POR pulse, if
MCLR is kept low long enough, the PWRT will expire.
Bringing MCLR high will begin execution immediately
(Figure 5-5). This is useful for testing purposes or for
synchronizing more than one PIC18 device operating
in parallel.
FIGURE 5-3: TIME-OUT SEQUENCE ON POWER-UP (MCLR TIED TO VDD, VDD RISE < TPWRT)
VDD
MCLR
INTERNAL POR
TPWRT
PWRT TIME-OUT
INTERNAL RESET

PIC18F87K90 FAMILY
FIGURE 5-4: TIME-OUT SEQUENCE ON POWER-UP (MCLR NOT TIED TO VDD): CASE 1
VDD
MCLR
INTERNAL POR
TPWRT
PWRT TIME-OUT
INTERNAL RESET
FIGURE 5-5: TIME-OUT SEQUENCE ON POWER-UP (MCLR NOT TIED TO VDD): CASE 2
VDD
MCLR
INTERNAL POR
TPWRT
PWRT TIME-OUT
INTERNAL RESET
FIGURE 5-6: SLOW RISE TIME (MCLR TIED TO VDD, VDD RISE > TPWRT)
3.3V
VDD 0V 1V
MCLR
INTERNAL POR
TPWRT
PWRT TIME-OUT
INTERNAL RESET

PIC18F87K90 FAMILY
5.7 Reset State of Registers different Reset situations, as indicated in Table 5-1.
These bits are used in software to determine the nature
Most registers are unaffected by a Reset. Their status of the Reset.
is unknown on POR and unchanged by all other
Resets. The other registers are forced to a “Reset Table 5-2 describes the Reset states for all of the
state” depending on the type of Reset that occurred. Special Function Registers. These are categorized by
Power-on and Brown-out Resets, Master Clear and
Most registers are not affected by a WDT wake-up, WDT Resets, and WDT wake-ups.
since this is viewed as the resumption of normal
operation. Status bits from the RCON register (CM, RI,
TO, PD, POR and BOR) are set or cleared differently in
TABLE 5-1: STATUS BITS, THEIR SIGNIFICANCE AND THE INITIALIZATION CONDITION FOR
RCON REGISTER
Program RCON Register STKPTR Register

Condition
Counter(1) CM RI TO PD POR BOR STKFUL STKUNF
Power-on Reset 0000h 1 1 1 1 0 0 0 0
RESET instruction 0000h u 0 u u u u u u
Brown-out Reset 0000h 1 1 1 1 u 0 u u
Configuration Mismatch Reset 0000h 0 u u u u u u u
MCLR Reset during 0000h u u 1 u u u u u
power-managed Run modes
MCLR Reset during power- 0000h u u 1 0 u u u u
managed Idle modes and
Sleep mode
MCLR Reset during full-power 0000h u u u u u u u u
execution
Stack Full Reset (STVREN = 1) 0000h u u u u u u 1 u
Stack Underflow Reset 0000h u u u u u u u 1
(STVREN = 1)
Stack Underflow Error (not an 0000h u u u u u u u 1
actual Reset, STVREN = 0)
WDT time-out during full-power 0000h u u 0 u u u u u
or power-managed Run modes
WDT time-out during PC + 2 u u 0 0 u u u u
power-managed Idle or Sleep
modes
Interrupt exit from PC + 2 u u u 0 u u u u
power-managed modes
Legend: u = unchanged
Note 1: When the wake-up is due to an interrupt and the GIEH or GIEL bit is set, the PC is loaded with the interrupt
vector (0008h or 0018h).

PIC18F87K90 FAMILY
TABLE 5-2: INITIALIZATION CONDITIONS FOR ALL REGISTERS
MCLR Resets,
WDT Reset,
Power-on Reset, Wake-up via WDT
Register Applicable Devices RESET Instruction,
Brown-out Reset or Interrupt
Stack Resets,
CM Resets
TOSU PIC18F6XK90 PIC18F8XK90 ---0 0000 ---0 0000 ---0 uuuu(1)
TOSH PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu(1)
TOSL PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu(1)
STKPTR PIC18F6XK90 PIC18F8XK90 00-0 0000 uu-0 0000 uu-u uuuu(1)
PCLATU PIC18F6XK90 PIC18F8XK90 ---0 0000 ---0 0000 ---u uuuu
PCLATH PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PCL PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 PC + 2(2)
TBLPTRU PIC18F6XK90 PIC18F8XK90 --00 0000 --00 0000 --uu uuuu
TBLPTRH PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
TBLPTRL PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
TABLAT PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PRODH PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
PRODL PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
INTCON PIC18F6XK90 PIC18F8XK90 0000 000x 0000 000u uuuu uuuu(3)
INTCON2 PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu(3)
INTCON3 PIC18F6XK90 PIC18F8XK90 1100 0000 1100 0000 uuuu uuuu(3)
INDF0 PIC18F6XK90 PIC18F8XK90 N/A N/A N/A
POSTINC0 PIC18F6XK90 PIC18F8XK90 N/A N/A N/A
POSTDEC0 PIC18F6XK90 PIC18F8XK90 N/A N/A N/A
PREINC0 PIC18F6XK90 PIC18F8XK90 N/A N/A N/A
PLUSW0 PIC18F6XK90 PIC18F8XK90 N/A N/A N/A
FSR0H PIC18F6XK90 PIC18F8XK90 ---- 0000 ---- 0000 ---- uuuu
FSR0L PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
WREG PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
BSR PIC18F6XK90 PIC18F8XK90 ---- 0000 ---- 0000 ---- uuuu
Legend: u = unchanged, x = unknown, - = unimplemented bit, read as ‘0’, q = value depends on condition.
Shaded cells indicate conditions do not apply for the designated device.
Note 1: When the wake-up is due to an interrupt, and the GIEL or GIEH bit is set, the TOSU, TOSH and TOSL are
updated with the current value of the PC. The STKPTR is modified to point to the next location in the
hardware stack.
2: When the wake-up is due to an interrupt and the GIEL or GIEH bit is set, the PC is loaded with the interrupt
3: One or more bits in the INTCONx or PIRx registers will be affected (to cause wake-up).
4: See Table 5-1 for the Reset value for a specific condition.

PIC18F87K90 FAMILY
TABLE 5-2: INITIALIZATION CONDITIONS FOR ALL REGISTERS (CONTINUED)
MCLR Resets,
WDT Reset,
Stack Resets,
CM Resets
STATUS(4) PIC18F6XK90 PIC18F8XK90 ---x xxxx ---u uuuu ---u uuuu
TMR0H PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
TMR0L PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
T0CON PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
SPBRGH1 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
OSCCON PIC18F6XK90 PIC18F8XK90 0110 q000 0110 q000 uuuu quuu
IPR5 PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
WDTCON PIC18F6XK90 PIC18F8XK90 0-x0 -000 0-x0 -000 u-uu -uuu
RCON PIC18F6XK90 PIC18F8XK90 0111 11qq 0uqq qquu uuuu qquu
TMR1H PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
T1CON PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
TMR2 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PR2 PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
T2CON PIC18F6XK90 PIC18F8XK90 -000 0000 -000 0000 -uuu uuuu
SSP1BUF PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
SSP1ADD PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
SSP1STAT PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
SSP1CON1 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
ADRESH PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
ADRESL PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
ADCON0 PIC18F6XK90 PIC18F8XK90 -000 0000 -000 0000 -uuu uuuu
ADCON1 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
ADCON2 PIC18F6XK90 PIC18F8XK90 0-00 0000 0-00 0000 u-uu uuuu
ECCP1AS PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
hardware stack.

PIC18F87K90 FAMILY
MCLR Resets,
WDT Reset,
Stack Resets,
CM Resets
ECCP1DEL PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
CCPR1H PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
CCPR1L PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
CCP1CON PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PIR5 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PIE5 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu(1)
PIR4 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu(1)
PIE4 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
CVRCON PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
CMSTAT PIC18F6XK90 PIC18F8XK90 111- ---- 111- ---- uuu- ----
T3CON PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0x00 0000 0x00
T3GCON PIC18F6XK90 PIC18F8XK90 0000 0x00 0000 0x00 uuuu uuuu
SPBRG1 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
RCREG1 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
TXREG1 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
TXSTA1 PIC18F6XK90 PIC18F8XK90 0000 0010 0000 0010 uuuu uuuu
RCSTA1 PIC18F6XK90 PIC18F8XK90 0000 000x 0000 000x uuuu uuuu
IPR6 PIC18F6XK90 PIC18F8XK90 ---1 -111 ---1 -111 ---u -uuu
HLVDCON PIC18F6XK90 PIC18F8XK90 0000 0101 0000 0101 uuuu uuuu
PIR6 PIC18F6XK90 PIC18F8XK90 ---0 -000 ---0 -000 ---u -uuu
PIR3 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PIE3 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
IPR2 PIC18F6XK90 PIC18F8XK90 1-11 1111 1-11 1111 u-uu uuuu
PIR2 PIC18F6XK90 PIC18F8XK90 0-00 0000 0-00 0000 u-uu uuuu
PIE2 PIC18F6XK90 PIC18F8XK90 0-00 0000 0-00 0000 u-uu uuuu
IPR1 PIC18F6XK90 PIC18F8XK90 -111 1111 -111 1111 -uuu uuuu
PIR1 PIC18F6XK90 PIC18F8XK90 -000 0000 -000 0000 -uuu uuuu
PIE1 PIC18F6XK90 PIC18F8XK90 -000 0000 -000 0000 -uuu uuuu
hardware stack.

PIC18F87K90 FAMILY
MCLR Resets,
WDT Reset,
Stack Resets,
CM Resets
PSTR1CON PIC18F6XK90 PIC18F8XK90 00-0 0001 00-0 0001 uu-u uuuu
OSCTUNE PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
TRISJ PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
TRISH PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
TRISG PIC18F6XK90 PIC18F8XK90 ---1 1111 ---1 1111 ---u uuuu
TRISF PIC18F6XK90 PIC18F8XK90 1111 111- 1111 111- uuuu uuu-
TRISE PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
TRISD PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
TRISC PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
TRISB PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
TRISA PIC18F6XK90 PIC18F8XK90 1111 1111 1111 1111 uuuu uuuu
LATJ PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
LATH PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
LATG PIC18F6XK90 PIC18F8XK90 ---x xxxx ---u uuuu ---u uuuu
LATF PIC18F6XK90 PIC18F8XK90 xxxx xxx- uuuu uuu- uuuu uuu-
LATE PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
LATD PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
LATC PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
LATB PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
LATA PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
PORTJ PIC18F6XK90 PIC18F8XK90 xxxx xxxx xxxx xxxx uuuu uuuu
PORTH PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PORTG PIC18F6XK90 PIC18F8XK90 --x0 000x --x0 000x --uu uuuu
PORTF PIC18F6XK90 PIC18F8XK90 0000 000- 0000 000- uuuu uuu-
PORTE PIC18F6XK90 PIC18F8XK90 xxxx xxxx xxxx xxxx uuuu uuuu
PORTD PIC18F6XK90 PIC18F8XK90 xxxx xxxx xxxx xxxx uuuu uuuu
PORTC PIC18F6XK90 PIC18F8XK90 xxxx xxxx xxxx xxxx uuuu uuuu
PORTB PIC18F6XK90 PIC18F8XK90 xxxx xxxx xxxx xxxx uuuu uuuu
PORTA PIC18F6XK90 PIC18F8XK90 xx0x 0000 uu0u 0000 uuuu uuuu
hardware stack.

PIC18F87K90 FAMILY
MCLR Resets,
WDT Reset,
Stack Resets,
CM Resets
EECON1 PIC18F6XK90 PIC18F8XK90 xx-0 x000 uu-0 u000 uu-u uuuu
EECON2 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 0000 0000
LCDDATA23 PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
LCDDATA22 PIC18F6XK90 PIC18F8XK90 ---- ---x ---- ---u ---- ---u
LCDDATA16 PIC18F6XK90 PIC18F8XJ90 ---- ---x ---- ---u ---- ---u
LCDDATA10 PIC18F6XK90 PIC18F8XK90 ---- ---x ---- ---u ---- ---u
LCDDATA10 PIC18F6XK90 PIC18F8XJ90 xxxx xxxx uuuu uuuu uuuu uuuu
LCDDATA4 PIC18F6XK90 PIC18F8XJ90 ---- ---x ---- ---u ---- ---u
BAUDCON1 PIC18F6XK90 PIC18F8XK90 0100 0-00 0100 0-00 uuuu u-uu
OSCCON2 PIC18F6XK90 PIC18F8XK90 -0-- 0-x0 -0-- 0-u0 -u-- u-uu
EEADRH PIC18F6XK90 PIC18F8XK90 ---- --00 ---- --00 ---- --uu
hardware stack.

PIC18F87K90 FAMILY
MCLR Resets,
WDT Reset,
Stack Resets,
CM Resets
EEADR PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
EEDATA PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PIE6 PIC18F6XK90 PIC18F8XK90 ---0 -000 ---0 -000 ---u -uuu
RTCCFG PIC18F6XK90 PIC18F8XK90 0-00 0000 u-uu uuuu u-uu uuuu
RTCCAL PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
RTCVALH PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
RTCVALL PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
ALRMCFG PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
ALRMRPT PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
ALRMVALH PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
ALRMVALL PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
CTMUCONH PIC18F6XK90 PIC18F8XK90 0-00 0000 0-00 0000 u-uu uuuu
CTMUCONL PIC18F6XK90 PIC18F8XK90 0000 0000 0000 00xx uuuu uuuu
CTMUICON PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
CM1CON PIC18F6XK90 PIC18F8XK90 0001 1111 0001 1111 uuuu uuuu
PADCFG1 PIC18F6XK90 PIC18F8XK90 000- -00- uuu- -uu- uuu- -uu-
CCP8CON PIC18F6XK90 PIC18F8XK90 --00 0000 --00 0000 --uu uuuu
hardware stack.

PIC18F87K90 FAMILY
MCLR Resets,
WDT Reset,
Stack Resets,
CM Resets
T7CON PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu -uuu
CCPTMRS0 PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
CCPTMRS1 PIC18F6XK90 PIC18F8XK90 00-0 -000 uu-u -uuu uu-u -uuu
CCPTMRS2 PIC18F6XK90 PIC18F8XK90 ---0 -000 ---u -uuu ---u -uuu
REFOCON PIC18F6XK90 PIC18F8XK90 0-00 0000 u-uu uuuu u-uu uuuu
ODCON1 PIC18F6XK90 PIC18F8XK90 000- ---0 uuu- ---u uuu- ---u
ODCON2 PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
ODCON3 PIC18F6XK90 PIC18F8XK90 00-- ---0 uu-- ---u uu-- ---u
ANCON0 PIC18F6XK90 PIC18F8XK90 1111 1111 uuuu uuuu uuuu uuuu
RCSTA2 PIC18F6XK90 PIC18F8XK90 0000 000x 0000 000x uuuu uuuu
TXSTA2 PIC18F6XK90 PIC18F8XK90 0000 0010 0000 0010 uuuu uuuu
BAUDCON2 PIC18F6XK90 PIC18F8XK90 0100 0-00 0100 0-00 uuuu u-uu
hardware stack.

PIC18F87K90 FAMILY
MCLR Resets,
WDT Reset,
Stack Resets,
CM Resets
SPBRGH2 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
SPBRG2 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
RCREG2 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
TXREG2 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PMD0 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
PMD1 PIC18F6XK90 PIC18F8XK90 -000 000- -000 000- -uuu uuu-
T5CON PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
PR4 PIC18F6XK90 PIC18F8XK90 1111 1111 uuuu uuuu uuuu uuuu
SSP2BUF PIC18F6XK90 PIC18F8XK90 xxxx xxxx uuuu uuuu uuuu uuuu
SSP2ADD PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
SSP2STAT PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
hardware stack.

PIC18F87K90 FAMILY
MCLR Resets,
WDT Reset,
Stack Resets,
CM Resets
LCDREF PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
LCDRL PIC18F6XK90 PIC18F8XK90 0000 -000 0000 -000 uuuu -uuu
LCDSE5 PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
LCDSE4 PIC18F6XK90 PIC18F8XK90 ---- ---0 ---- ---u ---- ---u
LCDSE4 PIC18F6XK90 PIC18F8XK90 0000 0000 uuuu uuuu uuuu uuuu
LCDPS PIC18F6XK90 PIC18F8XK90 0000 0000 0000 0000 uuuu uuuu
LCDCON PIC18F6XK90 PIC18F8XK90 000- 0000 000- 0000 uuu- uuuu
hardware stack.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
6.0 MEMORY ORGANIZATION The data EEPROM, for practical purposes, can be
regarded as a peripheral device because it is
PIC18F87K90 family devices have these types of
addressed and accessed through a set of control
memory:
registers.
• Program Memory
Additional detailed information on the operation of the
• Data RAM Flash program memory is provided in Section 7.0
• Data EEPROM “Flash Program Memory”. The data EEPROM is
As Harvard architecture devices, the data and program discussed separately in Section 8.0 “Data EEPROM
memories use separate busses. This enables Memory”.
concurrent access of the two memory spaces.
FIGURE 6-1: MEMORY MAPS FOR PIC18F87K90 FAMILY DEVICES
PC<20:0>
CALL, CALLW, RCALL, 21
RETURN, RETFIE, RETLW,
ADDULNK, SUBULNK
Stack Level 1


Stack Level 31
PIC18FX5K90 PIC18FX6K90 PIC18FX7K90

000000h
On-Chip On-Chip On-Chip
Memory Memory Memory
007FFFh
00FFFFh
User Memory Space

01FFFFh
Unimplemented Unimplemented Unimplemented

Read as ‘0’ Read as ‘0’ Read as ‘0’
1FFFFFh
Note: Sizes of memory areas are not to scale. The sizes of program memory areas are enhanced to show detail.

PIC18F87K90 FAMILY
6.1 Program Memory Organization FIGURE 6-2: HARD VECTOR FOR
PIC18 microcontrollers implement a 21-bit Program PIC18F87K90 FAMILY
Counter that is capable of addressing a 2-Mbyte DEVICES
program memory space. Accessing a location between
the upper boundary of the physically implemented Reset Vector 0000h
memory and the 2-Mbyte address will return all ‘0’s (a
High-Priority Interrupt Vector 0008h
NOP instruction).
The entire PIC18F87K90 family offers a range of Low-Priority Interrupt Vector 0018h
on-chip Flash program memory sizes, from 32 Kbytes
(up to 16,384 single-word instructions) to 128 Kbytes
(65,536 single-word instructions).
• PIC18F65K90 and PIC18F85K90 – 32 Kbytes of On-Chip
Flash memory, storing up to 16,384 single-word Program Memory
instructions
• PIC18F66K90 and PIC18F86K90 – 64 Kbytes of
Flash memory, storing up to 32,768 single-word
instructions
• PIC18F67K90 and PIC18F87K90 – 128 Kbytes of
Flash memory, storing up to 65,536 single-word
instructions
The program memory maps for individual family
members are shown in Figure 6-1.
6.1.1 HARD MEMORY VECTORS

Read ‘0’
All PIC18 devices have a total of three hard-coded
return vectors in their program memory space. The
Reset vector address is the default value to which the
Program Counter returns on all device Resets; it is
located at 0000h. 1FFFFFh
PIC18 devices also have two interrupt vector Legend: (Top of Memory) represents upper boundary
addresses for handling high-priority and low-priority of on-chip program memory space (see
interrupts. The high-priority interrupt vector is located at Figure 6-1 for device-specific values).
0008h and the low-priority interrupt vector is at 0018h. Shaded area represents unimplemented
The locations of these vectors are shown, in relation to memory. Areas are not shown to scale.
the program memory map, in Figure 6-2.

PIC18F87K90 FAMILY
6.1.2 PROGRAM COUNTER The stack operates as a 31-word by 21-bit RAM and a
5-bit Stack Pointer, STKPTR. The stack space is not
The Program Counter (PC) specifies the address of the
part of either program or data space. The Stack Pointer
instruction to fetch for execution. The PC is 21 bits wide
is readable and writable and the address on the top of
and contained in three separate 8-bit registers.
the stack is readable and writable through the
The low byte, known as the PCL register, is both Top-of-Stack Special Function Registers. Data can also
readable and writable. The high byte, or PCH register, be pushed to, or popped from the stack, using these
contains the PC<15:8> bits and is not directly readable registers.
or writable. Updates to the PCH register are performed
A CALL type instruction causes a push onto the stack.
through the PCLATH register. The upper byte is called
The Stack Pointer is first incremented and the location
PCU. This register contains the PC<20:16> bits; it is also
pointed to by the Stack Pointer is written with the
not directly readable or writable. Updates to the PCU
contents of the PC (already pointing to the instruction
register are performed through the PCLATU register.
following the CALL). A RETURN type instruction causes
The contents of PCLATH and PCLATU are transferred to a pop from the stack. The contents of the location
the Program Counter by any operation that writes PCL. pointed to by the STKPTR are transferred to the PC
Similarly, the upper two bytes of the Program Counter and then the Stack Pointer is decremented.
are transferred to PCLATH and PCLATU by an operation
The Stack Pointer is initialized to ‘00000’ after all
that reads PCL. This is useful for computed offsets to the
Resets. There is no RAM associated with the location
PC (see Section 6.1.5.1 “Computed GOTO”).
corresponding to a Stack Pointer value of ‘00000’; this
The PC addresses bytes in the program memory. To is only a Reset value. Status bits indicate if the stack is
prevent the PC from becoming misaligned with word full, has overflowed or has underflowed.
instructions, the Least Significant bit (LSb) of PCL is
fixed to a value of ‘0’. The PC increments by two to 6.1.3.1 Top-of-Stack Access
address sequential instructions in the program
Only the top of the return address stack (TOS) is
memory.
readable and writable. A set of three registers,
The CALL, RCALL, GOTO and program branch TOSU:TOSH:TOSL, holds the contents of the stack
instructions write to the Program Counter directly. For location pointed to by the STKPTR register
these instructions, the contents of PCLATH and (Figure 6-3). This allows users to implement a software
PCLATU are not transferred to the Program Counter. stack, if necessary. After a CALL, RCALL or interrupt (or
ADDULNK and SUBULNK instructions, if the extended
6.1.3 RETURN ADDRESS STACK instruction set is enabled), the software can read the
The return address stack enables execution of any pushed value by reading the TOSU:TOSH:TOSL regis-
combination of up to 31 program calls and interrupts. ters. These values can be placed on a user-defined
The PC is pushed onto the stack when a CALL or software stack. At return time, the software can return
RCALL instruction is executed, or an interrupt is these values to TOSU:TOSH:TOSL and do a return.
Acknowledged. The PC value is pulled off the stack on While accessing the stack, users must disable the
a RETURN, RETLW or a RETFIE instruction. The value Global Interrupt Enable bits to prevent inadvertent
also is pulled off the stack on ADDULNK and SUBULNK stack corruption.
instructions, if the extended instruction set is enabled.
PCLATU and PCLATH are not affected by any of the
RETURN or CALL instructions.
FIGURE 6-3: RETURN ADDRESS STACK AND ASSOCIATED REGISTERS
Return Address Stack <20:0>

Top-of-Stack Registers Stack Pointer
11111
TOSU TOSH TOSL 11110 STKPTR<4:0>
00h 1Ah 34h 11101 00010
00011
Top-of-Stack 001A34h 00010
000D58h 00001
00000

PIC18F87K90 FAMILY
6.1.3.2 Return Stack Pointer (STKPTR) When the stack has been popped enough times to
unload the stack, the next pop will return a value of zero
The STKPTR register (Register 6-1) contains the Stack
to the PC and sets the STKUNF bit while the Stack
Pointer value, the STKFUL (Stack Full) status bit and
Pointer remains at zero. The STKUNF bit will remain
the STKUNF (Stack Underflow) status bits. The value
set until cleared by software or until a POR occurs.
of the Stack Pointer can be 0 through 31. The Stack
Pointer increments before values are pushed onto the Note: Returning a value of zero to the PC on an
stack and decrements after values are popped off the underflow has the effect of vectoring the
stack. On Reset, the Stack Pointer value will be zero. program to the Reset vector, where the
The user may read and write the Stack Pointer value. stack conditions can be verified and
This feature can be used by a Real-Time Operating appropriate actions can be taken. This is
System (RTOS) for return stack maintenance. not the same as a Reset, as the contents
of the SFRs are not affected.
After the PC is pushed onto the stack 31 times (without
popping any values off the stack), the STKFUL bit is
6.1.3.3 PUSH and POP Instructions
set. The STKFUL bit is cleared by software or by a
POR. Since the Top-of-Stack is readable and writable, the
ability to push values onto the stack and pull values off
What happens when the stack becomes full depends
the stack, without disturbing normal program execu-
on the state of the STVREN (Stack Overflow Reset
tion, is a desirable feature. The PIC18 instruction set
Enable) Configuration bit. (For a description of the
includes two instructions, PUSH and POP, that permit
device Configuration bits, see Section 28.1 “Configu-
the TOS to be manipulated under software control.
ration Bits”.) If STVREN is set (default), the 31st push
TOSU, TOSH and TOSL can be modified to place data
will push the (PC + 2) value onto the stack, set the
or a return address on the stack.
STKFUL bit and reset the device. The STKFUL bit will
remain set and the Stack Pointer will be set to zero. The PUSH instruction places the current PC value onto
the stack. This increments the Stack Pointer and loads
If STVREN is cleared, the STKFUL bit will be set on the
the current PC value onto the stack.
31st push and the Stack Pointer will increment to 31.
Any additional pushes will not overwrite the 31st push The POP instruction discards the current TOS by
and the STKPTR will remain at 31. decrementing the Stack Pointer. The previous value
pushed onto the stack then becomes the TOS value.
REGISTER 6-1: STKPTR: STACK POINTER REGISTER

R/C-0 R/C-0 U-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0
STKFUL(1) STKUNF(1) — SP4 SP3 SP2 SP1 SP0
bit 7 bit 0
Legend: C = Clearable bit

bit 7 STKFUL: Stack Full Flag bit(1)

1 = Stack became full or overflowed
0 = Stack has not become full or overflowed
bit 6 STKUNF: Stack Underflow Flag bit(1)
1 = Stack underflow occurred
0 = Stack underflow did not occur
bit 4-0 SP<4:0>: Stack Pointer Location bits
Note 1: Bit 7 and bit 6 are cleared by user software or by a POR.

PIC18F87K90 FAMILY
6.1.3.4 Stack Full and Underflow Resets 6.1.5 LOOK-UP TABLES IN PROGRAM
Device Resets on stack overflow and stack underflow MEMORY
conditions are enabled by setting the STVREN bit There may be programming situations that require the
(CONFIG4L<0>). When STVREN is set, a full or under- creation of data structures, or look-up tables, in
flow condition will set the appropriate STKFUL or program memory. For PIC18 devices, look-up tables
STKUNF bit and then cause a device Reset. When can be implemented in two ways:
STVREN is cleared, a full or underflow condition will set
• Computed GOTO
the appropriate STKFUL or STKUNF bit, but not cause
a device Reset. The STKFUL or STKUNF bits are • Table Reads
cleared by the user software or a Power-on Reset.
6.1.5.1 Computed GOTO
6.1.4 FAST REGISTER STACK A computed GOTO is accomplished by adding an offset
A Fast Register Stack is provided for the STATUS, to the Program Counter. An example is shown in
WREG and BSR registers to provide a “fast return” Example 6-2.
option for interrupts. This stack is only one level deep A look-up table can be formed with an ADDWF PCL
and is neither readable nor writable. It is loaded with the instruction and a group of RETLW nn instructions. The
current value of the corresponding register when the W register is loaded with an offset into the table before
processor vectors for an interrupt. All interrupt sources executing a call to that table. The first instruction of the
will push values into the Stack registers. The values in called routine is the ADDWF PCL instruction. The next
the registers are then loaded back into the working instruction executed will be one of the RETLW nn
registers if the RETFIE, FAST instruction is used to instructions that returns the value, ‘nn’, to the calling
return from the interrupt. function.
If both low and high-priority interrupts are enabled, the The offset value (in WREG) specifies the number of
Stack registers cannot be used reliably to return from bytes that the Program Counter should advance and
low-priority interrupts. If a high-priority interrupt occurs should be multiples of two (LSb = 0).
while servicing a low-priority interrupt, the Stack In this method, only one data byte may be stored in
register values stored by the low-priority interrupt will each instruction location and room on the return
be overwritten. In these cases, users must save the key address stack is required.
registers in software during a low-priority interrupt.
If interrupt priority is not used, all interrupts may use the EXAMPLE 6-2: COMPUTED GOTO USING
Fast Register Stack for returns from interrupt. If no AN OFFSET VALUE
interrupts are used, the Fast Register Stack can be MOVF OFFSET, W
used to restore the STATUS, WREG and BSR registers CALL TABLE
at the end of a subroutine call. To use the Fast Register ORG nn00h
Stack for a subroutine call, a CALL label, FAST TABLE ADDWF PCL
instruction must be executed to save the STATUS, RETLW nnh
WREG and BSR registers to the Fast Register Stack. A RETLW nnh
RETURN, FAST instruction is then executed to restore RETLW nnh
these registers from the Fast Register Stack. .
.
Example 6-1 shows a source code example that uses .
the Fast Register Stack during a subroutine call and
return.
6.1.5.2 Table Reads
EXAMPLE 6-1: FAST REGISTER STACK A better method of storing data in program memory
CODE EXAMPLE allows two bytes of data to be stored in each instruction
location.
CALL SUB1, FAST ;STATUS, WREG, BSR
;SAVED IN FAST REGISTER Look-up table data may be stored, two bytes per
;STACK program word, while programming. The Table Pointer
 (TBLPTR) specifies the byte address and the Table
 Latch (TABLAT) contains the data that is read from the
program memory. Data is transferred from program
SUB1 
memory, one byte at a time.

RETURN FAST ;RESTORE VALUES SAVED The table read operation is discussed further in
;IN FAST REGISTER STACK Section 7.1 “Table Reads and Table Writes”.

PIC18F87K90 FAMILY
6.2 PIC18 Instruction Cycle 6.2.2 INSTRUCTION FLOW/PIPELINING
An “Instruction Cycle” consists of four Q cycles, Q1
6.2.1 CLOCKING SCHEME through Q4. The instruction fetch and execute are pipe-
The microcontroller clock input, whether from an lined in such a manner that a fetch takes one instruction
internal or external source, is internally divided by four cycle, while the decode and execute take another
to generate four non-overlapping, quadrature clocks instruction cycle. However, due to the pipelining, each
(Q1, Q2, Q3 and Q4). Internally, the Program Counter instruction effectively executes in one cycle. If an
is incremented on every Q1, with the instruction instruction (such as GOTO) causes the Program
fetched from the program memory and latched into the Counter to change, two cycles are required to complete
Instruction Register (IR) during Q4. the instruction. (See Example 6-3.)
The instruction is decoded and executed during the A fetch cycle begins with the Program Counter (PC)
following Q1 through Q4. The clocks and instruction incrementing in Q1.
execution flow are shown in Figure 6-4. In the execution cycle, the fetched instruction is latched
into the Instruction Register (IR) in cycle Q1. This
instruction is then decoded and executed during the
Q2, Q3 and Q4 cycles. Data memory is read during Q2
(operand read) and written during Q4 (destination
write).
FIGURE 6-4: CLOCK/INSTRUCTION CYCLE
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
OSC1
Q1
Q2 Internal
Phase
Q3 Clock
Q4
PC PC PC + 2 PC + 4
OSC2/CLKO
(RC mode)
Execute INST (PC – 2)
Fetch INST (PC) Execute INST (PC)
Fetch INST (PC + 2) Execute INST (PC + 2)
Fetch INST (PC + 4)
EXAMPLE 6-3: INSTRUCTION PIPELINE FLOW
TCY0 TCY1 TCY2 TCY3 TCY4 TCY5

1. MOVLW 55h Fetch 1 Execute 1
2. MOVWF PORTB Fetch 2 Execute 2
3. BRA SUB_1 Fetch 3 Execute 3
4. BSF PORTA, BIT3 (Forced NOP) Fetch 4 Flush (NOP)
5. Instruction @ address SUB_1 Fetch SUB_1 Execute SUB_1
All instructions are single cycle, except for any program branches. These take two cycles since the fetch instruction
is “flushed” from the pipeline while the new instruction is being fetched and then executed.

PIC18F87K90 FAMILY
6.2.3 INSTRUCTIONS IN PROGRAM The CALL and GOTO instructions have the absolute
MEMORY program memory address embedded into the instruc-
tion. Since instructions are always stored on word
The program memory is addressed in bytes. Instructions
boundaries, the data contained in the instruction is a
are stored as two or four bytes in program memory. The
word address. The word address is written to PC<20:1>
Least Significant Byte (LSB) of an instruction word is
which accesses the desired byte address in program
always stored in a program memory location with an
memory. Instruction #2 in Figure 6-5 shows how the
even address (LSB = 0). To maintain alignment with
instruction, GOTO 0006h, is encoded in the program
instruction boundaries, the PC increments in steps of
memory. Program branch instructions, which encode a
two and the LSB will always read ‘0’ (see Section 6.1.2
relative address offset, operate in the same manner. The
“Program Counter”).
offset value stored in a branch instruction represents the
Figure 6-5 shows an example of how instruction words number of single-word instructions that the PC will be
are stored in the program memory. offset by. For more details on the instruction set, see
Section 29.0 “Instruction Set Summary”.
FIGURE 6-5: INSTRUCTIONS IN PROGRAM MEMORY

Word Address
LSB = 1 LSB = 0 
Program Memory 000000h
Byte Locations  000002h
000004h
000006h
Instruction 1: MOVLW 055h 0Fh 55h 000008h
Instruction 2: GOTO 0006h EFh 03h 00000Ah
F0h 00h 00000Ch
Instruction 3: MOVFF 123h, 456h C1h 23h 00000Eh
F4h 56h 000010h
000012h
000014h
6.2.4 TWO-WORD INSTRUCTIONS used by the instruction sequence. If the first word is
skipped for some reason, and the second word is
The standard PIC18 instruction set has four two-word
executed by itself, a NOP is executed instead. This is
instructions: CALL, MOVFF, GOTO and LSFR. In all
necessary for cases when the two-word instruction is
cases, the second word of the instructions always has
preceded by a conditional instruction that changes the
‘1111’ as its four Most Significant bits. The other 12 bits
PC. Example 6-4 shows how this works.
are literal data, usually a data memory address.
The use of ‘1111’ in the 4 MSbs of an instruction Note: For information on two-word instructions
specifies a special form of NOP. If the instruction is in the extended instruction set, see
executed in proper sequence, immediately after the Section 6.5 “Program Memory and the
first word, the data in the second word is accessed and Extended Instruction Set”.
EXAMPLE 6-4: TWO-WORD INSTRUCTIONS

CASE 1:
Object Code Source Code
0110 0110 0000 0000 TSTFSZ REG1 ; is RAM location 0?
1100 0001 0010 0011 MOVFF REG1, REG2 ; No, skip this word
1111 0100 0101 0110 ; Execute this word as a NOP
0010 0100 0000 0000 ADDWF REG3 ; continue code
CASE 2:
Object Code Source Code
0110 0110 0000 0000 TSTFSZ REG1 ; is RAM location 0?
1100 0001 0010 0011 MOVFF REG1, REG2 ; Yes, execute this word
1111 0100 0101 0110 ; 2nd word of instruction
0010 0100 0000 0000 ADDWF REG3 ; continue code

PIC18F87K90 FAMILY
6.3 Data Memory Organization 6.3.1 BANK SELECT REGISTER
Large areas of data memory require an efficient
Note: The operation of some aspects of data
addressing scheme to make possible rapid access to
memory are changed when the PIC18
any address. Ideally, this means that an entire address
extended instruction set is enabled. See
does not need to be provided for each read or write
Section 6.6 “Data Memory and the
operation. For PIC18 devices, this is accomplished with
Extended Instruction Set” for more
a RAM banking scheme. This divides the memory
information.
space into 16 contiguous banks of 256 bytes. Depend-
The data memory in PIC18 devices is implemented as ing on the instruction, each location can be addressed
static RAM. Each register in the data memory has a directly by its full 12-bit address, or an 8-bit, low-order
12-bit address, allowing up to 4,096 bytes of data address and a 4-bit Bank Pointer.
memory. The memory space is divided into as many as Most instructions in the PIC18 instruction set make use
16 banks that contain 256 bytes each. PIC18FX6K90 of the Bank Pointer, known as the Bank Select Register
and PIC18FX7K90 devices implement all 16 complete (BSR). This SFR holds the four Most Significant bits of
banks, for a total of 4 Kbytes. PIC18FX5K90 devices a location’s address. The instruction itself includes the
implement only the first eight complete banks, for a eight Least Significant bits. Only the four lower bits of
total of 2 Kbytes. the BSR are implemented (BSR<3:0>). The upper four
Figure 6-6 and Figure 6-7 show the data memory bits are unused, always read as ‘0’ and cannot be
organization for the devices. written to. The BSR can be loaded directly by using the
MOVLB instruction.
The data memory contains Special Function Registers
(SFRs) and General Purpose Registers (GPRs). The The value of the BSR indicates the bank in data
SFRs are used for control and status of the controller memory. The eight bits in the instruction show the loca-
and peripheral functions, while GPRs are used for data tion in the bank and can be thought of as an offset from
storage and scratchpad operations in the user’s the bank’s lower boundary. The relationship between
application. Any read of an unimplemented location will the BSR’s value and the bank division in data memory
read as ‘0’s. is shown in Figure 6-7.
The instruction set and architecture allow operations Since up to 16 registers may share the same low-order
across all banks. The entire data memory may be address, the user must always be careful to ensure that
accessed by Direct, Indirect or Indexed Addressing the proper bank is selected before performing a data
modes. Addressing modes are discussed later in this read or write. For example, writing what should be
section. program data to an 8-bit address of F9h while the BSR
is 0Fh, will end up resetting the Program Counter.
To ensure that commonly used registers (select SFRs
and select GPRs) can be accessed in a single cycle, While any bank can be selected, only those banks that
PIC18 devices implement an Access Bank. This is a are actually implemented can be read or written to.
256-byte memory space that provides fast access to Writes to unimplemented banks are ignored, while
select SFRs and the lower portion of GPR Bank 0 with- reads from unimplemented banks will return ‘0’s. Even
out using the Bank Select Register. For details on the so, the STATUS register will still be affected as if the
Access RAM, see Section 6.3.2 “Access Bank”. operation was successful. The data memory map in
Figure 6-6 indicates which banks are implemented.
In the core PIC18 instruction set, only the MOVFF
instruction fully specifies the 12-bit address of the
source and target registers. When this instruction
executes, it ignores the BSR completely. All other
instructions include only the low-order address as an
operand and must use either the BSR or the Access
Bank to locate their target registers.

PIC18F87K90 FAMILY
FIGURE 6-6: DATA MEMORY MAP FOR PIC18FX5K90 AND PIC18FX7K90 DEVICES
BSR<3:0> Data Memory Map
00h 000h
= 0000 Access RAM
05Fh When a = 0:
Bank 0 060h
GPR The BSR is ignored and the
FFh 0FFh
Access Bank is used.
00h 100h
= 0001 The first 96 bytes are general
Bank 1 GPR
purpose RAM (from Bank 0).
FFh 1FFh
00h 200h The second 160 bytes are
= 0010 Special Function Registers
Bank 2 GPR
(from Bank 15).
FFh 2FFh
00h 300h
= 0011 When a = 1:
Bank 3 GPR
FFh 3FFh The BSR specifies the bank
00h 400h used by the instruction.
= 0100
Bank 4 GPR
FFh 4FFh
00h 500h
= 0101
Bank 5 GPR
FFh 5FFh
00h 600h
= 0110
Bank 6 GPR
FFh 6FFh
00h 700h
= 0111 Access Bank
Bank 7 GPR
FFh 00h
7FFh Access RAM Low
00h 800h 5Fh
= 1000
GPR(2) Access RAM High 60h
Bank 8 (SFRs)
FFh 8FFh FFh
= 1001 00h 900h
Bank 9 GPR(2)
FFh 9FFh
= 1010 00h A00h
Bank 10 GPR(2)
FFh AFFh
= 1011 00h B00h
Bank 11 GPR(2)
FFh BFFh
= 1100 00h C00h
Bank 12 GPR(2)
FFh CFFh
= 1101 00h D00h
Bank 13 GPR(2)
FFh DFFh
00h E00h
= 1110
Bank 14 GPR(1,2)
FFh EFFh
= 1111 00h GPR(1,2) F00h
Bank 15 F5Fh
SFR F60h
FFh FFFh
Note 1: Addresses, EF4h through F5Fh, are also used by SFRs, but are not part of the Access RAM. Users must
always use the complete address, or load the proper BSR value, to access these registers.
2: These addresses are unused for devices with 32 Kbytes of program memory (PIC18FX5K90). For those
devices, read these addresses at 00h.

PIC18F87K90 FAMILY
FIGURE 6-7: USE OF THE BANK SELECT REGISTER (DIRECT ADDRESSING)
BSR(1) Data Memory From Opcode(2)

7 0 000h 00h 7 0
0 0 0 0 0 0 1 0 Bank 0 1 1 11 1 11 1 11 11
100h FFh
00h
Bank 1
Bank Select(2) 200h FFh
00h
Bank 2
300h FFh
00h
Bank 3
through
Bank 13
FFh
E00h
00h
Bank 14
F00h FFh
00h
Bank 15
FFFh FFh
Note 1: The Access RAM bit of the instruction can be used to force an override of the selected bank (BSR<3:0>)
to the registers of the Access Bank.
2: The MOVFF instruction embeds the entire 12-bit address in the instruction.
6.3.2 ACCESS BANK Using this “forced” addressing allows the instruction to
operate on a data address in a single cycle without
While the use of the BSR, with an embedded 8-bit
updating the BSR first. For 8-bit addresses of 60h and
address, allows users to address the entire range of data
above, this means that users can evaluate and operate
memory, it also means that the user must ensure that the
on SFRs more efficiently. The Access RAM below 60h
correct bank is selected. If not, data may be read from,
is a good place for data values that the user might need
or written to, the wrong location. This can be disastrous
to access rapidly, such as immediate computational
if a GPR is the intended target of an operation, but an
results or common program variables.
SFR is written to instead. But verifying and/or changing
the BSR for each read or write to data memory can Access RAM also allows for faster and more code
become very inefficient. efficient context saving and switching of variables.
To streamline access for the most commonly used data The mapping of the Access Bank is slightly different
memory locations, the data memory is configured with when the extended instruction set is enabled (XINST
an Access Bank, which allows users to access a Configuration bit = 1). This is discussed in more detail
mapped block of memory without specifying a BSR. in Section 6.6.3 “Mapping the Access Bank in
The Access Bank consists of the first 96 bytes of Indexed Literal Offset Mode”.
memory (00h-5Fh) in Bank 0 and the last 160 bytes of
memory (60h-FFh) in Bank 15. The lower half is known 6.3.3 GENERAL PURPOSE
as the “Access RAM” and is composed of GPRs. The REGISTER FILE
upper half is where the device’s SFRs are mapped. PIC18 devices may have banked memory in the GPR
These two areas are mapped contiguously in the area. This is data RAM which is available for use by all
Access Bank and can be addressed in a linear fashion instructions. GPRs start at the bottom of Bank 0
by an 8-bit address (Figure 6-6). (address 000h) and grow upwards towards the bottom of
The Access Bank is used by core PIC18 instructions the SFR area. GPRs are not initialized by a Power-on
that include the Access RAM bit (the ‘a’ parameter in Reset and are unchanged on all other Resets.
the instruction). When ‘a’ is equal to ‘1’, the instruction
uses the BSR and the 8-bit address included in the
opcode for the data memory address. When ‘a’ is ‘0’,
however, the instruction is forced to use the Access
Bank address map. In that case, the current value of
the BSR is ignored entirely.

PIC18F87K90 FAMILY
6.3.4 SPECIAL FUNCTION REGISTERS The SFRs can be classified into two sets: those
associated with the “core” device functionality (ALU,
The Special Function Registers (SFRs) are registers Resets and interrupts) and those related to the
used by the CPU and peripheral modules for controlling peripheral functions. The Reset and Interrupt registers
the desired operation of the device. These registers are are described in their respective chapters, while the
implemented as static RAM. SFRs start at the top of ALU’s STATUS register is described later in this section.
data memory (FFFh) and extend downward to occupy Registers related to the operation of the peripheral
all of Bank 15 (F00h to FFFh) and the top part of features are described in the chapter for that peripheral.
Bank 14 (EF4h to EFFh).
The SFRs are typically distributed among the
A list of these registers is given in Table 6-1 and peripherals whose functions they control. Unused SFR
Table 6-2. locations are unimplemented and read as ‘0’s.
TABLE 6-1: PIC18F87K90 FAMILY SPECIAL FUNCTION REGISTER MAP(5)
Addr. Name Addr. Name Addr. Name Addr. Name Addr. Name Addr. Name
FFFh TOSU FDFh INDF2(1) FBFh ECCP1AS F9Fh IPR1 F7Fh EECON1 F5Fh RTCCFG
FFEh TOSH FDEh POSTINC2(1) FBEh ECCP1DEL F9Eh PIR1 F7Eh EECON2 F5Eh RTCCAL
FFDh TOSL FDDh POSTDEC2(1) FBDh CCPR1H F9Dh PIE1 F7Dh LCDDATA23(3) F5Dh RTCVALH
FFCh STKPTR FDCh PREINC2(1) FBCh CCPR1L F9Ch PSTR1CON F7Ch LCDDATA22(3) F5Ch RTCVALL
(1)
FFBh PCLATU FDBh PLUSW2 FBBh CCP1CON F9Bh OSCTUNE F7Bh LCDDATA21 F5Bh ALRMCFG
FFAh PCLATH FDAh FSR2H FBAh PIR5 F9Ah TRISJ(3) F7Ah LCDDATA20 F5Ah ALRMRPT
(3)
FF9h PCL FD9h FSR2L FB9h PIE5 F99h TRISH F79h LCDDATA19 F59h ALRMVALH
FF8h TBLPTRU FD8h STATUS FB8h IPR4 F98h TRISG F78h LCDDATA18 F58h ALRMVALL
FF7h TBLPTRH FD7h TMR0H FB7h PIR4 F97h TRISF F77h LCDDATA17(3) F57h CTMUCONH
FF6h TBLPTRL FD6h TMR0L FB6h PIE4 F96h TRISE F76h LCDDATA16(3) F56h CTMUCONL
FF5h TABLAT FD5h T0CON FB5h CVRCON F95h TRISD F75h LCDDATA15 F55h CTMUICON
FF4h PRODH FD4h SPBRGH1 FB4h CMSTAT F94h TRISC F74h LCDDATA14 F54h CMCON1
FF3h PRODL FD3h OSCCON FB3h TMR3H F93h TRISB F73h LCDDATA13 F53h PADCFG1
FF2h INTCON FD2h IPR5 FB2h TMR3L F92h TRISA F72h LCDDATA12 F52h ECCP2AS
FF1h INTCON2 FD1h WDTCON FB1h T3CON F91h LATJ(3) F71h LCDDATA11(3) F51h ECCP2DEL
FF0h INTCON3 FD0h RCON FB0h T3GCON F90h LATH(3) F70h LCDDATA10(3) F50h CCPR2H
FEFh INDF0(1) FCFh TMR1H FAFh SPBRG1 F8Fh LATG F6Fh LCDDATA9 F4Fh CCPR2L
FEEh POSTINC0(1) FCEh TMR1L FAEh RCREG1 F8Eh LATF F6Eh LCDDATA8 F4Eh CCP2CON
FEDh POSTDEC0(1) FCDh T1CON FADh TXREG1 F8Dh LATE F6Dh LCDDATA7 F4Dh ECCP3AS
FECh PREINC0(1) FCCh TMR2 FACh TXSTA1 F8Ch LATD F6Ch LCDDATA6 F4Ch ECCP3DEL
FEBh PLUSW0(1) FCBh PR2 FABh RCSTA1 F8Bh LATC F6Bh LCDDATA5(3) F4Bh CCPR3H
FEAh FSR0H FCAh T2CON FAAh T1GCON F8Ah LATB F6Ah LCDDATA4(3) F4Ah CCPR3L
FE9h FSR0L FC9h SSP1BUF FA9h IPR6 F89h LATA F69h LCDDATA3 F49h CCP3CON
FE8h WREG FC8h SSP1ADD FA8h HLVDCON F88h PORTJ(3) F68h LCDDATA2 F48h CCPR8H
(1) (2) (3)
FE7h INDF1 FC7h SSP1STAT FA7h — F87h PORTH F67h LCDDATA1 F47h CCPR8L
FE6h POSTINC1(1) FC6h SSP1CON1 FA6h PIR6 F86h PORTG F66h LCDDATA0 F46h CCP8CON
FE5h POSTDEC1(1) FC5h SSP1CON2 FA5h IPR3 F85h PORTF F65h BAUDCON1 F45h CCPR9H(4)
FE4h PREINC1(1) FC4h ADRESH FA4h PIR3 F84h PORTE F64h OSCCON2 F44h CCPR9L(4)
FE3h PLUSW1(1) FC3h ADRESL FA3h PIE3 F83h PORTD F63h EEADRH F43h CCP9CON(4)
FE2h FSR1H FC2h ADCON0 FA2h IPR2 F82h PORTC F62h EEADR F42h CCPR10H(4)
FE1h FSR1L FC1h ADCON1 FA1h PIR2 F81h PORTB F61h EEDATA F41h CCPR10L(4)
FE0h BSR FC0h ADCON2 FA0h PIE2 F80h PORTA F60h PIE6 F40h CCP10CON(4)
(4) (4)
F3Fh TMR7H F32h TMR12 F25h ANCON0 F18h PMD1 F0Bh CCPR6H EFEh SSP2CON2
Note 1: This is not a physical register.

2: Unimplemented registers are read as ‘0’.
3: This register is not available in 64-pin devices (PIC18F6XK90).
4: This register is not available in devices with a program memory of 32 Kbytes (PIC18FX5K90).
5: Addresses, EF4h through F5Fh, are also used by SFRs, but are not part of the Access RAM. Users must always load
the proper BSR value to access these registers.

PIC18F87K90 FAMILY
TABLE 6-1: PIC18F87K90 FAMILY SPECIAL FUNCTION REGISTER MAP(5) (CONTINUED)
Addr. Name Addr. Name Addr. Name Addr. Name Addr. Name Addr. Name
F3Eh TMR7L(4) F31h PR12(4) F24h ANCON1 F17h PMD2 F0Ah CCPR6L EFDh LCDREF
F3Dh T7CON(4) F30h T12CON(4) F23h ANCON2 F16h PMD3 F09h CCP6CON EFCh LCDRL
F3Ch T7GCON(4) F2Fh CM2CON F22h RCSTA2 F15h TMR5H F08h CCPR7H EFBh LCDSE5(3)
F3Bh TMR6 F2Eh CM3CON F21h TXSTA2 F14h TMR5L F07h CCPR7L EFAh LCDSE4
F3Ah PR6 F2Dh CCPTMRS0 F20h BAUDCON2 F13h T5CON F06h CCP7CON EF9h LCDSE3
F39H T6CON F2Ch CCPTMRS1 F1Fh SPBRGH2 F12h T5GCON F05h TMR4 EF8h LCDSE2
F38h TMR8 F2Bh CCPTMRS2 F1Eh SPBRG2 F11h CCPR4H F04h PR4 EF7h LCDSE1
F37h PR8 F2Ah REFOCON F1Dh RCREG2 F10h CCPR4L F03h T4CON EF6h LCDSE0
F36h T8CON F29H ODCON1 F1Ch TXREG2 F0Fh CCP4CON F02h SSP2BUF EF5h LCDPS
F35h TMR10(4) F28h ODCON2 F1Bh PSTR2CON F0Eh CCPR5H F01h SSP2ADD EF4h LCDCON
F34h PR10(4) F27h ODCON3 F1Ah PSTR3CON F0Dh CCPR5L F00h SSP2STAT
F33h T10CON(4) F26h — F19h PMD0 F0Ch CCP5CON EFFh SSP2CON1
Note 1: This is not a physical register.

2: Unimplemented registers are read as ‘0’.
3: This register is not available in 64-pin devices (PIC18F6XK90).
4: This register is not available in devices with a program memory of 32 Kbytes (PIC18FX5K90).
5: Addresses, EF4h through F5Fh, are also used by SFRs, but are not part of the Access RAM. Users must always load
the proper BSR value to access these registers.

PIC18F87K90 FAMILY
TABLE 6-2: PIC18F87K90 FAMILY REGISTER FILE SUMMARY
Value on
Address File Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
POR, BOR
EF4h LCDCON LCDEN SLPEN WERR — CS1 CS0 LMUX1 LMUX0 000- 0000
EF5h LCDPS WFT BIASMD LCDA WA LP3 LP2 LP1 LP0 0000 0000
EF6h LCDSE0 SE07 SE06 SE05 SE04 SE03 SE02 SE01 SE00 0000 0000
EFAh LCDSE4 SE39 SE38 S37 SE36 SE35 SE34 SE33 SE32 0000 0000
EFBh LCDSE5(2) SE47 SE46 SE45 SE44 SE43 SE42 SE41 SE40 0000 0000
EFCh LCDRL LRLAP1 LRLAP0 LRLBP1 LRLBP0 — LRLAT2 LRLAT1 LRLAT0 0000 -000
EFDh LCDREF LCDIRE LCDIRS LCDCST2 LCDCST1 LCDCST0 VLCD3PE VLCD2PE VLCD1PE 0000 0000
EFEh SSP2CON2 GCEN ACKSTAT ACKDT ACKEN RCEN PEN RSEN SEN 0000 0000
EFFh SSP2CON1 WCOL SSPOV SSPEN CKP SSPM3 SSPM2 SSPM1 SSPM0 0000 0000
F00h SSP2STAT SMP CKE D/A P S R/W UA BF 0000 0000
F01h SSP2ADD MSSP Address Register in I2C™ Slave Mode. SSP1 Baud Rate Reload Register in I2C Master Mode 0000 0000
F02h SSP2BUF MSSP Receive Buffer/Transmit Register xxxx xxxx
F03h T4CON — T4OUTPS3 T4OUTPS2 T4OUTPS1 T4OUTPS0 TMR4ON T4CKPS1 T4CKPS0 -000 0000
F04h PR4 Timer4 Period Register 0000 0000
F05h TMR4 Timer4 Register 1111 1111
F06h CCP7CON — — DC7B1 DC7B0 CCP7M3 CCP7M2 CCP7M1 CCP7M0 --00 0000
F07h CCPR7L Capture/Compare/PWM Register 7 Low Byte xxxx xxxx
F08h CCPR7H Capture/Compare/PWM Register7 High Byte xxxx xxxx
F0Ah CCPR6L Capture/Compare/PWM Register 6 Low Byte xxxx xxxx
F0Bh CCPR6H Capture/Compare/PWM Register6 High Byte xxxx xxxx
F0Ch CCP5CON — — DC5B1 DC5B0 CCP5M3 CCP5M2 CCP5M1 CCP5M0 --00 0000
F0Dh CCPR5L Capture/Compare/PWM Register 5 Low Byte xxxx xxxx
F0Eh CCPR5H Capture/Compare/PWM Register 5 High Byte xxxx xxxx
F0Fh CCP4CON — — DC4B1 DC4B0 CCP4M3 CCP4M2 CCP4M1 CCP4M0 --00 0000
F11h CCPR4H Capture/Compare/PWM Register 4 High Byte xxxx xxxx
F12h T5GCON TMR5GE T5GPOL T5GTM T5GSPM T5GGO/ T5GVAL T5GSS1 T5GSS0 0000 0000
T5DONE
F13h T5CON TMR5CS1 TMR5CS0 T5CKPS1 T5CKPS0 SOSCEN T5SYNC RD16 TMR5ON 0000 0000
F14h TMR5L Timer5 Register Low Byte 0000 0000
F15h TMR5H Timer5 Register High Byte xxxx xxxx
F16h PMD3 CCP10MD(3) CCP9MD(3) CCP8MD CCP7MD CCP6MD CCP5MD CCP4MD TMR12MD(3) 0000 0000
F17h PMD2 TMR10MD(3) TMR8MD TMR7MD(3) TMR6MD TMR5MD CMP3MD CMP2MD CMP1MD 0000 0000
F18h PMD1 — CTMUMD RTCCMD TMR4MD TMR3MD TMR2MD TMR1MD — -000 000-
F19h PMD0 CCP3MD CCP2MD CCP1MD UART2MD UART1MD SSP2MD SSP1MD ADCMD 0000 0000
F1Ah PSTR3CON CMPL1 CMPL0 — STRSYNC STRD STRC STRB STRA 00-0 0001
F1Bh PSTR2CON CMPL1 CMPL0 — STRSYNC STRD STRC STRB STRA 00-0 0001
F1Ch TXREG2 Transmit Data FIFO xxxx xxxx
F1Dh RCREG2 Receive Data FIFO 0000 0000
F1Eh SPBRG2 USART2 Baud Rate Generator Low Byte 0000 0000
F1Fh SPBRGH2 USART2 Baud Rate Generator High Byte 0000 0000
F20h BAUDCON2 ABDOVF RCIDL RXDTP TXCKP BRG16 — WUE ABDEN 0100 0-00
F21h TXSTA2 CSRC TX9 TXEN SYNC SENDB BRGH TRMT TX9D 0000 0010
F22h RCSTA2 SPEN RX9 SREN CREN ADDEN FERR OERR RX9D 0000 000x
F23h ANCON2 ANSEL23 ANSEL22 ANSEL21 ANSEL20 ANSEL19 ANSEL18 ANSEL17 ANSEL16 1111 1111
Note 1: This bit is available when Master Clear is disabled (MCLRE = 0). When MCLRE is set, the bit is unimplemented.
2: Unimplemented in 64-pin devices (PIC18F6XK90).
3: Unimplemented in devices with a program memory of 32 Kbytes (PIC18FX5K90).

PIC18F87K90 FAMILY
TABLE 6-2: PIC18F87K90 FAMILY REGISTER FILE SUMMARY (CONTINUED)
Value on
POR, BOR
F26h — — — — — — — — — —
F27h ODCON3 U2OD U1OD — — — — — CTMUDS 00-- ---0
F28h ODCON2 CCP10OD(3) CCP9OD(3) CCP8OD CCP7OD CCP6OD CCP5OD CCP4OD CCP3OD 0000 0000
F29H ODCON1 SSP1OD CCP2OD CCP1OD — — — — SSP2OD 000- ---0
F2Ah REFOCON ROON — ROSSLP ROSEL RODIV3 RODIV2 RODIV1 RODIV0 0-00 0000
F2Bh CCPTMRS2 — — — C10TSEL0 — C9TSEL0 C8TSEL1 C8TSEL0 ---0 -000
F2Ch CCPTMRS1 C7TSEL1 C7TSEL0 — C6TSEL0 — C5TSEL0 C4TSEL1 C4TSEL0 00-0 -000
F2Dh CCPTMRS0 C3TSEL1 C3TSEL0 C2TSEL2 C2TSEL1 C2TSEL0 C1TSEL2 C1TSEL1 C1TSEL0 0000 0000
F2Eh CM3CON CON COE CPOL EVPOL1 EVPOL0 CREF CCH1 CCH0 0001 1111
F2Fh CM2CON CON COE CPOL EVPOL1 EVPOL0 CREF CCH1 CCH0 0001 1111
F32h TMR12 TMR12 Register 0000 0000
F33h T10CON(3) — T10OUTPS3 T10OUTPS2 T10OUTPS1 T10OUTPS0 TMR10ON T10CKPS1 T10CKPS0 -000 0000
F35h TMR10 TMR10 Register 0000 0000
F38h TMR8 Timer8 Register 0000 0000
F39H T6CON — T6OUTPS3 T6OUTPS2 T6OUTPS1 T6OUTPS0 TMR6ON T6CKPS1 T6CKPS0 -000 0000
F3Ah PR6 Timer6 Period Register 1111 1111
F3Bh TMR6 Timer6 Register 0000 0000
F3Ch T7GCON(3) TMR7GE T7GPOL T7GTM T7GSPM T7GGO/ T7GVAL T7GSS1 T7GSS0 0000 0x00
T7DONE
F3Dh T7CON(3) TMR7CS1 TMR7CS0 T7CKPS1 T7CKPS0 — T7SYNC RD16 TMR7ON 0000 0x00
F3Eh TMR7L(3) Timer7 Register Low Byte xxxx xxxx
F3Fh TMR7H(3) Timer7 Register High Byte xxxx xxxx
F40h CCP10CON(3) — — DC10B1 DC10B0 CCP10M3 CCP10M2 CCP10M1 CCP10M0 --00 0000
F41h CCPR10L(3) Capture/Compare/PWM Register 10 Low Byte xxxx xxxx
F42h CCPR10H(3) Capture/Compare/PWM Register 10 High Byte xxxx xxxx
F43h CCP9CON(3) — — DC9B1 DC9B0 CCP9M3 CCP9M2 CCP9M1 CCP9M0 --00 0000
F44h CCPR9L(3) Capture/Compare/PWM Register 9 Low Byte xxxx xxxx
F45h CCPR9H(3) Capture/Compare/PWM Register 9 High Byte xxxx xxxx
F49h CCP3CON P3M1 P3M0 DC3B1 DC3B0 CCP3M3 CCP3M2 CCP3M1 CCP3M0 0000 0000
F4Ah CCPR3L Capture/Compare/PWM Register 3 Low Byte xxxx xxxx
F4Bh CCPR3H Capture/Compare/PWM Register 3 High Byte xxxx xxxx
F4Ch ECCP3DEL P3RSEN P3DC6 P3DC5 P3DC4 P3DC3 P3DC2 P3DC1 P3DC0 0000 0000
F4Dh ECCP3AS ECCP3ASE ECCP3AS2 ECCP3AS1 ECCP3AS0 PSS3AC1 PSS3AC0 PSS3BD1 PSS3BD0 0000 0000
F4Eh CCP2CON P2M1 P2M0 DC2B1 DC2B0 CCP2M3 CCP2M2 CCP2M1 CCP2M0 0000 0000
F4Fh CCPR2L Capture/Compare/PWM Register 2 Low Byte xxxx xxxx

PIC18F87K90 FAMILY
Value on
POR, BOR

F51h ECCP2DEL P2RSEN P2DC6 P2DC5 P2DC4 P2DC3 P2DC2 P2DC1 P2DC0 0000 0000
F52h ECCP2AS ECCP2ASE ECCP2AS2 ECCP2AS1 ECCP2AS0 PSS2AC1 PSS2AC0 PSS2BD1 PSS2BD0 0000 0000
F53h PADCFG1 RDPU REPU RJPU(2) — — RTSECSEL1 RTSECSEL0 — 000- -00-
F54h CM1CON CON COE CPOL EVPOL1 EVPOL0 CREF CCH1 CCH0 0001 1111
F55h CTMUICON ITRIM5 ITRIM4 ITRIM3 ITRIM2 ITRIM1 ITRIM0 IRNG1 IRNG1 0000 0000
F56h CTMUCONL EDG2POL EDG2SEL1 EDG2SEL0 EDG1POL EDG1SEL1 EDG1SEL0 EDG2STAT EDG1STAT 0000 0000
F57h CTMUCONH CTMUEN — CTMUSIDL TGEN EDGEN EDGSEQEN IDISSEN CTTRIG 0-00 0000
F58h ALRMVALL Alarm Value High Register Window based on APTR<1:0> 0000 0000
F59h ALRMVALH Alarm Value High Register Window based on APTR<1:0> xxxx xxxx
F5Ah ALRMRPT ARPT7 ARPT6 ARPT5 ARPT4 ARPT3 ARPT2 ARPT1 ARPT0 0000 0000
F5Bh ALRMCFG ALRMEN CHIME AMASK3 AMASK2 AMASK1 AMASK0 ALRMPTR1 ALRMPTR0 0000 0000
F5Ch RTCVALL RTCC Value Low Register Window based on RTCPTR<1:0> 0000 0000
F5Dh RTCVALH RTCC Value High Register Window based on RTCPTR<1:0> xxxx xxxx
F5Eh RTCCAL CAL7 CAL6 CAL5 CAL4 CAL3 CAL2 CAL1 CAL0 xxxx xxxx
F5Fh RTCCFG RTCEN — RTCWREN RTCSYNC HALFSEC RTCOE RTCPTR1 RTCPTR0 0-00 0000
F60h PIE6 — — — EEIE — CMP3IE CMP2IE CMP1IE ---0 -000
F61h EEDATA EEPROM Data Register 0000 0000
F62h EEADR EEPROM Address Register Low Byte 0000 0000
F63h EEADRH EEPROM Address Register High Byte ---- --00
F64h OSCCON2 — SOSCRUN — — SOSCGO — MFIOFS MFIOSEL -0-- 0-x0
F65h BAUDCON1 ABDOVF RCIDL RXDTP TXCKP BRG16 — WUE ABDEN 0000 0-x0
F66h LCDDATA0 S07C0 S06C0 S05C0 S04C0 S03C0 S02C0 S01C0 S00C0 xxxx xxxx
F6Ah LCDDATA4 S39C0 S38C0 S37C0 S36C0 S35C0 S34C0 S33C0 S32C0 xxxx xxxx
F6Bh LCDDATA5 S47C0 S46C0 S45C0 S44C0 S43C0 S42C0 S41C0 S40C0 xxxx xxxx
F6Ch LCDDATA6 S07C1 S06C1 S05C1 S04C1 S03C1 S02C1 S01C1 S00C1 xxxx xxxx
F6Dh LCDDATA7 S15C1 S14C1 S13C1 S12C1 S11C1 S10C1 S09C1 S08C1 xxxx xxxx
F6Eh LCDDATA8 S23C1 S22C1 S21C1 S20C1 S19C1 S18C1 S17C1 S16C1 xxxxxxxx
F6Fh LCDDATA9 S31C1 S30C1 S29C1 S28C1 S27C1 S26C1 S25C1 S24C1 xxxx xxxx
F70h LCDDATA10(2) S39C1(2) S38C1(2) S37C1(2) S36C1(2) S35C1(2) S34C1(2) S33C1(2) S32C1 xxxx xxxx
F71h LCDDATA11(2) S47C1 S46C1 S45C1 S44C1 S43C1 S42C1 S41C1 S40C1 xxxx xxxx
F76h LCDDATA16(2) S39C2(2) S38C2(2) S37C2(2) S36C2(2) S35C2(2) S34C2(2) S33C2(2) S32C2 xxxx xxxx
F77h LCDDATA17(2) S47C2 S46C2 S45C2 S44C2 S43C2 S42C2 S41C2 S40C2 xxxx xxxx
F7Ah LCDDATA20 S23C3 S22C3 S21C3 S20C3 S19C3 S18C3 S17C3 S16C3 xxxx xxxx
F7Bh LCDDATA21 S31C3 S30C3 S29C3 S28C3 S27C3 S26C3 S25C3 S24C3 xxxx xxxx
F7Ch LCDDATA22 S39C3(2) S38C3(2) S37C3(2) S36C3(2) S35C3(2) S34C3(2) S33C3(2) S32C3 xxxx xxxx
F7Dh LCDDATA23(2) S47C3 S46C3 S45C3 S44C3 S43C3 S42C3 S41C3 S40C3 xxxx xxxx
F7Eh EECON2 EEPROM Control Register 2 (not a physical register) ---- ----
F7Fh EECON1 EEPGD CFGS — FREE WRERR WREN WR RD xx-0 x000
F80h PORTA RA7 RA6 RA5 RA4 RA3 RA2 RA1 RA0 xxxx xxxx

PIC18F87K90 FAMILY
Value on
POR, BOR
F81h PORTB RB7 RB6 RB5 RB4 RB3 RB2 RB1 RB0 xxxx xxxx
F82h PORTC RC7 RC6 RC5 RC4 RC3 RC2 RC1 RC0 xxxx xxxx
F83h PORTD RD7 RD6 RD5 RD4 RD3 RD2 RD1 RD0 xxxx xxxx
F84h PORTE RE7 RE6 RE5 RE4 RE3 RE2 RE1 RE0 xxxx xxxx
F85h PORTF RF7 RF6 RF5 RF4 RF3 RF2 RF1 — xxxx xxx-
F86h PORTG — — RG5(1) RG4 RG3 RG2 RG1 RG0 --xx xxxx
F87h PORTH(2) RH7 RH6 RH5 RH4 RH3 RH2 RH1 RH0 xxxx xxxx
F88h PORTJ(2) RJ7 RJ6 RJ5 RJ4 RJ3 RJ2 RJ1 RJ0 xxxx xxxx
F89h LATA LATA7 LATA6 LATA5 LATA4 LATA3 LATA2 LATA1 LATA0 xxxx xxxx
F8Ah LATB LATB7 LATB6 LATB5 LATB4 LATB3 LATB2 LATB1 LATB0 xxxx xxxx
F8Bh LATC LATC7 LATC6 LATC5 LATC4 LATC3 LATC2 LATC1 LATC0 xxxx xxxx
F8Ch LATD LATD7 LATD6 LATD5 LATD4 LATD3 LATD2 LATD1 LATD0 xxxx xxxx
F8Dh LATE LATE7 LATE6 LATE5 LATE4 LATE3 LATE2 LATE1 LATE0 xxxx xxxx
F8Eh LATF LATF7 LATF6 LATF5 LATF4 LATF3 LATF2 LATF1 — xxxx xxx-
F8Fh LATG — — — LATG4 LATG3 LATG2 LATG1 LATG0 ---x xxxx
F90h LATH(2) LATH7 LATH6 LATH5 LATH4 LATH3 LATH2 LATH1 LATH0 xxxx xxxx
F91h LATJ(2) LATJ7 LATJ6 LATJ5 LATJ4 LATJ3 LATJ2 LATJ1 LATJ0 xxxx xxxx
F92h TRISA TRISA7 TRISA6 TRISA5 TRISA4 TRISA3 TRISA2 TRISA1 TRISA0 1111 1111
F93h TRISB TRISB7 TRISB6 TRISB5 TRISB4 TRISB3 TRISB2 TRISB1 TRISB0 1111 1111
F94h TRISC TRISC7 TRISC6 TRISC5 TRISC4 TRISC3 TRISC2 TRISC1 TRISC0 1111 1111
F95h TRISD TRISD7 TRISD6 TRISD5 TRISD4 TRISD3 TRISD2 TRISD1 TRISD0 1111 1111
F96h TRISE TRISE7 TRISE6 TRISE5 TRISE4 TRISE3 TRISE2 TRISE1 TRISE0 1111 1111
F97h TRISF TRISF7 TRISF6 TRISF5 TRISF4 TRISF3 TRISF2 TRISF1 — 1111 111-
F98h TRISG — — — TRISG4 TRISG3 TRISG2 TRISG1 TRISG0 ---1 1111
F99h TRISH(2) TRISH7 TRISH6 TRISH5 TRISH4 TRISH3 TRISH2 TRISH1 TRISH0 1111 1111
F9Ah TRISJ(2) TRISJ7 TRISJ6 TRISJ5 TRISJ4 TRISJ3 TRISJ2 TRISJ1 TRISJ0 1111 1111
F9Bh OSCTUNE INTSRC PLLEN TUN5 TUN4 TUN3 TUN2 TUN1 TUN0 0000 0000
F9Ch PSTR1CON CMPL1 CMPL0 — STRSYNC STRD STRC STRB STRA 00-0 0001
F9Dh PIE1 — ADIE RC1IE TX1IE SSP1IE TMR1GIE TMR2IE TMR1IE -000 0000
F9Eh PIR1 — ADIF RC1IF TX1IF SSP1IF TMR1GIF TMR2IF TMR1IF -000 0000
F9Fh IPR1 — ADIP RC1IP TX1IP SSP1IP TMR1GIP TMR2IP TMR1IP -111 1111
FA0h PIE2 OSCFIE — SSP2IE BCL2IE BCL1IE HLVDIE TMR3IE TMR3GIE 0-10 0000
FA1h PIR2 OSCFIF — SSP2IF BCL2IF BCL1IF HLVDIF TMR3IF TMR3GIF 0-10 0000
FA2h IPR2 OSCFIP — SSP2IP BCL2IP BCL1IP HLVDIP TMR3IP TMR3GIP 1-00 1110
FA3h PIE3 TMR5GIE LCDIE RC2IE TX2IE CTMUIE CCP2IE CCP1IE RTCCIE 0000 0000
FA4h PIR3 TMR5GIF LCDIF RC2IF TX2IF CTMUIF CCP2IF CCP1IF RTCCIF 0000 0000
FA5h IPR3 TMR5GIP LCDIP RC2IP TX2IP CTMUIP CCP2IP CCP1IP RTCCIP 1111 1111
FA6h PIR6 — — — EEIF — CMP3IF CMP2IF CMP1IF ---0 -000
FA7h — — — — — — — — — ---- ----
FA8h HLVDCON VDIRMAG BGVST IRVST HLVDEN HLVDL3 HLVDL2 HLVDL1 HLVDL0 0000 0000
FA9h IPR6 — — — EEIP — CMP3IP CMP2IP CMP1IP ---1 -111
FAAh T1GCON TMR1GE T1GPOL T1GTM T1GSPM T1GGO/ T1GVAL T1GSS1 T1GSS0 0000 0x00
T1DONE
FABh RCSTA1 SPEN RX9 SREN CREN ADDEN FERR OERR RX9D 0000 000x
FACh TXSTA1 CSRC TX9 TXEN SYNC SENDB BRGH TRMT TX9D 0000 0010
FADh TXREG1 USART1 Transmit Register xxxx xxxx
FAEh RCREG1 USART1 Receive Register 0000 0000
FAFh SPBRG1 USART1 Baud Rate Generator 0000 0000

PIC18F87K90 FAMILY
Value on
POR, BOR
FB0h T3GCON TMR3GE T3GPOL T3GTM T3GSPM T3GGO/ T3GVAL T3GSS1 T3GSS0 0000 0x00
T3DONE
FB1h T3CON TMR3CS1 TMR3CS0 T3CKPS1 T3CKPS0 SOSCEN T3SYNC RD16 TMR3ON 0000 0000
FB2h TMR3L Timer3 Register Low Byte xxxx xxxx
FB3h TMR3H Timer3 Register High Byte xxxx xxxx
FB4h CMSTAT CMP3OUT CMP2OUT CMP1OUT — — — — — 111- ----
FB5h CVRCON CVREN CVROE CVRSS CVR4 CVR3 CVR2 CVR1 CVR0 0000 0000
FB6h PIE4 CCP10IE(3) CCP9IE(3) CCP8IE CCP7IE(3) CCP6IE CCP5IE CCP4IE CCP3IE 0000 0000
FB7h PIR4 CCP10IF(3) CCP9IF(3) CCP8IF CCP7IF(3) CCP6IF CCP5IF CCP4IF CCP3IF 0000 0000
FB8h IPR4 CCP10IP(3) CCP9IP(3) CCP8IP CCP7IP(3) CCP6IP CCP5IP CCP4IP CCP3IP 1111 1111
FB9h PIE5 TMR7GIE(3) TMR12IE(3) TMR10IE(3) TMR8IE TMR7IE(3) TMR6IE TMR5IE TMR4IE 0000 0000
(3) (3) (3)
FBAh PIR5 TMR7GIF TMR12IF TMR10IF TMR8IF TMR7IF(3) TMR6IF TMR5IF TMR4IF 0000 0000
FBBh CCP1CON P1M1 P1M0 DC1B1 DC1B0 CCP1M3 CCP1M2 CCP1M1 CCP1M0 0000 0000
FBCh CCPR1L Capture/Compare/PWM Register 1 Low Byte xxxx xxxx
FBDh CCPR1H Capture/Compare/PWM Register 1 High Byte xxxx xxxx
FBEh ECCP1DEL P1RSEN P1DC6 P1DC5 P1DC4 P1DC3 P1DC2 P1DC1 P1DC0 0000 0000
FBFh ECCP1AS ECCP1ASE ECCP1AS2 ECCP1AS1 ECCP1AS0 PSS1AC1 PSS1AC0 PSS1BD1 PSS1BD0 0000 0000
FC0h ADCON2 ADFM — ACQT2 ACQT1 ACQT0 ADCS2 ADCS1 ADCS0 0—00 0000
FC1h ADCON1 TRIGSEL1 TRIGSEL0 VCFG1 VCFG0 VNCFG CHSN2 CHSN1 CHSN0 0000 0000
FC2h ADCON0 — CHS4 CHS3 CHS2 CHS1 CHS0 GO/DONE ADON -000 0000
FC3h ADRESL A/D Result Register Low Byte xxxx xxxx
FC4h ADRESH A/D Result Register High Byte xxxx xxxx
FC5h SSP1CON2 GCEN ACKSTAT ACKDT ACKEN RCEN PEN RSEN SEN 0000 0000
FC6h SSP1CON1 WCOL SSPOV SSPEN CKP SSPM3 SSPM2 SSPM1 SSPM0 0000 0000
FC7h SSP1STAT SMP CKE D/A P S R/W UA BF 0000 0000
FC8h SSP1ADD MSSP Address Register in I2C™ Slave Mode. SSP1 Baud Rate Reload Register in I2C Master Mode 0000 0000
FC9h SSP1BUF MSSP Receive Buffer/Transmit Register xxxx xxxx
FCAh T2CON — T2OUTPS3 T2OUTPS2 T2OUTPS1 T2OUTPS0 TMR2ON T2CKPS1 T2CKPS0 —000 0000
FCBh PR2 Timer2 Period Register 1111 1111
FCCh TMR2 Timer2 Register 0000 0000
FCDh T1CON TMR1CS1 TMR1CS0 T1CKPS1 T1CKPS0 SOSCEN T1SYNC RD16 TMR1ON 0000 0000
FCEh TMR1L Timer1 Register Low Byte xxxx xxxx
FCFh TMR1H Timer1 Register High Byte xxxx xxxx
FD0h RCON IPEN SBOREN CM RI TO PD POR BOR 0111 11qq
FD1h WDTCON REGSLP — ULPLVL SRETEN — ULPEN ULPSINK SWDTEN 0—x0 —000
FD2h IPR5 TMR7GIP(3) TMR12IP(3) TMR10I(3) P TMR8IP TMR7IP(3) TMR6IP TMR5IP TMR4IP 1111 1111
FD3h OSCCON IDLEN IRCF2 IRCF1 IRCF0 OSTS HFIOFS SCS1 SCS0 0110 q000
FD4h SPBRGH1 USART1 Baud Rate Generator High Byte 0000 0000
FD5h T0CON TMR0ON T08BIT T0CS T0SE PSA TOPS2 TOPS1 TOPS0 1111 1111
FD6h TMR0L Timer0 Register Low Byte xxxx xxxx
FD7h TMR0H Timer0 Register High Byte 0000 0000
FD8h STATUS — — — N OV Z DC C ---x xxxx
FD9h FSR2L Indirect Data Memory Address Pointer 2 Low Byte xxxx xxxx
FDAh FSR2H — — — — Indirect Data Memory Address Pointer 2 High Byte ---- xxxx
FDBh PLUSW2 Uses contents of FSR2 to address data memory – value of FSR2 pre-incremented (not a physical register) – ---- ----
value of FSR2 offset by W
FDCh PREINC2 Uses contents of FSR2 to address data memory – value of FSR2 pre--incremented (not a physical register) ---- ----
FDDh POSTDEC2 Uses contents of FSR2 to address data memory – value of FSR2 post-decremented (not a physical register) ---- ----
FDEh POSTINC2 Uses contents of FSR2 to address data memory – value of FSR2 post-incremented (not a physical register) ---- ----

PIC18F87K90 FAMILY
Value on
POR, BOR
FDFh INDF2 Uses contents of FSR2 to address data memory – value of FSR2 not changed (not a physical register) ---- ----
FE0h BSR — — — — Bank Select Register ---- 0000
FE1h FSR1L Indirect Data Memory Address Pointer 1 Low Byte xxxx xxxx
FE2h FSR1H — — — — Indirect Data Memory Address Pointer 1 High Byte ---- xxxx
FE3h PLUSW1 Uses contents of FSR1 to address data memory – value of FSR1 pre-incremented (not a physical register) – ---- ----
FE4h PREINC1 Uses contents of FSR1 to address data memory – value of FSR1 pre-incremented (not a physical register) ---- ----
FE5h POSTDEC1 Uses contents of FSR1 to address data memory – value of FSR1 post-decremented (not a physical register) ---- ----
FE6h POSTINC1 Uses contents of FSR1 to address data memory – value of FSR1 post-incremented (not a physical register) ---- ----
FE7h INDF1 Uses contents of FSR1 to address data memory – value of FSR1 not changed (not a physical register) ---- ----
FE8h WREG Working Register xxxx xxxx
FE9h FSR0L Indirect Data Memory Address Pointer 0 Low Byte xxxx xxxx
FEAh FSR0H — — — — Indirect Data Memory Address Pointer 0 High Byte ---- xxxx
FEBh PLUSW0 Uses contents of FSR0 to address data memory – value of FSR0 pre-incremented (not a physical register) – ---- ----
FECh PREINC0 Uses contents of FSR0 to address data memory – value of FSR0 pre-incremented (not a physical register) ---- ----
FEDh POSTDEC0 Uses contents of FSR0 to address data memory – value of FSR0 post-decremented (not a physical register) ---- ----
FEEh POSTINC0 Uses contents of FSR0 to address data memory – value of FSR0 post-incremented (not a physical register) ---- ----
FEFh INDF0 Uses contents of FSR0 to address data memory – value of FSR0 not changed (not a physical register) ---- ----
FF0h INTCON3 INT2IP INT1IP INT3IE INT2IE INT1IE INT3IF INT2IF INT1IF 1100 0000
FF1h INTCON2 RBPU INTEDG0 INTEDG1 INTEDG2 INTEDG3 TMR0IP INT3IP RBIP 1111 1111
FF2h INTCON GIE/GIEH PEIE/GIEL TMR0IE INT0IE RBIE TMR0IF INT0IF RBIF 0000 000x
FF3h PRODL Product Register Low Byte xxxx xxxx
FF4h PRODH Product Register High Byte xxxxxxxx
FF5h TABLAT Program Memory Table Latch 0000 0000
FF6h TBLPTRL Program Memory Table Pointer Low Byte (TBLPTR<7:0>) 0000 0000
FF7h TBLPTRH Program Memory Table Pointer High Byte (TBLPTR<15:8>) 0000 0000
FF8h TBLPTRU — — bit 21 Program Memory Table Pointer Upper Byte (TBLPTR<20:16>) --00 0000
FF9h PCL PC Low Byte (PC<7:0>) 0000 0000
FFAh PCLATH Holding Register for PC<15:8> 0000 0000
FFBh PCLATU — — — Holding Register for PC<20:16> ---0 0000
FFCh STKPTR STKFUL STKUNF — Return Stack Pointer uu-0 0000
FFDh TOSL Top-of-Stack Low Byte (TOS<7:0>) 0000 0000
FFEh TOSH Top-of-Stack High Byte (TOS<15:8>) 0000 0000
FFFh TOSU — — — Top-of-Stack Upper Byte (TOS<20:16>) ---0 0000

PIC18F87K90 FAMILY
6.3.5 STATUS REGISTER It is recommended, therefore, that only BCF, BSF,
SWAPF, MOVFF and MOVWF instructions be used to
The STATUS register, shown in Register 6-2, contains
alter the STATUS register because these instructions
the arithmetic status of the ALU. The STATUS register
do not affect the Z, C, DC, OV or N bits in the STATUS
can be the operand for any instruction, as with any
register.
other register. If the STATUS register is the destination
for an instruction that affects the Z, DC, C, OV or N bits, For other instructions not affecting any Status bits, see
the write to these five bits is disabled. the instruction set summaries in Table 29-2 and
Table 29-3.
These bits are set or cleared according to the device
logic. Therefore, the result of an instruction with the Note: The C and DC bits operate in subtraction,
STATUS register as destination may be different than as borrow and digit borrow bits, respectively.
intended. For example, CLRF STATUS will set the Z bit
but leave the other bits unchanged. The STATUS
register then reads back as ‘000u u1uu’.
REGISTER 6-2: STATUS REGISTER

U-0 U-0 U-0 R/W-x R/W-x R/W-x R/W-x R/W-x
— — — N OV Z DC(1) C(2)
bit 7 bit 0
Legend:

bit 4 N: Negative bit
This bit is used for signed arithmetic (2’s complement). It indicates whether the result was negative
(ALU MSB = 1).
1 = Result was negative
0 = Result was positive
bit 3 OV: Overflow bit
This bit is used for signed arithmetic (2’s complement). It indicates an overflow of the 7-bit magnitude
which causes the sign bit (bit 7) to change state.
1 = Overflow occurred for signed arithmetic (in this arithmetic operation)
0 = No overflow occurred
bit 2 Z: Zero bit
1 = The result of an arithmetic or logic operation is zero
0 = The result of an arithmetic or logic operation is not zero
bit 1 DC: Digit Carry/Borrow bit(1)
For ADDWF, ADDLW, SUBLW and SUBWF instructions:
1 = A carry-out from the 4th low-order bit of the result occurred
0 = No carry-out from the 4th low-order bit of the result
bit 0 C: Carry/Borrow bit(2)
For ADDWF, ADDLW, SUBLW and SUBWF instructions:
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
Note 1: For borrow, the polarity is reversed. A subtraction is executed by adding the 2’s complement of the second
operand.
2: For borrow, the polarity is reversed. A subtraction is executed by adding the 2’s complement of the second
operand.

PIC18F87K90 FAMILY
6.4 Data Addressing Modes of data RAM (see Section 6.3.3 “General Purpose
Register File”) or a location in the Access Bank (see
Note: The execution of some instructions in the Section 6.3.2 “Access Bank”).
core PIC18 instruction set are changed
The Access RAM bit ‘a’ determines how the address is
when the PIC18 extended instruction set is
interpreted. When ‘a’ is ‘1’, the contents of the BSR
enabled. For more information, see
(Section 6.3.1 “Bank Select Register”) are used with
Section 6.6 “Data Memory and the
the address to determine the complete 12-bit address
Extended Instruction Set”.
of the register. When ‘a’ is ‘0’, the address is interpreted
While the program memory can be addressed in only as being a register in the Access Bank. Addressing that
one way, through the Program Counter, information in uses the Access RAM is sometimes also known as
the data memory space can be addressed in several Direct Forced Addressing mode.
ways. For most instructions, the addressing mode is A few instructions, such as MOVFF, include the entire
fixed. Other instructions may use up to three modes, 12-bit address (either source or destination) in their
depending on which operands are used and whether or opcodes. In these cases, the BSR is ignored entirely.
not the extended instruction set is enabled.
The destination of the operation’s results is determined
The addressing modes are: by the destination bit, ‘d’. When ‘d’ is ‘1’, the results are
• Inherent stored back in the source register, overwriting its origi-
nal contents. When ‘d’ is ‘0’, the results are stored in
• Literal
the W register. Instructions without the ‘d’ argument
• Direct have a destination that is implicit in the instruction,
• Indirect either the target register is being operated on or the W
An additional addressing mode, Indexed Literal Offset, register.
is available when the extended instruction set is
enabled (XINST Configuration bit = 1). For details on 6.4.3 INDIRECT ADDRESSING
this mode’s operation, see Section 6.6.1 “Indexed Indirect Addressing allows the user to access a location
Addressing with Literal Offset”. in data memory without giving a fixed address in the
instruction. This is done by using File Select Registers
6.4.1 INHERENT AND LITERAL (FSRs) as pointers to the locations to be read or written
ADDRESSING to. Since the FSRs are themselves located in RAM as
Many PIC18 control instructions do not need any Special Function Registers, they can also be directly
argument at all. They either perform an operation that manipulated under program control. This makes FSRs
globally affects the device or they operate implicitly on very useful in implementing data structures such as
one register. This addressing mode is known as Inherent tables and arrays in data memory.
Addressing. Examples of this mode include SLEEP, The registers for Indirect Addressing are also
RESET and DAW. implemented with Indirect File Operands (INDFs) that
Other instructions work in a similar way, but require an permit automatic manipulation of the pointer value with
additional explicit argument in the opcode. This method auto-incrementing, auto-decrementing or offsetting
is known as the Literal Addressing mode because the with another value. This allows for efficient code using
instructions require some literal value as an argument. loops, such as the example of clearing an entire RAM
Examples of this include ADDLW and MOVLW, which bank in Example 6-5. It also enables users to perform
respectively, add or move a literal value to the W Indexed Addressing and other Stack Pointer
register. Other examples include CALL and GOTO, operations for program memory in data memory.
which include a 20-bit program memory address.
EXAMPLE 6-5: HOW TO CLEAR RAM
6.4.2 DIRECT ADDRESSING (BANK 1) USING
Direct Addressing specifies all or part of the source INDIRECT ADDRESSING
and/or destination address of the operation within the LFSR FSR0, 100h ;
opcode itself. The options are specified by the NEXT CLRF POSTINC0 ; Clear INDF
arguments accompanying the instruction. ; register then
; inc pointer
In the core PIC18 instruction set, bit-oriented and BTFSS FSR0H, 1 ; All done with
byte-oriented instructions use some version of Direct ; Bank1?
Addressing by default. All of these instructions include BRA NEXT ; NO, clear next
some 8-bit literal address as their Least Significant CONTINUE ; YES, continue
Byte. This address specifies the instruction’s data
source as either a register address in one of the banks

PIC18F87K90 FAMILY
6.4.3.1 FSR Registers and the are mapped in the SFR space, but are not physically
INDF Operand implemented. Reading or writing to a particular INDF
register actually accesses its corresponding FSR
At the core of Indirect Addressing are three sets of
register pair. A read from INDF1, for example, reads
registers: FSR0, FSR1 and FSR2. Each represents a
the data at the address indicated by FSR1H:FSR1L.
pair of 8-bit registers: FSRnH and FSRnL. The four
upper bits of the FSRnH register are not used, so each Instructions that use the INDF registers as operands
FSR pair holds a 12-bit value. This represents a value actually use the contents of their corresponding FSR as
that can address the entire range of the data memory a pointer to the instruction’s target. The INDF operand
in a linear fashion. The FSR register pairs, then, serve is just a convenient way of using the pointer.
as pointers to data memory locations. Because Indirect Addressing uses a full 12-bit address,
Indirect Addressing is accomplished with a set of data RAM banking is not necessary. Thus, the current
Indirect File Operands, INDF0 through INDF2. These contents of the BSR and the Access RAM bit have no
can be thought of as “virtual” registers. The operands effect on determining the target address.
FIGURE 6-8: INDIRECT ADDRESSING

000h
Using an instruction with one of the ADDWF, INDF1, 1 Bank 0
Indirect Addressing registers as the 100h
operand.... Bank 1
200h
Bank 2
300h
...uses the 12-bit address stored in FSR1H:FSR1L
the FSR pair associated with that
7 0 7 0
register....
x x x x 1 1 1 1 1 1 0 0 1 1 0 0 Bank 3
through
Bank 13
...to determine the data memory

location to be used in that operation.
In this case, the FSR1 pair contains E00h
FCCh. This means the contents of Bank 14
location, FCCh, will be added to that F00h
of the W register and stored back in Bank 15
FCCh. FFFh
Data Memory

PIC18F87K90 FAMILY
6.4.3.2 FSR Registers and POSTINC, 6.4.3.3 Operations by FSRs on FSRs
POSTDEC, PREINC and PLUSW Indirect Addressing operations that target other FSRs
In addition to the INDF operand, each FSR register pair or virtual registers represent special cases. For
also has four additional indirect operands. Like INDF, example, using an FSR to point to one of the virtual
these are “virtual” registers that cannot be indirectly registers will not result in successful operations.
read or written to. Accessing these registers actually As a specific case, assume that the FSR0H:FSR0L
accesses the associated FSR register pair, but also registers contain FE7h, the address of INDF1.
performs a specific action on its stored value. Attempts to read the value of the INDF1, using INDF0
These operands are: as an operand, will return 00h. Attempts to write to
INDF1, using INDF0 as the operand, will result in a
• POSTDEC – Accesses the FSR value, then
NOP.
automatically decrements it by ‘1’ afterwards
• POSTINC – Accesses the FSR value, then On the other hand, using the virtual registers to write to
automatically increments it by ‘1’ afterwards an FSR pair may not occur as planned. In these cases,
the value will be written to the FSR pair, but without any
• PREINC – Increments the FSR value by ‘1’, then
incrementing or decrementing. Thus, writing to INDF2
uses it in the operation
or POSTDEC2 will write the same value to the
• PLUSW – Adds the signed value of the W register FSR2H:FSR2L.
(range of -127 to 128) to that of the FSR and uses
the new value in the operation Since the FSRs are physical registers mapped in the
SFR space, they can be manipulated through all direct
In this context, accessing an INDF register uses the operations. Users should proceed cautiously when
value in the FSR registers without changing them. working on these registers, however, particularly if their
Similarly, accessing a PLUSW register gives the FSR code uses Indirect Addressing.
value, offset by the value in the W register – with
neither value actually changed in the operation. Similarly, operations by Indirect Addressing are
Accessing the other virtual registers changes the value generally permitted on all other SFRs. Users should
of the FSR registers. exercise the appropriate caution, so that they do not
inadvertently change settings that might affect the
Operations on the FSRs with POSTDEC, POSTINC operation of the device.
and PREINC affect the entire register pair. Rollovers of
the FSRnL register, from FFh to 00h, carry over to the
FSRnH register. On the other hand, results of these
6.5 Program Memory and the
operations do not change the value of any flags in the Extended Instruction Set
STATUS register (for example, Z, N and OV bits). The operation of program memory is unaffected by the
The PLUSW register can be used to implement a form use of the extended instruction set.
of Indexed Addressing in the data memory space. By Enabling the extended instruction set adds five
manipulating the value in the W register, users can additional two-word commands to the existing PIC18
reach addresses that are fixed offsets from pointer instruction set: ADDFSR, CALLW, MOVSF, MOVSS and
addresses. In some applications, this can be used to SUBFSR. These instructions are executed as described
implement some powerful program control structure, in Section 6.2.4 “Two-Word Instructions”.
such as software stacks, inside of data memory.

PIC18F87K90 FAMILY
6.6 Data Memory and the Extended Under these conditions, the file address of the
Instruction Set instruction is not interpreted as the lower byte of an
address (used with the BSR in Direct Addressing) or as
Enabling the PIC18 extended instruction set (XINST an 8-bit address in the Access Bank. Instead, the value
Configuration bit = 1) significantly changes certain is interpreted as an offset value to an Address Pointer
aspects of data memory and its addressing. Using the specified by FSR2. The offset and the contents of FSR2
Access Bank for many of the core PIC18 instructions are added to obtain the target address of the operation.
introduces a new addressing mode for the data memory
space. This mode also alters the behavior of Indirect 6.6.2 INSTRUCTIONS AFFECTED BY
Addressing using FSR2 and its associated operands. INDEXED LITERAL OFFSET MODE
What does not change is just as important. The size of Any of the core PIC18 instructions that can use Direct
the data memory space is unchanged, as well as its Addressing are potentially affected by the Indexed
linear addressing. The SFR map remains the same. Literal Offset Addressing mode. This includes all
Core PIC18 instructions can still operate in both Direct byte-oriented and bit-oriented instructions, or almost
and Indirect Addressing mode. Inherent and literal one-half of the standard PIC18 instruction set. Instruc-
instructions do not change at all. Indirect Addressing tions that only use Inherent or Literal Addressing
with FSR0 and FSR1 also remains unchanged. modes are unaffected.
Additionally, byte-oriented and bit-oriented instructions
6.6.1 INDEXED ADDRESSING WITH
are not affected if they do not use the Access Bank
LITERAL OFFSET
(Access RAM bit = 1) or include a file address of 60h or
Enabling the PIC18 extended instruction set changes above. Instructions meeting these criteria will continue
the behavior of Indirect Addressing using the FSR2 to execute as before. A comparison of the different
register pair and its associated file operands. Under the possible addressing modes, when the extended
proper conditions, instructions that use the Access instruction set is enabled, is shown in Figure 6-9.
Bank – that is, most bit-oriented and byte-oriented
Those who desire to use byte-oriented or bit-oriented
instructions – can invoke a form of Indexed Addressing
instructions in the Indexed Literal Offset mode should
using an offset specified in the instruction. This special
note the changes to assembler syntax for this mode.
addressing mode is known as Indexed Addressing with
This is described in more detail in Section 29.2.1
Literal Offset or the Indexed Literal Offset mode.
“Extended Instruction Syntax”.
When using the extended instruction set, this
addressing mode requires the following:
• Use of the Access Bank (‘a’ = 0)
• A file address argument that is less than or equal
to 5Fh

PIC18F87K90 FAMILY
FIGURE 6-9: COMPARING ADDRESSING OPTIONS FOR BIT-ORIENTED AND
BYTE-ORIENTED INSTRUCTIONS (EXTENDED INSTRUCTION SET ENABLED)
EXAMPLE INSTRUCTION: ADDWF, f, d, a (Opcode: 0010 01da ffff ffff)
000h
When a = 0 and f  60h:
The instruction executes in 060h
Direct Forced mode. ‘f’ is Bank 0
interpreted as a location in the 100h
Access RAM, between 060h 00h
and FFFh. This is the same as Bank 1 60h
locations, F60h to FFFh through
Bank 14 Valid Range
(Bank 15), of data memory. for ‘f’
Locations below 060h are not FFh
F00h
available in this addressing Access RAM
Bank 15
mode.
F40h
SFRs
FFFh
Data Memory
When a = 0 and f5Fh: 000h

The instruction executes in Bank 0
Indexed Literal Offset mode. ‘f’ 060h
is interpreted as an offset to the

100h
address value in FSR2. The 001001da ffffffff
two are added together to
Bank 1
obtain the address of the target through
register for the instruction. The Bank 14
address can be anywhere in FSR2H FSR2L
the data memory space.
F00h
Note that in this mode, the Bank 15
correct syntax is now: F40h
ADDWF [k], d SFRs
where ‘k’ is the same as ‘f’. FFFh
Data Memory
BSR
When a = 1 (all values of f): 000h 00000000
Bank 0
The instruction executes in
060h
Direct mode (also known as
Direct Long mode). ‘f’ is 100h
interpreted as a location in
one of the 16 banks of the data Bank 1 001001da ffffffff
memory space. The bank is through
Bank 14
designated by the Bank Select
Register (BSR). The address
can be in any implemented F00h
bank in the data memory Bank 15
space. F40h
SFRs
FFFh
Data Memory

PIC18F87K90 FAMILY
6.6.3 MAPPING THE ACCESS BANK IN Remapping the Access Bank applies only to operations
INDEXED LITERAL OFFSET MODE using the Indexed Literal Offset mode. Operations that
use the BSR (Access RAM bit = 1) will continue to use
The use of Indexed Literal Offset Addressing mode
Direct Addressing as before. Any Indirect or Indexed
effectively changes how the lower part of Access RAM
Addressing operation that explicitly uses any of the
(00h to 5Fh) is mapped. Rather than containing just the
indirect file operands (including FSR2) will continue to
contents of the bottom part of Bank 0, this mode maps
operate as standard Indirect Addressing. Any
the contents from Bank 0 and a user-defined “window”
instruction that uses the Access Bank, but includes a
that can be located anywhere in the data memory
register address of greater than 05Fh, will use Direct
space.
Addressing and the normal Access Bank map.
The value of FSR2 establishes the lower boundary of
the addresses mapped into the window, while the 6.6.4 BSR IN INDEXED LITERAL
upper boundary is defined by FSR2 plus 95 (5Fh). OFFSET MODE
Addresses in the Access RAM above 5Fh are mapped
Although the Access Bank is remapped when the
as previously described. (See Section 6.3.2 “Access
extended instruction set is enabled, the operation of the
Bank”.) An example of Access Bank remapping in this
BSR remains unchanged. Direct Addressing, using the
addressing mode is shown in Figure 6-10.
BSR to select the data memory bank, operates in the
same manner as previously described.
FIGURE 6-10: REMAPPING THE ACCESS BANK WITH INDEXED LITERAL

OFFSET ADDRESSING
Example Situation:
ADDWF f, d, a 000h
Not Accessible
FSR2H:FSR2L = 120h 05Fh
Locations in the region Bank 0

from the FSR2 Pointer 100h
(120h) to the pointer plus 120h
05Fh (17Fh) are mapped Window
17Fh 00h
to the bottom of the Bank 1
Access RAM (000h-05Fh). 200h Bank 1 “Window”
Special Function Registers 5Fh
60h
at F60h through FFFh are
mapped to 60h through Bank 2
FFh, as usual. through SFRs
Bank 0 addresses below Bank 14
5Fh are not available in FFh
this mode. They can still Access Bank
be addressed by using the F00h
BSR.
Bank 15
F60h
SFRs
FFFh
Data Memory

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
7.0 FLASH PROGRAM MEMORY 7.1 Table Reads and Table Writes
The Flash program memory is readable, writable and In order to read and write program memory, there are
erasable during normal operation over the entire VDD two operations that allow the processor to move bytes
range. between the program memory space and the data RAM:
A read from program memory is executed on one byte • Table Read (TBLRD)
at a time. For execution of a write to, or erasure of, • Table Write (TBLWT)
program memory: The program memory space is 16 bits wide, while the
• Memory of 32 Kbytes and 64 Kbytes data RAM space is 8 bits wide. Table reads and table
(PIC18FX5K90 and PIC18FX6K90 devices) – writes move data between these two memory spaces
Blocks of 64 bytes through an 8-bit register (TABLAT).
• Memory of 128 Kbytes (PIC18FX7K90 devices) – Table read operations retrieve data from program
Blocks of 128 bytes memory and place it into the data RAM space.
Writing or erasing program memory will cease Figure 7-1 shows the operation of a table read with
instruction fetches until the operation is complete. The program memory and data RAM.
program memory cannot be accessed during the write Table write operations store data from the data memory
or erase, therefore, code cannot execute. An internal space into holding registers in program memory. The
programming timer terminates program memory writes procedure to write the contents of the holding registers
and erases. into program memory is detailed in Section 7.5 “Writing
A value written to program memory does not need to be to Flash Program Memory”. Figure 7-2 shows the
a valid instruction. Executing a program memory operation of a table write with program memory and data
location that forms an invalid instruction results in a RAM.
NOP. Table operations work with byte entities. A table block
containing data, rather than program instructions, is not
required to be word-aligned. Therefore, a table block can
start and end at any byte address. If a table write is being
used to write executable code into program memory,
program instructions will need to be word-aligned.
FIGURE 7-1: TABLE READ OPERATION
Instruction: TBLRD*
Program Memory
Table Pointer(1)
Table Latch (8-bit)
TBLPTRU TBLPTRH TBLPTRL
TABLAT
Program Memory
(TBLPTR)
Note 1: The Table Pointer register points to a byte in program memory.

PIC18F87K90 FAMILY
FIGURE 7-2: TABLE WRITE OPERATION
Instruction: TBLWT*
Program Memory
Holding Registers
Table Pointer(1) Table Latch (8-bit)
TBLPTRU TBLPTRH TBLPTRL TABLAT
Program Memory
(TBLPTR)
Note 1: The Table Pointer actually points to one of 64 holding registers; the address of which is determined by
TBLPTRL<5:0>. The process for physically writing data to the program memory array is discussed in
Section 7.5 “Writing to Flash Program Memory”.
7.2 Control Registers The FREE bit, when set, allows a program memory
erase operation. When FREE is set, the erase
Several control registers are used in conjunction with operation is initiated on the next WR command. When
the TBLRD and TBLWT instructions. These include the: FREE is clear, only writes are enabled.
• EECON1 register The WREN bit, when set, allows a write operation. On
• EECON2 register power-up, the WREN bit is clear. The WRERR bit is set
• TABLAT register in hardware when the WR bit is set and cleared when
• TBLPTR registers the internal programming timer expires and the write
operation is complete.
7.2.1 EECON1 AND EECON2 REGISTERS
Note: During normal operation, the WRERR is
The EECON1 register (Register 7-1) is the control read as ‘1’. This can indicate that a write
register for memory accesses. The EECON2 register, operation was prematurely terminated by
not a physical register, is used exclusively in the a Reset, or a write operation was
memory write and erase sequences. Reading attempted improperly.
EECON2 will read all ‘0’s.
The EEPGD control bit determines if the access is a The WR control bit initiates write operations. The bit
program or data EEPROM memory access. When cannot be cleared, only set, in software. It is cleared in
clear, any subsequent operations operate on the data hardware at the completion of the write operation.
EEPROM memory. When set, any subsequent Note: The EEIF interrupt flag bit (PIR6<4>) is
operations operate on the program memory. set when the write is complete. It must be
The CFGS control bit determines if the access is to the cleared in software.
Configuration/Calibration registers or to program
memory/data EEPROM memory. When set,
subsequent operations operate on Configuration
registers regardless of EEPGD (see Section 28.0
“Special Features of the CPU”). When clear, memory
selection access is determined by EEPGD.

PIC18F87K90 FAMILY
REGISTER 7-1: EECON1: EEPROM CONTROL REGISTER 1
R/W-x R/W-x U-0 R/W-0 R/W-x R/W-0 R/S-0 R/S-0
EEPGD CFGS — FREE WRERR(1) WREN WR RD
bit 7 bit 0
Legend: S = Settable bit

bit 7 EEPGD: Flash Program or Data EEPROM Memory Select bit

1 = Access Flash program memory
0 = Access data EEPROM memory
bit 6 CFGS: Flash Program/Data EEPROM or Configuration Select bit
1 = Access Configuration registers
0 = Access Flash program or data EEPROM memory
bit 4 FREE: Flash Block Erase Enable bit
1 = Erase the program memory row addressed by TBLPTR on the next WR command
(cleared by completion of erase operation)
0 = Perform write-only
bit 3 WRERR: Flash Program/Data EEPROM Error Flag bit(1)
1 = A write operation is prematurely terminated (any Reset during self-timed programming in normal
operation or an improper write attempt)
0 = The write operation completed
bit 2 WREN: Flash Program/Data EEPROM Write Enable bit
1 = Allows write cycles to Flash program/data EEPROM
0 = Inhibits write cycles to Flash program/data EEPROM
bit 1 WR: Write-Control bit
1 = Initiates a data EEPROM erase/write cycle, or a program memory erase cycle or write cycle
(The operation is self-timed and the bit is cleared by hardware once write is complete.
The WR bit can only be set (not cleared) in software.)
0 = Write cycle to the EEPROM is complete
bit 0 RD: Read Control bit
1 = Initiates an EEPROM read (Read takes one cycle. RD is cleared in hardware. The RD bit can only
be set (not cleared) in software. The RD bit cannot be set when EEPGD = 1 or CFGS = 1.)
0 = Does not initiate an EEPROM read
Note 1: When a WRERR occurs, the EEPGD and CFGS bits are not cleared. This allows tracing of the error
condition.

PIC18F87K90 FAMILY
7.2.2 TABLAT – TABLE LATCH REGISTER 7.2.4 TABLE POINTER BOUNDARIES
The Table Latch (TABLAT) is an 8-bit register mapped The TBLPTR is used in reads, writes and erases of the
into the SFR space. The Table Latch register is used to Flash program memory.
hold 8-bit data during data transfers between program When a TBLRD is executed, all 22 bits of the TBLPTR
memory and data RAM. determine which byte is read from program memory
into the TABLAT.
7.2.3 TBLPTR – TABLE POINTER
REGISTER When a TBLWT is executed, the six LSbs of the Table
Pointer register (TBLPTR<5:0>) determine which of
The Table Pointer (TBLPTR) register addresses a byte the 64 program memory holding registers is written to.
within the program memory. The TBLPTR is comprised When the timed write to program memory begins (via
of three SFR registers: Table Pointer Upper Byte, Table the WR bit), the 16 MSbs of the TBLPTR
Pointer High Byte and Table Pointer Low Byte (TBLPTR<21:6>) determine which program memory
(TBLPTRU:TBLPTRH:TBLPTRL). These three regis- block of 64 bytes is written to. For more details, see
ters join to form a 22-bit wide pointer. The low-order Section 7.5 “Writing to Flash Program Memory”.
21 bits allow the device to address up to 2 Mbytes of
program memory space. The 22nd bit allows access to When an erase of program memory is executed, the
the Device ID, the User ID and the Configuration bits. 16 MSbs of the Table Pointer register (TBLPTR<21:6>)
point to the 64-byte block that will be erased. The Least
The Table Pointer register, TBLPTR, is used by the Significant bits (TBLPTR<5:0>) are ignored.
TBLRD and TBLWT instructions. These instructions can
update the TBLPTR in one of four ways, based on the Figure 7-3 describes the relevant boundaries of
table operation. These operations are shown in TBLPTR based on Flash program memory operations.
Table 7-1 and only affect the low-order 21 bits.
TABLE 7-1: TABLE POINTER OPERATIONS WITH TBLRD AND TBLWT INSTRUCTIONS
Example Operation on Table Pointer
TBLRD*
TBLPTR is not modified
TBLWT*
TBLRD*+
TBLPTR is incremented after the read/write
TBLWT*+
TBLRD*-
TBLPTR is decremented after the read/write
TBLWT*-
TBLRD+*
TBLPTR is incremented before the read/write
TBLWT+*
FIGURE 7-3: TABLE POINTER BOUNDARIES BASED ON OPERATION
21 TBLPTRU 16 15 TBLPTRH 8 7 TBLPTRL 0
TABLE ERASE/WRITE TABLE WRITE

TBLPTR<21:6> TBLPTR<5:0>
TABLE READ – TBLPTR<21:0>

PIC18F87K90 FAMILY
7.3 Reading the Flash Program The TBLPTR points to a byte address in program
Memory memory space. Executing TBLRD places the byte
pointed to into TABLAT. In addition, TBLPTR can be
The TBLRD instruction is used to retrieve data from modified automatically for the next table read
program memory and places it into data RAM. Table operation.
reads from program memory are performed one byte at
The internal program memory is typically organized by
a time.
words. The Least Significant bit of the address selects
between the high and low bytes of the word. Figure 7-4
shows the interface between the internal program
memory and the TABLAT.
FIGURE 7-4: READS FROM FLASH PROGRAM MEMORY
Program Memory
(Even Byte Address) (Odd Byte Address)
TBLPTR = xxxxx1 TBLPTR = xxxxx0
Instruction Register TABLAT

(IR)
FETCH TBLRD Read Register
EXAMPLE 7-1: READING A FLASH PROGRAM MEMORY WORD

BCF EECON1, CFGS ; point to Flash program memory
BSF EECON1, EEPGD ; access Flash program memory
MOVLW CODE_ADDR_UPPER ; Load TBLPTR with the base
MOVWF TBLPTRU ; address of the word
MOVLW CODE_ADDR_HIGH
MOVWF TBLPTRH
MOVLW CODE_ADDR_LOW
MOVWF TBLPTRL
READ_WORD
TBLRD*+ ; read into TABLAT and increment
MOVF TABLAT, W ; get data
MOVWF WORD_EVEN
TBLRD*+ ; read into TABLAT and increment
MOVF WORD_ODD

PIC18F87K90 FAMILY
7.4 Erasing Flash Program Memory 7.4.1 FLASH PROGRAM MEMORY
ERASE SEQUENCE
The erase block is 32 words or 64 bytes for the
PIC18FX5K90 and PIC18FX6K90 devices, and The sequence of events for erasing a block of internal
64 words or 128 bytes for the PIC18FX7K90 devices. program memory location is:
Word erase in the Flash array is not supported. 1. Load the Table Pointer register with the address
When initiating an erase sequence from the micro- of the row to be erased.
controller itself, a block of 64 or 128 bytes of program 2. Set the EECON1 register for the erase operation:
memory is erased. The Most Significant 16 bits of the • Set the EEPGD bit to point to program memory
TBLPTR<21:6> point to the block being erased. The • Clear the CFGS bit to access program memory
TBLPTR<5:0> bits are ignored.
• Set the WREN bit to enable writes
The EECON1 register commands the erase operation. • Set the FREE bit to enable the erase
The EEPGD bit must be set to point to the Flash
3. Disable the interrupts.
program memory. The WREN bit must be set to enable
write operations. The FREE bit is set to select an erase 4. Write 0x55 to EECON2.
operation. 5. Write 0xAA to EECON2.
For protection, the write initiate sequence for EECON2 6. Set the WR bit.
must be used. This begins the row erase cycle.
A long write is necessary for erasing the internal Flash. The CPU will stall for the duration of the erase
Instruction execution is halted while in a long write for TIW. (See Parameter D133A.)
cycle. The long write will be terminated by the internal 7. Re-enable interrupts.
programming timer.
EXAMPLE 7-2: ERASING A FLASH PROGRAM MEMORY ROW

MOVLW CODE_ADDR_UPPER ; load TBLPTR with the base
MOVWF TBLPTRU ; address of the memory block
MOVWF TBLPTRH
MOVLW CODE_ADDR_LOW
MOVWF TBLPTRL
ERASE_ROW
BSF EECON1, EEPGD ; point to Flash program memory
BCF EECON1, CFGS ; access Flash program memory
BSF EECON1, WREN ; enable write to memory
BSF EECON1, FREE ; enable Row Erase operation
BCF INTCON, GIE ; disable interrupts
Required MOVLW 0x55
Sequence MOVWF EECON2 ; write 55h
MOVLW 0xAA
MOVWF EECON2 ; write 0AAh
BSF EECON1, WR ; start erase (CPU stall)
BSF INTCON, GIE ; re-enable interrupts

PIC18F87K90 FAMILY
7.5 Writing to Flash Program Memory The long write is necessary for programming the inter-
nal Flash. Instruction execution is halted while in a long
The programming block is 32 words or 64 bytes for write cycle. The long write is terminated by the internal
PIC18FX5K90 and PIC18FX6K90 devices, and programming timer.
64 words or 128 bytes for PIC18FX7K90 devices.
Word or byte programming is not supported. The EEPROM on-chip timer controls the write time.
The write/erase voltages are generated by an on-chip
Table writes are used internally to load the holding charge pump, rated to operate over the voltage range
registers needed to program the Flash memory. There of the device.
are 64 holding registers for PIC18FX5K90 and
PIC18FX6K90 devices and 128 holding registers for Note: The default value of the holding registers on
PIC18FX7K90 used by the table writes for programming. device Resets, and after write operations, is
FFh. A write of FFh to a holding register does
Since the Table Latch (TABLAT) is only a single byte, the
not modify that byte. This means that individ-
TBLWT instruction may need to be executed 64 times for
ual bytes of program memory may be
each programming operation. All of the table write oper-
modified, provided that the change does not
ations will essentially be short writes because only the
attempt to change any bit from a ‘0’ to a ‘1’.
holding registers are written. At the end of updating the
64 or 128 holding registers, the EECON1 register must When modifying individual bytes, it is not nec-
be written to in order to start the programming operation essary to load all 64 or 128 holding registers
with a long write. before executing a write operation.
FIGURE 7-5: TABLE WRITES TO FLASH PROGRAM MEMORY
TABLAT
Write Register
8 8 8 8
TBLPTR = xxxxx0 TBLPTR = xxxxx1 TBLPTR = xxxxx2 TBLPTR = xxxx3F
Holding Register Holding Register Holding Register Holding Register
Program Memory
7.5.1 FLASH PROGRAM MEMORY WRITE 9. Write 0x55 to EECON2.

SEQUENCE 10. Write 0xAA to EECON2.
11. Set the WR bit. This will begin the write cycle.
The sequence of events for programming an internal
program memory location should be: The CPU will stall for the duration of the write for
TIW. (See Parameter D133A.)
1. Read the 64 or 128 bytes into RAM. 12. Re-enable the interrupts.
2. Update the data values in RAM as necessary. 13. Verify the memory (table read).
3. Load the Table Pointer register with the address
An example of the required code is shown in
being erased.
Example 7-3.
4. Execute the row erase procedure.
5. Load the Table Pointer register with the address Note: Before setting the WR bit, the Table
of the first byte being written. Pointer address needs to be within the
6. Write the 64 or 128 bytes into the holding intended address range of 64 or 128 bytes
registers with auto-increment. in the holding register.
7. Set the EECON1 register for the write operation:
• Set the EEPGD bit to point to program memory Note: Self-write execution to Flash and
• Clear the CFGS bit to access program memory EEPROM memory cannot be done while
• Set WREN to enable byte writes running in LP Oscillator mode (Low-Power
8. Disable the interrupts. mode). Therefore, executing a self-write
will put the device into High-Power mode.

PIC18F87K90 FAMILY
EXAMPLE 7-3: WRITING TO FLASH PROGRAM MEMORY
MOVLW SIZE_OF_BLOCK ; number of bytes in erase block
MOVWF COUNTER
MOVLW BUFFER_ADDR_HIGH ; point to buffer
MOVWF FSR0H
MOVLW BUFFER_ADDR_LOW
MOVWF FSR0L
MOVLW CODE_ADDR_UPPER ; Load TBLPTR with the base
MOVWF TBLPTRH
MOVLW CODE_ADDR_LOW
MOVWF TBLPTRL
READ_BLOCK
TBLRD*+ ; read into TABLAT, and inc
MOVWF POSTINC0 ; store data
DECFSZ COUNTER ; done?
BRA READ_BLOCK ; repeat
MODIFY_WORD
MOVLW DATA_ADDR_HIGH ; point to buffer
MOVWF FSR0H
MOVLW DATA_ADDR_LOW
MOVWF FSR0L
MOVLW NEW_DATA_LOW ; update buffer word
MOVWF POSTINC0
MOVLW NEW_DATA_HIGH
MOVWF INDF0
ERASE_BLOCK
MOVLW CODE_ADDR_UPPER ; load TBLPTR with the base
MOVWF TBLPTRH
MOVLW CODE_ADDR_LOW
MOVWF TBLPTRL
BSF EECON1, FREE ; enable Row Erase operation
MOVLW 0x55
Required MOVWF EECON2 ; write 55h
Sequence MOVLW 0xAA
BSF EECON1, WR ; start erase (CPU stall)
TBLRD*- ; dummy read decrement
MOVLW BUFFER_ADDR_HIGH ; point to buffer
MOVWF FSR0H
MOVLW BUFFER_ADDR_LOW
MOVWF FSR0L
WRITE_BUFFER_BACK
MOVLW SIZE_OF_BLOCK ; number of bytes in holding register
MOVWF COUNTER
WRITE_BYTE_TO_HREGS
MOVFF POSTINC0, WREG ; get low byte of buffer data
MOVWF TABLAT ; present data to table latch
TBLWT+* ; write data, perform a short write
; to internal TBLWT holding register.
DECFSZ COUNTER ; loop until buffers are full
GOTO WRITE_BYTE_TO_HREGS

PIC18F87K90 FAMILY
EXAMPLE 7-3: WRITING TO FLASH PROGRAM MEMORY (CONTINUED)
PROGRAM_MEMORY
MOVLW 0x55
Required MOVWF EECON2 ; write 55h
Sequence MOVLW 0xAA
BSF EECON1, WR ; start program (CPU stall)
BCF EECON1, WREN ; disable write to memory
7.5.2 WRITE VERIFY 7.5.4 PROTECTION AGAINST

Depending on the application, good programming SPURIOUS WRITES
practice may dictate that the value written to the To protect against spurious writes to Flash program
memory should be verified against the original value. memory, the write initiate sequence must also be
This should be used in applications where excessive followed. See Section 28.0 “Special Features of the
writes can stress bits near the specification limit. CPU” for more details.
7.5.3 UNEXPECTED TERMINATION OF 7.6 Flash Program Operation During

WRITE OPERATION
Code Protection
If a write is terminated by an unplanned event, such as
loss of power or an unexpected Reset, the memory See Section 28.6 “Program Verification and Code
location just programmed should be verified and repro- Protection” for details on code protection of Flash
grammed if needed. If the write operation is interrupted program memory.
by a MCLR Reset or a WDT Time-out Reset during
normal operation, the user can check the WRERR bit
and rewrite the location(s) as needed.
TABLE 7-2: REGISTERS ASSOCIATED WITH PROGRAM FLASH MEMORY

Reset
Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Values on
Page:
TBLPTRU — — bit 21(1) Program Memory Table Pointer Upper Byte (TBLPTR<20:16>) 75
TBPLTRH Program Memory Table Pointer High Byte (TBLPTR<15:8>) 75
TBLPTRL Program Memory Table Pointer Low Byte (TBLPTR<7:0>) 75
TABLAT Program Memory Table Latch 75
INTCON GIE/GIEH PEIE/GIEL TMR0IE INT0IE RBIE TMR0IF INT0IF RBIF 75
EECON2 EEPROM Control Register 2 (not a physical register) 79
EECON1 EEPGD CFGS — FREE WRERR WREN WR RD 79
IPR6 — — — EEIP — CMP3IP CMP2IP CMP1IP 77
PIR6 — — — EEIF — CMP3IF CMP2IF CMP1IF 77
PIE6 — — — EEIE — CMP3IE CMP2IE CMP1IE 80
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used during Flash/EEPROM access.
Note 1: Bit 21 of TBLPTRU allows access to the device Configuration bits.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
8.0 DATA EEPROM MEMORY 8.2 EECON1 and EECON2 Registers
The data EEPROM is a nonvolatile memory array, Access to the data EEPROM is controlled by two
separate from the data RAM and program memory, that registers: EECON1 and EECON2. These are the same
is used for long-term storage of program data. The registers which control access to the program memory
PIC18F87K90 family of devices has a 1024-byte data and are used in a similar manner for the data
EEPROM. It is not directly mapped in either the register EEPROM.
file or program memory space, but is indirectly The EECON1 register (Register 8-1) is the control
addressed through the Special Function Registers register for data and program memory access. Control
(SFRs). The EEPROM is readable and writable during bit, EEPGD, determines if the access will be to program
normal operation over the entire VDD range. memory or data EEPROM memory. When clear,
Five SFRs are used to read and write to the data operations will access the data EEPROM memory.
EEPROM, as well as the program memory. They are: When set, program memory is accessed.
• EECON1 Control bit, CFGS, determines if the access will be to
• EECON2 the Configuration registers or to program memory/data
• EEDATA EEPROM memory. When set, subsequent operations
• EEADR access Configuration registers. When CFGS is clear,
the EEPGD bit selects either program Flash or data
• EEADRH
EEPROM memory.
The data EEPROM allows byte read and write. When
The WREN bit, when set, will allow a write operation.
interfacing to the data memory block, EEDATA holds
On power-up, the WREN bit is clear. The WRERR bit is
the 8-bit data for read/write and the EEADRH:EEADR
set in hardware when the WREN bit is set, and cleared,
register pair holds the address of the EEPROM location
when the internal programming timer expires and the
being accessed.
write operation is complete.
The EEPROM data memory is rated for high erase/write
cycle endurance. A byte write automatically erases the Note: During normal operation, the WRERR is
location and writes the new data (erase-before-write). read as ‘1’. This can indicate that a write
The write time is controlled by an on-chip timer; it will operation was prematurely terminated by
vary with voltage and temperature, as well as from chip- a Reset, or a write operation was
to-chip. Please refer to Parameter D122 (Table 31-1 in attempted improperly.
Section 31.0 “Electrical Characteristics”) for exact The WR control bit initiates write operations. The bit
limits. cannot be cleared, only set, in software; it is cleared in
hardware at the completion of the write operation.
8.1 EEADR and EEADRH Registers
Note: The EEIF interrupt flag bit (PIR6<4>) is
The EEADRH:EEADR register pair is used to address set when the write is complete. It must be
the data EEPROM for read and write operations. cleared in software.
EEADRH holds the two MSbs of the address; the upper
Control bits, RD and WR, start read and erase/write
6 bits are ignored. The 10-bit range of the pair can
operations, respectively. These bits are set by firmware
address a memory range of 1024 bytes (00h to 3FFh).
and cleared by hardware at the completion of the
operation.
The RD bit cannot be set when accessing program
memory (EEPGD = 1). Program memory is read using
table read instructions. See Section 7.1 “Table Reads
and Table Writes” regarding table reads.
The EECON2 register is not a physical register. It is
used exclusively in the memory write and erase
sequences. Reading EECON2 will read all ‘0’s.

PIC18F87K90 FAMILY
REGISTER 8-1: EECON1: DATA EEPROM CONTROL REGISTER 1
R/W-x R/W-x U-0 R/W-0 R/W-x R/W-0 R/S-0 R/S-0
EEPGD CFGS — FREE WRERR(1) WREN WR RD
bit 7 bit 0
Legend: S = Settable bit

bit 7 EEPGD: Flash Program or Data EEPROM Memory Select bit

1 = Access Flash program memory
0 = Access data EEPROM memory
bit 6 CFGS: Flash Program/Data EEPROM or Configuration Select bit
1 = Access Configuration registers
0 = Access Flash program or data EEPROM memory
bit 4 FREE: Flash Row Erase Enable bit
1 = Erase the program memory row addressed by the TBLPTR on the next WR command
(cleared by completion of an erase operation)
0 = Perform write-only
bit 3 WRERR: Flash Program/Data EEPROM Error Flag bit(1)
1 = A write operation is prematurely terminated (any Reset during self-timed programming in normal
operation or an improper write attempt)
0 = The write operation completed
bit 2 WREN: Flash Program/Data EEPROM Write Enable bit
1 = Allows write cycles to Flash program/data EEPROM
0 = Inhibits write cycles to Flash program/data EEPROM
bit 1 WR: Write-Control bit
1 = Initiates a data EEPROM erase/write cycle, or a program memory erase cycle or write cycle
(The operation is self-timed and the bit is cleared by hardware once the write is complete. The
WR bit can only be set (not cleared) in software.)
0 = Write cycle to the EEPROM is complete
bit 0 RD: Read Control bit
1 = Initiates an EEPROM read (Read takes one cycle. RD is cleared in hardware. The RD bit can only
be set (not cleared) in software. The RD bit cannot be set when EEPGD = 1 or CFGS = 1.)
0 = Does not initiate an EEPROM read
Note 1: When a WRERR occurs, the EEPGD and CFGS bits are not cleared. This allows tracing of the error
condition.

PIC18F87K90 FAMILY
8.3 Reading the Data EEPROM Memory execution (i.e., runaway programs). The WREN bit
should be kept clear at all times, except when updating
To read a data memory location, the user must write the the EEPROM. The WREN bit is not cleared by hardware.
address to the EEADRH:EEADR register pair, clear the
EEPGD control bit (EECON1<7>) and then set control After a write sequence has been initiated, EECON1,
bit, RD (EECON1<0>). After one cycle, the data is EEADRH:EEADR and EEDATA cannot be modified.
available in the EEDATA register; therefore, it can be The WR bit will be inhibited from being set unless the
read after one NOP instruction. EEDATA will hold this WREN bit is set. The WREN bit must be set on a
value until another read operation, or until it is written to previous instruction. Both WR and WREN cannot be
by the user (during a write operation). The basic set with the same instruction.
process is shown in Example 8-1. At the completion of the write cycle, the WR bit is
cleared in hardware and the EEPROM Interrupt Flag bit
8.4 Writing to the Data EEPROM (EEIF) is set. The user may either enable this interrupt,
Memory or poll this bit. EEIF must be cleared by software.
To write an EEPROM data location, the address must first 8.5 Write Verify
be written to the EEADRH:EEADR register pair and the
data written to the EEDATA register. The sequence in Depending on the application, good programming
Example 8-2 must be followed to initiate the write cycle. practice may dictate that the value written to the
memory should be verified against the original value.
The write will not begin if this sequence is not exactly
This should be used in applications where excessive
followed (write 0x55 to EECON2, write 0xAA to
writes can stress bits near the specification limit.
EECON2, then set WR bit) for each byte. It is strongly
recommended that interrupts be disabled during this Note: Self-write execution to Flash and
code segment. EEPROM memory cannot be done while
Additionally, the WREN bit in EECON1 must be set to running in LP Oscillator mode (Low-Power
enable writes. This mechanism prevents accidental mode). Therefore, executing a self-write
writes to data EEPROM due to unexpected code will put the device into High-Power mode.
EXAMPLE 8-1: DATA EEPROM READ

MOVLW DATA_EE_ADDRH ;
MOVWF EEADRH ; Upper bits of Data Memory Address to read
MOVLW DATA_EE_ADDR ;
MOVWF EEADR ; Lower bits of Data Memory Address to read
BCF EECON1, EEPGD ; Point to DATA memory
BCF EECON1, CFGS ; Access EEPROM
BSF EECON1, RD ; EEPROM Read
NOP
MOVF EEDATA, W ; W = EEDATA
EXAMPLE 8-2: DATA EEPROM WRITE

MOVLW DATA_EE_ADDRH ;
MOVWF EEADRH ; Upper bits of Data Memory Address to write
MOVLW DATA_EE_ADDR ;
MOVWF EEADR ; Lower bits of Data Memory Address to write
MOVLW DATA_EE_DATA ;
MOVWF EEDATA ; Data Memory Value to write
BCF EECON1, EEPGD ; Point to DATA memory
BCF EECON1, CFGS ; Access EEPROM
BSF EECON1, WREN ; Enable writes
BCF INTCON, GIE ; Disable Interrupts

MOVLW 0x55 ;
Required MOVWF EECON2 ; Write 55h
Sequence MOVLW 0xAA ;
MOVWF EECON2 ; Write 0AAh
BTFSC EECON1, WR ; Wait for write to complete
GOTO $-2
BSF INTCON, GIE ; Enable Interrupts
; User code execution

BCF EECON1, WREN ; Disable writes on write complete (EEIF set)

PIC18F87K90 FAMILY
8.6 Operation During Code-Protect 8.8 Using the Data EEPROM
Data EEPROM memory has its own code-protect bits in The data EEPROM is a high-endurance, byte address-
the Configuration Words. External read and write able array that has been optimized for the storage of
operations are disabled if code protection is enabled. frequently changing information (e.g., program
The microcontroller itself can both read and write to the variables or other data that are updated often).
internal data EEPROM, regardless of the state of the Frequently changing values will typically be updated
code-protect Configuration bit. Refer to Section 28.0 more often than Specification D124. If this is the case,
“Special Features of the CPU” for additional an array refresh must be performed. For this reason,
information. variables that change infrequently (such as constants,
IDs, calibration, etc.) should be stored in Flash program
memory.
8.7 Protection Against Spurious Write
A simple data EEPROM refresh routine is shown in
There are conditions when the device may not want to Example 8-3.
write to the data EEPROM memory. To protect against
spurious EEPROM writes, various mechanisms have Note: If data EEPROM is only used to store
been implemented. On power-up, the WREN bit is constants and/or data that changes often,
cleared. In addition, writes to the EEPROM are an array refresh is likely not required. See
blocked during the Power-up Timer period (TPWRT, Specification D124 in Table 31-1.
Parameter 33 in Section 31.3 “DC Characteristics:
PIC18F87K90 Family (Industrial/Extended)”).
The write initiate sequence, and the WREN bit
together, help prevent an accidental write during
brown-out, power glitch or software malfunction. The
WREN bit is not cleared by hardware.
EXAMPLE 8-3: DATA EEPROM REFRESH ROUTINE

CLRF EEADR ; Start at address 0
CLRF EEADRH ;
BCF EECON1, CFGS ; Set for memory
BCF EECON1, EEPGD ; Set for Data EEPROM
BCF INTCON, GIE ; Disable interrupts
BSF EECON1, WREN ; Enable writes
LOOP ; Loop to refresh array
BSF EECON1, RD ; Read current address
MOVLW 0x55 ;
MOVWF EECON2 ; Write 55h
MOVLW 0xAA ;
MOVWF EECON2 ; Write 0AAh
BSF EECON1, WR ; Set WR bit to begin write
BTFSC EECON1, WR ; Wait for write to complete
BRA $-2
INCFSZ EEADR, F ; Increment address
BRA LOOP ; Not zero, do it again
INCFSZ EEADRH, F ; Increment the high address
BRA LOOP ; Not zero, do it again
BCF EECON1, WREN ; Disable writes

BSF INTCON, GIE ; Enable interrupts

PIC18F87K90 FAMILY
TABLE 8-1: REGISTERS ASSOCIATED WITH DATA EEPROM MEMORY
Reset
Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Values
on Page:
EEADRH — — — — — — EEPROM Address 79
Register High Byte
EEADR EEPROM Address Register Low Byte 80
EEDATA EEPROM Data Register 80
EECON2 EEPROM Control Register 2 (not a physical register) 79
EECON1 EEPGD CFGS — FREE WRERR WREN WR RD 79
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used during Flash/EEPROM access.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
9.0 8 x 8 HARDWARE MULTIPLIER EXAMPLE 9-1: 8 x 8 UNSIGNED
MULTIPLY ROUTINE
9.1 Introduction MOVF ARG1, W ;
MULWF ARG2 ; ARG1 * ARG2 ->
All PIC18 devices include an 8 x 8 hardware multiplier ; PRODH:PRODL
as part of the ALU. The multiplier performs an unsigned
operation and yields a 16-bit result that is stored in the
product register pair, PRODH:PRODL. The multiplier’s
operation does not affect any flags in the STATUS EXAMPLE 9-2: 8 x 8 SIGNED MULTIPLY
register. ROUTINE
Making multiplication a hardware operation allows it to MOVF ARG1, W
MULWF ARG2 ; ARG1 * ARG2 ->
be completed in a single instruction cycle. This has the
; PRODH:PRODL
advantages of higher computational throughput and
BTFSC ARG2, SB ; Test Sign Bit
reduced code size for multiplication algorithms and SUBWF PRODH, F ; PRODH = PRODH
allows PIC18 devices to be used in many applications ; - ARG1
previously reserved for digital-signal processors. A MOVF ARG2, W
comparison of various hardware and software multiply BTFSC ARG1, SB ; Test Sign Bit
operations, along with the savings in memory and SUBWF PRODH, F ; PRODH = PRODH
execution time, is shown in Table 9-1. ; - ARG2
9.2 Operation
Example 9-1 shows the instruction sequence for an 8 x 8
unsigned multiplication. Only one instruction is required
when one of the arguments is already loaded in the
WREG register.
Example 9-2 shows the sequence to do an 8 x 8 signed
multiplication. To account for the sign bits of the argu-
ments, each argument’s Most Significant bit (MSb) is
tested and the appropriate subtractions are done.
TABLE 9-1: PERFORMANCE COMPARISON FOR VARIOUS MULTIPLY OPERATIONS

Program Time
Cycles
Routine Multiply Method Memory
(Max) @ 64 MHz @ 48 MHz @ 10 MHz @ 4 MHz
(Words)
Without Hardware Multiply 13 69 4.3 s 5.7 s 27.6 s 69 s
8 x 8 Unsigned
Hardware Multiply 1 1 62.5 ns 83.3 ns 400 ns 1 s
8 x 8 Signed
Hardware Multiply 6 6 375 ns 500 ns 2.4 s 6 s
16 x 16 Without Hardware Multiply 21 242 15.1 s 20.1 s 96.8 s 242 s
Unsigned Hardware Multiply 28 28 1.7 s 2.3 s 11.2 s 28 s
16 x 16 Signed
Hardware Multiply 35 40 2.5 s 3.3 s 16.0 s 40 s

PIC18F87K90 FAMILY
Example 9-3 shows the sequence to do a 16 x 16 EQUATION 9-2: 16 x 16 SIGNED
unsigned multiplication. Equation 9-1 shows the MULTIPLICATION
algorithm that is used. The 32-bit result is stored in four ALGORITHM
registers (RES3:RES0).
RES3:RES0 = ARG1H:ARG1L  ARG2H:ARG2L
= (ARG1H  ARG2H  216) +
EQUATION 9-1: 16 x 16 UNSIGNED (ARG1H  ARG2L  28) +
MULTIPLICATION (ARG1L  ARG2H  28) +
ALGORITHM (ARG1L  ARG2L) +
(-1  ARG2H<7>  ARG1H:ARG1L  216) +
RES3:RES0 = ARG1H:ARG1L  ARG2H:ARG2L
(-1  ARG1H<7>  ARG2H:ARG2L  216)
= (ARG1H  ARG2H  216) +
(ARG1H  ARG2L  28) +
(ARG1L  ARG2H  28) + EXAMPLE 9-4: 16 x 16 SIGNED
(ARG1L  ARG2L)
MULTIPLY ROUTINE
MOVF ARG1L, W
EXAMPLE 9-3: 16 x 16 UNSIGNED MULWF ARG2L ; ARG1L * ARG2L ->
MULTIPLY ROUTINE ; PRODH:PRODL
MOVFF PRODH, RES1 ;
MOVF ARG1L, W MOVFF PRODL, RES0 ;
MULWF ARG2L ; ARG1L * ARG2L-> ;
; PRODH:PRODL MOVF ARG1H, W
MOVFF PRODH, RES1 ; MULWF ARG2H ; ARG1H * ARG2H ->
MOVFF PRODL, RES0 ; ; PRODH:PRODL
; MOVFF PRODH, RES3 ;
MOVF ARG1H, W MOVFF PRODL, RES2 ;
MULWF ARG2H ; ARG1H * ARG2H-> ;
; PRODH:PRODL MOVF ARG1L, W
MOVFF PRODH, RES3 ; MULWF ARG2H ; ARG1L * ARG2H ->
MOVFF PRODL, RES2 ; ; PRODH:PRODL
; MOVF PRODL, W ;
MOVF ARG1L, W ADDWF RES1, F ; Add cross
MULWF ARG2H ; ARG1L * ARG2H-> MOVF PRODH, W ; products
; PRODH:PRODL ADDWFC RES2, F ;
CLRF WREG ;
MOVF PRODL, W ;
ADDWFC RES3, F ;
ADDWF RES1, F ; Add cross
;
MOVF PRODH, W ; products
MOVF ARG1H, W ;
ADDWFC RES2, F ;
MULWF ARG2L ; ARG1H * ARG2L ->
CLRF WREG ; ; PRODH:PRODL
ADDWFC RES3, F ; MOVF PRODL, W ;
; ADDWF RES1, F ; Add cross
MOVF ARG1H, W ; MOVF PRODH, W ; products
MULWF ARG2L ; ARG1H * ARG2L-> ADDWFC RES2, F ;
; PRODH:PRODL CLRF WREG ;
MOVF PRODL, W ; ADDWFC RES3, F ;
ADDWF RES1, F ; Add cross ;
MOVF PRODH, W ; products BTFSS ARG2H, 7 ; ARG2H:ARG2L neg?
ADDWFC RES2, F ; BRA SIGN_ARG1 ; no, check ARG1
CLRF WREG ; MOVF ARG1L, W ;
ADDWFC RES3, F ; SUBWF RES2 ;
MOVF ARG1H, W ;
SUBWFB RES3 ;
Example 9-4 shows the sequence to do a 16 x 16 SIGN_ARG1
signed multiply. Equation 9-2 shows the algorithm BTFSS ARG1H, 7 ; ARG1H:ARG1L neg?
used. The 32-bit result is stored in four registers BRA CONT_CODE ; no, done
(RES3:RES0). To account for the sign bits of the MOVF ARG2L, W ;
arguments, the MSb for each argument pair is tested SUBWF RES2 ;
MOVF ARG2H, W ;
and the appropriate subtractions are done.
SUBWFB RES3
;
CONT_CODE
:

PIC18F87K90 FAMILY
10.0 INTERRUPTS When the IPEN bit is cleared (default state), the
interrupt priority feature is disabled and interrupts are
Members of the PIC18F87K90 family of devices have compatible with PIC® mid-range devices. In
multiple interrupt sources and an interrupt priority Compatibility mode, the interrupt priority bits for each
feature that allows most interrupt sources to be source have no effect. INTCON<6> is the PEIE bit that
assigned a high-priority level or a low-priority level. The enables/disables all peripheral interrupt sources.
high-priority interrupt vector is at 0008h and the INTCON<7> is the GIE bit that enables/disables all
low-priority interrupt vector is at 0018h. High-priority interrupt sources. All interrupts branch to address,
interrupt events will interrupt any low-priority interrupts 0008h, in Compatibility mode.
that may be in progress.
When an interrupt is responded to, the Global Interrupt
The registers for controlling interrupt operation are: Enable bit is cleared to disable further interrupts. If the
• RCON IPEN bit is cleared, this is the GIE bit. If interrupt priority
• INTCON levels are used, this will be either the GIEH or GIEL bit.
High-priority interrupt sources can interrupt a
• INTCON2
low-priority interrupt. Low-priority interrupts are not
• INTCON3 processed while high-priority interrupts are in progress.
• PIR1, PIR2, PIR3
The return address is pushed onto the stack and the
• PIE1, PIE2, PIE3 PC is loaded with the interrupt vector address (0008h
• IPR1, IPR2, IPR3 or 0018h). Once in the Interrupt Service Routine (ISR),
It is recommended that the Microchip header files the source(s) of the interrupt can be determined by poll-
supplied with MPLAB® IDE be used for the symbolic bit ing the interrupt flag bits. The interrupt flag bits must be
names in these registers. This allows the cleared in software, before re-enabling interrupts, to
assembler/compiler to automatically take care of the avoid recursive interrupts.
placement of these bits within the specified register. The “return from interrupt” instruction, RETFIE, exits
In general, interrupt sources have three bits to control the interrupt routine and sets the GIE bit (GIEH or GIEL
their operation. They are: if priority levels are used) that re-enables interrupts.
• Flag bit – Indicating that an interrupt event For external interrupt events, such as the INTx pins or
occurred the PORTB input change interrupt, the interrupt latency
will be three to four instruction cycles. The exact
• Enable bit – Enabling program execution to
latency is the same for one or two-cycle instructions.
branch to the interrupt vector address when the
Individual interrupt flag bits are set regardless of the
flag bit is set
status of their corresponding enable bit or the GIE bit.
• Priority bit – Specifying high priority or low priority
Note: Do not use the MOVFF instruction to
The interrupt priority feature is enabled by setting the
modify any of the Interrupt Control regis-
IPEN bit (RCON<7>). When interrupt priority is
ters while any interrupt is enabled. Doing
enabled, there are two bits that enable interrupts
so may cause erratic microcontroller
globally. Setting the GIEH bit (INTCON<7>) enables all
behavior.
interrupts that have the priority bit set (high priority).
Setting the GIEL bit (INTCON<6>) and GIEH bit
(INTCON<7>) enables all interrupts that have the
priority bit cleared (low priority). When the interrupt flag,
enable bit and appropriate Global Interrupt Enable bit
are set, the interrupt will vector immediately to address
0008h or 0018h, depending on the priority bit setting.
Individual interrupts can be disabled through their
corresponding enable bits.

PIC18F87K90 FAMILY
FIGURE 10-1: PIC18F87K90 FAMILY INTERRUPT LOGIC
PIR1<6:0>
PIE1<6:0> TMR0IF Wake-up if in
IPR1<6:0> TMR0IE Idle or Sleep modes
TMR0IP
RBIF
PIR2<7,5:0> RBIE
PIE2<7,5:0> RBIP
IPR2<7:7,5:0> INT0IF
INT0IE
PIR3<6:0> INT1IF
PIE3<6:0> INT1IE Interrupt to CPU
IPR3<6:0> INT1IP Vector to Location
INT2IF
INT2IE 0008h
PIR3<7:0>
PIE3<7:0> INT2IP
IPR3<7:0> INT3IF
INT3IE
INT3IP
GIE/GIEH
PIR5<7:0>
PIE5<7:0>
IPR5<7:0> IPEN
PIR6<4,2:0> IPEN
PIE6<4,2:0> PEIE/GIEL
IPR6<4,2:0>
IPEN
High-Priority Interrupt Generation
Low-Priority Interrupt Generation
PIR1<6:0>
PIE1<6:0>
IPR1<6:0>
PIR2<7,5:0>
PIE2<7,5:0>
IPR2<7,5:0> Interrupt to CPU
TMR0IF Vector to Location
TMR0IE IPEN
PIR3<7:0> 0018h
PIE3<7:0> TMR0IP
IPR3<7:0>
RBIF
PIR4<7:0> RBIE
PIE4<7:0> RBIP GIE/GIEH
IPR4<7:0> PEIE/GIEL
INT1IF
INT1IE
PIR5<7:0> INT1IP
PIE5<7:0> INT2IF
IPR5<7:0> INT2IE
INT2IP
PIR6<4,2:0> INT3IF
PIE6<4,2:0> INT3IE
IPR6<4,2:0> INT3IP

PIC18F87K90 FAMILY
10.1 INTCON Registers Note: Interrupt flag bits are set when an interrupt
The INTCON registers are readable and writable condition occurs, regardless of the state of
registers that contain various enable, priority and flag its corresponding enable bit or the Global
bits. Interrupt Enable bit. User software should
ensure the appropriate interrupt flag bits
are clear prior to enabling an interrupt.
This feature allows for software polling.
REGISTER 10-1: INTCON: INTERRUPT CONTROL REGISTER

R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-x
GIE/GIEH PEIE/GIEL TMR0IE INT0IE RBIE TMR0IF INT0IF RBIF(1)
bit 7 bit 0
Legend:
bit 7 GIE/GIEH: Global Interrupt Enable bit

When IPEN = 0:
1 = Enables all unmasked interrupts
0 = Disables all interrupts
When IPEN = 1:
1 = Enables all high-priority interrupts
0 = Disables all interrupts
bit 6 PEIE/GIEL: Peripheral Interrupt Enable bit
When IPEN = 0:
1 = Enables all unmasked peripheral interrupts
0 = Disables all peripheral interrupts
When IPEN = 1:
1 = Enables all low-priority peripheral interrupts
0 = Disables all low-priority peripheral interrupts
bit 5 TMR0IE: TMR0 Overflow Interrupt Enable bit
1 = Enables the TMR0 overflow interrupt
0 = Disables the TMR0 overflow interrupt
bit 4 INT0IE: INT0 External Interrupt Enable bit
1 = Enables the INT0 external interrupt
0 = Disables the INT0 external interrupt
bit 3 RBIE: RB Port Change Interrupt Enable bit
1 = Enables the RB port change interrupt
0 = Disables the RB port change interrupt
bit 2 TMR0IF: TMR0 Overflow Interrupt Flag bit
1 = TMR0 register has overflowed (must be cleared in software)
0 = TMR0 register did not overflow
bit 1 INT0IF: INT0 External Interrupt Flag bit
1 = The INT0 external interrupt occurred (must be cleared in software)
0 = The INT0 external interrupt did not occur
bit 0 RBIF: RB Port Change Interrupt Flag bit(1)
1 = At least one of the RB<7:4> pins changed state (must be cleared in software)
0 = None of the RB<7:4> pins have changed state
Note 1: A mismatch condition will continue to set this bit. Reading PORTB, and then waiting one additional instruction
cycle, will end the mismatch condition and allow the bit to be cleared.

PIC18F87K90 FAMILY
REGISTER 10-2: INTCON2: INTERRUPT CONTROL REGISTER 2
RBPU INTEDG0 INTEDG1 INTEDG2 INTEDG3 TMR0IP INT3IP RBIP
bit 7 bit 0
Legend:
bit 7 RBPU: PORTB Pull-up Enable bit

1 = All PORTB pull-ups are disabled
0 = PORTB pull-ups are enabled by individual TRIS register values
bit 6 INTEDG0: External Interrupt 0 Edge Select bit
1 = Interrupt on rising edge
0 = Interrupt on falling edge
bit 2 TMR0IP: TMR0 Overflow Interrupt Priority bit
1 = High priority
0 = Low priority
bit 1 INT3IP: INT3 External Interrupt Priority bit
1 = High priority
0 = Low priority
bit 0 RBIP: RB Port Change Interrupt Priority bit
1 = High priority
0 = Low priority
Note: Interrupt flag bits are set when an interrupt condition occurs, regardless of the state of its corresponding
enable bit or the Global Interrupt Enable bit. User software should ensure the appropriate interrupt flag bits
are clear prior to enabling an interrupt. This feature allows for software polling.

PIC18F87K90 FAMILY
REGISTER 10-3: INTCON3: INTERRUPT CONTROL REGISTER 3
INT2IP INT1IP INT3IE INT2IE INT1IE INT3IF INT2IF INT1IF
bit 7 bit 0
Legend:

1 = High priority
0 = Low priority
1 = High priority
0 = Low priority
Note: Interrupt flag bits are set when an interrupt condition occurs, regardless of the state of its corresponding
enable bit or the Global Interrupt Enable bit. User software should ensure the appropriate interrupt flag bits
are clear prior to enabling an interrupt. This feature allows for software polling.

PIC18F87K90 FAMILY
10.2 PIR Registers Note 1: Interrupt flag bits are set when an
The PIR registers contain the individual flag bits for the interrupt condition occurs, regardless of
peripheral interrupts. Due to the number of peripheral the state of its corresponding enable bit or
interrupt sources, there are six Peripheral Interrupt the Global Interrupt Enable bit, GIE
Request (Flag) registers (PIR1 through PIR6). (INTCON<7>).
2: User software should ensure the
appropriate interrupt flag bits are cleared
prior to enabling an interrupt and after
servicing that interrupt.
REGISTER 10-4: PIR1: PERIPHERAL INTERRUPT REQUEST (FLAG) REGISTER 1

U-0 R/W-0 R-0 R-0 R/W-0 R/W-0 R/W-0 R/W-0
— ADIF RC1IF TX1IF SSP1IF TMR1GIF TMR2IF TMR1IF
bit 7 bit 0
Legend:

bit 6 ADIF: A/D Converter Interrupt Flag bit
1 = An A/D conversion completed (must be cleared in software)
0 = The A/D conversion is not complete
bit 5 RC1IF: EUSART Receive Interrupt Flag bit
1 = The EUSART receive buffer, RCREG1, is full (cleared when RCREG1 is read)
0 = The EUSART receive buffer is empty
bit 4 TX1IF: EUSART Transmit Interrupt Flag bit
1 = The EUSART transmit buffer, TXREG1, is empty (cleared when TXREG1 is written)
0 = The EUSART transmit buffer is full
bit 3 SSP1IF: Master Synchronous Serial Port Interrupt Flag bit
1 = The transmission/reception is complete (must be cleared in software)
0 = Waiting to transmit/receive
bit 2 TMR1GIF: Timer1 Gate Interrupt Flag bit
1 = Timer gate interrupt has occurred (must be cleared in software)
0 = No timer gate interrupt has occurred
bit 1 TMR2IF: TMR2 to PR2 Match Interrupt Flag bit
1 = TMR2 to PR2 match has occurred (must be cleared in software)
0 = No TMR2 to PR2 match has occurred
1 = TMR1 register overflowed (must be cleared in software)

PIC18F87K90 FAMILY
OSCFIF — SSP2IF BCL2IF BCL1IF HLVDIF TMR3IF TMR3GIF
bit 7 bit 0
Legend:
bit 7 OSCFIF: Oscillator Fail Interrupt Flag bit

1 = Device oscillator failed, clock input has changed to INTOSC (must be cleared in software)
0 = Device clock operating
bit 5 SSP2IF: Master Synchronous Serial Port Interrupt Flag bit
1 = The transmission/reception has been completed (must be cleared in software)
0 = Waiting to transmit/receive
bit 4 BCL2IF: Bus Collision Interrupt Flag bit
1 = A bus collision occurred (must be cleared in software)
0 = No bus collision occurred
bit 3 BCL1IF: Bus Collision Interrupt Flag bit
1 = A bus collision occurred (must be cleared in software)
0 = No bus collision occurred
bit 2 HLVDIF: High/Low-Voltage Detect Interrupt Flag bit
1 = A high/low-voltage condition occurred (must be cleared in software)
0 = The device voltage is above the regulator’s low-voltage trip point
bit 0 TMR3GIF: TMR3 Gate Interrupt Flag bit
1 = Timer gate interrupt occurred (must be cleared in software)
0 = No timer gate interrupt occurred

PIC18F87K90 FAMILY
R/W-0 R/W-0 R-0 R-0 R/W-0 R/W-0 R/W-0 R/W-0
TMR5GIF LCDIF RC2IF TX2IF CTMUIF CCP2IF CCP1IF RTCCIF
bit 7 bit 0
Legend:
bit 7 TMR5GIF: Timer5 Gate Interrupt Flag bit

1 = Timer gate interrupt occurred (must be cleared in software)
0 = No timer gate interrupt occurred
bit 6 LCDIF: LCD Interrupt Flag bit (valid when Type-B waveform with Non-Static mode is selected)
1 = LCD data of all COMs is output (must be cleared in software)
0 = LCD data of all COMs is not yet output
bit 5 RC2IF: EUSART Receive Interrupt Flag bit
1 = The EUSART receive buffer, RCREG2, is full (cleared when RCREG2 is read)
0 = The EUSART receive buffer is empty
bit 4 TX2IF: EUSART Transmit Interrupt Flag bit
1 = The EUSART transmit buffer, TXREG2, is empty (cleared when TXREG2 is written)
0 = The EUSART transmit buffer is full
bit 3 CTMUIF: CTMU Interrupt Flag bit
1 = CTMU interrupt occurred (must be cleared in software)
0 = No CTMU interrupt occurred
bit 2 CCP2IF: ECCP2 Interrupt Flag bit
Capture mode:
1 = A TMR register capture occurred (must be cleared in software)
0 = No TMR register capture occurred
Compare mode:
1 = A TMR register compare match occurred (must be cleared in software)
0 = No TMR register compare match occurred
PWM mode:
Unused in this mode.
bit 1 CCP1IF: ECCP1 Interrupt Flag bit
Capture mode:
Compare mode:
PWM mode:
bit 0 RTCCIF: RTCC Interrupt Flag bit
1 = RTCC interrupt occurred (must be cleared in software)
0 = No RTCC interrupt occurred

PIC18F87K90 FAMILY
REGISTER 10-7: PIR4: PERIPHERAL INTERRUPT FLAG REGISTER 4
(1)
CCP10IF CCP9IF(1) CCP8IF CCP7IF CCP6IF CCP5IF CCP4IF CCP3IF
bit 7 bit 0
Legend:
bit 7-1 CCP10IF:CCP4IF: CCP<10:4> Interrupt Flag bits(1)

Capture Mode
Compare Mode
PWM Mode
Not used in PWM mode.
bit 0 CCP3IF: ECCP3 Interrupt Flag bits
Capture Mode
Compare Mode
PWM Mode
Not used in PWM mode.

PIC18F87K90 FAMILY
(1) (1) (1) (1)
TMR7GIF TMR12IF TMR10IF TMR8IF TMR7IF TMR6IF TMR5IF TMR4IF
bit 7 bit 0
Legend:
bit 7 TMR7GIF: TMR7 Gate Interrupt Flag bits(1)

1 = TMR gate interrupt occurred (must be cleared in software)
0 = No TMR gate interrupt occurred
bit 6 TMR12IF: TMR12 to PR12 Match Interrupt Flag bit(1)
1 = TMR12 to PR12 match occurred (must be cleared in software)
0 = No TMR12 to PR12 match occurred
bit 5 TMR10IF: TMR10 to PR10 Match Interrupt Flag bit(1)
bit 3 TMR7IF: TMR7 Overflow Interrupt Flag bit(1)

PIC18F87K90 FAMILY
U-0 U-0 U-0 R/W-0 U-0 R/W-0 R/W-0 R/W-0
— — — EEIF — CMP3IF CMP2IF CMP1IF
bit 7 bit 0
Legend:

bit 4 EEIF: Data EEDATA/Flash Write Operation Interrupt Flag bit
1 = The write operation is complete (must be cleared in software)
0 = The write operation is not complete, or has not been started
bit 2 CMP3IF: CMP3 Interrupt Flag bit
1 = CMP3 interrupt occurred (must be cleared in software)
0 = No CMP3 interrupt occurred
bit 1 CMP2IF: CMP2 Interrupt Flag bit
bit 0 CMP1IF: CM1 Interrupt Flag bit

PIC18F87K90 FAMILY
10.3 PIE Registers
The PIE registers contain the individual enable bits for
the peripheral interrupts. Due to the number of
peripheral interrupt sources, there are six Peripheral
Interrupt Enable registers (PIE1 through PIE6). When
IPEN (RCON<7>) = 0, the PEIE bit must be set to
enable any of these peripheral interrupts.
REGISTER 10-10: PIE1: PERIPHERAL INTERRUPT ENABLE REGISTER 1

U-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0
— ADIE RC1IE TX1IE SSP1IE TMR1GIE TMR2IE TMR1IE
bit 7 bit 0
Legend:

bit 6 ADIE: A/D Converter Interrupt Enable bit
1 = Enables the A/D interrupt
0 = Disables the A/D interrupt
bit 5 RC1IE: EUSART Receive Interrupt Enable bit
1 = Enables the EUSART receive interrupt
0 = Disables the EUSART receive interrupt
bit 4 TX1IE: EUSART Transmit Interrupt Enable bit
1 = Enables the EUSART transmit interrupt
0 = Disables the EUSART transmit interrupt
bit 3 SSP1IE: Master Synchronous Serial Port Interrupt Enable bit
1 = Enables the MSSP interrupt
0 = Disables the MSSP interrupt
bit 2 TMR1GIE: TMR1 Gate Interrupt Enable bit
1 = Enables the gate
0 = Disables the gate
bit 1 TMR2IE: TMR2 to PR2 Match Interrupt Enable bit
1 = Enables the TMR2 to PR2 match interrupt
0 = Disables the TMR2 to PR2 match interrupt

PIC18F87K90 FAMILY
OSCFIE — SSP2IE BCL2IE BCL1IE HLVDIE TMR3IE TMR3GIE
bit 7 bit 0
Legend:
bit 7 OSCFIE: Oscillator Fail Interrupt Enable bit

1 = Enabled
0 = Disabled
bit 5 SSP2IE: Master Synchronous Serial Port 2 Interrupt Enable bit
1 = Enables the MSSP interrupt
0 = Disables the MSSP interrupt
bit 4 BCL2IE: Bus Collision Interrupt Enable bit
1 = Enables the bus collision interrupt
0 = Disables the bus collision interrupt
bit 3 BCL1IE: Bus Collision Interrupt Enable bit
1 = Enabled
0 = Disabled
bit 2 HLVDIE: High/Low-Voltage Detect Interrupt Enable bit
1 = Enabled
0 = Disabled
1 = Enabled
0 = Disabled
bit 0 TMR3GIE: Timer3 Gate Interrupt Enable bit
1 = Enabled
0 = Disabled

PIC18F87K90 FAMILY
R/W-0 R/W-0 R-0 R-0 R/W-0 R/W-0 R/W-0 R/W-0
(1)
TMR5GIE LCDIE RC2IE TX2IE CTMUIE CCP2IE CCP1IE RTCCIE
bit 7 bit 0
Legend:
bit 7 TMR5GIE: Timer5 Gate Interrupt Enable bit

1 = Enabled
0 = Disabled
bit 6 LCDIE: LCD Interrupt Enable bit(1)
1 = Enabled
0 = Disabled
bit 5 RC2IE: AUSART Receive Interrupt Enable bit
1 = Enabled
0 = Disabled
bit 4 TX2IE: AUSART Transmit Interrupt Enable bit
1 = Enabled
0 = Disabled
bit 3 CTMUIE: CTMU Interrupt Enable bit
1 = Enabled
0 = Disabled
bit 2 CCP2IE: ECCP2 Interrupt Enable bit
1 = Enabled
0 = Disabled
bit 1 CCP1IE: ECCP1 Interrupt Enable bit
1 = Enabled
0 = Disabled
bit 0 RTCCIE: RTCC Interrupt Enable bit
1 = Enabled
0 = Disabled
Note 1: This bit is valid when the Type-B waveform with Non-Static mode is selected.

(1) CCP9IE(1)
CCP10IE CCP8IE CCP7IE CCP6IE CCP5IE CCP4IE CCP3IE
bit 7 bit 0
Legend:
bit 7-0 CCP10IE:CCP3IE: CCP<10:3> Interrupt Enable bits(1)

1 = Enabled
0 = Disabled
Note 1: CCP10IE and CCP9IE are unimplemented in devices with a program memory of 32 Kbytes (PIC18FX5K90).

PIC18F87K90 FAMILY
TMR7GIE(1) TMR12IE(1) TMR10IE(1) TMR8IE TMR7IE(1) TMR6IE TMR5IE TMR4IE
bit 7 bit 0
Legend:
bit 7 TMR7GIE: TMR7 Gate Interrupt Enable bit(1)

1 = Enabled
0 = Disabled
bit 6 TMR12IE: TMR12 to PR12 Match Interrupt Enable bit(1)
bit 5 TMR10IE: TMR10 to PR10 Match Interrupt Enable bit(1)
bit 3 TMR7IE: TMR7 Overflow Interrupt Enable bit(1)

PIC18F87K90 FAMILY
U-0 U-0 U-0 R/W-0 U-0 R/W-0 R/W-0 R/W-0
— — — EEIE — CMP3IE CMP2IE CMP1IE
bit 7 bit 0
Legend:

bit 4 EEIE: Data EEDATA/Flash Write Operation Enable bit
1 = Interrupt is enabled
0 = interrupt is disabled
bit 2 CMP3IE: CMP3 Enable bit
bit 1 CMP2E: CMP2 Enable bit
bit 0 CMP1IE: CMP1 Enable bit

PIC18F87K90 FAMILY
10.4 IPR Registers
The IPR registers contain the individual priority bits for
the peripheral interrupts. Due to the number of
peripheral interrupt sources, there are six Peripheral
Interrupt Priority registers (IPR1 through IPR6). Using
the priority bits requires that the Interrupt Priority
Enable (IPEN) bit (RCON<7>) be set.
REGISTER 10-16: IPR1: PERIPHERAL INTERRUPT PRIORITY REGISTER 1

— ADIP RC1IP TX1IP SSP1IP TMR1GIP TMR2IP TMR1IP
bit 7 bit 0
Legend:

bit 6 ADIP: A/D Converter Interrupt Priority bit
1 = High priority
0 = Low priority
bit 5 RC1IP: EUSART Receive Interrupt Priority bit
1 = High priority
0 = Low priority
bit 4 TX1IP: EUSART Transmit Interrupt Priority bit
1 = High priority
0 = Low priority
bit 3 SSP1IP: Master Synchronous Serial Port Interrupt Priority bit
1 = High priority
0 = Low priority
bit 2 TMR1GIP: Timer1 Gate Interrupt Priority bit
1 = High priority
0 = Low priority
bit 1 TMR2IP: TMR2 to PR2 Match Interrupt Priority bit
1 = High priority
0 = Low priority
1 = High priority
0 = Low priority

PIC18F87K90 FAMILY
OSCFIP — SSP2IP BCL2IP BCL1IP HLVDIP TMR3IP TMR3GIP
bit 7 bit 0
Legend:
bit 7 OSCFIP: Oscillator Fail Interrupt Priority bit

1 = High priority
0 = Low priority
bit 5 SSP2IP: Master Synchronous Serial Port 2 Interrupt Priority bit
1 = High priority
0 = Low priority
bit 4 BCL2IP: Bus Collision Interrupt priority bit (MSSP)
1 = High priority
0 = Low priority
bit 3 BCL1IP: Bus Collision Interrupt Priority bit
1 = High priority
0 = Low priority
bit 2 HLVDIP: High/Low-Voltage Detect Interrupt Priority bit
1 = High priority
0 = Low priority
1 = High priority
0 = Low priority
bit 0 TMR3GIP: TMR3 Gate Interrupt Priority bit
1 = High priority
0 = Low priority

PIC18F87K90 FAMILY
R/W-1 R/W-1 R-1 R-1 R/W-1 R/W-1 R/W-1 R/W-1
TMR5GIP LCDIP RC2IP TX2IP CTMUIP CCP2IP CCP1IP RTCCIP
bit 7 bit 0
Legend:
bit 7 TMR5GIP: Timer5 Gate interrupt Priority bit

1 = High priority
0 = Low priority
bit 6 LCDIP: LCD Interrupt Priority bit (valid when the Type-B waveform with Non-Static mode is selected)
1 = High priority
0 = Low priority
bit 5 RC2IP: AUSART Receive Priority Flag bit
1 = High priority
0 = Low priority
bit 4 TX2IP: AUSART Transmit Interrupt Priority bit
1 = High priority
0 = Low priority
bit 3 CTMUIP: CTMU Interrupt Priority bit
1 = High priority
0 = Low priority
bit CCP2IP: ECCP2 Interrupt Priority bit
1 = High priority
0 = Low priority
bit CCP1IP: ECCP1 Interrupt Priority bit
1 = High priority
0 = Low priority
bit 0 RTCCIP: RTCC Interrupt Priority bit
1 = High priority
0 = Low priority

CCP10IP(1) CCP9IP (1)
CCP8IP CCP7IP CCP6IP CCP5IP CCP4IP CCP3IP
bit 7 bit 0
Legend:
bit 7-0 CCP10IP:CCP3IP: CCP<10:3> Interrupt Priority bits(1)

1 = High priority
0 = Low priority
Note 1: CCP10IP and CCP9IP are unimplemented in devices with a program memory of 32 Kbytes (PIC18FX5K90).

PIC18F87K90 FAMILY
TMR7GIP(1) TMR12IP(1) TMR10IP(1) TMR8IP TMR7IP(1) TMR6IP TMR5IP TMR4IP
bit 7 bit 0
Legend:
bit 7 TMR7GIP: TMR7 Gate Interrupt Priority bit(1)

1 = High priority
0 = Low priority
bit 6 TMR12IP: TMR12 to PR12 Match Interrupt Priority bit(1)
1 = High priority
0 = Low priority
bit 5 TMR10IP: TMR10 to PR10 Match Interrupt Priority bit(1)
1 = High priority
0 = Low priority
1 = High priority
0 = Low priority
bit 3 TMR7IP: TMR7 Overflow Interrupt Priority bit(1)
1 = High priority
0 = Low priority
1 = High priority
0 = Low priority
1 = High priority
0 = Low priority
1 = High priority
0 = Low priority

PIC18F87K90 FAMILY
U-0 U-0 U-0 R/W-1 U-0 R/W-1 R/W-1 R/W-1
— — — EEIP — CMP3IP CMP2IP CMP1IP
bit 7 bit 0
Legend:

bit 4 EEIP: EE Interrupt Priority bit
1 = High priority
0 = Low priority
bit 3 SBOREN: Read as ‘0’
bit 2 CMP3IP: CMP3 Interrupt Priority bit
1 = High priority
0 = Low priority
1 = High priority
0 = Low priority
1 = High priority
0 = Low priority

PIC18F87K90 FAMILY
10.5 RCON Register
The RCON register contains the bits used to determine
the cause of the last Reset, or wake-up from Idle or
Sleep modes. RCON also contains the bit that enables
interrupt priorities (IPEN).
REGISTER 10-22: RCON: RESET CONTROL REGISTER

R/W-0 R/W-1 R/W-1 R/W-1 R-1 R-1 R/W-0 R/W-0
IPEN SBOREN CM RI TO PD POR BOR
bit 7 bit 0
Legend:
bit 7 IPEN: Interrupt Priority Enable bit

1 = Enable priority levels on interrupts
0 = Disable priority levels on interrupts (PIC16CXXX Compatibility mode)
bit 6 SBOREN: BOR Software Enable bit
If BOREN<1:0> = 01:
1 = BOR is enabled
0 = BOR is disabled
If BOREN<1:0> = 00, 10 or 11:
Bit is disabled and read as ‘0’.
bit 5 CM: Configuration Mismatch Flag bit
1 = A Configuration Mismatch Reset has not occurred
0 = A Configuration Mismatch Reset has occurred (must be subsequently set in software)
bit 4 RI: RESET Instruction Flag bit
For details of bit operation, see Register 5-1.
bit 3 TO: Watchdog Timer Time-out Flag bit
bit 2 PD: Power-Down Detection Flag bit
bit 1 POR: Power-on Reset Status bit
bit 0 BOR: Brown-out Reset Status bit

PIC18F87K90 FAMILY
10.6 INTx Pin Interrupts The interrupt can be enabled/disabled by setting/clearing
enable bit, TMR0IE (INTCON<5>). Interrupt priority for
External interrupts on the RB0/INT0, RB1/INT1, Timer0 is determined by the value contained in the
RB2/INT2 and RB3/INT3 pins are edge-triggered. If the interrupt priority bit, TMR0IP (INTCON2<2>). For further
corresponding INTEDGx bit in the INTCON2 register is details on the Timer0 module, see Section 12.0 “Timer0
set (= 1), the interrupt is triggered by a rising edge. If Module”.
that bit is clear, the trigger is on the falling edge.
When a valid edge appears on the RBx/INTx pin, the 10.8 PORTB Interrupt-on-Change
corresponding flag bit, INTxIF, is set. This interrupt can
be disabled by clearing the corresponding enable bit, An input change on PORTB<7:4> sets flag bit, RBIF
INTxIE. Before re-enabling the interrupt, the flag bit (INTCON<0>). The interrupt can be enabled/disabled
(INTxIF) must be cleared in software in the Interrupt by setting/clearing enable bit, RBIE (INTCON<3>).
Service Routine. Interrupt priority for PORTB interrupt-on-change is
All external interrupts (INT0, INT1, INT2 and INT3) can determined by the value contained in the interrupt
wake up the processor from the power-managed priority bit, RBIP (INTCON2<0>).
modes if bit, INTxIE, was set prior to going into the
power-managed modes. If the Global Interrupt Enable 10.9 Context Saving During Interrupts
bit (GIE) is set, the processor will branch to the interrupt
During interrupts, the return PC address is saved on
vector following wake-up.
the stack. Additionally, the WREG, STATUS and BSR
The interrupt priority for INT1, INT2 and INT3 is registers are saved on the Fast Return Stack.
determined by the value contained in the Interrupt
If a fast return from interrupt is not used (see
Priority bits, INT1IP (INTCON3<6>), INT2IP
Section 6.3 “Data Memory Organization”), the user
(INTCON3<7>) and INT3IP (INTCON2<1>).
may need to save the WREG, STATUS and BSR regis-
There is no priority bit associated with INT0. It is always ters on entry to the Interrupt Service Routine (ISR).
a high-priority interrupt source. Depending on the user’s application, other registers
may also need to be saved.
10.7 TMR0 Interrupt Example 10-1 saves and restores the WREG, STATUS
In 8-bit mode (the default), an overflow in the TMR0 and BSR registers during an Interrupt Service Routine.
register (FFh  00h) will set flag bit, TMR0IF. In 16-bit
mode, an overflow in the TMR0H:TMR0L register pair
(FFFFh  0000h) will set TMR0IF.
EXAMPLE 10-1: SAVING STATUS, WREG AND BSR REGISTERS IN RAM

MOVWF W_TEMP ; W_TEMP is in virtual bank
MOVFF STATUS, STATUS_TEMP ; STATUS_TEMP located anywhere
MOVFF BSR, BSR_TEMP ; BSR_TMEP located anywhere
;
; USER ISR CODE
;
MOVFF BSR_TEMP, BSR ; Restore BSR
MOVF W_TEMP, W ; Restore WREG
MOVFF STATUS_TEMP, STATUS ; Restore STATUS

PIC18F87K90 FAMILY
TABLE 10-1: SUMMARY OF REGISTERS ASSOCIATED WITH INTERRUPTS
Reset
on Page:
INTCON2 RBPU INTEDG0 INTEDG1 INTEDG2 INTEDG3 TMR0IP INT3IP RBIP 75
INTCON3 INT2IP INT1IP INT3IE INT2IE INT1IE INT3IF INT2IF INT1IF 75
PIR1 — ADIF RC1IF TX1IF SSP1IF TMR1GIF TMR2IF TMR1IF 77
PIR2 OSCFIF — SSP2IF BCL2IF BCL1IF HLVDIF TMR3IF TMR3GIF 77
PIR3 TMR5GIF LCDIF RC2IF TX2IF CTMUIF CCP2IF CCP1IF RTCCIF 77
PIR4 CCP10IF(1) CCP9IF(1) CCP8IF CCP7IF CCP6IF CCP5IF CCP4IF CCP3IF 77
(1) (1) (1) (1)
PIR5 TMR7GIF TMR12IF TMR10IF TMR8IF TMR7IF TMR6IF TMR5IF TMR4IF 77
PIE1 — ADIE RC1IE TX1IE SSP1IE TMR1GIE TMR2IE TMR1IE 77
PIE2 OSCFIE — SSP2IE BCL2IE BCL1IE HLVDIE TMR3IE TMR3GIE 77
PIE3 TMR5GIE LCDIE RC2IE TX2IE CTMUIE CCP2IE CCP1IE RTCCIE 77
(1) CCP9IE(1)
PIE4 CCP10IE CCP8IE CCP7IE CCP6IE CCP5IE CCP4IE CCP3IE 77
PIE5 TMR7GIE(1) TMR12IE(1) TMR10IE(1) TMR8IE TMR7IE(1) TMR6IE TMR5IE TMR4IE 77
IPR1 — ADIP RC1IP TX1IP SSP1IP TMR1GIP TMR2IP TMR1IP 77
IPR2 OSCFIP — SSP2IP BCL2IP BCL1IP HLVDIP TMR3IP TMR3GIP 77
IPR3 TMR5GIP LCDIP RC2IP TX2IP CTMUIP CCP2IP CCP1IP RTCCIP 77
(1) CCP9IP(1)
IPR4 CCP10IP CCP8IP CCP7IP CCP6IP CCP5IP CCP4IP CCP3IP 77
IPR5 TMR7GIP(1) TMR12IP(1) TMR10IP(1) TMR8IP TMR7IP(1) TMR6IP TMR5IP TMR4IP 76
RCON IPEN SBOREN CM RI TO PD POR BOR 76
Legend: Shaded cells are not used by the interrupts.

PIC18F87K90 FAMILY
11.0 I/O PORTS 11.1 I/O Port Pin Capabilities
Depending on the device selected and features When developing an application, the capabilities of the
enabled, there are up to nine ports available. Some port pins must be considered. Outputs on some pins
pins of the I/O ports are multiplexed with an alternate have higher output drive strength than others. Similarly,
function from the peripheral features on the device. In some pins can tolerate higher than VDD input levels.
general, when a peripheral is enabled, that pin may not All of the digital ports are 5.5V input tolerant. The ana-
be used as a general purpose I/O pin. log ports have the same tolerance, having clamping
Each port has three memory mapped registers for its diodes implemented internally.
operation:
11.1.1 PIN OUTPUT DRIVE
• TRIS register (Data Direction register)
• PORT register (reads the levels on the pins of the When used as digital I/O, the output pin drive strengths
device) vary, according to the pins’ grouping, to meet the needs
for a variety of applications. In general, there are two
• LAT register (Output Latch register)
classes of output pins, in terms of drive capability:
Reading the PORT register reads the current status of • Outputs designed to drive higher current loads,
the pins, whereas writing to the PORT register writes to such as LEDs:
the Output Latch (LAT) register.
- PORTA - PORTB
Setting a TRIS bit (= 1) makes the corresponding port
- PORTC
pin an input (putting the corresponding output driver in
a High-Impedance mode). Clearing a TRIS bit (= 0) • Outputs with lower drive levels, but capable of
makes the corresponding port pin an output (i.e., puts driving normal digital circuit loads with a high input
the contents of the corresponding LAT bit on the impedance. Also, able to drive LEDs, but only
selected pin). those with smaller current requirements:
The Output Latch (LAT register) is useful for - PORTD - PORTE
read-modify-write operations on the value that the I/O - PORTF - PORTG
pins are driving. Read-modify-write operations on the - PORTH(†) - PORTJ(†)
LAT register, read and write the latched output value for † These ports are not available in 64-pin devices.
the PORT register.
For more details, see “Absolute Maximum Ratings” in
A simplified model of a generic I/O port, without the Section 31.0 “Electrical Characteristics”.
interfaces to other peripherals, is shown in Figure 11-1.
Regardless of its port, all output pins in LCD Segment
or common-mode have sufficient output to directly
FIGURE 11-1: GENERIC I/O PORT drive a display.
OPERATION
11.1.2 PULL-UP CONFIGURATION
RD LAT Four of the I/O ports (PORTB, PORTD, PORTE and
PORTJ) implement configurable weak pull-ups on all
Data
Bus pins. These are internal pull-ups that allow floating
D Q
digital input signals to be pulled to a consistent level
WR LAT I/O Pin
or PORT without the use of external resistors.
CKx
The pull-ups are enabled with a single bit for each of the
Data Latch
ports: RBPU (INTCON2<7>) for PORTB, and RDPU,
D Q REPU and RJPU (PADCFG1<7:5>) for the other ports.
By setting RDPU, REPU and RJPU, each of the
WR TRIS pull-ups on these ports can be enabled. The pull-ups
CKx
TRIS Latch are disabled on a POR event.
Input
Buffer
RD TRIS
Q D
ENEN
RD PORT
Note: I/O pins have diode protection to VDD and VSS.

PIC18F87K90 FAMILY
11.1.3 OPEN-DRAIN OUTPUTS FIGURE 11-2: USING THE OPEN-DRAIN
The output pins for several peripherals are also OUTPUT (USART SHOWN
equipped with a configurable, open-drain output option. AS EXAMPLE)
This allows the peripherals to communicate with
3.3V +5V
external digital logic, operating at a higher voltage
level, without the use of level translators. PIC18F67K90
The open-drain option is implemented on port pins

specifically associated with the data and clock outputs
3.3V
of the USARTs, the MSSP module (in SPI mode) and VDD TXX 5V
the CCP modules. This option is selectively enabled by (at logic ‘1’)
setting the open-drain control bits in the registers:
ODCON1, ODCON2 and ODCON3.
When the open-drain option is required, the output pin
must also be tied through an external pull-up resistor
provided by the user to a higher voltage level, up to 5V
(Figure 11-2). When a digital logic high signal is output,
it is pulled up to the higher voltage level.
REGISTER 11-1: ODCON1: PERIPHERAL OPEN-DRAIN CONTROL REGISTER 1

R/W-0 R/W-0 R/W-0 U-0 U-0 U-0 U-0 R/W-0
SSP1OD CCP2OD CCP1OD — — — — SSP2OD
bit 7 bit 0
Legend:
bit 7 SSP1OD: SPI1 Open-Drain Output Enable bit

1 = Open-drain capability is enabled
0 = Open-drain capability is disabled
bit 6 CCP2OD: ECCP2 Open-Drain Output Enable bit
bit 0 SSP2OD: SPI2 Open-Drain Output Enable bit

PIC18F87K90 FAMILY
(1)
CCP10OD CCP9OD(1) CCP8OD CCP7OD CCP6OD CCP5OD CCP4OD CCP3OD
bit 7 bit 0
Legend:
bit 7 CCP10OD: CCP10 Open-Drain Output Enable bit(1)

bit 6 CCP9OD: CCP9 Open-Drain Output Enable bit(1)
bit 5 CCP8OD: CCP8 Open-Drain Output Enable bit

PIC18F87K90 FAMILY
R/W-0 R/W-0 U-0 U-0 U-0 U-0 U-0 R/W-0
U2OD U1OD — — — — — CTMUDS
bit 7 bit 0
Legend:
bit 7 U2OD: EUSART2 Open-Drain Output Enable bit

bit 6 U1OD: EUSART1 Open-Drain Output Enable bit
bit 0 CTMUDS: CTMU Pulse Delay Enable bit
1 = Pulse delay input for CTMU is enabled on pin, RF1
0 = Pulse delay input for CTMU is disabled on pin, RF1
11.1.4 ANALOG AND DIGITAL PORTS Setting these registers makes the corresponding pins
analog and clearing the registers makes the ports digi-
Many of the ports multiplex analog and digital function-
tal. For details on these registers, see Section 23.0
ality, providing a lot of flexibility for hardware designers.
“12-Bit Analog-to-Digital Converter (A/D) Module”.
PIC18F87K90 family devices can make any analog pin,
analog or digital, depending on an application’s needs.
The ports’ analog/digital functionality is controlled by
the registers: ANCON0, ANCON1 and ANCON2.

PIC18F87K90 FAMILY
11.2 PORTA, TRISA and OSC2/CLKO/RA6 and OSC1/CLKI/RA7 normally
LATA Registers serve as the external circuit connections for the exter-
nal (primary) oscillator circuit (HS Oscillator modes) or
PORTA is an 8-bit wide, bidirectional port. The corre- the external clock input and output (EC Oscillator
sponding Data Direction and Output Latch registers are modes). In these cases, RA6 and RA7 are not available
TRISA and LATA. as digital I/O and their corresponding TRIS and LAT
RA4/T0CKI is a Schmitt Trigger input. All other PORTA bits are read as ‘0’. When the device is configured to
pins have TTL input levels and full CMOS output use HF-INTOSC, MF-INTOSC or LF-INTOSC as the
drivers. default oscillator mode, RA6 and RA7 are automatically
configured as digital I/O; the oscillator and clock
The RA4 pin is multiplexed with the Timer0 clock input in/clock out functions are disabled.
and one of the LCD segment drives. RA5 and RA<3:0>
are multiplexed with analog inputs for the A/D RA1, RA4 and RA5 are multiplexed with LCD segment
Converter. RA1 is multiplexed with analog as well as drives that are controlled by bits in the LCDSE1 and
the LCD segment drive. LCDSE2 registers. I/O port functionality is only
available when the LCD segments are disabled.
The operation of the analog inputs as A/D Converter
inputs is selected by clearing or setting the RA5 has additional functionality for Timer1 and Timer3.
ANSEL<3:0> control bits in the ANCON1 register. The It can be configured as the Timer1 clock input or the
corresponding TRISA bits control the direction of these Timer3 external clock gate input.
pins, even when they are being used as analog inputs.
The user must ensure the bits in the TRISA register are EXAMPLE 11-1: INITIALIZING PORTA
maintained set when using them as analog inputs. CLRF PORTA ; Initialize PORTA by
; clearing output latches
Note: RA5 and RA<3:0> are configured as CLRF LATA ; Alternate method to
analog inputs on any Reset and are read ; clear output data latches
as ‘0’. RA4 is configured as a digital input. BANKSEL ANCON1
MOVLW 00h ; Configure A/D
MOVWF ANCON1 ; for digital inputs
MOVLW 0BFh ; Value used to initialize
; data direction
MOVWF TRISA ; Set RA<7, 5:0> as inputs,
; RA<6> as output

PIC18F87K90 FAMILY
TABLE 11-1: PORTA FUNCTIONS
TRIS I/O
Pin Name Function I/O Description
Setting Type
RA0/AN0/ULPWU RA0 0 O DIG LATA<0> data output; not affected by analog input.
1 I TTL PORTA<0> data input; disabled when analog input is enabled.
AN0 1 I ANA A/D Input Channel 0. Default input configuration on POR; does not
affect digital output.
ULPWU 1 I ANA Ultra Low-Power Wake-up (ULPWU) input.
RA1/AN1/SEG18 RA1 0 O DIG LATA<1> data output; not affected by analog input.
SEG18 1 O ANA LCD Segment 18 output; disables all other pin functions.
RA2/AN2/VREF- RA2 0 O DIG LATA<2> data output; not affected by analog input.
1 I TTL PORTA<2> data input; disabled when analog functions are enabled.
AN2 1 I ANA A/D Input Channel 2. Default input configuration on POR.
VREF- 1 I ANA A/D and comparator low reference voltage input.
RA3/AN3/VREF+ RA3 0 O DIG LATA<3> data output; not affected by analog input.
AN3 1 I ANA A/D Input Channel 3. Default input configuration on POR.
VREF+ 1 I ANA A/D and comparator high reference voltage input.
RA4/T0CKI/ RA4 0 O DIG LATA<4> data output.
SEG14 1 I ST PORTA<4> data input. Default configuration on POR.
T0CKI x I ST Timer0 clock input.
RA5/AN4/SEG15/ RA5 0 O DIG LATA<5> data output; not affected by analog input.
T1CKI/T3G/ 1 I TTL PORTA<5> data input; disabled when analog input is enabled.
HLVDIN
AN4 1 I ANA A/D Input Channel 4. Default configuration on POR.
T3G x I ST Timer3 external clock gate input.
HLVDIN 1 I ANA High/Low-Voltage Detect (HLVD) external trip point input.
OSC2/CLKO/RA6 OSC2 x O ANA Main oscillator feedback output connection (HS, XT and LP modes).
CLKO x O DIG System cycle clock output (FOSC/4, EC and INTOSC modes).
RA6 0 O DIG LATA<6> data output; disabled when OSC2 Configuration bit is set.
1 I TTL PORTA<6> data input; disabled when OSC2 Configuration bit is set.
OSC1/CLKI/RA7 OSC1 x I ANA Main oscillator input connection (HS, XT and LP modes).
CLKI x I ANA Main external clock source input (EC modes).
RA7 0 O DIG LATA<7> data output; disabled when OSC2 Configuration bit is set.
1 I TTL PORTA<7> data input; disabled when OSC2 Configuration bit is set.
Legend: O = Output, I = Input, ANA = Analog Signal, DIG = Digital Output, ST = Schmitt Trigger Buffer Input,
TTL = TTL Buffer Input, x = Don’t care (TRIS bit does not affect port direction or is overridden for this option).

PIC18F87K90 FAMILY
TABLE 11-2: SUMMARY OF REGISTERS ASSOCIATED WITH PORTA
Reset
Page:
PORTA RA7(1) RA6(1) RA5 RA4 RA3 RA2 RA1 RA0 78
(1)
LATA LATA7 LATA6(1) LATA5 LATA4 LATA3 LATA2 LATA1 LATA0 78
TRISA TRISA7(1) TRISA6(1) TRISA5 TRISA4 TRISA3 TRISA2 TRISA1 TRISA0 78
ANCON1 ANSEL15 ANSEL14 ANSEL13 ANSEL12 ANSEL11 ANSEL10 ANSEL9 ANSEL8 81
LCDSE1 SE15 SE14 SE13 SE12 SE11 SE10 SE09 SE08 83
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by PORTA.
Note 1: These bits are enabled depending on the oscillator mode selected. When not enabled as PORTA pins, they
are disabled and read as ‘x’.

PIC18F87K90 FAMILY
11.3 PORTB, TRISB and Four of the PORTB pins (RB<7:4>) have an
LATB Registers interrupt-on-change feature. Only pins configured as
inputs can cause this interrupt to occur. Any RB<7:4>
PORTB is an 8-bit wide, bidirectional port. The pin configured as an output will be excluded from the
corresponding Data Direction and Output Latch registers interrupt-on-change comparison.
are TRISB and LATB. All pins on PORTB are digital only.
Comparisons with the input pins (of RB<7:4>) are
made with the old value latched on the last read of
EXAMPLE 11-2: INITIALIZING PORTB PORTB. The “mismatch” outputs of RB<7:4> are ORed
CLRF PORTB ; Initialize PORTB by together to generate the RB Port Change Interrupt with
; clearing output Flag bit, RBIF (INTCON<0>).
; data latches
CLRF LATB ; Alternate method
This interrupt can wake the device from
; to clear output power-managed modes. To clear the interrupt in the
; data latches Interrupt Service Routine (ISR):
MOVLW 0CFh ; Value used to a) Any read or write of PORTB (except with the
; initialize data MOVFF (ANY), PORTB instruction). This will end
; direction
the mismatch condition.
MOVWF TRISB ; Set RB<3:0> as inputs
; RB<5:4> as outputs b) Wait one instruction cycle (such as executing a
; RB<7:6> as inputs NOP instruction).
c) Clear flag bit, RBIF.
Each of the PORTB pins has a weak internal pull-up. A A mismatch condition will continue to set flag bit, RBIF.
single control bit can turn on all the pull-ups. This is Reading PORTB will end the mismatch condition and
performed by clearing bit, RBPU (INTCON2<7>). The allow flag bit, RBIF, to be cleared after one TCY delay.
weak pull-up is automatically turned off when the port
pin is configured as an output. The pull-ups are The interrupt-on-change feature is recommended for
disabled on a Power-on Reset. wake-up on key depression operation and operations
where PORTB is only used for the interrupt-on-change
feature. Polling of PORTB is not recommended while
using the interrupt-on-change feature.
The RB<3:2> pins are multiplexed as CTMU edge
inputs. RB5 has an additional function for Timer3 and
Timer1. It can be configured for the Timer3 clock input
or Timer1 external clock gate input.
The RB<5:0> pins also are multiplexed with LCD seg-
ment drives that are controlled by bits in the registers,
LCDSE1 and LCDSE3. I/O port functionality is only
available when the LCD segments are disabled.

PIC18F87K90 FAMILY
TABLE 11-3: PORTB FUNCTIONS
TRIS I/O
Setting Type
RB0/INT0/SEG30/ RB0 0 O DIG LATB<0> data output.
FLT0 1 I TTL PORTB<0> data input; weak pull-up when RBPU bit is cleared.
INT0 1 I ST External Interrupt 0 input.
FLT0 x I ST Enhanced PWM Fault input for ECCPx.
RB1/INT1/SEG8 RB1 0 O DIG LATB<1> data output.
1 I TTL PORTB<1> data input; weak pull-up when RBPU bit is cleared.
CTED1
CTED1 x I ST CTMU Edge 1 input.
CTED2/ECCP2/
P2A
CTED2 x I ST CTMU Edge 2 input.
ECCP2 0 O DIG ECCP2 compare output and ECCP2 PWM output. Takes priority
over port data.
1 I ST ECCP2 capture input.
P2A 0 O DIG ECCP2 Enhanced PWM output, Channel A. May be configured for
tri-state during Enhanced PWM shutdown events. Takes priority
over port data.
RB4/KBI0/SEG11 RB4 0 O DIG LATB<4> data output.
KBI0 1 I TTL Interrupt-on-pin change.
RB5/KBI1/SEG29/ RB5 0 O DIG LATB<5> data output.
T3CKI/T1G
RB6/KBI2/PGC RB6 0 O DIG LATB<6> data output.
PGC x I ST Serial execution (ICSP™) clock input for ICSP and ICD operations.
RB7/KBI3/PGD RB7 0 O DIG LATB<7> data output.
PGD x O DIG Serial execution data output for ICSP and ICD operations.
x I ST Serial execution data input for ICSP and ICD operations.

PIC18F87K90 FAMILY
TABLE 11-4: SUMMARY OF REGISTERS ASSOCIATED WITH PORTB
Reset
Page:
PORTB RB7 RB6 RB5 RB4 RB3 RB2 RB1 RB0 78
LATB LATB7 LATB6 LATB5 LATB4 LATB3 LATB2 LATB1 LATB0 78
TRISB TRISB7 TRISB6 TRISB5 TRISB4 TRISB3 TRISB2 TRISB1 TRISB0 78
INTCON2 RBPU INTEDG0 INTEDG1 INTEDG2 INTEDG3 TMR0IP INT3IP RBIP 75
INTCON3 INT2IP INT1IP INT3IE INT2IE INT1IE INT3IF INT2IF INT1IF 75
Legend: Shaded cells are not used by PORTB.

PIC18F87K90 FAMILY
11.4 PORTC, TRISC and The contents of the TRISC register are affected by
LATC Registers peripheral overrides. Reading TRISC always returns
the current contents, even though a peripheral device
PORTC is an 8-bit wide, bidirectional port. The may be overriding one or more of the pins.
corresponding Data Direction and Output Latch registers
The RC<7:1> pins are multiplexed with LCD segment
are TRISC and LATC. Only PORTC pins, RC2 through
drives that are controlled by bits in the registers:
RC7, are digital only pins.
LCDSE1, LCDSE2, LCDSE3 and LCDSE4.
PORTC is multiplexed with ECCP, MSSP and EUSART
RC0 and RC1 pins serve as the input pins for the
peripheral functions (Table 11-5). The pins have
SOSC oscillator. On a power-up, these pins are defined
Schmitt Trigger input buffers. The pins for ECCP, SPI
as SOSC pins. In order to make these ports have digital
and EUSART are also configurable for open-drain out-
I/O port functionality, the CONFI1L<4:3> should be set
put whenever these functions are active. Open-drain
to ‘10’ (Digital SCLKI mode). I/O port functionality is
configuration is selected by setting the SSP1OD,
only available when the LCD segments are disabled.
CCPxOD and U1OD control bits in the registers,
ODCON1 and ODCON3.
EXAMPLE 11-3: INITIALIZING PORTC
RC1 is normally configured as the default peripheral
CLRF PORTC ; Initialize PORTC by
pin for the ECCP2 module. The assignment of ECCP2
; clearing output
is controlled by Configuration bit, CCP2MX (default
; data latches
state, CCP2MX = 1). CLRF LATC ; Alternate method
When enabling peripheral functions, use care in defining ; to clear output
TRIS bits for each PORTC pin. Some peripherals can ; data latches
override the TRIS bit to make a pin an output or input. MOVLW 0CFh ; Value used to
Consult the corresponding peripheral section for the ; initialize data
; direction
correct TRIS bit settings.
MOVWF TRISC ; Set RC<3:0> as inputs
Note: These pins are configured as digital inputs ; RC<5:4> as outputs
on any device Reset. ; RC<7:6> as inputs

PIC18F87K90 FAMILY
TABLE 11-5: PORTC FUNCTIONS
TRIS
Pin Name Function I/O I/O Type Description
Setting
RC0/SOSCO/ RC0 0 O DIG LATC<0> data output.
SCLKI/ 1 I ST PORTC<0> data input.
SOSCO 1 I ST SOSC oscillator output.
SCLKI x O ANA Digital clock input; enabled when SOSC oscillator is disabled.
RC1/SOSCI/ RC1 0 O DIG LATC<1> data output.
ECCP2/P2A/ 1 I ST PORTC<1> data input.
SEG32
SOSCI x I ANA SOSC oscillator input.
ECCP2(1) 0 O DIG ECCP2 compare output and ECCP2 PWM output; takes priority over port data.
P2A 0 O DIG ECCP2 Enhanced PWM output, Channel A. May be configured for tri-state
during Enhanced PWM shutdown events; takes priority over port data.
RC2/ECCP1/ RC2 0 O DIG LATC<2> data output.
P1A/SEG13 1 I ST PORTC<2> data input.
ECCP1 0 O DIG ECCP1 compare output and ECCP1 PWM output; takes priority over port data.
P1A 0 O DIG ECCP1 Enhanced PWM output, Channel A. May be configured for tri-state
during Enhanced PWM shutdown events; takes priority over port data.
RC3/SCK1/ RC3 0 O DIG LATC<3> data output.
SCL1/SEG17 1 I ST PORTC<3> data input.
SCK1 0 O DIG SPI clock output (MSSP module); takes priority over port data.
1 I ST SPI clock input (MSSP module).
SCL1 0 O DIG I2C™ clock output (MSSP module); takes priority over port data.
1 I I2C I2C clock input (MSSP module); input type depends on module setting.
RC4/SDI1/ RC4 0 O DIG LATC<4> data output.
SDA1/SEG16 1 I ST PORTC<4> data input.
SDI1 I ST SPI data input (MSSP module).
SDA1 1 O DIG I2C data output (MSSP module); takes priority over port data.
1 I I2C I2C data input (MSSP module); input type depends on module setting.
RC5/SDO1/ RC5 0 O DIG LATC<5> data output.
SEG12 1 I ST PORTC<5> data input.
SDO1 0 O DIG SPI data output (MSSP module).
RC6/TX1/CK1/ RC6 0 O DIG LATC<6> data output.
TX1 1 O DIG Synchronous serial data output (EUSART module); takes priority over port data.
CK1 1 O DIG Synchronous serial data input (EUSART module); user must configure as an input.
1 I ST Synchronous serial clock input (EUSART module).
Legend: O = Output, I = Input, ANA = Analog Signal, DIG = Digital Output, ST = Schmitt Trigger Buffer Input, TTL = TTL Buffer Input,
I2C = I2C Buffer Input, x = Don’t care (TRIS bit does not affect port direction or is overridden for this option).

PIC18F87K90 FAMILY
TABLE 11-5: PORTC FUNCTIONS (CONTINUED)
TRIS
Setting
RC7/RX1/DT1/ RC7 0 O DIG LATC<7> data output.
RX1 1 I ST Asynchronous serial receive data input (EUSART module).
DT1 1 O DIG Synchronous serial data output (EUSART module); takes priority over port data.
1 I ST Synchronous serial data input (EUSART module); user must configure as an input.
Legend: O = Output, I = Input, ANA = Analog Signal, DIG = Digital Output, ST = Schmitt Trigger Buffer Input, TTL = TTL Buffer Input,
TABLE 11-6: SUMMARY OF REGISTERS ASSOCIATED WITH PORTC

Reset
on Page:
PORTC RC7 RC6 RC5 RC4 RC3 RC2 RC1 RC0 78
LATC LATC7 LATBC6 LATC5 LATCB4 LATC3 LATC2 LATC1 LATC0 78
TRISC TRISC7 TRISC6 TRISC5 TRISC4 TRISC3 TRISC2 TRISC1 TRISC0 78
LCDSE4 SE39(1) SE38(1) SE37(1) SE36(1) SE35(1) SE34(1) SE33(1) SE32 83
ODCON1 SSP1OD CCP2OD CCP1OD — — — — SSP2OD 81
Legend: Shaded cells are not used by PORTC.
Note 1: This bit is unimplemented in PIC18F6XK90 devices, read as ‘0’.

PIC18F87K90 FAMILY
11.5 PORTD, TRISD and I/O port functionality is only available when the LCD
LATD Registers segments are disabled.
The PORTD also has the I2C and SPI functionality on
PORTD is an 8-bit wide, bidirectional port. The
RD4, RD5 and RD6. The pins for SPI are also configu-
rable for open-drain output. Open-drain configuration is
are TRISD and LATD.
selected by setting the SSPxOD control bits in the
All pins on PORTD are implemented with Schmitt ODCON1 register.
Trigger input buffers. Each pin is individually
RD0 has a CTMU functionality. RD1 has the functionality
configurable as an input or output.
for a Timer5 clock input and also Timer7 has functionality
Note: These pins are configured as digital inputs for an external clock gate input.
on any device Reset.
EXAMPLE 11-4: INITIALIZING PORTD
Each of the PORTD pins has a weak internal pull-up. A
single control bit can turn off all the pull-ups. This is CLRF PORTD ; Initialize PORTD by
performed by clearing bit, RDPU (PADCFG1<7>). The ; clearing output
; data latches
weak pull-up is automatically turned off when the port
CLRF LATD ; Alternate method
pin is configured as an output. The pull-ups are
; to clear output
disabled on all device Resets. ; data latches
All of the PORTD pins are multiplexed with LCD MOVLW 0CFh ; Value used to
segment drives that are controlled by bits in the ; initialize data
LCDSE0 register. RD0 is multiplexed with the CTMU ; direction
MOVWF TRISD ; Set RD<3:0> as inputs
pulse generator output.
; RD<5:4> as outputs
; RD<7:6> as inputs

PIC18F87K90 FAMILY
TABLE 11-7: PORTD FUNCTIONS
TRIS
Setting
RD0/SEG0/ RD0 0 O DIG LATD<0> data output.
CTPLS 1 I ST PORTD<0> data input.
CTPLS x O DIG CTMU pulse generator output.
T5CKI/T7G 1 I ST PORTD<1> data input.
RD2/SEG2 RD2 0 O DIG LATD<2> data output.
1 I ST PORTD<2> data input.
RD3/SEG3 RD3 0 O DIG LATD<3> data output.
1 I ST PORTD<3> data input.
SDO2 1 I ST PORTD<4> data input.
SDO2 0 P DOG SPI data output (MSSP module).
SDI2/SDA2 1 I ST PORTD<5> data input.
SDI2 1 I ST SPI data input (MSSP module).
SDA2 0 O I2C I2C™ data input (MSSP module). Input type depends on module setting.
1 I ANA LCD Segment 5 output; disables all other pin functions.
SCK2/SCL2 1 I ST PORTD<6> data input.
SCK2 0 O DIG SPI clock output (MSSP module); takes priority over port data.
1 I ST SPI clock input (MSSP module).
SCL2 0 O DIG I2C clock output (MSSP module); takes priority over port data.
1 I I2C I2C clock input (MSSP module). Input type depends on module setting.
SS2 1 I ST PORTD<7> data input.
SEG7 1 I ANA LCD Segment 7 output; disables all other pin functions.
SS2 1 I TTL Slave select input for MSSP module.

PIC18F87K90 FAMILY
TABLE 11-8: SUMMARY OF REGISTERS ASSOCIATED WITH PORTD
Reset
Page:
PORTD RD7 RD6 RD5 RD4 RD3 RD2 RD1 RD0 78
LATD LATD7 LATD6 LATD5 LATD4 LATD3 LATD2 LATD1 LATD0 78
TRISD TRISD7 TRISD6 TRISD5 TRISD4 TRISD3 TRISD2 TRISD1 TRISD0 78
PADCFG1 RDPU REPU RJPU(1) — — RTSECSEL1 RTSECSEL0 — 80
Legend: Shaded cells are not used by PORTD.
Note 1: This bit is not available in 64-pin devices.

PIC18F87K90 FAMILY
11.6 PORTE, TRISE and Pins, RE2, RE1 and RE0, are multiplexed with the
LATE Registers functions of LCDBIAS3, LCDBIAS2 and LCDBIAS1.
When LCD bias generation is required (in any applica-
PORTE is an 8-bit wide, bidirectional port. The tion where the device is connected to an external LCD),
corresponding Data Direction and Output Latch registers these pins cannot be used as digital I/O. These pins
are TRISE and LATE. can be used as digital I/O, however, when the internal
All pins on PORTE are implemented with Schmitt resistor ladder is used for bias generation.
Trigger input buffers. Each pin is individually PORTE is also multiplexed with the Enhanced PWM
configurable as an input or output. The RE7 pin is also Outputs B and C for ECCP1 and ECCP3, and Outputs
configurable for open-drain output when ECCP2 is B, C and D for ECCP2. For all devices, their default
active on this pin. Open-drain configuration is selected assignments are on PORTE<6:0>. On 80-pin devices,
by setting the CCP2OD control bit (ODCON1<6>) the multiplexing for the outputs of ECCP1 and ECCP3 is
controlled by the ECCPMX Configuration bit. Clearing
Note: These pins are configured as digital inputs
this bit reassigns the P1B/P1C and P3B/P3C outputs to
on any device Reset.
PORTH.
Each of the PORTE pins has a weak internal pull-up. A RE7 is multiplexed with the LCD segment drive
single control bit can turn off all the pull-ups. This is (SEG31) that is controlled by the LCDSE3<7> bit. I/O
performed by clearing bit, REPU (PADCFG1<6>). The port function is only available when the segment is dis-
weak pull-up is automatically turned off when the port abled. RE7 can also be configured as the alternate
pin is configured as an output. The pull-ups are peripheral pin for the ECCP2 module. This is done by
disabled on any device Reset. clearing the CCP2MX Configuration bit.
Pins, RE<6:3>, are multiplexed with the LCD common RE3 can also be configured as the Reference Clock
drives. I/O port functions are available only on those Output (REFO) from the system clock. For further details,
PORTE pins according to which commons are active. refer to Section 3.7 “Reference Clock Output”.
The configuration is determined by the LMUX<1:0>
control bits (LCDCON<1:0>). The availability is
EXAMPLE 11-5: INITIALIZING PORTE
summarized in Table 11-9.
CLRF PORTE ; Initialize PORTE by
; clearing output
TABLE 11-9: PORTE PINS AVAILABLE IN ; data latches
DIFFERENT LCD DRIVE CLRF LATE ; Alternate method
CONFIGURATIONS(1) ; to clear output
; data latches
LCDCON Active LCD PORTE Pins MOVLW 03h ; Value used to
<1:0> Commons Available for I/O ; initialize data
; direction
00 COM0 RE6, RE5, RE4 MOVWF TRISE ; Set RE<1:0> as inputs
01 COM0, COM1 RE6, RE5 ; RE<7:2> as outputs
10 COM0, COM1 RE6
and COM2
11 All (COM0 None
through COM3)
Note 1: If the LCD bias voltages are generated
using the internal resistor ladder, the
LCDBIASx pins are also available as I/O
ports (RE0, RE1 and RE2).

PIC18F87K90 FAMILY
TABLE 11-10: PORTE FUNCTIONS
TRIS I/O
Setting Type
RE0/LCDBIAS1/ RE0 0 O DIG LATE<0> data output.
P2D 1 I ST PORTE<0> data input.
LCDBIAS1 — I ANA LCD module bias voltage input.
P2D 0 O — ECCP2 PWM Output D. May be configured for tri-state during
Enhanced PWM shutdown events.
P2C 1 I ST PORTE<1> data input.
LCDBIAS2 — I ANA LCD module bias voltage input.
P2C 0 O — ECCP2 PWM Output C.
May be configured for tri-state during Enhanced PWM shutdown events.
P2B 1 I ST PORTE<2> data input.
LCDBIAS3 x I ANA LCD module bias voltage input.
P2B 0 O — ECCP2 PWM Output B. May be configured for tri-state during
RE3/COM0/ RE3 0 O DIG LATE<3> data output.
P3C/CCP9/ 1 I ST PORTE<3> data input.
REFO
COM0 x O ANA LCD Common 0 output; disables all other outputs.
P3C 0 O — ECCP3 PWM Output C. May be configured for tri-state during
CCP9(2) 0 O DIG CCP9 compare/PWM output; takes priority over port data.
1 I ST CCP9 capture input.
REFO x O DIG Reference output clock.
P3B/CCP8 1 I ST PORTE<4> data input.
CCP8 0 O DIG CCP8 Compare/PWM output; takes priority over port data.
P1C/CCP7 1 I ST PORTE<5> data input.
P1B/CCP6 1 I ST PORTE<6> data input.
x = Don’t care (TRIS bit does not affect port direction or is overridden for this option).
Note 1: Alternate assignment for ECCP2 when the CCP2MX Configuration bit is cleared.
2: This bit is unimplemented in PIC18FX5K90 devices.

PIC18F87K90 FAMILY
TABLE 11-10: PORTE FUNCTIONS (CONTINUED)
TRIS I/O
Setting Type
RE7/ECCP2/ RE7 0 O DIG LATE<7> data output.
P2A/SEG31 1 I ST PORTE<7> data input.
ECCP2(1) 0 O DIG ECCP2 compare/PWM output; takes priority over port data.
P2A 0 O — ECCP2 PWM Output A. May be configured for tri-state during
Enhanced PWM shutdown event.
SEG31 1 O ANA Segment 31 analog output for LCD; disables digital output.
Note 1: Alternate assignment for ECCP2 when the CCP2MX Configuration bit is cleared.
2: This bit is unimplemented in PIC18FX5K90 devices.
TABLE 11-11: SUMMARY OF REGISTERS ASSOCIATED WITH PORTE

Reset
on Page:
PORTE RE7 RE6 RE5 RE4 RE3 RE2 RE1 RE0 78
LATE LATE7 LATE6 LATE5 LATE4 LATE3 LATE2 LATE1 LATE0 78
TRISE TRISE7 TRISE6 TRISE5 TRISE4 TRISE3 TRISE2 TRISE1 TRISE0 78
LCDCON LCDEN SLPEN WERR — CS1 CS0 LMUX1 LMUX0 83
ODCON2 CCP10OD(2) CCP9OD(2) CCP8OD CCP7OD CCP6OD CCP5OD CCP4OD CCP3OD 81
Legend: Shaded cells are not used by PORTE.
Note 1: This bit is not available in 64-pin devices.

PIC18F87K90 FAMILY
11.7 PORTF, LATF and TRISF Registers PORTF is also multiplexed with LCD segment drives
controlled by bits in the LCDSE2 and LCDSE3
PORTF is a 7-bit wide, bidirectional port. The registers. I/O port functions are only available when the
corresponding Data Direction and Output Latch registers segments are disabled.
are TRISF and LATF. All pins on PORTF are
implemented with Schmitt Trigger input buffers. Each pin
EXAMPLE 11-6: INITIALIZING PORTF
is individually configurable as an input or output.
CLRF PORTF ; Initialize PORTF by
PORTF is multiplexed with analog peripheral functions, ; clearing output
as well as LCD segments. Pins, RF1 through RF6, may ; data latches
be used as comparator inputs or outputs by setting the CLRF LATF ; Alternate method
appropriate bits in the CMCON register. To use ; to clear output
RF<7:1> as digital inputs, it is also necessary to turn off ; data latches
the comparators. BANKSEL ANCON1
MOVLW 01Fh ; Make AN6, AN7 and AN5 digital
Note 1: On device Resets, pins, RF<7:1>, are MOVWF ANCON1 ;
configured as analog inputs and are read MOVLW 0F0h ; Make AN8, AN9, AN10 and AN11
as ‘0’. digital
MOVWF ANCON2 ; Set PORTF as digital I/O
2: To configure PORTF as a digital I/O, turn MOVLW 0CEh ; Value used to
off the comparators and clear ANCON1 ; initialize data
and ANCON2 to digital. ; direction
MOVWF TRISF ; Set RF3:RF1 as inputs
; RF5:RF4 as outputs
; RF7:RF6 as inputs

PIC18F87K90 FAMILY
TABLE 11-12: PORTF FUNCTIONS
TRIS I/O
Setting Type
RF1/AN6/C2OUT/ RF1 0 O DIG LATF<1> data output; not affected by analog input.
SEG19/CTDIN 1 I ST PORTF<1> data input; disabled when analog input is enabled.
AN6 1 I ANA A/D Input Channel 6; default configuration on POR.
C2OUT 0 O DIG Comparator 2 output; takes priority over port data.
CTDIN 1 I ST CTMU pulse delay input.
RF2/AN7/C1OUT/ RF2 0 O DIG LATF<2> data output; not affected by analog input.
SEG20 1 I ST PORTF<2> data input; disabled when analog input is enabled.
AN7 1 I ANA A/D Input Channel 7; default configuration on POR.
C1OUT 0 O DIG Comparator 1 output; takes priority over port data.
RF3/AN8/SEG21/ RF3 0 O DIG LATF<3> data output; not affected by analog input.
C2INB/CTMUI 1 I ST PORTF<3> data input; disabled when analog input is enabled.
AN8 1 I ANA A/D Input Channel 8 and Comparator C2+ input. Default input
configuration on POR; not affected by analog output.
C2INB 1 I ANA Comparator 2 Input B.
CTMUI x O — CTMU pulse generator charger for the C2INB comparator input.
C2INA 1 I ST PORTF<4> data input; disabled when analog input is enabled.
AN9 1 I ANA A/D Input Channel 9 and Comparator C2- input. Default input
configuration on POR; does not affect digital output.
C2INA 1 I ANA Comparator 2 Input A.
RF5/AN10/CVREF/ RF5 0 O DIG LATF<5> data output; not affected by analog input. Disabled when
SEG23/C1INB CVREF output is enabled.
1 I ST PORTF<5> data input; disabled when analog input is enabled.
Disabled when CVREF output is enabled.
AN10 1 I ANA A/D Input Channel 10 and Comparator C1+ input; default input
configuration on POR.
CVREF x O ANA Comparator voltage reference output. Enabling this feature disables
digital I/O.
C1INB 1 I ANA Comparator 1 Input B.
C1INA 1 I ST PORTF<6> data input; disabled when analog input is enabled.
AN11 1 I ANA A/D Input Channel 11 and Comparator C1- input. Default input
configuration on POR; does not affect digital output.
C1INA 1 I ANA Comparator 1 Input A.
RF7/AN5/SS1/ RF7 0 O DIG LATF<7> data output; not affected by analog input.
SEG25 1 I ST PORTF<7> data input; disabled when analog input is enabled.
AN5 1 I ANA A/D Input Channel 5. Default configuration on POR.
SS1 1 I TTL Slave select input for MSSP module.

PIC18F87K90 FAMILY
TABLE 11-13: SUMMARY OF REGISTERS ASSOCIATED WITH PORTF
Reset
on Page:
PORTF RF7 RF6 RF5 RF4 RF3 RF2 RF1 — 78
LATF LATF7 LATF6 LATF5 LATF4 LATF3 LATF2 LATF1 — 78
TRISF TRISF7 TRISF6 TRISF5 TRISF4 TRISF3 TRISF2 TRISF1 — 78
CMSTAT CMP3OUT CMP2OUT CMP1OUT — — — — — 77
CVRCON CVREN CVROE CVRR CVRSS CVR3 CVR2 CVR1 CVR0 77
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by PORTF.

PIC18F87K90 FAMILY
11.8 PORTG, TRISG and When enabling peripheral functions, care should be
LATG Registers taken in defining TRIS bits for each PORTG pin. Some
peripherals override the TRIS bit to make a pin an
PORTG is a 5-bit wide, bidirectional port. The output, while other peripherals override the TRIS bit to
corresponding Data Direction and Output Latch registers make a pin an input. The user should refer to the
are TRISG and LATG. corresponding peripheral section for the correct TRIS bit
PORTG is multiplexed with EUSART, LCD and settings. The pin override value is not loaded into the
CCP/ECCP/Analog/Comparator/RTCC/Timer input func- TRIS register. This allows read-modify-write of the TRIS
tions (Table 11-14). When operating as I/O, all PORTG register without concern due to peripheral overrides.
pins have Schmitt Trigger input buffers. The open-drain
functionality for the CCPx and UART can be configured EXAMPLE 11-7: INITIALIZING PORTG
using ODCONx. CLRF PORTG ; Initialize PORTG by
; clearing output
RG4 is multiplexed with LCD segment drives controlled
; data latches
by bits in the LCDSE2 register and as the
BCF CM1CON, CON ; disable
RG4/SEG26/RTCC/T7CKI/T5G/CCP5/AN16/P1D/C3INC ; comparator 1
pin. The I/O port function is only available when the CLRF LATG ; Alternate method
segments are disabled. ; to clear output
The RG5 pin is multiplexed with the MCLR pin and is ; data latches
BANKSEL ANCON2
available only as an input port. To configure this port for
MOVLW 0F0h ; make AN16 to AN19
input only, set the MCLRE pin (CONFIG3H<7>).
; digital
MOVWF ANCON2
MOVLW 04h ; Value used to
; initialize data
; direction
MOVWF TRISG ; Set RG1:RG0 as
; outputs
; RG2 as input
; RG4:RG3 as inputs
TABLE 11-14: PORTG FUNCTIONS

TRIS I/O
Setting Type
RG0/ECCP3/ RG0 0 O DIG LATG<0> data output.
P3A 1 I ST PORTG<0> data input.
ECCP3 0 O DIG ECCP3 compare output and ECCP3 PWM output; takes priority over
port data.
P3A 0 O — ECCP3 PWM Output A. May be configured for tri-state during
RG1/TX2/CK2/ RG1 0 O DIG LATG<1> data output.
AN19/C3OUT 1 I ST PORTG<1> data input.
TX2 1 O DIG Synchronous serial data output (EUSART module); takes priority over
port data.
CK2 1 O DIG Synchronous serial data input (EUSART module); user must configure
as an input.
1 I ST Synchronous serial clock input (EUSART module).
AN19 1 I ANA A/D Input Channel 19. Default input configuration on POR. Does not
C3OUT x O DIG Comparator 3 output.

PIC18F87K90 FAMILY
TABLE 11-14: PORTG FUNCTIONS (CONTINUED)
TRIS I/O
Setting Type
RG2/RX2/DT2/ RG2 0 O DIG LATG<2> data output.
AN18/C3INA 1 I ST PORTG<2> data input.
RX2 1 I ST Asynchronous serial receive data input (EUSART module).
DT2 1 O DIG Synchronous serial data output (EUSART module); takes priority over
port data.
1 I ST Synchronous serial data input (EUSART module); user must configure
as an input.
C3INA x I ANA Comparator 3 Input A.
RG3/CCP4/AN17/ RG3 0 O DIG LATG<3> data output.
P3D/C3INB 1 I ST PORTG<3> data input.
CCP4 0 O DIG CCP4 compare/PWM output; takes priority over port data.
AN17 1 I ANA A/D Input Channel 17. Default input configuration on PR; does not
C3INB x I ANA Comparator 3 Input B.
Enhanced PWM.
RG4/SEG26/ RG4 0 O DIG LATG<4> data output.
RTCC/T7CKI/ 1 I ST PORTG<4> data input.
T5G/CCP5/
AN16/P1D/
C3INC RTCC x O DIG RTCC output.
C3INC x I ANA Comparator 3 Input C.
Enhanced PWM.
RG5 I ST See the MCLR/RG5 pin.
TABLE 11-15: SUMMARY OF REGISTERS ASSOCIATED WITH PORTG

Reset Values
Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
on Page:
PORTG — — RG5(1) RG4 RG3 RG2 RG1 RG0 78
TRISG — — — TRISG4 TRISG3 TRISG2 TRISG1 TRISG0 78
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by PORTG.

PIC18F87K90 FAMILY
11.9 PORTH, LATH and EXAMPLE 11-8: INITIALIZING PORTH
TRISH Registers CLRF PORTH ; Initialize PORTH by
; clearing output
Note: PORTH is available only on the 80-pin ; data latches
devices. CLRF LATH ; Alternate method
; to clear output
PORTH is an 8-bit wide, bidirectional I/O port. The ; data latches
corresponding Data Direction and Output Latch registers BANKSEL ANCON2
are TRISH and LATH. MOVLW 0Fh ; Configure PORTH as
MOVWF ANCON2 ; digital I/O
All pins on PORTH are implemented with Schmitt
MOVLW 0Fh ; Configure PORTH as
Trigger input buffers. Each pin is individually
MOVWF ANCON1 ; digital I/O
configurable as an input or output. MOVLW 0CFh ; Value used to
All PORTH pins are multiplexed with the ; initialize data
ADC/CCP/Comparator and LCD segment drives ; direction
controlled by the LCDSE5 register. I/O port functions are MOVWF TRISH ; Set RH3:RH0 as inputs
; RH5:RH4 as outputs
only available when the segments are disabled.
; RH7:RH6 as inputs

PIC18F87K90 FAMILY
TABLE 11-16: PORTH FUNCTIONS
TRIS I/O
Setting Type
RH0/SEG47/ RH0 0 O DIG LATH<0> data output.
AN23 1 I ST PORTH<0> data input.
affect digital input.
AN20 1 I ANA A/D Input Channel 20.
Default input configuration on POR; does not affect digital input.
CCP9/P3C/ 1 I ST PORTH<4> data input.
AN12/C2INC
Enhanced PWM.
CCP8/P3B/ 1 I ST PORTH<5> data input.
AN13/C2IND
Enhanced PWM.
C2IND x I ANA Comparator 2 Input D.

PIC18F87K90 FAMILY
TABLE 11-16: PORTH FUNCTIONS (CONTINUED)
TRIS I/O
Setting Type
CCP7/P1C/ 1 I ST PORTH<6> data input.
AN14/C1INC
Enhanced PWM.
CCP6/P1B/ 1 I ST PORTH<7> data input.
AN15
Enhanced PWM.
TABLE 11-17: SUMMARY OF REGISTERS ASSOCIATED WITH PORTH

Reset
on Page:
PORTH RH7 RH6 RH5 RH4 RH3 RH2 RH1 RH0 78
LATH LATH7 LATH6 LATH5 LATH4 LATH3 LATH2 LATH1 LATH0 78
TRISH TRISH7 TRISH6 TRISH5 TRISH4 TRISH3 TRISH2 TRISH1 TRISH0 78

PIC18F87K90 FAMILY
11.10 PORTJ, TRISJ and Each of the PORTJ pins has a weak internal pull-up. A
LATJ Registers single control bit can turn off all the pull-ups. This is
performed by clearing bit RJPU (PADCFG1<5>). The
Note: PORTJ is available only on 80-pin devices. weak pull-up is automatically turned off when the port
pin is configured as an output. The pull-ups are
PORTJ is an 8-bit wide, bidirectional port. The disabled on any device Reset.
are TRISJ and LATJ. EXAMPLE 11-9: INITIALIZING PORTJ
All pins on PORTJ are implemented with Schmitt CLRF PORTJ ; Initialize PORTJ by
Trigger input buffers. Each pin is individually ; clearing output latches
configurable as an input or output. CLRF LATJ ; Alternate method
; to clear output latches
Note: These pins are configured as digital inputs MOVLW 0CFh ; Value used to
on any device Reset. ; initialize data
All PORTJ pins, except RJ0, are multiplexed with LCD ; direction
MOVWF TRISJ ; Set RJ3:RJ0 as inputs
segment drives controlled by the LCDSE4 register. I/O
; RJ5:RJ4 as output
port functions are only available on these pins when the ; RJ7:RJ6 as inputs
segments are disabled.

PIC18F87K90 FAMILY
TABLE 11-18: PORTJ FUNCTIONS
TRIS I/O
Setting Type
RJ0 RJ0 0 O DIG LATJ<0> data output.
1 I ST PORTJ<0> data input.
RJ1/SEG33 RJ1 0 O DIG LATJ<1> data output.
TABLE 11-19: SUMMARY OF REGISTERS ASSOCIATED WITH PORTJ

Reset
on Page:
PORTJ RJ7 RJ6 RJ5 RJ4 RJ3 RJ2 RJ1 RJ0 78
LATJ LATJ7 LATJ6 LATJ5 LATJ4 LATJ3 LATJ2 LATJ1 LATJ0 78
TRISJ TRISJ7 TRISJ6 TRISJ5 TRISJ4 TRISJ3 TRISJ2 TRISJ1 TRISJ0 78
Legend: Shaded cells are not used by PORTJ.
Note 1: Unimplemented in PIC18F6XK90 devices, read as ‘0’.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
12.0 TIMER0 MODULE The T0CON register (Register 12-1) controls all
aspects of the module’s operation, including the
The Timer0 module incorporates the following features: prescale selection. It is both readable and writable.
• Software-selectable operation as a timer or Figure 12-1 provides a simplified block diagram of the
counter in both 8-bit or 16-bit modes Timer0 module in 8-bit mode. Figure 12-2 provides a
• Readable and writable registers simplified block diagram of the Timer0 module in 16-bit
• Dedicated 8-bit, software programmable mode.
prescaler
• Selectable clock source (internal or external)
• Edge select for external clock
• Interrupt-on-overflow
REGISTER 12-1: T0CON: TIMER0 CONTROL REGISTER

TMR0ON T08BIT T0CS T0SE PSA T0PS2 T0PS1 T0PS0
bit 7 bit 0
Legend:
bit 7 TMR0ON: Timer0 On/Off Control bit

1 = Enables Timer0
0 = Stops Timer0
bit 6 T08BIT: Timer0 8-Bit/16-Bit Control bit
1 = Timer0 is configured as an 8-bit timer/counter
0 = Timer0 is configured as a 16-bit timer/counter
bit 5 T0CS: Timer0 Clock Source Select bit
1 = Transition on T0CKI pin input edge
0 = Internal clock (FOSC/4)
bit 4 T0SE: Timer0 Source Edge Select bit
1 = Increment on high-to-low transition on the T0CKI pin
0 = Increment on low-to-high transition on the T0CKI pin
bit 3 PSA: Timer0 Prescaler Assignment bit
1 = Timer0 prescaler is not assigned; Timer0 clock input bypasses the prescaler
0 = Timer0 prescaler is assigned; Timer0 clock input comes from the prescaler output
bit 2-0 T0PS<2:0>: Timer0 Prescaler Select bits
111 = 1:256 Prescale value

PIC18F87K90 FAMILY
12.1 Timer0 Operation internal phase clock (TOSC). There is a delay between
synchronization and the onset of incrementing the
Timer0 can operate as either a timer or a counter. The timer/counter.
mode is selected with the T0CS bit (T0CON<5>). In
Timer mode (T0CS = 0), the module increments on 12.2 Timer0 Reads and Writes in
every clock by default unless a different prescaler value 16-Bit Mode
is selected (see Section 12.3 “Prescaler”). If the
TMR0 register is written to, the increment is inhibited TMR0H is not the actual high byte of Timer0 in 16-bit
for the following two instruction cycles. The user can mode. It is actually a buffered version of the real high
work around this by writing an adjusted value to the byte of Timer0, which is not directly readable nor
TMR0 register. writable (see Figure 12-2). TMR0H is updated with the
The Counter mode is selected by setting the T0CS bit contents of the high byte of Timer0 during a read of
(= 1). In this mode, Timer0 increments, either on every TMR0L. This provides the ability to read all 16 bits of
rising edge or falling edge, of the T0CKI pin. The Timer0 without having to verify that the read of the high
incrementing edge is determined by the Timer0 Source and low byte were valid, due to a rollover between
Edge Select bit, T0SE (T0CON<4>); clearing this bit successive reads of the high and low byte.
selects the rising edge. Restrictions on the external Similarly, a write to the high byte of Timer0 must also
clock input are discussed below. take place through the TMR0H Buffer register. The high
An external clock source can be used to drive Timer0; byte is updated with the contents of TMR0H when a
however, it must meet certain requirements to ensure write occurs to TMR0L. This allows all 16 bits of Timer0
that the external clock can be synchronized with the to be updated at once.
FIGURE 12-1: TIMER0 BLOCK DIAGRAM (8-BIT MODE)
FOSC/4 0
1
Sync with Set
1 Internal TMR0L TMR0IF
Programmable Clocks on Overflow
T0CKI Pin 0
Prescaler
T0SE (2 TCY Delay)
T0CS 8
3
T0PS<2:0>
8
PSA Internal Data Bus
Note: Upon Reset, Timer0 is enabled in 8-bit mode with clock input from T0CKI max. prescale.
FIGURE 12-2: TIMER0 BLOCK DIAGRAM (16-BIT MODE)
FOSC/4 0
1
Sync with TMR0 Set
1 Internal TMR0L High Byte TMR0IF
T0CKI Pin Programmable 0 Clocks on Overflow
Prescaler 8
T0SE (2 TCY Delay)
T0CS 3 Read TMR0L
T0PS<2:0>
Write TMR0L
PSA
8
8
TMR0H
8
8
Internal Data Bus
Note: Upon Reset, Timer0 is enabled in 8-bit mode with clock input from T0CKI max. prescale.

PIC18F87K90 FAMILY
12.3 Prescaler 12.3.1 SWITCHING PRESCALER
ASSIGNMENT
An 8-bit counter is available as a prescaler for the Timer0
module. The prescaler is not directly readable or writable. The prescaler assignment is fully under software
Its value is set by the PSA and T0PS<2:0> bits control and can be changed “on-the-fly” during program
(T0CON<3:0>), which determine the prescaler execution.
assignment and prescale ratio.
Clearing the PSA bit assigns the prescaler to the 12.4 Timer0 Interrupt
Timer0 module. When it is assigned, prescale values The TMR0 interrupt is generated when the TMR0
from 1:2 through 1:256, in power-of-two increments, register overflows from FFh to 00h in 8-bit mode, or
are selectable. from FFFFh to 0000h in 16-bit mode. This overflow sets
When assigned to the Timer0 module, all instructions the TMR0IF flag bit. The interrupt can be masked by
writing to the TMR0 register (for example, CLRF TMR0, clearing the TMR0IE bit (INTCON<5>). Before re-
MOVWF TMR0, BSF TMR0) clear the prescaler count. enabling the interrupt, the TMR0IF bit must be cleared
in software by the Interrupt Service Routine (ISR).
Note: Writing to TMR0 when the prescaler is
assigned to Timer0 will clear the prescaler Since Timer0 is shut down in Sleep mode, the TMR0
count but will not change the prescaler interrupt cannot awaken the processor from Sleep.
assignment.
TABLE 12-1: REGISTERS ASSOCIATED WITH TIMER0

Reset
on Page:
TMR0L Timer0 Register Low Byte 76
TMR0H Timer0 Register High Byte 76
T0CON TMR0ON T08BIT T0CS T0SE PSA T0PS2 T0PS1 T0PS0 76
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by Timer0.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
13.0 TIMER1 MODULE Figure 13-1 displays a simplified block diagram of the
Timer1 module.
The Timer1 timer/counter module incorporates these
The SOSC oscillator can also be used as a low-power
features:
clock source for the microcontroller in power-managed
• Software-selectable operation as a 16-bit timer or operation. Timer1 can also work on the SOSC oscillator.
counter
Timer1 is controlled through the T1CON Control
• Readable and writable 8-bit registers (TMR1H register (Register 13-1), which also contains the SOSC
and TMR1L) Oscillator Enable bit (SOSCEN). Timer1 can be
• Selectable clock source (internal or external) with enabled or disabled by setting or clearing control bit,
device clock or SOSC oscillator internal options TMR1ON (T1CON<0>).
• Interrupt-on-overflow The FOSC clock source should not be used with the
• Reset on ECCP Special Event Trigger ECCP capture/compare features. If the timer will be
• Timer with gated control used with the capture or compare features, always
select one of the other timer clocking options.
TMR1CS1 TMR1CS0 T1CKPS1 T1CKPS0 SOSCEN T1SYNC RD16 TMR1ON
bit 7 bit 0
Legend:
bit 7-6 TMR1CS<1:0>: Timer1 Clock Source Select bits

10 = The Timer1 clock source is either a pin or an oscillator depending on the SOSCEN bit.
SOSCEN = 0:
External clock is from the T1CKI pin (on the rising edge).
SOSCEN = 1:
Crystal oscillator is on the SOSCI/SOSCO pins or an extended clock on SCKLI (depends on SOSCEL
fuse, CONFIG1L<4:3>)
01 = Timer1 clock source is the system clock (FOSC)(1)
00 = Timer1 clock source is the instruction clock (FOSC/4)
bit 5-4 T1CKPS<1:0>: Timer1 Input Clock Prescale Select bits
bit 3 SOSCEN: SOSC Oscillator Enable bit
1 = SOSC is enabled for Timer1 (based on SOSCSEL fuses)
0 = SOSC is disabled for Timer1
The oscillator inverter and feedback resistor are turned off to eliminate power drain.
bit 2 T1SYNC: Timer1 External Clock Input Synchronization Select bit
TMR1CS<1:0> = 10:
1 = Do not synchronize the external clock input
0 = Synchronize the external clock input
TMR1CS<1:0> = 0x:
This bit is ignored. Timer1 uses the internal clock when TMR1CS<1:0> = 1x.
bit 1 RD16: 16-Bit Read/Write Mode Enable bit
1 = Enables register read/write of Timer1 in one 16-bit operation
0 = Enables register read/write of Timer1 in two 8-bit operations
bit 0 TMR1ON: Timer1 On bit
1 = Enables Timer1
0 = Stops Timer1
Note 1: The FOSC clock source should not be selected if the timer will be used with the ECCP capture/compare features.

PIC18F87K90 FAMILY
13.1 Timer1 Gate Control Register
The Timer1 Gate Control register (T1GCON),
displayed in Register 13-2, is used to control the
Timer1 gate.
REGISTER 13-2: T1GCON: TIMER1 GATE CONTROL REGISTER(1)

R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R-x R/W-0 R/W-0
TMR1GE T1GPOL T1GTM T1GSPM T1GGO/T1DONE T1GVAL T1GSS1 T1GSS0
bit 7 bit 0
Legend:
bit 7 TMR1GE: Timer1 Gate Enable bit

If TMR1ON = 0:
This bit is ignored.
If TMR1ON = 1:
1 = Timer1 counting is controlled by the Timer1 gate function
0 = Timer1 counts regardless of the Timer1 gate function
bit 6 T1GPOL: Timer1 Gate Polarity bit
1 = Timer1 gate is active-high (Timer1 counts when gate is high)
0 = Timer1 gate is active-low (Timer1 counts when gate is low)
bit 5 T1GTM: Timer1 Gate Toggle Mode bit
1 = Timer1 Gate Toggle mode is enabled
0 = Timer1 Gate Toggle mode is disabled and toggle flip-flop is cleared
Timer1 gate flip-flop toggles on every rising edge.
bit 4 T1GSPM: Timer1 Gate Single Pulse Mode bit
1 = Timer1 Gate Single Pulse mode is enabled and is controlling Timer1 gate
0 = Timer1 Gate Single Pulse mode is disabled
bit 3 T1GGO/T1DONE: Timer1 Gate Single Pulse Acquisition Status bit
1 = Timer1 gate single pulse acquisition is ready, waiting for an edge
0 = Timer1 gate single pulse acquisition has completed or has not been started
This bit is automatically cleared when T1GSPM is cleared.
bit 2 T1GVAL: Timer1 Gate Current State bit
Indicates the current state of the Timer1 gate that could be provided to TMR1H:TMR1L; unaffected by
the Timer1 Gate Enable (TMR1GE) bit.
bit 1-0 T1GSS<1:0>: Timer1 Gate Source Select bits
11 = Comparator 2 output
01 = TMR2 to match PR2 output
00 = Timer1 gate pin
Note 1: Programming the T1GCON register prior to T1CON is recommended.

PIC18F87K90 FAMILY
13.2 Timer1 Operation 13.3.2 EXTERNAL CLOCK SOURCE
The Timer1 module is an 8 or 16-bit incrementing When the external clock source is selected, the Timer1
counter that is accessed through the TMR1H:TMR1L module may work as a timer or a counter.
register pair. When enabled to count, Timer1 is incremented on the
When used with an internal clock source, the module is rising edge of the external clock input, T1CKI. Either of
a timer and increments on every instruction cycle. these external clock sources can be synchronized to the
When used with an external clock source, the module microcontroller system clock or they can run
can be used as either a timer or counter. It increments asynchronously.
on every selected edge of the external source. When used as a timer with a clock oscillator, an
Timer1 is enabled by configuring the TMR1ON and external, 32.768 kHz crystal can be used in conjunction
TMR1GE bits in the T1CON and T1GCON registers, with the dedicated internal oscillator circuit.
respectively. Note: In Counter mode, a falling edge must be
When SOSC is selected as a Crystal mode (by registered by the counter prior to the first
SOSCEL), the RC1/SOSCI/ECCP2/P2A/SEG32 and incrementing rising edge after any one or
RC0/SOSCO/SCLKI pins become inputs. This means more of the following conditions:
the values of TRISC<1:0> are ignored and the pins are • Timer1 is enabled after a POR Reset
read as ‘0’. • Write to TMR1H or TMR1L
• Timer1 is disabled
13.3 Clock Source Selection • Timer1 is disabled (TMR1ON = 0)
When T1CKI is high, Timer1 is enabled
The TMR1CS<1:0> and SOSCEN bits of the T1CON
(TMR1ON = 1) when T1CKI is low.
register are used to select the clock source for Timer1.
Register 13-1 displays the clock source selections.
13.3.1 INTERNAL CLOCK SOURCE

When the internal clock source is selected, the
TMR1H:TMR1L register pair will increment on multiples
of FOSC, as determined by the Timer1 prescaler.
TABLE 13-1: TIMER1 CLOCK SOURCE SELECTION

TMR1CS1 TMR1CS0 SOSCEN Clock Source
0 1 x Clock Source (FOSC)
0 0 x Instruction Clock (FOSC/4)
1 0 0 External Clock on T1CKI Pin
1 0 1 Oscillator Circuit on SOSCI/SOSCO Pins

PIC18F87K90 FAMILY
FIGURE 13-1: TIMER1 BLOCK DIAGRAM
T1GSS<1:0>
T1G 00 T1GSPM
From Timer2 01 T1G_IN 0

Match PR2 0 T1GVAL Data Bus
D Q
From Comp. 1 10 Single Pulse RD
Output 1 T1GCON
Acq. Control Q1 EN
D Q 1
From Comp. 2
11
Output Q T1GGO/T1DONE Interrupt Set
TMR1ON CK
det TMR1GIF
R
T1GPOL T1GTM
TMR1GE
Set Flag bit TMR1ON
TMR1IF on
Overflow TMR1(2)
EN Synchronized
0 Clock Input
TMR1H TMR1L T1CLK
Q D
1
TMR1CS<1:0>
T1SYNC
SOSCO/T1CKI OUT
T1OSC Prescaler Synchronize(3)

1 1, 2, 4, 8 det
SOSCI 10
EN
2
0 FOSC T1CKPS<1:0>
Internal 01
Clock
SOSCEN FOSC/2
Internal Sleep Input
FOSC/4
Internal 00 Clock
(1) Clock
T1CKI
Note 1: The ST buffer is a high-speed type when using T1CKI.

2: Timer1 register increments on the rising edge.
3: Synchronization does not operate while in Sleep.

PIC18F87K90 FAMILY
13.4 Timer1 16-Bit Read/Write Mode TABLE 13-2: CAPACITOR SELECTION FOR
THE TIMER
Timer1 can be configured for 16-bit reads and writes.
OSCILLATOR(2,3,4,5)
When the RD16 control bit (T1CON<1>) is set, the
address for TMR1H is mapped to a buffer register for Oscillator
Freq. C1 C2
the high byte of Timer1. A read from TMR1L loads the Type
contents of the high byte of Timer1 into the Timer1 High LP 32 kHz 12 pF(1) 12 pF(1)
Byte Buffer register. This provides the user with the
ability to accurately read all 16 bits of Timer1 without Note 1: Microchip suggests these values as a
having to determine whether a read of the high byte, starting point in validating the oscillator
followed by a read of the low byte, has become invalid circuit.
due to a rollover between reads.
2: Higher capacitance increases the stabil-
A write to the high byte of Timer1 must also take place ity of the oscillator, but also increases the
through the TMR1H Buffer register. The Timer1 high start-up time.
byte is updated with the contents of TMR1H when a
3: Since each resonator/crystal has its own
write occurs to TMR1L. This allows a user to write all
characteristics, the user should consult
16 bits at once to both the high and low bytes of Timer1.
the resonator/crystal manufacturer for
The high byte of Timer1 is not directly readable or appropriate values of external
writable in this mode. All reads and writes must take components.
place through the Timer1 High Byte Buffer register.
4: Capacitor values are for design
Writes to TMR1H do not clear the Timer1 prescaler; the
guidance only. Values listed would be
prescaler is only cleared on writes to TMR1L.
typical of a CL = 10 pF rated crystal
when SOSCSEL<1:0> = 11.
13.5 SOSC Oscillator
5: Incorrect capacitance value may result in
An on-chip crystal oscillator circuit is incorporated a frequency not meeting the crystal
between pins, SOSCI (input) and SOSCO (amplifier manufacturer’s tolerance specification.
output). It is enabled by setting one of five bits: any of the
four SOSCEN bits in the TxCON registers (TxCON<3>) The SOSC crystal oscillator drive level is determined
or the SOSCGO bit in the OSCCON2 register based on the SOSCSEL<1:0> (CONFIG1L<4:3>)
(OSCCON2<3>). The oscillator is a low-power circuit, Configuration bits. The High Drive Level mode,
rated for 32 kHz crystals. It will continue to run during all SOSCSEL<1:0> = 11, is intended to drive a wide
power-managed modes. The circuit for a typical LP variety of 32.768 kHz crystals with a variety of load
oscillator is depicted in Figure 13-2. Table 13-2 provides capacitance (CL) ratings.
the capacitor selection for the SOSC oscillator. The Low Drive Level mode is highly optimized for
The user must provide a software time delay to ensure extremely low-power consumption. It is not intended to
proper start-up of the SOSC oscillator. drive all types of 32.768 kHz crystals. In the Low Drive
Level mode, the crystal oscillator circuit may not work
correctly if excessively large discrete capacitors are
FIGURE 13-2: EXTERNAL COMPONENTS
placed on the SOSCO and SOSCI pins. This mode is
FOR THE SOSC designed to work only with discrete capacitances of
OSCILLATOR approximately 3 pF-10 pF on each pin.
C1
PIC18F87K90 Crystal manufacturers usually specify a CL (Capaci-
12 pF tance Load) rating for their crystals. This value is
SOSCI related to, but not necessarily the same as, the values
that should be used for C1 and C2 in Figure 13-2.
XTAL
32.768 kHz For more details on selecting the optimum C1 and C2
for a given crystal, see the crystal manufacture’s
SOSCO applications information. The optimum value depends,
C2 in part, on the amount of parasitic capacitance in the
12 pF circuit, which is often unknown. For that reason, it is
highly recommended that thorough testing and
Note: See the Notes with Table 13-2 for additional
validation of the oscillator be performed after values
information about capacitor selection.
have been selected.

PIC18F87K90 FAMILY
13.5.1 USING SOSC AS A FIGURE 13-3: OSCILLATOR CIRCUIT
CLOCK SOURCE WITH GROUNDED
The SOSC oscillator is also available as a clock source GUARD RING
in power-managed modes. By setting the System VDD
Clock Select bits, SCS<1:0> (OSCCON<1:0>), to ‘01’,
the device switches to SEC_RUN mode, and both the VSS
CPU and peripherals are clocked from the SOSC oscil-
lator. If the IDLEN bit (OSCCON<7>) is cleared and a OSC1
SLEEP instruction is executed, the device enters
OSC2
SEC_IDLE mode. Additional details are available in
Section 4.0 “Power-Managed Modes”.
Whenever the SOSC oscillator is providing the clock
source, the SOSC System Clock Status Flag, SOSC- RC0
RUN (OSCCON2<6>), is set. This can be used to
RC1
determine the controller’s current clocking mode. It can
also indicate the clock source currently being used by
the Fail-Safe Clock Monitor (FSCM). RC2
If the Clock Monitor is enabled and the SOSC oscillator
Note: Not drawn to scale.
fails while providing the clock, polling the SOCSRUN
bit will indicate whether the clock is being provided by
the SOSC oscillator or another source. In the Low Drive Level mode, SOSCSEL<1:0> = 01, it
is critical that RC2 I/O pin signals be kept away from the
13.5.2 SOSC OSCILLATOR LAYOUT oscillator circuit. Configuring RC2 as a digital output,
CONSIDERATIONS and toggling it, can potentially disturb the oscillator cir-
cuit, even with a relatively good PCB layout. If possible,
The SOSC oscillator circuit draws very little power either leave RC2 unused or use it as an input pin with
during operation. Due to the low-power nature of the a slew rate limited signal source. If RC2 must be used
oscillator, it may also be sensitive to rapidly changing as a digital output, it may be necessary to use the High
signals in close proximity. This is especially true when Drive Level Oscillator mode (SOSCSEL<1:0> = 11)
the oscillator is configured for extremely low-power with many PCB layouts.
mode (CONFIG1L<4:3> (SOSCSEL) = 01).
Even in the High Drive Level mode, careful layout
The oscillator circuit, displayed in Figure 13-2, should procedures should still be followed when designing the
be located as close as possible to the microcontroller. oscillator circuit.
There should be no circuits passing within the oscillator
circuit boundaries other than VSS or VDD. In addition to dV/dt induced noise considerations, it is
important to ensure that the circuit board is clean. Even
If a high-speed circuit must be located near the oscillator, a very small amount of conductive soldering flux
it may help to have a grounded guard ring around the residue can cause PCB leakage currents that can
oscillator circuit. The guard, as displayed in Figure 13-3, overwhelm the oscillator circuit.
could be used on a single-sided PCB or in addition to a
ground plane. (Examples of a high-speed circuit include
the ECCP1 pin, in Output Compare or PWM mode, or 13.6 Timer1 Interrupt
the primary oscillator using the OSC2 pin.)
The TMR1 register pair (TMR1H:TMR1L) increments
from 0000h to FFFFh and rolls over to 0000h. The
Timer1 interrupt, if enabled, is generated on overflow
which is latched in the Timer1 Overflow Interrupt Flag
bit, TMR1IF (PIR1<0>). This interrupt can be enabled
or disabled by setting or clearing the Timer1 Interrupt
Enable bit, TMR1IE (PIE1<0>).

PIC18F87K90 FAMILY
13.7 Resetting Timer1 Using the ECCP 13.8.1 TIMER1 GATE COUNT ENABLE
Special Event Trigger The Timer1 Gate Enable mode is enabled by setting
the TMR1GE bit of the T1GCON register. The polarity
If ECCP modules are configured to use Timer1 and to
of the Timer1 Gate Enable mode is configured using
generate a Special Event Trigger in Compare mode
the T1GPOL bit (T1GCON<6>).
(CCPxM<3:0> = 1011), this signal will reset Timer1. The
trigger from ECCP2 will also start an A/D conversion, if When Timer1 Gate Enable mode is enabled, Timer1
the A/D module is enabled. (For more information, see will increment on the rising edge of the Timer1 clock
Section 19.3.4 “Special Event Trigger”.) source. When Timer1 Gate Enable mode is disabled,
no incrementing will occur and Timer1 will hold the
To take advantage of this feature, the module must be
current count. See Figure 13-4 for timing details.
configured as either a timer or a synchronous counter.
When used this way, the CCPRxH:CCPRxL register
pair effectively becomes a Period register for Timer1. TABLE 13-3: TIMER1 GATE ENABLE
SELECTIONS
If Timer1 is running in Asynchronous Counter mode,
this Reset operation may not work. T1CLK(†)
T1GPOL
T1G Pin Timer1 Operation
(T1GCON<6>)
In the event that a write to Timer1 coincides with a
Special Event Trigger, the write operation will take  0 0 Counts
precedence.  0 1 Holds Count
 1 0 Holds Count
Note: The Special Event Trigger from the
ECCPx module will only clear the TMR1  1 1 Counts
register’s content, but not set the TMR1IF † The clock on which TMR1 is running. For more
interrupt flag bit (PIR1<0>). information, see Figure 13-1.
13.8 Timer1 Gate Note: The CCP and ECCP modules use Timers,
Timer1 can be configured to count freely or the count can 1 through 8, for some modes. The assign-
be enabled and disabled using the Timer1 gate circuitry. ment of a particular timer to a CCP/ECCP
This is also referred to as Timer1 gate count enable. module is determined by the Timer to CCP
enable bits in the CCPTMRSx registers.
Timer1 gate can also be driven by multiple selectable
For more details, see Register 18-2,
sources.
Register 18-3 and Register 19-2
FIGURE 13-4: TIMER1 GATE COUNT ENABLE MODE
TMR1GE
T1GPOL
T1G_IN
T1CKI
T1GVAL
Timer1 N N+1 N+2 N+3 N+4

PIC18F87K90 FAMILY
13.8.2 TIMER1 GATE SOURCE 13.8.2.2 Timer2 Match Gate Operation
SELECTION The TMR2 register will increment until it matches the
The Timer1 gate source can be selected from one of value in the PR2 register. On the very next increment
four sources. Source selection is controlled by the cycle, TMR2 will be reset to 00h. When this Reset
T1GSSx bits, T1GCON<1:0> (see Table 13-4). occurs, a low-to-high pulse will automatically be gener-
ated and internally supplied to the Timer1 gate circuitry.
TABLE 13-4: TIMER1 GATE SOURCES The pulse will remain high for one instruction cycle and
will return back to a low state until the next match.
T1GSS<1:0> Timer1 Gate Source
Depending on T1GPOL, Timer1 increments differently
00 Timer1 Gate Pin when TMR2 matches PR2. When T1GPOL = 1, Timer1
01 TMR2 to Match PR2 increments for a single instruction cycle following a
(TMR2 increments to match PR2) TMR2 match with PR2. When T1GPOL = 0, Timer1
increments continuously, except for the cycle following
10 Comparator 1 Output
the match, when the gate signal goes from low-to-high.
(Comparator logic high output)
11 Comparator 2 Output 13.8.2.3 Comparator 1 Output Gate Operation
(Comparator logic high output)
The output of Comparator 1 can be internally supplied
The polarity for each available source is also selectable, to the Timer1 gate circuitry. After setting up
controlled by the T1GPOL bit (T1GCON<6>). Comparator 1 with the CM1CON register, Timer1 will
increment depending on the transition of the
13.8.2.1 T1G Pin Gate Operation CMP1OUT (CMSTAT<5>) bit.
The T1G pin is one source for Timer1 gate control. It
can be used to supply an external source to the Timer1 13.8.2.4 Comparator 2 Output Gate Operation
gate circuitry. The output of Comparator 2 can be internally supplied
to the Timer1 gate circuitry. After setting up
Comparator 2 with the CM2CON register, Timer1 will
increment depending on the transition of the
CMP2OUT (CMSTAT<6>) bit.

PIC18F87K90 FAMILY
13.8.3 TIMER1 GATE TOGGLE MODE The T1GVAL bit (T1GCON<2>) indicates when the
Toggled mode is active and the timer is counting.
When Timer1 Gate Toggle mode is enabled, it is
possible to measure the full cycle length of a Timer1 The Timer1 Gate Toggle mode is enabled by setting the
gate signal, as opposed to the duration of a single level T1GTM bit (T1GCON<5>). When T1GTM is cleared,
pulse. the flip-flop is cleared and held clear. This is necessary
in order to control which edge is measured.
The Timer1 gate source is routed through a flip-flop that
changes state on every incrementing edge of the
signal. (For timing details, see Figure 13-5.)
FIGURE 13-5: TIMER1 GATE TOGGLE MODE
TMR1GE
T1GPOL
T1GTM
T1G_IN
T1CKI
T1GVAL
Timer1 N N+1 N+2 N+3 N+4 N+5 N+6 N+7 N+8

PIC18F87K90 FAMILY
13.8.4 TIMER1 GATE SINGLE PULSE Clearing the T1GSPM bit of the T1GCON register will
MODE also clear the T1GGO/T1DONE bit. (For timing details,
see Figure 13-6.)
When Timer1 Gate Single Pulse mode is enabled, it is
possible to capture a single pulse gate event. Timer1 Simultaneously enabling the Toggle and Single Pulse
Gate Single Pulse mode is enabled by setting the modes will permit both sections to work together. This
T1GSPM bit (T1GCON<4>) and the T1GGO/T1DONE allows the cycle times on the Timer1 gate source to be
bit (T1GCON<3>). The Timer1 will be fully enabled on measured. (For timing details, see Figure 13-7.)
the next incrementing edge.
13.8.5 TIMER1 GATE VALUE STATUS
On the next trailing edge of the pulse, the T1GGO/
T1DONE bit will automatically be cleared. No other When the Timer1 gate value status is utilized, it is
gate events will be allowed to increment Timer1 until possible to read the most current level of the gate
the T1GGO/T1DONE bit is once again set in software. control value. The value is stored in the T1GVAL bit
(T1GCON<2>). This bit is valid even when the Timer1
gate is not enabled (TMR1GE bit is cleared).
FIGURE 13-6: TIMER1 GATE SINGLE PULSE MODE
TMR1GE
T1GPOL
T1GSPM
Cleared by Hardware on
T1GGO/ Set by Software Falling Edge of T1GVAL
T1DONE
Counting Enabled on
Rising Edge of T1G
T1G_IN
T1CKI
T1GVAL
Timer1 N N+1 N+2
Cleared by
RTCCIF Cleared by Software Set by Hardware on Software
Falling Edge of T1GVAL

PIC18F87K90 FAMILY
FIGURE 13-7: TIMER1 GATE SINGLE PULSE AND TOGGLE COMBINED MODE
TMR1GE
T1GPOL
T1GSPM
T1GTM
T1GGO/ Set by Software Falling Edge of T1GVAL
T1DONE Counting Enabled on
Rising Edge of T1G
T1G_IN
T1CKI
T1GVAL
Timer1 N N+1 N+2 N+3 N+4
Set by Hardware on Cleared by

RTCCIF Cleared by Software Falling Edge of T1GVAL Software
TABLE 13-5: REGISTERS ASSOCIATED WITH TIMER1 AS A TIMER/COUNTER

Reset
on Page:
T1CON TMR1CS1 TMR1CS0 T1CKPS1 T1CKPS0 SOSCEN T1SYNC RD16 TMR1ON 76
T1GCON TMR1GE T1GPOL T1GTM T1GSPM T1GGO/ T1GVAL T1GSS1 T1GSS0 77
T1DONE
OSCCON2 — SOSCRUN — — SOSCGO — MFIOFS MFIOSEL 79
CCPTMRS0 C3TSEL1 C3TSEL0 C2TSEL2 C2TSEL1 C2TSEL0 C1TSEL2 C1TSEL1 C1TSEL0 81
CCPTMRS1 C7TSEL1 C7TSEL0 — C6TSEL0 — C5TSEL0 C4TSEL1 C4TSEL0 81
CCPTMRS2 — — — C10TSEL0 — C9TSEL0 C8TSEL1 C8TSEL0 81
Legend: Shaded cells are not used by the Timer1 module.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
14.0 TIMER2 MODULE The value of TMR2 is compared to that of the Period reg-
ister, PR2, on each clock cycle. When the two values
The Timer2 module incorporates the following features: match, the comparator generates a match signal as the
• 8-bit Timer and Period registers (TMR2 and PR2, timer output. This signal also resets the value of TMR2
respectively) to 00h on the next cycle and drives the output counter/
• Both registers are readable and writable postscaler. (See Section 14.2 “Timer2 Interrupt”.)
• Software programmable prescaler The TMR2 and PR2 registers are both directly readable
(1:1, 1:4 and 1:16) and writable. The TMR2 register is cleared on any
• Software programmable postscaler device Reset, while the PR2 register initializes at FFh.
(1:1 through 1:16) Both the prescaler and postscaler counters are cleared
• Interrupt on TMR2 to PR2 match on the following events:
• Optional use as the shift clock for the • A write to the TMR2 register
MSSP modules • A write to the T2CON register
This module is controlled through the T2CON register • Any device Reset (Power-on Reset (POR), MCLR
(Register 14-1) that enables or disables the timer, and Reset, Watchdog Timer Reset (WDTR) or
configures the prescaler and postscaler. Timer2 can be Brown-out Reset [BOR])
shut off by clearing control bit, TMR2ON (T2CON<2>),
TMR2 is not cleared when T2CON is written.
to minimize power consumption.
A simplified block diagram of the module is shown in Note: The CCP and ECCP modules use Timers,
Figure 14-1. 1 through 8, for some modes. The assign-
ment of a particular timer to a CCP/ECCP
module is determined by the Timer to CCP
14.1 Timer2 Operation enable bits in the CCPTMRSx registers.
In normal operation, TMR2 is incremented from 00h on For more details, see Register 18-2,
each clock (FOSC/4). A 4-bit counter/prescaler on the Register 18-3 and Register 19-2.
clock input gives the prescale options of direct input,
divide-by-4 or divide-by-16. These are selected by the
prescaler control bits, T2CKPS<1:0> (T2CON<1:0>).

— T2OUTPS3 T2OUTPS2 T2OUTPS1 T2OUTPS0 TMR2ON T2CKPS1 T2CKPS0
bit 7 bit 0
Legend:

bit 6-3 T2OUTPS<3:0>: Timer2 Output Postscale Select bits
0000 = 1:1 Postscale
•
•
•
bit 2 TMR2ON: Timer2 On bit
1 = Timer2 is on
0 = Timer2 is off
bit 1-0 T2CKPS<1:0>: Timer2 Clock Prescale Select bits
00 = Prescaler is 1
01 = Prescaler is 4
1x = Prescaler is 16

PIC18F87K90 FAMILY
14.2 Timer2 Interrupt 14.3 Timer2 Output
Timer2 can also generate an optional device interrupt. The unscaled output of TMR2 is available primarily to
The Timer2 output signal (TMR2 to PR2 match) pro- the ECCP modules, where it is used as a time base for
vides the input for the 4-bit output counter/postscaler. operations in PWM mode.
This counter generates the TMR2 match interrupt flag, Timer2 can optionally be used as the shift clock source
which is latched in TMR2IF (PIR1<1>). The interrupt is for the MSSP modules operating in SPI mode.
enabled by setting the TMR2 Match Interrupt Enable Additional information is provided in Section 21.0
bit, TMR2IE (PIE1<1>). “Master Synchronous Serial Port (MSSP) Module”.
A range of 16 postscaler options (from 1:1 through 1:16
inclusive) can be selected with the postscaler control
bits, T2OUTPS<3:0> (T2CON<6:3>).
FIGURE 14-1: TIMER2 BLOCK DIAGRAM
4 1:1 to 1:16
T2OUTPS<3:0> Set TMR2IF
Postscaler
2
T2CKPS<1:0> TMR2 Output
(to PWM or MSSPx)
TMR2/PR2
Reset Match
1:1, 1:4, 1:16
FOSC/4 TMR2 Comparator PR2
Prescaler
8 8
8
Internal Data Bus
TABLE 14-1: REGISTERS ASSOCIATED WITH TIMER2 AS A TIMER/COUNTER

Reset
on Page:
TMR2 Timer2 Register 76
T2CON — T2OUTPS3 T2OUTPS2 T2OUTPS1 T2OUTPS0 TMR2ON T2CKPS1 T2CKPS0 76
PR2 Timer2 Period Register 76
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by the Timer2 module.

PIC18F87K90 FAMILY
15.0 TIMER3/5/7 MODULES A simplified block diagram of the Timer3/5/7 module is
shown in Figure 15-1.
The Timer3/5/7 timer/counter modules incorporate
The Timer3/5/7 module is controlled through the
these features:
TxCON register (Register 15-1). It also selects the
• Software-selectable operation as a 16-bit timer or clock source options for the ECCP modules. (For more
counter information, see Section 19.1.1 “ECCP Module and
• Readable and writable 8-bit registers (TMRxH Timer Resources”.)
and TMRxL) The FOSC clock source should not be used with the
• Selectable clock source (internal or external) with ECCP capture/compare features. If the timer will be
device clock or SOSC oscillator internal options used with the capture or compare features, always
• Interrupt-on-overflow select one of the other timer clocking options.
• Module Reset on ECCP Special Event Trigger
Timer7 is unimplemented for devices with a program
memory of 32 Kbytes (PIC18FX5K90).
Note: Throughout this section, generic references
are used for register and bit names that are the
same, except for an ‘x’ variable that indicates
the item’s association with the Timer3, Timer5
or Timer7 module. For example, the control
register is named TxCON and refers to
T3CON, T5CON and T7CON.

PIC18F87K90 FAMILY
REGISTER 15-1: TxCON: TIMER3/5/7 CONTROL REGISTER
TMRxCS1 TMRxCS0 TxCKPS1 TxCKPS0 SOSCEN TxSYNC RD16 TMRxON
bit 7 bit 0
Legend:
bit 7-6 TMRxCS<1:0>: Timerx Clock Source Select bits

10 = The Timer1 clock source is either a pin or an oscillator depending on the SOSCEN bit.
SOSCEN = 0:
External clock is from the T1CKI pin (on the rising edge).
SOSCEN = 1:
Crystal oscillator is on the SOSCI/SOSCO pins.
01 = Timerx clock source is the system clock (FOSC)(1)
00 = Timerx clock source is the instruction clock (FOSC/4)
bit 5-4 TxCKPS<1:0>: Timerx Input Clock Prescale Select bits
bit 3 SOSCEN: SOSC Oscillator Enable bit
1 = SOSC is enabled for Timerx (based on SOSCSEL fuses)
0 = SOSC is disabled for Timerx
bit 2 TxSYNC: Timerx External Clock Input Synchronization Control bit
(Not usable if the device clock comes from Timer1/Timer3.)
When TMRxCS<1:0> = 10:
1 = Do not synchronize external clock input
0 = Synchronize external clock input
When TMRxCS<1:0> = 0x:
This bit is ignored; Timer3 uses the internal clock.
bit 1 RD16: 16-Bit Read/Write Mode Enable bit
1 = Enables register read/write of Timerx in one 16-bit operation
0 = Enables register read/write of Timerx in two 8-bit operations
bit 0 TMRxON: Timerx On bit
1 = Enables Timerx
0 = Stops Timerx
Note 1: The FOSC clock source should not be selected if the timer will be used with the ECCP capture/compare
features.

PIC18F87K90 FAMILY
15.1 Timer3/5/7 Gate Control Register
The Timer3/5/7 Gate Control register (TxGCON),
provided in Register 14-2, is used to control the Timerx
gate.
REGISTER 15-2: TxGCON: TIMER3/5/7 GATE CONTROL REGISTER(1)

R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R-x R/W-0 R/W-0
TMRxGE TxGPOL TxGTM TxGSPM TxGGO/TxDONE TxGVAL TxGSS1 TxGSS0
bit 7 bit 0
Legend:
bit 7 TMRxGE: Timerx Gate Enable bit

If TMRxON = 0:
This bit is ignored.
If TMRxON = 1:
1 = Timerx counting is controlled by the Timerx gate function
0 = Timerx counts regardless of the Timerx gate function
bit 6 TxGPOL: Timerx Gate Polarity bit
1 = Timerx gate is active-high (Timerx counts when the gate is high)
0 = Timerx gate is active-low (Timerx counts when the gate is low)
bit 5 TxGTM: Timerx Gate Toggle Mode bit
1 = Timerx Gate Toggle mode is enabled.
0 = Timerx Gate Toggle mode is disabled and toggle flip-flop is cleared
Timerx gate flip-flop toggles on every rising edge.
bit 4 TxGSPM: Timerx Gate Single Pulse Mode bit
1 = Timerx Gate Single Pulse mode is enabled and is controlling the Timerx gate
0 = Timerx Gate Single Pulse mode is disabled
bit 3 TxGGO/TxDONE: Timerx Gate Single Pulse Acquisition Status bit
1 = Timerx gate single pulse acquisition is ready, waiting for an edge
0 = Timerx gate single pulse acquisition has completed or has not been started
This bit is automatically cleared when TxGSPM is cleared.
bit 2 TxGVAL: Timerx Gate Current State bit
Indicates the current state of the Timerx gate that could be provided to TMRxH:TMRxL. Unaffected by the
Timerx Gate Enable (TMRxGE) bit.
bit 1-0 TxGSS<1:0>: Timerx Gate Source Select bits
01 = TMR(x + 1) to match PR(x + 1) output(2)
00 = Timer1 gate pin
Watchdog Timer oscillator is turned on if TMRxGE = 1, regardless of the state of TMRxON.
Note 1: Programming the TxGCON prior to TxCON is recommended.

2: Timer(x+1) will be Timer4/6/8 or Timerx Timer3/5/7, respectively.

PIC18F87K90 FAMILY
REGISTER 15-3: OSCCON2: OSCILLATOR CONTROL REGISTER 2
U-0 R-0 U-0 U-0 R/W-0 U-0 R-x R/W-0
— SOSCRUN — — SOSCGO — MFIOFS MFIOSEL
bit 7 bit 0
Legend:

bit 6 SOSCRUN: SOSC Run Status bit
1 = System clock comes from a secondary SOSC
0 = System clock comes from an oscillator other than SOSC
bit 3 SOSCGO: Oscillator Start Control bit
1 = Oscillator is running even if no other sources are requesting it
0 = Oscillator is shut off if no other sources are requesting it (When the SOSC is selected to run from
a digital clock input, rather than an external crystal, this bit has no effect.)
bit 1 MFIOFS: MF-INTOSC Frequency Stable bit
1 = MF-INTOSC is stable
0 = MF-INTOSC is not stable
bit 0 MFIOSEL: MF-INTOSC Select bit
1 = MF-INTOSC is used in place of HF-INTOSC frequencies of 500 kHz, 250 kHz and 31.25 kHz
0 = MF-INTOSC is not used

PIC18F87K90 FAMILY
15.2 Timer3/5/7 Operation The operating mode is determined by the clock select
bits, TMRxCSx (TxCON<7:6>). When the TMRxCSx bits
Timer3, Timer5 and Timer7 can operate in these are cleared (= 00), Timer3/5/7 increments on every inter-
modes: nal instruction cycle (FOSC/4). When TMRxCSx = 01, the
• Timer Timer3/5/7 clock source is the system clock (FOSC), and
• Synchronous Counter when it is ‘10’, Timer3/5/7 works as a counter from the
external clock on the TxCKI pin (on the rising edge after
• Asynchronous Counter
the first falling edge) or the SOSC oscillator.
• Timer with Gated Control
FIGURE 15-1: TIMER3/5/7 BLOCK DIAGRAM
TxGSS<1:0>
TxG 00 TxGSPM
From TMR(x + 1)
01 TxG_IN 0
Match PR(x + 1) Data Bus
0 TxGVAL
D Q
From Comp. 1 Single Pulse RD
Output 10 1
Acq. Control Q1 EN T3GCON
D Q 1
From Comp. 2 11
Output Q TxGGO/TxDONE Interrupt Set
TMRxON CK
det TMRxGIF
R
TxGPOL TxGTM
TMRxGE
Set Flag bit TMRxON
TMRxIF on
Overflow TMRx(2)
EN Synchronized
0 Clock Input
TMRxH TMRxL TxCLK
Q D
1
TMRxCS<1:0> TxSYNC
SOSCO OUT
TxOSC Prescaler Synchronize(3)

1 1, 2, 4, 8 det
SOSCI 10
EN
2
0
FOSC TxCKPS<1:0>
Internal 01
Clock FOSC/2
SOSCEN Sleep Input
Internal
FOSC/4 Clock
Internal 00
(1) Clock
TxCKI
Note 1: The ST buffer is high-speed type when using TxCKI.

2: Timerx registers increment on the rising edge.
3: Synchronization does not operate while in Sleep.

PIC18F87K90 FAMILY
15.3 Timer3/5/7 16-Bit Read/Write Mode 15.5 Timer3/5/7 Gates
Timer3/5/7 can be configured for 16-bit reads and Timer3/5/7 can be configured to count freely or the
writes (see Figure 15.3). When the RD16 control bit count can be enabled and disabled using the Timer3/
(TxCON<1>) is set, the address for TMRxH is mapped 5/7 gate circuitry. This is also referred to as the
to a buffer register for the high byte of Timer3/5/7. A Timer3/5/7 gate count enable.
read from TMRxL will load the contents of the high byte The Timer3/5/7 gate can also be driven by multiple
of Timer3/5/7 into the Timerx High Byte Buffer register. selectable sources.
This provides users with the ability to accurately read
all 16 bits of Timer3/5/7 without having to determine 15.5.1 TIMER3/5/7 GATE COUNT ENABLE
whether a read of the high byte, followed by a read of
the low byte, has become invalid due to a rollover The Timerx Gate Enable mode is enabled by setting
between reads. the TMRxGE bit (TxGCON<7>). The polarity of the
Timerx Gate Enable mode is configured using the
A write to the high byte of Timer3/5/7 must also take TxGPOL bit (TxGCON<6>).
place through the TMRxH Buffer register. The Timer3/
5/7 high byte is updated with the contents of TMRxH When Timerx Gate Enable mode is enabled, Timer3/5/7
when a write occurs to TMRxL. This allows users to will increment on the rising edge of the Timer3/5/7 clock
write all 16 bits to both the high and low bytes of source. When Timerx Gate Enable mode is disabled, no
Timer3/5/7 at once. incrementing will occur and Timer3/5/7 will hold the
current count. See Figure 15-2 for timing details.
The high byte of Timer3/5/7 is not directly readable or
writable in this mode. All reads and writes must take
TABLE 15-1: TIMER3/5/7 GATE ENABLE
place through the Timerx High Byte Buffer register.
SELECTIONS
Writes to TMRxH do not clear the Timer3/5/7 prescaler.
TxGPOL
The prescaler is only cleared on writes to TMRxL. TxCLK(†) TxG Pin Timerx Operation
(TxGCON<6>)
15.4 Using the SOSC Oscillator as the  0 0 Counts
Timer3/5/7 Clock Source  0 1 Holds Count

 1 0 Holds Count
The SOSC internal oscillator may be used as the clock
1 1 Counts
source for Timer3/5/7. The SOSC oscillator is enabled by
setting one of five bits: any of the four SOSCEN bits in the † The clock on which TMR3/5/7 is running. For more
information, see TxCLK in Figure 15-1.
TxCON registers (TxCON<3>) or the SOSCGO bit in the
OSCCON2 register (OSCCON2<3>). To use it as the
Timer3/5/7 clock source, the TMRxCS bit must also be
set. As previously noted, this also configures Timer3/5/7
to increment on every rising edge of the oscillator source.
The SOSC oscillator is described in Section 13.0
“Timer1 Module”.
FIGURE 15-2: TIMER3/5/7 GATE COUNT ENABLE MODE
TMRxGE
TxGPOL
TxG_IN
TxCKI
TxGVAL
Timer3/5/7 N N+1 N+2 N+3 N+4

PIC18F87K90 FAMILY
15.5.2 TIMER3/5/7 GATE SOURCE Depending on TxGPOL, Timerx increments differently
SELECTION when TMR(x + 1) matches PR(x + 1). When
TxGPOL = 1, Timerx increments for a single instruction
The Timer3/5/7 gate source can be selected from one
cycle following a TMR(x + 1) match with PR(x + 1).
of four different sources. Source selection is controlled
When TxGPOL = 0, Timerx increments continuously
by the TxGSS<1:0> bits (TxGCON<1:0>). The polarity
except for the cycle following the match when the gate
for each available source is also selectable and is
signal goes from low-to-high.
controlled by the TxGPOL bit (TxGCON <6>).
15.5.2.3 Comparator 1 Output Gate Operation
TABLE 15-2: TIMER3/5/7 GATE SOURCES
The output of Comparator 1 can be internally supplied
TxGSS<1:0> Timerx Gate Source to the Timerx gate circuitry. After setting up
Comparator 1 with the CM1CON register, Timerx will
00 Timerx Gate Pin
increment depending on the transitions of the
TMR(x + 1) to Match PR(x + 1) CMP1OUT (CMSTAT<5>) bit.
01 (TMR(x + 1) increments to match
PR(x + 1) 15.5.2.4 Comparator 2 Output Gate Operation
Comparator 1 Output The output of Comparator 2 can be internally supplied
10
(Comparator logic high output) to the Timerx gate circuitry. After setting up
Comparator 2 Output Comparator 2 with the CM2CON register, Timerx will
11
(Comparator logic high output) increment depending on the transitions of the
CMP2OUT (CMSTAT<6>) bit.
15.5.2.1 TxG Pin Gate Operation
15.5.3 TIMER3/5/7 GATE TOGGLE MODE
The TxG pin is one source for Timer3/5/7 gate control.
It can be used to supply an external source to the When Timer3/5/7 Gate Toggle mode is enabled, it is
Timerx gate circuitry. possible to measure the full cycle length of a Timer3/5/
7 gate signal, as opposed to the duration of a single
15.5.2.2 Timer4/6/8 Match Gate Operation level pulse.
The Timer4/6/8 register will increment until it matches The Timerx gate source is routed through a flip-flop that
the value in the PRx register. On the very next increment changes state on every incrementing edge of the
cycle, TMRx will be reset to 00h. When this Reset signal. (For timing details, see Figure 15-3.)
occurs, a low-to-high pulse will automatically be gener- The TxGVAL bit will indicate when the Toggled mode is
ated and internally supplied to the Timerx gate circuitry. active and the timer is counting.
The pulse will remain high for one instruction cycle and
will return back to a low state until the next match. Timer3/5/7 Gate Toggle mode is enabled by setting the
TxGTM bit (TxGCON<5>). When the TxGTM bit is
cleared, the flip-flop is cleared and held clear. This is
necessary in order to control which edge is measured.
FIGURE 15-3: TIMER3/5/7 GATE TOGGLE MODE
TMRxGE
TxGPOL
TxGTM
TxG_IN
TxCKI
TxGVAL
Timer3/5/7 N N+1 N+2 N+3 N+4 N+5 N+6 N+7 N+8

PIC18F87K90 FAMILY
15.5.4 TIMER3/5/7 GATE SINGLE PULSE No other gate events will be allowed to increment
MODE Timer3/5/7 until the TxGGO/TxDONE bit is once again
set in software.
When Timer3/5/7 Gate Single Pulse mode is enabled,
it is possible to capture a single pulse gate event. Clearing the TxGSPM bit also will clear the TxGGO/
Timer3/5/7 Gate Single Pulse mode is first enabled by TxDONE bit. (For timing details, see Figure 15-4.)
setting the TxGSPM bit (TxGCON<4>). Next, the Simultaneously enabling the Toggle mode and the
TxGGO/TxDONE bit (TxGCON<3>) must be set. Single Pulse mode will permit both sections to work
The Timer3/5/7 will be fully enabled on the next incre- together. This allows the cycle times on the Timer3/5/7
menting edge. On the next trailing edge of the pulse, gate source to be measured. (For timing details, see
the TxGGO/TxDONE bit will automatically be cleared. Figure 15-5.)
FIGURE 15-4: TIMER3/5/7 GATE SINGLE PULSE MODE
TMRxGE
TxGPOL
TxGSPM
TxGGO/ Set by Software Falling Edge of TxGVAL
TxDONE
Counting Enabled on
Rising Edge of TxG
TxG_IN
T1CKI
TxGVAL
Timer3/5/7 N N+1 N+2
Cleared by
Cleared by Software Set by Hardware on Software
TMRxGIF
Falling Edge of TxGVAL

PIC18F87K90 FAMILY
FIGURE 15-5: TIMER3/5/7 GATE SINGLE PULSE AND TOGGLE COMBINED MODE
TMRxGE
TxGPOL
TxGSPM
TxGTM
TxGGO/ Set by Software Falling Edge of TxGVAL
TxDONE
Counting Enabled on
Rising Edge of TxG
TxG_IN
TxCKI
TxGVAL
Timer3/5/7 N N+1 N+2 N+3 N+4
Set by Hardware on Cleared by

Cleared by Software Falling Edge of TxGVAL Software
TMRxGIF
15.5.5 TIMER3/5/7 GATE VALUE STATUS 15.5.6 TIMER3/5/7 GATE EVENT

When Timer3/5/7 gate value status is utilized, it is INTERRUPT
possible to read the most current level of the gate con- When the Timer3/5/7 gate event interrupt is enabled, it
trol value. The value is stored in the TxGVAL bit is possible to generate an interrupt upon the comple-
(TxGCON<2>). The TxGVAL bit is valid even when the tion of a gate event. When the falling edge of TxGVAL
Timer3/5/7 gate is not enabled (TMRxGE bit is occurs, the TMRxGIF flag bit in the PIRx register will be
cleared). set. If the TMRxGIE bit in the PIEx register is set, then
an interrupt will be recognized.
The TMRxGIF flag bit operates even when the Timer3/
5/7 gate is not enabled (TMRxGE bit is cleared).

PIC18F87K90 FAMILY
15.6 Timer3/5/7 Interrupt 15.7 Resetting Timer3/5/7 Using the
The TMRx register pair (TMRxH:TMRxL) increments
ECCP Special Event Trigger
from 0000h to FFFFh and overflows to 0000h. The If the ECCP modules are configured to use Timerx and
Timerx interrupt, if enabled, is generated on overflow to generate a Special Event Trigger in Compare mode
and is latched in the interrupt flag bit, TMRxIF. (CCPxM<3:0> = 1011), this signal will reset Timerx. The
Table 15-3 gives each module’s flag bit. trigger from ECCP2 will also start an A/D conversion if
TABLE 15-3: TIMER3/5/7 INTERRUPT the A/D module is enabled. (For more information, see
FLAG BITS Section 19.3.4 “Special Event Trigger”.)
The module must be configured as either a timer or
Timer Module Flag Bit
synchronous counter to take advantage of this feature.
3 PIR2<1> When used this way, the CCPRxH:CCPRxL register
5 PIR5<1> pair effectively becomes a Period register for Timerx.
7 PIR5<3> If Timerx is running in Asynchronous Counter mode,
the Reset operation may not work.
This interrupt can be enabled or disabled by setting or
In the event that a write to Timerx coincides with a
clearing the TMRxIE bit, respectively. Table 15-4 gives
Special Event Trigger from an ECCP module, the write
each module’s enable bit.
will take precedence.
TABLE 15-4: TIMER3/5/7 INTERRUPT
Note: The Special Event Triggers from the
ENABLE BITS
ECCPx module will only clear the TMR3
Timer Module Flag Bit register’s content, but not set the TMR3IF
interrupt flag bit (PIR1<0>).
3 PIE2<1>
5 PIE5<1>
7 PIE5<3> Note: The CCP and ECCP modules use Timers,
1 through 8, for some modes. The assign-
Register 18-2 and Register 18-3.

PIC18F87K90 FAMILY
TABLE 15-5: REGISTERS ASSOCIATED WITH TIMER3/5/7 AS A TIMER/COUNTER
Reset
on Page:
PIR5 TMR7GIF(1) TMR12IF(1) TMR10IF(1) TMR8IF TMR7IF(1) TMR6IF TMR5IF TMR4IF 77
PIE5 TMR7GIE(1) TMR12IE(1) TMR10IE(1) TMR8IE TMR7IE(1) TMR6IE TMR5IE TMR4IE 77
PIR3 TMR5GIF(1) LCDIF RC2IF TX2IF CTMUIF CCP2IF CCP1IF RTCCIF 77
PIE3 TMR5GIE(1) LCDIE RC2IE TX2IE CTMUIE CCP2IE CCP1IE RTCCIE 77
T3DONE
T5DONE
(1)
TMR7L(1) Timer7 Register Low Byte 81
(1)
T7DONE
T7CON(1) TMR7CS1 TMR7CS0 T7CKPS1 T7CKPS0 SOSCEN T7SYNC RD16 TMR7ON 81
OSCCON2 — SOSCRUN — — SOSCGO — MFIOFS MFIOSEL 79
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by the Timer3/5/7 modules.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
16.0 TIMER4/6/8/10/12 MODULES 16.1 Timer4/6/8/10/12 Operation
The Timer4/6/8/10/12 timer modules have the following Timer4/6/8/10/12 can be used as the PWM time base
features: for the PWM mode of the ECCP modules. The TMRx
registers are readable and writable, and are cleared on
• 8-Bit Timer register (TMRx)
any device Reset. The input clock (FOSC/4) has a
• 8-Bit Period register (PRx) prescale option of 1:1, 1:4 or 1:16, selected by control
• Readable and writable (all registers) bits, TxCKPS<1:0> (TxCON<1:0>). The match output
• Software programmable prescaler (1:1, 1:4, 1:16) of TMRx goes through a 4-bit postscaler (that gives a
• Software programmable postscaler (1:1 to 1:16) 1:1 to 1:16 inclusive scaling) to generate a TMRx
• Interrupt on TMRx match of PRx interrupt, latched in the flag bit, TMRxIF. Table 16-1
shows each module’s flag bit.
Timer10 and Timer12 are unimplemented for devices
TABLE 16-1: TIMER4/6/8/10/12 FLAG BITS
with a program memory of 32 Kbytes (PIC18FX5K90).
Timer Flag Bit Timer Flag Bit
Note: Throughout this section, generic references
Module PIR5<x> Module PIR5<x>
are used for register and bit names that are the
same, except for an ‘x’ variable that indicates 4 0 10 5
the item’s association with the Timer4, Timer6, 6 2 12 6
Timer8, Timer10 or Timer12 module. For
8 4
example, the control register is named TxCON
and refers to T4CON, T6CON, T8CON,
T10CON and T12CON. The interrupt can be enabled or disabled by setting or
clearing the Timerx Interrupt Enable bit (TMRxIE),
The Timer4/6/8/10/12 modules have a control register, shown in Table 16-2.
which is shown in Register 16-1. Timer4/6/8/10/12 can
TABLE 16-2: TIMER4/6/8/10/12 INTERRUPT
be shut off by clearing control bit, TMRxON (TxCON<2>),
ENABLE BITS
to minimize power consumption. The prescaler and post-
scaler selection of Timer4/6/8/10/12 are also controlled Timer Flag Bit Timer Flag Bit
by this register. Figure 16-1 is a simplified block diagram Module PIE5<x> Module PIE5<x>
of the Timer4/6/8/10/12 modules.
4 0 10 5
6 2 12 6
8 4
The prescaler and postscaler counters are cleared
when any of the following occurs:
• A write to the TMRx register
• A write to the TxCON register
• Any device Reset (Power-on Reset (POR), MCLR
Reset, Watchdog Timer Reset (WDTR) or
Brown-out Reset (BOR))
A TMRx is not cleared when a TxCON is written.
Note: The CCP and ECCP modules use Timers,
1 through 8, for some modes. The assign-
Register 18-2 and Register 18-3.

PIC18F87K90 FAMILY
REGISTER 16-1: TxCON: TIMER4/6/8/10/12 CONTROL REGISTER
— TxOUTPS3 TxOUTPS2 TxOUTPS1 TxOUTPS0 TMRxON TxCKPS1 TxCKPS0
bit 7 bit 0
Legend:

bit 6-3 TxOUTPS<3:0>: Timerx Output Postscale Select bits
•
•
•
bit 2 TMRxON: Timerx On bit
1 = Timerx is on
0 = Timerx is off
bit 1-0 TxCKPS<1:0>: Timerx Clock Prescale Select bits
00 = Prescaler is 1
01 = Prescaler is 4
1x = Prescaler is 16

PIC18F87K90 FAMILY
16.2 Timer4/6/8/10/12 Interrupt 16.3 Output of TMRx
The Timer4/6/8/10/12 modules have 8-bit Period The outputs of TMRx (before the postscaler) are used
registers, PRx, that are both readable and writable. only as a PWM time base for the ECCP modules. They
Timer4/6/8/10/12 increment from 00h until they match are not used as baud rate clocks for the MSSP
PR4/6/8/10/12 and then reset to 00h on the next modules as is the Timer2 output.
increment cycle. The PRx registers are initialized to
FFh upon Reset.
FIGURE 16-1: TIMER4/6/8/10/12 BLOCK DIAGRAM
4 1:1 to 1:16
TxOUTPS<3:0> Set TMRxIF
Postscaler
2
TxCKPS<1:0> TMRx Output
(to PWM)
TMRx/PRx
Reset Match
FOSC/4 1:1, 1:4, 1:16 Comparator
TMRx PRx
Prescaler
8 8
8
Internal Data Bus
TABLE 16-3: REGISTERS ASSOCIATED WITH TIMER4/6/8/10/12 AS A TIMER/COUNTER

Reset
on Page:
IPR5 TMR7GIP(1) TMR12IP(1) TMR10IP(1) TMR8IP TMR7IP(1) TMR6IP TMR5IP TMR4IP 76
PIR5 TMR7GIF(1) TMR12IF(1) TMR10IF(1) TMR8IF TMR7IF(1) TMR6IF TMR5IF TMR4IF 77
(1) (1) (1) (1)
PIE5 TMR7GIE TMR12IE TMR10IE TMR8IE TMR7IE TMR6IE TMR5IE TMR4IE 77
CCPTMRS2 — — — C10TSEL0(1) — C9TSEL0(1) C8TSEL1 C8TSEL0 81
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by the Timer4/6/8/10/12 module.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
17.0 REAL-TIME CLOCK AND The RTCC module is intended for applications where
accurate time must be maintained for an extended
CALENDAR (RTCC) period with minimum to no intervention from the CPU.
The key features of the Real-Time Clock and Calendar The module is optimized for low-power usage in order
(RTCC) module are: to provide extended battery life while keeping track of
time.
• Time: hours, minutes and seconds
• Twenty-four hour format (military time) The module is a 100-year clock and calendar with auto-
matic leap year detection. The range of the clock is
• Calendar: weekday, date, month and year
from 00:00:00 (midnight) on January 1, 2000 to
• Alarm configurable 23:59:59 on December 31, 2099.
• Year range: 2000 to 2099
Hours are measured in 24-hour (military time) format.
• Leap year correction The clock provides a granularity of one second with
• BCD format for compact firmware half-second visibility to the user.
• Optimized for low-power operation
• User calibration with auto-adjust
• Calibration range: 2.64 seconds error per month
• Requirements: external 32.768 kHz clock crystal
• Alarm pulse or seconds clock output on RTCC pin
FIGURE 17-1: RTCC BLOCK DIAGRAM
RTCC Clock Domain CPU Clock Domain
32.768 kHz Input

from SOSC Oscillator RTCCFG
RTCC Prescalers
Internal RC ALRMRPT
(LF-INTOSC) YEAR
0.5s
MTHDY
RTCC Timer RTCVALx
WKDYHR
Alarm
Event MINSEC
Comparator
ALMTHDY
Compare Registers
ALRMVALx ALWDHR
with Masks
ALMINSEC
Repeat Counter
RTCC Interrupt
RTCC Interrupt Logic
Alarm Pulse
RTCC Pin
RTCOE

PIC18F87K90 FAMILY
17.1 RTCC MODULE REGISTERS Alarm Value Registers
The RTCC module registers are divided into the • ALRMVALH
following categories: • ALRMVALL
Both registers access the following registers:
RTCC Control Registers - ALRMMNTH
• RTCCFG - ALRMDAY
• RTCCAL - ALRMWD
• PADCFG1 - ALRMHR
• ALRMCFG - ALRMMIN
• ALRMRPT - ALRMSEC
Note: The RTCVALH and RTCVALL registers
RTCC Value Registers can be accessed through RTCRPT<1:0>
(RTCCFG<1:0>). ALRMVALH and
• RTCVALH ALRMVALL can be accessed through
• RTCVALL ALRMPTR<1:0> (ALRMCFG<1:0>).
Both registers access the following registers:
- YEAR
- MONTH
- DAY
- WEEKDAY
- HOUR
- MINUTE
- SECOND

PIC18F87K90 FAMILY
17.1.1 RTCC CONTROL REGISTERS
REGISTER 17-1: RTCCFG: RTCC CONFIGURATION REGISTER(1)
R/W-0 U-0 R/W-0 R-0 R-0 R/W-0 R/W-0 R/W-0

(2) (4) (3)
RTCEN — RTCWREN RTCSYNC HALFSEC RTCOE RTCPTR1 RTCPTR0
bit 7 bit 0
Legend:
bit 7 RTCEN: RTCC Enable bit(2)

1 = RTCC module is enabled
0 = RTCC module is disabled
bit 5 RTCWREN: RTCC Value Registers Write Enable bit(4)
1 = RTCVALH and RTCVALL registers can be written to by the user
0 = RTCVALH and RTCVALL registers are locked out from being written to by the user
bit 4 RTCSYNC: RTCC Value Registers Read Synchronization bit
1 = RTCVALH, RTCVALL and ALRMRPT registers can change while reading if a rollover ripple results
in an invalid data read. If the register is read twice and results in the same data, the data can be
assumed to be valid.
0 = RTCVALH, RTCVALL and ALCFGRPT registers can be read without concern over a rollover ripple
bit 3 HALFSEC: Half-Second Status bit(3)
1 = Second half period of a second
0 = First half period of a second
bit 2 RTCOE: RTCC Output Enable bit
1 = RTCC clock output is enabled
0 = RTCC clock output is disabled
bit 1-0 RTCPTR<1:0>: RTCC Value Register Window Pointer bits
Points to the corresponding RTCC Value registers when reading the RTCVALH and RTCVALL registers.
The RTCPTR<1:0> value decrements on every read or write of RTCVALH<15:8> until it reaches ‘00’.
RTCVALH:
00 = Minutes
01 = Weekday
10 = Month
11 = Reserved
RTCVALL:
00 = Seconds
01 = Hours
10 = Day
11 = Year
Note 1: The RTCCFG register is only affected by a POR.

2: A write to the RTCEN bit is only allowed when RTCWREN = 1.
3: This bit is read-only; it is cleared to ‘0’ on a write to the lower half of the MINSEC register.
4: The RTCWREN bit can only be written with the unlock sequence (see Example 17-1).

PIC18F87K90 FAMILY
REGISTER 17-2: RTCCAL: RTCC CALIBRATION REGISTER

CAL7 CAL6 CAL5 CAL4 CAL3 CAL2 CAL1 CAL0
bit 7 bit 0
Legend:
bit 7-0 CAL<7:0>: RTC Drift Calibration bits

01111111 = Maximum positive adjustment. Adds 508 RTC clock pulses every minute.
.
.
.
00000001 = Minimum positive adjustment. Adds four RTC clock pulses every minute.
00000000 = No adjustment
11111111 = Minimum negative adjustment. Subtracts four RTC clock pulses every minute.
.
.
.
10000000 = Maximum negative adjustment. Subtracts 512 RTC clock pulses every minute.
REGISTER 17-3: PADCFG1: PAD CONFIGURATION REGISTER

R/W-0 R/W-0 R/W-0 U-0 U-0 R/W-0 R/W-0 U-0
RDPU REPU RJPU(2) — — RTSECSEL1(1) RTSECSEL0(1) —
bit 7 bit 0
Legend:
bit 7 RDPU: PORTD Pull-up Enable bit

1 = PORTD pull-up resistors are enabled by individual port latch values
0 = All PORTD pull-up resistors are disabled
bit 6 REPU: PORTE Pull-up Enable bit
1 = PORTE pull-up resistors are enabled by individual port latch values
0 = All PORTE pull-up resistors are disabled
bit 5 RJPU: PORTJ Pull-up Enable bit(2)
1 = PORTJ pull-up resistors are enabled by individual port latch values
0 = All PORTJ pull-up resistors are disabled
bit 2-1 RTSECSEL<1:0>: RTCC Seconds Clock Output Select bits(1)
11 = Reserved; do not use
10 = RTCC source clock is selected for the RTCC pin (the pin can be LF-INTOSC or SOSC, depending
on the RTCOSC (CONFIG3L<1>) bit setting)
01 = RTCC seconds clock is selected for the RTCC pin
00 = RTCC alarm pulse is selected for the RTCC pin
Note 1: To enable the actual RTCC output, the RTCOE (RTCCFG<2>) bit must be set.
2: Available only in 80-pin parts.

PIC18F87K90 FAMILY
REGISTER 17-4: ALRMCFG: ALARM CONFIGURATION REGISTER
ALRMEN CHIME AMASK3 AMASK2 AMASK1 AMASK0 ALRMPTR1 ALRMPTR0
bit 7 bit 0
Legend:
bit 7 ALRMEN: Alarm Enable bit

1 = Alarm is enabled (cleared automatically after an alarm event whenever ALRMPTR<1:0> = 00
and CHIME = 0)
0 = Alarm is disabled
bit 6 CHIME: Chime Enable bit
1 = Chime is enabled; ALRMPTR<1:0> bits are allowed to roll over from 00h to FFh
0 = Chime is disabled; ALRMPTR<1:0> bits stop once they reach 00h
bit 5-2 AMASK<3:0>: Alarm Mask Configuration bits
0000 = Every half second
0001 = Every second
0010 = Every 10 seconds
0011 = Every minute
0100 = Every 10 minutes
0101 = Every hour
0110 = Once a day
0111 = Once a week
1000 = Once a month
1001 = Once a year (except when configured for February 29th, once every four years)
101x = Reserved – Do not use
11xx = Reserved – Do not use
bit 1-0 ALRMPTR<1:0>: Alarm Value Register Window Pointer bits
Points to the corresponding Alarm Value registers when reading the ALRMVALH and ALRMVALL
registers. The ALRMPTR<1:0> value decrements on every read or write of ALRMVALH until it reaches
‘00’.
ALRMVALH:
00 = ALRMMIN
01 = ALRMWD
10 = ALRMMNTH
11 = Unimplemented
ALRMVALL:
00 = ALRMSEC
01 = ALRMHR
10 = ALRMDAY
11 = Unimplemented

PIC18F87K90 FAMILY
REGISTER 17-5: ALRMRPT: ALARM REPEAT REGISTER
ARPT7 ARPT6 ARPT5 ARPT4 ARPT3 ARPT2 ARPT1 ARPT0
bit 7 bit 0
Legend:
bit 7-0 ARPT<7:0>: Alarm Repeat Counter Value bits

11111111 = Alarm will repeat 255 more times
.
.
.
00000000 = Alarm will not repeat
The counter decrements on any alarm event. The counter is prevented from rolling over from 00h to
FFh unless CHIME = 1.
17.1.2 RTCVALH AND RTCVALL

REGISTER MAPPINGS
REGISTER 17-6: RESERVED REGISTER

U-0 U-0 U-0 U-0 U-0 U-0 U-0 U-0
— — — — — — — —
bit 7 bit 0
Legend:

PIC18F87K90 FAMILY
REGISTER 17-7: YEAR: YEAR VALUE REGISTER(1)
R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x
YRTEN3 YRTEN2 YRTEN1 YRTEN0 YRONE3 YRONE2 YRONE1 YRONE0
bit 7 bit 0
Legend:
bit 7-4 YRTEN<3:0>: Binary Coded Decimal Value of Year’s Tens Digit bits
Contains a value from 0 to 9.
bit 3-0 YRONE<3:0>: Binary Coded Decimal Value of Year’s Ones Digit bits
Note 1: A write to the YEAR register is only allowed when RTCWREN = 1.
REGISTER 17-8: MONTH: MONTH VALUE REGISTER(1)

— — — MTHTEN0 MTHONE3 MTHONE2 MTHONE1 MTHONE0
bit 7 bit 0
Legend:

bit 4 MTHTEN0: Binary Coded Decimal Value of Month’s Tens Digit bits
Contains a value of ‘0’ or ‘1’.
bit 3-0 MTHONE<3:0>: Binary Coded Decimal Value of Month’s Ones Digit bits
Note 1: A write to this register is only allowed when RTCWREN = 1.

PIC18F87K90 FAMILY
REGISTER 17-9: DAY: DAY VALUE REGISTER(1)
U-0 U-0 R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x
— — DAYTEN1 DAYTEN0 DAYONE3 DAYONE2 DAYONE1 DAYONE0
bit 7 bit 0
Legend:

bit 5-4 DAYTEN<1:0>: Binary Coded Decimal value of Day’s Tens Digit bits
bit 3-0 DAYONE<3:0>: Binary Coded Decimal Value of Day’s Ones Digit bits
REGISTER 17-10: WEEKDAY: WEEKDAY VALUE REGISTER(1)

U-0 U-0 U-0 U-0 U-0 R/W-x R/W-x R/W-x
— — — — — WDAY2 WDAY1 WDAY0
bit 7 bit 0
Legend:

bit 2-0 WDAY<2:0>: Binary Coded Decimal Value of Weekday Digit bits

PIC18F87K90 FAMILY
REGISTER 17-11: HOUR: HOUR VALUE REGISTER(1)
— — HRTEN1 HRTEN0 HRONE3 HRONE2 HRONE1 HRONE0
bit 7 bit 0
Legend:

bit 5-4 HRTEN<1:0>: Binary Coded Decimal Value of Hour’s Tens Digit bits
bit 3-0 HRONE<3:0>: Binary Coded Decimal Value of Hour’s Ones Digit bits
REGISTER 17-12: MINUTE: MINUTE VALUE REGISTER

U-0 R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x R/W-x
— MINTEN2 MINTEN1 MINTEN0 MINONE3 MINONE2 MINONE1 MINONE0
bit 7 bit 0
Legend:

bit 6-4 MINTEN<2:0>: Binary Coded Decimal Value of Minute’s Tens Digit bits
bit 3-0 MINONE<3:0>: Binary Coded Decimal Value of Minute’s Ones Digit bits
REGISTER 17-13: SECOND: SECOND VALUE REGISTER

— SECTEN2 SECTEN1 SECTEN0 SECONE3 SECONE2 SECONE1 SECONE0
bit 7 bit 0
Legend:

bit 6-4 SECTEN<2:0>: Binary Coded Decimal Value of Second’s Tens Digit bits
bit 3-0 SECONE<3:0>: Binary Coded Decimal Value of Second’s Ones Digit bits

PIC18F87K90 FAMILY
17.1.3 ALRMVALH AND ALRMVALL
REGISTER MAPPINGS
REGISTER 17-14: ALRMMNTH: ALARM MONTH VALUE REGISTER(1)

— — — MTHTEN0 MTHONE3 MTHONE2 MTHONE1 MTHONE0
bit 7 bit 0
Legend:

bit 4 MTHTEN0: Binary Coded Decimal Value of Month’s Tens Digit bits
Contains a value of ‘0’ or ‘1’.
bit 3-0 MTHONE<3:0>: Binary Coded Decimal Value of Month’s Ones Digit bits
REGISTER 17-15: ALRMDAY: ALARM DAY VALUE REGISTER(1)

— — DAYTEN1 DAYTEN0 DAYONE3 DAYONE2 DAYONE1 DAYONE0
bit 7 bit 0
Legend:

bit 5-4 DAYTEN<1:0>: Binary Coded Decimal Value of Day’s Tens Digit bits
bit 3-0 DAYONE<3:0>: Binary Coded Decimal Value of Day’s Ones Digit bits

PIC18F87K90 FAMILY
REGISTER 17-16: ALRMWD: ALARM WEEKDAY VALUE REGISTER(1)
U-0 U-0 U-0 U-0 U-0 R/W-x R/W-x R/W-x
— — — — — WDAY2 WDAY1 WDAY0
bit 7 bit 0
Legend:

bit 2-0 WDAY<2:0>: Binary Coded Decimal Value of Weekday Digit bits
REGISTER 17-17: ALRMHR: ALARM HOURS VALUE REGISTER(1)

— — HRTEN1 HRTEN0 HRONE3 HRONE2 HRONE1 HRONE0
bit 7 bit 0
Legend:

bit 5-4 HRTEN<1:0>: Binary Coded Decimal Value of Hour’s Tens Digit bits
bit 3-0 HRONE<3:0>: Binary Coded Decimal Value of Hour’s Ones Digit bits

PIC18F87K90 FAMILY
REGISTER 17-18: ALRMMIN: ALARM MINUTES VALUE REGISTER
— MINTEN2 MINTEN1 MINTEN0 MINONE3 MINONE2 MINONE1 MINONE0
bit 7 bit 0
Legend:

bit 6-4 MINTEN<2:0>: Binary Coded Decimal Value of Minute’s Tens Digit bits
bit 3-0 MINONE<3:0>: Binary Coded Decimal Value of Minute’s Ones Digit bits
REGISTER 17-19: ALRMSEC: ALARM SECONDS VALUE REGISTER

— SECTEN2 SECTEN1 SECTEN0 SECONE3 SECONE2 SECONE1 SECONE0
bit 7 bit 0
Legend:

bit 6-4 SECTEN<2:0>: Binary Coded Decimal Value of Second’s Tens Digit bits
bit 3-0 SECONE<3:0>: Binary Coded Decimal Value of Second’s Ones Digit bits

PIC18F87K90 FAMILY
17.1.4 RTCEN BIT WRITE 17.2 Operation
RTCWREN (RTCCFG<5>) must be set before a write
to RTCEN can take place. Any write to the RTCEN bit, 17.2.1 REGISTER INTERFACE
while RTCWREN = 0, will be ignored. The register interface for the RTCC and alarm values is
Like the RTCEN bit, the RTCVALH and RTCVALL implemented using the Binary Coded Decimal (BCD)
registers can only be written to when RTCWREN = 1. format. This simplifies the firmware when using the
A write to these registers, while RTCWREN = 0, will be module as each of the digits is contained within its own
ignored. 4-bit value (see Figure 17-2 and Figure 17-3).
FIGURE 17-2: TIMER DIGIT FORMAT
Year Month Day Day of Week
0-9 0-9 0-1 0-9 0-3 0-9 0-6
Hours 1/2 Second Bit

(24-hour format) Minutes Seconds (binary format)
0-2 0-9 0-5 0-9 0-5 0-9 0/1
FIGURE 17-3: ALARM DIGIT FORMAT
Month Day Day of Week
0-1 0-9 0-3 0-9 0-6
Hours
(24-hour format) Minutes Seconds
0-2 0-9 0-5 0-9 0-5 0-9

PIC18F87K90 FAMILY
17.2.2 CLOCK SOURCE Calibration of the crystal can be done through this
module to yield an error of 3 seconds or less per month.
As previously mentioned, the RTCC module is intended
(For further details, see Section 17.2.9 “Calibration”.)
to be clocked by an external Real-Time Clock (RTC)
crystal oscillating at 32.768 kHz, but an internal oscillator
can be used. The RTCC clock selection is decided by
the RTCOSC bit (CONFIG3L<0>).
FIGURE 17-4: CLOCK SOURCE MULTIPLEXING
32.768 kHz XTAL

Half Second
from SOSC
1:16384 Clock One Second Clock
Half Second(1)
Clock Prescaler(1)
Internal RC
CONFIG3L<0>
Day
Second Hour:Minute Month Year
Day of Week
Note 1: Writing to the lower half of the MINSEC register resets all counters, allowing fraction of a second
synchronization; clock prescaler is held in Reset when RTCEN = 0.
17.2.2.1 Real-Time Clock Enable TABLE 17-1: DAY OF WEEK SCHEDULE

The RTCC module can be clocked by an external Day of Week
32.768 kHz crystal (SOSC oscillator), or the LF-INTOSC
oscillator, which can be selected in CONFIG3L<0>. Sunday 0
Monday 1
If the external clock is used, the SOSC oscillator should
be enabled via the SOSCGO bit (OSCCON2<3>). If Tuesday 2
LF-INTOSC is providing the clock, the INTOSC clock Wednesday 3
can be brought out to the RTCC pin by the RTSEC- Thursday 4
SEL<1:0> bits (PADCFG<2:1>).
Friday 5
17.2.3 DIGIT CARRY RULES Saturday 6
This section explains which timer values are affected
when there is a rollover: TABLE 17-2: DAY-TO-MONTH ROLLOVER
• Time of Day: From 23:59:59 to 00:00:00 with a SCHEDULE
carry to the Day field Month Maximum Day Field
• Month: From 12/31 to 01/01 with a carry to the
01 (January) 31
Year field
02 (February) 28 or 29(1)
• Day of Week: From 6 to 0 with no carry (see
Table 17-1) 03 (March) 31
• Year Carry: From 99 to 00; this also surpasses the 04 (April) 30
use of the RTCC 05 (May) 31
For the day-to-month rollover schedule, see Table 17-2. 06 (June) 30
Because the following values are in BCD format, the 07 (July) 31
carry to the upper BCD digit occurs at the count of 10, 08 (August) 31
not 16 (SECONDS, MINUTES, HOURS, WEEKDAY,
09 (September) 30
DAYS and MONTHS).
10 (October) 31
11 (November) 30
12 (December) 31
Note 1: See Section 17.2.4 “Leap Year”.

PIC18F87K90 FAMILY
17.2.4 LEAP YEAR 17.2.7 WRITE LOCK
Since the year range on the RTCC module is 2000 to In order to perform a write to any of the RTCC Timer
2099, the leap year calculation is determined by any year registers, the RTCWREN bit (RTCCFG<5>) must be set.
divisible by four in the above range. Only February is To avoid accidental writes to the RTCC Timer register,
affected in a leap year. it is recommended that the RTCWREN bit
February will have 29 days in a leap year and 28 days in (RTCCFG<5>) be kept clear when not writing to the
any other year. register. For the RTCWREN bit to be set, there is only
one instruction cycle time window allowed between the
17.2.5 GENERAL FUNCTIONALITY 55h/AA sequence and the setting of RTCWREN. For
All Timer registers containing a time value of seconds or that reason, it is recommended that users follow the
greater are writable. The user configures the time by code example in Example 17-1.
writing the required year, month, day, hour, minutes and
seconds to the Timer registers, via Register Pointers. EXAMPLE 17-1: SETTING THE
(See Section 17.2.8 “Register Mapping”.) RTCWREN BIT
The timer uses the newly written values and proceeds movlw 0x55
with the count from the required starting point. movwf EECON2
movlw 0xAA
The RTCC is enabled by setting the RTCEN bit movwf EECON2
(RTCCFG<7>). If enabled, while adjusting these regis- bsf RTCCFG,RTCWREN
ters, the timer still continues to increment. However, any
time the MINSEC register is written to, both of the timer 17.2.8 REGISTER MAPPING
prescalers are reset to ‘0’. This allows fraction of a
To limit the register interface, the RTCC Timer and
second synchronization.
Alarm Timer registers are accessed through
The Timer registers are updated in the same cycle as corresponding Register Pointers. The RTCC Value
the write instruction’s execution by the CPU. The user register window (RTCVALH and RTCVALL) uses the
must ensure that when RTCEN = 1, the updated RTCPTRx bits (RTCCFG<1:0>) to select the required
registers will not be incremented at the same time. This Timer register pair.
can be accomplished in several ways:
By reading or writing to the RTCVALH register, the
• By checking the RTCSYNC bit (RTCCFG<4>) RTCC Pointer value (RTCPTR<1:0>) decrements by ‘1’
• By checking the preceding digits from which a until it reaches ‘00’. When ‘00’ is reached, the
carry can occur MINUTES and SECONDS value is accessible through
• By updating the registers immediately following RTCVALH and RTCVALL until the pointer value is
the seconds pulse (or an alarm interrupt) manually changed.
The user has visibility to the half-second field of the

TABLE 17-3: RTCVALH AND RTCVALL
counter. This value is read-only and can be reset only
by writing to the lower half of the SECONDS register. REGISTER MAPPING
RTCC Value Register Window
17.2.6 SAFETY WINDOW FOR REGISTER RTCPTR<1:0>
READS AND WRITES RTCVALH RTCVALL
The RTCSYNC bit indicates a time window during 00 MINUTES SECONDS

which the RTCC Clock Domain registers can be safely 01 WEEKDAY HOURS
read and written without concern about a rollover.
10 MONTH DAY
When RTCSYNC = 0, the registers can be safely
accessed by the CPU. 11 — YEAR
Whether RTCSYNC = 1 or 0, the user should employ a The Alarm Value register windows (ALRMVALH and
firmware solution to ensure that the data read did not ALRMVALL) use the ALRMPTR bits (ALRMCFG<1:0>)
fall on a rollover boundary, resulting in an invalid or to select the desired alarm register pair.
partial read. This firmware solution would consist of
By reading or writing to the ALRMVALH register, the
reading each register twice and then comparing the two
Alarm Pointer value, ALRMPTR<1:0>, decrements by
values. If the two values matched, then a rollover did
one until it reaches ‘00’. When it reaches ‘00’, the
not occur.
ALRMMIN and ALRMSEC value is accessible through
ALRMVALH and ALRMVALL until the pointer value is
manually changed.

PIC18F87K90 FAMILY
TABLE 17-4: ALRMVAL REGISTER Writes to the RTCCAL register should occur only when
MAPPING the timer is turned off or immediately after the rising
edge of the seconds pulse.
Alarm Value Register Window
ALRMPTR<1:0> Note: In determining the crystal’s error value, it
ALRMVALH ALRMVALL is the user’s responsibility to include the
00 ALRMMIN ALRMSEC crystal’s initial error from drift due to
temperature or crystal aging.
01 ALRMWD ALRMHR
10 ALRMMNTH ALRMDAY
17.3 Alarm
11 — —
The Alarm features and characteristics are:
17.2.9 CALIBRATION • Configurable from half a second to one year
The real-time crystal input can be calibrated using the • Enabled using the ALRMEN bit (ALRMCFG<7>,
periodic auto-adjust feature. When properly calibrated, Register 17-4)
the RTCC can provide an error of less than three • Offers one-time and repeat alarm options
seconds per month.
To perform this calibration, find the number of error 17.3.1 CONFIGURING THE ALARM
clock pulses and store the value into the lower half of The alarm feature is enabled using the ALRMEN bit.
the RTCCAL register. The 8-bit, signed value, loaded This bit is cleared when an alarm is issued. The bit will
into RTCCAL, is multiplied by four and will be either not be cleared if the CHIME bit = 1 or if ALRMRPT  0.
added or subtracted from the RTCC timer, once every
The interval selection of the alarm is configured
minute.
through the ALRMCFG bits (AMASK<3:0>) (see
To calibrate the RTCC module: Figure 17-5). These bits determine which, and how
1. Use another timer resource on the device to find many, digits of the alarm must match the clock value for
the error of the 32.768 kHz crystal. the alarm to occur.
2. Convert the number of error clock pulses per The alarm can also be configured to repeat based on a
minute (see Equation 17-1). preconfigured interval. The number of times this
occurs, after the alarm is enabled, is stored in the
EQUATION 17-1: CONVERTING ERROR
ALRMRPT register.
CLOCK PULSES
(Ideal Frequency (32,758) – Measured Frequency) * 60 = Note: While the alarm is enabled (ALRMEN = 1),
Error Clocks per Minute changing any of the registers, other than
the RTCCAL, ALRMCFG and ALRMRPT
• If the oscillator is faster than ideal (negative registers and the CHIME bit, can result in a
result from Step 2), the RCFGCALL register false alarm event leading to a false alarm
value needs to be negative. This causes the interrupt. To avoid this, only change the
specified number of clock pulses to be timer and alarm values while the alarm is
subtracted from the timer counter, once every disabled (ALRMEN = 0). It is recommended
minute. that the ALRMCFG and ALRMRPT regis-
• If the oscillator is slower than ideal (positive ters and CHIME bit be changed when
result from Step 2), the RCFGCALL register RTCSYNC = 0.
value needs to be positive. This causes the
specified number of clock pulses to be added to
the timer counter, once every minute.
3. Load the RTCCAL register with the correct
value.

PIC18F87K90 FAMILY
FIGURE 17-5: ALARM MASK SETTINGS
Alarm Mask Setting Day of the
AMASK<3:0> Week Month Day Hours Minutes Seconds
0000 – Every half second
0001 – Every second
0010 – Every 10 seconds s
0011 – Every minute s s
0100 – Every 10 minutes m s s
0101 – Every hour m m s s
0110 – Every day h h m m s s
0111 – Every week d h h m m s s
1000 – Every month d d h h m m s s
1001 – Every year(1) m m d d h h m m s s
Note 1: Annually, except when configured for February 29.
When ALRMCFG = 00 and the CHIME bit = 0 17.3.2 ALARM INTERRUPT

(ALRMCFG<6>), the repeat function is disabled and
At every alarm event, an interrupt is generated.
only a single alarm will occur. The alarm can be
Additionally, an alarm pulse output is provided that
repeated up to 255 times by loading the ALRMRPT
operates at half the frequency of the alarm.
register with FFh.
The alarm pulse output is completely synchronous with
After each alarm is issued, the ALRMRPT register is
the RTCC clock and can be used as a trigger clock to
decremented by one. Once the register has reached
other peripherals. This output is available on the RTCC
‘00’, the alarm will be issued one last time.
pin. The output pulse is a clock with a 50% duty cycle
After the alarm is issued a last time, the ALRMEN bit is and a frequency half that of the alarm event (see
cleared automatically and the alarm is turned off. Indefi- Figure 17-6).
nite repetition of the alarm can occur if the CHIME bit = 1.
The RTCC pin also can output the seconds clock. The
When CHIME = 1, the alarm is not disabled when the user can select between the alarm pulse, generated by
ALRMRPT register reaches ‘00’, but it rolls over to FF the RTCC module, or the seconds clock output.
and continues counting indefinitely.
The RTSECSEL<1:0> bits (PADCFG1<2:1>) select
between these two outputs:
• Alarm pulse – RTSECSEL<1:0> = 00
• Seconds clock – RTSECSEL<1:0> = 01

PIC18F87K90 FAMILY
FIGURE 17-6: TIMER PULSE GENERATION
RTCEN bit
ALRMEN bit
RTCC Alarm Event
RTCC Pin
17.4 Sleep Mode 17.5.2 POWER-ON RESET (POR)

The timer and alarm continue to operate while in Sleep The RTCCFG and ALRMRPT registers are reset only
mode. The operation of the alarm is not affected by on a POR. Once the device exits the POR state, the
Sleep, as an alarm event can always wake up the CPU. clock registers should be reloaded with the desired
values.
The Idle mode does not affect the operation of the timer
or alarm. The timer prescaler values can be reset only by writing
to the SECONDS register. No device Reset can affect
the prescalers.
17.5 Reset
17.5.1 DEVICE RESET
When a device Reset occurs, the ALRMRPT register is
forced to its Reset state, causing the alarm to be
disabled (if enabled prior to the Reset). If the RTCC
was enabled, it will continue to operate when a basic
device Reset occurs.

PIC18F87K90 FAMILY
17.6 Register Maps
Table 17-5, Table 17-6 and Table 17-7 summarize the
registers associated with the RTCC module.
TABLE 17-5: RTCC CONTROL REGISTERS

All
File Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Resets
on Page:
RTCCFG RTCEN — RTCWREN RTCSYNC HALFSEC RTCOE RTCPTR1 RTCPTR0 80

RTCCAL CAL7 CAL6 CAL5 CAL4 CAL3 CAL2 CAL1 CAL0 80
(1)
PADCFG1 RDPU REPU RJPU — — RTSECSEL1 RTSECSEL0 — 80
ALRMCFG ALRMEN CHIME AMASK3 AMASK2 AMASK1 AMASK0 ALRMPTR1 ALRMPTR0 80
ALRMRPT ARPT7 ARPT6 ARPT5 ARPT4 ARPT3 ARPT2 ARPT1 ARPT0 80
Legend: — = unimplemented, read as ‘0’. Reset values are shown in hexadecimal for 80-pin devices.
Note 1: Not available on 64-pin devices.
TABLE 17-6: RTCC VALUE REGISTERS

All Resets
File Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
on Page:
RTCVALH RTCC Value High Register Window Based on RTCPTR<1:0> 80

RTCVALL RTCC Value Low Register Window Based on RTCPTR<1:0> 80
TABLE 17-7: ALARM VALUE REGISTERS

All Resets
on Page:
ALRMVALH Alarm Value High Register Window Based on ALRMPTR<1:0> 80

ALRMVALL Alarm Value Low Register Window Based on ALRMPTR<1:0> 80

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
18.0 CAPTURE/COMPARE/PWM Each CCP module contains a 16-bit register that can
operate as a 16-bit Capture register, a 16-bit Compare
(CCP) MODULES
register or a PWM Master/Slave Duty Cycle register.
PIC18F87K90 family devices have seven CCP For the sake of clarity, all CCP module operation in the
(Capture/Compare/PWM) modules, designated CCP4 following sections is described with respect to CCP4,
through CCP10. All the modules implement standard but is equally applicable to CCP5 through CCP10.
Capture, Compare and Pulse-Width Modulation
Note: The CCP9 and CCP10 modules are
(PWM) modes.
disabled on the devices with 32 Kbytes of
Note: Throughout this section, generic references program memory (PIC18FX5K90).
are used for register and bit names that are
the same, except for an ‘x’ variable that
indicates the item’s association with the
specific CCP module. For example, the
control register is named CCPxCON and
refers to CCP4CON through CCP10CON.
REGISTER 18-1: CCPxCON: CCPx CONTROL REGISTER (CCP4-CCP10 MODULES)(1)

U-0 U-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0
— — DCxB1 DCxB0 CCPxM3 CCPxM2 CCPxM1 CCPxM0
bit 7 bit 0
Legend:

bit 5-4 DCxB<1:0>: PWM Duty Cycle for CCPx Module bits (bit 1, bit 0)
Capture mode:
Unused.
Compare mode:
Unused.
PWM mode:
These bits are the two Least Significant bits (bit 1 and bit 0) of the 10-bit PWM duty cycle. The eight
Most Significant bits (DCxB<9:2>) of the duty cycle are found in CCPRxL.
bit 3-0 CCPxM<3:0>: CCPx Module Mode Select bits
0000 = Capture/Compare/PWM disabled (resets CCPx module)
0001 = Reserved
0010 = Compare mode: toggle output on match (CCPxIF bit is set)
0011 = Reserved
0100 = Capture mode: every falling edge
0101 = Capture mode: every rising edge
0110 = Capture mode: every 4th rising edge
1000 = Compare mode: initialize CCPx pin low; on compare match, force CCPx pin high (CCPxIF bit
is set)
1001 = Compare mode: initialize CCPx pin high; on compare match, force CCPx pin low (CCPxIF bit
is set)
1010 = Compare mode: generate software interrupt on compare match (CCPxIF bit is set, CCPx pin
reflects I/O state)
1011 = Compare mode: Special Event Trigger; reset timer on CCPx match (CCPxIF bit is set)(2)
11xx = PWM mode
Note 1: The CCP9 and CCP10 modules are not available on devices with 32 Kbytes of program memory
(PIC18FX5K90).
2: CCPxM<3:0> = 1011 will only reset the timer and not start the A/D conversion on a CCPx match.

PIC18F87K90 FAMILY
REGISTER 18-2: CCPTMRS1: CCPx TIMER SELECT REGISTER 1
R/W-0 R/W-0 U-0 R/W-0 U-0 R/W-0 R/W-0 R/W-0
C7TSEL1 C7TSEL0 — C6TSEL0 — C5TSEL0 C4TSEL1 C4TSEL0
bit 7 bit 0
Legend:
bit 7-6 C7TSEL<1:0>: CCP7 Timer Selection bits

00 = CCP7 is based off of TMR1/TMR2
bit 4 C6TSEL0: CCP6 Timer Selection bit
bit 2 C5TSEL0: CCP5 Timer Selection bit

PIC18F87K90 FAMILY
REGISTER 18-3: CCPTMRS2: CCPx TIMER SELECT REGISTER 2
U-0 U-0 U-0 R/W-0 U-0 R/W-0 R/W-0 R/W-0
— — — C10TSEL0(1) — C9TSEL0(1) C8TSEL1 C8TSEL0
bit 7 bit 0
Legend:

bit 4 C10TSEL0: CCP10 Timer Selection bit(1)
bit 2 C9TSEL0: CCP9 Timer Selection bit(1)
On non 32-Kbyte device variants:
On 32-Kbyte device variants (PIC18F85K90/65K90:
Note 1: This bit is unimplemented and reads as ‘0’ on devices with 32 Kbytes of program memory (PIC18FX5K90).

PIC18F87K90 FAMILY
REGISTER 18-4: CCPRxL: CCPx PERIOD LOW BYTE REGISTER
CCPRxL7 CCPRxL6 CCPRxL5 CCPRxL4 CCPRxL3 CCPRxL2 CCPRxL1 CCPRxL0
bit 7 bit 0
Legend:
bit 7-0 CCPRxL<7:0>: CCPx Period Register Low Byte bits

Capture Mode:
Capture register low byte.
Compare Mode:
Compare register low byte.
PWM Mode:
Duty Cycle register low byte.
REGISTER 18-5: CCPRxH: CCPx PERIOD HIGH BYTE REGISTER

CCPRxH7 CCPRxH6 CCPRxH5 CCPRxH4 CCPRxH3 CCPRxH2 CCPRxH1 CCPRxH0
bit 7 bit 0
Legend:
bit 7-0 CCPRxH<7:0>: CCPx Period Register High Byte bits

Capture Mode:
Capture register high byte.
Compare Mode:
Compare register high byte.
PWM Mode:
Duty Cycle Buffer register high byte.

PIC18F87K90 FAMILY
18.1 CCP Module Configuration TABLE 18-1: CCP MODE – TIMER
RESOURCE
Each Capture/Compare/PWM module is associated
with a control register (generically, CCPxCON) and a CCP Mode Timer Resource
data register (CCPRx). The data register, in turn, is
Capture
comprised of two 8-bit registers: CCPRxL (low byte) Timer1, Timer3, Timer 5 or Timer7
and CCPRxH (high byte). All registers are both Compare
readable and writable. PWM Timer2, Timer4, Timer 6 or Timer8
18.1.1 CCP MODULES AND TIMER The assignment of a particular timer to a module is
RESOURCES determined by the Timer to CCP enable bits in the
CCPTMRSx registers (see Register 18-2 and
The CCP modules utilize Timers, 1 through 8, which Register 18-3). All of the modules may be active at
varies with the selected mode. Various timers are avail- once and may share the same timer resource if they
able to the CCP modules in Capture, Compare or PWM are configured to operate in the same mode
modes, as shown in Table 18-1. (Capture/Compare or PWM) at the same time.
The CCPTMRS1 register selects the timers for CCP
modules, 7, 6, 5 and 4, and the CCPTMRS2 register
selects the timers for CCP modules, 10, 9 and 8. The
possible configurations are shown in Table 18-2 and
Table 18-3.
TABLE 18-2: TIMER ASSIGNMENTS FOR CCP MODULES 4, 5, 6 AND 7

CCPTMRS1 Register
CCP4 CCP5 CCP6 CCP7

Capture/ Capture/ Capture/ Capture/
C4TSEL PWM PWM PWM C7TSEL PWM
Compare C5TSEL0 Compare C6TSEL0 Compare Compare
<1:0> Mode Mode Mode <1:0> Mode
Mode Mode Mode Mode
0 0 TMR1 TMR2 0 TMR1 TMR2 0 TMR1 TMR2 0 0 TMR1 TMR2
0 1 TMR3 TMR4 1 TMR5 TMR4 1 TMR5 TMR2 0 1 TMR5 TMR4
1 0 TMR3 TMR6 1 0 TMR5 TMR6
1 1 Reserved(1) 1 1 TMR5 TMR8
Note 1: Do not use the reserved bits.
TABLE 18-3: TIMER ASSIGNMENTS FOR CCP MODULES 8, 9 AND 10

CCPTMRS2 Register
CCP8
CCP8 CCP9(1) CCP10(1)
Devices with 32 Kbytes(1)
Capture/ Capture/ Capture/ Capture/

C8TSEL PWM C8TSEL PWM PWM PWM
Compare Compare C9TSEL0 Compare C10TSEL0 Compare
<1:0> Mode <1:0> Mode Mode Mode
Mode Mode Mode Mode
0 0 TMR1 TMR2 0 0 TMR1 TMR2 0 TMR1 TMR2 0 TMR1 TMR2
0 1 TMR7 TMR4 0 1 TMR1 TMR4 1 TMR7 TMR4 1 TMR7 TMR2
1 0 TMR7 TMR6 1 0 TMR1 TMR6
1 1 Reserved(2) 1 1 Reserved(2)
Note 1: The module is not available for devices with 32 Kbytes of program memory.
2: Do not use the reserved bits.

PIC18F87K90 FAMILY
18.1.2 OPEN-DRAIN OUTPUT OPTION 18.2 Capture Mode
When operating in Output mode (the Compare or PWM In Capture mode, the CCPR4H:CCPR4L register pair
modes), the drivers for the CCPx pins can be optionally captures the 16-bit value of the TMR1 or TMR3 register
configured as open-drain outputs. This feature allows when an event occurs on the CCP4 pins. An event is
the voltage level on the pin to be pulled to a higher level defined as one of the following:
through an external pull-up resistor and allows the
output to communicate with external circuits without the • Every falling edge
need for additional level shifters. • Every rising edge
The open-drain output option is controlled by the • Every 4th rising edge
CCPxOD bits (ODCON2<7:2>). Setting the appropriate • Every 16th rising edge
bit configures the pin for the corresponding module for The event is selected by the mode select bits,
open-drain operation. CCP4M<3:0> (CCP4CON<3:0>). When a capture is
made, the interrupt request flag bit, CCP4IF (PIR4<1>),
18.1.3 PIN ASSIGNMENT FOR CCP6, is set. (It must be cleared in software.) If another
CCP7, CCP8 AND CCP9 capture occurs before the value in CCPR4 is read, the
The pin assignment for CCP6/7/8/9 (Capture input, old captured value is overwritten by the new captured
Compare and PWM output) can change, based on the value.
device configuration. Figure 18-1 shows the Capture mode block diagram.
The ECCPMX Configuration bit (CONFIG3H<1>)
determines the pin to which CCP6/7/8/9 is multiplexed. 18.2.1 CCP PIN CONFIGURATION
The pin assignments for these CCP modules are given In Capture mode, the appropriate CCPx pin should be
in Table 18-4. configured as an input by setting the corresponding
TRIS direction bit.
TABLE 18-4: CCP PIN ASSIGNMENT
Note: If RC1 or RE7 is configured as a CCP4
ECCPMX Pin Mapped To output, a write to the PORT causes a
Value capture condition.
CCP6 CCP7 CCP8 CC9
1 18.2.2 TIMER1/3/5/7 MODE SELECTION
RE6 RE5 RE4 RE3
(Default) For the available timers (1/3/5/7) to be used for the cap-
0 RH7 RH6 RH5 RH4 ture feature, the used timers must be running in Timer
mode or Synchronized Counter mode. In Asynchronous
Counter mode, the capture operation may not work.
The timer to be used with each CCP module is selected
in the CCPTMRSx registers. (See Section 18.1.1 “CCP
Modules and Timer Resources”.)
Details of the timer assignments for the CCP modules
are given in Table 18-2 and Table 18-3.

PIC18F87K90 FAMILY
FIGURE 18-1: CAPTURE MODE OPERATION BLOCK DIAGRAM
TMR5H TMR5L
Set CCP5IF
C5TSEL0 TMR5
CCP5 Pin Enable
Prescaler and CCPR5H CCPR5L
 1, 4, 16 Edge Detect
TMR1
C5TSEL0 Enable
4 TMR1H TMR1L
CCP5CON<3:0> Set CCP4IF
4
Q1:Q4
4
CCP4CON<3:0>
C4TSEL1 TMR3H TMR3L
C4TSEL0
TMR3
Enable
CCP4 Pin
TMR1
Enable
C4TSEL0
TMR1H TMR1L
C4TSEL1
Note: This block diagram uses CCP4 and CCP5, and their appropriate timers, as an example. For details on all
of the CCP modules and their timer assignments, see Table 18-2 and Table 18-3.
18.2.3 SOFTWARE INTERRUPT Switching from one capture prescaler to another may
generate an interrupt. Doing that also will not clear the
When the Capture mode is changed, a false capture
prescaler counter – meaning the first capture may be
interrupt may be generated. The user should keep the
from a non-zero prescaler.
CCP4IE bit (PIE4<1>) clear to avoid false interrupts
and should clear the flag bit, CCP4IF, following any Example 18-1 shows the recommended method for
such change in operating mode. switching between capture prescalers. This example
also clears the prescaler counter and will not generate
18.2.4 CCP PRESCALER the “false” interrupt.
There are four prescaler settings in Capture mode.
They are specified as part of the operating mode EXAMPLE 18-1: CHANGING BETWEEN
selected by the mode select bits (CCP4M<3:0>). CAPTURE PRESCALERS
Whenever the CCP module is turned off, or the CCP CLRF CCP4CON ; Turn CCP module off
module is not in Capture mode, the prescaler counter MOVLW NEW_CAPT_PS ; Load WREG with the
is cleared. This means that any Reset will clear the ; new prescaler mode
prescaler counter. ; value and CCP ON
MOVWF CCP4CON ; Load CCP4CON with
; this value

PIC18F87K90 FAMILY
18.3 Compare Mode 18.3.3 SOFTWARE INTERRUPT MODE
In Compare mode, the 16-bit CCPR4 register value is When the Generate Software Interrupt mode is chosen
constantly compared against either the TMR1 or TMR3 (CCP4M<3:0> = 1010), the CCP4 pin is not affected.
register pair value. When a match occurs, the CCP4 Only a CCP interrupt is generated, if enabled, and the
pin can be: CCP4IE bit is set.
• Driven high 18.3.4 SPECIAL EVENT TRIGGER

• Driven low
Both CCP modules are equipped with a Special Event
• Toggled (high-to-low or low-to-high) Trigger. This is an internal hardware signal generated
• Unchanged (that is, reflecting the state of the I/O in Compare mode to trigger actions by other modules.
latch) The Special Event Trigger is enabled by selecting
The action on the pin is based on the value of the mode the Compare Special Event Trigger mode
select bits (CCP4M<3:0>). At the same time, the (CCP4M<3:0> = 1011).
interrupt flag bit, CCP4IF, is set. For either CCP module, the Special Event Trigger resets
Figure 18-2 shows the Compare mode block diagram the timer register pair for whichever timer resource is
currently assigned as the module’s time base. This
18.3.1 CCP PIN CONFIGURATION allows the CCPRx registers to serve as a programmable
Period register for either timer.
The user must configure the CCPx pin as an output by
clearing the appropriate TRIS bit. The Special Event Trigger for CCP4 cannot start an
A/D conversion.
Note: Clearing the CCP4CON register will force
the RC1 or RE7 compare output latch Note: The Special Event Trigger of ECCP1 can
(depending on device configuration) to the start an A/D conversion, but the A/D
default low level. This is not the PORTC or Converter needs to be enabled. For
PORTE I/O data latch. more information, see Section 19.0
“Enhanced Capture/Compare/PWM
18.3.2 TIMER1/3/5/7 MODE SELECTION (ECCP) Module”.
If the CCP module is using the compare feature in
conjunction with any of the Timer1/3/5/7 timers, the tim-
ers must be running in Timer mode or Synchronized
Counter mode. In Asynchronous Counter mode, the
compare operation may not work.
Note: Details of the timer assignments for the
CCP modules are given in Table 18-2 and
Table 18-3.

PIC18F87K90 FAMILY
FIGURE 18-2: COMPARE MODE OPERATION BLOCK DIAGRAM
Special Event Trigger
Set CCP5IF (Timer1/5 Reset)
CCPR5H CCPR5L
CCP5 Pin
Compare Output S Q
Comparator
Match Logic
R
TRIS
4 Output Enable
CCP5CON<3:0>
TMR1H TMR1L 0
TMR5H TMR5L 1
C5TSEL0
0 TMR1H TMR1L
1 TMR3H TMR3L
(Timer1/Timer3 Reset, A/D Trigger)
C4TSEL1
C4TSEL0
Set CCP4IF CCP4 Pin
Compare Output S Q
Comparator
Match Logic
R
TRIS
4 Output Enable
CCPR4H CCPR4L
CCP4CON<3:0>
Note: This block diagram uses CCP4 and CCP5, and their appropriate timers, as an example. For details on all
of the CCP modules and their timer assignments, see Table 18-2 and Table 18-3.

PIC18F87K90 FAMILY
TABLE 18-5: REGISTERS ASSOCIATED WITH CAPTURE, COMPARE, TIMER1/3/5/7
Reset
on Page:
PIE4 CCP10IE(1) CCP9IE (1)
CCP8IE CCP7IE CCP6IE CCP5IE CCP4IE CCP3IE 77
IPR4 CCP10IP(1) CCP9IP(1) CCP8IP CCP7IP CCP6IP CCP5IP CCP4IP CCP3IP 77
TRISH(2) TRISH7 TRISH6 TRISH5 TRISH4 TRISH3 TRISH2 TRISH1 TRISH0 78
TMR7L(1) Timer7 Register Low Byte 81
(1)
T7CON(1) TMR7CS1 TMR7CS0 T7CKPS1 T7CKPS0 SOSCEN T7SYNC RD16 TMR7ON 81
CCPR4L Capture/Compare/PWM Register 4 Low Byte 82
CCPR4H Capture/Compare/PWM Register 4 High Byte 82
CCPR9L(1) Capture/Compare/PWM Register 9 Low Byte 80
(1)
CCPR10H(1) Capture/Compare/PWM Register 10 High Byte 80
CCP4CON — — DC4B1 DC4B0 CCP4M3 CCP4M2 CCP4M1 CCP4M0 82
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by Capture/Compare or Timer1/3/5/7.
2: Unimplemented in 64-pin devices.

PIC18F87K90 FAMILY
TABLE 18-5: REGISTERS ASSOCIATED WITH CAPTURE, COMPARE, TIMER1/3/5/7 (CONTINUED)
Reset
on Page:
(1)
CCP10CON(1) — — DC10B1 DC10B0 CCP10M3 CCP10M2 CCP10M1 CCP10M0 81
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by Capture/Compare or Timer1/3/5/7.
2: Unimplemented in 64-pin devices.
18.4 PWM Mode FIGURE 18-3: SIMPLIFIED PWM BLOCK

DIAGRAM
In Pulse-Width Modulation (PWM) mode, the CCP4 pin
produces up to a 10-bit resolution PWM output. Since Duty Cycle Registers CCP4CON<5:4>
the CCP4 pin is multiplexed with a PORTC or PORTE
data latch, the appropriate TRIS bit must be cleared to CCPR4L (Note 2)
make the CCP4 pin an output.
Note: Clearing the CCP4CON register will force
the RC1 or RE7 output latch (depending CCPR4H (Slave) (Note 2)
on device configuration) to the default low
level. This is not the PORTC or PORTE
Comparator R Q
I/O data latch.
RC2/ECCP1
Figure 18-3 shows a simplified block diagram of the TMR2 (Note 1)
ECCP1 module in PWM mode. S
For a step-by-step procedure on how to set up the CCP

Comparator TRISC<2>
module for PWM operation, see Section 18.4.3
Clear Timer,
“Setup for PWM Operation”. ECCP1 Pin and
PR2 Latch D.C.
Note 1: The 8-bit TMR2 value is concatenated with the 2-bit

internal Q clock, or 2 bits of the prescaler, to create
the 10-bit time base.
2: CCP4 and its appropriate timers are used as an
example. For details on all of the CCP modules and
their timer assignments, see Table 18-2 and
Table 18-3.

PIC18F87K90 FAMILY
A PWM output (Figure 18-4) has a time base (period) 18.4.2 PWM DUTY CYCLE
and a time that the output stays high (duty cycle). The
The PWM duty cycle is specified by writing to the
frequency of the PWM is the inverse of the period
CCPR4L register (using CCP4 as an example) and to
(1/period).
the CCP4CON<5:4> bits. Up to 10-bit resolution is avail-
able. The CCPR4L contains the eight MSbs and the
FIGURE 18-4: PWM OUTPUT CCP4CON<5:4> bits contain the two LSbs. This 10-bit
Period value is represented by CCPR4L:CCP4CON<5:4>. The
following equation is used to calculate the PWM duty
cycle in time:
Duty Cycle
EQUATION 18-2:
TMR2 = PR2 PWM Duty Cycle = (CCPR4L:CCP4CON<5:4>) •
TOSC • (TMR2 Prescale Value)
TMR2 = Duty Cycle
TMR2 = PR2 CCPR4L and CCP4CON<5:4> can be written to at any

time, but the duty cycle value is not latched into
CCPR4H until after a match between PR2 and TMR2
18.4.1 PWM PERIOD occurs (that is, the period is complete). In PWM mode,
The PWM period is specified by writing to the PR2 CCPR4H is a read-only register.
register. The PWM period can be calculated using the The CCPR4H register and a 2-bit internal latch are
following formula: used to double-buffer the PWM duty cycle. This
double-buffering is essential for glitchless PWM
EQUATION 18-1: operation.
PWM Period = [(PR2) + 1] • 4 • TOSC • When the CCPR4H and 2-bit latch match TMR2,
(TMR2 Prescale Value) concatenated with an internal 2-bit Q clock or two bits
of the TMR2 prescaler, the CCP4 pin is cleared.
PWM frequency is defined as 1/[PWM period].
The maximum PWM resolution (bits) for a given PWM
When TMR2 is equal to PR2, the following three events frequency is given by Equation 18-3:
occur on the next increment cycle:
• TMR2 is cleared EQUATION 18-3:
• The CCP4 pin is set F OSC
log  ---------------
(An exception: If PWM duty cycle = 0%, the CCP4  F PWM
PWM Resolution (max) = -----------------------------bits
pin will not be set) log  2 
• The PWM duty cycle is latched from CCPR4L into
CCPR4H
Note: If the PWM duty cycle value is longer than
Note: The Timer2 postscalers (see the PWM period, the CCP4 pin will not be
Section 14.0 “Timer2 Module”) are not cleared.
used in the determination of the PWM
frequency. The postscaler could be used
to have a servo update rate at a different
frequency than the PWM output.
TABLE 18-6: EXAMPLE PWM FREQUENCIES AND RESOLUTIONS AT 40 MHz

PWM Frequency 2.44 kHz 9.77 kHz 39.06 kHz 156.25 kHz 312.50 kHz 416.67 kHz
Timer Prescaler (1, 4, 16) 16 4 1 1 1 1
PR2 Value FFh FFh FFh 3Fh 1Fh 17h
Maximum Resolution (bits) 14 12 10 8 7 6.58

PIC18F87K90 FAMILY
18.4.3 SETUP FOR PWM OPERATION 3. Make the CCP4 pin an output by clearing the
appropriate TRIS bit.
To configure the CCP module for PWM operation (with
CCP4 as an example): 4. Set the TMR2 prescale value, then enable
Timer2 by writing to T2CON.
1. Set the PWM period by writing to the PR2
5. Configure the CCP4 module for PWM operation.
register.
2. Set the PWM duty cycle by writing to the
CCPR4L register and CCP4CON<5:4> bits.
TABLE 18-7: REGISTERS ASSOCIATED WITH PWM AND TIMERS

Reset
on Page:
PIR4 CCP10IF(1) CCP9IF (1)
CCP8IF CCP7IF CCP6IF CCP5IF CCP4IF CCP3IF 77
PIE4 CCP10IE(1) CCP9IE(1) CCP8IE CCP7IE CCP6IE CCP5IE CCP4IE CCP3IE 77
TRISH TRISH7 TRISH6 TRISH5 TRISH4 TRISH3 TRISH2 TRISH1 TRISH0 78
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by PWM or Timer2/4/6/8.

PIC18F87K90 FAMILY
TABLE 18-7: REGISTERS ASSOCIATED WITH PWM AND TIMERS (CONTINUED)
Reset
on Page:
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by PWM or Timer2/4/6/8.

PIC18F87K90 FAMILY
19.0 ENHANCED ECCP1, ECCP2 and ECCP3 are implemented as
standard CCP modules with Enhanced PWM
CAPTURE/COMPARE/PWM
capabilities. These include:
(ECCP) MODULE
• Provision for two or four output channels
PIC18F87K90 family devices have three Enhanced • Output Steering modes
Capture/Compare/PWM (ECCP) modules: ECCP1,
• Programmable polarity
ECCP2 and ECCP3. These modules contain a 16-bit
register, which can operate as a 16-bit Capture register, • Programmable dead-band control
a 16-bit Compare register or a PWM Master/Slave Duty • Automatic shutdown and restart
Cycle register. These ECCP modules are upward The enhanced features are discussed in detail in
compatible with CCP Section 19.4 “PWM (Enhanced Mode)”.
Note: Throughout this section, generic references The ECCP1, ECCP2 and ECCP3 modules use the
are used for register and bit names that are control registers, CCP1CON, CCP2CON and
the same, except for an ‘x’ variable that indi- CCP3CON. The control registers, CCP4CON through
cates the item’s association with the CCP10CON, are for the CCP4 through CCP10
ECCP1, ECCP2 or ECCP3 module. For modules.
example, the control register is named
CCPxCON and refers to CCP1CON,
CCP2CON and CCP3CON.

PIC18F87K90 FAMILY
REGISTER 19-1: CCPxCON: ENHANCED CAPTURE/COMPARE/PWM x CONTROL
PxM1 PxM0 DCxB1 DCxB0 CCPxM3 CCPxM2 CCPxM1 CCPxM0
bit 7 bit 0
Legend:
bit 7-6 PxM<1:0>: Enhanced PWM Output Configuration bits

If CCPxM<3:2> = 00, 01, 10:
xx = PxA is assigned as a capture/compare input/output; PxB, PxC and PxD are assigned as PORT pins
If CCPxM<3:2> = 11:
00 = Single output: PxA, PxB, PxC and PxD are controlled by steering (see Section 19.4.7 “Pulse
Steering Mode”)
01 = Full-bridge output forward: PxD is modulated; PxA is active; PxB, PxC are inactive
10 = Half-bridge output: PxA, PxB are modulated with dead-band control; PxC and PxD are
assigned as PORT pins
11 = Full-bridge output reverse: PxB is modulated; PxC is active; PxA and PxD are inactive
bit 5-4 DCxB<1:0>: PWM Duty Cycle Bit 1 and Bit 0
Capture mode:
Unused.
Compare mode:
Unused.
PWM mode:
These bits are the two LSbs of the 10-bit PWM duty cycle. The eight MSbs of the duty cycle are found
in CCPRxL.
bit 3-0 CCPxM<3:0>: ECCPx Mode Select bits
0000 = Capture/Compare/PWM off (resets ECCPx module)
0001 = Reserved
0010 = Compare mode: toggle output on match
0011 = Capture mode
0100 = Capture mode: every falling edge
0101 = Capture mode: every rising edge
0110 = Capture mode: every fourth rising edge
1000 = Compare mode: initialize the ECCPx pin low; set the output on a compare match (set CCPxIF)
1001 = Compare mode: initialize the ECCPx pin high; clear the output on a compare match (set CCPxIF)
1010 = Compare mode: generate a software interrupt only; ECCPx pin reverts to an I/O state
1011 = Compare mode: trigger special event (ECCPx resets TMR1 or TMR3, starts A/D conversion,
sets CCxIF bit)
1100 = PWM mode: PxA and PxC are active-high; PxB and PxD are active-high
1101 = PWM mode: PxA and PxC are active-high; PxB and PxD are active-low
1110 = PWM mode: PxA and PxC are active-low; PxB and PxD are active-high
1111 = PWM mode: PxA and PxC are active-low; PxB and PxD are active-low

PIC18F87K90 FAMILY
REGISTER 19-2: CCPTMRS0: CCP TIMER SELECT 0 REGISTER
C3TSEL1 C3TSEL0 C2TSEL2 C2TSEL1 C2TSEL0 C1TSEL2 C1TSEL1 C1TSEL0
bit 7 bit 0
Legend:
bit 7-6 C3TSEL<1:0>: ECCP3 Timer Selection bits

00 = ECCP3 is based off of TMR1/TMR2
100 = ECCP2 is based off of TMR3/TMR10; option is reserved on the 32-Kbyte device variant; do not use

PIC18F87K90 FAMILY
In addition to the expanded range of modes available The assignment of a particular timer to a module is
through the CCPxCON and ECCPxAS registers, the determined by the timer to ECCP enable bits in the
ECCP modules have two additional registers associated CCPTMRSx register (Register 19-2). The interactions
with Enhanced PWM operation and auto-shutdown between the two modules are depicted in Figure 19-1.
features. They are: Capture operations are designed to be used when the
• ECCPxDEL – Enhanced PWM Control timer is configured for Synchronous Counter mode.
Capture operations may not work as expected if the
• PSTRxCON – Pulse Steering Control
associated timer is configured for Asynchronous Counter
mode.
19.1 ECCP Outputs and Configuration
19.1.2 ECCP PIN ASSIGNMENT
The Enhanced CCP module may have up to four PWM
outputs, depending on the selected operating mode. The pin assignment for ECCPx (capture input,
The CCPxCON register is modified to allow control compare and PWM output) can change, based on
over four PWM outputs: ECCPx/PxA, PxB, PxC and device configuration. The ECCPMX (CONFIG3H<1>)
PxD. Applications can use one, two or four of these Configuration bit determines which pins, ECCP1 and
outputs. ECCP3, are multiplexed to.
The outputs that are active depend on the selected • Default/ECCPMX = 1:
ECCP operating mode. The pin assignments are - ECCP1 (P1B/P1C) is multiplexed onto RE6
summarized in Table 19-3. and RE5
To configure the I/O pins as PWM outputs, the proper - ECCP3 (P3B/P3C) is multiplexed onto RE4
PWM mode must be selected by setting the PxM<1:0> and RE3
and CCPxM<3:0> bits. The appropriate TRIS direction • ECCPMX = 0:
bits for the PORT pins must also be set as outputs. - ECCP1 (P1B/P1C) is multiplexed onto RH7
and RH6
19.1.1 ECCP MODULE AND TIMER
- ECCP3 (P3B/P3C) is multiplexed onto RH5
RESOURCES
and RH4.
The ECCP modules use Timers, 1, 2, 3, 4, 6, 8, 10 or 12,
The pin assignment for ECCP2 (capture input,
depending on the mode selected. These timers are
compare and PWM output) can change, based on
available to CCP modules in Capture, Compare or PWM
device configuration.
modes, as shown in Table 19-1.
The CCP2MX Configuration bit (CONFIG3H<0>)
TABLE 19-1: ECCP MODE – TIMER determines which pin, ECCP2, is multiplexed to.
RESOURCE • If CCP2MX = 1 (default) – ECCP2 is multiplexed
to RC1
ECCP Mode Timer Resource
• If CCP2MX = 0 – ECCP2 is multiplexed to:
Capture Timer1 or Timer3 - RE7 is the ECCP2 pin with CCP2MX = 0
Compare Timer1 or Timer3
PWM Timer2, Timer4, Timer6, Timer8,
Timer10 or Timer12

PIC18F87K90 FAMILY
19.2 Capture Mode 19.2.2 TIMER1/TIMER3 MODE SELECTION
In Capture mode, the CCPRxH:CCPRxL register pair The timers that are to be used with the capture feature
captures the 16-bit value of the TMR1 or TMR3 (Timer1 and/or Timer3) must be running in Timer mode
registers when an event occurs on the corresponding or Synchronized Counter mode. In Asynchronous
ECCPx pin. An event is defined as one of the following: Counter mode, the capture operation may not work.
The timer to be used with each ECCP module is
• Every falling edge selected in the CCPTMRS0 register (Register 19-2).
• Every rising edge
• Every fourth rising edge 19.2.3 SOFTWARE INTERRUPT
• Every 16th rising edge When the Capture mode is changed, a false capture
The event is selected by the mode select bits, interrupt may be generated. The user should keep the
CCPxM<3:0> (CCPxCON register<3:0>). When a CCPxIE interrupt enable bit clear to avoid false interrupts.
capture is made, the interrupt request flag bit, CCPxIF, The interrupt flag bit, CCPxIF, should also be cleared
is set (see Table 19-2). The flag must be cleared by following any such change in operating mode.
software. If another capture occurs before the value in
the CCPRxH/L register is read, the old captured value 19.2.4 ECCP PRESCALER
is overwritten by the new captured value. There are four prescaler settings in Capture mode; they
are specified as part of the operating mode selected by
TABLE 19-2: ECCP1/2/3 INTERRUPT FLAG
the mode select bits (CCPxM<3:0>). Whenever the
BITS
ECCP module is turned off, or Capture mode is
ECCP Module Flag Bit disabled, the prescaler counter is cleared. This means
that any Reset will clear the prescaler counter.
1 PIR3<1>
Switching from one capture prescaler to another may
2 PIR3<2>
generate an interrupt. Also, the prescaler counter will
3 PIR4<0> not be cleared; therefore, the first capture may be from
a non-zero prescaler. Example 19-1 provides the
19.2.1 ECCP PIN CONFIGURATION recommended method for switching between capture
In Capture mode, the appropriate ECCPx pin should be prescalers. This example also clears the prescaler
configured as an input by setting the corresponding counter and will not generate the “false” interrupt.
TRIS direction bit.
EXAMPLE 19-1: CHANGING BETWEEN
Note: If the ECCPx pin is configured as an
output, a write to the PORT can cause a CAPTURE PRESCALERS
capture condition. CLRF CCP1CON ; Turn ECCP module off
MOVLW NEW_CAPT_PS ; Load WREG with the
; new prescaler mode
; value and ECCP ON
MOVWF CCP1CON ; Load CCP1CON with
; this value
FIGURE 19-1: CAPTURE MODE OPERATION BLOCK DIAGRAM
TMR3H TMR3L
Set CCP1IF
C1TSEL0
C1TSEL1 TMR3
ECCP1 Pin C1TSEL2 Enable
C1TSEL0 TMR1
C1TSEL1 Enable
C1TSEL2
4 TMR1H TMR1L
CCP1CON<3:0>
4
Q1:Q4

PIC18F87K90 FAMILY
19.3 Compare Mode 19.3.2 TIMER1/TIMER3 MODE SELECTION
In Compare mode, the 16-bit CCPRx register value is Timer1 and/or Timer3 must be running in Timer mode
constantly compared against either the TMR1 or TMR3 or Synchronized Counter mode if the ECCP module is
register pair value. When a match occurs, the ECCPx using the compare feature. In Asynchronous Counter
pin can be: mode, the compare operation will not work reliably.
• Driven high 19.3.3 SOFTWARE INTERRUPT MODE

• Driven low
When the Generate Software Interrupt mode is chosen
• Toggled (high-to-low or low-to-high) (CCPxM<3:0> = 1010), the ECCPx pin is not affected;
• Unchanged (that is, reflecting the state of the I/O only the CCPxIF interrupt flag is affected.
latch)
The action on the pin is based on the value of the mode 19.3.4 SPECIAL EVENT TRIGGER
select bits (CCPxM<3:0>). At the same time, the The ECCP module is equipped with a Special Event
interrupt flag bit, CCPxIF, is set. Trigger. This is an internal hardware signal generated
in Compare mode to trigger actions by other modules.
19.3.1 ECCP PIN CONFIGURATION The Special Event Trigger is enabled by selecting
Users must configure the ECCPx pin as an output by the Compare Special Event Trigger mode
clearing the appropriate TRIS bit. (CCPxM<3:0> = 1011).
The Special Event Trigger resets the Timer register pair
Note: Clearing the CCPxCON register will force
for whichever timer resource is currently assigned as the
the ECCPx compare output latch
module’s time base. This allows the CCPRx registers to
(depending on device configuration) to the
serve as a programmable Period register for either timer.
default low level. This is not the PORTx
I/O data latch. The Special Event Trigger can also start an A/D conver-
sion. In order to do this, the A/D Converter must
already be enabled.
FIGURE 19-2: COMPARE MODE OPERATION BLOCK DIAGRAM
0 TMR1H TMR1L
1 TMR3H TMR3L

C1TSEL0 (Timer1/Timer3 Reset, A/D Trigger)
C1TSEL1
C1TSEL2
Set CCP1IF ECCP1 Pin
Compare Output S Q
Comparator
Match Logic
R
TRIS
4 Output Enable
CCPR1H CCPR1L
CCP1CON<3:0>

PIC18F87K90 FAMILY
19.4 PWM (Enhanced Mode) The PWM outputs are multiplexed with I/O pins and are
designated: PxA, PxB, PxC and PxD. The polarity of the
The Enhanced PWM mode can generate a PWM signal PWM pins is configurable and is selected by setting the
on up to four different output pins with up to 10 bits of CCPxM bits in the CCPxCON register appropriately.
resolution. It can do this through four different PWM
Output modes: Table 19-1 provides the pin assignments for each
Enhanced PWM mode.
• Single PWM
Figure 19-3 provides an example of a simplified block
• Half-Bridge PWM
diagram of the Enhanced PWM module.
• Full-Bridge PWM, Forward mode
• Full-Bridge PWM, Reverse mode Note: To prevent the generation of an
incomplete waveform when the PWM is
To select an Enhanced PWM mode, the PxM bits of the first enabled, the ECCP module waits until
CCPxCON register must be set appropriately. the start of a new PWM period before
generating a PWM signal.
FIGURE 19-3: EXAMPLE SIMPLIFIED BLOCK DIAGRAM OF THE ENHANCED PWM MODE
DC1B<1:0> PxM<1:0> CCPxM<3:0>

Duty Cycle Registers
2 4
CCPR1L
ECCPx/PxA ECCP1/Output Pin
TRIS
CCPR1H (Slave)
PxB Output Pin
Output TRIS
Comparator R Q Controller
PxC Output Pin
TMR2 (Note 1)
TRIS
S
PxD Output Pin

Comparator
Clear Timer2, TRIS
Toggle PWM Pin and
PR2 Latch Duty Cycle
ECCP1DEL
Note 1: The 8-bit TMR2 register is concatenated with the 2-bit internal Q clock, or 2 bits of the prescaler to create
the 10-bit time base.
Note: The TRIS register value for each PWM output must be configured appropriately.
Any pin not used by an Enhanced PWM mode is available for alternate pin functions.

PIC18F87K90 FAMILY
TABLE 19-3: EXAMPLE PIN ASSIGNMENTS FOR VARIOUS PWM ENHANCED MODES
ECCP Mode PxM<1:0> PxA PxB PxC PxD
Single 00 Yes(1) Yes(1) Yes(1) Yes(1)
Half-Bridge 10 Yes Yes No No
Full-Bridge, Forward 01 Yes Yes Yes Yes
Full-Bridge, Reverse 11 Yes Yes Yes Yes
Note 1: Outputs are enabled by pulse steering in Single mode (see Register 19-5).
FIGURE 19-4: EXAMPLE PWM (ENHANCED MODE) OUTPUT RELATIONSHIPS

(ACTIVE-HIGH STATE)
Pulse Width PR2 + 1

PxM<1:0> Signal 0
Period
00 (Single Output) PxA Modulated

Delay(1) Delay(1)
PxA Modulated
10 (Half-Bridge) PxB Modulated
PxA Active
(Full-Bridge, PxB Inactive

01 Forward)
PxC Inactive
PxD Modulated
PxA Inactive
(Full-Bridge, PxB Modulated

11
Reverse)
PxC Active
PxD Inactive
Relationships:
• Period = 4 * TOSC * (PR2 + 1) * (TMR2 Prescale Value)
• Pulse Width = TOSC * (CCPRxL<7:0>:CCPxCON<5:4>) * (TMR2 Prescale Value)
• Delay = 4 * TOSC * (ECCPxDEL<6:0>)
Note 1: Dead-band delay is programmed using the ECCPxDEL register (see Section 19.4.6 “Programmable
Dead-Band Delay Mode”).

PIC18F87K90 FAMILY
FIGURE 19-5: EXAMPLE ENHANCED PWM OUTPUT RELATIONSHIPS (ACTIVE-LOW STATE)
Pulse PR2 + 1
PxM<1:0> Signal 0
Width
Period
00 (Single Output) PxA Modulated
PxA Modulated
Delay(1) Delay(1)
10 (Half-Bridge) PxB Modulated
PxA Active
(Full-Bridge, PxB Inactive

01 Forward)
PxC Inactive
PxD Modulated
PxA Inactive
(Full-Bridge, PxB Modulated

11
Reverse)
PxC Active
PxD Inactive
Relationships:
• Period = 4 * TOSC * (PR2 + 1) * (TMR2 Prescale Value)
• Pulse Width = TOSC * (CCPRxL<7:0>:CCPxCON<5:4>) * (TMR2 Prescale Value)
• Delay = 4 * TOSC * (ECCPxDEL<6:0>)
Note 1: Dead-band delay is programmed using the ECCP1DEL register (see Section 19.4.6 “Programmable Dead-Band
Delay Mode”).

PIC18F87K90 FAMILY
19.4.1 HALF-BRIDGE MODE Since the PxA and PxB outputs are multiplexed with the
PORT data latches, the associated TRIS bits must be
In Half-Bridge mode, two pins are used as outputs to
cleared to configure PxA and PxB as outputs.
drive push-pull loads. The PWM output signal is output
on the PxA pin, while the complementary PWM output
signal is output on the PxB pin (see Figure 19-6). This FIGURE 19-6: EXAMPLE OF
mode can be used for half-bridge applications, as HALF-BRIDGE PWM
shown in Figure 19-7, or for full-bridge applications, OUTPUT
where four power switches are being modulated with Period Period
two PWM signals.
Pulse Width
In Half-Bridge mode, the programmable dead-band delay
can be used to prevent shoot-through current in PxA(2)
half-bridge power devices. The value of the PxDC<6:0> td
bits of the ECCPxDEL register sets the number of td
instruction cycles before the output is driven active. If the PxB(2)
value is greater than the duty cycle, the corresponding
output remains inactive during the entire cycle. For more (1) (1) (1)
details on the dead-band delay operations, see
Section 19.4.6 “Programmable Dead-Band Delay td = Dead-Band Delay
Mode”. Note 1: At this time, the TMR2 register is equal to the
PR2 register.
2: Output signals are shown as active-high.
FIGURE 19-7: EXAMPLE OF HALF-BRIDGE APPLICATIONS
Standard Half-Bridge Circuit (“Push-Pull”)
FET
Driver +
PxA
-
Load
FET
Driver
+
PxB
-
Half-Bridge Output Driving a Full-Bridge Circuit
V+
FET FET
Driver Driver
PxA
Load
FET FET
Driver Driver
PxB

PIC18F87K90 FAMILY
19.4.2 FULL-BRIDGE MODE In the Reverse mode, the PxC pin is driven to its active
state and the PxB pin is modulated, while the PxA and
In Full-Bridge mode, all four pins are used as outputs.
PxD pins are driven to their inactive state, as provided
An example of a full-bridge application is provided in
Figure 19-9.
Figure 19-8.
The PxA, PxB, PxC and PxD outputs are multiplexed
In the Forward mode, the PxA pin is driven to its active
with the PORT data latches. The associated TRIS bits
state and the PxD pin is modulated, while the PxB and
must be cleared to configure the PxA, PxB, PxC and
PxC pins are driven to their inactive state, as provided in
PxD pins as outputs.
Figure 19-9.
FIGURE 19-8: EXAMPLE OF FULL-BRIDGE APPLICATION

V+
FET QA QC FET
Driver Driver
PxA
Load
PxB
FET FET
Driver Driver
PxC
QB QD
V-
PxD

PIC18F87K90 FAMILY
FIGURE 19-9: EXAMPLE OF FULL-BRIDGE PWM OUTPUT
Forward Mode
Period
(2)
PxA
Pulse Width
PxB(2)
PxC(2)
PxD(2)
(1) (1)
Reverse Mode
Period
Pulse Width
PxA(2)
PxB(2)
PxC(2)
PxD(2)
(1) (1)
Note 1: At this time, the TMR2 register is equal to the PR2 register.
2: The output signal is shown as active-high.

PIC18F87K90 FAMILY
19.4.2.1 Direction Change in Full-Bridge Mode • The direction of the PWM output changes when
the duty cycle of the output is at or near 100%.
In the Full-Bridge mode, the PxM1 bit in the CCPxCON
register allows users to control the forward/reverse • The turn-off time of the power switch, including
direction. When the application firmware changes this the power device and driver circuit, is greater than
direction control bit, the module will change to the new the turn-on time.
direction on the next PWM cycle. Figure 19-11 shows an example of the PWM direction
A direction change is initiated in software by changing changing from forward to reverse, at a near 100% duty
the PxM1 bit of the CCPxCON register. The following cycle. In this example, at time, t1, the PxA and PxD
sequence occurs prior to the end of the current PWM outputs become inactive, while the PxC output
period: becomes active. Since the turn-off time of the power
devices is longer than the turn-on time, a shoot-through
• The modulated outputs (PxB and PxD) are placed current will flow through power devices, QC and QD
in their inactive state. (see Figure 19-8), for the duration of ‘t’. The same
• The associated unmodulated outputs (PxA and phenomenon will occur to power devices, QA and QB,
PxC) are switched to drive in the opposite for PWM direction change from reverse to forward.
direction.
If changing PWM direction at high duty cycle is required
• PWM modulation resumes at the beginning of the for an application, two possible solutions for eliminating
next period. the shoot-through current are:
For an illustration of this sequence, see Figure 19-10. • Reduce PWM duty cycle for one PWM period
The Full-Bridge mode does not provide a dead-band before changing directions.
delay. As one output is modulated at a time, a • Use switch drivers that can drive the switches off
dead-band delay is generally not required. There is a faster than they can drive them on.
situation where a dead-band delay is required. This
Other options to prevent shoot-through current may
situation occurs when both of the following conditions
exist.
are true:
FIGURE 19-10: EXAMPLE OF PWM DIRECTION CHANGE
Period(1) Period
Signal
PxA (Active-High)
PxB (Active-High)
Pulse Width
PxC (Active-High)
(2)
PxD (Active-High)
Pulse Width
Note 1: The direction bit, PxM1 of the CCPxCON register, is written any time during the PWM cycle.
2: When changing directions, the PxA and PxC signals switch before the end of the current PWM cycle. The
modulated PxB and PxD signals are inactive at this time. The length of this time is:
(1/FOSC) • TMR2 Prescale Value.

PIC18F87K90 FAMILY
FIGURE 19-11: EXAMPLE OF PWM DIRECTION CHANGE AT NEAR 100% DUTY CYCLE(1)
Forward Period t1 Reverse Period
PxA
PxB
PW
PxC
PxD
PW
TON(2)
External Switch C
TOFF(3)
External Switch D
Potential T = TOFF – TON(2,3)

Shoot-Through Current
Note 1: All signals are shown as active-high.

2: TON is the turn-on delay of power switch, QC, and its driver.
3: TOFF is the turn-off delay of power switch, QD, and its driver.
19.4.3 START-UP CONSIDERATIONS complete a full PWM cycle before enabling the PWM
pin output drivers. The completion of a full PWM cycle
When any PWM mode is used, the application
is indicated by the TMR2IF or TMR4IF bit of the PIR1
hardware must use the proper external pull-up and/or
or PIR5 register being set as the second PWM period
pull-down resistors on the PWM output pins.
begins.
Note: When the microcontroller is released from
Reset, all of the I/O pins are in the 19.4.4 ENHANCED PWM
High-Impedance state. The external AUTO-SHUTDOWN MODE
circuits must keep the power switch The PWM mode supports an Auto-Shutdown mode that
devices in the OFF state until the micro- will disable the PWM outputs when an external
controller drives the I/O pins with the shutdown event occurs. Auto-Shutdown mode places
proper signal levels or activates the PWM the PWM output pins into a predetermined state. This
output(s). mode is used to help prevent the PWM from damaging
The CCPxM<1:0> bits of the CCPxCON register allow the application.
the user to choose whether the PWM output signals are The auto-shutdown sources are selected using the
active-high or active-low for each pair of PWM output ECCPxAS<2:0> bits (ECCPxAS<6:4>). A shutdown
pins (PxA/PxC and PxB/PxD). The PWM output event may be generated by:
polarities must be selected before the PWM pin output
• A logic ‘0’ on the pin that is assigned the FLT0
drivers are enabled. Changing the polarity configura-
input function
tion while the PWM pin output drivers are enabled is
not recommended since it may result in damage to the • Comparator C1
application circuits. • Comparator C2
The PxA, PxB, PxC and PxD output latches may not be • Setting the ECCPxASE bit in firmware
in the proper states when the PWM module is A shutdown condition is indicated by the ECCPxASE
initialized. Enabling the PWM pin output drivers at the (Auto-Shutdown Event Status) bit (ECCPxAS<7>). If
same time as the Enhanced PWM modes may cause the bit is a ‘0’, the PWM pins are operating normally. If
damage to the application circuit. The Enhanced PWM the bit is a ‘1’, the PWM outputs are in the shutdown
modes must be enabled in the proper Output mode and state.

PIC18F87K90 FAMILY
When a shutdown event occurs, two things happen: Each pin pair may be placed into one of three states:
• The ECCPxASE bit is set to ‘1’. The ECCPxASE • Drive logic ‘1’
will remain set until cleared in firmware or an • Drive logic ‘0’
auto-restart occurs. (See Section 19.4.5 • Tri-state (high-impedance)
“Auto-Restart Mode”.)
• The enabled PWM pins are asynchronously
placed in their shutdown states. The PWM output
pins are grouped into pairs: PxA/PxC and
PxB/PxD. The state of each pin pair is determined
by the PSSxAC and PSSxBD bits
(ECCPxAS<3:2> and <1:0>, respectively).
REGISTER 19-3: ECCPxAS: ECCPx AUTO-SHUTDOWN CONTROL REGISTER

ECCPxASE ECCPxAS2 ECCPxAS1 ECCPxAS0 PSSxAC1 PSSxAC0 PSSxBD1 PSSxBD0
bit 7 bit 0
Legend:
bit 7 ECCPxASE: ECCP Auto-Shutdown Event Status bit

1 = A shutdown event has occurred; ECCP outputs are in a shutdown state
0 = ECCP outputs are operating
bit 6-4 ECCPxAS<2:0>: ECCP Auto-Shutdown Source Select bits
000 = Auto-shutdown is disabled
001 = Comparator C1OUT output is high
010 = Comparator C2OUT output is high
011 = Either Comparator C1OUT or C2OUT is high
100 = VIL on FLT0 pin
101 = VIL on FLT0 pin or Comparator C1OUT output is high
110 = VIL on FLT0 pin or Comparator C2OUT output is high
111 = VIL on FLT0 pin, Comparator C1OUT or Comparator C2OUT is high
bit 3-2 PSSxAC<1:0>: Pins PxA and PxC Shutdown State Control bits
00 = Drive the PxA and PxC pins to ‘0’
01 = Drive the PxA and PxC pins to ‘1’
1x = PxA and PxC pins tri-state
bit 1-0 PSSxBD<1:0>: Pins PxB and PxD Shutdown State Control bits
00 = Drive the PxB and PxD pins to ‘0’
01 = Drive the PxB and PxD pins to ‘1’
1x = PxB and PxD pins tri-state
Note: The auto-shutdown condition is a level-based signal, not an edge-based signal. As long as the level is
present, the auto-shutdown will persist.
Writing to the ECCPxASE bit is disabled while an auto-shutdown condition persists. Once the
auto-shutdown condition has been removed and the PWM restarted (either through firmware or
auto-restart), the PWM signal will always restart at the beginning of the next PWM period.

PIC18F87K90 FAMILY
FIGURE 19-12: PWM AUTO-SHUTDOWN WITH FIRMWARE RESTART (PxRSEN = 0)
PWM Period
Shutdown Event
ECCPxASE bit
PWM Activity
Normal PWM
ECCPxASE
Cleared by
Start of Firmware
Shutdown Shutdown PWM
PWM Period Event Occurs Event Clears Resumes
19.4.5 AUTO-RESTART MODE The module will wait until the next PWM period begins,
however, before re-enabling the output pin. This behav-
The Enhanced PWM can be configured to automatically
ior allows the auto-shutdown with auto-restart features
restart the PWM signal once the auto-shutdown condi-
to be used in applications based on current mode of
tion has been removed. Auto-restart is enabled by
PWM control.
setting the PxRSEN bit (ECCPxDEL<7>).
If auto-restart is enabled, the ECCPxASE bit will
remain set as long as the auto-shutdown condition is
active. When the auto-shutdown condition is removed,
the ECCPxASE bit will be cleared via hardware and
normal operation will resume.
FIGURE 19-13: PWM AUTO-SHUTDOWN WITH AUTO-RESTART ENABLED (PxRSEN = 1)
PWM Period
Shutdown Event
ECCPxASE bit
PWM Activity
Normal PWM
Start of Shutdown Shutdown PWM

PWM Period Event Occurs Event Clears Resumes

PIC18F87K90 FAMILY
19.4.6 PROGRAMMABLE DEAD-BAND FIGURE 19-14: EXAMPLE OF
DELAY MODE HALF-BRIDGE PWM
In half-bridge applications, where all power switches OUTPUT
are modulated at the PWM frequency, the power Period Period
switches normally require more time to turn off than to
Pulse Width
turn on. If both the upper and lower power switches are
switched at the same time (one turned on and the other PxA(2)
turned off), both switches may be on for a short period td

until one switch completely turns off. During this brief
td
interval, a very high current (shoot-through current) will PxB(2)
flow through both power switches, shorting the bridge
supply. To avoid this potentially destructive (1) (1) (1)
shoot-through current from flowing during switching,
turning on either of the power switches is normally td = Dead-Band Delay
delayed to allow the other switch to completely turn off.
In Half-Bridge mode, a digitally programmable Note 1: At this time, the TMR2 register is equal to the
dead-band delay is available to avoid shoot-through PR2 register.
current from destroying the bridge power switches. The 2: Output signals are shown as active-high.
delay occurs at the signal transition from the non-active
state to the active state. For an illustration, see
Figure 19-14. The lower seven bits of the associated
ECCPxDEL register (Register 19-4) set the delay
period in terms of microcontroller instruction cycles
(TCY or 4 TOSC).
FIGURE 19-15: EXAMPLE OF HALF-BRIDGE APPLICATIONS

V+
Standard Half-Bridge Circuit (“Push-Pull”)
FET
Driver +
PxA V
-
Load
FET
Driver
+
PxB V
-
V-

PIC18F87K90 FAMILY
REGISTER 19-4: ECCPxDEL: ENHANCED PWM CONTROL REGISTER
PxRSEN PxDC6 PxDC5 PxDC4 PxDC3 PxDC2 PxDC1 PxDC0
bit 7 bit 0
Legend:
bit 7 PxRSEN: PWM Restart Enable bit

1 = Upon auto-shutdown, the ECCPxASE bit clears automatically once the shutdown event goes
away; the PWM restarts automatically
0 = Upon auto-shutdown, ECCPxASE must be cleared by software to restart the PWM
bit 6-0 PxDC<6:0>: PWM Delay Count bits
PxDCn = Number of FOSC/4 (4 * TOSC) cycles between the scheduled time when a PWM signal
should transition active and the actual time it does transition active.
19.4.7 PULSE STEERING MODE While the PWM Steering mode is active, the
CCPxM<1:0> bits (CCPxCON<1:0>) select the PWM
In Single Output mode, pulse steering allows any of the
output polarity for the Px<D:A> pins.
PWM pins to be the modulated signal. Additionally, the
same PWM signal can simultaneously be available on The PWM auto-shutdown operation also applies to the
multiple pins. PWM Steering mode, as described in Section 19.4.4
“Enhanced PWM Auto-shutdown mode”. An
Once the Single Output mode is selected
auto-shutdown event will only affect pins that have
(CCPxM<3:2> = 11 and PxM<1:0> = 00 of the
PWM outputs enabled.
CCPxCON register), the user firmware can bring out
the same PWM signal to one, two, three or four output
pins by setting the appropriate STR<D:A> bits
(PSTRxCON<3:0>), as provided in Table 19-3.
Note: The associated TRIS bits must be set to
output (‘0’) to enable the pin output driver
in order to see the PWM signal on the pin.

PIC18F87K90 FAMILY
REGISTER 19-5: PSTRxCON: PULSE STEERING CONTROL(1)
R/W-0 R/W-0 U-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-1
CMPL1 CMPL0 — STRSYNC STRD STRC STRB STRA
bit 7 bit 0
Legend:
bit 7-6 CMPL<1:0>: Complementary Mode Output Assignment Steering Sync bits
00 = See STRD:STRA
01 = PA and PB are selected as the complementary output pair
10 = PA and PC are selected as the complementary output pair
11 = PA and PD are selected as the complementary output pair
bit 4 STRSYNC: Steering Sync bit
1 = Output steering update occurs on the next PWM period
0 = Output steering update occurs at the beginning of the instruction cycle boundary
bit 3 STRD: Steering Enable Bit D
1 = PxD pin has the PWM waveform with polarity control from CCPxM<1:0>
0 = PxD pin is assigned to a PORT pin
bit 2 STRC: Steering Enable Bit C
1 = PxC pin has the PWM waveform with polarity control from CCPxM<1:0>
0 = PxC pin is assigned to a PORT pin
bit 1 STRB: Steering Enable Bit B
1 = PxB pin has the PWM waveform with polarity control from CCPxM<1:0>
0 = PxB pin is assigned to a PORT pin
bit 0 STRA: Steering Enable Bit A
1 = PxA pin has the PWM waveform with polarity control from CCPxM<1:0>
0 = PxA pin is assigned to a PORT pin
Note 1: The PWM Steering mode is available only when the CCPxCON register bits, CCPxM<3:2> = 11 and
PxM<1:0> = 00.

PIC18F87K90 FAMILY
FIGURE 19-16: SIMPLIFIED STEERING 19.4.7.1 Steering Synchronization
BLOCK DIAGRAM The STRSYNC bit of the PSTRxCON register gives the
STRA user two choices for when the steering event will
PxA Signal happen. When the STRSYNC bit is ‘0’, the steering
Output Pin
CCPxM1 1 event will happen at the end of the instruction that
writes to the PSTRxCON register. In this case, the out-
PORT Data
0 put signal at the Px<D:A> pins may be an incomplete
TRIS PWM waveform. This operation is useful when the user
STRB
firmware needs to immediately remove a PWM signal
from the pin.
CCPxM0 1 Output Pin
When the STRSYNC bit is ‘1’, the effective steering
PORT Data update will happen at the beginning of the next PWM
0
TRIS period. In this case, steering on/off the PWM output will
STRC always produce a complete PWM waveform.
Figures 19-17 and 19-18 illustrate the timing diagrams
Output Pin
CCPxM1 1 of the PWM steering depending on the STRSYNC
setting.
PORT Data
0
TRIS
STRD
CCPxM0 1 Output Pin
PORT Data 0
TRIS
Note 1: PORT outputs are configured as displayed
when the CCPxCON register bits,
PxM<1:0> = 00 and CCP1Mx<3:2> = 11.
2: Single PWM output requires setting at least
one of the STRx bits.
FIGURE 19-17: EXAMPLE OF STEERING EVENT AT END OF INSTRUCTION (STRSYNC = 0)
PWM Period
PWM
STRn
P1<D:A> PORT Data PORT Data
P1n = PWM
FIGURE 19-18: EXAMPLE OF STEERING EVENT AT BEGINNING OF INSTRUCTION (STRSYNC = 1)
PWM
STRn
P1<D:A> PORT Data PORT Data
P1n = PWM

PIC18F87K90 FAMILY
19.4.8 OPERATION IN POWER-MANAGED 19.4.8.1 Operation with Fail-Safe
MODES Clock Monitor (FSCM)
In Sleep mode, all clock sources are disabled. If the Fail-Safe Clock Monitor (FSCM) is enabled, a clock
Timer2/4/6/8 will not increment and the state of the failure will force the device into the power-managed
module will not change. If the ECCPx pin is driving a RC_RUN mode and the OSCFIF bit of the PIR2/4/6/8
value, it will continue to drive that value. When the register will be set. The ECCPx will then be clocked from
device wakes up, it will continue from this state. If the internal oscillator clock source, which may have a
Two-Speed Start-ups are enabled, the initial start-up different clock frequency than the primary clock.
frequency from HF-INTOSC and the postscaler may
not be immediately stable. 19.4.9 EFFECTS OF A RESET
In PRI_IDLE mode, the primary clock will continue to Both Power-on Reset and subsequent Resets will force
clock the ECCPx module without change. all ports to Input mode and the ECCP registers to their
Reset states. This forces the ECCP module to reset to
a state compatible with previous, non-Enhanced CCP
modules used on other PIC18 and PIC16 devices.

PIC18F87K90 FAMILY
TABLE 19-4: REGISTERS ASSOCIATED WITH ECCP1/2/3 MODULE AND TIMER1/2/3/4/6/8/10/12
Reset
File
Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Values on
Name
Page:
PIE4 CCP10IE(1) CCP9IE(1) CCP8IE CCP7IE CCP6IE CCP5IE CCP4IE CCP3IE 77
TMR10(1) TMR10 Register 81
TMR12(1) TMR10 Register 81
T10CON(1) — T10OUTPS3 T10OUTPS2 T10OUTPS1 T10OUTPS0 TMR10ON T10CKPS1 T10CKPS0 81
T12CON(1) — T12OUTPS3 T12OUTPS2 T12OUTPS1 T12OUTPS0 TMR12ON T12CKPS1 T12CKPS0 81
CCPR1H Capture/Compare/PWM Register1 High Byte 77
CCPR1L Capture/Compare/PWM Register1 Low Byte 77
CCP1CON P1M1 P1M0 DC1B1 DC1B0 CCP1M3 CCP1M2 CCP1M1 CCP1M0 77
2: Unimplemented in PIC18F6XK90 devices.

PIC18F87K90 FAMILY
20.0 LIQUID CRYSTAL DISPLAY The LCD driver module supports:
(LCD) DRIVER MODULE • Direct driving of LCD panel
• Three LCD clock sources with selectable prescaler
The Liquid Crystal Display (LCD) driver module
generates the timing control to drive a static or • Up to four commons:
multiplexed LCD panel. In the 80-pin devices - Static (One common)
(PIC18F8XK90), the module drives the panels of up to - 1/2 multiplex (two commons)
four commons and up to 48 segments and in the 64-pin - 1/3 multiplex (three commons)
devices (PIC18F6XK90), the module drives the panels - 1/4 multiplex (four commons)
of up to four commons and up to 33 segments. It also
• Up to 48 (in 80-pin devices), 32 (in 64-pin
provides control of the LCD pixel data.
devices) segments
• Static, 1/2 or 1/3 LCD bias
• Internal resistors for bias voltage generation
• Software contrast control for LCD using the
internal biasing
A simplified block diagram of the module is shown in
Figure 20-1.
FIGURE 20-1: LCD DRIVER MODULE BLOCK DIAGRAM
LCDDATAx SE<47:0>
Data Bus Registers 192-to-48
MUX To I/O Pads
24 x 8
(= 4 x 48)
Timing Control
LCDCON COM<3:0>
To I/O Pads
LCDPS
LCDSEx
FOSC/4
SOSC Clock Source

Select and
Prescaler
LF-INTOSC Oscillator

PIC18F87K90 FAMILY
20.1 LCD Registers The LCDCON register, shown in Register 20-1,
controls the overall operation of the module. Once the
The LCD driver module has 32 registers: module is configured, the LCDEN (LCDCON<7>) bit is
• LCD Control Register (LCDCON) used to enable or disable the LCD module. The LCD
• LCD Phase Register (LCDPS) panel can also operate during Sleep by clearing the
SLPEN (LCDCON<6>) bit.
• LCD Reference Ladder Register (LCDRL)
• LCD Reference Voltage Control Register The LCDPS register, shown in Register 20-2,
(LCDREF) configures the LCD clock source prescaler and the type
of waveform, Type-A or Type-B. For details on these
• Six LCD Segment Enable Registers
features, see Section 20.2 “LCD Clock Source
(LCDSE5:LCDSE0)
Selection”, Section 20.3 “LCD Bias Types” and
• 24 LCD Data Registers Section 20.8 “LCD Waveform Generation”.
(LCDDATA23:LCDDATA0)
REGISTER 20-1: LCDCON: LCD CONTROL REGISTER

R/W-0 R/W-0 R/C-0 U-0 R/W-0 R/W-0 R/W-0 R/W-0
LCDEN SLPEN WERR — CS1 CS0 LMUX1 LMUX0
bit 7 bit 0

bit 7 LCDEN: LCD Driver Enable bit

1 = LCD driver module is enabled
0 = LCD driver module is disabled
bit 6 SLPEN: LCD Driver Enable in Sleep mode bit
1 = LCD driver module is disabled in Sleep mode
0 = LCD driver module is enabled in Sleep mode
bit 5 WERR: LCD Write Failed Error bit
1 = LCDDATAx register is written while WA (LCDPS<4>) = 0 (must be cleared in software)
0 = No LCD write error
bit 3-2 CS<1:0>: Clock Source Select bits
00 = (FOSC/4)/8192
01 = SOSC oscillator/32
1x = INTRC (31.25 kHz)/32
bit 1-0 LMUX<1:0>: Commons Select bits
Maximum Maximum
LMUX<1:0> Multiplex Number of Pixels Number of Pixels Bias
(PIC18F6X90) (PIC18F8X90)
00 Static (COM0) 33 48 Static
01 1/2 (COM<1:0>) 66 96 1/2 or 1/3
10 1/3 (COM<2:0>) 99 144 1/2 or 1/3
11 1/4 (COM<3:0>) 132 192 1/3

PIC18F87K90 FAMILY
REGISTER 20-2: LCDPS: LCD PHASE REGISTER
R/W-0 R/W-0 R-0 R-0 R/W-0 R/W-0 R/W-0 R/W-0
WFT BIASMD LCDA WA LP3 LP2 LP1 LP0
bit 7 bit 0
Legend:
bit 7 WFT: Waveform Type Select bit

1 = Type-B waveform (phase changes on each frame boundary)
0 = Type-A waveform (phase changes within each common type)
bit 6 BIASMD: Bias Mode Select bit
When LMUX<1:0> = 00:
0 = Static Bias mode (do not set this bit to ‘1’)
1 = 1/2 Bias mode
0 = 1/3 Bias mode
1 = 1/2 Bias mode
0 = 1/3 Bias mode
0 = 1/3 Bias mode (do not set this bit to ‘1’)
bit 5 LCDA: LCD Active Status bit
1 = LCD driver module is active
0 = LCD driver module is inactive
bit 4 WA: LCD Write Allow Status bit
1 = Write into the LCDDATAx registers is allowed
0 = Write into the LCDDATAx registers is not allowed
bit 3-0 LP<3:0>: LCD Prescaler Select bits
1111 = 1:16
1110 = 1:15
1101 = 1:14
1100 = 1:13
1011 = 1:12
1010 = 1:11
1001 = 1:10
1000 = 1:9
0111 = 1:8
0110 = 1:7
0101 = 1:6
0100 = 1:5
0011 = 1:4
0010 = 1:3
0001 = 1:2
0000 = 1:1

PIC18F87K90 FAMILY
REGISTER 20-3: LCDREF: LCD REFERENCE VOLTAGE CONTROL REGISTER
LCDIRE LCDIRS LCDCST2 LCDCST1 LCDCST0 VLCD3PE VLCD2PE VLCD1PE
bit 7 bit 0
Legend:
bit 7 LCDIRE: LCD Internal Reference Enable bit

1 = Internal LCD reference is enabled and connected to the internal contrast control circuit
0 = Internal LCD reference is disabled
bit 6 LCDIRS: LCD Internal Reference Source bit
If LCDIRE = 1:
1 = Internal LCD contrast control is powered by VDDCORE (3.3V) voltage
0 = Internal LCD contrast control is powered by VDD
If LCDIRE = 0:
Internal LCD contrast control is unconnected. LCD band gap buffer is disabled.
bit 5-3 LCDCST<2:0>: LCD Contrast Control bits
Selects the Resistance of the LCD Contrast Control Resistor Ladder:
111 = Resistor ladder is at maximum resistance (minimum contrast)
110 = Resistor ladder is at 6/7th of maximum resistance
000 = Minimum resistance (maximum contrast); resistor ladder is shorted
bit 2 VLCD3PE: Bias 3 Pin Enable bit
1 = Bias 3 level is connected to the external pin, LCDBIAS3
0 = Bias 3 level is internal (internal resistor ladder)

PIC18F87K90 FAMILY
REGISTER 20-4: LCDRL: LCD REFERENCE LADDER CONTROL REGISTER
R/W-0 R/W-0 R/W-0 R/W-0 U-0 R/W-0 R/W-0 R/W-0
LRLAP1 LRLAP0 LRLBP1 LRLBP0 —(1) LRLAT2 LRLAT1 LRLAT0
bit 7 bit 0
Legend:
bit 7-6 LRLAP<1:0>: LCD Reference Ladder A Time Power Control bits
During Time Interval A:
11 = Internal LCD reference ladder is powered in High-Power mode
10 = Internal LCD reference ladder is powered in Medium Power mode
01 = Internal LCD reference ladder is powered in Low-Power mode
00 = Internal LCD reference ladder is powered down and unconnected
bit 5-4 LRLBP<1:0>: LCD Reference Ladder B Time Power Control bits
During Time Interval B:
11 = Internal LCD reference ladder is powered in High-Power mode
10 = Internal LCD reference ladder is powered in Medium Power mode
01 = Internal LCD reference ladder is powered in Low-Power mode
00 = Internal LCD reference ladder is powered down and unconnected
bit 3 Unimplemented: Read as ‘0’(1)
bit 2-0 LRLAT<2:0>: LCD Reference Ladder A Time Interval Control bits
Sets the number of 32 clock counts when the A Time Interval Power mode is active.
For Type-A Waveforms (WFT = 0):
000 = Internal LCD reference ladder is always in B Power mode
001 = Internal LCD reference ladder is in A Power mode for 1 clock and B Power mode for 15 clocks
010 = Internal LCD reference ladder is in A Power mode for 2 clocks and B Power mode for 14 clocks
For Type-B Waveforms (WFT = 1):
000 = Internal LCD reference ladder is always in B Power mode
001 = Internal LCD reference ladder is in A Power mode for 1 clock and B Power mode for 31 clocks
Note 1: LCDRL<3> should be maintained as ‘0’.

PIC18F87K90 FAMILY
The LCDSE5:LCDSE0 registers configure the Note: The LCDSE5:LCDSE4 registers are not
functions of the port pins. Setting the segment enable implemented in PIC18F6XK90 devices.
bit for a particular segment configures that pin as an
LCD driver. There are six LCD Segment Enable Once the module is initialized for the LCD panel, the
registers, as shown in Table 20-1. The prototype individual bits of the LCDDATA23:LCDDATA0 registers
LCDSEx register is shown in Register 20-5. are cleared or set to represent a clear or dark pixel,
respectively.
TABLE 20-1: LCDSE REGISTERS AND Specific sets of LCDDATA registers are used with
ASSOCIATED SEGMENTS specific segments and common signals. Each bit
represents a unique combination of a specific segment
Register Segments connected to a specific common.
LCDSE0 7:0 (RD<7:0>) Individual LCDDATA bits are named by the convention,
LCDSE1 15:8 (RA<5:4>, RC2, RC5, “SxxCy”, with “xx” as the segment number and “y” as
RB<4:1>) the common number. The relationship is summarized
LCDSE2 23:16 (RF<5:1>, RA1, in Table 20-2. The prototype LCDDATAx register is
RC<4:3>) shown in Register 20-6.
LCDSE3 31:24 (RE7, RB0, RB5, Note: In PIC18F6XK90 devices, writing into the
RC<7:6>, RG4, RF<7:6>) registers, LCDDATA4, LCDDATA5,
LCDSE4 39:32 (RJ<4:7>, RJ<3:1>, RC1) LCDDATA10, LCDDATA11, LCDDATA16,
LCDDATA17, LCDDATA22 and
LCDSE5 47:40 (RH<0:3>, RH<7:4>) LCDDATA23, will not affect the status of
any pixel. These registers can be used as
general purpose registers.
REGISTER 20-5: LCDSEx: LCD SEGMENTx ENABLE REGISTER

SE(n + 7) SE(n + 6) SE(n + 5) SE(n + 4) SE(n + 3) SE(n + 2) SE(n + 1) SE(n)
bit 7 bit 0
Legend:
bit 7-0 SE(n + 7):SE(n): Segment Enable bits

For LCDSE0: n = 0
For LCDSE1: n = 8
For LCDSE2: n = 16
For LCDSE3: n = 24
For LCDSE4: n = 32
For LCDSE5: n = 40
1 = Segment function of the pin is enabled, digital I/O is disabled
0 = I/O function of the pin is enabled

PIC18F87K90 FAMILY
TABLE 20-2: LCDDATA REGISTERS AND BITS FOR SEGMENT AND COM COMBINATIONS
COM Lines
Segments
0 1 2 3
LCDDATA0 LCDDATA6 LCDDATA12 LCDDATA18
0 through 7
S00C0:S07C0 S00C1:S07C1 S00C2:S07C2 S00C3:S07C3
8 through 15
16 through 23
24 through 31
LCDDATA4(1) LCDDATA10(1) LCDDATA16(1) LCDDATA22(1)
32 through 39
LCDDATA5(2) LCDDATA11(2) LCDDATA17(2) LCDDATA23(2)
40 through 47
Note 1: Bits<7:1> of these registers are not implemented in PIC18F6XK90 devices. Bit 0 of these registers
(SEG32Cy) is always implemented.
2: These registers are not implemented in PIC18F6XK90 devices.
REGISTER 20-6: LCDDATAx: LCD DATAx REGISTER

S(n + 7)Cy S(n + 6)Cy S(n + 5)Cy S(n + 4)Cy S(n + 3)Cy S(n + 2)Cy S(n + 1)Cy S(n)Cy
bit 7 bit 0
Legend:
bit 7-0 S(n + 7)Cy:S(n)Cy: Pixel On bits

For registers, LCDDATA0 through LCDDATA5: n = (8x), y = 0
For registers, LCDDATA6 through LCDDATA11: n = (8(x – 6)), y = 1
1 = Pixel on (dark)
0 = Pixel off (clear)

PIC18F87K90 FAMILY
20.2 LCD Clock Source Selection The third clock source is a 31.25 kHz internal RC
oscillator/32 that provides approximately 1 kHz output.
The LCD driver module has three possible clock
sources: The second and third clock sources may be used to
continue running the LCD while the processor is in
• (FOSC/4)/8192 Sleep.
• SOSC Clock/32
These clock sources are selected through the bits
• INTRC/32 CS<1:0> (LCDCON<3:2>).
The first clock source is the system clock divided by
8,192 ((FOSC/4)/8192). This divider ratio is chosen to 20.2.1 LCD PRESCALER
provide about 1 kHz output when the system clock is A 16-bit counter is available as a prescaler for the LCD
8 MHz. The divider is not programmable. Instead, the clock. The prescaler is not directly readable or writable.
LCD prescaler bits, LCDPS<3:0>, are used to set the Its value is set by the LP<3:0> bits (LCDPS<3:0>) that
LCD frame clock rate. determines the prescaler assignment and prescale ratio.
The second clock source is the SOSC oscillator/32. Selectable prescale values are from 1:1 through
This also outputs about 1 kHz when a 32.768 kHz 1:32,768, in power-of-2 increments.
crystal is used with the SOSC oscillator. To use the
SOSC oscillator as a clock source, set the SOSCEN
(T1CON<3>) bit.
FIGURE 20-2: LCD CLOCK GENERATION
COM0
COM1
COM2
COM3
System Clock
(FOSC/4) ÷8192
÷4 STAT
SOSC 32 kHz ÷1, 2, 3, 4
÷32 ÷2 DUP 4-Bit Prog Prescaler
Crystal Oscillator Ring Counter
TRIP
LF-INTOSC Oscillator QUAD
÷32 LP<3:0>
Nom FRC = 31.25 kHz LMUX<1:0>
(LCDPS<3:0>)
(LCDCON<1:0>)
CS<1:0> LMUX<1:0>
(LCDCON<3:2>) (LCDCON<1:0>)

PIC18F87K90 FAMILY
20.3 LCD Bias Types If the internal reference ladder is used to generate bias
voltages, it also can provide software contrast control
The LCD module can be configured in one of three bias (using LCDCST<2:0>). An external resistor ladder can
types: not do this.
• Static bias (two voltage levels: VSS and VDD)
• 1/2 bias (three voltage levels: VSS, 1/2 VDD and 20.3.1 EXTERNAL RESISTOR BIASING
VDD) The external resistor ladder should be connected to the
• 1/3 bias (four voltage levels: VSS, 1/3 VDD, VLCD1 pin (Bias 1), VLCD2 pin (Bias 2), VLCD3 pin
2/3 VDD and VDD) (Bias 3) and VSS. The VLCD3 pin is used to set the
highest voltage to the LCD glass and can be connected
LCD bias voltages can be generated with an internal or
to VDD or a lower voltage.
external resistor ladder. The internal resistor ladder
eliminates the external solution’s use of up to three Figure 20-3 shows the proper way to connect the
pins. resistor ladder to the Bias pins.
FIGURE 20-3: LCD BIAS EXTERNAL RESISTOR LADDER CONNECTION DIAGRAM
Static
1/2 Bias 1/3 Bias
Bias
VLCD0 AVSS AVSS AVSS
VLCD3 To VLCD1 — 1/2 AVDD 1/3 AVDD
VLCD2 LCD
VLCD1 Driver VLCD2 — 1/2 AVDD 2/3 AVDD
VLCD0 VLCD3 AVDD AVDD AVDD
LCD Bias 3 LCD Bias 2 LCD Bias 1 Connections for External R-ladder
AVDD* Static Bias
1/2 Bias
AVDD* 10 k* 10 k*
AVSS
AVDD* 10 k* 10 k* 10 k* 1/3 Bias

AVSS
* These values are provided for design guidance only and should be optimized for the application by the designer.

PIC18F87K90 FAMILY
20.3.2 INTERNAL RESISTOR BIASING Table 20-3 shows the total resistance of each of the
ladders. Figure 20-4 shows the internal resister ladder
This mode does not use external resistors, but rather
connections. When the internal resistor ladder is
internal resistor ladders that are configured to generate
selected, the bias voltage can either be from VDD or
the bias voltage.
from VDDCORE, depending on the LCDIRS setting.
The internal reference ladder actually consists of three
separate ladders. Disabling the internal reference TABLE 20-3: INTERNAL RESISTANCE
ladder disconnects all of the ladders, allowing external LADDER POWER MODES
voltages to be supplied.
Nominal
Depending on the total resistance of the resistor Power Mode Resistance of IDD
ladders, the biasing can be classified as low, medium Entire Ladder
or high power.
Low 3 M 1 A
Medium 300 k 10 A
High 30 k 100 A
FIGURE 20-4: LCD BIAS INTERNAL RESISTOR LADDER CONNECTION DIAGRAM

VVDD
DD
3x VDDCORE
Band Gap
LCDIRS
LCDIRE
LCDCST<2:0>
VLCD3PE
LCDBIAS3
VLCD2PE
LCDBIAS2
VLCD1PE
LCDBIAS1
Low Medium High

Resistor Resistor Resistor
Ladder Ladder Ladder
A Power Mode
B Power Mode
LRLAT<2:0>
LRLAP<1:0> LRLBP<1:0>

PIC18F87K90 FAMILY
There are two power modes designated as “Mode A” 20.3.2.1 Automatic Power Mode Switching
and “Mode B”. Mode A is set by the bits, LRLAP<1:0>
As an LCD segment is electrically only a capacitor, cur-
and Mode B by LRLB<1:0>. The resistor ladder to use
rent is drawn only during the interval when the voltage
for Modes A and B are selected by the bits,
is switching. To minimize total device current, the LCD
LRLAP<1:0> and LRLBP<1:0>, respectively
reference ladder can be operated in a different power
Each ladder has a matching contrast control ladder, mode for the transition portion of the duration. This is
tuned to the nominal resistance of the reference ladder. controlled by the LCDRL register.
This contrast control resistor can be controlled by
Mode A Power mode is active for a programmable time,
LCDREF<5:3> (LCDCST<2:0>). Disabling the internal
beginning at the time when the LCD segment waveform
reference ladder results in all of the ladders being
is transitioning. The LCDRL<2:1> (LRLAT<2:0>) bits
disconnected, allowing external voltages to be
select how long, or if the Mode A is active. Mode B
supplied.
Power mode is active for the remaining time before the
To get additional current in High-Power mode, when segments or commons change again.
LCDRL<7:6> (LRLAP<1:0>) = 11, both the medium
As shown in Figure 20-5, there are 32 counts in a single
and high-power resistor ladders are activated.
segment time. Type-A can be chosen during the time
Whenever the LCD module is inactive (LCDA when the wave form is in transition. Type-B can be
(LCDPS<5>) = 0), the reference ladder will be turned used when the clock is stable or not in transition.
off.
By using this feature of automatic power switching,
using Type-A/Type-B, the power consumption can be
optimized for a given contrast.
FIGURE 20-5: LCD REFERENCE LADDER POWER MODE SWITCHING DIAGRAM
Single Segment Time
lcd_32x_clk
cnt<4:0> 'H00 'H01 'H02 'H03 'H04 'H05 'H06 'H07 'H1E 'H1F 'H00 'H01
lcd_clk
LRLAT<2:0> 'H3
Segment Data
LRLAT<2:0>
Power Mode Power Mode A Power Mode B Mode A

PIC18F87K90 FAMILY
20.3.2.2 Contrast Control
The LCD contrast control circuit consists of a 7-tap
resistor ladder, controlled by the LCDCST bits (see
Figure 20-6.).
FIGURE 20-6: INTERNAL REFERENCE AND CONTRAST CONTROL BLOCK DIAGRAM
VDD 7 Stages
R R R R
Analog
MUX
7
To Top of
Reference Ladder
0
LCDCST<2:0>
3
Internal Reference Contrast Control

PIC18F87K90 FAMILY
20.3.2.3 Internal Reference If the pin is a digital I/O, the corresponding TRIS bit
controls the data direction. If the pin is a COM drive, the
Under firmware control, an internal reference for the
TRIS setting of that pin is overridden.
LCD bias voltages can be enabled. When enabled, the
source of this voltage can be VDD. Note: On a Power-on Reset, the LMUX<1:0>
When no internal reference is selected, the LCD bits are ‘00’.
contrast control circuit is disabled and LCD bias must
be provided externally. Whenever the LCD module is TABLE 20-4: PORTE<6:4> FUNCTION
inactive (LCDA = 0), the internal reference will be
LMUX<1:0> PORTE<6> PORTE<5> PORTE<4>
turned off.
00 Digital I/O Digital I/O Digital I/O
20.3.2.4 VLCDx Pins 01 Digital I/O Digital I/O COM1 Driver
The VLCD3, VLCD2 and VLCD1 pins provide the 10 Digital I/O COM2 Driver COM1 Driver
ability for an external LCD bias network to be used
11 COM3 Driver COM2 Driver COM1 Driver
instead of the internal ladder. Use of the VLCDx pins
does not prevent use of the internal ladder.
20.5 Segment Enables
Each VLCD pin has an independent control in the
LCDREF register, allowing access to any or all of the The LCDSEx registers are used to select the pin
LCD bias signals. function for each segment pin. The selection allows
each pin to operate as either an LCD segment driver or
This architecture allows for maximum flexibility in
different applications. The VLCDx pins could be used a digital only pin. To configure the pin as a segment pin,
the corresponding bits in the LCDSEx registers must
to add capacitors to the internal reference ladder for
be set to ‘1’.
increasing the drive capacity. For applications where
the internal contrast control is insufficient, the firmware If the pin is a digital I/O, the corresponding TRIS bit
can choose to enable only the VLCD3 pin, allowing an controls the data direction. Any bit set in the LCDSEx
external contrast control circuit to use the internal registers overrides any bit settings in the corresponding
reference divider. TRIS register.
Note: On a Power-on Reset, these pins are
20.4 LCD Multiplex Types configured as digital I/O.
The LCD driver module can be configured into four
multiplex types: 20.6 Pixel Control
• Static (only COM0 used) The LCDDATAx registers contain bits that define the
• 1/2 multiplex (COM0 and COM1 are used) state of each pixel. Each bit defines one unique pixel.
• 1/3 multiplex (COM0, COM1 and COM2 are used) Table 20-2 shows the correlation of each bit in the
• 1/4 multiplex (COM0, COM1, COM2 and COM3 are LCDDATAx registers to the respective common and
used) segment signals.
The LMUX<1:0> setting (LCDCON<1:0>) decides the Any LCD pixel location not being used for display can
function of the PORTE<6:4> bits. (For details, see be used as general purpose RAM.
Table 20-4.)

PIC18F87K90 FAMILY
20.7 LCD Frame Frequency 20.8 LCD Waveform Generation
The rate at which the COM and SEG outputs change is LCD waveform generation is based on the philosophy
called the LCD frame frequency. that the net AC voltage across the dark pixel should be
maximized and the net AC voltage across the clear
TABLE 20-5: FRAME FREQUENCY pixel should be minimized. The net DC voltage across
FORMULAS any pixel should be zero.
Multiplex Frame Frequency = The COM signal represents the time slice for each
common, while the SEG contains the pixel data.
Static Clock Source/(4 x 1 x (LP<3:0> + 1))
The pixel signal (COM-SEG) will have no DC
1/2 Clock Source/(2 x 2 x (LP<3:0> + 1)) component and can take only one of the two rms values.
1/3 Clock Source/(1 x 3 x (LP<3:0> + 1)) The higher rms value will create a dark pixel and a lower
1/4 Clock Source/(1 x 4 x (LP<3:0> + 1)) rms value will create a clear pixel.
Note: Clock source is (FOSC/4)/8192, As the number of commons increases, the delta
Timer1 Osc/32 or INTRC/32. between the two rms values decreases. The delta
represents the maximum contrast that the display can
have.
TABLE 20-6: APPROXIMATE FRAME
FREQUENCY (IN Hz) USING The LCDs can be driven by two types of waveforms:
FOSC AT 32 MHz, TIMER1 AT Type-A and Type-B. In a Type-A waveform, the phase
changes within each common type, whereas a Type-B
32.768 kHz OR INTRC OSC
waveform’s phase changes on each frame boundary.
LP<3:0> Static 1/2 1/3 1/4 Thus, Type-A waveforms maintain 0 VDC over a single
frame, whereas Type-B waveforms take two frames.
1 125 125 167 125
2 83 83 111 83 Note 1: If Sleep has to be executed with
LCD Sleep enabled (SLPEN
3 62 62 83 62
(LCDCON<6>) = 1), care must be taken
4 50 50 67 50 to execute Sleep only when VDC on all
5 42 42 56 42 the pixels is ‘0’.
6 36 36 48 36 2: When the LCD clock source is (FOSC/4)/
7 31 31 42 31 8192, if Sleep is executed irrespective of
the LCDCON<SLPEN> setting, the LCD
goes into Sleep. Thus, take care to see
that VDC on all pixels is ‘0’ when Sleep is
executed.
Figure 20-7 through Figure 20-17 provide waveforms
for static, half-multiplex, one-third multiplex and quarter
multiplex drives for Type-A and Type-B waveforms.

PIC18F87K90 FAMILY
FIGURE 20-7: TYPE-A/TYPE-B WAVEFORMS IN STATIC DRIVE
V1
COM0
COM0 V0
V1
SEG0
V0
V1
SEG1
V0
SEG7
SEG6
SEG5
SEG4
SEG3
SEG2
SEG1
SEG0
V1
COM0-SEG0 V0
-V1
COM0-SEG1 V0
1 Frame

PIC18F87K90 FAMILY
FIGURE 20-8: TYPE-A WAVEFORMS IN 1/2 MUX, 1/2 BIAS DRIVE
V2
COM0 V1
V0
COM1
V2
COM0
COM1 V1
V0
V2
SEG0 V1
V0
V2
SEG3
SEG2
SEG1
SEG0
SEG1 V1
V0
V2
V1
COM0-SEG0 V0
-V1
-V2
V2
V1
COM0-SEG1 V0
-V1
-V2
1 Frame

PIC18F87K90 FAMILY
FIGURE 20-9: TYPE-B WAVEFORMS IN 1/2 MUX, 1/2 BIAS DRIVE
V2
COM0 V1
COM1
V0
COM0
V2
COM1 V1
V0
V2
SEG0
V1
V0
V2
SEG1
SEG3
SEG2
SEG1
SEG0
V1
V0
V2
V1
COM0-SEG0 V0
-V1
-V2
V2
V1
COM0-SEG1 V0
-V1
-V2
2 Frames

PIC18F87K90 FAMILY
V3
V2
COM0
V1
COM1
V0
V3
COM0
V2
COM1
V1
V0
V3
V2
SEG0
V1
V0
V3
V2
SEG1
SEG3
SEG2
SEG1
SEG0
V1
V0
V3
V2
V1
COM0-SEG0 V0
-V1
-V2
-V3
V3
V2
V1
COM0-SEG1 V0
-V1
-V2
1 Frame
-V3

PIC18F87K90 FAMILY
V3
V2
COM0
V1
COM1
V0
V3
COM0
V2
COM1
V1
V0
V3
V2
SEG0
V1
V0
V3
V2
SEG1
SEG3
SEG2
SEG1
SEG0
V1
V0
V3
V2
V1
COM0-SEG0 V0
-V1
-V2
-V3
V3
V2
V1
COM0-SEG1 V0
-V1
-V2
2 Frames
-V3

PIC18F87K90 FAMILY
V2
COM0
V1
V0
COM2 V2
COM1 V1
COM1 V0
COM0
V2
COM2 V1
V0
V2
SEG0
V1
SEG2
V0
SEG2
SEG1
SEG0
V2
SEG1 V1
V0
V2
V1
COM0-SEG0 V0
-V1
-V2
V2
V1
COM0-SEG1 V0
-V1
-V2
1 Frame

PIC18F87K90 FAMILY
V2
COM0
V1
V0
COM2
V2
COM1
V1
COM1
V0
COM0
V2
COM2
V1
V0
V2
SEG0
V1
V0
SEG2
SEG1
SEG0
V2
SEG1
V1
V0
V2
V1
COM0-SEG0 V0
-V1
-V2
V2
V1
COM0-SEG1 V0
-V1
-V2
2 Frames

PIC18F87K90 FAMILY
V3
V2
COM0
V1
V0
COM2 V3
V2
COM1
V1
COM1
V0
COM0
V3
V2
COM2
V1
V0
V3
V2
SEG0
SEG2 V1
V0
SEG2
SEG1
SEG0
V3
V2
SEG1
V1
V0
V3
V2
V1
COM0-SEG0 V0
-V1
-V2
-V3
V3
V2
V1
COM0-SEG1 V0
-V1
-V2
-V3
1 Frame

PIC18F87K90 FAMILY
V3
V2
COM0
V1
V0
COM2 V3
V2
COM1
V1
COM1
V0
COM0
V3
V2
COM2
V1
V0
V3
V2
SEG0
V1
V0
SEG2
SEG1
SEG0
V3
V2
SEG1
V1
V0
V3
V2
V1
COM0-SEG0 V0
-V1
-V2
-V3
V3
V2
V1
COM0-SEG1 V0
-V1
-V2
-V3
2 Frames

PIC18F87K90 FAMILY
COM3
V3
COM2 V2
COM0 V1
V0
V3
COM1 V2
COM1 V1
COM0 V0
V3
V2
COM2 V1
V0
V3
V2
COM3 V1
V0
V3
V2
SEG0 V1
V0
SEG1
SEG0
V3
V2
SEG1 V1
V0
V3
V2
V1
COM0-SEG0 V0
-V1
-V2
-V3
V3
V2
V1
COM0-SEG1 V0
-V1
-V2
-V3
1 Frame

PIC18F87K90 FAMILY
COM3
V3
COM2 V2
COM0 V1
V0
V3
COM1 V2
COM1 V1
COM0 V0
V3
V2
COM2 V1
V0
V3
V2
COM3 V1
V0
V3
V2
SEG0 V1
V0
SEG1
SEG0
V3
V2
SEG1 V1
V0
V3
V2
V1
COM0-SEG0 V0
-V1
-V2
-V3
V3
V2
V1
COM0-SEG1 V0
-V1
-V2
-V3
2 Frames

PIC18F87K90 FAMILY
20.9 LCD Interrupts When the LCD driver is running with Type-B waveforms
and the LMUX<1:0> bits are not equal to ‘00’, there are
The LCD timing generation provides an interrupt that some additional issues.
defines the LCD frame timing. This interrupt can be
used to coordinate the writing of the pixel data with the Since the DC voltage on the pixel takes two frames to
start of a new frame, which produces a visually crisp maintain 0V, the pixel data must not change between
transition of the image. subsequent frames. If the pixel data were allowed to
change, the waveform for the odd frames would not
This interrupt can also be used to synchronize external necessarily be the complement of the waveform gener-
events to the LCD. For example, the interface to an ated in the even frames and a DC component would be
external segment driver can be synchronized for introduced into the panel.
segment data updates to the LCD frame.
Because of this, using Type-B waveforms requires
A new frame is defined as beginning at the leading synchronizing the LCD pixel updates to occur within a
edge of the COM0 common signal. The interrupt will be subframe after the frame interrupt.
set immediately after the LCD controller completes
accessing all pixel data required for a frame. This will To correctly sequence writing in Type-B, the interrupt
occur at a fixed interval before the frame boundary only occurs on complete phase intervals. If the user
(TFINT), as shown in Figure 20-18. attempts to write when the write is disabled, the WERR
bit (LCDCON<5>) is set.
The LCD controller will begin to access data for the
next frame within the interval from the interrupt to when Note: The interrupt is not generated when the
the controller begins accessing data after the interrupt Type-A waveform is selected and when
(TFWR). New data must be written within TFWR, as this the Type-B with no multiplex (static) is
is when the LCD controller will begin to access the data selected.
for the next frame.
FIGURE 20-18: EXAMPLE WAVEFORMS AND INTERRUPT TIMING IN QUARTER DUTY

CYCLE DRIVE
LCD Controller Accesses
Interrupt Next Frame Data
Occurs
V3
V2
COM0 V1
V0
V3
V2
COM1 V1
V0
V3
V2
COM2 V1
V0
COM3 V3
V2
V1
V0
2 Frames
TFINT
TFWR Frame
Frame Frame
Boundary Boundary Boundary
TFWR = TFRAME/2*(LMUX<1:0> + 1) + TCY/2

TFINT = (TFWR/2 – (2 TCY + 40 ns)) minimum = 1.5(TFRAME/4) – (2 TCY + 40 ns)
(TFWR/2 – (1 TCY + 40 ns)) maximum = 1.5(TFRAME/4) – (1 TCY + 40 ns)

PIC18F87K90 FAMILY
20.10 Operation During Sleep interrupt can be used to determine the frame boundary.
For the formulas to calculate the delay, see
The LCD module can operate during Sleep. Setting the Section 20.9 “LCD Interrupts”.
SLPEN bit (LCDCON<6>) allows the LCD module to go
to Sleep. Clearing this bit allows the module to continue If a SLEEP instruction is executed and SLPEN = 0, the
operating during Sleep. module will continue to display the current contents of
the LCDDATA registers. The LCD data cannot be
If a SLEEP instruction is executed and SLPEN = 1, the changed.
LCD module will cease all functions and go into a very
low-current consumption mode. The module will stop To allow the module to continue operation while in
operation immediately and drive the minimum LCD volt- Sleep, the clock source must be either the internal RC
age on both segment and common lines. Figure 20-19 oscillator or Timer1 external oscillator.
shows this operation. If the system clock is selected and the module is
The LCD module current consumption will not programmed to not Sleep, the module will ignore the
decrease in this mode, but the overall consumption of SLPEN bit and stop operation immediately. The
the device will be lower due to shut down of the core minimum LCD voltage then will be driven onto the
and other peripheral functions. segments and commons.
To ensure that no DC component is introduced on the Note: The internal RC oscillator or external
panel, the SLEEP instruction should be executed SOSC oscillator must be used to operate
immediately after an LCD frame boundary. The LCD the LCD module during Sleep.
FIGURE 20-19: SLEEP ENTRY/EXIT WHEN SLPEN = 1 OR CS<1:0> = 00

V3
V2
V1
COM0 V0
V3
V2
V1
COM1 V0
V3
V2
V1
COM2 V0
V3
V2
V1
SEG0 V0
2 Frames
SLEEP Instruction Execution Wake-up

PIC18F87K90 FAMILY
20.11 Configuring the LCD Module 4. Configure the following LCD module functions
using the LCDCON register:
To configure the LCD module.
• Multiplex and Bias mode – LMUX<1:0> bits
1. Select the frame clock prescale, using bits, • Timing Source – CS<1:0> bits
LP<3:0> (LCDPS<3:0>).
• Sleep mode – SLPEN bit
2. Configure the appropriate pins to function as
5. Write initial values to the pixel data registers,
segment drivers using the LCDSEx registers.
LCDDATA0 through LCDDATA23.
3. If using the internal reference resistors for
6. Clear the LCD Interrupt Flag, LCDIF (PIR3<6>),
biasing, enable the internal reference ladder
and if desired, enable the interrupt by setting bit,
and:
LCDIE (PIE3<6>).
• Define the Mode A and Mode B interval by
7. Enable the LCD module by setting bit, LCDEN
using the LRLAT<2:0> bits (LCDRL<2:0>)
(LCDCON<7>).
• Define the low, medium or high ladder for
Mode A and Mode B by using the LRLAP<1:0>
bits (LCDRL<7:6>) and the LRLBP<1:0> bits
(LCDRL<5:4>), respectively
• Set the VLCDxPE bits and enable the
LCDIRE bit (LCDREF<7>)

PIC18F87K90 FAMILY
TABLE 20-7: REGISTERS ASSOCIATED WITH LCD OPERATION
Reset
on Page:
(1)
LCDDATA23 S47C3 S46C3 S45C3 S44C3 S43C3 S42C3 S41C3 S40C3 79
LCDDATA22(1) S39C3 S38C3 S37C3 S36C3 S35C3 S34C3 S33C3 S32C3 79
(1)
(1)
(1)
(2)
LCDSE4(2) SE39 SE38 SE37 SE36 SE35 SE34 SE33 SE32 83
LCDCON LCDEN SLPEN WERR — CS1 CS0 LMUX1 LMUX0 83
LCDPS WFT BIASMD LCDA WA LP3 LP2 LP1 LP0 83
LCDREF LCDIRE LCDIRS LCDCST2 LCDCST1 LCDCST0 VLCD3PE VLCD2PE VLCD1PE 83
LCDRL LRLAP1 LRLAP0 LRLBP1 LRLBP0 — LRLAT2 LRLAT1 LRLAT0 83
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used for LCD operations.
Note 1: These registers are implemented, but unused on 64-pin devices, and may be used as general purpose data
RAM.
2: These registers are unimplemented in 64-pin devices.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
21.0 MASTER SYNCHRONOUS 21.3 SPI Mode
SERIAL PORT (MSSP) The SPI mode allows 8 bits of data to be synchronously
MODULE transmitted and received simultaneously. All four
modes of SPI are supported. To accomplish
21.1 Master SSP (MSSP) Module communication, typically three pins are used:
Overview • Serial Data Out (SDOx) – RC5/SDO1/SEG12 or
RD4/SEG4/SDO2
The Master Synchronous Serial Port (MSSP) module is
• Serial Data In (SDIx) – RC4/SDI1/SDA1/SEG16
a serial interface, useful for communicating with other
or RD5/SEG5/SDI2/SDA2
peripheral or microcontroller devices. These peripheral
devices may be serial EEPROMs, shift registers, • Serial Clock (SCKx) – RC3/SCK1/SCL1/SEG17
display drivers, A/D Converters, etc. The MSSP or RD6/SEG6/SCK2/SCL2
module can operate in one of two modes: Additionally, a fourth pin may be used when in a Slave
• Serial Peripheral Interface (SPI) mode of operation:
• Inter-Integrated Circuit (I2C™) • Slave Select (SSx) – RF7/AN5/SS1/SEG25 or
- Full Master mode RD7/SEG7/SS2
- Slave mode (with general address call) Figure 21-1 shows the block diagram of the MSSP
module when operating in SPI mode.
The I2C interface supports the following modes in
hardware:
FIGURE 21-1: MSSPx BLOCK DIAGRAM
• Master mode (SPI MODE)
• Multi-Master mode
Internal
• Slave mode with 5-bit and 7-bit address masking Data Bus
(with address masking for both 10-bit and 7-bit
Read Write
addressing)
All members of the PIC18F87K90 family have two SSPxBUF reg
MSSP modules, designated as MSSP1 and MSSP2.
Each module operates independently of the other.
Note: Throughout this section, generic refer- SDIx
ences to an MSSP module in any of its SSPxSR reg
operating modes may be interpreted as SDOx bit 0 Shift
being equally applicable to MSSP1 or Clock
MSSP2. Register names and module I/O
signals use the generic designator ‘x’ to
indicate the use of a numeral to distinguish
a particular module when required. Control
SSx SSx Control
bit names are not individuated.
Enable
21.2 Control Registers Edge

Select
Each MSSP module has three associated control regis-
ters. These include a status register (SSPxSTAT) and 2
two control registers (SSPxCON1 and SSPxCON2). The Clock Select
use of these registers and their individual configuration
SSPM<3:0>
bits differ significantly depending on whether the MSSP
module is operated in SPI or I2C mode.
SCKx
SMP:CKE 4
2 (
TMR2 Output
2
)
Additional details are provided under the individual Edge
sections. Select Prescaler TOSC
4, 16, 64
Note: In devices with more than one MSSP
module, it is very important to pay close Data to TXx/RXx in SSPxSR
attention to SSPxCON register names. TRIS bit
SSP1CON1 and SSP1CON2 control
different operational aspects of the same Note: Only port I/O names are used in this diagram for
module, while SSP1CON1 and the sake of brevity. Refer to the text for a full list of
SSP2CON1 control the same features for multiplexed functions.
two different modules.

PIC18F87K90 FAMILY
21.3.1 REGISTERS In receive operations, SSPxSR and SSPxBUF
together create a double-buffered receiver. When
Each MSSP module has four registers for SPI mode
SSPxSR receives a complete byte, it is transferred to
operation. These are:
SSPxBUF and the SSPxIF interrupt is set.
• MSSPx Control Register 1 (SSPxCON1)
During transmission, the SSPxBUF is not
• MSSPx Status Register (SSPxSTAT) double-buffered. A write to SSPxBUF will write to both
• Serial Receive/Transmit Buffer Register SSPxBUF and SSPxSR.
(SSPxBUF)
Note: The SSPxBUF register cannot be used
• MSSPx Shift Register (SSPxSR) – Not directly
with read-modify-write instructions, such
accessible
as BCF, COMF, etc.
SSPxCON1 and SSPxSTAT are the control and status
To avoid lost data in Master mode, a read of
registers in SPI mode operation. The SSPxCON1
the SSPxBUF must be performed to clear the
register is readable and writable. The lower 6 bits of
Buffer Full (BF) detect bit (SSPxSTAT<0>)
the SSPxSTAT are read-only. The upper two bits of the
between each transmission.
SSPxSTAT are read/write.
SSPxSR is the shift register used for shifting data in or
out. SSPxBUF is the buffer register to which data
bytes are written to or read from.
REGISTER 21-1: SSPxSTAT: MSSPx STATUS REGISTER (SPI MODE)

R/W-0 R/W-0 R-0 R-0 R-0 R-0 R-0 R-0
(1)
SMP CKE D/A P S R/W UA BF
bit 7 bit 0
Legend:
bit 7 SMP: Sample bit

SPI Master mode:
1 = Input data sampled at the end of data output time
0 = Input data sampled at the middle of data output time
SPI Slave mode:
SMP must be cleared when SPI is used in Slave mode.
bit 6 CKE: SPI Clock Select bit(1)
1 = Transmit occurs on transition from active to Idle clock state
0 = Transmit occurs on transition from Idle to active clock state
bit 5 D/A: Data/Address bit
Used in I2C™ mode only.
bit 4 P: Stop bit
Used in I2C mode only. This bit is cleared when the MSSPx module is disabled; SSPEN is cleared.
bit 3 S: Start bit
Used in I2C mode only.
bit 2 R/W: Read/Write Information bit
bit 1 UA: Update Address bit
bit 0 BF: Buffer Full Status bit (Receive mode only)
1 = Receive complete, SSPxBUF is full
0 = Receive not complete, SSPxBUF is empty
Note 1: Polarity of the clock state is set by the CKP bit (SSPxCON1<4>).

PIC18F87K90 FAMILY
REGISTER 21-2: SSPxCON1: MSSPx CONTROL REGISTER 1 (SPI MODE)
WCOL SSPOV(1) SSPEN(2) CKP SSPM3(3) SSPM2(3) SSPM1(3) SSPM0(3)
bit 7 bit 0
Legend:
bit 7 WCOL: Write Collision Detect bit

1 = The SSPxBUF register is written while it is still transmitting the previous word (must be cleared in
software)
0 = No collision
bit 6 SSPOV: Receive Overflow Indicator bit(1)
SPI Slave mode:
1 = A new byte is received while the SSPxBUF register is still holding the previous data. In case of over-
flow, the data in SSPxSR is lost. Overflow can only occur in Slave mode. The user must read the
SSPxBUF, even if only transmitting data, to avoid setting overflow (must be cleared in software).
0 = No overflow
bit 5 SSPEN: Master Synchronous Serial Port Enable bit(2)
1 = Enables serial port and configures SCKx, SDOx, SDIx and SSx as serial port pins
0 = Disables serial port and configures these pins as I/O port pins
bit 4 CKP: Clock Polarity Select bit
1 = Idle state for clock is a high level
0 = Idle state for clock is a low level
bit 3-0 SSPM<3:0>: Master Synchronous Serial Port Mode Select bits(3)
1010 = SPI Master mode: Clock = FOSC/8
0101 = SPI Slave mode: Clock = SCKx pin; SSx pin control is disabled; SSx can be used as an I/O pin
0100 = SPI Slave mode: Clock = SCKx pin; SSx pin control is enabled
0011 = SPI Master mode: Clock = TMR2 Output/2
Note 1: In Master mode, the overflow bit is not set since each new reception (and transmission) is initiated by
writing to the SSPxBUF register.
2: When enabled, these pins must be properly configured as inputs or outputs.
3: Bit combinations not specifically listed here are either reserved or implemented in I2C™ mode only.

PIC18F87K90 FAMILY
21.3.2 OPERATION When the application software is expecting to receive
valid data, the SSPxBUF should be read before the next
When initializing the SPI, several options need to be
byte of data to transfer is written to the SSPxBUF. The
specified. This is done by programming the appropriate
Buffer Full bit, BF (SSPxSTAT<0>), indicates when
control bits (SSPxCON1<5:0> and SSPxSTAT<7:6>).
SSPxBUF has been loaded with the received data
These control bits allow the following to be specified:
(transmission is complete). When the SSPxBUF is read,
• Master mode (SCKx is the clock output) the BF bit is cleared. This data may be irrelevant if the
• Slave mode (SCKx is the clock input) SPI is only a transmitter. Generally, the MSSP interrupt
• Clock Polarity (Idle state of SCKx) is used to determine when the transmission/reception
• Data Input Sample Phase (middle or end of data has completed. If the interrupt method is not going to be
output time) used, then software polling can be done to ensure that a
write collision does not occur. Example 21-1 shows the
• Clock Edge (output data on rising/falling edge of
loading of the SSPxBUF (SSPxSR) for data
SCKx)
transmission.
• Clock Rate (Master mode only)
The SSPxSR is not directly readable or writable and
• Slave Select mode (Slave mode only)
can only be accessed by addressing the SSPxBUF
Each MSSP module consists of a Transmit/Receive register. Additionally, the SSPxSTAT register indicates
Shift register (SSPxSR) and a Serial Receive Transmit the various status conditions.
Buffer register (SSPxBUF). The SSPxSR shifts the
data in and out of the device, MSb first. The SSPxBUF 21.3.3 OPEN-DRAIN OUTPUT OPTION
holds the data that was written to the SSPxSR until the The drivers for the SDOx output and SCKx clock pins
received data is ready. Once the 8 bits of data have can be optionally configured as open-drain outputs.
been received, that byte is moved to the SSPxBUF This feature allows the voltage level on the pin to be
register. Then, the Buffer Full detect bit, BF pulled to a higher level through an external pull-up
(SSPxSTAT<0>), and the interrupt flag bit, SSPxIF, are resistor, and allows the output to communicate with
set. This double-buffering of the received data external circuits without the need for additional level
(SSPxBUF) allows the next byte to start reception before shifters. For more information, see Section 11.1.3
reading the data that was just received. Any write to the “Open-Drain Outputs”.
SSPxBUF register during transmission/reception of data
will be ignored and the Write Collision Detect bit, WCOL The open-drain output option is controlled by the
(SSPxCON1<7>), will be set. User software must clear SSP2OD (ODCON1<0>) and SSP1OD bits
the WCOL bit so that it can be determined if the following (ODCON1<7>). Setting an SSPxOD bit configures the
write(s) to the SSPxBUF register completed SDOx and SCKx pins for the corresponding module for
successfully. open-drain operation.
Note: To avoid lost data in Master mode, a
read of the SSPxBUF must be per-
formed to clear the Buffer Full (BF)
detect bit (SSPxSTAT<0>) between
each transmission.
EXAMPLE 21-1: LOADING THE SSP1BUF (SSP1SR) REGISTER

LOOP BTFSS SSP1STAT, BF ;Has data been received (transmit complete)?
BRA LOOP ;No
MOVF SSP1BUF, W ;WREG reg = contents of SSP1BUF
MOVWF RXDATA ;Save in user RAM, if data is meaningful
MOVF TXDATA, W ;W reg = contents of TXDATA
MOVWF SSP1BUF ;New data to xmit

PIC18F87K90 FAMILY
21.3.4 ENABLING SPI I/O Any serial port function that is not desired may be
overridden by programming the corresponding Data
To enable the serial port, MSSP Enable bit, SSPEN
Direction (TRIS) register to the opposite value.
(SSPxCON1<5>), must be set. To reset or reconfigure
SPI mode, clear the SSPEN bit, reinitialize the
21.3.5 TYPICAL CONNECTION
SSPxCON registers and then set the SSPEN bit. This
configures the SDIx, SDOx, SCKx and SSx pins as Figure 21-2 shows a typical connection between two
serial port pins. For the pins to behave as the serial port microcontrollers. The master controller (Processor 1)
function, some must have their data direction bits (in initiates the data transfer by sending the SCKx signal.
the TRIS register) appropriately programmed as Data is shifted out of both shift registers on their pro-
follows: grammed clock edge and latched on the opposite edge
of the clock. Both processors should be programmed to
• SDIx must have TRISC<4> or TRISD<5> bit set
the same Clock Polarity (CKP), then both controllers
• SDOx must have the TRISC<5> or TRISD<4> bit would send and receive data at the same time.
cleared Whether the data is meaningful (or dummy data)
• SCKx (Master mode) must have the TRISC<3> or depends on the application software. This leads to
TRISD<6>bit cleared three scenarios for data transmission:
• SCKx (Slave mode) must have the TRISC<3> or • Master sends data–Slave sends dummy data
TRISD<6> bit set
• Master sends data–Slave sends data
• SSx must have the TRISF<7> or TRISD<7> bit set
• Master sends dummy data–Slave sends data
FIGURE 21-2: SPI MASTER/SLAVE CONNECTION
SPI Master SSPM<3:0> = 00xxb SPI Slave SSPM<3:0> = 010xb
SDOx SDIx
Serial Input Buffer Serial Input Buffer

(SSPxBUF) (SSPxBUF)
SDIx SDOx
Shift Register Shift Register
(SSPxSR) (SSPxSR)
MSb LSb MSb LSb
Serial Clock
SCKx SCKx
PROCESSOR 1 PROCESSOR 2

PIC18F87K90 FAMILY
21.3.6 MASTER MODE shown in Figure 21-3, Figure 21-5 and Figure 21-6,
where the MSB is transmitted first. In Master mode, the
The master can initiate the data transfer at any time
SPI clock rate (bit rate) is user-programmable to be one
because it controls the SCKx. The master determines
of the following:
when the slave (Processor 1, Figure 21-2) is to
broadcast data by the software protocol. • FOSC/4 (or TCY)
In Master mode, the data is transmitted/received as • FOSC/16 (or 4 • TCY)
soon as the SSPxBUF register is written to. If the SPI • FOSC/64 (or 16 • TCY)
is only going to receive, the SDOx output could be dis- • Timer2 output/2
abled (programmed as an input). The SSPxSR register
This allows a maximum data rate (at 40 MHz) of
will continue to shift in the signal present on the SDIx
10.00 Mbps.
pin at the programmed clock rate. As each byte is
received, it will be loaded into the SSPxBUF register as Figure 21-3 shows the waveforms for Master mode.
if a normal received byte (interrupts and status bits When the CKE bit is set, the SDOx data is valid before
appropriately set). This could be useful in receiver there is a clock edge on SCKx. The change of the input
applications as a “Line Activity Monitor” mode. sample is shown based on the state of the SMP bit. The
time when the SSPxBUF is loaded with the received
The clock polarity is selected by appropriately
data is shown.
programming the CKP bit (SSPxCON1<4>). This, then,
would give waveforms for SPI communication as
FIGURE 21-3: SPI MODE WAVEFORM (MASTER MODE)
Write to
SSPxBUF
SCKx
(CKP = 0
CKE = 0)
SCKx
(CKP = 1
CKE = 0)
4 Clock
SCKx Modes
(CKP = 0
CKE = 1)
SCKx
(CKP = 1
CKE = 1)
SDOx bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0

(CKE = 0)

(CKE = 1)
SDIx
(SMP = 0) bit 7 bit 0
Input
Sample
(SMP = 0)
SDIx
(SMP = 1) bit 0
bit 7
Input
Sample
(SMP = 1)
SSPxIF
Next Q4 Cycle
SSPxSR to after Q2
SSPxBUF

PIC18F87K90 FAMILY
21.3.7 SLAVE MODE transmitted byte and becomes a floating output.
External pull-up/pull-down resistors may be desirable
In Slave mode, the data is transmitted and received as
depending on the application.
the external clock pulses appear on SCKx. When the
last bit is latched, the SSPxIF interrupt flag bit is set. Note 1: When the SPI is in Slave mode, with
While in Slave mode, the external clock is supplied by the SSx pin control enabled
the external clock source on the SCKx pin. This exter- (SSPxCON1<3:0> = 0100), the SPI
nal clock must meet the minimum high and low times module will reset if the SSx pin is set to
as specified in the electrical specifications. VDD.
While in Sleep mode, the slave can transmit/receive 2: If the SPI is used in Slave mode, with CKE
data. When a byte is received, the device can be set, then the SSx pin control must be
configured to wake-up from Sleep. enabled.
When the SPI module resets, the bit counter is forced
21.3.8 SLAVE SELECT
to ‘0’. This can be done by either forcing the SSx pin to
SYNCHRONIZATION a high level or clearing the SSPEN bit.
The SSx pin allows a Synchronous Slave mode. The To emulate two-wire communication, the SDOx pin can
SPI must be in Slave mode with the SSx pin control be connected to the SDIx pin. When the SPI needs to
enabled (SSPxCON1<3:0> = 04h). When the SSx pin operate as a receiver, the SDOx pin can be configured
is low, transmission and reception are enabled and the as an input. This disables transmissions from the
SDOx pin is driven. When the SSx pin goes high, the SDOx. The SDIx can always be left as an input (SDIx
SDOx pin is no longer driven, even if in the middle of a function) since it cannot create a bus conflict.
FIGURE 21-4: SLAVE SYNCHRONIZATION WAVEFORM
SSx
SCKx
(CKP = 0
CKE = 0)
SCKx
(CKP = 1
CKE = 0)
Write to
SSPxBUF
SDOx bit 7 bit 6 bit 7 bit 0
SDIx bit 0
Input
Sample
(SMP = 0)
SSPxIF
Interrupt
Flag
Next Q4 Cycle
SSPxBUF

PIC18F87K90 FAMILY
FIGURE 21-5: SPI MODE WAVEFORM (SLAVE MODE WITH CKE = 0)
SSx
Optional
SCKx
(CKP = 0
CKE = 0)
SCKx
(CKP = 1
CKE = 0)
Write to
SSPxBUF
SDIx
Input
Sample
(SMP = 0)
SSPxIF
Interrupt
Flag
Next Q4 Cycle
SSPxBUF
FIGURE 21-6: SPI MODE WAVEFORM (SLAVE MODE WITH CKE = 1)
SSx
Not Optional
SCKx
(CKP = 0
CKE = 1)
SCKx
(CKP = 1
CKE = 1)
Write to
SSPxBUF
SDIx
(SMP = 0) bit 0
bit 7
Input
Sample
(SMP = 0)
SSPxIF
Interrupt
Flag
Next Q4 Cycle
after Q2
SSPxSR to
SSPxBUF

PIC18F87K90 FAMILY
21.3.9 OPERATION IN POWER-MANAGED 21.3.11 BUS MODE COMPATIBILITY
MODES Table 21-1 shows the compatibility between the
In SPI Master mode, module clocks may be operating standard SPI modes and the states of the CKP and
at a different speed than when in full-power mode; in CKE control bits.
the case of the Sleep mode, all clocks are halted.
TABLE 21-1: SPI BUS MODES
In Idle modes, a clock is provided to the peripherals.
That clock can be from the primary clock source, the Standard SPI Mode Control Bits State
secondary clock (SOSC oscillator) or the INTOSC Terminology
source. See Section 3.3 “Clock Sources and CKP CKE
Oscillator Switching” for additional information. 0, 0 0 1
In most cases, the speed that the master clocks SPI 0, 1 0 0
data is not important; however, this should be 1, 0 1 1
evaluated for each system.
1, 1 1 0
If MSSP interrupts are enabled, they can wake the
controller from Sleep mode, or one of the Idle modes, There is also an SMP bit which controls when the data
when the master completes sending data. If an exit is sampled.
from Sleep or Idle mode is not desired, MSSP
interrupts should be disabled. 21.3.12 SPI CLOCK SPEED AND MODULE
INTERACTIONS
If the Sleep mode is selected, all module clocks are
halted and the transmission/reception will remain in Because MSSP1 and MSSP2 are independent
that state until the device wakes. After the device modules, they can operate simultaneously at different
returns to Run mode, the module will resume data rates. Setting the SSPM<3:0> bits of the
transmitting and receiving data. SSPxCON1 register determines the rate for the
corresponding module.
In SPI Slave mode, the SPI Transmit/Receive Shift
register operates asynchronously to the device. This An exception is when both modules use Timer2 as a
allows the device to be placed in any power-managed time base in Master mode. In this instance, any
mode and data to be shifted into the SPI changes to the Timer2 module’s operation will affect
Transmit/Receive Shift register. When all 8 bits have both MSSP modules equally. If different bit rates are
been received, the MSSP interrupt flag bit will be set, required for each module, the user should select one of
and if enabled, will wake the device. the other three time base options for one of the
modules.
21.3.10 EFFECTS OF A RESET
A Reset disables the MSSP module and terminates the
current transfer.

PIC18F87K90 FAMILY
TABLE 21-2: REGISTERS ASSOCIATED WITH SPI OPERATION
Reset
on Page:
IPR2 OSCFIP — SSP2IP BCL2IP BCL1IP HLVDIP TMR3IP TMR3GIP 77
SSP1BUF MSSP1 Receive Buffer/Transmit Register 82
SSP1CON1 WCOL SSPOV SSPEN CKP SSPM3 SSPM2 SSPM1 SSPM0 76
SSP1CON2 GCEN ACKSTAT ACKDT ACKEN RCEN PEN RSEN SEN 76
SSP1STAT SMP CKE D/A P S R/W UA BF 76
SSP2CON2 GCEN ACKSTAT ACKDT ACKEN RCEN PEN RSEN SEN 83
ODCON3 U2OD U1OD — — — — — CTMUDS 81
Legend: Shaded cells are not used by the MSSP module in SPI mode.

PIC18F87K90 FAMILY
21.4 I2C™ Mode 21.4.1 REGISTERS
The MSSP module in I 2C mode fully implements all The MSSP module has seven registers for I2C
master and slave functions (including general call operation. These are:
support), and provides interrupts on Start and Stop bits • MSSPx Control Register 1 (SSPxCON1)
in hardware to determine a free bus (multi-master • MSSPx Control Register 2 (SSPxCON2)
function). The MSSP module implements the standard
• MSSPx Status Register (SSPxSTAT)
mode specifications, as well as 7-bit and 10-bit
addressing. • Serial Receive/Transmit Buffer Register
(SSPxBUF)
Two pins are used for data transfer:
• MSSPx Shift Register (SSPxSR) – Not directly
• Serial Clock (SCLx) – RC3/SCK1/SCL1/SEG17 accessible
or RD6/SEG6/SCK2/SCL2 • MSSPx Address Register (SSPxADD)
• Serial Data (SDAx) – RC4/SDI1/SDA1/SEG16 or • I2C Slave Address Mask Register (SSPxMSK)
RD5/SEG5/SDI2/SDA2
SSPxCON1, SSPxCON2 and SSPxSTAT are the
The user must configure these pins as inputs by setting control and status registers in I2C mode operation. The
the associated TRIS bits. SSPxCON1 and SSPxCON2 registers are readable and
writable. The lower 6 bits of the SSPxSTAT are
FIGURE 21-7: MSSPx BLOCK DIAGRAM read-only. The upper two bits of the SSPxSTAT are
(I2C™ MODE) read/write.
SSPxSR is the shift register used for shifting data in or
Internal
Data Bus out. SSPxBUF is the buffer register to which data
bytes are written to or read from.
Read Write
SSPxADD contains the slave device address when the
SSPxBUF reg
MSSP is configured in I2C Slave mode. When the
SCLx MSSP is configured in Master mode, the lower seven
bits of SSPxADD act as the Baud Rate Generator
Shift
Clock reload value.
SSPxSR reg SSPxMSK holds the slave address mask value when
SDAx MSb LSb the module is configured for 7-Bit Address Masking
mode. While it is a separate register, it shares the same
SFR address as SSPxADD. It is only accessible when
Match Detect Addr Match
the SSPM<3:0> bits are specifically set to permit
Address Mask access. Additional details are provided in
Section 21.4.3.4 “7-Bit Address Masking Mode”.
SSPxADD reg In receive operations, SSPxSR and SSPxBUF
together, create a double-buffered receiver. When
SSPxSR receives a complete byte, it is transferred to
SSPxBUF and the SSPxIF interrupt is set.
Start and Set, Reset
Stop bit Detect S, P bits During transmission, the SSPxBUF is not
(SSPxSTAT reg) double-buffered. A write to SSPxBUF will write to both
SSPxBUF and SSPxSR.
Note: Only port I/O names are used in this diagram for
the sake of brevity. Refer to the text for a full list of
multiplexed functions.

PIC18F87K90 FAMILY
REGISTER 21-3: SSPxSTAT: MSSPx STATUS REGISTER (I2C™ MODE)
R/W-0 R/W-0 R-0 R-0 R-0 R-0 R-0 R-0
SMP CKE D/A P(1) S(1) R/W(2,3) UA BF
bit 7 bit 0
Legend:
bit 7 SMP: Slew Rate Control bit

In Master or Slave mode:
1 = Slew rate control is disabled for Standard Speed mode (100 kHz and 1 MHz)
0 = Slew rate control is enabled for High-Speed mode (400 kHz)
bit 6 CKE: SMBus Select bit
In Master or Slave mode:
1 = Enable SMBus-specific inputs
0 = Disable SMBus-specific inputs
bit 5 D/A: Data/Address bit
In Master mode:
Reserved.
In Slave mode:
1 = Indicates that the last byte received or transmitted was data
0 = Indicates that the last byte received or transmitted was an address
bit 4 P: Stop bit(1)
1 = Indicates that a Stop bit has been detected last
0 = Stop bit was not detected last
bit 3 S: Start bit(1)
1 = Indicates that a Start bit has been detected last
0 = Start bit was not detected last
bit 2 R/W: Read/Write Information bit(2,3)
In Slave mode:
1 = Read
0 = Write
In Master mode:
1 = Transmit is in progress
0 = Transmit is not in progress
bit 1 UA: Update Address bit (10-Bit Slave mode only)
1 = Indicates that the user needs to update the address in the SSPxADD register
0 = Address does not need to be updated
bit 0 BF: Buffer Full Status bit
In Transmit mode:
1 = SSPxBUF is full
0 = SSPxBUF is empty
In Receive mode:
1 = SSPxBUF is full (does not include the ACK and Stop bits)
0 = SSPxBUF is empty (does not include the ACK and Stop bits)
Note 1: This bit is cleared on Reset and when SSPEN is cleared.

2: This bit holds the R/W bit information following the last address match. This bit is only valid from the
address match to the next Start bit, Stop bit or not ACK bit.
3: ORing this bit with SEN, RSEN, PEN, RCEN or ACKEN will indicate if the MSSPx is in Active mode.

PIC18F87K90 FAMILY
REGISTER 21-4: SSPxCON1: MSSPx CONTROL REGISTER 1 (I2C™ MODE)
(1)
WCOL SSPOV SSPEN CKP SSPM3(2) SSPM2(2) SSPM1(2) SSPM0(2)
bit 7 bit 0
Legend:
bit 7 WCOL: Write Collision Detect bit

In Master Transmit mode:
1 = A write to the SSPxBUF register was attempted while the I2C conditions were not valid for a
transmission to be started (must be cleared in software)
0 = No collision
In Slave Transmit mode:
1 = The SSPxBUF register is written while it is still transmitting the previous word (must be cleared in
software)
0 = No collision
In Receive mode (Master or Slave modes):
This is a “don’t care” bit.
bit 6 SSPOV: Receive Overflow Indicator bit
In Receive mode:
1 = A byte is received while the SSPxBUF register is still holding the previous byte (must be cleared in
software)
0 = No overflow
In Transmit mode:
This is a “don’t care” bit in Transmit mode.
bit 5 SSPEN: Master Synchronous Serial Port Enable bit(1)
1 = Enables the serial port and configures the SDAx and SCLx pins as the serial port pins
0 = Disables the serial port and configures these pins as I/O port pins
bit 4 CKP: SCKx Release Control bit
In Slave mode:
1 = Releases clock
0 = Holds clock low (clock stretch); used to ensure data setup time
In Master mode:
bit 3-0 SSPM<3:0>: Master Synchronous Serial Port Mode Select bits(2)
1111 = I2C Slave mode: 10-bit address with Start and Stop bit interrupts enabled
1110 = I2C Slave mode: 7-bit address with Start and Stop bit interrupts enabled
1011 = I2C Firmware Controlled Master mode (slave Idle)
1001 = Load the SSPMSK register at the SSPxADD SFR address(3,4)
1000 = I2C Master mode: Clock = FOSC/(4 * (SSPxADD + 1))
0111 = I2C Slave mode: 10-bit address
0110 = I2C Slave mode: 7-bit address
Note 1: When enabled, the SDAx and SCLx pins must be configured as inputs.
2: Bit combinations not specifically listed here are either reserved or implemented in SPI mode only.
3: When SSPM<3:0> = 1001, any reads or writes to the SSPxADD SFR address actually access the
SSPxMSK register.
4: This mode is only available when 7-Bit Address Masking mode is selected (MSSPMSK Configuration bit is ‘1’).

PIC18F87K90 FAMILY
REGISTER 21-5: SSPxCON2: MSSPx CONTROL REGISTER 2 (I2C™ MASTER MODE)
GCEN ACKSTAT ACKDT(1) ACKEN(2) RCEN(2) PEN(2) RSEN(2) SEN(2)
bit 7 bit 0
Legend:
bit 7 GCEN: General Call Enable bit

Unused in Master mode.
bit 6 ACKSTAT: Acknowledge Status bit (Master Transmit mode only)
1 = Acknowledge was not received from slave
0 = Acknowledge was received from slave
bit 5 ACKDT: Acknowledge Data bit (Master Receive mode only)(1)
1 = Not Acknowledge
0 = Acknowledge
bit 4 ACKEN: Acknowledge Sequence Enable bit(2)
1 = Initiates Acknowledge sequence on SDAx and SCLx pins and transmits ACKDT data bit;
automatically cleared by hardware
0 = Acknowledge sequence is Idle
bit 3 RCEN: Receive Enable bit (Master Receive mode only)(2)
1 = Enables Receive mode for I2C
0 = Receive is Idle
bit 2 PEN: Stop Condition Enable bit(2)
1 = Initiates Stop condition on SDAx and SCLx pins; automatically cleared by hardware
0 = Stop condition is Idle
bit 1 RSEN: Repeated Start Condition Enable bit(2)
1 = Initiates Repeated Start condition on SDAx and SCLx pins; automatically cleared by hardware
0 = Repeated Start condition is Idle
bit 0 SEN: Start Condition Enable bit(2)
1 = Initiates Start condition on SDAx and SCLx pins; automatically cleared by hardware
0 = Start condition is Idle
Note 1: The value that will be transmitted when the user initiates an Acknowledge sequence at the end of a
receive.
2: If the I2C module is active, these bits may not be set (no spooling) and the SSPxBUF may not be written to
(or writes to the SSPxBUF are disabled).

PIC18F87K90 FAMILY
REGISTER 21-6: SSPxCON2: MSSPx CONTROL REGISTER 2 (I2C™ SLAVE MODE)
GCEN ACKSTAT ADMSK5 ADMSK4 ADMSK3 ADMSK2 ADMSK1 SEN(1)
bit 7 bit 0
Legend:
bit 7 GCEN: General Call Enable bit

1 = Enables interrupt when a general call address (0000h) is received in the SSPxSR
0 = General call address is disabled
bit 6 ACKSTAT: Acknowledge Status bit
Unused in Slave mode.
bit 5-2 ADMSK<5:2>: Slave Address Mask Select bits (5-Bit Address Masking mode)
1 = Masking of corresponding bits of SSPxADD is enabled
0 = Masking of corresponding bits of SSPxADD is disabled
bit 1 ADMSK1: Slave Address Least Significant bit(s) Mask Select bit
In 7-Bit Addressing mode:
1 = Masking of SSPxADD<1> only is enabled
0 = Masking of SSPxADD<1> only is disabled
In 10-Bit Addressing mode:
1 = Masking of SSPxADD<1:0> is enabled
0 = Masking of SSPxADD<1:0> is disabled
bit 0 SEN: Stretch Enable bit(1)
1 = Clock stretching is enabled for both slave transmit and slave receive (stretch enabled)
0 = Clock stretching is disabled
Note 1: If the I2C module is active, this bit may not be set (no spooling) and the SSPxBUF may not be written to
(or writes to the SSPxBUF are disabled).
REGISTER 21-7: SSPxMSK: I2C™ SLAVE ADDRESS MASK REGISTER (7-BIT MASKING MODE)(1)
MSK7 MSK6 MSK5 MSK4 MSK3 MSK2 MSK1 MSK0(2)
bit 7 bit 0
Legend:
bit 7-0 MSK<7:0>: Slave Address Mask Select bit

1 = Masking of the corresponding bit of SSPxADD is enabled
0 = Masking of the corresponding bit of SSPxADD is disabled
Note 1: This register shares the same SFR address as SSPxADD and is only addressable in select MSSPx
operating modes. See Section 21.4.3.4 “7-Bit Address Masking Mode” for more details.
2: MSK0 is not used as a mask bit in 7-bit addressing.

PIC18F87K90 FAMILY
21.4.2 OPERATION 21.4.3.1 Addressing
The MSSP module functions are enabled by setting the Once the MSSP module has been enabled, it waits for
MSSP Enable bit, SSPEN (SSPxCON1<5>). a Start condition to occur. Following the Start condition,
The SSPxCON1 register allows control of the I2C the 8 bits are shifted into the SSPxSR register. All
operation. Four mode selection bits (SSPxCON1<3:0>) incoming bits are sampled with the rising edge of the
allow one of the following I2C modes to be selected: clock (SCLx) line. The value of register, SSPxSR<7:1>,
is compared to the value of the SSPxADD register. The
• I2C Master mode, clock address is compared on the falling edge of the eighth
• I 2C Slave mode (7-bit address) clock (SCLx) pulse. If the addresses match and the BF
• I 2C Slave mode (10-bit address) and SSPOV bits are clear, the following events occur:
• I 2C Slave mode (7-bit address) with Start and 1. The SSPxSR register value is loaded into the
Stop bit interrupts enabled SSPxBUF register.
• I 2C Slave mode (10-bit address) with Start and 2. The Buffer Full bit, BF, is set.
Stop bit interrupts enabled 3. An ACK pulse is generated.
• I 2C Firmware Controlled Master mode, slave is 4. The MSSP Interrupt Flag bit, SSPxIF, is set (and
Idle an interrupt is generated, if enabled) on the
Selection of any I 2C mode with the SSPEN bit set falling edge of the ninth SCLx pulse.
forces the SCLx and SDAx pins to be open-drain, In 10-Bit Addressing mode, two address bytes need to
provided these pins are programmed as inputs by be received by the slave. The five Most Significant bits
setting the appropriate TRISC or TRISD bits. To ensure (MSbs) of the first address byte specify if this is a 10-bit
proper operation of the module, pull-up resistors must address. The R/W bit (SSPxSTAT<2>) must specify a
be provided externally to the SCLx and SDAx pins. write so the slave device will receive the second
address byte. For a 10-bit address, the first byte would
21.4.3 SLAVE MODE equal ‘11110 A9 A8 0’, where ‘A9’ and ‘A8’ are the
In Slave mode, the SCLx and SDAx pins must be two MSbs of the address. The sequence of events for
configured as inputs (TRISC<4:3> set). The MSSP 10-bit addressing is as follows, with Steps, 7 through 9,
module will override the input state with the output data for the slave-transmitter:
when required (slave-transmitter). 1. Receive first (high) byte of address (bits,
The I 2C Slave mode hardware will always generate an SSPxIF, BF and UA, are set on address match).
interrupt on an address match. Address masking will 2. Update the SSPxADD register with second (low)
allow the hardware to generate an interrupt for more byte of address (clears bit, UA, and releases the
than one address (up to 31 in 7-bit addressing and up SCLx line).
to 63 in 10-bit addressing). Through the mode select 3. Read the SSPxBUF register (clears bit, BF) and
bits, the user can also choose to interrupt on Start and clear flag bit, SSPxIF.
Stop bits.
4. Receive second (low) byte of address (bits,
When an address is matched, or the data transfer after SSPxIF, BF and UA, are set).
an address match is received, the hardware auto- 5. Update the SSPxADD register with the first
matically will generate the Acknowledge (ACK) pulse (high) byte of address. If match releases the
and load the SSPxBUF register with the received value SCLx line, this will clear bit, UA.
currently in the SSPxSR register.
6. Read the SSPxBUF register (clears bit, BF) and
Any combination of the following conditions will cause clear flag bit, SSPxIF.
the MSSP module not to give this ACK pulse: 7. Receive Repeated Start condition.
• The Buffer Full bit, BF (SSPxSTAT<0>), was set 8. Receive first (high) byte of address (bits,
before the transfer was received. SSPxIF and BF, are set).
• The overflow bit, SSPOV (SSPxCON1<6>), was 9. Read the SSPxBUF register (clears bit, BF) and
set before the transfer was received. clear flag bit, SSPxIF.
In this case, the SSPxSR register value is not loaded
into the SSPxBUF, but bit, SSPxIF, is set. The BF bit is
cleared by reading the SSPxBUF register, while bit,
SSPOV, is cleared through software.
The SCLx clock input must have a minimum high and
low for proper operation. The high and low times of the
I2C specification, as well as the requirement of the
MSSP module, are shown in timing Parameter 100 and
Parameter 101.

PIC18F87K90 FAMILY
21.4.3.2 Address Masking Modes Acknowledge up to 31 addresses when using 7-bit
addressing, or 63 addresses with 10-bit addressing
Masking an address bit causes that bit to become a
(see Example 21-2). This Masking mode is selected
“don’t care”. When one address bit is masked, two
when the MSSPMSK Configuration bit is programmed
addresses will be Acknowledged and cause an
(‘0’).
interrupt. It is possible to mask more than one address
bit at a time, which greatly expands the number of The address mask in this mode is stored in the
addresses Acknowledged. SSPxCON2 register, which stops functioning as a
control register in I2C Slave mode (Register 21-6). In
The I2C slave behaves the same way whether address
7-Bit Address Masking mode, address mask bits,
masking is used or not. However, when address
ADMSK<5:1> (SSPxCON2<5:1>), mask the corre-
masking is used, the I2C slave can Acknowledge
sponding address bits in the SSPxADD register. For
multiple addresses and cause interrupts. When this
any ADMSK bits that are set (ADMSK<n> = 1), the cor-
occurs, it is necessary to determine which address
responding address bit is ignored (SSPxADD<n> = x).
caused the interrupt by checking the SSPxBUF.
For the module to issue an address Acknowledge, it is
The PIC18F87K90 family of devices is capable of using sufficient to match only on addresses that do not have
two different Address Masking modes in I2C slave an active address mask.
operation: 5-Bit Address Masking and 7-Bit Address
In 10-Bit Address Masking mode, bits, ADMSK<5:2>,
Masking. The Masking mode is selected at device
mask the corresponding address bits in the SSPxADD
configuration using the MSSPMSK Configuration bit.
register. In addition, ADMSK1 simultaneously masks
The default device configuration is 7-Bit Address
the two LSbs of the address (SSPxADD<1:0>). For any
Masking.
ADMSK bits that are active (ADMSK<n> = 1), the cor-
Both Masking modes, in turn, support address masking responding address bit is ignored (SPxADD<n> = x).
of 7-bit and 10-bit addresses. The combination of Also note, that although in 10-Bit Address Masking
Masking modes and addresses provide different mode, the upper address bits reuse part of the
ranges of Acknowledgable addresses for each SSPxADD register bits. The address mask bits do not
combination. interact with those bits; they only affect the lower
While both Masking modes function in roughly the address bits.
same manner, the way they use address masks are Note 1: ADMSK1 masks the two Least Significant
different. bits of the address.
21.4.3.3 5-Bit Address Masking Mode 2: The two Most Significant bits of the
address are not affected by address
As the name implies, 5-Bit Address Masking mode
masking.
uses an address mask of up to 5 bits to create a range
of addresses to be Acknowledged, using bits, 5 through
1, of the incoming address. This allows the module to
EXAMPLE 21-2: ADDRESS MASKING EXAMPLES IN 5-BIT MASKING MODE

7-Bit Addressing:
SSPxADD<7:1>= A0h (1010000) (SSPxADD<0> is assumed to be ‘0’)
ADMSK<5:1> = 00111
Addresses Acknowledged: A0h, A2h, A4h, A6h, A8h, AAh, ACh, AEh
10-Bit Addressing:
SSPxADD<7:0> = A0h (10100000) (The two MSb of the address are ignored in this example, since
they are not affected by masking)
ADMSK<5:1> = 00111
Addresses Acknowledged: A0h, A1h, A2h, A3h, A4h, A5h, A6h, A7h, A8h, A9h, AAh, ABh, ACh, ADh,
AEh, AFh

PIC18F87K90 FAMILY
21.4.3.4 7-Bit Address Masking Mode Setting or clearing mask bits in SSPxMSK behaves in
the opposite manner of the ADMSK bits in 5-Bit
Unlike 5-bit masking, 7-Bit Address Masking mode
Address Masking mode. That is, clearing a bit in
uses a mask of up to 8 bits (in 10-bit addressing) to
SSPxMSK causes the corresponding address bit to be
define a range of addresses that can be Acknowl-
masked; setting the bit requires a match in that
edged, using the lowest bits of the incoming address.
position. SSPxMSK resets to all ‘1’s upon any Reset
This allows the module to Acknowledge up to 127 dif-
condition and, therefore, has no effect on the standard
ferent addresses with 7-bit addressing, or 255 with
MSSP operation until written with a mask value.
10-bit addressing (see Example 21-3). This mode is
the default configuration of the module and is selected With 7-bit addressing, SSPxMSK<7:1> bits mask the
when MSSPMSK is unprogrammed (‘1’). corresponding address bits in the SSPxADD register.
For any SSPxMSK bits that are active
The address mask for 7-Bit Address Masking mode is
(SSPxMSK<n> = 0), the corresponding SSPxADD
stored in the SSPxMSK register, instead of the
address bit is ignored (SSPxADD<n> = x). For the
SSPxCON2 register. SSPxMSK is a separate hard-
module to issue an address Acknowledge, it is
ware register within the module, but it is not directly
sufficient to match only on addresses that do not have
addressable. Instead, it shares an address in the SFR
an active address mask.
space with the SSPxADD register. To access the
SSPxMSK register, it is necessary to select MSSP With 10-bit addressing, SSPxMSK<7:0> bits mask the
mode, ‘1001’ (SSPxCON1<3:0> = 1001) and then corresponding address bits in the SSPxADD register.
read or write to the location of SSPxADD. For any SSPxMSK bits that are active (= 0), the
corresponding SSPxADD address bit is ignored
To use 7-Bit Address Masking mode, it is necessary to
(SSPxADD<n> = x).
initialize SSPxMSK with a value before selecting the
I2C Slave Addressing mode. Thus, the required Note: The two Most Significant bits of the
sequence of events is: address are not affected by address
1. Select SSPxMSK Access mode masking.
(SSPxCON2<3:0> = 1001).
2. Write the mask value to the appropriate
SSPxADD register address (FC8h for MSSP1,
F6Eh for MSSP2).
3. Set the appropriate I2C Slave mode
(SSPxCON2<3:0> = 0111 for 10-bit addressing,
‘0110’ for 7-bit addressing).
EXAMPLE 21-3: ADDRESS MASKING EXAMPLES IN 7-BIT MASKING MODE

7-Bit Addressing:
SSPxADD<7:1> = 1010 000
SSPxMSK<7:1> = 1111 001
Addresses Acknowledged = ACh, A8h, A4h, A0h
10-Bit Addressing:
SSPxADD<7:0> = 1010 0000
SSPxMSK<7:0> = 1111 0011
Addresses Acknowledged = ACh, A8h, A4h, A0h

PIC18F87K90 FAMILY
21.4.3.5 Reception 21.4.3.6 Transmission
When the R/W bit of the address byte is clear and an When the R/W bit of the incoming address byte is set
address match occurs, the R/W bit of the SSPxSTAT and an address match occurs, the R/W bit of the
register is cleared. The received address is loaded into SSPxSTAT register is set. The received address is
the SSPxBUF register and the SDAx line is held low loaded into the SSPxBUF register. The ACK pulse will
(ACK). be sent on the ninth bit and pin, SCLx, is held low
When the address byte overflow condition exists, then regardless of SEN (see Section 21.4.4 “Clock
the no Acknowledge (ACK) pulse is given. An overflow Stretching” for more details). By stretching the clock,
condition is defined as either bit, BF (SSPxSTAT<0>), the master will be unable to assert another clock pulse
is set or bit, SSPOV (SSPxCON1<6>), is set. until the slave is done preparing the transmit data. The
transmit data must be loaded into the SSPxBUF regis-
An MSSP interrupt is generated for each data transfer ter which also loads the SSPxSR register. Then, pin,
byte. The interrupt flag bit, SSPxIF, must be cleared in SCLx, should be enabled by setting bit, CKP
software. The SSPxSTAT register is used to determine (SSPxCON1<4>). The eight data bits are shifted out on
the status of the byte. the falling edge of the SCLx input. This ensures that the
If SEN is enabled (SSPxCON2<0> = 1), SCLx will be SDAx signal is valid during the SCLx high time
held low (clock stretch) following each data transfer. The (Figure 21-10).
clock must be released by setting bit, CKP The ACK pulse from the master-receiver is latched on
(SSPxCON1<4>). See Section 21.4.4 “Clock the rising edge of the ninth SCLx input pulse. If the
Stretching” for more details. SDAx line is high (not ACK), then the data transfer is
complete. In this case, when the ACK is latched by the
slave, the slave logic is reset and the slave monitors for
another occurrence of the Start bit. If the SDAx line was
low (ACK), the next transmit data must be loaded into
the SSPxBUF register. Again, pin, SCLx, must be
enabled by setting bit, CKP.
An MSSP interrupt is generated for each data transfer
byte. The SSPxIF bit must be cleared in software and
the SSPxSTAT register is used to determine the status
of the byte. The SSPxIF bit is set on the falling edge of
the ninth clock pulse.

FIGURE 21-8:
DS39957D-page 322
Receiving Address R/W = 0 Receiving Data ACK Receiving Data ACK
SDAx A7 A6 A5 A4 A3 A2 A1 ACK D7 D6 D5 D4 D3 D2 D1 D0 D7 D6 D5 D4 D3 D2 D1 D0
SCLx 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
S P
SSPxIF (PIR1<3> or PIR3<7>)

Bus master
PIC18F87K90 FAMILY
terminates
transfer
BF (SSPxSTAT<0>)
Cleared in software
SSPxBUF is read
SSPOV (SSPxCON1<6>)
SSPOV is set
because SSPxBUF is
still full. ACK is not sent.
CKP (SSPxCON<4>)
(CKP does not reset to ‘0’ when SEN = 0)
I2C™ SLAVE MODE TIMING WITH SEN = 0 (RECEPTION, 7-BIT ADDRESS)
 2009-2011 Microchip Technology Inc.

FIGURE 21-9:

SDAx A7 A6 A5 X A3 X X ACK D7 D6 D5 D4 D3 D2 D1 D0 D7 D6 D5 D4 D3 D2 D1 D0
SCLx 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
S P

Bus master
terminates
(RECEPTION, 7-BIT ADDRESS)
transfer
BF (SSPxSTAT<0>)
Cleared in software
SSPxBUF is read
SSPOV (SSPxCON1<6>)
SSPOV is set
because SSPxBUF is
CKP (SSPxCON<4>)
Note 1: x = Don’t care (i.e., address bit can either be a ‘1’ or a ‘0’).
2: In this example, an address equal to A7.A6.A5.X.A3.X.X will be Acknowledged and cause an interrupt.
I2C™ SLAVE MODE TIMING WITH SEN = 0 AND ADMSK<5:1> = 01011
PIC18F87K90 FAMILY
DS39957D-page 323
FIGURE 21-10:
DS39957D-page 324
Receiving Address R/W = 1 Transmitting Data Transmitting Data
ACK ACK
SCLx
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
S
Data in SCLx held low P
sampled while CPU
responds to SSPxIF

PIC18F87K90 FAMILY
BF (SSPxSTAT<0>)
Cleared in software Cleared in software
From SSPxIF ISR From SSPxIF ISR
SSPxBUF is written in software SSPxBUF is written in software
Clear by reading
CKP (SSPxCON<4>)
CKP is set in software CKP is set in software

I2C™ SLAVE MODE TIMING (TRANSMISSION, 7-BIT ADDRESS)

FIGURE 21-11:
Clock is held low until Clock is held low until

update of SSPxADD has update of SSPxADD has
taken place taken place
Receive First Byte of Address Receive Second Byte of Address Receive Data Byte Receive Data Byte
R/W = 0 ACK
SDAx 1 1 1 1 0 A9 A8 ACK A7 A6 A5 X A3 A2 X X ACK D7 D6 D5 D4 D3 D2 D1 D0 ACK D7 D6 D5 D4 D3 D2 D1 D0

SCLx 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
S P
Bus master
terminates
SSPxIF (PIR1<3> or PIR3<7>) transfer
Cleared in software Cleared in software Cleared in software

Cleared in software
BF (SSPxSTAT<0>)
SSPxBUF is written with Dummy read of SSPxBUF

contents of SSPxSR to clear BF flag
(RECEPTION, 10-BIT ADDRESS)
SSPOV (SSPxCON1<6>)
SSPOV is set
because SSPxBUF is
UA (SSPxSTAT<1>)
UA is set indicating that Cleared by hardware Cleared by hardware when

the SSPxADD needs to be when SSPxADD is updated SSPxADD is updated with high
updated with low byte of address byte of address
UA is set indicating that

SSPxADD needs to be
updated
CKP (SSPxCON<4>)
Note 1: x = Don’t care (i.e., address bit can either be a ‘1’ or a ‘0’).
I2C™ SLAVE MODE TIMING WITH SEN = 0 AND ADMSK<5:1> = 01001
2: In this example, an address equal to A9.A8.A7.A6.A5.X.A3.A2.X.X will be Acknowledged and cause an interrupt.
3: Note that the Most Significant bits of the address are not affected by the bit masking.
PIC18F87K90 FAMILY
DS39957D-page 325
FIGURE 21-12:
DS39957D-page 326
update of SSPxADD has update of SSPxADD has
taken place taken place
R/W = 0 ACK
SDAx 1 1 1 1 0 A9 A8 ACK A7 A6 A5 A4 A3 A2 A1 A0 ACK D7 D6 D5 D4 D3 D2 D1 D0 ACK D7 D6 D5 D4 D3 D2 D1 D0
SCLx 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
S P
Bus master
terminates
SSPxIF (PIR1<3> or PIR3<7>) transfer

Cleared in software
BF (SSPxSTAT<0>)
PIC18F87K90 FAMILY
SSPxBUF is written with Dummy read of SSPxBUF

contents of SSPxSR to clear BF flag
SSPOV (SSPxCON1<6>)
SSPOV is set
because SSPxBUF is
UA (SSPxSTAT<1>)
UA is set indicating that Cleared by hardware Cleared by hardware when

the SSPxADD needs to be when SSPxADD is updated SSPxADD is updated with high
updated with low byte of address byte of address

SSPxADD needs to be
updated
CKP (SSPxCON<4>)

FIGURE 21-13:
Bus master
terminates
Clock is held low until Clock is held low until transfer
update of SSPxADD has update of SSPxADD has Clock is held low until
taken place taken place CKP is set to ‘1’
R/W = 0

Receive First Byte of Address Receive Second Byte of Address Receive First Byte of Address R/W = 1 Transmitting Data Byte ACK
SDAx 1 1 1 1 0 A9 A8 ACK A7 A6 A5 A4 A3 A2 A1 A0 ACK 1 1 1 1 0 A9 A8 ACK D7 D6 D5 D4 D3 D2 D1 D0
SCLx 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
S Sr P
BF (SSPxSTAT<0>)
SSPxBUF is written with Dummy read of SSPxBUF Dummy read of SSPxBUF

contents of SSPxSR to clear BF flag BF flag is clear Write of SSPxBUF Completion of
to clear BF flag initiates transmit data transmission
at the end of the
UA (SSPxSTAT<1>) third address sequence clears BF flag
UA is set indicating that Cleared by hardware when Cleared by hardware when

the SSPxADD needs to be SSPxADD is updated with low SSPxADD is updated with high
updated byte of address byte of address.

SSPxADD needs to be
updated
CKP (SSPxCON1<4>)
CKP is set in software
CKP is automatically cleared in hardware, holding SCLx low

I2C™ SLAVE MODE TIMING (TRANSMISSION, 10-BIT ADDRESS)
PIC18F87K90 FAMILY
DS39957D-page 327
PIC18F87K90 FAMILY
21.4.4 CLOCK STRETCHING 21.4.4.3 Clock Stretching for 7-Bit Slave
Both 7-Bit and 10-Bit Slave modes implement Transmit Mode
automatic clock stretching during a transmit sequence. The 7-Bit Slave Transmit mode implements clock
The SEN bit (SSPxCON2<0>) allows clock stretching stretching by clearing the CKP bit after the falling edge
to be enabled during receives. Setting SEN will cause of the ninth clock if the BF bit is clear. This occurs
the SCLx pin to be held low at the end of each data regardless of the state of the SEN bit.
receive sequence. The user’s ISR must set the CKP bit before transmis-
sion is allowed to continue. By holding the SCLx line
21.4.4.1 Clock Stretching for 7-Bit Slave low, the user has time to service the ISR and load the
Receive Mode (SEN = 1) contents of the SSPxBUF before the master device
In 7-Bit Slave Receive mode, on the falling edge of the can initiate another transmit sequence (see
ninth clock at the end of the ACK sequence, if the BF Figure 21-10).
bit is set, the CKP bit in the SSPxCON1 register is Note 1: If the user loads the contents of
automatically cleared, forcing the SCLx output to be SSPxBUF, setting the BF bit before the
held low. The CKP bit being cleared to ‘0’ will assert falling edge of the ninth clock, the CKP bit
the SCLx line low. The CKP bit must be set in the will not be cleared and clock stretching
user’s ISR before reception is allowed to continue. By will not occur.
holding the SCLx line low, the user has time to service
the ISR and read the contents of the SSPxBUF before 2: The CKP bit can be set in software
the master device can initiate another receive regardless of the state of the BF bit.
sequence. This will prevent buffer overruns from
occurring (see Figure 21-15). 21.4.4.4 Clock Stretching for 10-Bit Slave
Note 1: If the user reads the contents of the Transmit Mode
SSPxBUF before the falling edge of the In 10-Bit Slave Transmit mode, clock stretching is
ninth clock, thus clearing the BF bit, the controlled during the first two address sequences by
CKP bit will not be cleared and clock the state of the UA bit, just as it is in 10-Bit Slave
stretching will not occur. Receive mode. The first two addresses are followed
2: The CKP bit can be set in software by a third address sequence, which contains the
regardless of the state of the BF bit. The high-order bits of the 10-bit address and the R/W bit
user should be careful to clear the BF bit set to ‘1’. After the third address sequence is
in the ISR before the next receive performed, the UA bit is not set, the module is now
sequence in order to prevent an overflow configured in Transmit mode and clock stretching is
condition. controlled by the BF flag as in 7-Bit Slave Transmit
mode (see Figure 21-13).
21.4.4.2 Clock Stretching for 10-Bit Slave
Receive Mode (SEN = 1)
In 10-Bit Slave Receive mode, during the address
sequence, clock stretching automatically takes place
but CKP is not cleared. During this time, if the UA bit is
set after the ninth clock, clock stretching is initiated.
The UA bit is set after receiving the upper byte of the
10-bit address and following the receive of the second
byte of the 10-bit address with the R/W bit cleared to
‘0’. The release of the clock line occurs upon updating
SSPxADD. Clock stretching will occur on each data
receive sequence as described in 7-bit mode.
Note: If the user polls the UA bit and clears it by
updating the SSPxADD register before the
falling edge of the ninth clock occurs, and if
the user hasn’t cleared the BF bit by read-
ing the SSPxBUF register before that time,
then the CKP bit will still NOT be asserted
low. Clock stretching on the basis of the
state of the BF bit only occurs during a data
sequence, not an address sequence.

PIC18F87K90 FAMILY
21.4.4.5 Clock Synchronization and already asserted the SCLx line. The SCLx output will
the CKP bit remain low until the CKP bit is set and all other
devices on the I2C bus have deasserted SCLx. This
When the CKP bit is cleared, the SCLx output is forced
ensures that a write to the CKP bit will not violate the
to ‘0’. However, clearing the CKP bit will not assert the
minimum high time requirement for SCLx (see
SCLx output low until the SCLx output is already
Figure 21-14).
sampled low. Therefore, the CKP bit will not assert the
SCLx line until an external I2C master device has
FIGURE 21-14: CLOCK SYNCHRONIZATION TIMING
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
SDAx DX DX – 1
SCLx
Master Device
CKP Asserts Clock
Master Device
Deasserts Clock
WR
SSPxCON1

FIGURE 21-15:
DS39957D-page 330
Clock is not held low
because buffer full bit is
clear prior to falling edge Clock is held low until Clock is not held low
of 9th clock CKP is set to ‘1’ because ACK = 1
SCLx 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
S P

Bus master
PIC18F87K90 FAMILY
terminates
transfer
BF (SSPxSTAT<0>)
Cleared in software
SSPxBUF is read
SSPOV (SSPxCON1<6>)
SSPOV is set
because SSPxBUF is
CKP (SSPxCON<4>)
CKP
If BF is cleared written
prior to the falling to ‘1’ in
edge of the 9th clock, software
CKP will not be reset BF is set after falling
to ‘0’ and no clock edge of the 9th clock,
stretching will occur CKP is reset to ‘0’ and
clock stretching occurs

FIGURE 21-16:

update of SSPxADD has update of SSPxADD has Clock is not held low
Clock is held low until
taken place taken place because ACK = 1
CKP is set to ‘1’
R/W = 0 ACK
ACK ACK
SDAx 1 1 1 1 0 A9 A8 ACK A7 A6 A5 A4 A3 A2 A1 A0 D7 D6 D5 D4 D3 D2 D1 D0 D7 D6 D5 D4 D3 D2 D1 D0

SCLx 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
S P

Bus master
terminates
Cleared in software Cleared in software Cleared in software transfer
Cleared in software
BF (SSPxSTAT<0>)
SSPxBUF is written with Dummy read of SSPxBUF Dummy read of SSPxBUF
contents of SSPxSR to clear BF flag to clear BF flag
SSPOV (SSPxCON1<6>)
SSPOV is set
because SSPxBUF is
UA (SSPxSTAT<1>)
UA is set indicating that Cleared by hardware when Cleared by hardware when

the SSPxADD needs to be SSPxADD is updated with low SSPxADD is updated with high
updated byte of address after falling edge byte of address after falling edge
of ninth clock of ninth clock

SSPxADD needs to be
updated
CKP (SSPxCON<4>)
Note: An update of the SSPxADD
register before the falling
edge of the ninth clock will CKP written to ‘1’
have no effect on UA and in software
UA will remain set.
Note: An update of the SSPxADD register before

the falling edge of the ninth clock will have no
effect on UA and UA will remain set.
PIC18F87K90 FAMILY
DS39957D-page 331
PIC18F87K90 FAMILY
21.4.5 GENERAL CALL ADDRESS If the general call address matches, the SSPxSR is
SUPPORT transferred to the SSPxBUF, the BF flag bit is set
(eighth bit), and on the falling edge of the ninth bit (ACK
The addressing procedure for the I2C bus is such that
bit), the SSPxIF interrupt flag bit is set.
the first byte after the Start condition usually
determines which device will be the slave addressed by When the interrupt is serviced, the source for the
the master. The exception is the general call address interrupt can be checked by reading the contents of the
which can address all devices. When this address is SSPxBUF. The value can be used to determine if the
used, all devices should, in theory, respond with an address was device-specific or a general call address.
Acknowledge. In 10-Bit Addressing mode, the SSPxADD is required
The general call address is one of eight addresses to be updated for the second half of the address to
reserved for specific purposes by the I2C protocol. It match and the UA bit is set (SSPxSTAT<1>). If the gen-
consists of all ‘0’s with R/W = 0. eral call address is sampled when the GCEN bit is set,
while the slave is configured in 10-Bit Addressing
The general call address is recognized when the
mode, then the second half of the address is not
General Call Enable bit, GCEN, is enabled
necessary, the UA bit will not be set and the slave will
(SSPxCON2<7> set). Following a Start bit detect, 8 bits
begin receiving data after the Acknowledge
are shifted into the SSPxSR and the address is
(Figure 21-17).
compared against the SSPxADD. It is also compared to
the general call address and fixed in hardware.
FIGURE 21-17: SLAVE MODE GENERAL CALL ADDRESS SEQUENCE

(7 OR 10-BIT ADDRESSING MODE)
Address is Compared to General Call Address

after ACK, Set Interrupt
R/W = 0 Receiving Data ACK

SDAx General Call Address ACK D7 D6 D5 D4 D3 D2 D1 D0
SCLx 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
S
SSPxIF
BF (SSPxSTAT<0>)
Cleared in Software
SSPxBUF is Read
SSPOV (SSPxCON1<6>) ‘0’
GCEN (SSPxCON2<7>)
‘1’

PIC18F87K90 FAMILY
21.4.6 MASTER MODE Note: The MSSPx module, when configured in
Master mode is enabled by setting and clearing the I2C Master mode, does not allow queueing
appropriate SSPM bits in SSPxCON1 and by setting of events. For instance, the user is not
the SSPEN bit. In Master mode, the SCLx and SDAx allowed to initiate a Start condition and
lines are manipulated by the MSSP hardware if the immediately write the SSPxBUF register
TRIS bits are set. to initiate transmission before the Start
condition is complete. In this case, the
Master mode of operation is supported by interrupt
SSPxBUF will not be written to and the
generation on the detection of the Start and Stop
WCOL bit will be set, indicating that a write
conditions. The Stop (P) and Start (S) bits are cleared
to the SSPxBUF did not occur.
from a Reset or when the MSSP module is disabled.
Control of the I 2C bus may be taken when the P bit is The following events will cause the MSSPx Interrupt
set, or the bus is Idle, with both the S and P bits clear. Flag bit, SSPxIF, to be set (and MSSP interrupt, if
In Firmware Controlled Master mode, user code enabled):
conducts all I 2C bus operations based on Start and • Start condition
Stop bit conditions.
• Stop condition
Once Master mode is enabled, the user has six • Data transfer byte transmitted/received
options.
• Acknowledge transmitted
1. Assert a Start condition on SDAx and SCLx. • Repeated Start
2. Assert a Repeated Start condition on SDAx and
SCLx.
3. Write to the SSPxBUF register, initiating
transmission of data/address.
4. Configure the I2C port to receive data.
5. Generate an Acknowledge condition at the end
of a received byte of data.
6. Generate a Stop condition on SDAx and SCLx.
FIGURE 21-18: MSSP BLOCK DIAGRAM (I2C™ MASTER MODE)
Internal SSPM<3:0>
Data Bus SSPxADD<6:0>
Read Write
SSPxBUF Baud
Rate
Generator
SDAx Shift
Clock Arbitrate/WCOL Detect
SDAx In Clock
SSPxSR
(hold off clock source)
MSb LSb
Receive Enable
Start bit, Stop bit,

Clock Cntl
Acknowledge
Generate
SCLx
Start bit Detect

Stop bit Detect
SCLx In Write Collision Detect Set/Reset S, P (SSPxSTAT), WCOL (SSPxCON1);
Clock Arbitration Set SSPxIF, BCLxIF;
Bus Collision State Counter for Reset ACKSTAT, PEN (SSPxCON2)
End of XMIT/RCV

PIC18F87K90 FAMILY
21.4.6.1 I2C Master Mode Operation A typical transmit sequence would go as follows:
The master device generates all of the serial clock 1. The user generates a Start condition by setting
pulses and the Start and Stop conditions. A transfer is the Start Enable bit, SEN (SSPxCON2<0>).
ended with a Stop condition or with a Repeated Start 2. SSPxIF is set. The MSSPx module will wait the
condition. Since the Repeated Start condition is also required start time before any other operation
the beginning of the next serial transfer, the I2C bus will takes place.
not be released. 3. The user loads the SSPxBUF with the slave
In Master Transmitter mode, serial data is output address to transmit.
through SDAx while SCLx outputs the serial clock. The 4. Address is shifted out the SDAx pin until all 8 bits
first byte transmitted contains the slave address of the are transmitted.
receiving device (7 bits) and the Read/Write (R/W) bit. 5. The MSSPx module shifts in the ACK bit from
In this case, the R/W bit will be logic ‘0’. Serial data is the slave device and writes its value into the
transmitted, 8 bits at a time. After each byte is transmit- SSPxCON2 register (SSPxCON2<6>).
ted, an Acknowledge bit is received. Start and Stop
6. The MSSPx module generates an interrupt at
conditions are output to indicate the beginning and the
the end of the ninth clock cycle by setting the
end of a serial transfer.
SSPxIF bit.
In Master Receive mode, the first byte transmitted 7. The user loads the SSPxBUF with 8 bits of data.
contains the slave address of the transmitting device
8. Data is shifted out the SDAx pin until all 8 bits
(7 bits) and the R/W bit. In this case, the R/W bit will be
are transmitted.
logic ‘1’. Thus, the first byte transmitted is a 7-bit slave
address, followed by a ‘1’ to indicate the receive bit. 9. The MSSPx module shifts in the ACK bit from
Serial data is received via SDAx, while SCLx outputs the slave device and writes its value into the
the serial clock. Serial data is received, 8 bits at a time. SSPxCON2 register (SSPxCON2<6>).
After each byte is received, an Acknowledge bit is 10. The MSSPx module generates an interrupt at
transmitted. Start and Stop conditions indicate the the end of the ninth clock cycle by setting the
beginning and end of transmission. SSPxIF bit.
The Baud Rate Generator, used for the SPI mode 11. The user generates a Stop condition by setting
operation, is used to set the SCLx clock frequency for the Stop Enable bit, PEN (SSPxCON2<2>).
either 100 kHz, 400 kHz or 1 MHz I2C operation. See 12. An interrupt is generated once the Stop condition
Section 21.4.7 “Baud Rate” for more details. is complete.

PIC18F87K90 FAMILY
21.4.7 BAUD RATE 21.4.7.1 Baud Rate and Module
2
In I C Master mode, the Baud Rate Generator (BRG) Interdependence
reload value is placed in the lower 7 bits of the Because MSSP1 and MSSP2 are independent, they
SSPxADD register (Figure 21-19). When a write can operate simultaneously in I2C Master mode at
occurs to SSPxBUF, the Baud Rate Generator will different baud rates. This is done by using different
automatically begin counting. The BRG counts down to BRG reload values for each module.
0 and stops until another reload has taken place. The
Because this mode derives its basic clock source from
BRG count is decremented twice per instruction cycle
the system clock, any changes to the clock will affect
(TCY) on the Q2 and Q4 clocks. In I2C Master mode, the
both modules in the same proportion. It may be
BRG is reloaded automatically.
possible to change one or both baud rates back to a
Once the given operation is complete (i.e., transmis- previous value by changing the BRG reload value.
sion of the last data bit is followed by ACK), the internal
clock will automatically stop counting and the SCLx pin
will remain in its last state.
Table 21-3 demonstrates clock rates based on
instruction cycles and the BRG value loaded into
SSPxADD. The SSPxADD BRG value of ‘0x00’ is not
supported.
FIGURE 21-19: BAUD RATE GENERATOR BLOCK DIAGRAM
SSPM<3:0> SSPxADD<6:0>
SSPM<3:0> Reload Reload

SCLx Control
CLKO BRG Down Counter FOSC/4
TABLE 21-3: I2C™ CLOCK RATE w/BRG

FSCL
FOSC FCY FCY * 2 BRG Value
(2 Rollovers of BRG)
40 MHz 10 MHz 20 MHz 18h 400 kHz
40 MHz 10 MHz 20 MHz 1Fh 312.5 kHz
16 MHz 4 MHz 8 MHz 0Ch 308 kHz
16 MHz(1) 4 MHz 8 MHz 03h 1 MHz(1)
Note 1: A minimum of 16 MHz FOSC is required to get the 1 MHz I2C.

PIC18F87K90 FAMILY
21.4.7.2 Clock Arbitration SCLx pin is sampled high, the Baud Rate Generator is
reloaded with the contents of SSPxADD<6:0> and
Clock arbitration occurs when the master, during any
begins counting. This ensures that the SCLx high time
receive, transmit or Repeated Start/Stop condition,
will always be at least one BRG rollover count in the
deasserts the SCLx pin (SCLx allowed to float high).
event that the clock is held low by an external device
When the SCLx pin is allowed to float high, the Baud
(Figure 21-20).
Rate Generator (BRG) is suspended from counting
until the SCLx pin is actually sampled high. When the
FIGURE 21-20: BAUD RATE GENERATOR TIMING WITH CLOCK ARBITRATION
SDAx DX DX – 1
SCLx Deasserted but Slave Holds SCLx Allowed to Transition High

SCLx Low (clock arbitration)
SCLx
BRG Decrements on
Q2 and Q4 Cycles
BRG
03h 02h 01h 00h (hold off) 03h 02h
Value
SCLx is Sampled High, Reload Takes

Place and BRG Starts its Count
BRG
Reload

PIC18F87K90 FAMILY
21.4.8 I2C™ MASTER MODE START Note: If, at the beginning of the Start condition,
CONDITION TIMING the SDAx and SCLx pins are already
To initiate a Start condition, the user sets the Start sampled low, or if during the Start condi-
Enable bit, SEN (SSPxCON2<0>). If the SDAx and tion, the SCLx line is sampled low before
SCLx pins are sampled high, the Baud Rate Generator the SDAx line is driven low, a bus collision
is reloaded with the contents of SSPxADD<6:0> and occurs, the Bus Collision Interrupt Flag,
starts its count. If SCLx and SDAx are both sampled BCLxIF, is set, the Start condition is
high when the Baud Rate Generator times out (TBRG), aborted and the I2C module is reset into its
the SDAx pin is driven low. The action of the SDAx Idle state.
being driven low while SCLx is high is the Start condi-
tion and causes the S bit (SSPxSTAT<3>) to be set. 21.4.8.1 WCOL Status Flag
Following this, the Baud Rate Generator is reloaded If the user writes the SSPxBUF when a Start sequence
with the contents of SSPxADD<6:0> and resumes its is in progress, the WCOL bit is set and the contents of
count. When the Baud Rate Generator times out the buffer are unchanged (the write doesn’t occur).
(TBRG), the SEN bit (SSPxCON2<0>) will be
Note: Because queueing of events is not
automatically cleared by hardware. The Baud Rate
allowed, writing to the lower 5 bits of
Generator is suspended, leaving the SDAx line held low
SSPxCON2 is disabled until the Start
and the Start condition is complete.
condition is complete.
FIGURE 21-21: FIRST START BIT TIMING
Set S bit (SSPxSTAT<3>)

Write to SEN bit Occurs Here
SDAx = 1, At Completion of Start bit,
SCLx = 1
Hardware Clears SEN bit
and Sets SSPxIF bit
TBRG TBRG Write to SSPxBUF Occurs Here
1st bit 2nd bit

SDAx
TBRG
SCLx
TBRG
S

PIC18F87K90 FAMILY
21.4.9 I2C™ MASTER MODE REPEATED Note 1: If RSEN is programmed while any other
START CONDITION TIMING event is in progress, it will not take effect.
A Repeated Start condition occurs when the RSEN bit 2: A bus collision during the Repeated Start
(SSPxCON2<1>) is programmed high and the I2C logic condition occurs if:
module is in the Idle state. When the RSEN bit is set,
• SDAx is sampled low when SCLx
the SCLx pin is asserted low. When the SCLx pin is
goes from low-to-high.
sampled low, the Baud Rate Generator is loaded with
the contents of SSPxADD<5:0> and begins counting. • SCLx goes low before SDAx is
The SDAx pin is released (brought high) for one Baud asserted low. This may indicate that
Rate Generator count (TBRG). When the Baud Rate another master is attempting to
Generator times out, and if SDAx is sampled high, the transmit a data ‘1’.
SCLx pin will be deasserted (brought high). When
Immediately following the SSPxIF bit getting set, the
SCLx is sampled high, the Baud Rate Generator is
user may write the SSPxBUF with the 7-bit address in
reloaded with the contents of SSPxADD<6:0> and
7-bit mode or the default first address in 10-bit mode.
begins counting. SDAx and SCLx must be sampled
After the first eight bits are transmitted and an ACK is
high for one TBRG. This action is then followed by
received, the user may then transmit an additional eight
assertion of the SDAx pin (SDAx = 0) for one TBRG
bits of address (10-bit mode) or eight bits of data (7-bit
while SCLx is high. Following this, the RSEN bit
mode).
(SSPxCON2<1>) will be automatically cleared and the
Baud Rate Generator will not be reloaded, leaving the 21.4.9.1 WCOL Status Flag
SDAx pin held low. As soon as a Start condition is
detected on the SDAx and SCLx pins, the S bit If the user writes the SSPxBUF when a Repeated Start
(SSPxSTAT<3>) will be set. The SSPxIF bit will not be sequence is in progress, the WCOL is set and the
set until the Baud Rate Generator has timed out. contents of the buffer are unchanged (the write doesn’t
occur).
Note: Because queueing of events is not
allowed, writing of the lower 5 bits of
SSPxCON2 is disabled until the Repeated
Start condition is complete.
FIGURE 21-22: REPEATED START CONDITION WAVEFORM
S bit Set by Hardware

SDAx = 1, At Completion of Start bit,
Write to SSPxCON2 Occurs Here: SDAx = 1, SCLx = 1 Hardware Clears RSEN bit
SCLx (no change). and Sets SSPxIF
TBRG TBRG TBRG
SDAx 1st bit
RSEN bit Set by Hardware

on Falling Edge of Ninth Clock,
Write to SSPxBUF Occurs Here
End of XMIT
TBRG
SCLx
TBRG
Sr = Repeated Start

PIC18F87K90 FAMILY
21.4.10 I2C™ MASTER MODE 2 TCY after the SSPxBUF write. If SSPxBUF is rewritten
TRANSMISSION within 2 TCY, the WCOL bit is set and SSPxBUF is
updated. This may result in a corrupted transfer.
Transmission of a data byte, a 7-bit address or the
other half of a 10-bit address, is accomplished by The user should verify that the WCOL bit is clear after
simply writing a value to the SSPxBUF register. This each write to SSPxBUF to ensure the transfer is correct.
action will set the Buffer Full flag bit, BF, and allow the In all cases, WCOL must be cleared in software.
Baud Rate Generator to begin counting and start the
next transmission. Each bit of address/data will be 21.4.10.3 ACKSTAT Status Flag
shifted out onto the SDAx pin after the falling edge of In Transmit mode, the ACKSTAT bit (SSPxCON2<6>)
SCLx is asserted (see data hold time specification is cleared when the slave has sent an Acknowledge
Parameter 106). SCLx is held low for one Baud Rate (ACK = 0) and is set when the slave does not Acknowl-
Generator rollover count (TBRG). Data should be valid edge (ACK = 1). A slave sends an Acknowledge when
before SCLx is released high (see data setup time it has recognized its address (including a general call),
specification Parameter 107). When the SCLx pin is or when the slave has properly received its data.
released high, it is held that way for TBRG. The data on
the SDAx pin must remain stable for that duration and 21.4.11 I2C MASTER MODE RECEPTION
some hold time after the next falling edge of SCLx. Master mode reception is enabled by programming the
After the eighth bit is shifted out (the falling edge of the Receive Enable bit, RCEN (SSPxCON2<3>).
eighth clock), the BF flag is cleared and the master
releases SDAx. This allows the slave device being Note: The MSSP module must be in an inactive
addressed to respond with an ACK bit during the ninth state before the RCEN bit is set or the
bit time if an address match occurred, or if data was RCEN bit will be disregarded.
received properly. The status of ACK is written into the
The Baud Rate Generator begins counting, and on
ACKDT bit on the falling edge of the ninth clock. If the
each rollover, the state of the SCLx pin changes
master receives an Acknowledge, the Acknowledge
(high-to-low/low-to-high) and data is shifted into the
Status bit, ACKSTAT, is cleared; if not, the bit is set.
SSPxSR. After the falling edge of the eighth clock, the
After the ninth clock, the SSPxIF bit is set and the
receive enable flag is automatically cleared, the con-
master clock (Baud Rate Generator) is suspended until
tents of the SSPxSR are loaded into the SSPxBUF, the
the next data byte is loaded into the SSPxBUF, leaving
BF flag bit is set, the SSPxIF flag bit is set and the Baud
SCLx low and SDAx unchanged (Figure 21-23).
Rate Generator is suspended from counting, holding
After the write to the SSPxBUF, each bit of the address SCLx low. The MSSP is now in Idle state awaiting the
will be shifted out on the falling edge of SCLx until all next command. When the buffer is read by the CPU,
seven address bits and the R/W bit are completed. On the BF flag bit is automatically cleared. The user can
the falling edge of the eighth clock, the master will then send an Acknowledge bit at the end of reception
deassert the SDAx pin, allowing the slave to respond by setting the Acknowledge Sequence Enable bit,
with an Acknowledge. On the falling edge of the ninth ACKEN (SSPxCON2<4>).
clock, the master will sample the SDAx pin to see if the
address was recognized by a slave. The status of the 21.4.11.1 BF Status Flag
ACK bit is loaded into the ACKSTAT status bit
In receive operation, the BF bit is set when an address
(SSPxCON2<6>). Following the falling edge of the
or data byte is loaded into SSPxBUF from SSPxSR. It
ninth clock transmission of the address, the SSPxIF
is cleared when the SSPxBUF register is read.
flag is set, the BF flag is cleared and the Baud Rate
Generator is turned off until another write to the 21.4.11.2 SSPOV Status Flag
SSPxBUF takes place, holding SCLx low and allowing
SDAx to float. In receive operation, the SSPOV bit is set when 8 bits
are received into the SSPxSR and the BF flag bit is
21.4.10.1 BF Status Flag already set from a previous reception.
In Transmit mode, the BF bit (SSPxSTAT<0>) is set 21.4.11.3 WCOL Status Flag
when the CPU writes to SSPxBUF and is cleared when
all 8 bits are shifted out. If the user writes the SSPxBUF when a receive is
already in progress (i.e., SSPxSR is still shifting in a
21.4.10.2 WCOL Status Flag data byte), the WCOL bit is set and the contents of the
buffer are unchanged (the write doesn’t occur).
If the user writes the SSPxBUF when a transmit is
already in progress (i.e., SSPxSR is still shifting out a
data byte), the WCOL bit is set and the contents of the
buffer are unchanged (the write doesn’t occur) after

FIGURE 21-23:
DS39957D-page 340
Write SSPxCON2<0> (SEN = 1), ACKSTAT in
Start condition begins SSPxCON2 = 1
From slave, clear ACKSTAT bit (SSPxCON2<6>)
SEN = 0
Transmitting Data or Second Half
Transmit Address to Slave R/W = 0 ACK
of 10-bit Address
SDAx A7 A6 A5 A4 A3 A2 A1 ACK = 0 D7 D6 D5 D4 D3 D2 D1 D0
SSPxBUF written with 7-bit address and R/W,

start transmit
SCLx 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
S P
SCLx held low
while CPU
responds to SSPxIF
SSPxIF
PIC18F87K90 FAMILY
Cleared in software service routine

Cleared in software from MSSP interrupt
Cleared in software
BF (SSPxSTAT<0>)
SSPxBUF written SSPxBUF is written in software

SEN
After Start condition, SEN cleared by hardware
PEN
R/W
I 2C™ MASTER MODE WAVEFORM (TRANSMISSION, 7 OR 10-BIT ADDRESS)

FIGURE 21-24:
Write to SSPxCON2<4>
to start Acknowledge sequence,
SDAx = ACKDT (SSPxCON2<5>) = 0
Write to SSPxCON2<0> (SEN = 1),
begin Start condition ACK from master, Set ACKEN, start Acknowledge sequence,
Master configured as a receiver SDAx = ACKDT = 0 SDAx = ACKDT = 1
SEN = 0 by programming SSPxCON2<3> (RCEN = 1)
PEN bit = 1
Write to SSPxBUF occurs here, RCEN cleared RCEN = 1, start RCEN cleared
ACK from Slave next receive automatically written here
start XMIT automatically

Transmit Address to Slave R/W = 1 Receiving Data from Slave Receiving Data from Slave
SDAx A7 A6 A5 A4 A3 A2 A1 ACK D7 D6 D5 D4 D3 D2 D1 D0 ACK D7 D6 D5 D4 D3 D2 D1 D0 ACK
Bus master
ACK is not sent terminates
transfer
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
SCLx S P
Data shifted in on falling edge of CLK Set SSPxIF at end
of receive Set SSPxIF interrupt
Set SSPxIF interrupt at end of Acknowledge
Set SSPxIF interrupt sequence
at end of receive
at end of Acknowledge
SSPxIF sequence
Set P bit
Cleared in software Cleared in software Cleared in software Cleared in software (SSPxSTAT<4>)
Cleared in
SDAx = 0, SCLx = 1, software and SSPxIF
while CPU
responds to SSPxIF
BF
(SSPxSTAT<0>) Last bit is shifted into SSPxSR and
contents are unloaded into SSPxBUF
SSPOV
SSPOV is set because

SSPxBUF is still full
ACKEN
I 2C™ MASTER MODE WAVEFORM (RECEPTION, 7-BIT ADDRESS)
PIC18F87K90 FAMILY
DS39957D-page 341
PIC18F87K90 FAMILY
21.4.12 ACKNOWLEDGE SEQUENCE 21.4.13 STOP CONDITION TIMING
TIMING A Stop bit is asserted on the SDAx pin at the end of a
An Acknowledge sequence is enabled by setting the receive/transmit by setting the Stop Sequence Enable
Acknowledge Sequence Enable bit, ACKEN bit, PEN (SSPxCON2<2>). At the end of a
(SSPxCON2<4>). When this bit is set, the SCLx pin is receive/transmit, the SCLx line is held low after the
pulled low and the contents of the Acknowledge data bit falling edge of the ninth clock. When the PEN bit is set,
are presented on the SDAx pin. If the user wishes to the master will assert the SDAx line low. When the
generate an Acknowledge, then the ACKDT bit should SDAx line is sampled low, the Baud Rate Generator is
be cleared. If not, the user should set the ACKDT bit reloaded and counts down to 0. When the Baud Rate
before starting an Acknowledge sequence. The Baud Generator times out, the SCLx pin will be brought high
Rate Generator then counts for one rollover period and one TBRG (Baud Rate Generator rollover count)
(TBRG) and the SCLx pin is deasserted (pulled high). later, the SDAx pin will be deasserted. When the SDAx
When the SCLx pin is sampled high (clock arbitration), pin is sampled high while SCLx is high, the P bit
the Baud Rate Generator counts for TBRG; the SCLx pin (SSPxSTAT<4>) is set. A TBRG later, the PEN bit is
is then pulled low. Following this, the ACKEN bit is auto- cleared and the SSPxIF bit is set (see Figure 21-26).
matically cleared, the Baud Rate Generator is turned off
and the MSSP module then goes into an inactive state 21.4.13.1 WCOL Status Flag
(Figure 21-25). If the user writes the SSPxBUF when a Stop sequence
is in progress, then the WCOL bit is set and the
21.4.12.1 WCOL Status Flag contents of the buffer are unchanged (the write doesn’t
If the user writes the SSPxBUF when an Acknowledge occur).
sequence is in progress, then WCOL is set and the
contents of the buffer are unchanged (the write doesn’t
occur).
FIGURE 21-25: ACKNOWLEDGE SEQUENCE WAVEFORM

Acknowledge Sequence Starts Here, ACKEN Automatically Cleared
Write to SSPxCON2,
ACKEN = 1, ACKDT = 0
TBRG TBRG
SDAx D0 ACK
SCLx 8 9
SSPxIF
Cleared in
SSPxIF Set at Cleared in Software
the End of Receive Software SSPxIF Set at the End
of Acknowledge Sequence
Note: TBRG = one Baud Rate Generator period.
FIGURE 21-26: STOP CONDITION RECEIVE OR TRANSMIT MODE

Write to SSPxCON2, SCLx = 1 for TBRG, Followed by SDAx = 1 for TBRG
Set PEN After SDAx Sampled High. P bit (SSPxSTAT<4>) is Set.
Falling Edge of PEN bit (SSPxCON2<2>) is Cleared by

9th Clock Hardware and the SSPxIF bit is Set
TBRG
SCLx
SDAx ACK
P
TBRG TBRG TBRG
SCLx Brought High After TBRG
SDAx Asserted Low Before Rising Edge of Clock
to Set Up Stop Condition
Note: TBRG = one Baud Rate Generator period.

PIC18F87K90 FAMILY
21.4.14 SLEEP OPERATION 21.4.17 MULTI -MASTER COMMUNICATION,
2
While in Sleep mode, the I C module can receive BUS COLLISION AND BUS
addresses or data and when an address match or ARBITRATION
complete byte transfer occurs, wake the processor Multi-Master mode support is achieved by bus arbitra-
from Sleep (if the MSSP interrupt is enabled). tion. When the master outputs address/data bits onto
the SDAx pin, arbitration takes place when the master
21.4.15 EFFECTS OF A RESET outputs a ‘1’ on SDAx, by letting SDAx float high, and
A Reset disables the MSSP module and terminates the another master asserts a ‘0’. When the SCLx pin floats
current transfer. high, data should be stable. If the expected data on
SDAx is a ‘1’ and the data sampled on the SDAx
21.4.16 MULTI-MASTER MODE pin = 0, then a bus collision has taken place. The
In Multi-Master mode, the interrupt generation on the master will set the Bus Collision Interrupt Flag, BCLxIF,
detection of the Start and Stop conditions allows the and reset the I2C port to its Idle state (Figure 21-27).
determination of when the bus is free. The Stop (P) and If a transmit was in progress when the bus collision
Start (S) bits are cleared from a Reset or when the occurred, the transmission is halted, the BF flag is
MSSP module is disabled. Control of the I 2C bus may cleared, the SDAx and SCLx lines are deasserted and
be taken when the P bit (SSPxSTAT<4>) is set, or the the SSPxBUF can be written to. When the user services
bus is Idle, with both the S and P bits clear. When the the bus collision Interrupt Service Routine and if the I2C
bus is busy, enabling the MSSP interrupt will generate bus is free, the user can resume communication by
the interrupt when the Stop condition occurs. asserting a Start condition.
In multi-master operation, the SDAx line must be If a Start, Repeated Start, Stop or Acknowledge condition
monitored for arbitration to see if the signal level is the was in progress when the bus collision occurred, the con-
expected output level. This check is performed in dition is aborted, the SDAx and SCLx lines are
hardware with the result placed in the BCLxIF bit. deasserted and the respective control bits in the
The states where arbitration can be lost are: SSPxCON2 register are cleared. When the user services
the bus collision Interrupt Service Routine, and if the I2C
• Address Transfer bus is free, the user can resume communication by
• Data Transfer asserting a Start condition.
• A Start Condition The master will continue to monitor the SDAx and SCLx
• A Repeated Start Condition pins. If a Stop condition occurs, the SSPxIF bit will be set.
• An Acknowledge Condition A write to the SSPxBUF will start the transmission of
data at the first data bit regardless of where the
transmitter left off when the bus collision occurred.
In Multi-Master mode, the interrupt generation on the
detection of Start and Stop conditions allows the determi-
nation of when the bus is free. Control of the I2C bus can
be taken when the P bit is set in the SSPxSTAT register,
or the bus is Idle and the S and P bits are cleared.
FIGURE 21-27: BUS COLLISION TIMING FOR TRANSMIT AND ACKNOWLEDGE

Sample SDAx. While SCLx is High,
Data Changes SDAx Line Pulled Low
Data Doesn’t Match what is Driven
while SCLx = 0 by Another Source by the Master;
SDAx Released Bus Collision has Occurred.
by Master
SDAx
SCLx Set Bus Collision

Interrupt (BCLxIF)
BCLxIF

PIC18F87K90 FAMILY
21.4.17.1 Bus Collision During a Start If the SDAx pin is sampled low during this count, the
Condition BRG is reset and the SDAx line is asserted early
(Figure 21-30). If, however, a ‘1’ is sampled on the
During a Start condition, a bus collision occurs if:
SDAx pin, the SDAx pin is asserted low at the end of
a) SDAx or SCLx is sampled low at the beginning the BRG count. The Baud Rate Generator is then
of the Start condition (Figure 21-28). reloaded and counts down to 0. If the SCLx pin is
b) SCLx is sampled low before SDAx is asserted sampled as ‘0’ during this time, a bus collision does not
low (Figure 21-29). occur. At the end of the BRG count, the SCLx pin is
asserted low.
During a Start condition, both the SDAx and the SCLx
pins are monitored. Note: The reason that a bus collision is not a
If the SDAx pin is already low, or the SCLx pin is factor during a Start condition is that no two
already low, then all of the following occur: bus masters can assert a Start condition at
the exact same time. Therefore, one mas-
• The Start condition is aborted
ter will always assert SDAx before the
• The BCLxIF flag is set other. This condition does not cause a bus
• The MSSP module is reset to its inactive state collision because the two masters must be
(see Figure 21-28) allowed to arbitrate the first address
The Start condition begins with the SDAx and SCLx following the Start condition. If the address
pins deasserted. When the SDAx pin is sampled high, is the same, arbitration must be allowed to
the Baud Rate Generator is loaded from continue into the data portion, Repeated
SSPxADD<6:0> and counts down to 0. If the SCLx pin Start or Stop conditions.
is sampled low while SDAx is high, a bus collision
occurs because it is assumed that another master is
attempting to drive a data ‘1’ during the Start condition.
FIGURE 21-28: BUS COLLISION DURING START CONDITION (SDAx ONLY)
SDAx goes Low Before the SEN bit is Set.

Set BCLxIF,
S bit and SSPxIF Set because
SDAx = 0, SCLx = 1.
SDAx
SCLx
Set SEN, Enable Start SEN Cleared Automatically because of Bus Collision.
Condition if SDAx = 1, SCLx = 1 MSSP module Reset into Idle State.
SEN
SDAx Sampled Low before
Start Condition. Set BCLxIF.
S bit and SSPxIF Set because
BCLxIF SDAx = 0, SCLx = 1.
SSPxIF and BCLxIF are
Cleared in Software
SSPxIF
SSPxIF and BCLxIF are

Cleared in Software

PIC18F87K90 FAMILY
FIGURE 21-29: BUS COLLISION DURING START CONDITION (SCLx = 0)
SDAx = 0, SCLx = 1
TBRG TBRG
SDAx
SCLx Set SEN, Enable Start

Sequence if SDAx = 1, SCLx = 1
SCLx = 0 before SDAx = 0,
Bus Collision Occurs. Set BCLxIF.
SEN
SCLx = 0 before BRG Time-out,
Bus Collision Occurs. Set BCLxIF.
BCLxIF
Interrupt Cleared
in Software
S ‘0’ ‘0’
SSPxIF ‘0’ ‘0’
FIGURE 21-30: BRG RESET DUE TO SDAx ARBITRATION DURING START CONDITION
SDAx = 0, SCLx = 1
Set S Set SSPxIF
Less than TBRG
TBRG
SDAx SDAx Pulled Low by Other Master.

Reset BRG and Assert SDAx.
SCLx S
SCLx Pulled Low After BRG
Time-out
SEN
Set SEN, Enable Start
Sequence if SDAx = 1, SCLx = 1
BCLxIF ‘0’
SSPxIF
SDAx = 0, SCLx = 1, Interrupts Cleared
Set SSPxIF in Software

PIC18F87K90 FAMILY
21.4.17.2 Bus Collision During a Repeated If SDAx is low, a bus collision has occurred (i.e., another
Start Condition master is attempting to transmit a data ‘0’,
Figure 21-31). If SDAx is sampled high, the BRG is
During a Repeated Start condition, a bus collision
reloaded and begins counting. If SDAx goes from
occurs if:
high-to-low before the BRG times out, no bus collision
a) A low level is sampled on SDAx when SCLx occurs because no two masters can assert SDAx at
goes from a low level to a high level. exactly the same time.
b) SCLx goes low before SDAx is asserted low, If SCLx goes from high-to-low before the BRG times
indicating that another master is attempting to out and SDAx has not already been asserted, a bus
transmit a data ‘1’. collision occurs. In this case, another master is
When the user deasserts SDAx and the pin is allowed attempting to transmit a data ‘1’ during the Repeated
to float high, the BRG is loaded with SSPxADD<6:0> Start condition (see Figure 21-32).
and counts down to 0. The SCLx pin is then deasserted If, at the end of the BRG time-out, both SCLx and SDAx
and when sampled high, the SDAx pin is sampled. are still high, the SDAx pin is driven low and the BRG is
reloaded and begins counting. At the end of the count,
regardless of the status of the SCLx pin, the SCLx pin is
driven low and the Repeated Start condition is complete.
FIGURE 21-31: BUS COLLISION DURING A REPEATED START CONDITION (CASE 1)
SDAx
SCLx
Sample SDAx when SCLx goes High.

If SDAx = 0, Set BCLxIF and Release SDAx and SCLx.
RSEN
BCLxIF
Cleared in Software
S ‘0’
SSPxIF ‘0’
FIGURE 21-32: BUS COLLISION DURING REPEATED START CONDITION (CASE 2)
TBRG TBRG
SDAx
SCLx
SCLx goes Low Before SDAx,

BCLxIF Set BCLxIF. Release SDAx and SCLx.
Interrupt Cleared
in Software
RSEN
S ‘0’
SSPxIF

PIC18F87K90 FAMILY
21.4.17.3 Bus Collision During a Stop The Stop condition begins with SDAx asserted low.
Condition When SDAx is sampled low, the SCLx pin is allowed to
float. When the pin is sampled high (clock arbitration),
Bus collision occurs during a Stop condition if:
the Baud Rate Generator is loaded with
a) After the SDAx pin has been deasserted and SSPxADD<6:0> and counts down to 0. After the BRG
allowed to float high, SDAx is sampled low after times out, SDAx is sampled. If SDAx is sampled low, a
the BRG has timed out. bus collision has occurred. This is due to another
b) After the SCLx pin is deasserted, SCLx is master attempting to drive a data ‘0’ (Figure 21-33). If
sampled low before SDAx goes high. the SCLx pin is sampled low before SDAx is allowed to
float high, a bus collision occurs. This is another case
of another master attempting to drive a data ‘0’
(Figure 21-34).
FIGURE 21-33: BUS COLLISION DURING A STOP CONDITION (CASE 1)
TBRG TBRG TBRG SDAx Sampled

Low After TBRG,
Set BCLxIF
SDAx
SDAx Asserted Low

SCLx
PEN
BCLxIF
P ‘0’
SSPxIF ‘0’
FIGURE 21-34: BUS COLLISION DURING A STOP CONDITION (CASE 2)
TBRG TBRG TBRG
SDAx
SCLx goes Low Before SDAx goes High,
Assert SDAx
Set BCLxIF
SCLx
PEN
BCLxIF
P ‘0’
SSPxIF ‘0’

PIC18F87K90 FAMILY
TABLE 21-4: REGISTERS ASSOCIATED WITH I2C™ OPERATION
Reset
on Page:
PIR2 OSCFIF — SSP2IF BLC2IF BCL1IF HLVDIF TMR3IF TMR3GIF 77
PIE2 OSCFIE — SSP2IE BLC2IE BCL1IE HLVDIE TMR3IE TMR3GIE 77
IPR2 OSCFIP — SSP2IP BLC2IP BCL1IP HLVDIP TMR3IP TMR3GIP 77
2C™
SSP1ADD MSSP1 Address Register (I Slave mode), 76
MSSP1 Baud Rate Reload Register (I2C Master mode)
SSP1MSK(1) MSK7 MSK6 MSK5 MSK4 MSK3 MSK2 MSK1 MSK0 —
GCEN ACKSTAT ACKDT ACKEN RCEN PEN RSEN SEN
SSP1CON2 76
GCEN ACKSTAT ADMSK5(2) ADMSK4(2) ADMSK3(2) ADMSK2(2) ADMSK1(2) SEN
SSP2ADD MSSP2 Address Register (I2C Slave mode), 82
MSSP2 Baud Rate Reload Register (I2C Master mode)
SSP2MSK(1) MSK7 MSK6 MSK5 MSK4 MSK3 MSK2 MSK1 MSK0 —
GCEN ACKSTAT ACKDT ACKEN RCEN PEN RSEN SEN
SSP2CON2 83
GCEN ACKSTAT ADMSK5(2) ADMSK4(2) ADMSK3(2) ADMSK2(2) ADMSK1(2) SEN
2
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by the MSSP module in I C™ mode.
Note 1: SSPxMSK shares the same address in SFR space as SSPxADD, but is only accessible in certain I2C™
Slave operating modes in 7-Bit Masking mode. See Section 21.4.3.4 “7-Bit Address Masking Mode” for
more details.
2: Alternate bit definitions for use in I2C Slave mode operations only.

PIC18F87K90 FAMILY
22.0 ENHANCED UNIVERSAL The pins of EUSART1 and EUSART2 are multiplexed
with the functions of PORTC (RC6/TX1/CK1/
SYNCHRONOUS
SEG27 and RC7/RX1/DT1/SEG28) and PORTG
ASYNCHRONOUS RECEIVER (RG1/TX2/CK2/AN19/C3OUT and RG2/RX2/DT2/
TRANSMITTER (EUSART) AN18/C3INA), respectively. In order to configure
these pins as an EUSART:
The Enhanced Universal Synchronous Asynchronous
Receiver Transmitter (EUSART) module is one of two • For EUSART1:
serial I/O modules. (Generically, the EUSART is also - SPEN (RCSTA1<7>) bit must be set (= 1)
known as a Serial Communications Interface or SCI.) - TRISC<7> bit must be set (= 1)
The EUSART can be configured as a full-duplex, - TRISC<6> bit must be cleared (= 0) for
asynchronous system that can communicate with Asynchronous and Synchronous Master
peripheral devices, such as CRT terminals and modes
personal computers. It can also be configured as a
- TRISC<6> bit must be set (= 1) for
half-duplex synchronous system that can communicate
Synchronous Slave mode
with peripheral devices, such as A/D or D/A integrated
circuits, serial EEPROMs, etc. • For EUSART2:
- SPEN (RCSTA2<7>) bit must be set (= 1)
The Enhanced USART module implements additional
features, including automatic baud rate detection and - TRISG<2> bit must be set (= 1)
calibration, automatic wake-up on Sync Break recep- - TRISG<1> bit must be cleared (= 0) for
tion and 12-bit Break character transmit. These make it Asynchronous and Synchronous Master
ideally suited for use in Local Interconnect Network bus modes
(LIN/J2602 bus) systems. - TRISC<6> bit must be set (= 1) for
All members of the PIC18F87K90 family are equipped Synchronous Slave mode
with two independent EUSART modules, referred to as Note: The EUSART control will automatically
EUSART1 and EUSART2. They can be configured in reconfigure the pin from input to output as
the following modes: needed.
• Asynchronous (full duplex) with: The operation of each Enhanced USART module is
- Auto-wake-up on character reception controlled through three registers:
- Auto-baud calibration • Transmit Status and Control (TXSTAx)
- 12-bit Break character transmission • Receive Status and Control (RCSTAx)
• Synchronous – Master (half duplex) with • Baud Rate Control (BAUDCONx)
selectable clock polarity
These are detailed in Register 22-1, Register 22-2 and
• Synchronous – Slave (half duplex) with selectable Register 22-3, respectively, on the following pages.
clock polarity
Note: Throughout this section, references to
register and bit names that may be asso-
ciated with a specific EUSART module are
referred to generically by the use of ‘x’ in
place of the specific module number.
Thus, “RCSTAx” might refer to the
Receive Status register for either
EUSART1 or EUSART2.

PIC18F87K90 FAMILY
REGISTER 22-1: TXSTAx: TRANSMIT STATUS AND CONTROL REGISTER
R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R-1 R/W-0
CSRC TX9 TXEN(1) SYNC SENDB BRGH TRMT TX9D
bit 7 bit 0
Legend:
bit 7 CSRC: Clock Source Select bit

Asynchronous mode:
Don’t care.
Synchronous mode:
1 = Master mode (clock generated internally from BRG)
0 = Slave mode (clock from external source)
bit 6 TX9: 9-Bit Transmit Enable bit
1 = Selects 9-bit transmission
0 = Selects 8-bit transmission
bit 5 TXEN: Transmit Enable bit(1)
1 = Transmit is enabled
0 = Transmit is disabled
bit 4 SYNC: EUSART Mode Select bit
1 = Synchronous mode
0 = Asynchronous mode
bit 3 SENDB: Send Break Character bit
Asynchronous mode:
1 = Send Sync Break on next transmission (cleared by hardware upon completion)
0 = Sync Break transmission has completed
Synchronous mode:
Don’t care.
bit 2 BRGH: High Baud Rate Select bit
Asynchronous mode:
1 = High speed
0 = Low speed
Synchronous mode:
bit 1 TRMT: Transmit Shift Register Status bit
1 = TSR is empty
0 = TSR is full
bit 0 TX9D: 9th bit of Transmit Data
Can be an address/data bit or a parity bit.
Note 1: SREN/CREN overrides TXEN in Sync mode.

PIC18F87K90 FAMILY
REGISTER 22-2: RCSTAx: RECEIVE STATUS AND CONTROL REGISTER
R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R-0 R-0 R-x
SPEN RX9 SREN CREN ADDEN FERR OERR RX9D
bit 7 bit 0
Legend:
bit 7 SPEN: Serial Port Enable bit

1 = Serial port is enabled
0 = Serial is port disabled (held in Reset)
bit 6 RX9: 9-Bit Receive Enable bit
1 = Selects 9-bit reception
0 = Selects 8-bit reception
bit 5 SREN: Single Receive Enable bit
Asynchronous mode:
Don’t care.
Synchronous mode – Master:
1 = Enables single receive
0 = Disables single receive
This bit is cleared after reception is complete.
Synchronous mode – Slave:
Don’t care.
bit 4 CREN: Continuous Receive Enable bit
Asynchronous mode:
1 = Enables the receiver
0 = Disables the receiver
Synchronous mode:
1 = Enables continuous receive until enable bit, CREN, is cleared (CREN overrides SREN)
0 = Disables continuous receive
bit 3 ADDEN: Address Detect Enable bit
Asynchronous mode 9-Bit (RX9 = 1):
1 = Enables address detection, enables interrupt and loads the receive buffer when RSR<8> is set
0 = Disables address detection, all bytes are received and ninth bit can be used as parity bit
Asynchronous mode 8-Bit (RX9 = 0):
Don’t care.
bit 2 FERR: Framing Error bit
1 = Framing error (can be cleared by reading the RCREGx register and receiving the next valid byte)
0 = No framing error
bit 1 OERR: Overrun Error bit
1 = Overrun error (can be cleared by clearing bit, CREN)
0 = No overrun error
bit 0 RX9D: 9th bit of Received Data
This can be an address/data bit or a parity bit and must be calculated by user firmware.

PIC18F87K90 FAMILY
REGISTER 22-3: BAUDCONx: BAUD RATE CONTROL REGISTER
R/W-0 R-1 R/W-0 R/W-0 R/W-0 U-0 R/W-0 R/W-0
ABDOVF RCIDL RXDTP TXCKP BRG16 — WUE ABDEN
bit 7 bit 0
Legend:
bit 7 ABDOVF: Auto-Baud Acquisition Rollover Status bit

1 = A BRG rollover has occurred during Auto-Baud Rate Detect mode (must be cleared in software)
0 = No BRG rollover has occurred
bit 6 RCIDL: Receive Operation Idle Status bit
1 = Receive operation is Idle
0 = Receive operation is active
bit 5 RXDTP: Data/Receive Polarity Select bit
Asynchronous mode:
1 = Receive data (RXx) is inverted (active-low)
0 = Receive data (RXx) is not inverted (active-high)
Synchronous mode:
1 = Data (DTx) is inverted (active-low)
0 = Data (DTx) is not inverted (active-high)
bit 4 TXCKP: Synchronous Clock Polarity Select bit
Asynchronous mode:
1 = Idle state for transmit (TXx) is a low level
0 = Idle state for transmit (TXx) is a high level
Synchronous mode:
1 = Idle state for clock (CKx) is a high level
0 = Idle state for clock (CKx) is a low level
bit 3 BRG16: 16-Bit Baud Rate Register Enable bit
1 = 16-bit Baud Rate Generator – SPBRGHx and SPBRGx
0 = 8-bit Baud Rate Generator – SPBRGx only (Compatible mode), SPBRGHx value is ignored
bit 1 WUE: Wake-up Enable bit
Asynchronous mode:
1 = EUSART will continue to sample the RXx pin – interrupt is generated on the falling edge; bit is
cleared in hardware on the following rising edge
0 = RXx pin is not monitored or the rising edge detected
Synchronous mode:
bit 0 ABDEN: Auto-Baud Detect Enable bit
Asynchronous mode:
1 = Enable baud rate measurement on the next character. Requires reception of a Sync field (55h);
cleared in hardware upon completion.
0 = Baud rate measurement is disabled or completed
Synchronous mode:

PIC18F87K90 FAMILY
22.1 Baud Rate Generator (BRG) Writing a new value to the SPBRGHx:SPBRGx registers
causes the BRG timer to be reset (or cleared). This
The BRG is a dedicated, 8-bit or 16-bit generator that ensures the BRG does not wait for a timer overflow
supports both the Asynchronous and Synchronous before outputting the new baud rate. When operated in
modes of the EUSART. By default, the BRG operates Synchronous mode, SPBRGH:SPBRG values of 0000h
in 8-bit mode; setting the BRG16 bit (BAUDCONx<3>) and 0001h are not supported. In the Asynchronous
selects 16-bit mode. mode, all BRG values may be used.
The SPBRGHx:SPBRGx register pair controls the period
of a free-running timer. In Asynchronous mode, bits, 22.1.1 OPERATION IN POWER-MANAGED
BRGH (TXSTAx<2>) and BRG16 (BAUDCONx<3>), MODES
also control the baud rate. In Synchronous mode, BRGH The device clock is used to generate the desired baud
is ignored. Table 22-1 shows the formula for computation rate. When one of the power-managed modes is
of the baud rate for different EUSART modes which only entered, the new clock source may be operating at a
apply in Master mode (internally generated clock). different frequency. This may require an adjustment to
Given the desired baud rate and FOSC, the nearest the value in the SPBRGx register pair.
integer value for the SPBRGHx:SPBRGx registers can
be calculated using the formulas in Table 22-1. From this, 22.1.2 SAMPLING
the error in baud rate can be determined. An example The data on the RXx pin (either RC7/RX1/DT1/SEG28
calculation is shown in Example 22-1. Typical baud rates or RG2/RX2/DT2/AN18/C3INA) is sampled three times
and error values for the various Asynchronous modes by a majority detect circuit to determine if a high or a
are shown in Table 22-2. It may be advantageous to use low level is present at the RXx pin.
the high baud rate (BRGH = 1) or the 16-bit BRG to
reduce the baud rate error, or achieve a slow baud rate
for a fast oscillator frequency.
TABLE 22-1: BAUD RATE FORMULAS

Configuration Bits
BRG/EUSART Mode Baud Rate Formula
SYNC BRG16 BRGH
0 0 0 8-Bit/Asynchronous FOSC/[64 (n + 1)]
0 0 1 8-Bit/Asynchronous
FOSC/[16 (n + 1)]
1 0 x 8-Bit/Synchronous FOSC/[4 (n + 1)]
1 1 x 16-Bit/Synchronous
Legend: x = Don’t care, n = value of SPBRGHx:SPBRGx register pair

PIC18F87K90 FAMILY
EXAMPLE 22-1: CALCULATING BAUD RATE ERROR
For a device with FOSC of 16 MHz, desired baud rate of 9600, Asynchronous mode, and 8-bit BRG:
Desired Baud Rate = FOSC/(64 ([SPBRGHx:SPBRGx] + 1))
Solving for SPBRGHx:SPBRGx:
X = ((FOSC/Desired Baud Rate)/64) – 1
= ((16000000/9600)/64) – 1
= [25.042] = 25
Calculated Baud Rate = 16000000/(64 (25 + 1))
= 9615
Error = (Calculated Baud Rate – Desired Baud Rate)/Desired Baud Rate
= (9615 – 9600)/9600 = 0.16%
TABLE 22-2: REGISTERS ASSOCIATED WITH BAUD RATE GENERATOR

Reset Values
on Page:
TXSTA1 CSRC TX9 TXEN SYNC SENDB BRGH TRMT TX9D 77
RCSTA1 SPEN RX9 SREN CREN ADDEN FERR OERR RX9D 77
BAUDCON1 ABDOVF RCIDL RXDTP TXCKP BRG16 — WUE ABDEN 79
SPBRGH1 EUSART1 Baud Rate Generator Register High Byte 76
SPBRG1 EUSART1 Baud Rate Generator Register Low Byte 77
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by the BRG.

PIC18F87K90 FAMILY
TABLE 22-3: BAUD RATES FOR ASYNCHRONOUS MODES
SYNC = 0, BRGH = 0, BRG16 = 0
BAUD
FOSC = 40.000 MHz FOSC = 20.000 MHz FOSC = 10.000 MHz FOSC = 8.000 MHz
RATE
(K) Actual SPBRG Actual SPBRG Actual SPBRG Actual SPBRG
% % % %
Rate value Rate value Rate value Rate value
Error Error Error Error
(K) (decimal) (K) (decimal) (K) (decimal) (K) (decimal)
0.3 — — — — — — — — — — — —
1.2 — — — 1.221 1.73 255 1.202 0.16 129 1.201 -0.16 103
2.4 2.441 1.73 255 2.404 0.16 129 2.404 0.16 64 2.403 -0.16 51
9.6 9.615 0.16 64 9.766 1.73 31 9.766 1.73 15 9.615 -0.16 12
19.2 19.531 1.73 31 19.531 1.73 15 19.531 1.73 7 — — —
57.6 56.818 -1.36 10 62.500 8.51 4 52.083 -9.58 2 — — —
115.2 125.000 8.51 4 104.167 -9.58 2 78.125 -32.18 1 — — —
SYNC = 0, BRGH = 0, BRG16 = 0

BAUD
FOSC = 4.000 MHz FOSC = 2.000 MHz FOSC = 1.000 MHz
RATE
(K) Actual SPBRG Actual SPBRG Actual SPBRG
% % %
Rate value Rate value Rate value
Error Error Error
(K) (decimal) (K) (decimal) (K) (decimal)
0.3 0.300 0.16 207 0.300 -0.16 103 0.300 -0.16 51

1.2 1.202 0.16 51 1.201 -0.16 25 1.201 -0.16 12
2.4 2.404 0.16 25 2.403 -0.16 12 — — —
9.6 8.929 -6.99 6 — — — — — —
19.2 20.833 8.51 2 — — — — — —
57.6 62.500 8.51 0 — — — — — —
115.2 62.500 -45.75 0 — — — — — —
SYNC = 0, BRGH = 1, BRG16 = 0

BAUD
FOSC = 40.000 MHz FOSC = 20.000 MHz FOSC = 10.000 MHz FOSC = 8.000 MHz
RATE
% % % %
0.3 — — — — — — — — — — — —
1.2 — — — — — — — — — — — —
2.4 — — — — — — 2.441 1.73 255 2.403 -0.16 207
9.6 9.766 1.73 255 9.615 0.16 129 9.615 0.16 64 9.615 -0.16 51
19.2 19.231 0.16 129 19.231 0.16 64 19.531 1.73 31 19.230 -0.16 25
57.6 58.140 0.94 42 56.818 -1.36 21 56.818 -1.36 10 55.555 3.55 8
115.2 113.636 -1.36 21 113.636 -1.36 10 125.000 8.51 4 — — —
SYNC = 0, BRGH = 1, BRG16 = 0

BAUD
RATE
% % %
Error Error Error
0.3 — — — — — — 0.300 -0.16 207

1.2 1.202 0.16 207 1.201 -0.16 103 1.201 -0.16 51
2.4 2.404 0.16 103 2.403 -0.16 51 2.403 -0.16 25
9.6 9.615 0.16 25 9.615 -0.16 12 — — —
19.2 19.231 0.16 12 — — — — — —
57.6 62.500 8.51 3 — — — — — —
115.2 125.000 8.51 1 — — — — — —

PIC18F87K90 FAMILY
TABLE 22-3: BAUD RATES FOR ASYNCHRONOUS MODES (CONTINUED)
SYNC = 0, BRGH = 0, BRG16 = 1
BAUD FOSC = 40.000 MHz FOSC = 20.000 MHz FOSC = 10.000 MHz FOSC = 8.000 MHz
RATE
% % % %
0.3 0.300 0.00 8332 0.300 0.02 4165 0.300 0.02 2082 0.300 -0.04 1665
1.2 1.200 0.02 2082 1.200 -0.03 1041 1.200 -0.03 520 1.201 -0.16 415
2.4 2.402 0.06 1040 2.399 -0.03 520 2.404 0.16 259 2.403 -0.16 207
9.6 9.615 0.16 259 9.615 0.16 129 9.615 0.16 64 9.615 -0.16 51
19.2 19.231 0.16 129 19.231 0.16 64 19.531 1.73 31 19.230 -0.16 25
57.6 58.140 0.94 42 56.818 -1.36 21 56.818 -1.36 10 55.555 3.55 8
115.2 113.636 -1.36 21 113.636 -1.36 10 125.000 8.51 4 — — —
SYNC = 0, BRGH = 0, BRG16 = 1

BAUD
RATE
% % %
Error Error Error
0.3 0.300 0.04 832 0.300 -0.16 415 0.300 -0.16 207
1.2 1.202 0.16 207 1.201 -0.16 103 1.201 -0.16 51
2.4 2.404 0.16 103 2.403 -0.16 51 2.403 -0.16 25
9.6 9.615 0.16 25 9.615 -0.16 12 — — —
19.2 19.231 0.16 12 — — — — — —
57.6 62.500 8.51 3 — — — — — —
115.2 125.000 8.51 1 — — — — — —
SYNC = 0, BRGH = 1, BRG16 = 1 or SYNC = 1, BRG16 = 1

BAUD FOSC = 40.000 MHz FOSC = 20.000 MHz FOSC = 10.000 MHz FOSC = 8.000 MHz
RATE
% % % %
0.3 0.300 0.00 33332 0.300 0.00 16665 0.300 0.00 8332 0.300 -0.01 6665
1.2 1.200 0.00 8332 1.200 0.02 4165 1.200 0.02 2082 1.200 -0.04 1665
2.4 2.400 0.02 4165 2.400 0.02 2082 2.402 0.06 1040 2.400 -0.04 832
9.6 9.606 0.06 1040 9.596 -0.03 520 9.615 0.16 259 9.615 -0.16 207
19.2 19.193 -0.03 520 19.231 0.16 259 19.231 0.16 129 19.230 -0.16 103
57.6 57.803 0.35 172 57.471 -0.22 86 58.140 0.94 42 57.142 0.79 34
115.2 114.943 -0.22 86 116.279 0.94 42 113.636 -1.36 21 117.647 -2.12 16
SYNC = 0, BRGH = 1, BRG16 = 1 or SYNC = 1, BRG16 = 1

BAUD FOSC = 4.000 MHz FOSC = 2.000 MHz FOSC = 1.000 MHz
RATE
% % %
Error Error Error
0.3 0.300 0.01 3332 0.300 -0.04 1665 0.300 -0.04 832
1.2 1.200 0.04 832 1.201 -0.16 415 1.201 -0.16 207
2.4 2.404 0.16 415 2.403 -0.16 207 2.403 -0.16 103
9.6 9.615 0.16 103 9.615 -0.16 51 9.615 -0.16 25
19.2 19.231 0.16 51 19.230 -0.16 25 19.230 -0.16 12
57.6 58.824 2.12 16 55.555 3.55 8 — — —
115.2 111.111 -3.55 8 — — — — — —

PIC18F87K90 FAMILY
22.1.3 AUTO-BAUD RATE DETECT While the ABD sequence takes place, the EUSART
state machine is held in Idle. The RCxIF interrupt is set
The Enhanced USART module supports the automatic
once the fifth rising edge on RXx is detected. The value
detection and calibration of baud rate. This feature is
in the RCREGx needs to be read to clear the RCxIF
active only in Asynchronous mode and while the WUE
interrupt. The contents of RCREGx should be
bit is clear.
discarded.
The automatic baud rate measurement sequence
(Figure 22-1) begins whenever a Start bit is received Note 1: If the WUE bit is set with the ABDEN bit,
and the ABDEN bit is set. The calculation is Auto-Baud Rate Detection will occur on
self-averaging. the byte following the Break character.
In the Auto-Baud Rate Detect (ABD) mode, the clock to 2: It is up to the user to determine that the
the BRG is reversed. Rather than the BRG clocking the incoming character baud rate is within the
incoming RXx signal, the RXx signal is timing the BRG. range of the selected BRG clock source.
In ABD mode, the internal Baud Rate Generator is Some combinations of oscillator fre-
used as a counter to time the bit period of the incoming quency and EUSART baud rates are not
serial byte stream. possible due to bit error rates. Overall
system timing and communication baud
Once the ABDEN bit is set, the state machine will clear rates must be taken into consideration
the BRG and look for a Start bit. The Auto-Baud Rate when using the Auto-Baud Rate Detection
Detect must receive a byte with the value, 55h (ASCII feature.
“U”, which is also the LIN/J2602 bus Sync character), in
order to calculate the proper bit rate. The measurement 3: To maximize baud rate range, it is
is taken over both a low and a high bit time in order to recommended to set the BRG16
minimize any effects caused by asymmetry of the incom- (BAUDCONx<3>) bit if the auto-baud
ing signal. After a Start bit, the SPBRGx begins counting feature is used.
up, using the preselected clock source on the first rising
edge of RXx. After eight bits on the RXx pin or the fifth
TABLE 22-4: BRG COUNTER
rising edge, an accumulated value totalling the proper
CLOCK RATES
BRG period is left in the SPBRGHx:SPBRGx register
pair. Once the 5th edge is seen (this should correspond BRG16 BRGH BRG Counter Clock
to the Stop bit), the ABDEN bit is automatically cleared.
0 0 FOSC/512
If a rollover of the BRG occurs (an overflow from FFFFh
0 1 FOSC/128
to 0000h), the event is trapped by the ABDOVF status
bit (BAUDCONx<7>). It is set in hardware by BRG roll- 1 0 FOSC/128
overs and can be set or cleared by the user in software. 1 1 FOSC/32
ABD mode remains active after rollover events and the
ABDEN bit remains set (Figure 22-2). 22.1.3.1 ABD and EUSART Transmission
While calibrating the baud rate period, the BRG Since the BRG clock is reversed during ABD acquisi-
registers are clocked at 1/8th the preconfigured clock tion, the EUSART transmitter cannot be used during
rate. Note that the BRG clock can be configured by the ABD. This means that whenever the ABDEN bit is set,
BRG16 and BRGH bits. The BRG16 bit must be set to TXREGx cannot be written to. Users should also
use both SPBRG1 and SPBRGH1 as a 16-bit counter. ensure that ABDEN does not become set during a
This allows the user to verify that no carry occurred for transmit sequence. Failing to do this may result in
8-bit modes by checking for 00h in the SPBRGHx unpredictable EUSART operation.
register. Refer to Table 22-4 for counter clock rates to
the BRG.

PIC18F87K90 FAMILY
FIGURE 22-1: AUTOMATIC BAUD RATE CALCULATION
BRG Value XXXXh 0000h 001Ch
Edge #1 Edge #2 Edge #3 Edge #4 Edge #5

RXx Pin Start Bit 0 Bit 1 Bit 2 Bit 3 Bit 4 Bit 5 Bit 6 Bit 7 Stop Bit
BRG Clock
Set by User Auto-Cleared

ABDEN bit
RCxIF bit
(Interrupt)
Read
RCREGx
SPBRGx XXXXh 1Ch
SPBRGHx XXXXh 00h
Note: The ABD sequence requires the EUSART module to be configured in Asynchronous mode and WUE = 0.
FIGURE 22-2: BRG OVERFLOW SEQUENCE
BRG Clock
ABDEN bit
RXx Pin Start Bit 0
ABDOVF bit
FFFFh
BRG Value XXXXh 0000h 0000h

PIC18F87K90 FAMILY
22.2 EUSART Asynchronous Mode Once the TXREGx register transfers the data to the TSR
register (occurs in one TCY), the TXREGx register is
The Asynchronous mode of operation is selected by empty and the TXxIF flag bit is set. This interrupt can be
clearing the SYNC bit (TXSTAx<4>). In this mode, the enabled or disabled by setting or clearing the interrupt
EUSART uses standard Non-Return-to-Zero (NRZ) enable bit, TXxIE. TXxIF will be set regardless of the
format (one Start bit, eight or nine data bits and one Stop state of TXxIE; it cannot be cleared in software. TXxIF is
bit). The most common data format is 8 bits. An on-chip,
also not cleared immediately upon loading TXREGx, but
dedicated 8-bit/16-bit Baud Rate Generator can be used
becomes valid in the second instruction cycle following
to derive standard baud rate frequencies from the
the load instruction. Polling TXxIF immediately following
oscillator.
a load of TXREGx will return invalid results.
The EUSART transmits and receives the LSb first. The
EUSART’s transmitter and receiver are functionally While TXxIF indicates the status of the TXREGx regis-
independent but use the same data format and baud ter; another bit, TRMT (TXSTAx<1>), shows the status
rate. The Baud Rate Generator produces a clock, either of the TSR register. TRMT is a read-only bit which is set
x16 or x64 of the bit shift rate, depending on the BRGH when the TSR register is empty. No interrupt logic is
and BRG16 bits (TXSTAx<2> and BAUDCONx<3>). tied to this bit so the user has to poll this bit in order to
Parity is not supported by the hardware but can be determine if the TSR register is empty.
implemented in software and stored as the 9th data bit.
Note 1: The TSR register is not mapped in data
When operating in Asynchronous mode, the EUSART memory, so it is not available to the user.
module consists of the following important elements:
2: Flag bit, TXxIF, is set when enable bit,
• Baud Rate Generator TXEN, is set.
• Sampling Circuit
To set up an Asynchronous Transmission:
• Asynchronous Transmitter
• Asynchronous Receiver 1. Initialize the SPBRGHx:SPBRGx registers for
the appropriate baud rate. Set or clear the
• Auto-Wake-up on Sync Break Character
BRGH and BRG16 bits, as required, to achieve
• 12-Bit Break Character Transmit the desired baud rate.
• Auto-Baud Rate Detection 2. Enable the asynchronous serial port by clearing
bit, SYNC, and setting bit, SPEN.
22.2.1 EUSART ASYNCHRONOUS
3. If interrupts are desired, set enable bit, TXxIE.
TRANSMITTER
4. If 9-bit transmission is desired, set transmit bit,
The EUSART transmitter block diagram is shown in TX9; can be used as an address/data bit.
Figure 22-3. The heart of the transmitter is the Transmit 5. Enable the transmission by setting bit, TXEN,
(Serial) Shift Register (TSR). The Shift register obtains which will also set bit, TXxIF.
its data from the Read/Write Transmit Buffer register,
6. If 9-bit transmission is selected, the ninth bit
TXREGx. The TXREGx register is loaded with data in should be loaded in bit, TX9D.
software. The TSR register is not loaded until the Stop
7. Load data to the TXREGx register (starts
bit has been transmitted from the previous load. As
transmission).
soon as the Stop bit is transmitted, the TSR is loaded
8. If using interrupts, ensure that the GIE and PEIE
with new data from the TXREGx register (if available).
bits in the INTCON register (INTCON<7:6>) are
set.
FIGURE 22-3: EUSART TRANSMIT BLOCK DIAGRAM

Data Bus
TXxIF TXREGx Register

TXxIE
8
MSb LSb
(8)  0 Pin Buffer
and Control
TSR Register TXx Pin
Interrupt
TXEN Baud Rate CLK
TRMT SPEN
BRG16 SPBRGHx SPBRGx
TX9
Baud Rate Generator TX9D

PIC18F87K90 FAMILY
FIGURE 22-4: ASYNCHRONOUS TRANSMISSION
Write to TXREGx
Word 1
BRG Output
(Shift Clock)
TXx (pin)
Start bit bit 0 bit 1 bit 7/8 Stop bit
Word 1
TXxIF bit
(Transmit Buffer 1 TCY
Reg. Empty Flag)
Word 1
TRMT bit
Transmit Shift Reg
(Transmit Shift
Reg. Empty Flag)
FIGURE 22-5: ASYNCHRONOUS TRANSMISSION (BACK-TO-BACK)
Write to TXREGx
Word 1 Word 2
BRG Output
(Shift Clock)
TXx (pin)
Start bit bit 0 bit 1 bit 7/8 Stop bit Start bit bit 0
TXxIF bit 1 TCY Word 1 Word 2

(Interrupt Reg. Flag)
1 TCY
Word 1 Word 2
TRMT bit Transmit Shift Reg. Transmit Shift Reg.
(Transmit Shift
Reg. Empty Flag)
Note: This timing diagram shows two consecutive transmissions.

PIC18F87K90 FAMILY
TABLE 22-5: REGISTERS ASSOCIATED WITH ASYNCHRONOUS TRANSMISSION
Reset
on Page:
TXREG1 EUSART1 Transmit Register 77
Legend: — = unimplemented locations read as ‘0’. Shaded cells are not used for asynchronous transmission.

PIC18F87K90 FAMILY
22.2.2 EUSART ASYNCHRONOUS 22.2.3 SETTING UP 9-BIT MODE WITH
RECEIVER ADDRESS DETECT
The receiver block diagram is shown in Figure 22-6. This mode would typically be used in RS-485 systems.
The data is received on the RXx pin and drives the data To set up an Asynchronous Reception with Address
recovery block. The data recovery block is actually a Detect Enable:
high-speed shifter operating at x16 times the baud rate, 1. Initialize the SPBRGHx:SPBRGx registers for
whereas the main receive serial shifter operates at the the appropriate baud rate. Set or clear the
bit rate or at FOSC. This mode would typically be used BRGH and BRG16 bits, as required, to achieve
in RS-232 systems. the desired baud rate.
To set up an Asynchronous Reception: 2. Enable the asynchronous serial port by clearing
1. Initialize the SPBRGHx:SPBRGx registers for the SYNC bit and setting the SPEN bit.
the appropriate baud rate. Set or clear the 3. If interrupts are required, set the RCEN bit and
BRGH and BRG16 bits, as required, to achieve select the desired priority level with the RCxIP bit.
the desired baud rate. 4. Set the RX9 bit to enable 9-bit reception.
2. Enable the asynchronous serial port by clearing 5. Set the ADDEN bit to enable address detect.
bit, SYNC, and setting bit, SPEN. 6. Enable reception by setting the CREN bit.
3. If interrupts are desired, set enable bit, RCxIE. 7. The RCxIF bit will be set when reception is
4. If 9-bit reception is desired, set bit, RX9. complete. The interrupt will be Acknowledged if
5. Enable the reception by setting bit, CREN. the RCxIE and GIE bits are set.
6. Flag bit, RCxIF, will be set when reception is 8. Read the RCSTAx register to determine if any
complete and an interrupt will be generated if error occurred during reception, as well as read
enable bit, RCxIE, was set. bit 9 of data (if applicable).
7. Read the RCSTAx register to get the 9th bit (if 9. Read RCREGx to determine if the device is
enabled) and determine if any error occurred being addressed.
during reception. 10. If any error occurred, clear the CREN bit.
8. Read the 8-bit received data by reading the 11. If the device has been addressed, clear the
RCREGx register. ADDEN bit to allow all received data into the
9. If any error occurred, clear the error by clearing receive buffer and interrupt the CPU.
enable bit, CREN.
set.
FIGURE 22-6: EUSART RECEIVE BLOCK DIAGRAM
CREN OERR FERR
x64 Baud Rate CLK

 64 MSb RSR Register LSb
BRG16 SPBRGHx SPBRGx or
 16 
or Stop (8) 7 1 0 Start
Baud Rate Generator 4
RX9
Pin Buffer Data

and Control Recovery
RXx RX9D RCREGx Register
FIFO
SPEN
8
Interrupt RCxIF Data Bus

RCxIE

PIC18F87K90 FAMILY
FIGURE 22-7: ASYNCHRONOUS RECEPTION
Start Start Start
RXx (pin) bit bit 0 bit 1 bit 7/8 Stop bit bit 0 bit 7/8 Stop bit bit 7/8 Stop
bit bit bit
Rcv Shift Reg
Rcv Buffer Reg
Word 1 Word 2
RCREGx RCREGx
Read Rcv
Buffer Reg
RCREGx
RCxIF
(Interrupt Flag)
OERR bit
CREN
Note: This timing diagram shows three words appearing on the RXx input. The RCREGx (Receive Buffer) is read after the third word
causing the OERR (Overrun) bit to be set.
TABLE 22-6: REGISTERS ASSOCIATED WITH ASYNCHRONOUS RECEPTION

Reset
on Page:
RCREG1 EUSART1 Receive Register 77
Legend: — = unimplemented locations read as ‘0’. Shaded cells are not used for asynchronous reception.

PIC18F87K90 FAMILY
22.2.4 AUTO-WAKE-UP ON SYNC BREAK 22.2.4.1 Special Considerations Using
CHARACTER Auto-Wake-up
During Sleep mode, all clocks to the EUSART are Since auto-wake-up functions by sensing rising edge
suspended. Because of this, the Baud Rate Generator transitions on RXx/DTx, information with any state
is inactive and a proper byte reception cannot be per- changes before the Stop bit may signal a false
formed. The auto-wake-up feature allows the controller End-of-Character (EOC) and cause data or framing
to wake-up due to activity on the RXx/DTx line while the errors. To work properly, therefore, the initial character
EUSART is operating in Asynchronous mode. in the transmission must be all ‘0’s. This can be 00h
The auto-wake-up feature is enabled by setting the (8 bits) for standard RS-232 devices or 000h (12 bits)
WUE bit (BAUDCONx<1>). Once set, the typical for LIN/J2602 bus.
receive sequence on RXx/DTx is disabled and the Oscillator start-up time must also be considered,
EUSART remains in an Idle state, monitoring for a especially in applications using oscillators with longer
wake-up event independent of the CPU mode. A start-up intervals (i.e., HS or HSPLL mode). The Sync
wake-up event consists of a high-to-low transition on Break (or Wake-up Signal) character must be of
the RXx/DTx line. (This coincides with the start of a sufficient length and be followed by a sufficient interval
Sync Break or a Wake-up Signal character for the to allow enough time for the selected oscillator to start
LIN/J2602 protocol.) and provide proper initialization of the EUSART.
Following a wake-up event, the module generates an
RCxIF interrupt. The interrupt is generated synchro-
nously to the Q clocks in normal operating modes
(Figure 22-8) and asynchronously if the device is in
Sleep mode (Figure 22-9). The interrupt condition is
cleared by reading the RCREGx register.
The WUE bit is automatically cleared once a low-to-high
transition is observed on the RXx line following the
wake-up event. At this point, the EUSART module is in
Idle mode and returns to normal operation. This signals
to the user that the Sync Break event is over.

PIC18F87K90 FAMILY
22.2.4.2 Special Considerations Using The fact that the WUE bit has been cleared (or is still
the WUE Bit set) and the RCxIF flag is set should not be used as an
indicator of the integrity of the data in RCREGx. Users
The timing of WUE and RCxIF events may cause some
should consider implementing a parallel method in
confusion when it comes to determining the validity of
firmware to verify received data integrity.
received data. As noted, setting the WUE bit places the
EUSART in an Idle mode. The wake-up event causes a To assure that no actual data is lost, check the RCIDL
receive interrupt by setting the RCxIF bit. The WUE bit bit to verify that a receive operation is not in process. If
is cleared after this when a rising edge is seen on a receive operation is not occurring, the WUE bit may
RXx/DTx. The interrupt condition is then cleared by then be set just prior to entering Sleep mode.
reading the RCREGx register. Ordinarily, the data in
RCREGx will be dummy data and should be discarded.
FIGURE 22-8: AUTO-WAKE-UP BIT (WUE) TIMINGS DURING NORMAL OPERATION
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
OSC1
Bit Set by User Auto-Cleared
WUE bit(1)
RXx/DTx Line
RCxIF
Cleared due to User Read of RCREGx
Note 1: The EUSART remains in Idle while the WUE bit is set.
FIGURE 22-9: AUTO-WAKE-UP BIT (WUE) TIMINGS DURING SLEEP

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
OSC1
Bit Set by User Auto-Cleared
WUE bit(2)
RXx/DTx Line
Note 1
RCxIF
Cleared due to User Read of RCREGx
SLEEP Command Executed Sleep Ends
Note 1: If the wake-up event requires long oscillator warm-up time, the auto-clear of the WUE bit can occur before the oscillator is ready. This
sequence should not depend on the presence of Q clocks.
2: The EUSART remains in Idle while the WUE bit is set.

PIC18F87K90 FAMILY
22.2.5 BREAK CHARACTER SEQUENCE 1. Configure the EUSART for the desired mode.
The EUSART module has the capability of sending the 2. Set the TXEN and SENDB bits to set up the
special Break character sequences that are required by Break character.
the LIN/J2602 bus standard. The Break character 3. Load the TXREGx with a dummy character to
transmit consists of a Start bit, followed by twelve ‘0’ initiate transmission (the value is ignored).
bits and a Stop bit. The Frame Break character is sent 4. Write ‘55h’ to TXREGx to load the Sync
whenever the SENDB and TXEN bits (TXSTAx<3> and character into the transmit FIFO buffer.
TXSTAx<5>, respectively) are set while the Transmit 5. After the Break has been sent, the SENDB bit is
Shift Register is loaded with data. Note that the value reset by hardware. The Sync character now
of data written to TXREGx will be ignored and all ‘0’s transmits in the preconfigured mode.
will be transmitted.
When the TXREGx becomes empty, as indicated by
The SENDB bit is automatically reset by hardware after the TXxIF, the next data byte can be written to
the corresponding Stop bit is sent. This allows the user TXREGx.
to preload the transmit FIFO with the next transmit byte
following the Break character (typically, the Sync 22.2.6 RECEIVING A BREAK CHARACTER
character in the LIN/J2602 specification).
The Enhanced USART module can receive a Break
Note that the data value written to the TXREGx for the character in two ways.
Break character is ignored. The write simply serves the
The first method forces configuration of the baud rate
purpose of initiating the proper sequence.
at a frequency of 9/13 the typical speed. This allows for
The TRMT bit indicates when the transmit operation is the Stop bit transition to be at the correct sampling
active or Idle, just as it does during normal transmis- location (13 bits for Break versus Start bit and 8 data
sion. See Figure 22-10 for the timing of the Break bits for typical data).
character sequence.
The second method uses the auto-wake-up feature
22.2.5.1 Break and Sync Transmit Sequence described in Section 22.2.4 “Auto-Wake-up on Sync
Break Character”. By enabling this feature, the
The following sequence will send a message frame EUSART will sample the next two transitions on
header made up of a Break, followed by an Auto-Baud RXx/DTx, cause an RCxIF interrupt and receive the
Sync byte. This sequence is typical of a LIN/J2602 bus next data byte followed by another interrupt.
master.
Note that following a Break character, the user will
typically want to enable the Auto-Baud Rate Detect
feature. For both methods, the user can set the ABDEN
bit once the TXxIF interrupt is observed.
FIGURE 22-10: SEND BREAK CHARACTER SEQUENCE
Write to TXREGx
Dummy Write
BRG Output
(Shift Clock)
TXx (pin) Start Bit Bit 0 Bit 1 Bit 11 Stop Bit
Break
TXxIF bit
(Transmit Buffer
Reg. Empty Flag)
TRMT bit
(Transmit Shift
Reg. Empty Flag)
SENDB Sampled Here Auto-Cleared
SENDB bit
(Transmit Shift
Reg. Empty Flag)

PIC18F87K90 FAMILY
22.3 EUSART Synchronous Once the TXREGx register transfers the data to the
Master Mode TSR register (occurs in one TCY), the TXREGx is empty
and the TXxIF flag bit is set. The interrupt can be
The Synchronous Master mode is entered by setting enabled or disabled by setting or clearing the interrupt
the CSRC bit (TXSTAx<7>). In this mode, the data is enable bit, TXxIE. TXxIF is set regardless of the state
transmitted in a half-duplex manner (i.e., transmission of enable bit, TXxIE; it cannot be cleared in software. It
and reception do not occur at the same time). When will reset only when new data is loaded into the
transmitting data, the reception is inhibited and vice TXREGx register.
versa. Synchronous mode is entered by setting bit,
While flag bit, TXxIF, indicates the status of the TXREGx
SYNC (TXSTAx<4>). In addition, enable bit, SPEN
register, another bit, TRMT (TXSTAx<1>), shows the
(RCSTAx<7>), is set in order to configure the TXx and
status of the TSR register. TRMT is a read-only bit which
RXx pins to CKx (clock) and DTx (data) lines,
is set when the TSR is empty. No interrupt logic is tied to
respectively.
this bit, so the user must poll this bit in order to determine
The Master mode indicates that the processor trans- if the TSR register is empty. The TSR is not mapped in
mits the master clock on the CKx line. Clock polarity is data memory so it is not available to the user.
selected with the TXCKP bit (BAUDCONx<4>). Setting
To set up a Synchronous Master Transmission:
TXCKP sets the Idle state on CKx as high, while clear-
ing the bit sets the Idle state as low. This option is 1. Initialize the SPBRGHx:SPBRGx registers for the
provided to support Microwire devices with this module. appropriate baud rate. Set or clear the BRG16
bit, as required, to achieve the desired baud rate.
22.3.1 EUSART SYNCHRONOUS MASTER 2. Enable the synchronous master serial port by
TRANSMISSION setting bits, SYNC, SPEN and CSRC.
The EUSART transmitter block diagram is shown in 3. If interrupts are desired, set enable bit, TXxIE.
Figure 22-3. The heart of the transmitter is the Transmit 4. If 9-bit transmission is desired, set bit, TX9.
(Serial) Shift Register (TSR). The TSR register obtains 5. Enable the transmission by setting bit, TXEN.
its data from the Read/Write Transmit Buffer register, 6. If 9-bit transmission is selected, the ninth bit
TXREGx. The TXREGx register is loaded with data in should be loaded in bit, TX9D.
software. The TSR register is not loaded until the last
7. Start transmission by loading data to the
bit has been transmitted from the previous load. As
TXREGx register.
soon as the last bit is transmitted, the TSR is loaded
with new data from the TXREGx (if available). 8. If using interrupts, ensure that the GIE and PEIE
set.
FIGURE 22-11: SYNCHRONOUS TRANSMISSION
Q1 Q2 Q3Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q3 Q4 Q1 Q2 Q3 Q4 Q1Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
RC7/RX1/DT1/
SEG28 Pin bit 0 bit 1 bit 2 bit 7 bit 0 bit 1 bit 7
Word 1 Word 2
RC6/TX1/CK1/
SEG27 Pin
(TXCKP = 0)
RC6/TX1/CK1/
SEG27 Pin
(TXCKP = 1)
Write to
TXREG1 Reg Write Word 1 Write Word 2
TX1IF bit
(Interrupt Flag)
TRMT bit
TXEN bit ‘1’ ‘1’
Note: Sync Master mode, SPBRGx = 0, continuous transmission of two 8-bit words. This example is equally applicable to EUSART2
(RG1/TX2/CK2/AN19/C3OUT and RG2/RX2/DT2/AN18/C3INA).

PIC18F87K90 FAMILY
FIGURE 22-12: SYNCHRONOUS TRANSMISSION (THROUGH TXEN)
RC7/RX1/DT1/SEG28 Pin bit 0 bit 1 bit 2 bit 6 bit 7
RC6/TX1/CK1/SEG27 Pin
Write to
TXREG1 reg
TX1IF bit
TRMT bit
TXEN bit
Note: This example is equally applicable to EUSART2 (RG1/TX2/CK2/AN19/C3OUT and RG2/RX2/DT2/AN18/C3INA).
TABLE 22-7: REGISTERS ASSOCIATED WITH SYNCHRONOUS MASTER TRANSMISSION

Reset
on Page:
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used for synchronous master transmission.

PIC18F87K90 FAMILY
22.3.2 EUSART SYNCHRONOUS 3. Ensure bits, CREN and SREN, are clear.
MASTER RECEPTION 4. If interrupts are desired, set enable bit, RCxIE.
5. If 9-bit reception is desired, set bit, RX9.
Once Synchronous mode is selected, reception is
enabled by setting either the Single Receive Enable bit, 6. If a single reception is required, set bit, SREN.
SREN (RCSTAx<5>), or the Continuous Receive For continuous reception, set bit, CREN.
Enable bit, CREN (RCSTAx<4>). Data is sampled on 7. Interrupt flag bit, RCxIF, will be set when recep-
the RXx pin on the falling edge of the clock. tion is complete and an interrupt will be generated
if the enable bit, RCxIE, was set.
If enable bit, SREN, is set, only a single word is
8. Read the RCSTAx register to get the 9th bit (if
received. If enable bit, CREN, is set, the reception is enabled) and determine if any error occurred
continuous until CREN is cleared. If both bits are set,
during reception.
then CREN takes precedence.
9. Read the 8-bit received data by reading the
To set up a Synchronous Master Reception: RCREGx register.
1. Initialize the SPBRGHx:SPBRGx registers for the 10. If any error occurred, clear the error by clearing
appropriate baud rate. Set or clear the BRG16 bit CREN.
bit, as required, to achieve the desired baud rate. 11. If using interrupts, ensure that the GIE and PEIE bits
2. Enable the synchronous master serial port by in the INTCON register (INTCON<7:6>) are set.
setting bits, SYNC, SPEN and CSRC.
FIGURE 22-13: SYNCHRONOUS RECEPTION (MASTER MODE, SREN)
Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
RC7/RX1/DT1/
SEG28 Pin bit 0 bit 1 bit 2 bit 3 bit 4 bit 5 bit 6 bit 7
RC6/TX1/CK1/
SEG27 Pin
(TXCKP = 0)
RC6/TX1/CK1/
SEG27 Pin
(TXCKP = 1)
Write to
bit, SREN
SREN bit
CREN bit ‘0’ ‘0’
RC1IF bit
(Interrupt)
Read
RCREG1
Note: Timing diagram demonstrates Sync Master mode with bit, SREN = 1, and bit, BRGH = 0. This example is equally applicable to EUSART2
(RG1/TX2/CK2/AN19/C3OUT and RG2/RX2/DT2/AN18/C3INA).

PIC18F87K90 FAMILY
TABLE 22-8: REGISTERS ASSOCIATED WITH SYNCHRONOUS MASTER RECEPTION
Reset Values
on Page:
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used for synchronous master reception.

PIC18F87K90 FAMILY
22.4 EUSART Synchronous e) If enable bit, TXxIE, is set, the interrupt will wake
Slave Mode the chip from Sleep. If the global interrupt is
enabled, the program will branch to the interrupt
Synchronous Slave mode is entered by clearing bit, vector.
CSRC (TXSTAx<7>). This mode differs from the
To set up a Synchronous Slave Transmission:
Synchronous Master mode in that the shift clock is sup-
plied externally at the CKx pin (instead of being supplied 1. Enable the synchronous slave serial port by
internally in Master mode). This allows the device to setting bits, SYNC and SPEN, and clearing bit,
transfer or receive data while in any low-power mode. CSRC.
2. Clear bits, CREN and SREN.
22.4.1 EUSART SYNCHRONOUS 3. If interrupts are desired, set enable bit, TXxIE.
SLAVE TRANSMISSION 4. If 9-bit transmission is desired, set bit, TX9.
The operation of the Synchronous Master and Slave 5. Enable the transmission by setting enable bit,
modes is identical, except in the case of Sleep mode. TXEN.
If two words are written to the TXREGx and then the 6. If 9-bit transmission is selected, the ninth bit
SLEEP instruction is executed, the following will occur: should be loaded in bit, TX9D.
a) The first word will immediately transfer to the 7. Start transmission by loading data to the
TSR register and transmit. TXREGx register.
b) The second word will remain in the TXREGx 8. If using interrupts, ensure that the GIE and PEIE
register. bits in the INTCON register (INTCON<7:6>) are
c) Flag bit, TXxIF, will not be set. set.
d) When the first word has been shifted out of TSR,
the TXREGx register will transfer the second word
to the TSR and flag bit, TXxIF, will now be set.
TABLE 22-9: REGISTERS ASSOCIATED WITH SYNCHRONOUS SLAVE TRANSMISSION

Reset
on Page:
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used for synchronous slave transmission.

PIC18F87K90 FAMILY
22.4.2 EUSART SYNCHRONOUS SLAVE To set up a Synchronous Slave Reception:
RECEPTION 1. Enable the synchronous master serial port by
The operation of the Synchronous Master and Slave setting bits, SYNC and SPEN, and clearing bit,
modes is identical, except in the case of Sleep, or any CSRC.
Idle mode, and bit, SREN, which is a “don’t care” in 2. If interrupts are desired, set enable bit, RCxIE.
Slave mode. 3. If 9-bit reception is desired, set bit, RX9.
If receive is enabled by setting the CREN bit prior to 4. To enable reception, set enable bit, CREN.
entering Sleep or any Idle mode, then a word may be 5. Flag bit, RCxIF, will be set when reception is
received while in this low-power mode. Once the word complete. An interrupt will be generated if
is received, the RSR register will transfer the data to the enable bit, RCxIE, was set.
RCREGx register. If the RCxIE enable bit is set, the 6. Read the RCSTAx register to get the 9th bit (if
interrupt generated will wake the chip from the enabled) and determine if any error occurred
low-power mode. If the global interrupt is enabled, the during reception.
program will branch to the interrupt vector.
7. Read the 8-bit received data by reading the
RCREGx register.
8. If any error occurred, clear the error by clearing
bit, CREN.
set.
TABLE 22-10: REGISTERS ASSOCIATED WITH SYNCHRONOUS SLAVE RECEPTION

Reset
on Page:
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used for synchronous slave reception.

PIC18F87K90 FAMILY
23.0 12-BIT ANALOG-TO-DIGITAL 23.1 Differential A/D Converter
CONVERTER (A/D) MODULE The converter in PIC18F87K90 family devices is
The Analog-to-Digital (A/D) Converter module in the implemented as a differential A/D where the differential
PIC18F87K90 family of devices has 16 inputs for the voltage between two channels is measured and
64-pin devices and 24 inputs for the 80-pin devices. converted to digital values (see Figure 23-1).
This module allows conversion of an analog input The converter can also be configured to measure a
signal to a corresponding signed 12-bit digital number. voltage from a single input by clearing the CHSN bits
The module has these registers: (ADCON1<2:0>). With this configuration, the negative
channel input is connected internally to AVSS (see
• A/D Control Register 0 (ADCON0) Figure 23-2).
• A/D Control Register 1 (ADCON1)
• A/D Control Register 2 (ADCON2) FIGURE 23-1: DIFFERENTIAL CHANNEL
• A/D Port Configuration Register 0 (ANCON0) MEASUREMENT
• A/D Port Configuration Register 1 (ANCON1)
Positive input
•
•
A/D Port Configuration Register 2 (ANCON2)
ADRESH (the upper A/D Results register)
CHS<4:0> + ADC
• ADRESL (the lower A/D Results register)
Negative input
CHSN<2:0>
–
The ADCON0 register, shown in Register 23-1, con-
trols the operation of the A/D module. The ADCON1 Differential conversion feeds the two input channels to
register, shown in Register 23-2, configures the voltage a unity gain differential amplifier. The positive channel
reference and special trigger selection. The ADCON2 input is selected using the CHS bits (ADCON0<6:2>)
register, shown in Register 23-3, configures the A/D and the negative channel input is selected using the
clock source and programmed acquisition time and CHSN bits (ADCON1<2:0>).
justification.
The output from the amplifier is fed to the A/D convert,
as shown in Figure 23-1. The 12-bit result is available
on the ADRESH and ADRESL registers. There is also
a sign bit, along with the 12-bit result, indicating if the
result is a positive or negative value.
FIGURE 23-2: SINGLE CHANNEL

MEASUREMENT
Positive input
CHS<4:0> + ADC
CHSN<2:0>
= 000
–
AVSS
In the Single Channel Measurement mode, the

negative input is connected to AVSS by clearing the
CHSN bits (ADCON1<2:0>).

PIC18F87K90 FAMILY
23.2 A/D Registers
23.2.1 A/D CONTROL REGISTERS
REGISTER 23-1: ADCON0: A/D CONTROL REGISTER 0

— CHS4 CHS3 CHS2 CHS1 CHS0 GO/DONE ADON
bit 7 bit 0
Legend:

bit 6-2 CHS<4:0>: Analog Channel Select bits
00000 = Channel 00 (AN0) 10000 = Channel 16 (AN16)
00100 = Channel 04 (AN4) 10100 = Channel 20 (AN20)(1,2)
01000 = Channel 08 (AN8) 11000 = (Reserved)(2)
01100 = Channel 12 (AN12)(1,2) 11100 = Channel 28 (Reserved CTMU)
01101 = Channel 13 (AN13)(1,2) 11101 = Channel 29 (Internal temperature diode)
01110 = Channel 14 (AN14)(1,2) 11110 = Channel 30 (VDDCORE)
01111 = Channel 15 (AN15)(1,2) 11111 = Channel 31 (1.024V band gap)
bit 1 GO/DONE: A/D Conversion Status bit
1 = A/D (or calibration) cycle is in progress. Setting this bit starts an A/D conversion cycle. The bit is
cleared automatically by hardware when the A/D conversion is completed.
0 = A/D conversion is completed or is not in progress
bit 0 ADON: A/D On bit
1 = A/D Converter is operating
0 = A/D Converter module is shut off and consuming no operating current
Note 1: These channels are not implemented on 64-pin devices.

2: Performing a conversion on unimplemented channels will return random values.

PIC18F87K90 FAMILY
TRIGSEL1 TRIGSEL0 VCFG1 VCFG0 VNCFG CHSN2 CHSN1 CHSN0
bit 7 bit 0
Legend:
bit 7-6 TRIGSEL<1:0>: Special Trigger Select bits

11 = Selects the special trigger from the RTCC
10 = Selects the special trigger from the Timer1
01 = Selects the special trigger from the CTMU
00 = Selects the special trigger from the ECCP2
bit 5-4 VCFG<1:0>: A/D VREF+ Configuration bits
11 = Internal VREF+ (4.096V)
10 = Internal VREF+ (2.048V)
01 = External VREF+
00 = AVDD
bit 3 VNCFG: A/D VREF- Configuration bit
1 = External VREF
0 = AVSS
bit 2-0 CHSN<2:0>: Analog Negative Channel Select bits
111 = Channel 07 (AN6)
000 = Selecting ‘000’ chooses AVSS/external VREF- as a negative channel based on VNCFG

PIC18F87K90 FAMILY
ADFM — ACQT2 ACQT1 ACQT0 ADCS2 ADCS1 ADCS0
bit 7 bit 0
Legend:
bit 7 ADFM: A/D Result Format Select bit

1 = Right justified
0 = Left justified
bit 5-3 ACQT<2:0>: A/D Acquisition Time Select bits
111 = 20 TAD
110 = 16 TAD
101 = 12 TAD
100 = 8 TAD
011 = 6 TAD
010 = 4 TAD
001 = 2 TAD
000 = 0 TAD(1)
bit 2-0 ADCS<2:0>: A/D Conversion Clock Select bits
111 = FRC (clock derived from A/D RC oscillator)(1)
110 = FOSC/64
101 = FOSC/16
100 = FOSC/4
011 = FRC (clock derived from A/D RC oscillator)(1)
010 = FOSC/32
001 = FOSC/8
000 = FOSC/2
Note 1: If the A/D FRC clock source is selected, a delay of one TCY (instruction cycle) is added before the A/D
clock starts. This allows the SLEEP instruction to be executed before starting a conversion.

PIC18F87K90 FAMILY
23.2.2 A/D RESULT REGISTERS performed on the result. The results are represented as
a two's compliment binary value. This means that when
The ADRESH:ADRESL register pair is where the 12-bit
sign bits and magnitude bits are considered together in
A/D result and extended sign bits (ADSGN) are loaded
right justification, the ADRESH and ADRESL can be
at the completion of a conversion. This register pair is
read as a single signed integer value.
16 bits wide. The A/D module gives the flexibility of left
or right justifying the 12-bit result in the 16-Bit Result When the A/D Converter is disabled, these 8-bit registers
register. The A/D Format Select bit (ADFM) controls can be used as two general purpose registers.
this justification.
Figure 23-3 shows the operation of the A/D result
justification and location of the sign bit (ADSGN). The
extended sign bits allow for easier 16-bit math to be
FIGURE 23-3: A/D RESULT JUSTIFICATION
12-Bit Result Right Justified

Left Justified
ADFM = 0 ADFM = 1
ADRESH ADRESL ADRESH ADRESL
Result bits ADSGN bit
Two’s Complement Example Results Number Line

Left Justified Right Justified
Hex Decimal Hex Decimal

0xFFF0 4095 0x0FFF 4095
0xFFE0 4094 0x0FFE 4094
… … … …
0x0020 2 0x0002 2
0x0010 1 0x0001 1
0x0000 0 0x0000 0
0xFFFF -1 0xFFFF -1
0xFFEF -2 0xFFFE -2
… … … …
0x001F -4095 0xF001 -4095
0x000F -4096 0xF000 -4096

PIC18F87K90 FAMILY
REGISTER 23-4: ADRESH: A/D RESULT HIGH BYTE REGISTER, LEFT JUSTIFIED (ADFM = 0)
ADRES11 ADRES10 ADRES9 ADRES8 ADRES7 ADRES6 ADRES5 ADRES4
bit 7 bit 0
Legend:
bit 7-0 ADRES<11:4>: A/D Result High Byte bits
REGISTER 23-5: ADRESL: A/D RESULT LOW BYTE REGISTER, LEFT JUSTIFIED (ADFM = 0)
ADRES3 ADRES2 ADRES1 ADRES0 ADSGN ADSGN ADSGN ADSGN
bit 7 bit 0
Legend:
bit 7-4 ADRES<3:0>: A/D Result Low Byte bits

bit 3-0 ADSGN: A/D Result Sign bits
1 = A/D result is negative
0 = A/D result is positive

PIC18F87K90 FAMILY
REGISTER 23-6: ADRESH: A/D RESULT HIGH BYTE REGISTER, RIGHT JUSTIFIED (ADFM = 1)
ADSGN ADSGN ADSGN ADSGN ADRES11 ADRES10 ADRES9 ADRES8
bit 7 bit 0
Legend:
bit 7-4 ADSGN: A/D Result Sign bits

1 = A/D result is negative
0 = A/D result is positive
bit 3-0 ADRES<11:8>: A/D Result High Byte bits
REGISTER 23-7: ADRESL: A/D RESULT LOW BYTE REGISTER, RIGHT JUSTIFIED (ADFM = 1)
ADRES7 ADRES6 ADRES5 ADRES4 ADRES3 ADRES2 ADRES1 ADRES0
bit 7 bit 0
Legend:
bit 7-0 ADRES<7:0>: A/D Result Low Byte bits

PIC18F87K90 FAMILY
The ANCONx registers are used to configure the input for either the A/D Converter or the comparator
operation of the I/O pin associated with each analog module, with all digital peripherals disabled and digital
channel. Clearing a ANSELx bit configures the inputs read as ‘0’.
corresponding pin (ANx) to operate as a digital only I/O. As a rule, I/O pins that are multiplexed with analog
Setting a bit configures the pin to operate as an analog inputs default to analog operation on any device Reset.
REGISTER 23-8: ANCON0: A/D PORT CONFIGURATION REGISTER 0

ANSEL7 ANSEL6 ANSEL5 ANSEL4 ANSEL3 ANSEL2 ANSEL1 ANSEL0
bit 7 bit 0
Legend:
bit 7-0 ANSEL<7:0>: Analog Port Configuration bits (AN7 and AN0)
0 = Pin is configured as a digital port
1 = Pin is configured as an analog channel – digital input disabled and any inputs read as ‘0’

(1)
ANSEL15 ANSEL14(1) ANSEL13(1) ANSEL12(1) ANSEL11 ANSEL10 ANSEL9 ANSEL8
bit 7 bit 0
Legend:
bit 7-0 ANSEL<15:8>: Analog Port Configuration bits (AN15 through AN8)
0 = Pin is configured as a digital port
1 = Pin is configured as an analog channel – digital input is disabled and any inputs read as ‘0’
Note 1: AN12 through AN15, and AN20 to AN23, are implemented only on 80-pin devices. For 64-pin devices, the
corresponding ANSELx bits are still implemented for these channels, but have no effect.

PIC18F87K90 FAMILY
ANSEL23(1) ANSEL22(1) ANSEL21(1) ANSEL20(1) ANSEL19 ANSEL18 ANSEL17 ANSEL16
bit 7 bit 0
Legend:
bit 7-0 ANSEL<23:16>: Analog Port Configuration bits (AN23 through AN16)
0 = Pin configured as a digital port
1 = Pin configured as an analog channel — digital input disabled and any inputs read as ‘0’
Note 1: AN12 through AN15, and AN20 to AN23, are implemented only on 80-pin devices. For 64-pin devices, the
corresponding ANSELx bits are still implemented for these channels, but have no effect.
The analog reference voltage is software-selectable to Each port pin associated with the A/D Converter can be
either the device’s positive and negative supply voltage configured as an analog input or a digital I/O. The
(AVDD and AVSS) or the voltage level on the ADRESH and ADRESL registers contain the result of
RA3/AN3/VREF+ and RA2/AN2/VREF- pins. VREF+ has the A/D conversion. When the A/D conversion is com-
two additional internal voltage reference selections: plete, the result is loaded into the ADRESH:ADRESL
2.048V and 4.096V. register pair, the GO/DONE bit (ADCON0<1>) is
The A/D Converter can uniquely operate while the cleared and the A/D Interrupt Flag bit, ADIF (PIR1<6>),
device is in Sleep mode. To operate in Sleep, the A/D is set.
conversion clock must be derived from the A/D A device Reset forces all registers to their Reset state.
Converter’s internal RC oscillator. This forces the A/D module to be turned off and any
The output of the Sample-and-Hold (S/H) is the input conversion in progress is aborted. The value in the
into the converter, which generates the result via ADRESH:ADRESL register pair is not modified for a
successive approximation. Power-on Reset. These registers will contain unknown
data after a Power-on Reset.
The block diagram of the A/D module is shown in
Figure 23-4.

PIC18F87K90 FAMILY
FIGURE 23-4: A/D BLOCK DIAGRAM
CHS<4:0>
11111
1.024V Band Gap
11110
VDDCORE
11101 Reserved
Temperature Diode
11100
Reserved CTMU
11011
(Unimplemented)
11010
(Unimplemented)
11001
(Unimplemented)
11000
(Unimplemented)
10111
AN23(1)
12-Bit
A/D 10110
Converter AN22(1)
00100
AN4
00011
AN3
00010
AN2
00001
AN1
00000
AN0
Positive Input Voltage CHSN<2:0>
111
Negative Input Voltage AN6
110
AN5
Reference
Voltage
VCFG<1:0>
11 Internal VREF+ 001

(4.096V) AN0
VREF+ Internal VREF+ 000
10 AVSS
(2.048V)
01
AN3
VREF- VDD
00
VNCFG
AN2
VSS(2)
Note 1: Channels, AN12 through AN15, and AN20 through AN23, are not available on 64-pin devices.
2: I/O pins have diode protection to VDD and VSS.

PIC18F87K90 FAMILY
After the A/D module has been configured as desired, 2. Configure the A/D interrupt (if desired):
the selected channel must be acquired before the • Clear the ADIF bit (PIR1<6>)
conversion can start. The analog input channels must • Set the ADIE bit (PIE1<6>)
have their corresponding TRIS bits selected as inputs.
• Set the GIE bit (INTCON<7>)
To determine acquisition time, see Section 23.3 “A/D
Acquisition Requirements”. After this acquisition 3. Wait the required acquisition time (if required).
time has elapsed, the A/D conversion can be started. 4. Start the conversion:
An acquisition time can be programmed to occur • Set the GO/DONE bit (ADCON0<1>)
between setting the GO/DONE bit and the actual start 5. Wait for A/D conversion to complete, by either:
of the conversion.
• Polling for the GO/DONE bit to be cleared
To do an A/D conversion, follow these steps:
OR
1. Configure the A/D module:
• Waiting for the A/D interrupt
• Configure the required ADC pins as analog
6. Read A/D Result registers (ADRESH:ADRESL),
pins (ANCON0, ANCON1 and ANCON2)
and if required, clear bit, ADIF.
• Set the voltage reference (ADCON1)
7. For the next conversion, begin with Step 1 or 2,
• Select the A/D positive and negative input as required.
channels (ADCON0 and ADCON1)
The A/D conversion time per bit is defined as TAD.
• Select the A/D acquisition time (ADCON2)
Before the next acquisition starts, a minimum Wait
• Select the A/D conversion clock (ADCON2) of 2 TAD is required.
• Turn on the A/D module (ADCON0)
FIGURE 23-5: ANALOG INPUT MODEL

VDD
Sampling
Switch
VT = 0.6V
ANx RIC 1k SS RSS
RS
VAIN CPIN ILEAKAGE

VT = 0.6V CHOLD = 25 pF
5 pF ±100 nA
VSS
Legend: CPIN = Input Capacitance

VT = Threshold Voltage
ILEAKAGE = Leakage Current at the pin due to VDD
various junctions
RIC = Interconnect Resistance
SS = Sampling Switch
CHOLD = Sample/Hold Capacitance (from DAC)
RSS = Sampling Switch Resistance 1 2 3 4
Sampling Switch (k)

PIC18F87K90 FAMILY
23.3 A/D Acquisition Requirements To calculate the minimum acquisition time,
Equation 23-1 can be used. This equation assumes
For the A/D Converter to meet its specified accuracy, the that 1/2 LSb error is used (1,024 steps for the A/D). The
charge holding capacitor (CHOLD) must be allowed to 1/2 LSb error is the maximum error allowed for the A/D
fully charge to the input channel voltage level. The to meet its specified resolution.
analog input model is shown in Figure 23-5. The source
impedance (RS) and the internal sampling switch (RSS) Equation 23-3 shows the calculation of the minimum
impedance directly affect the time required to charge the required acquisition time, TACQ. This calculation is
capacitor CHOLD. The sampling switch (RSS) impedance based on the following application system
varies over the device voltage (VDD). assumptions:
The source impedance affects the offset voltage at the • CHOLD = 25 pF

analog input (due to pin leakage current). The maxi- • Rs = 2.5 k
mum recommended impedance for analog sources • Conversion Error  1/2 LSb
is 2.5 k. After the analog input channel is selected or • VDD = 3V  Rss = 2 k
changed, the channel must be sampled for at least the
• Temperature = 85C
minimum acquisition time before starting a conversion.
Note: When the conversion is started, the
holding capacitor is disconnected from the
input pin.
EQUATION 23-1: ACQUISITION TIME

TACQ = Amplifier Settling Time + Holding Capacitor Charging Time + Temperature Coefficient
= TAMP + TC + TCOFF
EQUATION 23-2: A/D MINIMUM CHARGING TIME

VHOLD = (VREF – (VREF/2048)) • (1 – e(-TC/CHOLD(RIC + RSS + RS)))
or
TC = -(CHOLD)(RIC + RSS + RS) ln(1/2048)
EQUATION 23-3: CALCULATING THE MINIMUM REQUIRED ACQUISITION TIME

TACQ = TAMP + TC + TCOFF
TAMP = 0.2 s
TCOFF = (Temp – 25C)(0.02 s/C)
(85C – 25C)(0.02 s/C)
1.2 s
Temperature coefficient is only required for temperatures > 25C. Below 25C, TCOFF = 0 ms.
TC = -(CHOLD)(RIC + RSS + RS) ln(1/2048) s
-(25 pF) (1 k + 2 k + 2.5 k) ln(0.0004883) s
1.05 s
TACQ = 0.2 s + 1.05 s + 1.2 s
2.45 s

PIC18F87K90 FAMILY
23.4 Selecting and Configuring TABLE 23-1: TAD vs. DEVICE OPERATING
Automatic Acquisition Time FREQUENCIES
The ADCON2 register allows the user to select an AD Clock Source (TAD) Maximum Device
acquisition time that occurs each time the GO/DONE Operation ADCS<2:0> Frequency
bit is set.
2 TOSC 000 2.50 MHz
When the GO/DONE bit is set, sampling is stopped and
a conversion begins. The user is responsible for ensur- 4 TOSC 100 5.00 MHz
ing the required acquisition time has passed between 8 TOSC 001 10.00 MHz
selecting the desired input channel and setting the 16 TOSC 101 20.00 MHz
GO/DONE bit.
32 TOSC 010 40.00 MHz
This occurs when the ACQT<2:0> bits
64 TOSC 110 64.00 MHz
(ADCON2<5:3>) remain in their Reset state (‘000’),
which is compatible with devices that do not offer RC(2) x11 1.00 MHz(1)
programmable acquisition times. Note 1: The RC source has a typical TAD time of
If desired, the ACQTx bits can be set to select a pro- 4 s.
grammable acquisition time for the A/D module. When 2: For device frequencies above 1 MHz, the
the GO/DONE bit is set, the A/D module continues to device must be in Sleep mode for the
sample the input for the selected acquisition time, then entire conversion or the A/D accuracy
automatically begins a conversion. Since the acquisi- may be out of specification.
tion time is programmed, there may be no need to wait
for an acquisition time between selecting a channel and
setting the GO/DONE bit.
23.6 Configuring Analog Port Pins
In either case, when the conversion is completed, the The ANCON0, ANCON1, ANCON2, TRISA, TRISF,
GO/DONE bit is cleared, the ADIF flag is set and the TRISG and TRISH registers control the operation of the
A/D begins sampling the currently selected channel A/D port pins. The port pins needed as analog inputs
again. If an acquisition time is programmed, there is must have their corresponding TRISx bits set (input). If
nothing to indicate if the acquisition time has ended or the TRISx bit is cleared (output), the digital output level
if the conversion has begun. (VOH or VOL) will be converted.
The A/D operation is independent of the state of the
23.5 Selecting the A/D Conversion CHS<3:0> bits and the TRISx bits.
Clock
Note 1: When reading the PORT register, all pins
The A/D conversion time per bit is defined as TAD. The configured as analog input channels will
A/D conversion requires 14 TAD per 12-bit conversion. read as cleared (a low level). Pins config-
The source of the A/D conversion clock is ured as digital inputs will convert an
software-selectable. analog input. Analog levels on a digitally
The possible options for TAD are: configured input will be accurately
converted.
• 2 TOSC
2: Analog levels on any pin defined as a
• 4 TOSC
digital input may cause the digital input
• 8 TOSC buffer to consume current out of the
• 16 TOSC device’s specification limits.
• 32 TOSC
• 64 TOSC
• Using the internal RC Oscillator
For correct A/D conversions, the A/D conversion clock
(TAD) must be as short as possible but greater than the
minimum TAD. (For more information, see
Parameter 130 in Table 31-26.)
Table 23-1 shows the resultant TAD times derived from
the device operating frequencies and the A/D clock
source selected.

PIC18F87K90 FAMILY
23.7 A/D Conversions ADRESH:ADRESL registers will continue to contain
the value of the last completed conversion (or the last
Figure 23-6 shows the operation of the A/D Converter value written to the ADRESH:ADRESL registers).
after the GO/DONE bit has been set and the
ACQT<2:0> bits are cleared. A conversion is started After the A/D conversion is completed or aborted, a
after the following instruction to allow entry into Sleep 2 TAD Wait is required before the next acquisition can
mode before the conversion begins. be started. After this Wait, acquisition on the selected
channel is automatically started.
Figure 23-7 shows the operation of the A/D Converter
after the GO/DONE bit has been set, the ACQT<2:0> Note: The GO/DONE bit should NOT be set in
bits set to ‘010’ and a 4 TAD acquisition time selected. the same instruction that turns on the A/D.
Clearing the GO/DONE 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. This means the
FIGURE 23-6: A/D CONVERSION TAD CYCLES (ACQT<2:0> = 000, TACQ = 0)

TCY - TAD TAD1 TAD2 TAD3 TAD4 TAD5 TAD6 TAD7 TAD8 TAD9 TAD10 TAD11 TAD12 TAD13
b11 b10 b9 b8 b7 b6 b5 b4 b3 b2 b1 b0
Conversion starts
Holding capacitor is disconnected from analog input (typically 100 ns)
Set GO/DONE bit
Next Q4: ADRESH:ADRESL is loaded, GO/DONE bit is cleared,

ADIF bit is set, holding capacitor is connected to analog input.
FIGURE 23-7: A/D CONVERSION TAD CYCLES (ACQT<2:0> = 010, TACQ = 4 TAD)
TACQT Cycles TAD Cycles
1 2 3 4 1 2 3 4 5 6 7 8 9 10 11 12 13
b11 b10 b9 b8 b7 b6 b5 b4 b3 b2 b1 b0
Automatic
Acquisition Conversion starts
Time (Holding capacitor is disconnected)
Set GO/DONE bit

(Holding capacitor continues
acquiring input) Next Q4: ADRESH:ADRESL is loaded, GO/DONE bit is cleared,
ADIF bit is set, holding capacitor is reconnected to analog input.

PIC18F87K90 FAMILY
23.8 Use of the Special Event Triggers 23.9 Operation in Power-Managed
A/D conversion can be started by the Special Event
Modes
Trigger of any of these modules: The selection of the automatic acquisition time and A/D
• ECCP2 – Requires CCP2M<3:0> bits conversion clock is determined, in part, by the clock
(CCP2CON<3:0>) set at ‘1011’ source and frequency while in a power-managed
• CTMU – Requires the setting of the CTTRIG bit mode.
(CTMUCONH<0>) If the A/D is expected to operate while the device is in
• Timer1 a power-managed mode, the ACQT<2:0> and
• RTCC ADCS<2:0> bits in ADCON2 should be updated in
accordance with the power-managed mode clock that
To start an A/D conversion: will be used.
• The A/D module must be enabled (ADON = 1) After the power-managed mode is entered (either of the
• The appropriate analog input channel is selected power-managed Run modes), an A/D acquisition or
• The minimum acquisition period is set in one of conversion may be started. Once an acquisition or
these ways: conversion is started, the device should continue to be
- Timing provided by the user clocked by the same power-managed mode clock source
until the conversion has been completed. If desired, the
- Selection made of an appropriate TACQ time
device may be placed into the corresponding
With these conditions met, the trigger sets the power-managed Idle mode during the conversion.
GO/DONE bit and the A/D acquisition starts.
If the power-managed mode clock frequency is less
If the A/D module is not enabled (ADON = 0), the than 1 MHz, the A/D RC clock source should be
module ignores the Special Event Trigger. selected.
Note: With an ECCP2 trigger, Timer1 or Timer3 Operation in Sleep mode requires that the A/D RC
is cleared. The timers reset to automati- clock be selected. If bits, ACQT<2:0>, are set to ‘000’
cally repeat the A/D acquisition period and a conversion is started, the conversion will be
with minimal software overhead (moving delayed one instruction cycle to allow execution of the
ADRESH:ADRESL to the desired loca- SLEEP instruction and entry into Sleep mode. The
tion). If the A/D module is not enabled, the IDLEN and SCS<1:0> bits in the OSCCON register
Special Event Trigger is ignored by the must have already been cleared prior to starting the
module, but the timer’s counter resets. conversion.

PIC18F87K90 FAMILY
TABLE 23-2: SUMMARY OF A/D REGISTERS
Reset
on Page:
ADRESH A/D Result Register High Byte 76
ADRESL A/D Result Register Low Byte 76
ADCON0 — CHS4 CHS3 CHS2 CHS1 CHS0 GO/DONE ADON 76
ADCON1 TRIGSEL1 TRIGSEL0 VCFG1 VCFG0 VNCFG CHSN2 CHSN1 CHSN0 76
ADCON2 ADFM — ACQT2 ACQT1 ACQT0 ADCS2 ADCS1 ADCS0 76
PORTA RA7(2) RA6(2) RA5 RA4 RA3 RA2 RA1 RA0 78
PORTG — — RG5(3) RG4 RG3 RG2 RG1 RG0 78
PORTH(1) RH7 RH6 RH5 RH4 RH3 RH2 RH1 RH0 78
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used for A/D conversion.
Note 1: This register is not implemented on 64-pin devices.
2: These bits are available only in certain oscillator modes, when the OSC2 Configuration bit = 0. If that
Configuration bit is cleared, this signal is not implemented.
3: This bit is available when Master Clear is disabled (MCLRE = 0). When MCLRE is set, the bit is
unimplemented.

PIC18F87K90 FAMILY
24.0 COMPARATOR MODULE 24.1 Registers
The analog comparator module contains three compar- The CMxCON registers (CM1CON, CM2CON and
ators that can be independently configured in a variety CM3CON) select the input and output configuration for
of ways. The inputs can be selected from the analog each comparator, as well as the settings for interrupt
inputs and two internal voltage references. The digital generation (see Register 24-1).
outputs are available at the pin level and can also be The CMSTAT register (Register 24-2) provides the out-
read through the control register. Multiple output and put results of the comparators. The bits in this register
interrupt event generation are also available. A generic are read-only.
single comparator from the module is shown in
Figure 24-1.
Key features of the module includes:
• Independent comparator control
• Programmable input configuration
• Output to both pin and register levels
• Programmable output polarity
• Independent interrupt generation for each
comparator with configurable interrupt-on-change
FIGURE 24-1: COMPARATOR SIMPLIFIED BLOCK DIAGRAM
CCH<1:0> CMPxOUT
(CMSTAT<7:5>)
CxINB 0
CxINC(2) 1
Interrupt
C2INB/C2IND (1,2) 2 CMPxIF
Logic
VBG 3
EVPOL<1:0>
CREF COE
VIN- CxOUT
Polarity
CxINA 0 VIN+ Cx Logic
CVREF 1
CON CPOL
Note 1: Comparators, 1 and 3, use C2INB as an input to the inverting terminal. Comparator 2 uses C2IND as an input to
the inverted terminal.
2: C1INC, C2INC and C2IND are all unavailable on 64-pin devices (PIC18F6XK90).

PIC18F87K90 FAMILY
REGISTER 24-1: CMxCON: COMPARATOR CONTROL x REGISTER
CON COE CPOL EVPOL1 EVPOL0 CREF CCH1 CCH0
bit 7 bit 0
Legend:
bit 7 CON: Comparator Enable bit

1 = Comparator is enabled
0 = Comparator is disabled
bit 6 COE: Comparator Output Enable bit
1 = Comparator output is present on the CxOUT pin
0 = Comparator output is internal only
bit 5 CPOL: Comparator Output Polarity Select bit
1 = Comparator output is inverted
0 = Comparator output is not inverted
bit 4-3 EVPOL<1:0>: Interrupt Polarity Select bits
11 = Interrupt generation on any change of the output(1)
10 = Interrupt generation only on high-to-low transition of the output
01 = Interrupt generation only on low-to-high transition of the output
00 = Interrupt generation is disabled
bit 2 CREF: Comparator Reference Select bit (non-inverting input)
1 = Non-inverting input connects to the internal CVREF voltage
0 = Non-inverting input connects to the CxINA pin
bit 1-0 CCH<1:0>: Comparator Channel Select bits
11 = Inverting input of the comparator connects to VBG
10 = Inverting input of the comparator connects to the C2INB or C2IND pin(2,3)
01 = Inverting input of the comparator connects to the CxINC pin(3)
00 = Inverting input of the comparator connects to the CxINB pin
Note 1: The CMPxIF bit is automatically set any time this mode is selected and must be cleared by the application
after the initial configuration.
2: Comparators, 1 and 3, use C2INB as an input to the inverting terminal; Comparator 2 uses C2IND.
3: C1INC, C2INC and C2IND are all unavailable for 64-pin devices (PIC18F6XK90).

PIC18F87K90 FAMILY
REGISTER 24-2: CMSTAT: COMPARATOR STATUS REGISTER
R-1 R-1 R-1 U-0 U-0 U-0 U-0 U-0
CMP3OUT CMP2OUT CMP1OUT — — — — —
bit 7 bit 0
Legend:
bit 7-5 CMPxOUT<3:1>: Comparator x Status bits

If CPOL (CMxCON<5>)= 0 (non-inverted polarity):
1 = Comparator x’s VIN+ > VIN-
0 = Comparator x’s VIN+ < VIN-
If CPOL = 1 (inverted polarity):
1 = Comparator x’s VIN+ < VIN-
0 = Comparator x’s VIN+ > VIN-

PIC18F87K90 FAMILY
24.2 Comparator Operation 24.3 Comparator Response Time
A single comparator is shown in Figure 24-2, along with Response time is the minimum time, after selecting a
the relationship between the analog input levels and new reference voltage or input source, before the com-
the digital output. When the analog input at VIN+ is less parator output has a valid level. The response time of
than the analog input, VIN-, the output of the compara- the comparator differs from the settling time of the
tor is a digital low level. When the analog input at VIN+ voltage reference. Therefore, both of these times must
is greater than the analog input, VIN-, the output of the be considered when determining the total response to
comparator is a digital high level. The shaded areas of a comparator input change. Otherwise, the maximum
the output of the comparator, in Figure 24-2, represent delay of the comparators should be used (see
the uncertainty due to input offsets and response time. Section 31.0 “Electrical Characteristics”).
FIGURE 24-2: SINGLE COMPARATOR 24.4 Analog Input Connection

Considerations
VIN- – A simplified circuit for an analog input is shown in
Output Figure 24-3. Since the analog pins are connected to a
VIN+ + digital output, they have reverse biased diodes to VDD
and VSS. The analog input, therefore, must be between
VSS and VDD. If the input voltage deviates from this
range by more than 0.6V in either direction, one of the
diodes is forward biased and a latch-up condition may
VIN-
occur.
VIN+ A maximum source impedance of 10 k is
recommended for the analog sources. Any external
component connected to an analog input pin, such as
Output a capacitor or a Zener diode, should have very little
leakage current.
FIGURE 24-3: COMPARATOR ANALOG INPUT MODEL
VDD
VT = 0.6V RIC
RS
Comparator
<10k Input
AIN
CPIN ILEAKAGE
VA VT = 0.6V ±100 nA
5 pF
VSS
Legend: CPIN = Input Capacitance

VT = Threshold Voltage
ILEAKAGE = Leakage Current at the pin due to various junctions
RIC = Interconnect Resistance
RS = Source Impedance
VA = Analog Voltage

PIC18F87K90 FAMILY
24.5 Comparator Control and operating mode, either an external or internal voltage
Configuration reference may be used. For external analog pins that
are unavailable in 64-pin devices (C1INC, C2INC and
Each comparator has up to eight possible combina- C2IND), the corresponding configurations that use
tions of inputs: up to four external analog inputs and them as inputs are unavailable.
one of two internal voltage references.
The analog signal present at VIN- is compared to the
All of the comparators allow a selection of the signal from signal at VIN+ and the digital output of the comparator
pin, CxINA, or the voltage from the Comparator Voltage is adjusted accordingly.
Reference (CVREF) on the non-inverting channel. This is
The external reference is used when CREF = 0
compared to either CxINB, CxINC, C2INB/C2IND or the
(CMxCON<2>) and VIN+ is connected to the CxINA
microcontroller’s fixed internal reference voltage (VBG,
pin. When external voltage references are used, the
1.024V nominal) on the inverting channel. The
comparator module can be configured to have the
comparator inputs and outputs are tied to fixed I/O pins,
reference sources externally. The reference signal
defined in Table 24-1. The available comparator config-
must be between VSS and VDD, and can be applied to
urations and their corresponding bit settings are shown
either pin of the comparator.
in Figure 24-4.
The comparator module also allows the selection of an
TABLE 24-1: COMPARATOR INPUTS AND internally generated voltage reference from the Compar-
ator Voltage Reference (CVREF) module. This module is
OUTPUTS
described in more detail in Section 25.0 “Comparator
Comparator Input or Output I/O Pin Voltage Reference Module”. The reference from the
comparator voltage reference module is only available
C1INA (VIN+) RF6
when CREF = 1. In this mode, the internal voltage
C1INB (VIN-) RF5 reference is applied to the comparator’s VIN+ pin.
1 C1INC(1) (VIN-) RH6
Note: The comparator input pin, selected by
C2INB (VIN-) RF3 CCH<1:0>, must be configured as an input
C1OUT RF2 by setting both the corresponding TRISF,
C2INA (VIN+) RF4 TRISG or TRISH bit and the corresponding
ANSELx bit in the ANCONx register.
C2INB (VIN-) RF3
2 (1) (VIN-)
C2INC RH4 24.5.2 COMPARATOR ENABLE AND
C2IND(1) (VIN-) RH5 OUTPUT SELECTION
C2OUT RF1 The comparator outputs are read through the CMSTAT
C3INA (VIN+) RG2 register. The CMSTAT<5> bit reads the Comparator 1
C3INB (VIN-) RG3 output, CMSTAT<6> reads Comparator 2 output and
CMSTAT<7> reads Comparator 3 output. These bits
3 C3INC (VIN-) RG4 are read-only.
C2INB (VIN-) RF3
The comparator outputs may also be directly output to
C3OUT RG1 the RF2, RF1 and RG1 I/O pins by setting the COE bit
Note 1: C1INC, C2INC and C2IND are all (CMxCON<6>). When enabled, multiplexers in the
unavailable for 64-pin devices output path of the pins switch to the output of the
(PIC18F6XK90). comparator. While in this mode, the TRISF<2:1> and
TRISG<1> bits still function as the digital output enable
24.5.1 COMPARATOR ENABLE AND bits for the RF2, RF1 and RG1 pins.
INPUT SELECTION By default, the comparator’s output is at logic high
whenever the voltage on VIN+ is greater than on VIN-.
Setting the CON bit of the CMxCON register
The polarity of the comparator outputs can be inverted
(CMxCON<7>) enables the comparator for operation.
using the CPOL bit (CMxCON<5>).
Clearing the CON bit disables the comparator, resulting
in minimum current consumption. The uncertainty of each of the comparators is related to
the input offset voltage and the response time given in
The CCH<1:0> bits in the CMxCON register
the specifications, as discussed in Section 24.2
(CMxCON<1:0>) direct either one of three analog input
“Comparator Operation”.
pins, or the Internal Reference Voltage (VBG), to the
comparator, VIN-. Depending on the comparator

PIC18F87K90 FAMILY
FIGURE 24-4: COMPARATOR CONFIGURATIONS
Comparator Off
CON = 0, CREF = x, CCH<1:0> = xx
COE
VIN-
VIN+ Cx
Off (Read as ‘0’) CxOUT
Pin
Comparator CxINB > CxINA Compare Comparator CxINC > CxINA Compare(2,3)
CON = 1, CREF = 0, CCH<1:0> = 00 CON = 1, CREF = 0, CCH<1:0> = 01
COE COE
CxINB VIN- CxINC VIN-

VIN+ Cx VIN+ Cx
CxINA CxOUT CxINA CxOUT
Pin Pin
Comparator CxIND > CxINA Compare(3) Comparator VIRV > CxINA Compare
COE COE
C2INB/ VIN- VIN-
C2IND VBG(1)
VIN+ Cx VIN+ Cx
CxINA CxOUT CxINA CxOUT
Pin Pin
Comparator CxINB > CVREF Compare Comparator CxINC > CVREF Compare(2,3)
COE COE
CxINB VIN- CxINC VIN-

VIN+ Cx VIN+ Cx
CVREF CxOUT CVREF CxOUT
Pin Pin
Comparator CxIND > CVREF Compare(3) Comparator VIRV > CVREF Compare
COE COE
CxINB/ VIN- VIN-
CxIND VBG(1)
VIN+ Cx VIN+ Cx
CVREF CxOUT CVREF
CxOUT
Pin Pin
Note 1: VBG is the Internal Reference Voltage (1.024V nominal).

2: Configuration is unavailable for CM1CON on 64-pin devices (PIC18F6XK90).
3: Configuration is unavailable for CM2CON on 64-pin devices (PIC18F6XK90).

PIC18F87K90 FAMILY
24.6 Comparator Interrupts When EVPOL<1:0> = 11, the comparator interrupt flag
is set whenever there is a change in the output value of
The comparator interrupt flag is set whenever any of either comparator. Software will need to maintain infor-
the following occurs: mation about the status of the output bits, as read from
• Low-to-high transition of the comparator output CMSTAT<7:5>, to determine the actual change that
• High-to-low transition of the comparator output occurred.
• Any change in the comparator output The CMPxIF bits (PIR6<2:0>) are the Comparator
Interrupt Flags. The CMPxIF bits must be reset by
The comparator interrupt selection is done by the
clearing them. Since it is also possible to write a ‘1’ to
EVPOL<1:0> bits in the CMxCON register
this register, a simulated interrupt may be initiated.
(CMxCON<4:3>).
Table 24-2 shows the interrupt generation with respect
In order to provide maximum flexibility, the output of the to comparator input voltages and EVPOL bit settings.
comparator may be inverted using the CPOL bit in the
Both the CMPxIE bits (PIE6<2:0>) and the PEIE bit
CMxCON register (CMxCON<5>). This is functionally
(INTCON<6>) must be set to enable the interrupt. In
identical to reversing the inverting and non-inverting
addition, the GIE bit (INTCON<7>) must also be set. If
inputs of the comparator for a particular mode.
any of these bits are clear, the interrupt is not enabled,
An interrupt is generated on the low-to-high or high-to- though the CMPxIF bits will still be set if an interrupt
low transition of the comparator output. This mode of condition occurs.
interrupt generation is dependent on EVPOL<1:0> in
A simplified diagram of the interrupt section is shown in
the CMxCON register. When EVPOL<1:0> = 01 or 10,
Figure 24-3.
the interrupt is generated on a low-to-high or high-to-
low transition of the comparator output. Once the Note: CMPxIF will not be set when EVPOL<1:0> = 00.
interrupt is generated, it is required to clear the interrupt
flag by software.
TABLE 24-2: COMPARATOR INTERRUPT GENERATION

Comparator Interrupt
CPOL EVPOL<1:0> CxOUT Transition
Input Change Generated
VIN+ > VIN- Low-to-High No
00
VIN+ < VIN- High-to-Low No
VIN+ > VIN- Low-to-High Yes
01
VIN+ < VIN- High-to-Low No
0
VIN+ > VIN- Low-to-High No
10
VIN+ < VIN- High-to-Low Yes
VIN+ > VIN- Low-to-High Yes
11
VIN+ < VIN- High-to-Low Yes
VIN+ > VIN- High-to-Low No
00
VIN+ < VIN- Low-to-High No
VIN+ > VIN- High-to-Low No
01
VIN+ < VIN- Low-to-High Yes
1
VIN+ > VIN- High-to-Low Yes
10
VIN+ < VIN- Low-to-High No
VIN+ > VIN- High-to-Low Yes
11
VIN+ < VIN- Low-to-High Yes

PIC18F87K90 FAMILY
24.7 Comparator Operation To minimize power consumption while in Sleep mode,
During Sleep turn off the comparators (CON = 0) before entering
Sleep. If the device wakes up from Sleep, the contents
When a comparator is active and the device is placed of the CMxCON register are not affected.
in Sleep mode, the comparator remains active and the
interrupt is functional, if enabled. This interrupt will 24.8 Effects of a Reset
wake up the device from Sleep mode, when enabled.
Each operational comparator will consume additional A device Reset forces the CMxCON registers to their
current. Reset state. This forces both comparators and the
voltage reference to the OFF state.
TABLE 24-3: REGISTERS ASSOCIATED WITH COMPARATOR MODULE

Reset
Page:
CM1CON CON COE CPOL EVPOL1 EVPOL0 CREF CCH1 CCH0 80
CVRCON CVREN CVROE CVRSS CVR4 CVR3 CVR2 CVR1 CVR0 77
CMSTAT CMP3OUT CMP2OUT CMP1OUT — — — — — 77
LATF LATF7 LATF6 LATF5 LATF4 LATF3 LATF2 LATF1 — 78
PORTG — — RG5 RG4 RG3 RG2 RG1 RG0 78
LATG — — — LATG4 LATG3 LATG2 LATG1 LATG0 78
(1)
PORTH RH7 RH6 RH5 RH4 RH3 RH2 RH1 RH0 78
LATH(1) LATH7 LATH6 LATH5 LATH4 LATH3 LATH2 LATH1 LATH0 78
Legend: — = unimplemented, read as ‘0’.
Note 1: This register is not implemented on 64-pin devices.

PIC18F87K90 FAMILY
25.0 COMPARATOR VOLTAGE EQUATION 25-1:
REFERENCE MODULE If CVRSS = 1:
The comparator voltage reference is a 32-tap resistor CVREF = (VREF-) + (CVR<4:0>/32) • (VREF+ – VREF-)
ladder network that provides a selectable reference
voltage. Although its primary purpose is to provide a
If CVRSS = 0:
reference for the analog comparators, it may also be
used independently of them. CVREF = (AVSS) + (CVR<4:0>/32) • (AVDD – AVSS)
A block diagram of the module is shown in Figure 25-1.
The resistor ladder is segmented to provide a range of The comparator reference supply voltage can come
CVREF values and has a power-down function to from either VDD and VSS, or the external VREF+ and
conserve power when the reference is not being used. VREF- that are multiplexed with RA3 and RA2. The
The module’s supply reference can be provided from voltage source is selected by the CVRSS bit
either device VDD/VSS or an external voltage reference. (CVRCON<5>).
The settling time of the comparator voltage reference
25.1 Configuring the Comparator must be considered when changing the CVREF
Voltage Reference output (see Table 31-2 in Section 31.0 “Electrical
Characteristics”).
The comparator voltage reference module is controlled
through the CVRCON register (Register 25-1). The
comparator voltage reference provides a range of
output voltage with 32 levels.
The CVR<4:0> selection bits (CVRCON<4:0>) offer a
range of output voltages. Equation 25-1 shows how the
comparator voltage reference is computed.
REGISTER 25-1: CVRCON: COMPARATOR VOLTAGE REFERENCE CONTROL REGISTER

CVREN CVROE CVRSS CVR4 CVR3 CVR2 CVR1 CVR0
bit 7 bit 0
Legend:
bit 7 CVREN: Comparator Voltage Reference Enable bit

1 = CVREF circuit is powered on
0 = CVREF circuit is powered down
bit 6 CVROE: Comparator VREF Output Enable bit
1 = CVREF voltage level is output on the CVREF pin
0 = CVREF voltage level is disconnected from the CVREF pin
bit 5 CVRSS: Comparator VREF Source Selection bit
1 = Comparator reference source: CVRSRC = VREF+ – VREF-
0 = Comparator reference source: CVRSRC = AVDD – AVSS
bit 4-0 CVR<4:0>: Comparator VREF Value Selection (0  CVR<4:0>  31) bits
When CVRSS = 1:
CVREF = (VREF-) + (CVR<4:0>/32)  (VREF+ – VREF-)
When CVRSS = 0:
CVREF = (AVSS) + (CVR<4:0>/32)  (AVDD – AVSS)

PIC18F87K90 FAMILY
FIGURE 25-1: COMPARATOR VOLTAGE REFERENCE BLOCK DIAGRAM
CVRSS = 1
VREF+
AVDD
CVRSS = 0
CVR<4:0>
CVREN
32-to-1 MUX
32 Steps CVREF
R
R
R
CVRSS = 1
VREF-
CVRSS = 0
25.2 Voltage Reference Accuracy/Error 25.4 Effects of a Reset

The full range of voltage reference cannot be realized A device Reset disables the voltage reference by
due to the construction of the module. The transistors clearing bit, CVREN (CVRCON<7>). This Reset also
on the top and bottom of the resistor ladder network disconnects the reference from the RF5 pin by clearing
(Figure 25-1) keep CVREF from approaching the refer- bit, CVROE (CVRCON<6>).
ence source rails. The voltage reference is derived
from the reference source; therefore, the CVREF output 25.5 Connection Considerations
changes with fluctuations in that source. The tested
absolute accuracy of the voltage reference can be The voltage reference module operates independently
found in Section 31.0 “Electrical Characteristics”. of the comparator module. The output of the reference
generator may be connected to the RF5 pin if the
25.3 Operation During Sleep CVROE bit is set. Enabling the voltage reference out-
put onto RF5, when it is configured as a digital input,
When the device wakes up from Sleep through an will increase current consumption. Connecting RF5 as
interrupt or a Watchdog Timer time-out, the contents of a digital output, with CVRSS enabled, will also increase
the CVRCON register are not affected. To minimize current consumption.
current consumption in Sleep mode, the voltage The RF5 pin can be used as a simple D/A output with
reference should be disabled. limited drive capability. Due to the limited current drive
capability, a buffer must be used on the voltage
reference output for external connections to VREF.
Figure 25-2 shows an example buffering technique.

PIC18F87K90 FAMILY
FIGURE 25-2: COMPARATOR VOLTAGE REFERENCE OUTPUT BUFFER EXAMPLE
PIC18F87K90
CVREF
R(1)
Module
+
Voltage RF5 CVREF Output
–
Reference
Output
Impedance
Note 1: R is dependent upon the Voltage Reference Configuration bits, CVRCON<3:0> and CVRCON<5>.
TABLE 25-1: REGISTERS ASSOCIATED WITH COMPARATOR VOLTAGE REFERENCE

Reset
Page:
CVRCON CVREN CVROE CVRSS CVR4 CVR3 CVR2 CVR1 CVR0 77
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used with the comparator voltage reference.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
26.0 HIGH/LOW-VOLTAGE DETECT The High/Low-Voltage Detect Control register
(Register 26-1) completely controls the operation of the
(HLVD)
HLVD module. This allows the circuitry to be “turned
The PIC18F87K90 family of devices has a High/Low- off” by the user under software control, which
Voltage Detect module (HLVD). This is a programmable minimizes the current consumption for the device.
circuit that sets both a device voltage trip point and the The module’s block diagram is shown in Figure 26-1.
direction of change from that point. If the device experi-
ences an excursion past the trip point in that direction, an
interrupt flag is set. If the interrupt is enabled, the
program execution branches to the interrupt vector
address and the software responds to the interrupt.
REGISTER 26-1: HLVDCON: HIGH/LOW-VOLTAGE DETECT CONTROL REGISTER

R/W-0 R-0 R-0 R/W-0 R/W-0 R/W-1 R/W-0 R/W-0
(1) (1) (1)
VDIRMAG BGVST IRVST HLVDEN HLVDL3 HLVDL2 HLVDL1 HLVDL0(1)
bit 7 bit 0
Legend:
bit 7 VDIRMAG: Voltage Direction Magnitude Select bit

1 = Event occurs when voltage equals or exceeds trip point (HLVDL<3:0>)
0 = Event occurs when voltage equals or falls below trip point (HLVDL<3:0>)
bit 6 BGVST: Band Gap Reference Voltages Stable Status Flag bit
1 = Internal band gap voltage references are stable
0 = Internal band gap voltage references are not stable
bit 5 IRVST: Internal Reference Voltage Stable Flag bit
1 = Indicates that the voltage detect logic will generate the interrupt flag at the specified voltage range
0 = Indicates that the voltage detect logic will not generate the interrupt flag at the specified voltage
range and the HLVD interrupt should not be enabled
bit 4 HLVDEN: High/Low-Voltage Detect Power Enable bit
1 = HLVD is enabled
0 = HLVD is disabled
bit 3-0 HLVDL<3:0>: Voltage Detection Limit bits(1)
1111 = External analog input is used (input comes from the HLVDIN pin)
1110 = Maximum setting
.
.
.
0000 = Minimum setting
Note 1: For the electrical specifications, see Parameter D420.

PIC18F87K90 FAMILY
The module is enabled by setting the HLVDEN bit trip point voltage. The “trip point” voltage is the voltage
(HLVDCON<4>). Each time the HLVD module is level at which the device detects a high or low-voltage
enabled, the circuitry requires some time to stabilize. event, depending on the configuration of the module.
The IRVST bit (HLVDCON<5>) is a read-only bit used When the supply voltage is equal to the trip point, the
to indicate when the circuit is stable. The module can voltage tapped off of the resistor array is equal to the
only generate an interrupt after the circuit is stable and internal reference voltage generated by the voltage
IRVST is set. reference module. The comparator then generates an
The VDIRMAG bit (HLVDCON<7>) determines the interrupt signal by setting the HLVDIF bit.
overall operation of the module. When VDIRMAG is The trip point voltage is software programmable to any of
cleared, the module monitors for drops in VDD below a 16 values. The trip point is selected by programming the
predetermined set point. When the bit is set, the HLVDL<3:0> bits (HLVDCON<3:0>).
module monitors for rises in VDD above the set point.
The HLVD module has an additional feature that allows
the user to supply the trip voltage to the module from an
26.1 Operation
external source. This mode is enabled when bits,
When the HLVD module is enabled, a comparator uses HLVDL<3:0>, are set to ‘1111’. In this state, the
an internally generated reference voltage as the set comparator input is multiplexed from the external input
point. The set point is compared with the trip point, pin, HLVDIN. This gives users the flexibility of configur-
where each node in the resistor divider represents a ing the High/Low-Voltage Detect interrupt to occur at
any voltage in the valid operating range.
FIGURE 26-1: HLVD MODULE BLOCK DIAGRAM (WITH EXTERNAL INPUT)
Externally Generated
Trip Point
VDD
VDD HLVDL<3:0> HLVDCON

Register
HLVDEN VDIRMAG
HLVDIN
16-to-1 MUX
Set
HLVDIF
HLVDEN
Internal Voltage
BOREN Reference
1.024V Typical

PIC18F87K90 FAMILY
26.2 HLVD Setup Depending on the application, the HLVD module does
not need to operate constantly. To reduce current
To set up the HLVD module: requirements, the HLVD circuitry may only need to be
1. Select the desired HLVD trip point by writing the enabled for short periods where the voltage is checked.
value to the HLVDL<3:0> bits. After such a check, the module could be disabled.
2. Set the VDIRMAG bit to detect high voltage
(VDIRMAG = 1) or low voltage (VDIRMAG = 0). 26.4 HLVD Start-up Time
3. Enable the HLVD module by setting the The internal reference voltage of the HLVD module,
HLVDEN bit. specified in electrical specification Parameter 37
4. Clear the HLVD interrupt flag (PIR2<2>), which (Section 31.0 “Electrical Characteristics”), may be
may have been set from a previous interrupt. used by other internal circuitry, such as the
5. If interrupts are desired, enable the HLVD programmable Brown-out Reset. If the HLVD or other
interrupt by setting the HLVDIE and GIE bits circuits using the voltage reference are disabled to
(PIE2<2> and INTCON<7>, respectively). lower the device’s current consumption, the reference
An interrupt will not be generated until the voltage circuit will require time to become stable before
IRVST bit is set. a low or high-voltage condition can be reliably
detected. This start-up time, TIRVST, is an interval that
Note: Before changing any module settings is independent of device clock speed. It is specified in
(VDIRMAG, HLVDL<3:0>), first disable the electrical specification Parameter 36 (Table 31-10).
module (HLVDEN = 0), make the changes
The HLVD interrupt flag is not enabled until TIRVST has
and re-enable the module. This prevents
expired and a stable reference voltage is reached. For
the generation of false HLVD events.
this reason, brief excursions beyond the set point may
not be detected during this interval (see Figure 26-2 or
26.3 Current Consumption Figure 26-3).
When the module is enabled, the HLVD comparator
and voltage divider are enabled and consume static
current. The total current consumption, when enabled,
is specified in electrical specification Parameter D022B
(Table 31-10).
FIGURE 26-2: LOW-VOLTAGE DETECT OPERATION (VDIRMAG = 0)

CASE 1: HLVDIF may Not be Set
VDD
VHLVD
HLVDIF
Enable HLVD
TIRVST
IRVST
HLVDIF Cleared in Software
Internal Reference is Stable
CASE 2:
VDD
VHLVD
HLVDIF
Enable HLVD
TIRVST
IRVST
Internal Reference is Stable HLVDIF Cleared in Software
HLVDIF Cleared in Software,
HLVDIF Remains Set since HLVD Condition still Exists

PIC18F87K90 FAMILY
FIGURE 26-3: HIGH-VOLTAGE DETECT OPERATION (VDIRMAG = 1)
CASE 1:
HLVDIF May Not be Set
VHLVD
VDD
HLVDIF
Enable HLVD
IRVST TIRVST

CASE 2:
VHLVD
VDD
HLVDIF
Enable HLVD
IRVST TIRVST

HLVDIF Cleared in Software,

HLVDIF remains Set since HLVD Condition still Exists
26.5 Applications FIGURE 26-4: TYPICAL LOW-VOLTAGE

DETECT APPLICATION
In many applications, it is desirable to detect a drop
below, or rise above, a particular voltage threshold. For
example, the HLVD module could be periodically
enabled to detect Universal Serial Bus (USB) attach or
detach. This assumes the device is powered by a lower
voltage source than the USB when detached. An attach VA
would indicate a High-Voltage Detect from, for VB
example, 3.3V to 5V (the voltage on USB) and vice
Voltage
versa for a detach. This feature could save a design a

few extra components and an attach signal (input pin).
For general battery applications, Figure 26-4 shows a
possible voltage curve. Over time, the device voltage
decreases. When the device voltage reaches voltage,
VA, the HLVD logic generates an interrupt at time, TA.
The interrupt could cause the execution of an ISR TA TB
Time
(Interrupt Service Routine), which would allow the
application to perform “housekeeping tasks” and a Legend: VA = HLVD trip point
controlled shutdown before the device voltage exits the VB = Minimum valid device
valid operating range at TB. This would give the appli- operating voltage
cation a time window, represented by the difference
between TA and TB, to safely exit.

PIC18F87K90 FAMILY
26.6 Operation During Sleep 26.7 Effects of a Reset
When enabled, the HLVD circuitry continues to operate A device Reset forces all registers to their Reset state.
during Sleep. If the device voltage crosses the trip This forces the HLVD module to be turned off.
point, the HLVDIF bit will be set and the device will
wake-up from Sleep. Device execution will continue
from the interrupt vector address if interrupts have
been globally enabled.
TABLE 26-1: REGISTERS ASSOCIATED WITH HIGH/LOW-VOLTAGE DETECT MODULE

Reset
Page:
HLVDCON VDIRMAG BGVST IRVST HLVDEN HLVDL3 HLVDL2 HLVDL1 HLVDL0 77
PIR2 OSCFIF — SSP2IF BLC2IF BCL1IF HLVDIF TMR3IF TMR3GIF 77
PIE2 OSCFIE — SSP2IE BLC2IE BCL1IE HLVDIE TMR3IE TMR3GIE 77
IPR2 OSCFIP — SSP2IP BLC2IP BCL1IP HLVDIP TMR3IP TMR3GIP 77
Legend: — = unimplemented, read as ‘0’. Shaded cells are unused by the HLVD module.
Note 1: PORTA<7:6> and their direction bits are individually configured as port pins based on various primary
oscillator modes. When disabled, these bits read as ‘0’.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
27.0 CHARGE TIME • Control of response to edges
MEASUREMENT UNIT (CTMU) • Time measurement resolution of 1 nanosecond
• High-precision time measurement
The Charge Time Measurement Unit (CTMU) is a
• Time delay of external or internal signal
flexible analog module that provides accurate differen-
asynchronous to system clock
tial time measurement between pulse sources, as well
as asynchronous pulse generation. By working with • Accurate current source suitable for capacitive
other on-chip analog modules, the CTMU can precisely measurement
measure time, capacitance and relative changes in The CTMU works in conjunction with the A/D Converter
capacitance or generate output pulses with a specific to provide up to 24 channels for time or charge
time delay. The CTMU is ideal for interfacing with measurement, depending on the specific device and
capacitive-based sensors. the number of A/D channels available. When config-
The module includes these key features: ured for time delay, the CTMU is connected to one of
the analog comparators. The level-sensitive input edge
• Up to 24 channels available for capacitive or time sources can be selected from four sources: two
measurement input external inputs or the ECCP1/ECCP2 Special Event
• On-chip precision current source Triggers.
• Four-edge input trigger sources The CTMU special event can trigger the Analog-to-Digital
• Polarity control for each edge source Converter module.
• Control of edge sequence Figure 27-1 provides a block diagram of the CTMU.
FIGURE 27-1: CTMU BLOCK DIAGRAM
CTMUCON
EDGEN CTMUICON
EDGSEQEN
ITRIM<5:0> TGEN
EDG1SELx
EDG1POL IRNG<1:0> IDISSEN
EDG2SELx EDG1STAT CTTRIG
EDG2POL EDG2STAT Current Source
CTED1 Edge
CTMU
Control Control A/D Trigger
CTED2 Logic Current Logic
Control
ECCP2
Pulse CTPLS
ECCP1 Generator
A/D Converter Comparator 2
Input
Comparator 2 Output

PIC18F87K90 FAMILY
27.1 CTMU Registers The CTMUCONH and CTMUCONL registers
(Register 27-1 and Register 27-2) contain control bits
The control registers for the CTMU are: for configuring the CTMU module edge source selec-
• CTMUCONH tion, edge source polarity selection, edge sequencing,
• CTMUCONL A/D trigger, analog circuit capacitor discharge and
enables. The CTMUICON register (Register 27-3) has
• CTMUICON
bits for selecting the current source range and current
source trim.
REGISTER 27-1: CTMUCONH: CTMU CONTROL HIGH REGISTER

CTMUEN — CTMUSIDL TGEN EDGEN EDGSEQEN IDISSEN CTTRIG
bit 7 bit 0
Legend:
bit 7 CTMUEN: CTMU Enable bit

1 = Module is enabled
0 = Module is disabled
bit 5 CTMUSIDL: Stop in Idle Mode bit
1 = Discontinue module operation when device enters Idle mode
0 = Continue module operation in Idle mode
bit 4 TGEN: Time Generation Enable bit
1 = Enables edge delay generation
0 = Disables edge delay generation
bit 3 EDGEN: Edge Enable bit
1 = Edges are not blocked
0 = Edges are blocked
bit 2 EDGSEQEN: Edge Sequence Enable bit
1 = Edge 1 event must occur before Edge 2 event can occur
0 = No edge sequence is needed
bit 1 IDISSEN: Analog Current Source Control bit
1 = Analog current source output is grounded
0 = Analog current source output is not grounded
bit 0 CTTRIG: Trigger Control bit
1 = Trigger output is enabled
0 = Trigger output is disabled

PIC18F87K90 FAMILY
REGISTER 27-2: CTMUCONL: CTMU CONTROL LOW REGISTER
EDG2POL EDG2SEL1 EDG2SEL0 EDG1POL EDG1SEL1 EDG1SEL0 EDG2STAT EDG1STAT
bit 7 bit 0
Legend:
bit 7 EDG2POL: Edge 2 Polarity Select bit

1 = Edge 2 is programmed for a positive edge response
0 = Edge 2 is programmed for a negative edge response
bit 6-5 EDG2SEL<1:0>: Edge 2 Source Select bits
11 = CTED1 pin
10 = CTED2 pin
01 = ECCP1 Special Event Trigger
bit 4 EDG1POL: Edge 1 Polarity Select bit
1 = Edge 1 is programmed for a positive edge response
0 = Edge 1 is programmed for a negative edge response
bit 3-2 EDG1SEL<1:0>: Edge 1 Source Select bits
11 = CTED1 pin
10 = CTED2 pin
bit 1 EDG2STAT: Edge 2 Status bit
1 = Edge 2 event has occurred
0 = Edge 2 event has not occurred
bit 0 EDG1STAT: Edge 1 Status bit
1 = Edge 1 event has occurred
0 = Edge 1 event has not occurred

PIC18F87K90 FAMILY
REGISTER 27-3: CTMUICON: CTMU CURRENT CONTROL REGISTER
ITRIM5 ITRIM4 ITRIM3 ITRIM2 ITRIM1 ITRIM0 IRNG1 IRNG0
bit 7 bit 0
Legend:
bit 7-2 ITRIM<5:0>: Current Source Trim bits

011111 = Maximum positive change from nominal current
011110
.
.
.
000001 = Minimum positive change from nominal current
000000 = Nominal current output specified by IRNG<1:0>
111111 = Minimum negative change from nominal current
.
.
.
100010
100001 = Maximum negative change from nominal current
bit 1-0 IRNG<1:0>: Current Source Range Select bits
11 = 100 x Base Current
10 = 10 x Base Current
01 = Base current level (0.55 A nominal)
00 = Current source disabled

PIC18F87K90 FAMILY
27.2 CTMU Operation Current trim is provided by the ITRIM<5:0> bits
The CTMU works by using a fixed current source to (CTMUICON<7:2>). These six bits allow trimming of
charge a circuit. The type of circuit depends on the type the current source in steps of approximately 2% per
of measurement being made. step. Half of the range adjusts the current source posi-
tively and the other half reduces the current source. A
In the case of charge measurement, the current is fixed value of ‘000000’ is the neutral position (no change). A
and the amount of time the current is applied to the cir- value of ‘100000’ is the maximum negative adjustment
cuit is fixed. The amount of voltage read by the A/D (approximately -62%) and ‘011111’ is the maximum
becomes a measurement of the circuit’s capacitance. positive adjustment (approximately +62%).
In the case of time measurement, the current, as well
as the capacitance of the circuit, is fixed. In this case, 27.2.3 EDGE SELECTION AND CONTROL
the voltage read by the A/D is representative of the CTMU measurements are controlled by edge events
amount of time elapsed from the time the current occurring on the module’s two input channels. Each
source starts and stops charging the circuit. channel, referred to as Edge 1 and Edge 2, can be con-
If the CTMU is being used as a time delay, both capaci- figured to receive input pulses from one of the edge
tance and current source are fixed, as well as the voltage input pins (CTED1 and CTED2) or CCPx Special Event
supplied to the comparator circuit. The delay of a signal Triggers. The input channels are level-sensitive,
is determined by the amount of time it takes the voltage responding to the instantaneous level on the channel
to charge to the comparator threshold voltage. rather than a transition between levels. The inputs are
selected using the EDG1SEL and EDG2SEL bit pairs
27.2.1 THEORY OF OPERATION (CTMUCONL<3:2, 6:5>).
The operation of the CTMU is based on the equation In addition to source, each channel can be configured for
for charge: event polarity using the EDGE2POL and EDGE1POL
dV bits (CTMUCONL<7,4>). The input channels can also
C=I• be filtered for an edge event sequence (Edge 1
dT
occurring before Edge 2) by setting the EDGSEQEN bit
More simply, the amount of charge measured in (CTMUCONH<2>).
coulombs in a circuit is defined as current in amperes
(I) multiplied by the amount of time in seconds that the 27.2.4 EDGE STATUS
current flows (t). Charge is also defined as the capaci- The CTMUCON register also contains two status bits,
tance in farads (C) multiplied by the voltage of the EDG2STAT and EDG1STAT (CTMUCONL<1:0>).
circuit (V). It follows that: Their primary function is to show if an edge response
I•t=C•V has occurred on the corresponding channel. The
CTMU automatically sets a particular bit when an edge
The CTMU module provides a constant, known current response is detected on its channel. The level-sensitive
source. The A/D Converter is used to measure (V) in nature of the input channels also means that the status
the equation, leaving two unknowns: capacitance (C) bits become set immediately if the channel’s configura-
and time (t). The above equation can be used to calcu- tion is changed and matches the channel’s current
late capacitance or time, by either the relationship state.
using the known fixed capacitance of the circuit:
The module uses the edge status bits to control the cur-
t = (C • V)/I rent source output to external analog modules (such as
the A/D Converter). Current is only supplied to external
or by: modules when only one (not both) of the status bits is
C = (I • t)/V set. Current is shut off when both bits are either set or
cleared. This allows the CTMU to measure current only
using a fixed time that the current source is applied to during the interval between edges. After both status
the circuit. bits are set, it is necessary to clear them before another
measurement is taken. Both bits should be cleared
27.2.2 CURRENT SOURCE simultaneously, if possible, to avoid re-enabling the
CTMU current source.
At the heart of the CTMU is a precision current source,
designed to provide a constant reference for measure- In addition to being set by the CTMU hardware, the
ments. The level of current is user-selectable across edge status bits can also be set by software. This per-
three ranges or a total of two orders of magnitude, with mits a user application to manually enable or disable
the ability to trim the output in ±2% increments the current source. Setting either (but not both) of the
(nominal). The current range is selected by the bits enables the current source. Setting or clearing both
IRNG<1:0> bits (CTMUICON<1:0>), with a value of bits at once disables the source.
‘00’ representing the lowest range.

PIC18F87K90 FAMILY
27.2.5 INTERRUPTS Depending on the type of measurement or pulse
generation being performed, one or more additional
The CTMU sets its interrupt flag (PIR3<3>) whenever
modules may also need to be initialized and configured
the current source is enabled, then disabled. An inter-
with the CTMU module:
rupt is generated only if the corresponding interrupt
enable bit (PIE3<3>) is also set. If edge sequencing is • Edge Source Generation: In addition to the
not enabled (i.e., Edge 1 must occur before Edge 2), it external edge input pins, CCPx Special Event
is necessary to monitor the edge status bits and Triggers can be used as edge sources for the
determine which edge occurred last and caused the CTMU.
interrupt. • Capacitance or Time Measurement: The CTMU
module uses the A/D Converter to measure the
27.3 CTMU Module Initialization voltage across a capacitor that is connected to one
of the analog input channels.
The following sequence is a general guideline used to
• Pulse Generation: When generating system clock
initialize the CTMU module:
independent, output pulses, the CTMU module
1. Select the current source range using the uses Comparator 2 and the associated
IRNGx bits (CTMUICON<1:0>). comparator voltage reference.
2. Adjust the current source trim using the ITRIMx
bits (CTMUICON<7:2>). 27.4 Calibrating the CTMU Module
3. Configure the edge input sources for Edge 1 and
The CTMU requires calibration for precise measure-
Edge 2 by setting the EDG1SEL and EDG2SEL
ments of capacitance and time, as well as for accurate
bits (CTMUCONL<3:2> and <6:5>, respectively).
time delay. If the application only requires measurement
4. Configure the input polarities for the edge inputs of a relative change in capacitance or time, calibration is
using the EDG2POL and EDG1POL bits usually not necessary. An example of a lesser precision
(CTMUCONL<7,4>). application is a capacitive touch switch, in which the
The default configuration is for negative edge touch circuit has a baseline capacitance and the added
polarity (high-to-low transitions). capacitance of the human body changes the overall
5. Enable edge sequencing using the EDGSEQEN capacitance of a circuit.
bit (CTMUCONH<2>). If actual capacitance or time measurement is required,
By default, edge sequencing is disabled. two hardware calibrations must take place:
6. Select the operating mode (Measurement or • The current source needs calibration to set it to a
Time Delay) with the TGEN bit. precise current.
• The circuit being measured needs calibration to
The default mode is the Time/Capacitance
measure or nullify any capacitance other than that
Measurement.
to be measured.
7. Configure the module to automatically trigger
an A/D conversion when the second edge 27.4.1 CURRENT SOURCE CALIBRATION
event has occurred using the CTTRIG bit The current source on board the CTMU module has a
(CTMUCONH<0>). range of ±60% nominal for each of three current
The conversion trigger is disabled by default. ranges. For precise measurements, it is possible to
8. Discharge the connected circuit by setting the measure and adjust this current source by placing a
IDISSEN bit (CTMUCONH<1>). high-precision resistor, RCAL, onto an unused analog
channel. An example circuit is shown in Figure 27-2.
9. After waiting a sufficient time for the circuit to
discharge, clear IDISSEN. To measure the current source:
10. Disable the module by clearing the CTMUEN bit 1. Initialize the A/D Converter.
(CTMUCONH<7>). 2. Initialize the CTMU.
11. Clear the Edge Status bits, EDG2STAT and 3. Enable the current source by setting EDG1STAT
EDG1STAT (CTMUCONL<1:0>). (CTMUCONL<0>).
12. Enable both edge inputs by setting the EDGEN 4. Issue the settling time delay.
bit (CTMUCONH<3>). 5. Perform the A/D conversion.
13. Enable the module by setting the CTMUEN bit. 6. Calculate the current source current using
I = V/RCAL, where RCAL is a high-precision
resistance and V is measured by performing an
A/D conversion.

PIC18F87K90 FAMILY
The CTMU current source may be trimmed with the A value of 70% of full-scale voltage is chosen to make
trim bits in CTMUICON using an iterative process to get sure that the A/D Converter was in a range that is well
the exact current desired. Alternatively, the nominal above the noise floor. If an exact current is chosen to
value without adjustment may be used. That value may incorporate the trimming bits from CTMUICON, the
be stored by software for use in all subsequent resistor value of RCAL may need to be adjusted accord-
capacitive or time measurements. ingly. RCAL may also be adjusted to allow for available
To calculate the value for RCAL, the nominal current resistor values. RCAL should be of the highest precision
must be chosen. Then, the resistance can be available, in light of the precision needed for the circuit
calculated. that the CTMU will be measuring. A recommended
minimum would be 0.1% tolerance.
For example, if the A/D Converter reference voltage is
3.3V, use 70% of full scale (or 2.31V) as the desired The following examples show a typical method for
approximate voltage to be read by the A/D Converter. If performing a CTMU current calibration.
the range of the CTMU current source is selected to be • Example 27-1 demonstrates how to initialize the
0.55 A, the resistor value needed is calculated as, A/D Converter and the CTMU.
RCAL = 2.31V/0.55 A, for a value of 4.2 MΩ. Similarly, This routine is typical for applications using both
if the current source is chosen to be 5.5 A, RCAL would modules.
be 420,000Ω, and 42,000Ω if the current source is set
to 55 A. • Example 27-2 demonstrates one method for the
actual calibration routine.
FIGURE 27-2: CTMU CURRENT SOURCE This method manually triggers the A/D Converter to
CALIBRATION CIRCUIT demonstrate the entire step-wise process. It is also
possible to automatically trigger the conversion by
PIC18F87K90 setting the CTMU’s CTTRIG bit (CTMUCONH<0>).
CTMU
Current Source
A/D
Trigger
A/D Converter
ANx
A/D
RCAL MUX

PIC18F87K90 FAMILY
EXAMPLE 27-1: SETUP FOR CTMU CALIBRATION ROUTINES
#include "p18cxxx.h"
/**************************************************************************/
/*Setup CTMU *****************************************************************/
/**************************************************************************/
void setup(void)
{ //CTMUCON - CTMU Control register
CTMUCONH = 0x00; //make sure CTMU is disabled

CTMUCONL = 0X90;
//CTMU continues to run when emulator is stopped,CTMU continues
//to run in idle mode,Time Generation mode disabled, Edges are blocked
//No edge sequence order, Analog current source not grounded, trigger
//output disabled, Edge2 polarity = positive level, Edge2 source =
//source 0, Edge1 polarity = positive level, Edge1 source = source 0,
// Set Edge status bits to zero
//CTMUICON - CTMU Current Control Register

CTMUICON = 0x01; //0.55uA, Nominal - No Adjustment
/**************************************************************************/
//Setup AD converter;
/**************************************************************************/
TRISA=0x04; //set channel 2 as an input
// Configured AN2 as an analog channel

// ANCON0
ANCON0 = 0X04;
// ANCON1
ANCON1 = 0XE0;
// ADCON1
ADCON2bits.ADFM=1; //Resulst format 1= Right justified
ADCON2bits.ACQT=1; //Acquition time 7 = 20TAD 2 = 4TAD 1=2TAD
ADCON2bits.ADCS=2; //Clock conversion bits 6= FOSC/64 2=FOSC/32
// ADCON0
ADCON1bits.VCFG0 =0; //Vref+ = AVdd
ADCON1bits.VCFG1 =0; //Vref+ = AVdd
ADCON1bits.VNCFG =0; //Vref- = AVss
ADCON0bits.CHS=2; //Select ADC channel
ADCON0bits.ADON=1; //Turn on ADC

PIC18F87K90 FAMILY
EXAMPLE 27-2: CURRENT CALIBRATION ROUTINE
#define COUNT 500 //@ 8MHz = 125uS.

#define DELAY for(i=0;i<COUNT;i++)
#define RCAL .027 //R value is 4200000 (4.2M)
//scaled so that result is in
//1/100th of uA
#define ADSCALE 1023 //for unsigned conversion 10 sig bits
#define ADREF 3.3 //Vdd connected to A/D Vr+
int main(void)
{
int i;
int j = 0; //index for loop
unsigned int Vread = 0;
double VTot = 0;
float Vavg=0, Vcal=0, CTMUISrc = 0; //float values stored for calcs
//assume CTMU and A/D have been setup correctly

//see Example 25-1 for CTMU & A/D setup
setup();
CTMUCONHbits.CTMUEN = 1; //Enable the CTMU

for(j=0;j<10;j++)
{
CTMUCONHbits.IDISSEN = 1; //drain charge on the circuit
DELAY; //wait 125us
CTMUCONHbits.IDISSEN = 0; //end drain of circuit
CTMUCONLbits.EDG1STAT = 1; //Begin charging the circuit

//using CTMU current source
DELAY; //wait for 125us
CTMUCONLbits.EDG1STAT = 0; //Stop charging circuit
PIR1bits.ADIF = 0; //make sure A/D Int not set

ADCON0bits.GO=1; //and begin A/D conv.
while(!PIR1bits.ADIF); //Wait for A/D convert complete
Vread = ADRES; //Get the value from the A/D

PIR1bits.ADIF = 0; //Clear A/D Interrupt Flag
VTot += Vread; //Add the reading to the total
}
Vavg = (float)(VTot/10.000); //Average of 10 readings

Vcal = (float)(Vavg/ADSCALE*ADREF);
CTMUISrc = Vcal/RCAL; //CTMUISrc is in 1/100ths of uA

PIC18F87K90 FAMILY
27.4.2 CAPACITANCE CALIBRATION This measured value is then stored and used for
calculations of time measurement or subtracted for
There is a small amount of capacitance from the
capacitance measurement. For calibration, it is
internal A/D Converter sample capacitor, as well as
expected that the capacitance of CSTRAY + CAD is
stray capacitance from the circuit board traces and
approximately known; CAD is approximately 4 pF.
pads that affect the precision of capacitance measure-
ments. A measurement of the stray capacitance can be An iterative process may be required to adjust the time,
taken by making sure the desired capacitance to be t, that the circuit is charged to obtain a reasonable volt-
measured has been removed. age reading from the A/D Converter. The value of t may
be determined by setting COFFSET to a theoretical value
After removing the capacitance to be measured:
and solving for t. For example, if CSTRAY is theoretically
1. Initialize the A/D Converter and the CTMU. calculated to be 11 pF, and V is expected to be 70% of
2. Set EDG1STAT (= 1). VDD or 2.31V, t would be:
3. Wait for a fixed delay of time, t.
(4 pF + 11 pF) • 2.31V/0.55 A
4. Clear EDG1STAT.
5. Perform an A/D conversion.
or 63 s.
6. Calculate the stray and A/D sample capacitances:
See Example 27-3 for a typical routine for CTMU
COFFSET = CSTRAY + CAD = (I • t)/V capacitance calibration.
Where:
• I is known from the current source measurement
step
• t is a fixed delay
• V is measured by performing an A/D conversion

PIC18F87K90 FAMILY
EXAMPLE 27-3: CAPACITANCE CALIBRATION ROUTINE
#define COUNT 25 //@ 8MHz INTFRC = 62.5 us.

#define ETIME COUNT*2.5 //time in uS
#define ADSCALE 1023 //for unsigned conversion 10 sig bits
#define ADREF 3.3 //Vdd connected to A/D Vr+
#define RCAL .027 //R value is 4200000 (4.2M)
//scaled so that result is in
//1/100th of uA
int main(void)
{
int i;
int j = 0; //index for loop
unsigned int Vread = 0;
float CTMUISrc, CTMUCap, Vavg, VTot, Vcal;

setup();

for(j=0;j<10;j++)
{
DELAY; //wait 125us



PIR1bits.ADIF = 0; //Clear A/D Interrupt Flag
VTot += Vread; //Add the reading to the total
}
Vavg = (float)(VTot/10.000); //Average of 10 readings

Vcal = (float)(Vavg/ADSCALE*ADREF);
CTMUISrc = Vcal/RCAL; //CTMUISrc is in 1/100ths of uA
CTMUCap = (CTMUISrc*ETIME/Vcal)/100;
}

PIC18F87K90 FAMILY
27.5 Measuring Capacitance with the 27.5.2 RELATIVE CHARGE
CTMU MEASUREMENT
There are two ways to measure capacitance with the Not all applications require precise capacitance
CTMU. The absolute method measures the actual measurements. When detecting a valid press of a
capacitance value. The relative method only measures capacitance-based switch, only a relative change of
for any change in the capacitance. capacitance needs to be detected.
In such an application, when the switch is open (or not
27.5.1 ABSOLUTE CAPACITANCE touched), the total capacitance is the capacitance of the
MEASUREMENT combination of the board traces, the A/D Converter and
For absolute capacitance measurements, both the other elements. A larger voltage will be measured by the
current and capacitance calibration steps, found in A/D Converter. When the switch is closed (or touched),
Section 27.4 “Calibrating the CTMU Module”, the total capacitance is larger due to the addition of the
should be followed. capacitance of the human body to the above listed
capacitances and a smaller voltage will be measured by
To perform these measurements: the A/D Converter.
1. Initialize the A/D Converter. To detect capacitance changes simply:
2. Initialize the CTMU.
1. Initialize the A/D Converter and the CTMU.
3. Set EDG1STAT.
2. Set EDG1STAT.
4. Wait for a fixed delay, T.
3. Wait for a fixed delay.
5. Clear EDG1STAT.
4. Clear EDG1STAT.
7. Calculate the total capacitance, CTOTAL = (I * T)/V,
where: The voltage measured by performing the A/D conver-
sion is an indication of the relative capacitance. In this
• I is known from the current source
case, no calibration of the current source or circuit
measurement step (Section 27.4.1 “Current
capacitance measurement is needed. (For a sample
Source Calibration”)
software routine for a capacitive touch switch, see
• T is a fixed delay Example 27-4.)
• V is measured by performing an A/D
conversion
8. Subtract the stray and A/D capacitance
(COFFSET from Section 27.4.2 “Capacitance
Calibration”) from CTOTAL to determine the
measured capacitance.

PIC18F87K90 FAMILY
EXAMPLE 27-4: ROUTINE FOR CAPACITIVE TOUCH SWITCH
#define COUNT 500 //@ 8MHz = 125uS.

#define OPENSW 1000 //Un-pressed switch value
#define TRIP 300 //Difference between pressed
//and un-pressed switch
#define HYST 65 //amount to change
//from pressed to un-pressed
#define PRESSED 1
#define UNPRESSED 0
int main(void)
{
unsigned int Vread; //storage for reading
unsigned int switchState;
int i;

setup();

DELAY; //wait 125us


if(Vread < OPENSW - TRIP)

{
switchState = PRESSED;
}
else if(Vread > OPENSW - TRIP + HYST)
{
switchState = UNPRESSED;
}
}

PIC18F87K90 FAMILY
27.6 Measuring Time with the CTMU It is assumed that the time measured is small enough
Module that the capacitance, COFFSET, provides a valid voltage
to the A/D Converter. For the smallest time measure-
Time can be precisely measured after the ratio (C/I) is ment, always set the A/D Channel Select register
measured from the current and capacitance calibration (AD1CHS) to an unused A/D channel, the correspond-
step. To do that: ing pin for which is not connected to any circuit board
1. Initialize the A/D Converter and the CTMU. trace. This minimizes added stray capacitance,
keeping the total circuit capacitance close to that of the
2. Set EDG1STAT.
A/D Converter itself (25 pF).
3. Set EDG2STAT.
To measure longer time intervals, an external capacitor
may be connected to an A/D channel and that channel
5. Calculate the time between edges as T = (C/I) • V, selected whenever making a time measurement.
where:
• I is calculated in the current calibration step
(Section 27.4.1 “Current Source Calibration”)
• C is calculated in the capacitance calibration step
(Section 27.4.2 “Capacitance Calibration”)
• V is measured by performing the A/D conversion
FIGURE 27-3: TYPICAL CONNECTIONS AND INTERNAL CONFIGURATION FOR TIME

MEASUREMENT
PIC18F87K90
CTMU
CTED1 EDG1
Current Source
CTED2 EDG2
Output Pulse
A/D Converter
ANX
CAD
RPR

PIC18F87K90 FAMILY
27.7 Creating a Delay with the CTMU An example use of the external capacitor feature is
Module interfacing with variable capacitive-based sensors,
such as a humidity sensor. As the humidity varies, the
A unique feature on board the CTMU module is its ability pulse-width output on CTPLS will vary. An example use
to generate system clock independent output pulses, of the CTDIN feature is interfacing with a digital sensor.
based on either an internal voltage or an external capac- The CTPLS output pin can be connected to an input
itor value. When using an external voltage, this is capture pin and the varying pulse width measured to
accomplished using the CTDIN input pin as a trigger for determine the humidity in the application.
the pulse delay. When using an internal capacitor
To use this feature:
value, this is accomplished using the internal compara-
tor voltage reference module and Comparator 2 input 1. If CTMUDS is cleared, initialize Comparator 2.
pin. The pulse is output onto the CTPLS pin. To enable 2. If CTMUDS is cleared, initialize the comparator
this mode, set the TGEN bit. voltage reference.
See Figure 27-4 for an example circuit. When 3. Initialize the CTMU and enable time delay
CTMUDS (ODCON3<0>) is cleared, the pulse delay is generation by setting the TGEN bit.
determined by the output of Comparator 2, and when it 4. Set EDG1STAT.
is set, the pulse delay is determined by the input of When CTMUDS is cleared, as soon as CDELAY charges
CTDIN. CDELAY is chosen by the user to determine the to the value of the voltage reference trip point, an output
output pulse width on CTPLS. The pulse width is calcu- pulse is generated on CTPLS. When CTMUDS is set, as
lated by T = (CDELAY/I) * V, where I is known from the soon as CTDIN is set, an output pulse is generated on
current source measurement step (Section 27.4.1 CTPLS.
“Current Source Calibration”) and V is the internal
reference voltage (CVREF).
FIGURE 27-4: TYPICAL CONNECTIONS AND INTERNAL CONFIGURATION FOR PULSE

DELAY GENERATION
PIC18F87K90
CTMU
CTED1 EDG1 CTPLS
Current Source
Comparator CTMUDS
CTMUI CTDIN
C2
CDELAY
CVREF
C1
External Reference
External Comparator

PIC18F87K90 FAMILY
27.8 Measuring Temperature Using the source the current to the diode. The ADC reading will
CTMU Module reflect the temperature. With the increase, the ADC
readings will go low. This can be used for low-cost
The CTMU, along with an internal diode, can be used temperature measurement applications.
to measure the temperature. The ADC can be con-
nected to the internal diode and the CTMU module can
EXAMPLE 27-5: ROUTINE FOR TEMPERATURE MEASUREMENT USING INTERNAL DIODE

//Initialize CTMU
CTMUICON=0x03;
CTMUCONHbits.CTMUEN=1;
CTMUCONLbits.EDG1STAT=1;
//Initialize ADC
ADCON0=0xE5; //ADCON and connect to Internal diode
ADCON1=0;
ADCON2=0xBE; //Right justified
ADCON0bits.GO=1;
while(ADCON0bits.GO==1);
Temp=ADRES; ;//read ADC results ( inversely proportional to temperature)
----------------------------------------------------------------------------------------------
Note: The temperature diode is not calibrated; the user will have to calibrate the diode to their application.

PIC18F87K90 FAMILY
27.9 Operation During Sleep/Idle Modes module is performing an operation when Idle mode is
invoked, in this case, the results will be similar to those
27.9.1 SLEEP MODE with Sleep mode.
When the device enters any Sleep mode, the CTMU
module current source is always disabled. If the CTMU 27.10 Effects of a Reset on CTMU
is performing an operation that depends on the current Upon Reset, all registers of the CTMU are cleared. This
source when Sleep mode is invoked, the operation may disables the CTMU module, turns off its current source
not terminate correctly. Capacitance and time and returns all configuration options to their default set-
measurements may return erroneous values. tings. The module needs to be re-initialized following
any Reset.
27.9.2 IDLE MODE
If the CTMU is in the process of taking a measurement at
The behavior of the CTMU in Idle mode is determined
the time of Reset, the measurement will be lost. A partial
by the CTMUSIDL bit (CTMUCONH<5>). If CTMUSIDL
charge may exist on the circuit that was being measured,
is cleared, the module will continue to operate in Idle
which should be properly discharged before the CTMU
mode. If CTMUSIDL is set, the module’s current source
makes subsequent attempts to make a measurement.
is disabled when the device enters Idle mode. If the
The circuit is discharged by setting and clearing the
IDISSEN bit (CTMUCONH<1>) while the A/D Converter
is connected to the appropriate channel.
TABLE 27-1: REGISTERS ASSOCIATED WITH CTMU MODULE

Reset
on Page:
CTMUCONH CTMUEN — CTMUSIDL TGEN EDGEN EDGSEQEN IDISSEN CTTRIG 80

CTMUCONL EDG2POL EDG2SEL1 EDG2SEL0 EDG1POL EDG1SEL1 EDG1SEL0 EDG2STAT EDG1STAT 80
CTMUICON ITRIM5 ITRIM4 ITRIM3 ITRIM2 ITRIM1 ITRIM0 IRNG1 IRNG0 80
Legend: — = unimplemented, read as ‘0’

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
28.0 SPECIAL FEATURES OF THE 28.1 Configuration Bits
CPU The Configuration bits can be programmed (read as
The PIC18F87K90 family of devices includes several ‘0’) or left unprogrammed (read as ‘1’) to select various
features intended to maximize reliability and minimize device configurations. These bits are mapped starting
cost through elimination of external components. at program memory location, 300000h.
These include: The user will note that address, 300000h, is beyond the
• Oscillator Selection user program memory space. In fact, it belongs to the
configuration memory space (300000h-3FFFFFh),
• Resets:
which can only be accessed using table reads and
- Power-on Reset (POR)
table writes.
- Power-up Timer (PWRT)
- Oscillator Start-up Timer (OST) Software programming the Configuration registers is
- Brown-out Reset (BOR) done in a manner similar to programming the Flash
• Interrupts memory. The WR bit in the EECON1 register starts a
• Watchdog Timer (WDT) and On-Chip Regulator self-timed write to the Configuration register. In normal
operation mode, a TBLWT instruction, with the TBLPTR
• Fail-Safe Clock Monitor
pointing to the Configuration register, sets up the
• Two-Speed Start-up
address and the data for the Configuration register write.
• Code Protection
Setting the WR bit starts a long write to the Configuration
• ID Locations register. The Configuration registers are written a byte at
• In-Circuit Serial Programming™ (ICSP™) a time. To write or erase a configuration cell, a TBLWT
The oscillator can be configured for the application instruction can write a ‘1’ or a ‘0’ into the cell. For
depending on frequency, power, accuracy and cost. All additional details on Flash programming, refer to
of the options are discussed in detail in Section 3.0 Section 7.5 “Writing to Flash Program Memory”.
“Oscillator Configurations”.
A complete discussion of device Resets and interrupts
is available in previous sections of this data sheet.
In addition to their Power-up and Oscillator Start-up
Timers provided for Resets, the PIC18F87K90 family of
devices has a Watchdog Timer, which is either perma-
nently enabled via the Configuration bits or software
controlled (if configured as disabled).
The inclusion of an internal RC (LF-INTOSC) oscillator
also provides the additional benefits of a Fail-Safe
Clock Monitor (FSCM) and Two-Speed Start-up. FSCM
provides for background monitoring of the peripheral
clock and automatic switchover in the event of its fail-
ure. Two-Speed Start-up enables code to be executed
almost immediately on start-up, while the primary clock
source completes its start-up delays.
All of these features are enabled and configured by
setting the appropriate Configuration register bits.

PIC18F87K90 FAMILY
TABLE 28-1: CONFIGURATION BITS AND DEVICE IDs
Default/
File Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Unprogrammed
Value
300000h CONFIG1L — XINST — SOSCSEL1 SOSCSEL0 INTOSCSEL — RETEN -1-1 1--1

300001h CONFIG1H IESO FCMEN — PLLCFG FOSC3 FOSC2 FOSC1 FOSC0 0000 1000
300002h CONFIG2L — BORPWR1 BORWPR0 BORV1 BORV0 BOREN1 BOREN0 PWRTEN -111 1111
300003h CONFIG2H — WDTPS4 WDTPS3 WDTPS2 WDTPS1 WDTPS0 WDTEN1 WDTEN0 -111 1111
300004h CONFIG3L — — — — — — — RTCOSC ---- ---1
300005h CONFIG3H MCLRE — — — MSSPMSK — ECCPMX(2) CCP2MX 1--- 1-11
300006h CONFIG4L DEBUG — — BBSIZ0 — — — STVREN 1--1 ---1
300008h CONFIG5L CP7(1) CP6(1) CP5(1) CP4(1) CP3 CP2 CP1 CP0 1111 1111
300009h CONFIG5H CPD CPB — — — — — — 11-- ----
30000Ah CONFIG6L WRT7(1) WRT6(1) WRT5(1) WRT4(1) WRT3 WRT2 WRT1 WRT0 1111 1111
30000Bh CONFIG6H WRTD WRTB WRTC — — — — — 111- ----
(1) (1) (1) (1)
30000Ch CONFIG7L EBTR7 EBTR6 EBTR5 EBTR4 EBTR3 EBTR2 EBTR1 EBTR0 1111 1111
30000Dh CONFIG7H — EBTRB — — — — — — -1-- ----
3FFFFEh DEVID1(3) DEV2 DEV1 DEV0 REV4 REV3 REV2 REV1 REV0 xxxx xxxx
3FFFFFh DEVID2(3) DEV10 DEV9 DEV8 DEV7 DEV6 DEV5 DEV4 DEV3 xxxx xxxx
Legend: x = unknown, u = unchanged, - = unimplemented, q = value depends on condition.
Shaded cells are unimplemented, read as ‘0’.
Note 1: Implemented in the PIC18F67K90 and PIC18F87K90 devices.
2: Implemented in the 80-pin devices (PIC18F8XK90).
3: See Register 28-14 for DEVID1 values. DEVID registers are read-only and cannot be programmed by the user.

PIC18F87K90 FAMILY
REGISTER 28-1: CONFIG1L: CONFIGURATION REGISTER 1 LOW (BYTE ADDRESS 300000h)
U-0 R/P-1 U-0 R/P-1 R/P-1 R/P-1 U-0 R/P-1
— XINST — SOSCSEL1 SOSCSEL0 INTOSCSEL0 — RETEN
bit 7 bit 0
Legend: P = Programmable bit


bit 6 XINST: Extended Instruction Set Enable bit
1 = Instruction set extension and Indexed Addressing mode are enabled
0 = Instruction set extension and Indexed Addressing mode are disabled (Legacy mode)
bit 4-3 SOSCSEL<1:0>: SOSC Power Selection and Mode Configuration bits
11 = High-power SOSC circuit is selected
10 = Digital (SCLKI) mode: I/O port functionality of RC0 and RC1 is enabled
01 = Low-power SOSC circuit is selected
00 = Reserved
bit 2 INTOSCSEL: LF-INTOSC Low-Power Enable bit
1 = LF-INTOSC is in High-Power mode during Sleep
0 = LF-INTOSC is in Low-Power mode during Sleep
bit 0 RETEN: VREG Sleep Enable bit
1 = Ultra low-power regulator is disabled. Regulator power in Sleep mode is controlled by VREGSLP
(WDTCON<7>)
0 = Ultra low-power regulator is enabled. Regulator power in Sleep mode is controlled by SRETEN
(WDTCON<4>).

PIC18F87K90 FAMILY
REGISTER 28-2: CONFIG1H: CONFIGURATION REGISTER 1 HIGH (BYTE ADDRESS 300001h)
R/P-0 R/P-0 U-0 U-0 R/P-1 R/P-0 R/P-0 R/P-0
IESO FCMEN — PLLCFG(1) FOSC3(2) FOSC2(2) FOSC1(2) FOSC0(2)
bit 7 bit 0

bit 7 IESO: Internal/External Oscillator Switchover bit

1 = Two-Speed Start-up is enabled
0 = Two-Speed Start-up is disabled
bit 6 FCMEN: Fail-Safe Clock Monitor Enable bit
1 = Fail-Safe Clock Monitor is enabled
0 = Fail-Safe Clock Monitor is disabled
bit 4 PLLCFG: 4x PLL Enable bit(1)
1 = Oscillator is multiplied by 4
0 = Oscillator is used directly
bit 3-0 FOSC<3:0>: Oscillator Selection bits(2)
1101 = EC1, EC oscillator (low power, DC-160 kHz)
1100 = EC1IO, EC oscillator with CLKOUT function on RA6 (low power, DC-160 kHz)
1011 = EC2, EC oscillator (medium power, 160 kHz-16 MHz)
1010 = EC2IO, EC oscillator with CLKOUT function on RA6 (medium power,160 kHz-16MHz)
0101 = EC3, EC oscillator (high power, 4 MHz-64 MHz)
0100 = EC3IO, EC oscillator with CLKOUT function on RA6 (high power, 4 MHz-64 MHz)
0011 = HS1, HS oscillator (medium power, 4 MHz-16 MHz)
0010 = HS2, HS oscillator (high power, 16 MHz-25 MHz)
0001 = XT oscillator
0000 = LP oscillator
0111 = RC, External RC oscillator
0110 = RCIO, External RC oscillator with CKLOUT function on RA6
1000 = INTIO2, Internal RC oscillator
1001 = INTIO1, Internal RC oscillator with CLKOUT function on RA6
Note 1: Not valid for the INTIOx PLL mode.

2: INTIO+PLL can only be enabled by the PLLEN bit (OSCTUNE<6>). Other PLL modes can be enabled by
either the PLLEN bit or the PLLCFG (CONFIG1H<4>) bit.

PIC18F87K90 FAMILY
U-0 R/P-1 R/P-1 R/P-1 R/P-1 R/P-1 R/P-1 R/P-1
(1) (1) (1) (1) (2) (2)
— BORPWR1 BORPWR0 BORV1 BORV0 BOREN1 BOREN0 PWRTEN(2)
bit 7 bit 0


bit 6-5 BORPWR<1:0>: BORMV Power Level bits(1)
11 = ZPBORVMV instead of BORMV is selected
10 = BORMV is set to high-power level
01 = BORMV is set to medium-power level
00 = BORMV is set to low-power level
bit 4-3 BORV<1:0>: Brown-out Reset Voltage bits(1)
11 = VBORMV is set to 1.8V
bit 2-1 BOREN<1:0>: Brown-out Reset Enable bits(2)
11 = Brown-out Reset is enabled in hardware only (SBOREN is disabled)
10 = Brown-out Reset is enabled in hardware only and disabled in Sleep mode (SBOREN is disabled)
01 = Brown-out Reset is enabled and controlled by software (SBOREN is enabled)
00 = Brown-out Reset is disabled in hardware and software
bit 0 PWRTEN: Power-up Timer Enable bit(2)
1 = PWRT is disabled
0 = PWRT is enabled
Note 1: For the specifications, see Section 31.1 “DC Characteristics: Supply Voltage PIC18F87K90 Family
(Industrial/Extended)”.
2: The Power-up Timer is decoupled from Brown-out Reset, allowing these features to be independently
controlled.

PIC18F87K90 FAMILY
U-0 R/P-1 R/P-1 R/P-1 R/P-1 R/P-1 R/P-1 R/P-1
— WDTPS4 WDTPS3 WDTPS2 WDTPS1 WDTPS0 WDTEN1 WDTEN0
bit 7 bit 0


bit 6-2 WDTPS<4:0>: Watchdog Timer Postscale Select bits
11111 = 1:1,048,576
10011 = 1:524,288
10010 = 1:262,144
10001 = 1:131,072
10000 = 1:65,536
01111 = 1:32,768
01110 = 1:16,384
01101 = 1:8,192
01100 = 1:4,096
01011 = 1:2,048
01010 = 1:1,024
01001 = 1:512
01000 = 1:256
00111 = 1:128
00110 = 1:64
00101 = 1:32
00100 = 1:16
00011 = 1:8
00010 = 1:4
00001 = 1:2
00000 = 1:1
bit 1-0 WDTEN<1:0>: Watchdog Timer Enable bits
11 = WDT is enabled in hardware; SWDTEN bit is disabled
10 = WDT is controlled by the SWDTEN bit setting
01 = WDT is enabled only while device is active and is disabled in Sleep mode; SWDTEN bit is disabled
00 = WDT is disabled in hardware; SWDTEN bit is disabled

PIC18F87K90 FAMILY
U-0 U-0 U-0 U-0 U-0 U-0 U-0 R/P-1
— — — — — — — RTCOSC
bit 7 bit 0


bit 0 RTCOSC: RTCC Reference Clock Select bit
1 = RTCC uses SOSC as a reference clock
0 = RTCC uses LF-INTOSC as a reference clock

R/P-1 U-0 U-0 U-0 R/P-1 U-0 R/P-1 R/P-1
MCLRE — — — MSSPMSK — ECCPMX(1) CCP2MX
bit 7 bit 0

bit 7 MCLRE: MCLR Pin Enable bit

1 = MCLR pin is enabled; RG5 input pin is disabled
0 = RG5 input pin is enabled; MCLR is disabled
bit 3 MSSPMSK: MSSP V3 7-Bit Address Masking Mode Enable bit
1 = 7-Bit Address Masking mode is enabled
0 = 5-Bit Address Masking mode is enabled
bit 1 ECCPMX: ECCP MUX bit(1)
1 = Enhanced ECCP1 (P1B/P1C) is multiplexed onto RE6 and RE5, CCP6 onto RE6 and CCP7 onto RE5
Enhanced ECCP3 (P3B/P3C) is multiplexed onto RE4 and RE3, CCP8 onto RE4 and CCP9 onto RE3
0 = Enhanced ECCP1 (P1B/P1C) is multiplexed onto RH7 and RH6, CCP6 onto RH7 and CCP7 onto RH6
Enhanced ECCP3 (P3B/P3C) is multiplexed onto RH5 and RH4, CCP8 onto RH5 and CCP9 onto RH4
bit 0 CCP2MX: ECCP2 MUX bit
1 = ECCP2 is multiplexed with RC1
0 = ECCP2 input/output is multiplexed with RE7(1)
Note 1: This feature is only available on 80-pin devices.

PIC18F87K90 FAMILY
R/P-1 U-0 U-0 R/P-0 U-0 R/P-0 U-0 R/P-1
DEBUG — — BBSIZ0 — — — STVREN
bit 7 bit 0

bit 7 DEBUG: Background Debugger Enable bit

1 = Background debugger is disabled, RB6 and RB7 are configured as general purpose I/O pins
0 = Background debugger is enabled, RB6 and RB7 are dedicated to In-Circuit Debug
bit 4 BBSIZ<0>: Boot Block Size Select bit
1 = 2 kW boot block size
0 = 1 kW boot block size
bit 0 STVREN: Stack Full/Underflow Reset Enable bit
1 = Stack full/underflow will cause a Reset
0 = Stack full/underflow will not cause a Reset

PIC18F87K90 FAMILY
REGISTER 28-8: CONFIG5L: CONFIGURATION REGISTER 5 LOW (BYTE ADDRESS 300008h)(2)
R/C-1 R/C-1 R/C-1 R/C-1 R/C-1 R/C-1 R/C-1 R/C-1
CP7(1) CP6(1) CP5(1) CP4(1) CP3 CP2 CP1 CP0
bit 7 bit 0

bit 7 CP7: Code Protection bit(1)

1 = Block 7 is not code-protected
0 = Block 7 is code-protected
bit 3 CP3: Code Protection bit
Note 1: This bit is only available on PIC18F67K90 and PIC18F87K90.

2: For the memory size of the blocks, refer to Figure 28-6.

PIC18F87K90 FAMILY
REGISTER 28-9: CONFIG5H: CONFIGURATION REGISTER 5 HIGH (BYTE ADDRESS 300009h)(1)
R/C-1 R/C-1 U-0 U-0 U-0 U-0 U-0 U-0
CPD CPB — — — — — —
bit 7 bit 0

bit 7 CPD: Data EEPROM Code Protection bit

1 = Data EEPROM is not code-protected
0 = Data EEPROM is code-protected
bit 6 CPB: Boot Block Code Protection bit
1 = Boot block is not code-protected
0 = Boot block is code-protected
Note 1: For the memory size of the blocks, refer to Figure 28-6.

PIC18F87K90 FAMILY
REGISTER 28-10: CONFIG6L: CONFIGURATION REGISTER 6 LOW (BYTE ADDRESS 30000Ah)(2)
WRT7(1) WRT6(1) WRT5(1) WRT4(1) WRT3 WRT2 WRT1 WRT0
bit 7 bit 0

bit 7 WRT7: Write Protection bit(1)

1 = Block 7 is not write-protected
0 = Block 7 is write-protected
bit 3 WRT3: Write Protection bit


PIC18F87K90 FAMILY
REGISTER 28-11: CONFIG6H: CONFIGURATION REGISTER 6 HIGH (BYTE ADDRESS 30000Bh)(2)
R/C-1 R/C-1 R-1 U-0 U-0 U-0 U-0 U-0
WRTD WRTB WRTC(1) — — — — —
bit 7 bit 0

bit 7 WRTD: Data EEPROM Write Protection bit

1 = Data EEPROM is not write-protected
0 = Data EEPROM is write-protected
bit 6 WRTB: Boot Block Write Protection bit
1 = Boot block is not write-protected
0 = Boot block is write-protected
bit 5 WRTC: Configuration Register Write Protection bit(1)
1 = Configuration registers are not write-protected
0 = Configuration registers are write-protected
Note 1: This bit is read-only in Normal Execution mode; it can be written only in Program mode.

PIC18F87K90 FAMILY
REGISTER 28-12: CONFIG7L: CONFIGURATION REGISTER 7 LOW (BYTE ADDRESS 30000Ch)(3)
EBTR7(1) EBTR6(1) EBTR5(1) EBTR4(1) EBTR3 EBTR2 EBTR1 EBTR0
bit 7 bit 0

bit 7 EBTR7: Table Read Protection bit(1)

1 = Block 7 is not protected from table reads executed in other blocks
0 = Block 7 is protected from table reads executed in other blocks
bit 3 EBTR3: Table Read Protection bit

2: This bit is only available on PIC18F66K90, PIC18F67K90, PIC18F86K90 and PIC18F87K90 devices.

PIC18F87K90 FAMILY
REGISTER 28-13: CONFIG7H: CONFIGURATION REGISTER 7 HIGH (BYTE ADDRESS 30000Dh)(1)
U-0 R/C-1 U-0 U-0 U-0 U-0 U-0 U-0
— EBTRB — — — — — —
bit 7 bit 0


bit 6 EBTRB: Boot Block Table Read Protection bit
1 = Boot block is not protected from table reads executed in other blocks
0 = Boot block is protected from table reads executed in other blocks
Note 1: For the memory size of the blocks, refer to Figure 28-6.

PIC18F87K90 FAMILY
REGISTER 28-14: DEVID1: DEVICE ID REGISTER 1 FOR THE PIC18F87K90 FAMILY
R R R R R R R R
DEV2 DEV1 DEV0 REV4 REV3 REV2 REV1 REV0
bit 7 bit 0
Legend:
bit 7-5 DEV<2:0>: Device ID bits

Devices with DEV<10:3> of 0101 0010 (see DEVID2):
010 = PIC18F65K90
000 = PIC18F66K90
101 = PIC18F85K90
011 = PIC18F86K90
Devices with DEV<10:3> of 0101 0001:
000 = PIC18F67K90
010 = PIC18F87K90
bit 4-0 REV<4:0>: Revision ID bits
These bits are used to indicate the device revision.
REGISTER 28-15: DEVID2: DEVICE ID REGISTER 2 FOR THE PIC18F87K90 FAMILY

R R R R R R R R
DEV10(1) DEV9(1) DEV8(1) DEV7(1) DEV6(1) DEV5(1) DEV4(1) DEV3(1)
bit 7 bit 0
Legend:
bit 7-0 DEV<10:3>: Device ID bits(1)

These bits are used with the DEV<2:0> bits in the Device ID Register 1 to identify the part number.
0101 0010 = PIC18F65K90, PIC18F66K90, PIC18F85K90 and PIC18F86K90
0101 0001 = PIC18F67K90 and PIC18F87K90
Note 1: These values for DEV<10:3> may be shared with other devices. The specific device is always identified by
using the entire DEV<10:0> bit sequence.

PIC18F87K90 FAMILY
28.2 Watchdog Timer (WDT) The WDT can be operated in one of four modes as
determined by the CONFIG2H bits (WDTEN<1:0>) The
For the PIC18F87K90 family of devices, the WDT is four modes are:
driven by the LF-INTOSC source. When the WDT is
enabled, the clock source is also enabled. The nominal • WDT Enabled
WDT period is 4 ms and has the same stability as the • WDT Disabled
LF-INTOSC oscillator. • WDT under Software Control (WDTCON<0>,
The 4 ms period of the WDT is multiplied by a 16-bit SWDTEN)
postscaler. Any output of the WDT postscaler is • WDT
selected by a multiplexer, controlled by bits in - Enabled during normal operation
Configuration Register 2H. Available periods range - Disabled during Sleep
from 4 ms to 4,194 seconds (about one hour). The
WDT and postscaler are cleared when any of the
following events occur: a SLEEP or CLRWDT instruction Note 1: The CLRWDT and SLEEP instructions
is executed, the IRCF bits (OSCCON<6:4>) are clear the WDT and postscaler counts
changed or a clock failure has occurred. when executed.
2: Changing the setting of the IRCF bits
(OSCCON<6:4>) clears the WDT and
postscaler counts.
3: When a CLRWDT instruction is executed,
the postscaler count will be cleared.
FIGURE 28-1: WDT BLOCK DIAGRAM
WDT Enabled,
SWDTEN Disabled
WDT Controlled with
SWDTEN bit Setting
WDT Enabled only While
Device Active, Disabled
WDT Disabled in Hardware,
SWDTEN Disabled
WDTEN1 Enable WDT

WDTEN0
WDT Counter
Wake-up from
INTRC Source 128
Power-Managed
Modes
Change on IRCF<2:0> bits
Programmable Postscaler Reset WDT
CLRWDT Reset
1:1 to 1:1,048,576
All Device Resets
WDTPS<3:0> 4
Sleep
SWDTEN Enable WDT

WDTEN<1:0>
INTRC Source

PIC18F87K90 FAMILY
28.2.1 CONTROL REGISTER
Register 28-16 shows the WDTCON register. This is a
readable and writable register which contains a control
bit that allows software to override the WDT Enable
Configuration bit, but only if the Configuration bit has
disabled the WDT.
REGISTER 28-16: WDTCON: WATCHDOG TIMER CONTROL REGISTER

R/W-0 U-0 R-x R/W-0 U-0 R/W-0 R/W-0 R/W-0
REGSLP — ULPLVL(3) SRETEN(2) — ULPEN ULPSINK(3) SWDTEN(1)
bit 7 bit 0
Legend:
bit 7 REGSLP: Regulator Voltage Sleep Enable bit

1 = Regulator goes into Low-Power mode when device’s Sleep mode is enabled
0 = Regulator stays in normal mode when device’s Sleep mode is activated
bit 5 ULPLVL: Ultra Low-Power Wake-up Output bit(3)
1 = Voltage on RA0 > ~0.5V
0 = Voltage on RA0 < ~0.5V
bit 4 SRETEN: Regulator Voltage Sleep Disable bit(2)
1 = If RETEN (CONFIG1L<0>) = 0 and the regulator is enabled, the device goes into Ultra
Low-Power mode in Sleep
0 = The regulator is on when the device’s Sleep mode is enabled and the Low-Power mode is
controlled by REGSLP
bit 2 ULPEN: Ultra Low-Power Wake-up (ULPWU) Module Enable bit
1 = Ultra Low-Power Wake-up module is enabled; ULPLVL bit indicates a comparator output
0 = Ultra Low-Power Wake-up module is disabled
bit 1 ULPSINK: Ultra Low-Power Wake-up Current Sink Enable bit(3)
1 = Ultra Low-Power Wake-up current sink is enabled
0 = Ultra Low-Power Wake-up current sink is disabled
bit 0 SWDTEN: Software Controlled Watchdog Timer Enable bit(1)
1 = Watchdog Timer is on
0 = Watchdog Timer is off
Note 1: This bit has no effect if the Configuration bits, WDTEN<1:0>, are enabled.
2: This bit is only available when ENVREG = 1 and RETEN = 0.
3: This bit is not valid unless ULPEN = 1.
TABLE 28-2: SUMMARY OF WATCHDOG TIMER REGISTERS

Reset
Page:
WDTCON REGSLP — ULPLVL SRETEN — ULPEN ULPSINK SWDTEN 76
Legend: — = unimplemented, read as ‘0’. Shaded cells are not used by the Watchdog Timer.

PIC18F87K90 FAMILY
28.3 On-Chip Voltage Regulator FIGURE 28-2: CONNECTIONS FOR THE
ON-CHIP REGULATOR
All of the PIC18F87K90 family devices power their core
digital logic at a nominal 3.3V. For designs that are Regulator Enabled (ENVREG tied to VDD):
required to operate at a higher typical voltage, such as
5V
5V, all family devices incorporate two on-chip regula-
PIC18F87K90
tors that allow the device to run its core logic from VDD.
Those regulators are: VDD
ENVREG
• Normal On-Chip Regulator
• Ultra Low-Power, On-Chip Regulator VDDCORE/VCAP
CF
The hardware configuration of these regulators is the VSS
same and is explained in Section 28.3.1 “Regulator
Enable/disable by Hardware”. The regulators’ only
differences relate to when the device enters Sleep, as
explained in Section 28.3.2.
Regulator Disabled (ENVREG tied to VSS):
28.3.1 REGULATOR ENABLE/DISABLE BY

3.3V(1)
HARDWARE
PIC18F87K90
The regulator can be enabled or disabled only by VDD
hardware. The regulator is controlled by the ENVREG
pin and the VDDCORE/VCAP pin. ENVREG
28.3.1.1 Regulator Enable Mode VDDCORE/VCAP

0.1 F
Tying VDD to the pin enables the regulator, which in turn,
VSS
provides power to the core from the other VDD pins.
When the regulator is enabled, a low-ESR filter capac-
itor must be connected to the VDDCORE/VCAP pin (see
Figure 28-2). This helps maintain the regulator’s Note 1: These are typical operating voltages. For the
stability. The recommended value for the filter capacitor full operating ranges of VDD and VDDCORE,
is given in Section 31.2 DC Characteristics. see Section 31.2 “DC Characteristics”.
28.3.1.2 Regulator Disable Mode

If ENVREG is tied to VSS, the regulator is disabled. In
this case, a 0.1 F capacitor should be connected to
the VDDCORE/VCAP pin (see Figure 28-2).
When the regulator is being used, the overall voltage
budget is very tight. The regulator should operate the
device down to 1.8V. When VDD drops below 3.3V, the
regulator no longer regulates, but the output voltage fol-
lows the input until VDD reaches 1.8V. Below this voltage,
the output of the regulator output may drop to 0V.

PIC18F87K90 FAMILY
28.3.2 OPERATION OF REGULATOR IN The various modes of regulator operation are shown in
SLEEP Table 28-3.
The difference in the two regulators’ operation arises When the ultra low-power regulator is in Sleep mode,
with Sleep mode. The ultra low-power regulator gives the internal reference voltages in the chip will be shut
the device the lowest current in the Regulator Enabled off and any interrupts referring to the internal reference
mode. will not wake up the device. If the BOR or LVD is
enabled, the regulator will keep the internal references
The on-chip regulator can go into a lower power mode,
on and the lowest possible current will not be achieved.
when the device goes to Sleep, by setting the REGSLP
bit (WDTCON<7>). This puts the regulator in a standby When using the ultra low-power regulator in Sleep
mode so that the device consumes much less current. mode, the device will take about 250 s, typical, to start
executing the code after it wakes up.
The on-chip regulator can also go into the Ultra Low-
Power mode, which consumes the lowest current
possible with the regulator enabled. This mode is
controlled by the RETEN bit (CONFIG1L<0>) and
SRETEN bit (WDTCON<4>).
TABLE 28-3: SLEEP MODE REGULATOR SETTINGS(1)

VREGSLP SRETEN RETEN
Regulator Power Mode
WDTCON<7> WDTCON<4> CONFIG1L<0>
Enabled Normal Operation (Sleep) 0 x 1
Enabled Low-Power mode (Sleep) 1 x 1
Enabled Normal Operation (Sleep) 0 0 x
Enabled Low-Power mode (Sleep) 1 0 x
Enabled Ultra Low-Power mode (Sleep) x 1 0
Note 1: x = Indicates that VIT status is invalid.

PIC18F87K90 FAMILY
28.4 Two-Speed Start-up In all other power-managed modes, Two-Speed Start-
up is not used. The device will be clocked by the
The Two-Speed Start-up feature helps to minimize the currently selected clock source until the primary clock
latency period, from oscillator start-up to code execution, source becomes available. The setting of the IESO bit
by allowing the microcontroller to use the INTOSC is ignored.
(LF-INTOSC, MF-INTOSC, HF-INTOSC) oscillator as a
clock source, until the primary clock source is available. 28.4.1 SPECIAL CONSIDERATIONS FOR
It is enabled by setting the IESO Configuration bit. USING TWO-SPEED START-UP
Two-Speed Start-up should be enabled only if the While using the INTOSC oscillator in Two-Speed Start-
primary oscillator mode is LP, XT or HS (Crystal-Based up, the device still obeys the normal command
modes). Other sources do not require an OST start-up sequences for entering power-managed modes,
delay; for these, Two-Speed Start-up should be disabled. including multiple SLEEP instructions (refer to
When enabled, Resets and wake-ups from Sleep mode Section 4.1.4 “Multiple Sleep Commands”). In
cause the device to configure itself to run from the practice, this means that user code can change the
internal oscillator block as the clock source, following SCS<1:0> bit settings or issue SLEEP instructions
the time-out of the Power-up Timer after a Power-on before the OST times out. This would allow an
Reset is enabled. This allows almost immediate code application to briefly wake-up, perform routine
execution while the primary oscillator starts and the “housekeeping” tasks and return to Sleep before the
OST is running. Once the OST times out, the device device starts to operate from the primary oscillator.
automatically switches to PRI_RUN mode. User code can also check if the primary clock source is
To use a higher clock speed on wake-up, the INTOSC currently providing the device clocking by checking the
or postscaler clock sources can be selected to provide status of the OSTS bit (OSCCON<3>). If the bit is set,
a higher clock speed by setting bits, IRCF<2:0>, the primary oscillator is providing the clock. Otherwise,
immediately after Reset. For wake-ups from Sleep, the the internal oscillator block is providing the clock during
INTOSC or postscaler clock sources can be selected wake-up from Reset or Sleep mode.
by setting the IRCF<2:0> bits prior to entering Sleep
mode.
FIGURE 28-3: TIMING TRANSITION FOR TWO-SPEED START-UP (INTOSC TO HSPLL)
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3
INTOSC
Multiplexer
OSC1
TOST(1) TPLL(1)
1 2 n-1 n
PLL Clock
Output
Clock
Transition(2)
CPU Clock
Peripheral
Clock
Program
Counter PC PC + 2 PC + 4 PC + 6
Wake from Interrupt Event OSTS bit Set
Note 1: TOST = 1024 TOSC; TPLL = 2 ms (approx). These intervals are not shown to scale.

PIC18F87K90 FAMILY
28.5 Fail-Safe Clock Monitor To use a higher clock speed on wake-up, the INTOSC
or postscaler clock sources can be selected to provide
The Fail-Safe Clock Monitor (FSCM) allows the micro- a higher clock speed by setting bits, IRCF<2:0>,
controller to continue operation in the event of an external immediately after Reset. For wake-ups from Sleep, the
oscillator failure by automatically switching the device INTOSC or postscaler clock sources can be selected
clock to the internal oscillator block. The FSCM function by setting the IRCF<2:0> bits prior to entering Sleep
is enabled by setting the FCMEN Configuration bit. mode.
When FSCM is enabled, the LF-INTOSC oscillator runs The FSCM will detect only failures of the primary or
at all times to monitor clocks to peripherals and provide secondary clock sources. If the internal oscillator block
a backup clock in the event of a clock failure. Clock fails, no failure would be detected nor would any action
monitoring (shown in Figure 28-4) is accomplished by be possible.
creating a sample clock signal, which is the output from
the LF-INTOSC, divided by 64. This allows ample time 28.5.1 FSCM AND THE WATCHDOG TIMER
between FSCM sample clocks for a peripheral clock
Both the FSCM and the WDT are clocked by the
edge to occur. The peripheral device clock and the
INTOSC oscillator. Since the WDT operates with a
sample clock are presented as inputs to the Clock
separate divider and counter, disabling the WDT has
Monitor (CM) latch. The CM is set on the falling edge of
no effect on the operation of the INTOSC oscillator
the device clock source, but cleared on the rising edge
when the FSCM is enabled.
of the sample clock.
As already noted, the clock source is switched to the
FIGURE 28-4: FSCM BLOCK DIAGRAM INTOSC clock when a clock failure is detected.
Depending on the frequency selected by the
Clock Monitor IRCF<2:0> bits, this may mean a substantial change in
Latch (CM)
the speed of code execution. If the WDT is enabled
(edge-triggered)
Peripheral with a small prescale value, a decrease in clock speed
S Q allows a WDT time-out to occur and a subsequent
Clock
device Reset. For this reason, Fail-Safe Clock events
also reset the WDT and postscaler, allowing it to start
INTRC timing from when execution speed was changed, and
÷ 64 C Q
Source decreasing the likelihood of an erroneous time-out.
(32 s) 488 Hz 28.5.2 EXITING FAIL-SAFE OPERATION
(2.048 ms)
The Fail-Safe condition is terminated by either a device
Clock Reset or by entering a power-managed mode. On
Failure Reset, the controller starts the primary clock source
Detected specified in Configuration Register 1H (with any
required start-up delays that are required for the
Clock failure is tested for on the falling edge of the oscillator mode, such as the OST or PLL timer). The
sample clock. If a sample clock falling edge occurs INTOSC multiplexer provides the device clock until the
while CM is still set, a clock failure has been detected primary clock source becomes ready (similar to a Two-
(Figure 28-5). This causes the following: Speed Start-up). The clock source is then switched to
• The FSCM generates an oscillator fail interrupt by the primary clock (indicated by the OSTS bit in the
setting bit, OSCFIF (PIR2<7>) OSCCON register becoming set). The Fail-Safe Clock
• The device clock source switches to the internal Monitor then resumes monitoring the peripheral clock.
oscillator block (OSCCON is not updated to show The primary clock source may never become ready
the current clock source – this is the Fail-Safe during start-up. In this case, operation is clocked by the
condition) INTOSC multiplexer. The OSCCON register will remain
• The WDT is reset in its Reset state until a power-managed mode is
entered.
During switchover, the postscaler frequency from the
internal oscillator block may not be sufficiently stable for
timing-sensitive applications. In these cases, it may be
desirable to select another clock configuration and enter
an alternate power-managed mode. This can be done to
attempt a partial recovery or execute a controlled shut-
down. See Section 4.1.4 “Multiple Sleep Commands”
and Section 28.4.1 “Special Considerations for
Using Two-Speed Start-up” for more details.

PIC18F87K90 FAMILY
FIGURE 28-5: FSCM TIMING DIAGRAM
Sample Clock
Device Oscillator
Clock Failure
Output
CM Output
(Q)
Failure
Detected
OSCFIF
CM Test CM Test CM Test

Note: The device clock is normally at a much higher frequency than the sample clock. The relative frequencies in this
example have been chosen for clarity.
28.5.3 FSCM INTERRUPTS IN For oscillator modes involving a crystal or resonator

POWER-MANAGED MODES (HS, HSPLL, LP or XT), the situation is somewhat
different. Since the oscillator may require a start-up
By entering a power-managed mode, the clock
time considerably longer than the FCSM sample clock
multiplexer selects the clock source selected by the
time, a false clock failure may be detected. To prevent
OSCCON register. Fail-Safe Monitoring of the power-
this, the internal oscillator block is automatically config-
managed clock source resumes in the power-managed
ured as the device clock and functions until the primary
mode.
clock is stable (when the OST and PLL timers have
If an oscillator failure occurs during power-managed timed out).
operation, the subsequent events depend on whether
This is identical to Two-Speed Start-up mode. Once the
or not the Oscillator Failure Interrupt Flag is enabled. If
primary clock is stable, the INTOSC returns to its role
enabled (OSCFIF = 1), code execution will be clocked
as the FSCM source.
by the INTOSC multiplexer. An automatic transition
back to the failed clock source will not occur. Note: The same logic that prevents false oscilla-
If the interrupt is disabled, subsequent interrupts while tor failure interrupts on POR, or wake from
in Idle mode will cause the CPU to begin executing Sleep, also prevents the detection of the
instructions while being clocked by the INTOSC oscillator’s failure to start at all following
source. these events. This can be avoided by
monitoring the OSTS bit and using a
28.5.4 POR OR WAKE FROM SLEEP timing routine to determine if the oscillator
is taking too long to start. Even so, no
The FSCM is designed to detect oscillator failure at any oscillator failure interrupt will be flagged.
point after the device has exited Power-on Reset
(POR) or low-power Sleep mode. When the primary As noted in Section 28.4.1 “Special Considerations
device clock is EC, RC or INTRC modes, monitoring for Using Two-Speed Start-up”, it is also possible to
can begin immediately following these events. select another clock configuration and enter an
alternate power-managed mode while waiting for the
primary clock to become stable. When the new power-
managed mode is selected, the primary clock is
disabled.

PIC18F87K90 FAMILY
28.6 Program Verification and Each of the blocks has three code protection bits
Code Protection associated with them. They are:
• Code-Protect bit (CPn)
The user program memory is divided into four blocks
• Write-Protect bit (WRTn)
for the PIC18FX5K90 device and PIC18FX6K90
devices, and eight blocks for PIC18FX7K90 devices. • External Block Table Read bit (EBTRn)
One of these is a boot block of 1 or 2 Kbytes. The Figure 28-6 shows the program memory organization for
remainder of the memory is divided into blocks on 48, 64, 96 and 128-Kbyte devices and the specific code
binary boundaries. protection bit associated with each block. The actual
locations of the bits are summarized in Table 28-4.
FIGURE 28-6: CODE-PROTECTED PROGRAM MEMORY FOR THE PIC18F87K90 FAMILY(1)
000000h
Code Memory
01FFFFh Device/Memory Size(2)
PIC18FX7K90 PIC18FX6K90 PIC18FX5K90
BBSIZ = 1 BBSIZ = 0 BBSIZ = 1 BBSIZ = 0 BBSIZ = 1 BBSIZ = 0 Address
Boot Boot Boot Boot Boot Block Boot 0000h

Block Block Block Block 2 kW Block
Unimplemented
2 kW 2 kW Block 0 Block 0 0800h
Read as ‘‘00’’
Read as
7 kW 3 kW
Block 0 Block 0 1000h
Block 0
Block 0 6 kW 2 kW 17FFh
7 kW
6 kW Block 1 Block 1 1800
4 kW 4 kW 3FFF
Block 1 Block 1 Block 2 Block 2 4000h
Block 1 Block 1 8 kW 8 kW 4 kW 4 kW 5FFFh
8 kW 8 kW Block 3 Block 3 6000h
200000h
4 kW 4 kW 7FFF
Block 2 Block 2 Block 2 Block 2 8000h
8 kW 8 kW 8 kW 8 kW BFFFh
Block 3 Block 3 Block 3 Block 3 C000h
8 kW 8 kW 8 kW 8 kW FFFFh
Configuration
and ID
8 kW 8 kW 13FFFh
Space
8 kW 8 kW 17FFFh
8 kW 8 kW 1BFFFh
Block 7 Block 7 1C000h
8 kW 8 kW 1FFFFh
3FFFFFh
Note 1: Sizes of memory areas are not to scale.

2: Boot block size is determined by the BBSIZ0 bit (CONFIG4L<4>).

PIC18F87K90 FAMILY
TABLE 28-4: SUMMARY OF CODE PROTECTION REGISTERS
300008h CONFIG5L CP7(1) CP6(1) CP5(1) CP4(1) CP3 CP2 CP1 CP0
300009h CONFIG5H CPD CPB — — — — — —
(1) (1) (1) (1)
30000Ah CONFIG6L WRT7 WRT6 WRT5 WRT4 WRT3 WRT2 WRT1 WRT0
30000Bh CONFIG6H WRTD WRTB WRTC — — — — —
30000Ch CONFIG7L EBRT7(1) EBRT6 (1)
EBTR5 (1)
EBTR4 (1)
EBTR3 EBTR2 EBTR1 EBTR0
30000Dh CONFIG7H — EBTRB — — — — — —
Legend: Shaded cells are unimplemented.
Note 1: This bit is available only on the PIC18F67K90 and PIC18F87K90 devices.
28.6.1 PROGRAM MEMORY location outside of that block is not allowed to read and
CODE PROTECTION will result in reading ‘0’s. Figures 28-7 through 28-9
illustrate table write and table read protection.
The program memory may be read to, or written from,
any location using the table read and table write Note: Code protection bits may only be written
instructions. The Device ID may be read with table to a ‘0’ from a ‘1’ state. It is not possible to
reads. The Configuration registers may be read and write a ‘1’ to a bit in the ‘0’ state. Code
written with the table read and table write instructions. protection bits are only set to ‘1’ by a full
In Normal Execution mode, the CPn bits have no direct chip erase or block erase function. The full
effect. CPn bits inhibit external reads and writes. A block chip erase and block erase functions can
of user memory may be protected from table writes if the only be initiated via ICSP or an external
WRTn Configuration bit is ‘0’. programmer. Refer to the device
programming specification for more
The EBTRn bits control table reads. For a block of user information.
memory with the EBTRn bit set to ‘0’, a table read
instruction that executes from within that block is allowed
to read. A table read instruction that executes from a
FIGURE 28-7: TABLE WRITE (WRTn) DISALLOWED

Register Values Program Memory Configuration Bit Settings
000000h
WRTB, EBTRB = 11
0007FFh
000800h
TBLPTR = 0008FFh
WRT0, EBTR0 = 01
PC = 003FFEh TBLWT* 003FFFh

004000h
WRT1, EBTR1 = 11
007FFFh
008000h
PC = 00BFFEh TBLWT* WRT2, EBTR2 = 11
00BFFFh
00C000h
WRT3, EBTR3 = 11
00FFFFh
Results: All table writes are disabled to Blockn whenever WRTn = 0.

PIC18F87K90 FAMILY
FIGURE 28-8: EXTERNAL BLOCK TABLE READ (EBTRn) DISALLOWED
000000h
WRTB, EBTRB = 11
0007FFh
000800h
TBLPTR = 0008FFh
WRT0, EBTR0 = 10
003FFFh
004000h
PC = 007FFEh TBLRD* WRT1, EBTR1 = 11
007FFFh
008000h
WRT2, EBTR2 = 11
00BFFFh
00C000h
WRT3, EBTR3 = 11
00FFFFh
Results: All table reads from external blocks to Blockn are disabled whenever EBTRn = 0.
The TABLAT register returns a value of ‘0’.
FIGURE 28-9: EXTERNAL BLOCK TABLE READ (EBTRn) ALLOWED

000000h
WRTB, EBTRB = 11
0007FFh
000800h
TBLPTR = 0008FFh WRT0, EBTR0 = 10
PC = 003FFEh TBLRD* 003FFFh

004000h
WRT1, EBTR1 = 11
007FFFh
008000h
WRT2, EBTR2 = 11
00BFFFh
00C000h
WRT3, EBTR3 = 11
00FFFFh
Results: Table reads are permitted within Blockn, even when EBTRBn = 0.
The TABLAT register returns the value of the data at the location, TBLPTR.

PIC18F87K90 FAMILY
28.6.2 DATA EEPROM 28.9 In-Circuit Debugger
CODE PROTECTION
When the DEBUG Configuration bit is programmed to
The entire data EEPROM is protected from external a ‘0’, the In-Circuit Debugger (ICD) functionality is
reads and writes by two bits: CPD and WRTD. CPD enabled. This function allows simple debugging
inhibits external reads and writes of data EEPROM. functions when used with MPLAB® IDE. When the
WRTD inhibits internal and external writes to data microcontroller has this feature enabled, some
EEPROM. The CPU can always read data EEPROM resources are not available for general use. Table 28-5
under normal operation, regardless of the protection bit shows which resources are required by the background
settings. debugger.
28.6.3 CONFIGURATION REGISTER

TABLE 28-5: DEBUGGER RESOURCES
PROTECTION
I/O Pins: RB6, RB7
The Configuration registers can be write-protected.
Stack: Two Levels
The WRTC bit controls protection of the Configuration
registers. In Normal Execution mode, the WRTC bit is Program Memory: 512 Bytes
readable only. WRTC can only be written via ICSP or Data Memory: 10 Bytes
an external programmer.
To use the In-Circuit Debugger function of the micro-
controller, the design must implement In-Circuit Serial
28.7 ID Locations
Programming connections to MCLR/RG5/VPP, VDD,
Eight memory locations (200000h-200007h) are VSS, RB7 and RB6. This will interface to the In-Circuit
designated as ID locations, where the user can store Debugger module, available from Microchip or one of
checksum or other code identification numbers. These the third party development tool companies.
locations are both readable and writable, during Nor-
mal Execution mode through the TBLRD and TBLWT
instructions, or during program/verify. The ID locations
can be read when the device is code-protected.
28.8 In-Circuit Serial Programming

The PIC18F87K90 family of devices can be serially
programmed while in the end application circuit. This is
simply done with two lines for clock and data, and three
other lines for power, ground and the programming
voltage. This allows customers to manufacture boards
with unprogrammed devices and then program the
microcontroller just before shipping the product. This
also allows the most recent firmware or a custom
firmware to be programmed. For the various
programming modes, please refer to the device
programming specification.

PIC18F87K90 FAMILY
29.0 INSTRUCTION SET SUMMARY The literal instructions may use some of the following
operands:
The PIC18F87K90 family of devices incorporates the
• A literal value to be loaded into a file register
standard set of 75 PIC18 core instructions, as well as
(specified by ‘k’)
an extended set of 8 new instructions for the optimiza-
tion of code that is recursive or that utilizes a software • The desired FSR register to load the literal value
stack. The extended set is discussed later in this into (specified by ‘f’)
section. • No operand required
(specified by ‘—’)
29.1 Standard Instruction Set The control instructions may use some of the following
operands:
The standard PIC18 MCU instruction set adds many
enhancements to the previous PIC® MCU instruction • A program memory address (specified by ‘n’)
sets, while maintaining an easy migration from these • The mode of the CALL or RETURN instructions
PIC MCU instruction sets. Most instructions are a (specified by ‘s’)
single program memory word (16 bits), but there are • The mode of the table read and table write
four instructions that require two program memory instructions (specified by ‘m’)
locations. • No operand required
Each single-word instruction is a 16-bit word divided (specified by ‘—’)
into an opcode, which specifies the instruction type and All instructions are a single word, except for four
one or more operands, which further specify the double-word instructions. These instructions were
operation of the instruction. made double-word to contain the required information
The instruction set is highly orthogonal and is grouped in 32 bits. In the second word, the 4 MSbs are ‘1’s. If
into four basic categories: this second word is executed as an instruction (by
• Byte-oriented operations itself), it will execute as a NOP.
• Bit-oriented operations All single-word instructions are executed in a single
• Literal operations instruction cycle, unless a conditional test is true or the
Program Counter is changed as a result of the instruc-
• Control operations
tion. In these cases, the execution takes two instruction
The PIC18 instruction set summary in Table 29-2 lists cycles with the additional instruction cycle(s) executed
byte-oriented, bit-oriented, literal and control as a NOP.
operations. Table 29-1 shows the opcode field
The double-word instructions execute in two instruction
descriptions.
cycles.
Most byte-oriented instructions have three operands:
One instruction cycle consists of four oscillator periods.
1. The file register (specified by ‘f’) Thus, for an oscillator frequency of 4 MHz, the normal
2. The destination of the result (specified by ‘d’) instruction execution time is 1 s. If a conditional test is
3. The accessed memory (specified by ‘a’) true, or the Program Counter is changed as a result of
an instruction, the instruction execution time is 2 s.
The file register designator, ‘f’, specifies which file reg- Two-word branch instructions (if true) would take 3 s.
ister is to be used by the instruction. The destination
designator, ‘d’, specifies where the result of the Figure 29-1 shows the general formats that the instruc-
operation is to be placed. If ‘d’ is zero, the result is tions can have. All examples use the convention ‘nnh’
placed in the WREG register. If ‘d’ is one, the result is to represent a hexadecimal number.
placed in the file register specified in the instruction. The Instruction Set Summary, shown in Table 29-2,
All bit-oriented instructions have three operands: lists the standard instructions recognized by the
Microchip MPASMTM Assembler.
1. The file register (specified by ‘f’)
Section 29.1.1 “Standard Instruction Set” provides
2. The bit in the file register (specified by ‘b’)
a description of each instruction.
3. The accessed memory (specified by ‘a’)
The bit field designator, ‘b’, selects the number of the bit
affected by the operation, while the file register desig-
nator, ‘f’, represents the number of the file in which the
bit is located.

PIC18F87K90 FAMILY
TABLE 29-1: OPCODE FIELD DESCRIPTIONS
Field Description
a RAM access bit:
a = 0: RAM location in Access RAM (BSR register is ignored)
a = 1: RAM bank is specified by BSR register
bbb Bit address within an 8-bit file register (0 to 7).
BSR Bank Select Register. Used to select the current RAM bank.
C, DC, Z, OV, N ALU Status bits: Carry, Digit Carry, Zero, Overflow, Negative.
d Destination select bit:
d = 0: store result in WREG
d = 1: store result in file register f
dest Destination: either the WREG register or the specified register file location.
f 8-bit register file address (00h to FFh), or 2-bit FSR designator (0h to 3h).
fs 12-bit register file address (000h to FFFh). This is the source address.
fd 12-bit register file address (000h to FFFh). This is the destination address.
GIE Global Interrupt Enable bit.
k Literal field, constant data or label (may be either an 8-bit, 12-bit or a 20-bit value).
label Label name.
mm The mode of the TBLPTR register for the table read and table write instructions.
Only used with table read and table write instructions:
* No Change to register (such as TBLPTR with table reads and writes)
*+ Post-Increment register (such as TBLPTR with table reads and writes)
*- Post-Decrement register (such as TBLPTR with table reads and writes)
+* Pre-Increment register (such as TBLPTR with table reads and writes)
n The relative address (2’s complement number) for relative branch instructions or the direct address for
Call/Branch and Return instructions.
PC Program Counter.
PCL Program Counter Low Byte.
PCH Program Counter High Byte.
PCLATH Program Counter High Byte Latch.
PCLATU Program Counter Upper Byte Latch.
PD Power-Down bit.
PRODH Product of Multiply High Byte.
PRODL Product of Multiply Low Byte.
s Fast Call/Return mode select bit:
s = 0: do not update into/from shadow registers
s = 1: certain registers loaded into/from shadow registers (Fast mode)
TBLPTR 21-bit Table Pointer (points to a Program Memory location).
TABLAT 8-bit Table Latch.
TO Time-out bit.
TOS Top-of-Stack.
u Unused or Unchanged.
WDT Watchdog Timer.
WREG Working register (accumulator).
x Don’t care (‘0’ or ‘1’). The assembler will generate code with x = 0. It is the recommended form of use for
compatibility with all Microchip software tools.
zs 7-bit offset value for Indirect Addressing of register files (source).
zd 7-bit offset value for Indirect Addressing of register files (destination).
{ } Optional argument.
[text] Indicates an Indexed Address.
(text) The contents of text.
[expr]<n> Specifies bit n of the register indicated by the pointer expr.
 Assigned to.
< > Register bit field.
 In the set of.
italics User-defined term (font is Courier New).

PIC18F87K90 FAMILY
FIGURE 29-1: GENERAL FORMAT FOR INSTRUCTIONS
Byte-oriented file register operations Example Instruction
15 10 9 8 7 0
OPCODE d a f (FILE #) ADDWF MYREG, W, B
d = 0 for result destination to be WREG register
d = 1 for result destination to be file register (f)
a = 0 to force Access Bank
a = 1 for BSR to select bank
f = 8-bit file register address
Byte to Byte move operations (2-word)

15 12 11 0
OPCODE f (Source FILE #) MOVFF MYREG1, MYREG2
15 12 11 0
1111 f (Destination FILE #)
Bit-oriented file register operations

15 12 11 9 8 7 0
OPCODE b (BIT #) a f (FILE #) BSF MYREG, bit, B
b = 3-bit position of bit in file register (f)

a = 0 to force Access Bank
a = 1 for BSR to select bank
Literal operations
15 8 7 0
OPCODE k (literal) MOVLW 7Fh
k = 8-bit immediate value
Control operations
CALL, GOTO and Branch operations
15 8 7 0
OPCODE n<7:0> (literal) GOTO Label
15 12 11 0
1111 n<19:8> (literal)
n = 20-bit immediate value
15 8 7 0
OPCODE S n<7:0> (literal) CALL MYFUNC
15 12 11 0
1111 n<19:8> (literal)
S = Fast bit
15 11 10 0
OPCODE n<10:0> (literal) BRA MYFUNC
15 8 7 0
OPCODE n<7:0> (literal) BC MYFUNC

PIC18F87K90 FAMILY
TABLE 29-2: PIC18F87K90 FAMILY INSTRUCTION SET
Mnemonic, 16-Bit Instruction Word Status
Description Cycles Notes
Operands MSb LSb Affected
BYTE-ORIENTED OPERATIONS
ADDWF f, d, a Add WREG and f 1 0010 01da ffff ffff C, DC, Z, OV, N 1, 2
ADDWFC f, d, a Add WREG and Carry bit to f 1 0010 00da ffff ffff C, DC, Z, OV, N 1, 2
ANDWF f, d, a AND WREG with f 1 0001 01da ffff ffff Z, N 1, 2
CLRF f, a Clear f 1 0110 101a ffff ffff Z 2
COMF f, d, a Complement f 1 0001 11da ffff ffff Z, N 1, 2
CPFSEQ f, a Compare f with WREG, Skip = 1 (2 or 3) 0110 001a ffff ffff None 4
CPFSGT f, a Compare f with WREG, Skip > 1 (2 or 3) 0110 010a ffff ffff None 4
CPFSLT f, a Compare f with WREG, Skip < 1 (2 or 3) 0110 000a ffff ffff None 1, 2
DECF f, d, a Decrement f 1 0000 01da ffff ffff C, DC, Z, OV, N 1, 2, 3, 4
DECFSZ f, d, a Decrement f, Skip if 0 1 (2 or 3) 0010 11da ffff ffff None 1, 2, 3, 4
DCFSNZ f, d, a Decrement f, Skip if Not 0 1 (2 or 3) 0100 11da ffff ffff None 1, 2
INCF f, d, a Increment f 1 0010 10da ffff ffff C, DC, Z, OV, N 1, 2, 3, 4
INCFSZ f, d, a Increment f, Skip if 0 1 (2 or 3) 0011 11da ffff ffff None 4
INFSNZ f, d, a Increment f, Skip if Not 0 1 (2 or 3) 0100 10da ffff ffff None 1, 2
IORWF f, d, a Inclusive OR WREG with f 1 0001 00da ffff ffff Z, N 1, 2
MOVF f, d, a Move f 1 0101 00da ffff ffff Z, N 1
MOVFF fs, fd Move fs (source) to 1st word 2 1100 ffff ffff ffff None
fd (destination) 2nd word 1111 ffff ffff ffff
MOVWF f, a Move WREG to f 1 0110 111a ffff ffff None
MULWF f, a Multiply WREG with f 1 0000 001a ffff ffff None 1, 2
NEGF f, a Negate f 1 0110 110a ffff ffff C, DC, Z, OV, N
RLCF f, d, a Rotate Left f through Carry 1 0011 01da ffff ffff C, Z, N 1, 2
RLNCF f, d, a Rotate Left f (No Carry) 1 0100 01da ffff ffff Z, N
RRCF f, d, a Rotate Right f through Carry 1 0011 00da ffff ffff C, Z, N
RRNCF f, d, a Rotate Right f (No Carry) 1 0100 00da ffff ffff Z, N
SETF f, a Set f 1 0110 100a ffff ffff None 1, 2
SUBFWB f, d, a Subtract f from WREG with 1 0101 01da ffff ffff C, DC, Z, OV, N
Borrow
SUBWF f, d, a Subtract WREG from f 1 0101 11da ffff ffff C, DC, Z, OV, N 1, 2
SUBWFB f, d, a Subtract WREG from f with 1 0101 10da ffff ffff C, DC, Z, OV, N
Borrow
SWAPF f, d, a Swap Nibbles in f 1 0011 10da ffff ffff None 4
TSTFSZ f, a Test f, Skip if 0 1 (2 or 3) 0110 011a ffff ffff None 1, 2
XORWF f, d, a Exclusive OR WREG with f 1 0001 10da ffff ffff Z, N
Note 1: When a PORT register is modified as a function of itself (e.g., MOVF PORTB, 1, 0), the value used will be that
value present on the pins themselves. For example, if the data latch is ‘1’ for a pin configured as input and is
driven low by an external device, the data will be written back with a ‘0’.
2: If this instruction is executed on the TMR0 register (and, where applicable, d = 1), the prescaler will be cleared if
assigned.
3: If the Program Counter (PC) is modified or a conditional test is true, the instruction requires two cycles. The
second cycle is executed as a NOP.
4: Some instructions are two-word instructions. The second word of these instructions will be executed as a NOP
unless the first word of the instruction retrieves the information embedded in these 16 bits. This ensures that all
program memory locations have a valid instruction.

PIC18F87K90 FAMILY
TABLE 29-2: PIC18F87K90 FAMILY INSTRUCTION SET (CONTINUED)
BIT-ORIENTED OPERATIONS
BCF f, b, a Bit Clear f 1 1001 bbba ffff ffff None 1, 2
BSF f, b, a Bit Set f 1 1000 bbba ffff ffff None 1, 2
BTFSC f, b, a Bit Test f, Skip if Clear 1 (2 or 3) 1011 bbba ffff ffff None 3, 4
BTFSS f, b, a Bit Test f, Skip if Set 1 (2 or 3) 1010 bbba ffff ffff None 3, 4
BTG f, b, a Bit Toggle f 1 0111 bbba ffff ffff None 1, 2
CONTROL OPERATIONS
BC n Branch if Carry 1 (2) 1110 0010 nnnn nnnn None
BN n Branch if Negative 1 (2) 1110 0110 nnnn nnnn None
BNC n Branch if Not Carry 1 (2) 1110 0011 nnnn nnnn None
BNN n Branch if Not Negative 1 (2) 1110 0111 nnnn nnnn None
BNOV n Branch if Not Overflow 1 (2) 1110 0101 nnnn nnnn None
BNZ n Branch if Not Zero 1 (2) 1110 0001 nnnn nnnn None
BOV n Branch if Overflow 1 (2) 1110 0100 nnnn nnnn None
BRA n Branch Unconditionally 2 1101 0nnn nnnn nnnn None
BZ n Branch if Zero 1 (2) 1110 0000 nnnn nnnn None
CALL n, s Call Subroutine 1st word 2 1110 110s kkkk kkkk None
2nd word 1111 kkkk kkkk kkkk
CLRWDT — Clear Watchdog Timer 1 0000 0000 0000 0100 TO, PD
DAW — Decimal Adjust WREG 1 0000 0000 0000 0111 C
GOTO n Go to Address 1st word 2 1110 1111 kkkk kkkk None
2nd word 1111 kkkk kkkk kkkk
NOP — No Operation 1 0000 0000 0000 0000 None
NOP — No Operation 1 1111 xxxx xxxx xxxx None 4
POP — Pop Top of Return Stack (TOS) 1 0000 0000 0000 0110 None
PUSH — Push Top of Return Stack (TOS) 1 0000 0000 0000 0101 None
RCALL n Relative Call 2 1101 1nnn nnnn nnnn None
RESET Software Device Reset 1 0000 0000 1111 1111 All
RETFIE s Return from Interrupt Enable 2 0000 0000 0001 000s GIE/GIEH,
PEIE/GIEL
RETLW k Return with Literal in WREG 2 0000 1100 kkkk kkkk None
RETURN s Return from Subroutine 2 0000 0000 0001 001s None
SLEEP — Go into Standby mode 1 0000 0000 0000 0011 TO, PD
assigned.

PIC18F87K90 FAMILY
TABLE 29-2: PIC18F87K90 FAMILY INSTRUCTION SET (CONTINUED)
LITERAL OPERATIONS
ADDLW k Add Literal and WREG 1 0000 1111 kkkk kkkk C, DC, Z, OV, N
ANDLW k AND Literal with WREG 1 0000 1011 kkkk kkkk Z, N
IORLW k Inclusive OR Literal with WREG 1 0000 1001 kkkk kkkk Z, N
LFSR f, k Move literal (12-bit) 2nd word 2 1110 1110 00ff kkkk None
to FSR(f) 1st word 1111 0000 kkkk kkkk
MOVLB k Move Literal to BSR<3:0> 1 0000 0001 0000 kkkk None
MOVLW k Move Literal to WREG 1 0000 1110 kkkk kkkk None
MULLW k Multiply Literal with WREG 1 0000 1101 kkkk kkkk None
RETLW k Return with Literal in WREG 2 0000 1100 kkkk kkkk None
SUBLW k Subtract WREG from Literal 1 0000 1000 kkkk kkkk C, DC, Z, OV, N
XORLW k Exclusive OR Literal with WREG 1 0000 1010 kkkk kkkk Z, N
DATA MEMORY  PROGRAM MEMORY OPERATIONS
TBLRD* Table Read 2 0000 0000 0000 1000 None
TBLRD*+ Table Read with Post-Increment 0000 0000 0000 1001 None
TBLRD*- Table Read with Post-Decrement 0000 0000 0000 1010 None
TBLRD+* Table Read with Pre-Increment 0000 0000 0000 1011 None
TBLWT* Table Write 2 0000 0000 0000 1100 None
TBLWT*+ Table Write with Post-Increment 0000 0000 0000 1101 None
TBLWT*- Table Write with Post-Decrement 0000 0000 0000 1110 None
TBLWT+* Table Write with Pre-Increment 0000 0000 0000 1111 None
assigned.

PIC18F87K90 FAMILY
29.1.1 STANDARD INSTRUCTION SET
ADDLW ADD Literal to W ADDWF ADD W to f
Syntax: ADDLW k Syntax: ADDWF f {,d {,a}}

Operands: 0  k  255 Operands: 0  f  255
Operation: (W) + k  W d  [0,1]
a  [0,1]
Status Affected: N, OV, C, DC, Z
Operation: (W) + (f)  dest
Encoding: 0000 1111 kkkk kkkk
Description: The contents of W are added to the
Encoding: 0010 01da ffff ffff
8-bit literal ‘k’ and the result is placed in
W. Description: Add W to register ‘f’. If ‘d’ is ‘0’, the
result is stored in W. If ‘d’ is ‘1’, the
Words: 1
result is stored back in register ‘f’.
Cycles: 1
If ‘a’ is ‘0’, the Access Bank is selected.
Q Cycle Activity: If ‘a’ is ‘1’, the BSR is used to select the
Q1 Q2 Q3 Q4 GPR bank.
Decode Read Process Write to If ‘a’ is ‘0’ and the extended instruction
literal ‘k’ Data W set is enabled, this instruction operates
in Indexed Literal Offset Addressing
Example: ADDLW 15h
mode whenever f 95 (5Fh). See
Section 29.2.3 “Byte-Oriented and
Before Instruction Bit-Oriented Instructions in Indexed
W = 10h Literal Offset Mode” for details.
After Instruction
Words: 1
W = 25h
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4
Decode Read Process Write to
register ‘f’ Data destination
Example: ADDWF REG, 0, 0

Before Instruction
W = 17h
REG = 0C2h
After Instruction
W = 0D9h
REG = 0C2h
Note: All PIC18 instructions may take an optional label argument preceding the instruction mnemonic for use in
symbolic addressing. If a label is used, the instruction format then becomes: {label} instruction argument(s).

PIC18F87K90 FAMILY
ADDWFC ADD W and Carry bit to f ANDLW AND Literal with W
Syntax: ADDWFC f {,d {,a}} Syntax: ANDLW k

Operands: 0  f  255 Operands: 0  k  255
d [0,1] Operation: (W) .AND. k  W
a [0,1]
Status Affected: N, Z
Operation: (W) + (f) + (C)  dest
Status Affected: N,OV, C, DC, Z
Description: The contents of W are ANDed with the
8-bit literal ‘k’. The result is placed in W.
Description: Add W, the Carry flag and data memory Words: 1
location ‘f’. If ‘d’ is ‘0’, the result is
placed in W. If ‘d’ is ‘1’, the result is Cycles: 1
placed in data memory location ‘f’. Q Cycle Activity:
If ‘a’ is ‘0’, the Access Bank is selected. Q1 Q2 Q3 Q4
If ‘a’ is ‘1’, the BSR is used to select the Decode Read literal Process Write to
GPR bank. ‘k’ Data W
If ‘a’ is ‘0’ and the extended instruction
set is enabled, this instruction operates Example: ANDLW 05Fh
mode whenever f 95 (5Fh). See Before Instruction
Section 29.2.3 “Byte-Oriented and W = A3h
Bit-Oriented Instructions in Indexed After Instruction
Literal Offset Mode” for details. W = 03h
Words: 1
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4
Example: ADDWFC REG, 0, 1

Before Instruction
Carry bit = 1
REG = 02h
W = 4Dh
After Instruction
Carry bit = 0
REG = 02h
W = 50h

PIC18F87K90 FAMILY
ANDWF AND W with f BC Branch if Carry
Syntax: ANDWF f {,d {,a}} Syntax: BC n

Operands: 0  f  255 Operands: -128  n  127
d [0,1] Operation: if Carry bit is ‘1’,
a [0,1]
(PC) + 2 + 2n  PC
Operation: (W) .AND. (f)  dest Status Affected: None
Status Affected: N, Z Encoding: 1110 0010 nnnn nnnn
Description: If the Carry bit is ’1’, then the program
Description: The contents of W are ANDed with will branch.
register ‘f’. If ‘d’ is ‘0’, the result is stored
The 2’s complement number ‘2n’ is
in W. If ‘d’ is ‘1’, the result is stored back added to the PC. Since the PC will have
in register ‘f’. incremented to fetch the next
If ‘a’ is ‘0’, the Access Bank is selected. instruction, the new address will be
If ‘a’ is ‘1’, the BSR is used to select the PC + 2 + 2n. This instruction is then a
GPR bank. two-cycle instruction.
If ‘a’ is ‘0’ and the extended instruction Words: 1
set is enabled, this instruction operates
Cycles: 1(2)
mode whenever f 95 (5Fh). See Q Cycle Activity:
Section 29.2.3 “Byte-Oriented and If Jump:
Bit-Oriented Instructions in Indexed Q1 Q2 Q3 Q4
Literal Offset Mode” for details. Decode Read literal Process Write to
Words: 1 ‘n’ Data PC
Cycles: 1 No No No No
operation operation operation operation
Q Cycle Activity: If No Jump:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Decode Read Process Write to Decode Read literal Process No
‘n’ Data operation
Example: ANDWF REG, 0, 0 Example: HERE BC 5

Before Instruction
Before Instruction
W = 17h PC = address (HERE)
REG = C2h
After Instruction
After Instruction
W = 02h If Carry = 1;
PC = address (HERE + 12)
REG = C2h If Carry = 0;

PIC18F87K90 FAMILY
BCF Bit Clear f BN Branch if Negative
Syntax: BCF f, b {,a} Syntax: BN n

0b7 Operation: if Negative bit is ‘1’,
a [0,1]
(PC) + 2 + 2n  PC
Operation: 0  f Status Affected: None
Status Affected: None Encoding: 1110 0110 nnnn nnnn
Encoding: 1001 bbba ffff ffff
Description: If the Negative bit is ‘1’, then the
Description: Bit ‘b’ in register ‘f’ is cleared. program will branch.
If ‘a’ is ‘0’, the Access Bank is selected. The 2’s complement number ‘2n’ is
If ‘a’ is ‘1’, the BSR is used to select the added to the PC. Since the PC will have
GPR bank. incremented to fetch the next
If ‘a’ is ‘0’ and the extended instruction instruction, the new address will be
set is enabled, this instruction operates PC + 2 + 2n. This instruction is then a
two-cycle instruction.
mode whenever f 95 (5Fh). See Words: 1
Cycles: 1(2)
Bit-Oriented Instructions in Indexed
Literal Offset Mode” for details. Q Cycle Activity:
If Jump:
Words: 1
Q1 Q2 Q3 Q4
Cycles: 1
Decode Read literal Process Write to
Q Cycle Activity: ‘n’ Data PC
Q1 Q2 Q3 Q4 No No No No
Decode Read Process Write operation operation operation operation
register ‘f’ Data register ‘f’ If No Jump:
Q1 Q2 Q3 Q4
Example: BCF FLAG_REG, 7, 0 Decode Read literal Process No
‘n’ Data operation
Before Instruction
FLAG_REG = C7h
After Instruction Example: HERE BN Jump
FLAG_REG = 47h
Before Instruction
PC = address (HERE)
After Instruction
If Negative = 1;
PC = address (Jump)
If Negative = 0;

PIC18F87K90 FAMILY
BNC Branch if Not Carry BNN Branch if Not Negative
Syntax: BNC n Syntax: BNN n

Operands: -128  n  127 Operands: -128  n  127
Operation: if Carry bit is ‘0’, Operation: if Negative bit is ‘0’,
(PC) + 2 + 2n  PC (PC) + 2 + 2n  PC
Status Affected: None Status Affected: None
Encoding: 1110 0011 nnnn nnnn Encoding: 1110 0111 nnnn nnnn
Description: If the Carry bit is ‘0’, then the program Description: If the Negative bit is ‘0’, then the
will branch. program will branch.
The 2’s complement number ‘2n’ is The 2’s complement number ‘2n’ is
added to the PC. Since the PC will have added to the PC. Since the PC will have
incremented to fetch the next incremented to fetch the next
instruction, the new address will be instruction, the new address will be
PC + 2 + 2n. This instruction is then a PC + 2 + 2n. This instruction is then a
two-cycle instruction. two-cycle instruction.
Words: 1 Words: 1
Cycles: 1(2) Cycles: 1(2)
Q Cycle Activity: Q Cycle Activity:
If Jump: If Jump:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Decode Read literal Process Write to Decode Read literal Process Write to
‘n’ Data PC ‘n’ Data PC
No No No No No No No No
operation operation operation operation operation operation operation operation
If No Jump: If No Jump:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Decode Read literal Process No Decode Read literal Process No
‘n’ Data operation ‘n’ Data operation
Example: HERE BNC Jump Example: HERE BNN Jump

Before Instruction Before Instruction
PC = address (HERE) PC = address (HERE)
After Instruction After Instruction
If Carry = 0; If Negative = 0;
PC = address (Jump) PC = address (Jump)
If Carry = 1; If Negative = 1;
PC = address (HERE + 2) PC = address (HERE + 2)

PIC18F87K90 FAMILY
BNOV Branch if Not Overflow BNZ Branch if Not Zero
Syntax: BNOV n Syntax: BNZ n

Operands: -128  n  127 Operands: -128  n  127
Operation: if Overflow bit is ‘0’, Operation: if Zero bit is ‘0’,
(PC) + 2 + 2n  PC (PC) + 2 + 2n  PC
Encoding: 1110 0101 nnnn nnnn Encoding: 1110 0001 nnnn nnnn
Description: If the Overflow bit is ‘0’, then the Description: If the Zero bit is ‘0’, then the program
program will branch. will branch.
The 2’s complement number ‘2n’ is The 2’s complement number ‘2n’ is
added to the PC. Since the PC will have added to the PC. Since the PC will have
incremented to fetch the next incremented to fetch the next
instruction, the new address will be instruction, the new address will be
PC + 2 + 2n. This instruction is then a PC + 2 + 2n. This instruction is then a
two-cycle instruction. two-cycle instruction.
Words: 1 Words: 1
If Jump: If Jump:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Decode Read literal Process Write to Decode Read literal Process Write to
‘n’ Data PC ‘n’ Data PC
If No Jump: If No Jump:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Decode Read literal Process No Decode Read literal Process No
‘n’ Data operation ‘n’ Data operation
Example: HERE BNOV Jump Example: HERE BNZ Jump

If Overflow = 0; If Zero = 0;
PC = address (Jump) PC = address (Jump)
If Overflow = 1; If Zero = 1;
PC = address (HERE + 2) PC = address (HERE + 2)

PIC18F87K90 FAMILY
BRA Unconditional Branch BSF Bit Set f
Syntax: BRA n Syntax: BSF f, b {,a}

Operands: -1024  n  1023 Operands: 0  f  255
Operation: (PC) + 2 + 2n  PC 0b7
a [0,1]
Status Affected: None
Operation: 1  f
Encoding: 1101 0nnn nnnn nnnn
Description: Add the 2’s complement number ‘2n’ to
the PC. Since the PC will have
incremented to fetch the next Description: Bit ‘b’ in register ‘f’ is set.
instruction, the new address will be
PC + 2 + 2n. This instruction is a If ‘a’ is ‘1’, the BSR is used to select the
two-cycle instruction. GPR bank.
Words: 1 If ‘a’ is ‘0’ and the extended instruction
Cycles: 2 set is enabled, this instruction operates
Q Cycle Activity: in Indexed Literal Offset Addressing
Q1 Q2 Q3 Q4 Section 29.2.3 “Byte-Oriented and
Decode Read literal Process Write to Bit-Oriented Instructions in Indexed
‘n’ Data PC Literal Offset Mode” for details.
No No No No
Words: 1
Cycles: 1
Q Cycle Activity:
Example: HERE BRA Jump Q1 Q2 Q3 Q4
Before Instruction Decode Read Process Write
PC = address (HERE) register ‘f’ Data register ‘f’
After Instruction
PC = address (Jump)
Example: BSF FLAG_REG, 7, 1
Before Instruction
FLAG_REG = 0Ah
After Instruction
FLAG_REG = 8Ah

PIC18F87K90 FAMILY
BTFSC Bit Test File, Skip if Clear BTFSS Bit Test File, Skip if Set
Syntax: BTFSC f, b {,a} Syntax: BTFSS f, b {,a}

Operands: 0  f  255 Operands: 0  f  255
0b7 0b<7
a [0,1] a [0,1]
Operation: skip if (f) = 0 Operation: skip if (f) = 1
Encoding: 1011 bbba ffff ffff Encoding: 1010 bbba ffff ffff
Description: If bit ‘b’ in register ‘f’ is ‘0’, then the next Description: If bit ‘b’ in register ‘f’ is ‘1’, then the next
instruction is skipped. If bit ‘b’ is ‘0’, then instruction is skipped. If bit ‘b’ is ‘1’, then
the next instruction fetched during the the next instruction fetched during the
current instruction execution is discarded current instruction execution is discarded
and a NOP is executed instead, making and a NOP is executed instead, making
this a two-cycle instruction. this a two-cycle instruction.
If ‘a’ is ‘0’, the Access Bank is selected. If If ‘a’ is ‘0’, the Access Bank is selected. If
‘a’ is ‘1’, the BSR is used to select the ‘a’ is ‘1’, the BSR is used to select the
GPR bank. GPR bank.
If ‘a’ is ‘0’ and the extended instruction set If ‘a’ is ‘0’ and the extended instruction
is enabled, this instruction operates in set is enabled, this instruction operates in
Indexed Literal Offset Addressing mode Indexed Literal Offset Addressing mode
whenever f 95 (5Fh). See whenever f 95 (5Fh). See
Section 29.2.3 “Byte-Oriented and Section 29.2.3 “Byte-Oriented and
Bit-Oriented Instructions in Indexed Bit-Oriented Instructions in Indexed
Literal Offset Mode” for details. Literal Offset Mode” for details.
Words: 1 Words: 1
Note: 3 cycles if skip and followed Note: 3 cycles if skip and followed
by a 2-word instruction. by a 2-word instruction.
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Decode Read Process No Decode Read Process No
register ‘f’ Data operation register ‘f’ Data operation
If skip: If skip:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
If skip and followed by 2-word instruction: If skip and followed by 2-word instruction:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Example: HERE BTFSC FLAG, 1, 0 Example: HERE BTFSS FLAG, 1, 0

FALSE : FALSE :
TRUE : TRUE :
If FLAG<1> = 0; If FLAG<1> = 0;
PC = address (TRUE) PC = address (FALSE)
If FLAG<1> = 1; If FLAG<1> = 1;
PC = address (FALSE) PC = address (TRUE)

PIC18F87K90 FAMILY
BTG Bit Toggle f BOV Branch if Overflow
Syntax: BTG f, b {,a} Syntax: BOV n

0b<7 Operation: if Overflow bit is ‘1’,
a [0,1]
(PC) + 2 + 2n  PC
Operation: (f)  f Status Affected: None
Status Affected: None Encoding: 1110 0100 nnnn nnnn
Description: If the Overflow bit is ‘1’, then the
Description: Bit ‘b’ in data memory location, ‘f’, is program will branch.
inverted.
The 2’s complement number ‘2n’ is
If ‘a’ is ‘0’, the Access Bank is selected. added to the PC. Since the PC will have
If ‘a’ is ‘1’, the BSR is used to select the incremented to fetch the next
GPR bank. instruction, the new address will be
If ‘a’ is ‘0’ and the extended instruction PC + 2 + 2n. This instruction is then a
in Indexed Literal Offset Addressing Words: 1
Cycles: 1(2)
Bit-Oriented Instructions in Indexed Q Cycle Activity:
Literal Offset Mode” for details. If Jump:
Words: 1 Q1 Q2 Q3 Q4
Decode Read literal Process Write to PC
Cycles: 1 ‘n’ Data
Q Cycle Activity: No No No No
Q1 Q2 Q3 Q4 operation operation operation operation
Decode Read Process Write If No Jump:
register ‘f’ Data register ‘f’ Q1 Q2 Q3 Q4
Decode Read literal Process No
Example: BTG PORTC, 4, 0 ‘n’ Data operation
Before Instruction:
PORTC = 0111 0101 [75h] Example: HERE BOV Jump
After Instruction:
Before Instruction
PORTC = 0110 0101 [65h] PC = address (HERE)
After Instruction
If Overflow = 1;
PC = address (Jump)
If Overflow = 0;

PIC18F87K90 FAMILY
BZ Branch if Zero CALL Subroutine Call
Syntax: BZ n Syntax: CALL k {,s}

Operands: -128  n  127 Operands: 0  k  1048575
Operation: if Zero bit is ‘1’, s [0,1]
(PC) + 2 + 2n  PC Operation: (PC) + 4  TOS,
Status Affected: None k  PC<20:1>;
if s = 1,
Encoding: 1110 0000 nnnn nnnn (W)  WS,
Description: If the Zero bit is ‘1’, then the program (STATUS)  STATUSS,
will branch. (BSR)  BSRS
The 2’s complement number ‘2n’ is Status Affected: None

added to the PC. Since the PC will have Encoding:
incremented to fetch the next 1st word (k<7:0>) 1110 110s k7kkk kkkk0
instruction, the new address will be 2nd word(k<19:8>) 1111 k19kkk kkkk kkkk8
PC + 2 + 2n. This instruction is then a
two-cycle instruction. Description: Subroutine call of entire 2-Mbyte
memory range. First, return address
Words: 1 (PC+ 4) is pushed onto the return stack.
Cycles: 1(2) If ‘s’ = 1, the W, STATUS and BSR
registers are also pushed into their
Q Cycle Activity:
respective shadow registers, WS,
If Jump: STATUSS and BSRS. If ‘s’ = 0, no
Q1 Q2 Q3 Q4 update occurs. Then, the 20-bit value ‘k’
Decode Read literal Process Write to is loaded into PC<20:1>. CALL is a
‘n’ Data PC two-cycle instruction.
No No No No Words: 2
Cycles: 2
If No Jump:
Q1 Q2 Q3 Q4 Q Cycle Activity:
Decode Read literal Process No Q1 Q2 Q3 Q4
‘n’ Data operation Decode Read literal Push PC to Read literal
‘k’<7:0>, stack ’k’<19:8>,
Write to PC
Example: HERE BZ Jump
No No No No
Before Instruction operation operation operation operation
PC = address (HERE)
After Instruction
Example: HERE CALL THERE,1
If Zero = 1;
PC = address (Jump) Before Instruction
If Zero = 0; PC = address (HERE)
After Instruction
PC = address (THERE)
TOS = address (HERE + 4)
WS = W
BSRS = BSR
STATUSS = STATUS

PIC18F87K90 FAMILY
CLRF Clear f CLRWDT Clear Watchdog Timer
Syntax: CLRF f {,a} Syntax: CLRWDT

Operands: 0  f  255 Operands: None
a [0,1] Operation: 000h  WDT,
Operation: 000h  f, 000h  WDT postscaler,
1Z 1  TO,
1  PD
Status Affected: Z
Encoding: 0110 101a ffff ffff Status Affected: TO, PD
Description: Clears the contents of the specified Encoding: 0000 0000 0000 0100
register. Description: CLRWDT instruction resets the
If ‘a’ is ‘0’, the Access Bank is selected. Watchdog Timer. It also resets the post-
If ‘a’ is ‘1’, the BSR is used to select the scaler of the WDT. Status bits, TO and
PD, are set.
GPR bank.
If ‘a’ is ‘0’ and the extended instruction Words: 1
set is enabled, this instruction operates Cycles: 1
in Indexed Literal Offset Addressing Q Cycle Activity:
Q1 Q2 Q3 Q4
Bit-Oriented Instructions in Indexed Decode No Process No
Literal Offset Mode” for details. operation Data operation
Words: 1
Example: CLRWDT
Cycles: 1
Before Instruction
Q Cycle Activity:
WDT Counter = ?
Q1 Q2 Q3 Q4 After Instruction
Decode Read Process Write WDT Counter = 00h
register ‘f’ Data register ‘f’ WDT Postscaler = 0
TO = 1
Example: CLRF FLAG_REG,1 PD = 1
Before Instruction
FLAG_REG = 5Ah
After Instruction
FLAG_REG = 00h

PIC18F87K90 FAMILY
COMF Complement f CPFSEQ Compare f with W, Skip if f = W
Syntax: COMF f {,d {,a}} Syntax: CPFSEQ f {,a}

d  [0,1] a  [0,1]
a  [0,1] Operation: (f) – (W),
Operation: f  dest skip if (f) = (W)
(unsigned comparison)
Encoding: 0110 001a ffff ffff
Description: The contents of register ‘f’ are Description: Compares the contents of data memory
complemented. If ‘d’ is ‘0’, the result is location ‘f’ to the contents of W by
stored in W. If ‘d’ is ‘1’, the result is performing an unsigned subtraction.
stored back in register ‘f’.
If ‘f’ = W, then the fetched instruction is
If ‘a’ is ‘0’, the Access Bank is selected. discarded and a NOP is executed
If ‘a’ is ‘1’, the BSR is used to select the instead, making this a two-cycle
GPR bank. instruction.
If ‘a’ is ‘0’ and the extended instruction If ‘a’ is ‘0’, the Access Bank is selected.
set is enabled, this instruction operates If ‘a’ is ‘1’, the BSR is used to select the
in Indexed Literal Offset Addressing GPR bank.
Section 29.2.3 “Byte-Oriented and If ‘a’ is ‘0’ and the extended instruction
Bit-Oriented Instructions in Indexed set is enabled, this instruction operates
Literal Offset Mode” for details. in Indexed Literal Offset Addressing
Words: 1 Section 29.2.3 “Byte-Oriented and
Cycles: 1 Bit-Oriented Instructions in Indexed
Literal Offset Mode” for details.
Q Cycle Activity:
Q1 Q2 Q3 Q4 Words: 1
Decode Read Process Write to Cycles: 1(2)
register ‘f’ Data destination Note: 3 cycles if skip and followed
by a 2-word instruction.
Q Cycle Activity:
Example: COMF REG, 0, 0
Q1 Q2 Q3 Q4
Before Instruction Decode Read Process No
REG = 13h register ‘f’ Data operation
After Instruction If skip:
REG = 13h
W = ECh Q1 Q2 Q3 Q4
No No No No
If skip and followed by 2-word instruction:
Q1 Q2 Q3 Q4
No No No No
No No No No
Example: HERE CPFSEQ REG, 0

NEQUAL :
EQUAL :
Before Instruction
PC Address = HERE
W = ?
REG = ?
After Instruction
If REG = W;
PC = Address (EQUAL)
If REG  W;
PC = Address (NEQUAL)

PIC18F87K90 FAMILY
CPFSGT Compare f with W, Skip if f > W CPFSLT Compare f with W, Skip if f < W
Syntax: CPFSGT f {,a} Syntax: CPFSLT f {,a}
a  [0,1] a  [0,1]
Operation: (f) –W),
Operation: (f) –W),
skip if (f) > (W) skip if (f) < (W)
(unsigned comparison) (unsigned comparison)
Description: Compares the contents of data memory
location ‘f’ to the contents of the W by Description: Compares the contents of data memory
performing an unsigned subtraction. location ‘f’ to the contents of W by
performing an unsigned subtraction.
If the contents of ‘f’ are greater than the
contents of WREG, then the fetched If the contents of ‘f’ are less than the
instruction is discarded and a NOP is contents of W, then the fetched
executed instead, making this a instruction is discarded and a NOP is
two-cycle instruction. executed instead, making this a
If ‘a’ is ‘1’, the BSR is used to select the If ‘a’ is ‘0’, the Access Bank is selected.
GPR bank. If ‘a’ is ‘1’, the BSR is used to select the
GPR bank.
set is enabled, this instruction operates Words: 1
in Indexed Literal Offset Addressing Cycles: 1(2)
mode whenever f 95 (5Fh). See Note: 3 cycles if skip and followed
Section 29.2.3 “Byte-Oriented and by a 2-word instruction.
Q Cycle Activity:
Q1 Q2 Q3 Q4
Words: 1
Decode Read Process No
Cycles: 1(2) register ‘f’ Data operation
Note: 3 cycles if skip and followed
If skip:
Q1 Q2 Q3 Q4
Q Cycle Activity:
No No No No
Q1 Q2 Q3 Q4
register ‘f’ Data operation If skip and followed by 2-word instruction:
If skip: Q1 Q2 Q3 Q4
Q1 Q2 Q3 Q4 No No No No
No No No No operation operation operation operation
operation operation operation operation No No No No
If skip and followed by 2-word instruction: operation operation operation operation
Q1 Q2 Q3 Q4
No No No No Example: HERE CPFSLT REG, 1
operation operation operation operation NLESS :
No No No No LESS :
operation operation operation operation Before Instruction
PC = Address (HERE)
Example: HERE CPFSGT REG, 0 W = ?
NGREATER : After Instruction
GREATER : If REG < W;
Before Instruction PC = Address (LESS)
If REG  W;
PC = Address (HERE) PC = Address (NLESS)
W = ?
After Instruction
If REG  W;
PC = Address (GREATER)
If REG  W;
PC = Address (NGREATER)

PIC18F87K90 FAMILY
DAW Decimal Adjust W Register DECF Decrement f
Syntax: DAW Syntax: DECF f {,d {,a}}

Operands: None Operands: 0  f  255
Operation: If [W<3:0> > 9] or [DC = 1], then d  [0,1]
a  [0,1]
(W<3:0>) + 6  W<3:0>;
else, Operation: (f) – 1  dest
(W<3:0>)  W<3:0>;
Status Affected: C, DC, N, OV, Z
If [W<7:4> > 9] or [C = 1], then Encoding: 0000 01da ffff ffff
(W<7:4>) + 6  W<7:4>; Description: Decrement register ‘f’. If ‘d’ is ‘0’, the
C =1; result is stored in W. If ‘d’ is ‘1’, the
else, result is stored back in register ‘f’.
(W<7:4>)  W<7:4>
Status Affected: C If ‘a’ is ‘1’, the BSR is used to select the
Encoding: 0000 0000 0000 0111 GPR bank.
Description: DAW adjusts the 8-bit value in W, If ‘a’ is ‘0’ and the extended instruction
resulting from the earlier addition of two set is enabled, this instruction operates
variables (each in packed BCD format) in Indexed Literal Offset Addressing
and produces a correct packed BCD mode whenever f 95 (5Fh). See
result. Section 29.2.3 “Byte-Oriented and
Words: 1
Cycles: 1 Words: 1
Q Cycle Activity: Cycles: 1
Q1 Q2 Q3 Q4
Q Cycle Activity:
Decode Read Process Write
register W Data W Q1 Q2 Q3 Q4
Example 1: DAW
Before Instruction
W = A5h Example: DECF CNT, 1, 0
C = 0 Before Instruction
DC = 0
CNT = 01h
After Instruction Z = 0
W = 05h After Instruction
C = 1
DC = 0 CNT = 00h
Z = 1
Example 2:
Before Instruction
W = CEh
C = 0
DC = 0
After Instruction
W = 34h
C = 1
DC = 0

PIC18F87K90 FAMILY
DECFSZ Decrement f, Skip if 0 DCFSNZ Decrement f, Skip if Not 0
Syntax: DECFSZ f {,d {,a}} Syntax: DCFSNZ f {,d {,a}}

d  [0,1] d  [0,1]
a  [0,1] a  [0,1]
Operation: (f) – 1  dest, Operation: (f) – 1  dest,
skip if result = 0 skip if result  0
Encoding: 0010 11da ffff ffff Encoding: 0100 11da ffff ffff
Description: The contents of register ‘f’ are Description: The contents of register ‘f’ are
decremented. If ‘d’ is ‘0’, the result is decremented. If ‘d’ is ‘0’, the result is
placed in W. If ‘d’ is ‘1’, the result is placed in W. If ‘d’ is ‘1’, the result is
placed back in register ‘f’. placed back in register ‘f’.
If the result is ‘0’, the next instruction If the result is not ‘0’, the next
which is already fetched is discarded instruction which is already fetched is
and a NOP is executed instead, making discarded and a NOP is executed
it a two-cycle instruction. instead, making it a two-cycle
instruction.
If ‘a’ is ‘0’ and the extended instruction GPR bank.
set is enabled, this instruction operates If ‘a’ is ‘0’ and the extended instruction
in Indexed Literal Offset Addressing set is enabled, this instruction operates
mode whenever f 95 (5Fh). See in Indexed Literal Offset Addressing
Section 29.2.3 “Byte-Oriented and mode whenever f 95 (5Fh). See
Bit-Oriented Instructions in Indexed Section 29.2.3 “Byte-Oriented and
Literal Offset Mode” for details. Bit-Oriented Instructions in Indexed
Words: 1
Cycles: 1(2) Words: 1
Note: 3 cycles if skip and followed Cycles: 1(2)
by a 2-word instruction. Note: 3 cycles if skip and followed
Q Cycle Activity:
Decode Read Process Write to Q1 Q2 Q3 Q4
register ‘f’ Data destination Decode Read Process Write to
If skip: register ‘f’ Data destination
Q1 Q2 Q3 Q4 If skip:
No No No No Q1 Q2 Q3 Q4
If skip and followed by 2-word instruction: operation operation operation operation
Q1 Q2 Q3 Q4 If skip and followed by 2-word instruction:
No No No No Q1 Q2 Q3 Q4
No No No No operation operation operation operation
Example: HERE DECFSZ CNT, 1, 1
GOTO LOOP Example: HERE DCFSNZ TEMP, 1, 0
CONTINUE ZERO :
NZERO :
Before Instruction
PC = Address (HERE) Before Instruction
After Instruction TEMP = ?
CNT = CNT – 1 After Instruction
If CNT = 0; TEMP = TEMP – 1,
PC = Address (CONTINUE) If TEMP = 0;
If CNT  0; PC = Address (ZERO)
PC = Address (HERE + 2) If TEMP  0;
PC = Address (NZERO)

PIC18F87K90 FAMILY
GOTO Unconditional Branch INCF Increment f
Syntax: GOTO k Syntax: INCF f {,d {,a}}

Operation: k  PC<20:1> d  [0,1]
a  [0,1]
Operation: (f) + 1  dest
Encoding:
1st word (k<7:0>) 1110 1111 k7kkk kkkk0 Status Affected: C, DC, N, OV, Z
2nd word(k<19:8>) 1111 k19kkk kkkk kkkk8 Encoding: 0010 10da ffff ffff
Description: GOTO allows an unconditional branch Description: The contents of register ‘f’ are
anywhere within entire 2-Mbyte memory incremented. If ‘d’ is ‘0’, the result is
range. The 20-bit value ‘k’ is loaded into placed in W. If ‘d’ is ‘1’, the result is
PC<20:1>. GOTO is always a two-cycle placed back in register ‘f’.
instruction.
Words: 2 If ‘a’ is ‘1’, the BSR is used to select the
GPR bank.
Cycles: 2
Q Cycle Activity: If ‘a’ is ‘0’ and the extended instruction
Q1 Q2 Q3 Q4
Decode Read literal No Read literal mode whenever f 95 (5Fh). See
‘k’<7:0>, operation ‘k’<19:8>, Section 29.2.3 “Byte-Oriented and
Write to PC Bit-Oriented Instructions in Indexed
No No No No Literal Offset Mode” for details.
Words: 1
Cycles: 1
Example: GOTO THERE
Q Cycle Activity:
After Instruction
Q1 Q2 Q3 Q4
PC = Address (THERE)
Example: INCF CNT, 1, 0

Before Instruction
CNT = FFh
Z = 0
C = ?
DC = ?
After Instruction
CNT = 00h
Z = 1
C = 1
DC = 1

PIC18F87K90 FAMILY
INCFSZ Increment f, Skip if 0 INFSNZ Increment f, Skip if Not 0
Syntax: INCFSZ f {,d {,a}} Syntax: INFSNZ f {,d {,a}}

d  [0,1] d  [0,1]
a  [0,1] a  [0,1]
Operation: (f) + 1  dest, Operation: (f) + 1  dest,

skip if result = 0 skip if result  0
Description: The contents of register ‘f’ are
Description: The contents of register ‘f’ are incremented. If ‘d’ is ‘0’, the result is
incremented. If ‘d’ is ‘0’, the result is placed in W. If ‘d’ is ‘1’, the result is
placed in W. If ‘d’ is ‘1’, the result is placed back in register ‘f’.
placed back in register ‘f’.
If the result is not ‘0’, the next
If the result is ‘0’, the next instruction instruction which is already fetched is
which is already fetched is discarded discarded and a NOP is executed
and a NOP is executed instead, making instead, making it a two-cycle
it a two-cycle instruction. instruction.
If ‘a’ is ‘0’, the Access Bank is selected. If ‘a’ is ‘0’, the Access Bank is selected.
If ‘a’ is ‘1’, the BSR is used to select the If ‘a’ is ‘1’, the BSR is used to select the
GPR bank. GPR bank.
If ‘a’ is ‘0’ and the extended instruction If ‘a’ is ‘0’ and the extended instruction
set is enabled, this instruction operates set is enabled, this instruction operates
in Indexed Literal Offset Addressing in Indexed Literal Offset Addressing
mode whenever f 95 (5Fh). See mode whenever f 95 (5Fh). See
Section 29.2.3 “Byte-Oriented and Section 29.2.3 “Byte-Oriented and
Bit-Oriented Instructions in Indexed Bit-Oriented Instructions in Indexed
Literal Offset Mode” for details. Literal Offset Mode” for details.
Words: 1 Words: 1
Note: 3 cycles if skip and followed Note: 3 cycles if skip and followed
by a 2-word instruction. by a 2-word instruction.
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Decode Read Process Write to Decode Read Process Write to
register ‘f’ Data destination register ‘f’ Data destination
If skip: If skip:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
If skip and followed by 2-word instruction: If skip and followed by 2-word instruction:
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Example: HERE INCFSZ CNT, 1, 0 Example: HERE INFSNZ REG, 1, 0

NZERO : ZERO
ZERO : NZERO
PC = Address (HERE) PC = Address (HERE)
CNT = CNT + 1 REG = REG + 1
If CNT = 0; If REG  0;
PC = Address (ZERO) PC = Address (NZERO)
If CNT  0; If REG = 0;
PC = Address (NZERO) PC = Address (ZERO)

PIC18F87K90 FAMILY
IORLW Inclusive OR Literal with W IORWF Inclusive OR W with f
Syntax: IORLW k Syntax: IORWF f {,d {,a}}

Operation: (W) .OR. k  W d  [0,1]
a  [0,1]
Operation: (W) .OR. (f)  dest
Description: The contents of W are ORed with the
8-bit literal ‘k’. The result is placed
in W. Description: Inclusive OR W with register ‘f’. If ‘d’ is
‘0’, the result is placed in W. If ‘d’ is ‘1’,
Words: 1
the result is placed back in register ‘f’.
Cycles: 1
Q Cycle Activity: If ‘a’ is ‘1’, the BSR is used to select the
Q1 Q2 Q3 Q4 GPR bank.
Decode Read Process Write to If ‘a’ is ‘0’ and the extended instruction
literal ‘k’ Data W set is enabled, this instruction operates
Example: IORLW 35h mode whenever f 95 (5Fh). See
Before Instruction Bit-Oriented Instructions in Indexed
W = 9Ah Literal Offset Mode” for details.
After Instruction
W = BFh Words: 1
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4
Example: IORWF RESULT, 0, 1

Before Instruction
RESULT = 13h
W = 91h
After Instruction
RESULT = 13h
W = 93h

PIC18F87K90 FAMILY
LFSR Load FSR MOVF Move f
Syntax: LFSR f, k Syntax: MOVF f {,d {,a}}

Operands: 0f2 Operands: 0  f  255
0  k  4095 d  [0,1]
a  [0,1]
Operation: k  FSRf
Status Affected: None Operation: f  dest
Encoding: 1110 1110 00ff k11kkk Status Affected: N, Z

1111 0000 k7kkk kkkk Encoding: 0101 00da ffff ffff
Description: The 12-bit literal ‘k’ is loaded into the Description: The contents of register ‘f’ are moved to
file select register pointed to by ‘f’. a destination dependent upon the
Words: 2 status of ‘d’. If ‘d’ is ‘0’, the result is
placed in W. If ‘d’ is ‘1’, the result is
Cycles: 2 placed back in register ‘f’. Location ‘f’
Q Cycle Activity: can be anywhere in the
Q1 Q2 Q3 Q4 256-byte bank.
Decode Read literal Process Write If ‘a’ is ‘0’, the Access Bank is selected.
‘k’ MSB Data literal ‘k’ If ‘a’ is ‘1’, the BSR is used to select the
MSB to GPR bank.
FSRfH If ‘a’ is ‘0’ and the extended instruction
Decode Read literal Process Write literal set is enabled, this instruction operates
‘k’ LSB Data ‘k’ to FSRfL in Indexed Literal Offset Addressing
Example: LFSR 2, 3ABh
After Instruction Literal Offset Mode” for details.
FSR2H = 03h
FSR2L = ABh Words: 1
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4
register ‘f’ Data W
Example: MOVF REG, 0, 0

Before Instruction
REG = 22h
W = FFh
After Instruction
REG = 22h
W = 22h

PIC18F87K90 FAMILY
MOVFF Move f to f MOVLB Move Literal to Low Nibble in BSR
Syntax: MOVFF fs,fd Syntax: MOVLB k

Operands: 0  fs  4095 Operands: 0  k  255
0  fd  4095 Operation: k  BSR
Operation: (fs)  fd Status Affected: None
Encoding: Description: The 8-bit literal ‘k’ is loaded into the
1st word (source) 1100 ffff ffff ffffs
Bank Select Register (BSR). The value
2nd word (destin.) 1111 ffff ffff ffffd
of BSR<7:4> always remains ‘0’
Description: The contents of source register ‘fs’ are regardless of the value of k7:k4.
moved to destination register ‘fd’. Words: 1
Location of source ‘fs’ can be anywhere
in the 4096-byte data space (000h to Cycles: 1
FFFh) and location of destination ‘fd’ Q Cycle Activity:
can also be anywhere from 000h to Q1 Q2 Q3 Q4
FFFh.
Decode Read Process Write literal
Either source or destination can be W literal ‘k’ Data ‘k’ to BSR
(a useful special situation).
MOVFF is particularly useful for Example: MOVLB 5
transferring a data memory location to a
peripheral register (such as the transmit Before Instruction
buffer or an I/O port). BSR Register = 02h
After Instruction
The MOVFF instruction cannot use the BSR Register = 05h
PCL, TOSU, TOSH or TOSL as the
destination register
Words: 2
Cycles: 2
Q Cycle Activity:
Q1 Q2 Q3 Q4
register ‘f’ Data operation
(src)
Decode No No Write
operation operation register ‘f’
No dummy (dest)
read
Example: MOVFF REG1, REG2

Before Instruction
REG1 = 33h
REG2 = 11h
After Instruction
REG1 = 33h
REG2 = 33h

PIC18F87K90 FAMILY
MOVLW Move Literal to W MOVWF Move W to f
Syntax: MOVLW k Syntax: MOVWF f {,a}

Operation: kW a  [0,1]
Status Affected: None Operation: (W)  f

Description: The 8-bit literal ‘k’ is loaded into W. Encoding: 0110 111a ffff ffff
Words: 1 Description: Move data from W to register ‘f’.

Location ‘f’ can be anywhere in the
Cycles: 1 256-byte bank.
Q Cycle Activity: If ‘a’ is ‘0’, the Access Bank is selected.
Q1 Q2 Q3 Q4 If ‘a’ is ‘1’, the BSR is used to select the
Decode Read Process Write to GPR bank.
literal ‘k’ Data W If ‘a’ is ‘0’ and the extended instruction
Example: MOVLW 5Ah in Indexed Literal Offset Addressing
After Instruction Section 29.2.3 “Byte-Oriented and
W = 5Ah Bit-Oriented Instructions in Indexed
Words: 1
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4
register ‘f’ Data register ‘f’
Example: MOVWF REG, 0

Before Instruction
W = 4Fh
REG = FFh
After Instruction
W = 4Fh
REG = 4Fh

PIC18F87K90 FAMILY
MULLW Multiply Literal with W MULWF Multiply W with f
Syntax: MULLW k Syntax: MULWF f {,a}

Operation: (W) x k  PRODH:PRODL a  [0,1]
Status Affected: None Operation: (W) x (f)  PRODH:PRODL

Description: An unsigned multiplication is carried Encoding: 0000 001a ffff ffff
out between the contents of W and the Description: An unsigned multiplication is carried out
8-bit literal ‘k’. The 16-bit result is between the contents of W and the
placed in the PRODH:PRODL register register file location ‘f’. The 16-bit result is
pair. PRODH contains the high byte. stored in the PRODH:PRODL register
W is unchanged. pair. PRODH contains the high byte. Both
W and ‘f’ are unchanged.
None of the Status flags are affected.
None of the Status flags are affected.
Note that neither Overflow nor Carry is
possible in this operation. A Zero result Note that neither Overflow nor Carry is
is possible but not detected. possible in this operation. A Zero result is
possible but not detected.
Words: 1
If ‘a’ is ‘0’, the Access Bank is selected. If
Cycles: 1 ‘a’ is ‘1’, the BSR is used to select the
Q Cycle Activity: GPR bank.
Q1 Q2 Q3 Q4 If ‘a’ is ‘0’ and the extended instruction set
Decode Read Process Write is enabled, this instruction operates in
literal ‘k’ Data registers Indexed Literal Offset Addressing mode
PRODH: whenever f 95 (5Fh). See
PRODL Section 29.2.3 “Byte-Oriented and
Example: MULLW 0C4h
Words: 1
Before Instruction
W = E2h Cycles: 1
PRODH = ? Q Cycle Activity:
PRODL = ?
After Instruction Q1 Q2 Q3 Q4
W = E2h Decode Read Process Write
PRODH = ADh register ‘f’ Data registers
PRODL = 08h PRODH:
PRODL
Example: MULWF REG, 1

Before Instruction
W = C4h
REG = B5h
PRODH = ?
PRODL = ?
After Instruction
W = C4h
REG = B5h
PRODH = 8Ah
PRODL = 94h

PIC18F87K90 FAMILY
NEGF Negate f NOP No Operation
Syntax: NEGF f {,a} Syntax: NOP

Operands: 0  f  255 Operands: None
a  [0,1] Operation: No operation
Operation: (f) + 1  f Status Affected: None
Encoding: 0000 0000 0000 0000
Encoding: 0110 110a ffff ffff 1111 xxxx xxxx xxxx
Description: Location ‘f’ is negated using two’s Description: No operation.
complement. The result is placed in the Words: 1
data memory location ‘f’.
Cycles: 1
If ‘a’ is ‘1’, the BSR is used to select the Q Cycle Activity:
GPR bank. Q1 Q2 Q3 Q4
If ‘a’ is ‘0’ and the extended instruction Decode No No No
set is enabled, this instruction operates operation operation operation
mode whenever f 95 (5Fh). See Example:
Bit-Oriented Instructions in Indexed None.
Words: 1
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4
Example: NEGF REG, 1

Before Instruction
REG = 0011 1010 [3Ah]
After Instruction
REG = 1100 0110 [C6h]

PIC18F87K90 FAMILY
POP Pop Top of Return Stack PUSH Push Top of Return Stack
Syntax: POP Syntax: PUSH

Operands: None Operands: None
Operation: (TOS)  bit bucket Operation: (PC + 2)  TOS
Encoding: 0000 0000 0000 0110 Encoding: 0000 0000 0000 0101
Description: The TOS value is pulled off the return Description: The PC + 2 is pushed onto the top of
stack and is discarded. The TOS value the return stack. The previous TOS
then becomes the previous value that value is pushed down on the stack.
was pushed onto the return stack. This instruction allows implementing a
This instruction is provided to enable software stack by modifying TOS and
the user to properly manage the return then pushing it onto the return stack.
stack to incorporate a software stack.
Words: 1
Words: 1 Cycles: 1
Cycles: 1 Q Cycle Activity:
Q Cycle Activity: Q1 Q2 Q3 Q4
Q1 Q2 Q3 Q4 Decode PUSH No No
Decode No POP TOS No PC + 2 onto operation operation
operation value operation return stack
Example: POP Example: PUSH

GOTO NEW
Before Instruction
Before Instruction TOS = 345Ah
TOS = 0031A2h PC = 0124h
Stack (1 level down) = 014332h
After Instruction
After Instruction PC = 0126h
TOS = 014332h TOS = 0126h
PC = NEW Stack (1 level down) = 345Ah

PIC18F87K90 FAMILY
RCALL Relative Call RESET Reset
Syntax: RCALL n Syntax: RESET

Operands: -1024  n  1023 Operands: None
Operation: (PC) + 2  TOS, Operation: Reset all registers and flags that are
(PC) + 2 + 2n  PC affected by a MCLR Reset.
Status Affected: None Status Affected: All
Encoding: 1101 1nnn nnnn nnnn Encoding: 0000 0000 1111 1111
Description: Subroutine call with a jump up to 1K Description: This instruction provides a way to
from the current location. First, return execute a MCLR Reset in software.
address (PC + 2) is pushed onto the
Words: 1
stack. Then, add the 2’s complement
number ‘2n’ to the PC. Since the PC will Cycles: 1
have incremented to fetch the next Q Cycle Activity:
instruction, the new address will be Q1 Q2 Q3 Q4
PC + 2 + 2n. This instruction is a
two-cycle instruction. Decode Start No No
reset operation operation
Words: 1
Cycles: 2 Example: RESET
Q Cycle Activity: After Instruction
Q1 Q2 Q3 Q4 Registers = Reset Value
Decode Read literal Process Write to PC Flags* = Reset Value
‘n’ Data
PUSH PC
to stack
No No No No
Example: HERE RCALL Jump

Before Instruction
PC = Address (HERE)
After Instruction
PC = Address (Jump)
TOS = Address (HERE + 2)

PIC18F87K90 FAMILY
RETFIE Return from Interrupt RETLW Return Literal to W
Syntax: RETFIE {s} Syntax: RETLW k

Operands: s  [0,1] Operands: 0  k  255
Operation: (TOS)  PC, Operation: k  W,
1  GIE/GIEH or PEIE/GIEL; (TOS)  PC,
if s = 1, PCLATU, PCLATH are unchanged
(WS)  W,
(STATUSS)  STATUS,
(BSRS)  BSR, Encoding: 0000 1100 kkkk kkkk
PCLATU, PCLATH are unchanged Description: W is loaded with the 8-bit literal ‘k’. The
Status Affected: GIE/GIEH, PEIE/GIEL. Program Counter is loaded from the top
of the stack (the return address). The
Encoding: 0000 0000 0001 000s high address latch (PCLATH) remains
Description: Return from interrupt. Stack is popped unchanged.
and Top-of-Stack (TOS) is loaded into Words: 1
the PC. Interrupts are enabled by
setting either the high or low-priority Cycles: 2
Global Interrupt Enable bit. If ‘s’ = 1, the Q Cycle Activity:
contents of the shadow registers WS, Q1 Q2 Q3 Q4
STATUSS and BSRS are loaded into
Decode Read Process POP PC
their corresponding registers W,
STATUS and BSR. If ‘s’ = 0, no update literal ‘k’ Data from stack,
of these registers occurs. write to W
No No No No
Words: 1 operation operation operation operation
Cycles: 2
Q Cycle Activity: Example:
Q1 Q2 Q3 Q4
Decode No No POP PC CALL TABLE ; W contains table
operation operation from stack ; offset value
; W now has
Set GIEH or
; table value
GIEL
:
No No No No TABLE
operation operation operation operation ADDWF PCL ; W = offset
RETLW k0 ; Begin table
Example: RETFIE 1 RETLW k1 ;
:
After Interrupt
:
PC = TOS RETLW kn ; End of table
W = WS
BSR = BSRS
STATUS = STATUSS Before Instruction
GIE/GIEH, PEIE/GIEL = 1
W = 07h
After Instruction
W = value of kn

PIC18F87K90 FAMILY
RETURN Return from Subroutine RLCF Rotate Left f through Carry
Syntax: RETURN {s} Syntax: RLCF f {,d {,a}}

Operands: s  [0,1] Operands: 0  f  255
Operation: (TOS)  PC; d  [0,1]
a  [0,1]
if s = 1,
(WS)  W, Operation: (f<n>)  dest<n + 1>,
(STATUSS)  STATUS, (f<7>)  C,
(BSRS)  BSR, (C)  dest<0>
PCLATU, PCLATH are unchanged Status Affected: C, N, Z
Status Affected: None Encoding: 0011 01da ffff ffff
Encoding: 0000 0000 0001 001s
Description: The contents of register ‘f’ are rotated
Description: Return from subroutine. The stack is one bit to the left through the Carry flag.
popped and the top of the stack (TOS) If ‘d’ is ‘0’, the result is placed in W. If ‘d’
is loaded into the Program Counter. If is ‘1’, the result is stored back in register
‘s’= 1, the contents of the shadow ‘f’.
registers WS, STATUSS and BSRS are
loaded into their corresponding If ‘a’ is ‘1’, the BSR is used to select the
registers W, STATUS and BSR. If GPR bank.
‘s’ = 0, no update of these registers
occurs. If ‘a’ is ‘0’ and the extended instruction
Words: 1
Cycles: 2 mode whenever f 95 (5Fh). See
Q Cycle Activity:
Q1 Q2 Q3 Q4 Literal Offset Mode” for details.
Decode No Process POP PC
operation Data from stack C register f
No No No No
operation operation operation operation Words: 1
Cycles: 1
Q Cycle Activity:
Example: RETURN Q1 Q2 Q3 Q4
After Instruction: Decode Read Process Write to
PC = TOS register ‘f’ Data destination
Example: RLCF REG, 0, 0

Before Instruction
REG = 1110 0110
C = 0
After Instruction
REG = 1110 0110
W = 1100 1100
C = 1

PIC18F87K90 FAMILY
RLNCF Rotate Left f (No Carry) RRCF Rotate Right f through Carry
Syntax: RLNCF f {,d {,a}} Syntax: RRCF f {,d {,a}}

d  [0,1] d  [0,1]
a  [0,1] a  [0,1]
Operation: (f<n>)  dest<n + 1>, Operation: (f<n>)  dest<n – 1>,
(f<7>)  dest<0> (f<0>)  C,
Status Affected: N, Z (C)  dest<7>
Status Affected: C, N, Z
Description: The contents of register ‘f’ are rotated Encoding: 0011 00da ffff ffff
one bit to the left. If ‘d’ is ‘0’, the result Description: The contents of register ‘f’ are rotated
is placed in W. If ‘d’ is ‘1’, the result is one bit to the right through the Carry
stored back in register ‘f’. flag. If ‘d’ is ‘0’, the result is placed in W.
If ‘a’ is ‘0’, the Access Bank is selected. If ‘d’ is ‘1’, the result is placed back in
If ‘a’ is ‘1’, the BSR is used to select the register ‘f’.
GPR bank. If ‘a’ is ‘0’, the Access Bank is selected.
If ‘a’ is ‘0’ and the extended instruction If ‘a’ is ‘1’, the BSR is used to select the
GPR bank.
in Indexed Literal Offset Addressing If ‘a’ is ‘0’ and the extended instruction
mode whenever f 95 (5Fh). See set is enabled, this instruction operates
Section 29.2.3 “Byte-Oriented and in Indexed Literal Offset Addressing
Bit-Oriented Instructions in Indexed mode whenever f 95 (5Fh). See
Literal Offset Mode” for details. Section 29.2.3 “Byte-Oriented and
register f
Words: 1 C register f
Cycles: 1
Words: 1
Q Cycle Activity:
Cycles: 1
Q1 Q2 Q3 Q4
Decode Read Process Write to Q Cycle Activity:
register ‘f’ Data destination Q1 Q2 Q3 Q4
Example: RLNCF REG, 1, 0 register ‘f’ Data destination
Before Instruction
REG = 1010 1011 Example: RRCF REG, 0, 0
After Instruction Before Instruction
REG = 0101 0111 REG = 1110 0110
C = 0
After Instruction
REG = 1110 0110
W = 0111 0011
C = 0

PIC18F87K90 FAMILY
RRNCF Rotate Right f (No Carry) SETF Set f
Syntax: RRNCF f {,d {,a}} Syntax: SETF f {,a}

d  [0,1] a [0,1]
a  [0,1]
Operation: FFh  f
Operation: (f<n>)  dest<n – 1>, Status Affected: None
(f<0>)  dest<7>
Description: The contents of the specified register
Encoding: 0100 00da ffff ffff are set to FFh.
Description: The contents of register ‘f’ are rotated
one bit to the right. If ‘d’ is ‘0’, the result If ‘a’ is ‘1’, the BSR is used to select the
is placed in W. If ‘d’ is ‘1’, the result is GPR bank.
placed back in register ‘f’.
If ‘a’ is ‘0’, the Access Bank will be set is enabled, this instruction operates
selected, overriding the BSR value. If ‘a’
is ‘1’, then the bank will be selected as mode whenever f 95 (5Fh). See
per the BSR value. Section 29.2.3 “Byte-Oriented and
If ‘a’ is ‘0’ and the extended instruction Bit-Oriented Instructions in Indexed
set is enabled, this instruction operates Literal Offset Mode” for details.
Words: 1
Section 29.2.3 “Byte-Oriented and Cycles: 1
Bit-Oriented Instructions in Indexed Q Cycle Activity:
Literal Offset Mode” for details. Q1 Q2 Q3 Q4
register f Decode Read Process Write
Words: 1
Cycles: 1 Example: SETF REG,1
Q Cycle Activity: Before Instruction
Q1 Q2 Q3 Q4 REG = 5Ah
Decode Read Process Write to After Instruction
REG = FFh
Example 1: RRNCF REG, 1, 0

Before Instruction
REG = 1101 0111
After Instruction
REG = 1110 1011
Example 2: RRNCF REG, 0, 0

Before Instruction
W = ?
REG = 1101 0111
After Instruction
W = 1110 1011
REG = 1101 0111

PIC18F87K90 FAMILY
SLEEP Enter Sleep Mode SUBFWB Subtract f from W with Borrow
Syntax: SLEEP Syntax: SUBFWB f {,d {,a}}

Operands: None Operands: 0 f 255
Operation: 00h  WDT, d  [0,1]
a  [0,1]
0  WDT postscaler,
1  TO, Operation: (W) – (f) – (C) dest
0  PD
Status Affected: TO, PD Encoding: 0101 01da ffff ffff
Encoding: 0000 0000 0000 0011
Description: Subtract register ‘f’ and Carry flag
Description: The Power-Down status bit (PD) is (borrow) from W (2’s complement
cleared. The Time-out status bit (TO) method). If ‘d’ is ‘0’, the result is stored in
is set. The Watchdog Timer and its W. If ‘d’ is ‘1’, the result is stored in
postscaler are cleared. register ‘f’.
The processor is put into Sleep mode If ‘a’ is ‘0’, the Access Bank is selected. If
with the oscillator stopped. ‘a’ is ‘1’, the BSR is used to select the
Words: 1 GPR bank.
Cycles: 1
set is enabled, this instruction operates in
Q Cycle Activity: Indexed Literal Offset Addressing mode
Q1 Q2 Q3 Q4 whenever f 95 (5Fh). See
Decode No Process Go to Section 29.2.3 “Byte-Oriented and
operation Data Sleep Bit-Oriented Instructions in Indexed
Example: SLEEP Words: 1
Before Instruction Cycles: 1

TO = ? Q Cycle Activity:
PD = ? Q1 Q2 Q3 Q4
After Instruction Decode Read Process Write to
TO = 1† register ‘f’ Data destination
PD = 0
Example 1: SUBFWB REG, 1, 0
† If WDT causes wake-up, this bit is cleared. Before Instruction
REG = 3
W = 2
C = 1
After Instruction
REG = FF
W = 2
C = 0
Z = 0
N = 1 ; result is negative
Before Instruction
REG = 2
W = 5
C = 1
After Instruction
REG = 2
W = 3
C = 1
Z = 0
N = 0 ; result is positive
Before Instruction
REG = 1
W = 2
C = 0
After Instruction
REG = 0
W = 2
C = 1
Z = 1 ; result is zero
N = 0

PIC18F87K90 FAMILY
SUBLW Subtract W from Literal SUBWF Subtract W from f
Syntax: SUBLW k Syntax: SUBWF f {,d {,a}}

Operands: 0 k 255 Operands: 0 f 255
Operation: k – (W) W d  [0,1]
a  [0,1]
Operation: (f) – (W) dest
Description: W is subtracted from the 8-bit
literal ‘k’. The result is placed in W.
Words: 1 Description: Subtract W from register ‘f’ (2’s
complement method). If ‘d’ is ‘0’, the
Cycles: 1 result is stored in W. If ‘d’ is ‘1’, the result
Q Cycle Activity: is stored back in register ‘f’.
Q1 Q2 Q3 Q4 If ‘a’ is ‘0’, the Access Bank is selected.
Decode Read Process Write to If ‘a’ is ‘1’, the BSR is used to select the
literal ‘k’ Data W GPR bank.
Example 1: SUBLW 02h
Before Instruction in Indexed Literal Offset Addressing
W = 01h mode whenever f 95 (5Fh). See
C = ? Section 29.2.3 “Byte-Oriented and
After Instruction Bit-Oriented Instructions in Indexed
W = 01h Literal Offset Mode” for details.
C = 1 ; result is positive
Z = 0 Words: 1
N = 0
Cycles: 1
Example 2: SUBLW 02h
Q Cycle Activity:
Before Instruction
Q1 Q2 Q3 Q4
W = 02h
C = ? Decode Read Process Write to
After Instruction register ‘f’ Data destination
W = 00h
C = 1 ; result is zero Example 1: SUBWF REG, 1, 0
Z = 1 Before Instruction
N = 0 REG = 3
Example 3: SUBLW 02h W = 2
C = ?
Before Instruction After Instruction
W = 03h REG = 1
C = ? W = 2
After Instruction C = 1 ; result is positive
W = FFh ; (2’s complement) Z = 0
C = 0 ; result is negative N = 0
Z = 0 Example 2: SUBWF REG, 0, 0
N = 1
Before Instruction
REG = 2
W = 2
C = ?
After Instruction
REG = 2
W = 0
C = 1 ; result is zero
Z = 1
N = 0
Example 3: SUBWF REG, 1, 0
Before Instruction
REG = 1
W = 2
C = ?
After Instruction
REG = FFh ;(2’s complement)
W = 2
C = 0 ; result is negative
Z = 0
N = 1

PIC18F87K90 FAMILY
SUBWFB Subtract W from f with Borrow SWAPF Swap f

Syntax: SUBWFB f {,d {,a}} Syntax: SWAPF f {,d {,a}}
d  [0,1] d  [0,1]
a  [0,1] a  [0,1]
Operation: (f) – (W) – (C) dest Operation: (f<3:0>)  dest<7:4>,
Status Affected: N, OV, C, DC, Z (f<7:4>)  dest<3:0>
Encoding: 0101 10da ffff ffff Status Affected: None
Description: Subtract W and the Carry flag (borrow) Encoding: 0011 10da ffff ffff
from register ‘f’ (2’s complement
method). If ‘d’ is ‘0’, the result is stored Description: The upper and lower nibbles of register
in W. If ‘d’ is ‘1’, the result is stored back ‘f’ are exchanged. If ‘d’ is ‘0’, the result
in register ‘f’. is placed in W. If ‘d’ is ‘1’, the result is
placed in register ‘f’.
GPR bank.
set is enabled, this instruction operates If ‘a’ is ‘0’ and the extended instruction
in Indexed Literal Offset Addressing set is enabled, this instruction operates
mode whenever f 95 (5Fh). See in Indexed Literal Offset Addressing
Section 29.2.3 “Byte-Oriented and mode whenever f 95 (5Fh). See
Bit-Oriented Instructions in Indexed Section 29.2.3 “Byte-Oriented and
Literal Offset Mode” for details. Bit-Oriented Instructions in Indexed
Words: 1
Words: 1
Cycles: 1
Q Cycle Activity: Cycles: 1
Decode Read Process Write to Q1 Q2 Q3 Q4
register ‘f’ Data destination Decode Read Process Write to
Example 1: SUBWFB REG, 1, 0 register ‘f’ Data destination
Before Instruction
REG = 19h (0001 1001) Example: SWAPF REG, 1, 0
W = 0Dh (0000 1101)
C = 1 Before Instruction
After Instruction REG = 53h
REG = 0Ch (0000 1011) After Instruction
W = 0Dh (0000 1101) REG = 35h
C = 1
Z = 0
N = 0 ; result is positive
Example 2: SUBWFB REG, 0, 0
Before Instruction
REG = 1Bh (0001 1011)
W = 1Ah (0001 1010)
C = 0
After Instruction
REG = 1Bh (0001 1011)
W = 00h
C = 1
Z = 1 ; result is zero
N = 0
Example 3: SUBWFB REG, 1, 0
Before Instruction
REG = 03h (0000 0011)
W = 0Eh (0000 1101)
C = 1
After Instruction
REG = F5h (1111 0100)
; [2’s comp]
W = 0Eh (0000 1101)
C = 0
Z = 0
N = 1 ; result is negative

PIC18F87K90 FAMILY
TBLRD Table Read TBLRD Table Read (Continued)
Syntax: TBLRD ( *; *+; *-; +*) Example 1: TBLRD *+ ;

Operands: None Before Instruction
Operation: if TBLRD *, TABLAT = 55h
TBLPTR = 00A356h
(Prog Mem (TBLPTR))  TABLAT, MEMORY(00A356h) = 34h
TBLPTR – No Change; After Instruction
if TBLRD *+, TABLAT = 34h
(Prog Mem (TBLPTR))  TABLAT, TBLPTR = 00A357h
(TBLPTR) + 1  TBLPTR; Example 2: TBLRD +* ;
if TBLRD *-,
(Prog Mem (TBLPTR))  TABLAT, Before Instruction
(TBLPTR) – 1  TBLPTR; TABLAT = AAh
if TBLRD +*, TBLPTR = 01A357h
MEMORY(01A357h) = 12h
(TBLPTR) + 1  TBLPTR, MEMORY(01A358h) = 34h
(Prog Mem (TBLPTR))  TABLAT After Instruction
Status Affected: None TABLAT = 34h
TBLPTR = 01A358h
Encoding: 0000 0000 0000 10nn
nn=0 *
=1 *+
=2 *-
=3 +*
Description: This instruction is used to read the contents
of Program Memory (P.M.). To address the
program memory, a pointer called Table
Pointer (TBLPTR) is used.
The TBLPTR (a 21-bit pointer) points to
each byte in the program memory. TBLPTR
has a 2-Mbyte address range.
TBLPTR<0> = 0:Least Significant Byte of
Program Memory Word
TBLPTR<0> = 1:Most Significant Byte of
Program Memory Word
The TBLRD instruction can modify the value
of TBLPTR as follows:
• no change
• post-increment
• post-decrement
• pre-increment
Words: 1
Cycles: 2
Q Cycle Activity:
Q1 Q2 Q3 Q4
Decode No No No
operation operation operation
No No operation No No operation
operation (Read Program operation (Write
Memory) TABLAT)

PIC18F87K90 FAMILY
TBLWT Table Write TBLWT Table Write (Continued)

Syntax: TBLWT ( *; *+; *-; +*) Example 1: TBLWT *+;
Operands: None Before Instruction
Operation: if TBLWT*, TABLAT = 55h
(TABLAT)  Holding Register, TBLPTR = 00A356h
HOLDING REGISTER
TBLPTR – No Change; (00A356h) = FFh
if TBLWT*+, After Instructions (table write completion)
(TABLAT)  Holding Register,
TABLAT = 55h
(TBLPTR) + 1  TBLPTR; TBLPTR = 00A357h
if TBLWT*-, HOLDING REGISTER
(TABLAT)  Holding Register, (00A356h) = 55h
(TBLPTR) – 1  TBLPTR; Example 2: TBLWT +*;
if TBLWT+*,
(TBLPTR) + 1  TBLPTR, Before Instruction
TABLAT = 34h
(TABLAT)  Holding Register TBLPTR = 01389Ah
Status Affected: None HOLDING REGISTER
(01389Ah) = FFh
Encoding: 0000 0000 0000 11nn HOLDING REGISTER
nn=0 * (01389Bh) = FFh
=1 *+ After Instruction (table write completion)
=2 *- TABLAT = 34h
=3 +* TBLPTR = 01389Bh
HOLDING REGISTER
Description: This instruction uses the 3 LSBs of (01389Ah) = FFh
TBLPTR to determine which of the HOLDING REGISTER
8 holding registers the TABLAT is written (01389Bh) = 34h
to. The holding registers are used to
program the contents of Program Memory
(P.M.). (Refer to Section 6.0 “Memory
Organization” for additional details on
programming Flash memory.)
The TBLPTR (a 21-bit pointer) points to
each byte in the program memory.
TBLPTR has a 2-Mbyte address range.
The LSb of the TBLPTR selects which
byte of the program memory location to
access.
TBLPTR<0> = 0:Least Significant Byte
of Program Memory
Word
TBLPTR<0> = 1:Most Significant Byte
of Program Memory
Word
The TBLWT instruction can modify the
value of TBLPTR as follows:
• no change
• post-increment
• post-decrement
• pre-increment
Words: 1
Cycles: 2
Q Cycle Activity:
Q1 Q2 Q3 Q4
Decode No No No
operation operation operation
No No No No
(Read (Write to
TABLAT) Holding
Register)

PIC18F87K90 FAMILY
TSTFSZ Test f, Skip if 0 XORLW Exclusive OR Literal with W
Syntax: TSTFSZ f {,a} Syntax: XORLW k

Operands: 0  f  255 Operands: 0 k 255
a  [0,1] Operation: (W) .XOR. k W
Operation: skip if f = 0 Status Affected: N, Z
Encoding: 0110 011a ffff ffff Description: The contents of W are XORed with
Description: If ‘f’ = 0, the next instruction fetched the 8-bit literal ‘k’. The result is placed
during the current instruction execution in W.
is discarded and a NOP is executed,
Words: 1
making this a two-cycle instruction.
Cycles: 1
If ‘a’ is ‘1’, the BSR is used to select the Q Cycle Activity:
GPR bank. Q1 Q2 Q3 Q4
If ‘a’ is ‘0’ and the extended instruction Decode Read Process Write to
set is enabled, this instruction operates literal ‘k’ Data W
mode whenever f 95 (5Fh). See Example: XORLW 0AFh
Bit-Oriented Instructions in Indexed Before Instruction
Literal Offset Mode” for details. W = B5h
After Instruction
Words: 1 W = 1Ah
Cycles: 1(2)
Note: 3 cycles if skip and followed
Q Cycle Activity:
Q1 Q2 Q3 Q4
register ‘f’ Data operation
If skip:
Q1 Q2 Q3 Q4
No No No No
If skip and followed by 2-word instruction:
Q1 Q2 Q3 Q4
No No No No
No No No No
Example: HERE TSTFSZ CNT, 1

NZERO :
ZERO :
Before Instruction
PC = Address (HERE)
After Instruction
If CNT = 00h,
PC = Address (ZERO)
If CNT  00h,
PC = Address (NZERO)

PIC18F87K90 FAMILY
XORWF Exclusive OR W with f
Syntax: XORWF f {,d {,a}}

Operands: 0  f  255
d  [0,1]
a  [0,1]
Operation: (W) .XOR. (f) dest
Description: Exclusive OR the contents of W with
register ‘f’. If ‘d’ is ‘0’, the result is stored
in W. If ‘d’ is ‘1’, the result is stored back
in the register ‘f’.
If ‘a’ is ‘1’, the BSR is used to select the
GPR bank.
Words: 1
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4
Example: XORWF REG, 1, 0

Before Instruction
REG = AFh
W = B5h
After Instruction
REG = 1Ah
W = B5h

PIC18F87K90 FAMILY
29.2 Extended Instruction Set A summary of the instructions in the extended instruc-
tion set is provided in Table 29-3. Detailed descriptions
In addition to the standard 75 instructions of the PIC18 are provided in Section 29.2.2 “Extended Instruction
instruction set, the PIC18F87K90 family of devices also Set”. The opcode field descriptions in Table 29-1
provides an optional extension to the core CPU func- (page 452) apply to both the standard and extended
tionality. The added features include eight additional PIC18 instruction sets.
instructions that augment Indirect and Indexed
Addressing operations and the implementation of Note: The instruction set extension and the
Indexed Literal Offset Addressing for many of the Indexed Literal Offset Addressing mode
standard PIC18 instructions. were designed for optimizing applications
written in C; the user may likely never use
The additional features of the extended instruction set
these instructions directly in assembler.
are enabled by default on unprogrammed devices.
The syntax for these commands is
Users must properly set or clear the XINST Configura-
provided as a reference for users who
tion bit during programming to enable or disable these
may be reviewing code that has been
features.
generated by a compiler.
The instructions in the extended set can all be
classified as literal operations, which either manipulate 29.2.1 EXTENDED INSTRUCTION SYNTAX
the File Select Registers, or use them for Indexed
Most of the extended instructions use indexed argu-
Addressing. Two of the instructions, ADDFSR and
ments, using one of the File Select Registers and some
SUBFSR, each have an additional special instantiation
offset to specify a source or destination register. When
for using FSR2. These versions (ADDULNK and
an argument for an instruction serves as part of
SUBULNK) allow for automatic return after execution.
Indexed Addressing, it is enclosed in square brackets
The extended instructions are specifically implemented (“[ ]”). This is done to indicate that the argument is used
to optimize re-entrant program code (that is, code that as an index or offset. The MPASM™ Assembler will
is recursive or that uses a software stack) written in flag an error if it determines that an index or offset value
high-level languages, particularly C. Among other is not bracketed.
things, they allow users working in high-level
When the extended instruction set is enabled, brackets
languages to perform certain operations on data
are also used to indicate index arguments in
structures more efficiently. These include:
byte-oriented and bit-oriented instructions. This is in
• Dynamic allocation and deallocation of software addition to other changes in their syntax. For more
stack space when entering and leaving details, see Section 29.2.3.1 “Extended Instruction
subroutines Syntax with Standard PIC18 Commands”.
• Function Pointer invocation
Note: In the past, square brackets have been
• Software Stack Pointer manipulation used to denote optional arguments in the
• Manipulation of variables located in a software PIC18 and earlier instruction sets. In this
stack text and going forward, optional
arguments are denoted by braces (“{ }”).
TABLE 29-3: EXTENSIONS TO THE PIC18 INSTRUCTION SET

Description Cycles
ADDFSR f, k Add Literal to FSR 1 1110 1000 ffkk kkkk None

ADDULNK k Add Literal to FSR2 and Return 2 1110 1000 11kk kkkk None
CALLW Call Subroutine using WREG 2 0000 0000 0001 0100 None
MOVSF zs, fd Move zs (source) to 1st word 2 1110 1011 0zzz zzzz None
fd (destination) 2nd word 1111 ffff ffff ffff
MOVSS zs, zd Move zs (source) to 1st word 2 1110 1011 1zzz zzzz None
zd (destination) 2nd word 1111 xxxx xzzz zzzz
PUSHL k Store Literal at FSR2, 1 1110 1010 kkkk kkkk None
Decrement FSR2
SUBFSR f, k Subtract Literal from FSR 1 1110 1001 ffkk kkkk None
SUBULNK k Subtract Literal from FSR2 and 2 1110 1001 11kk kkkk None
Return

PIC18F87K90 FAMILY
29.2.2 EXTENDED INSTRUCTION SET
ADDFSR Add Literal to FSR ADDULNK Add Literal to FSR2 and Return
Syntax: ADDFSR f, k Syntax: ADDULNK k
Operands: 0  k  63 Operands: 0  k  63
f  [ 0, 1, 2 ] Operation: FSR2 + k  FSR2,
Operation: FSR(f) + k  FSR(f) (TOS) PC
Encoding: 1110 1000 ffkk kkkk Encoding: 1110 1000 11kk kkkk
Description: The 6-bit literal ‘k’ is added to the Description: The 6-bit literal ‘k’ is added to the
contents of the FSR specified by ‘f’. contents of FSR2. A RETURN is then
Words: 1 executed by loading the PC with the
TOS.
Cycles: 1
Q Cycle Activity: The instruction takes two cycles to
execute; a NOP is performed during
Q1 Q2 Q3 Q4
the second cycle.
literal ‘k’ Data FSR This may be thought of as a special
case of the ADDFSR instruction,
where f = 3 (binary ‘11’); it operates
Example: ADDFSR 2, 23h only on FSR2.
Words: 1
Before Instruction
FSR2 = 03FFh Cycles: 2
After Instruction Q Cycle Activity:
FSR2 = 0422h Q1 Q2 Q3 Q4
literal ‘k’ Data FSR
No No No No
Operation Operation Operation Operation
Example: ADDULNK 23h

Before Instruction
FSR2 = 03FFh
PC = 0100h
After Instruction
FSR2 = 0422h
PC = (TOS)
Note: All PIC18 instructions may take an optional label argument preceding the instruction mnemonic for use in
symbolic addressing. If a label is used, the instruction format then becomes: {label} instruction argument(s).

PIC18F87K90 FAMILY
CALLW Subroutine Call Using WREG MOVSF Move Indexed to f
Syntax: CALLW Syntax: MOVSF [zs], fd

Operands: None Operands: 0  zs  127
Operation: (PC + 2)  TOS, 0  fd  4095
(W)  PCL, Operation: ((FSR2) + zs)  fd
(PCLATH)  PCH, Status Affected: None
(PCLATU)  PCU
Encoding:
Status Affected: None 1st word (source) 1110 1011 0zzz zzzzs
Encoding: 0000 0000 0001 0100 2nd word (destin.) 1111 ffff ffff ffffd
Description First, the return address (PC + 2) is Description: The contents of the source register are
pushed onto the return stack. Next, the moved to destination register ‘fd’. The
contents of W are written to PCL; the actual address of the source register is
existing value is discarded. Then, the determined by adding the 7-bit literal
contents of PCLATH and PCLATU are offset ‘zs’, in the first word, to the value
latched into PCH and PCU, of FSR2. The address of the destination
respectively. The second cycle is register is specified by the 12-bit literal
executed as a NOP instruction while the ‘fd’ in the second word. Both addresses
new next instruction is fetched. can be anywhere in the 4096-byte data
space (000h to FFFh).
Unlike CALL, there is no option to
update W, STATUS or BSR. The MOVSF instruction cannot use the
PCL, TOSU, TOSH or TOSL as the
Words: 1
destination register.
Cycles: 2
If the resultant source address points to
Q Cycle Activity: an Indirect Addressing register, the
Q1 Q2 Q3 Q4 value returned will be 00h.
Decode Read Push PC to No Words: 2
WREG stack operation
Cycles: 2
No No No No
operation operation operation operation Q Cycle Activity:
Q1 Q2 Q3 Q4
Decode Determine Determine Read
Example: HERE CALLW source addr source addr source reg
Before Instruction Decode No No Write
PC = address (HERE) operation operation register ‘f’
PCLATH = 10h (dest)
PCLATU = 00h No dummy
W = 06h read
After Instruction
PC = 001006h
TOS = address (HERE + 2) Example: MOVSF [05h], REG2
PCLATH = 10h
PCLATU = 00h Before Instruction
W = 06h FSR2 = 80h
Contents
of 85h = 33h
REG2 = 11h
After Instruction
FSR2 = 80h
Contents
of 85h = 33h
REG2 = 33h

PIC18F87K90 FAMILY
MOVSS Move Indexed to Indexed PUSHL Store Literal at FSR2, Decrement FSR2
Syntax: MOVSS [zs], [zd] Syntax: PUSHL k
Operands: 0  zs  127 Operands: 0k  255
0  zd  127
Operation: k  (FSR2),
Operation: ((FSR2) + zs)  ((FSR2) + zd) FSR2 – 1  FSR2
Encoding:
1st word (source) 1110 1011 1zzz zzzzs
2nd word (dest.) 1111 xxxx xzzz zzzzd Description: The 8-bit literal ‘k’ is written to the data
Description The contents of the source register are memory address specified by FSR2.
FSR2 is decremented by 1 after the
moved to the destination register. The
addresses of the source and destination operation.
registers are determined by adding the This instruction allows users to push
7-bit literal offsets, ‘zs’ or ‘zd’, values onto a software stack.
respectively, to the value of FSR2. Both Words: 1
registers can be located anywhere in
the 4096-byte data memory space Cycles: 1
(000h to FFFh). Q Cycle Activity:
The MOVSS instruction cannot use the Q1 Q2 Q3 Q4
PCL, TOSU, TOSH or TOSL as the Decode Read ‘k’ Process Write to
destination register. data destination
If the resultant source address points to
an Indirect Addressing register, the
value returned will be 00h. If the Example: PUSHL 08h
resultant destination address points to Before Instruction
an Indirect Addressing register, the FSR2H:FSR2L = 01ECh
instruction will execute as a NOP. Memory (01ECh) = 00h
Words: 2
After Instruction
Cycles: 2 FSR2H:FSR2L = 01EBh
Q Cycle Activity: Memory (01ECh) = 08h
Q1 Q2 Q3 Q4
Decode Determine Determine Read
source addr source addr source reg
Decode Determine Determine Write
dest addr dest addr to dest reg
Example: MOVSS [05h], [06h]

Before Instruction
FSR2 = 80h
Contents
of 85h = 33h
Contents
of 86h = 11h
After Instruction
FSR2 = 80h
Contents
of 85h = 33h
Contents
of 86h = 33h

PIC18F87K90 FAMILY
SUBFSR Subtract Literal from FSR SUBULNK Subtract Literal from FSR2 and Return
Syntax: SUBFSR f, k Syntax: SUBULNK k
Operands: 0  k  63 Operands: 0  k  63
f  [ 0, 1, 2 ] Operation: FSR2 – k  FSR2,
Operation: FSRf – k  FSRf (TOS) PC
Encoding: 1110 1001 ffkk kkkk Encoding: 1110 1001 11kk kkkk
Description: The 6-bit literal ‘k’ is subtracted from Description: The 6-bit literal ‘k’ is subtracted from the
the contents of the FSR specified contents of the FSR2. A RETURN is then
by ‘f’. executed by loading the PC with the
Words: 1 TOS.
Cycles: 1 The instruction takes two cycles to
Q Cycle Activity: execute; a NOP is performed during the
second cycle.
Q1 Q2 Q3 Q4
Decode Read Process Write to This may be thought of as a special case
of the SUBFSR instruction, where f = 3
(binary ‘11’); it operates only on FSR2.
Words: 1
Example: SUBFSR 2, 23h Cycles: 2
Before Instruction Q Cycle Activity:
FSR2 = 03FFh Q1 Q2 Q3 Q4
After Instruction Decode Read Process Write to
FSR2 = 03DCh register ‘f’ Data destination
No No No No
Operation Operation Operation Operation
Example: SUBULNK 23h

Before Instruction
FSR2 = 03FFh
PC = 0100h
After Instruction
FSR2 = 03DCh
PC = (TOS)

PIC18F87K90 FAMILY
29.2.3 BYTE-ORIENTED AND 29.2.3.1 Extended Instruction Syntax with
BIT-ORIENTED INSTRUCTIONS IN Standard PIC18 Commands
INDEXED LITERAL OFFSET MODE When the extended instruction set is enabled, the file
Note: Enabling the PIC18 instruction set exten- register argument, ‘f’, in the standard byte-oriented and
sion may cause legacy applications to bit-oriented commands is replaced with the literal offset
behave erratically or fail entirely. value, ‘k’. As already noted, this occurs only when ‘f’ is
less than or equal to 5Fh. When an offset value is used,
In addition to eight new commands in the extended set, it must be indicated by square brackets (“[ ]”). As with
enabling the extended instruction set also enables the extended instructions, the use of brackets indicates
Indexed Literal Offset Addressing (Section 6.6.1 to the compiler that the value is to be interpreted as an
“Indexed Addressing with Literal Offset”). This has index or an offset. Omitting the brackets, or using a
a significant impact on the way that many commands of value greater than 5Fh within the brackets, will
the standard PIC18 instruction set are interpreted. generate an error in the MPASM™ Assembler.
When the extended set is disabled, addresses embed- If the index argument is properly bracketed for Indexed
ded in opcodes are treated as literal memory locations: Literal Offset Addressing, the Access RAM argument is
either as a location in the Access Bank (a = 0) or in a never specified; it will automatically be assumed to be
GPR bank designated by the BSR (a = 1). When the ‘0’. This is in contrast to standard operation (extended
extended instruction set is enabled and a = 0, however, instruction set disabled), when ‘a’ is set on the basis of
a file register argument of 5Fh or less is interpreted as the target address. Declaring the Access RAM bit in
an offset from the pointer value in FSR2 and not as a this mode will also generate an error in the MPASM
literal address. For practical purposes, this means that Assembler.
all instructions that use the Access RAM bit as an
The destination argument, ‘d’, functions as before.
argument – that is, all byte-oriented and bit-oriented
instructions, or almost half of the core PIC18 instruc- In the latest versions of the MPASM Assembler,
tions – may behave differently when the extended language support for the extended instruction set must
instruction set is enabled. be explicitly invoked. This is done with either the
command line option, /y, or the PE directive in the
When the content of FSR2 is 00h, the boundaries of the
source listing.
Access RAM are essentially remapped to their original
values. This may be useful in creating backward
29.2.4 CONSIDERATIONS WHEN
compatible code. If this technique is used, it may be
ENABLING THE EXTENDED
necessary to save the value of FSR2 and restore it
when moving back and forth between C and assembly INSTRUCTION SET
routines in order to preserve the Stack Pointer. Users It is important to note that the extensions to the instruc-
must also keep in mind the syntax requirements of the tion set may not be beneficial to all users. In particular,
extended instruction set (see Section 29.2.3.1 users who are not writing code that uses a software
“Extended Instruction Syntax with Standard PIC18 stack may not benefit from using the extensions to the
Commands”). instruction set.
Although the Indexed Literal Offset mode can be very Additionally, the Indexed Literal Offset Addressing
useful for dynamic stack and pointer manipulation, it mode may create issues with legacy applications
can also be very annoying if a simple arithmetic opera- written to the PIC18 assembler. This is because
tion is carried out on the wrong register. Users who are instructions in the legacy code may attempt to address
accustomed to the PIC18 programming must keep in registers in the Access Bank below 5Fh. Since these
mind, that when the extended instruction set is addresses are interpreted as literal offsets to FSR2
enabled, register addresses of 5Fh or less are used for when the instruction set extension is enabled, the
Indexed Literal Offset Addressing. application may read or write to the wrong data
Representative examples of typical byte-oriented and addresses.
bit-oriented instructions in the Indexed Literal Offset When porting an application to the PIC18F87K90 family,
mode are provided on the following page to show how it is very important to consider the type of code. A large,
execution is affected. The operand conditions shown in re-entrant application that is written in C and would ben-
the examples are applicable to all instructions of these efit from efficient compilation will do well when using the
types. instruction set extensions. Legacy applications that
heavily use the Access Bank will most likely not benefit
from using the extended instruction set.

PIC18F87K90 FAMILY
ADD W to Indexed Bit Set Indexed

ADDWF BSF
(Indexed Literal Offset mode) (Indexed Literal Offset mode)
Syntax: ADDWF [k] {,d} Syntax: BSF [k], b

d  [0,1] 0b7
Operation: (W) + ((FSR2) + k)  dest Operation: 1  ((FSR2) + k)
Status Affected: N, OV, C, DC, Z Status Affected: None
Encoding: 0010 01d0 kkkk kkkk Encoding: 1000 bbb0 kkkk kkkk
Description: The contents of W are added to the Description: Bit ‘b’ of the register indicated by FSR2,
contents of the register indicated by offset by the value ‘k’, is set.
FSR2, offset by the value ‘k’. Words: 1
If ‘d’ is ‘0’, the result is stored in W. If ‘d’
Cycles: 1
is ‘1’, the result is stored back in
register ‘f’ (default). Q Cycle Activity:
Q1 Q2 Q3 Q4
Words: 1
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4 Example: BSF [FLAG_OFST], 7
Decode Read ‘k’ Process Write to Before Instruction
Data destination FLAG_OFST = 0Ah
FSR2 = 0A00h
Example: ADDWF [OFST] ,0
Contents
of 0A0Ah = 55h
Before Instruction After Instruction
W = 17h Contents
OFST = 2Ch of 0A0Ah = D5h
FSR2 = 0A00h
Contents
of 0A2Ch = 20h
After Instruction Set Indexed
SETF
W = 37h (Indexed Literal Offset mode)
Contents
of 0A2Ch = 20h Syntax: SETF [k]
Operands: 0  k  95
Operation: FFh  ((FSR2) + k)
Description: The contents of the register indicated by
FSR2, offset by ‘k’, are set to FFh.
Words: 1
Cycles: 1
Q Cycle Activity:
Q1 Q2 Q3 Q4
Decode Read ‘k’ Process Write
Data register
Example: SETF [OFST]

Before Instruction
OFST = 2Ch
FSR2 = 0A00h
Contents
of 0A2Ch = 00h
After Instruction
Contents
of 0A2Ch = FFh

PIC18F87K90 FAMILY
29.2.5 SPECIAL CONSIDERATIONS WITH To develop software for the extended instruction set,
MICROCHIP MPLAB® IDE TOOLS the user must enable support for the instructions and
the Indexed Addressing mode in their language tool(s).
The latest versions of Microchip’s software tools have Depending on the environment being used, this may be
been designed to fully support the extended instruction done in several ways:
set for the PIC18F87K90 family family. This includes
the MPLAB C18 C Compiler, MPASM assembly • A menu option or dialog box within the
language and MPLAB Integrated Development environment that allows the user to configure the
Environment (IDE). language tool and its settings for the project
• A command line option
When selecting a target device for software
development, MPLAB IDE will automatically set default • A directive in the source code
Configuration bits for that device. The default setting for These options vary between different compilers,
the XINST Configuration bit is ‘0’, disabling the assemblers and development environments. Users are
extended instruction set and Indexed Literal Offset encouraged to review the documentation accompany-
Addressing. For proper execution of applications ing their development systems for the appropriate
developed to take advantage of the extended information.
instruction set, XINST must be set during
programming.

PIC18F87K90 FAMILY
30.0 DEVELOPMENT SUPPORT 30.1 MPLAB Integrated Development
Environment Software
The PIC® microcontrollers and dsPIC® digital signal
controllers are supported with a full range of software The MPLAB IDE software brings an ease of software
and hardware development tools: development previously unseen in the 8/16/32-bit
• Integrated Development Environment microcontroller market. The MPLAB IDE is a Windows®
operating system-based application that contains:
- MPLAB® IDE Software
• Compilers/Assemblers/Linkers • A single graphical interface to all debugging tools
- MPLAB C Compiler for Various Device - Simulator
Families - Programmer (sold separately)
- HI-TECH C for Various Device Families - In-Circuit Emulator (sold separately)
- MPASMTM Assembler - In-Circuit Debugger (sold separately)
- MPLINKTM Object Linker/ • A full-featured editor with color-coded context
MPLIBTM Object Librarian • A multiple project manager
- MPLAB Assembler/Linker/Librarian for • Customizable data windows with direct edit of
Various Device Families contents
• Simulators • High-level source code debugging
- MPLAB SIM Software Simulator • Mouse over variable inspection
• Emulators • Drag and drop variables from source to watch
- MPLAB REAL ICE™ In-Circuit Emulator windows
• In-Circuit Debuggers • Extensive on-line help
- MPLAB ICD 3 • Integration of select third party tools, such as
- PICkit™ 3 Debug Express IAR C Compilers
• Device Programmers The MPLAB IDE allows you to:
- PICkit™ 2 Programmer • Edit your source files (either C or assembly)
- MPLAB PM3 Device Programmer • One-touch compile or assemble, and download to
• Low-Cost Demonstration/Development Boards, emulator and simulator tools (automatically
Evaluation Kits, and Starter Kits updates all project information)
• Debug using:
- Source files (C or assembly)
- Mixed C and assembly
- Machine code
MPLAB IDE supports multiple debugging tools in a
single development paradigm, from the cost-effective
simulators, through low-cost in-circuit debuggers, to
full-featured emulators. This eliminates the learning
curve when upgrading to tools with increased flexibility
and power.

PIC18F87K90 FAMILY
30.2 MPLAB C Compilers for Various 30.5 MPLINK Object Linker/
Device Families MPLIB Object Librarian
The MPLAB C Compiler code development systems The MPLINK Object Linker combines relocatable
are complete ANSI C compilers for Microchip’s PIC18, objects created by the MPASM Assembler and the
PIC24 and PIC32 families of microcontrollers and the MPLAB C18 C Compiler. It can link relocatable objects
dsPIC30 and dsPIC33 families of digital signal control- from precompiled libraries, using directives from a
lers. These compilers provide powerful integration linker script.
capabilities, superior code optimization and ease of The MPLIB Object Librarian manages the creation and
use. modification of library files of precompiled code. When
For easy source level debugging, the compilers provide a routine from a library is called from a source file, only
symbol information that is optimized to the MPLAB IDE the modules that contain that routine will be linked in
debugger. with the application. This allows large libraries to be
used efficiently in many different applications.
30.3 HI-TECH C for Various Device The object linker/library features include:
Families • Efficient linking of single libraries instead of many
The HI-TECH C Compiler code development systems smaller files
are complete ANSI C compilers for Microchip’s PIC • Enhanced code maintainability by grouping
family of microcontrollers and the dsPIC family of digital related modules together
signal controllers. These compilers provide powerful • Flexible creation of libraries with easy module
integration capabilities, omniscient code generation listing, replacement, deletion and extraction
and ease of use.
For easy source level debugging, the compilers provide 30.6 MPLAB Assembler, Linker and
symbol information that is optimized to the MPLAB IDE Librarian for Various Device
debugger. Families
The compilers include a macro assembler, linker, pre-
MPLAB Assembler produces relocatable machine
processor, and one-step driver, and can run on multiple
code from symbolic assembly language for PIC24,
platforms.
PIC32 and dsPIC devices. MPLAB C Compiler uses
the assembler to produce its object file. The assembler
30.4 MPASM Assembler generates relocatable object files that can then be
The MPASM Assembler is a full-featured, universal archived or linked with other relocatable object files and
macro assembler for PIC10/12/16/18 MCUs. archives to create an executable file. Notable features
of the assembler include:
The MPASM Assembler generates relocatable object
files for the MPLINK Object Linker, Intel® standard HEX • Support for the entire device instruction set
files, MAP files to detail memory usage and symbol • Support for fixed-point and floating-point data
reference, absolute LST files that contain source lines • Command line interface
and generated machine code and COFF files for • Rich directive set
debugging.
• Flexible macro language
The MPASM Assembler features include: • MPLAB IDE compatibility
• Integration into MPLAB IDE projects
• User-defined macros to streamline
assembly code
• Conditional assembly for multi-purpose
source files
• Directives that allow complete control over the
assembly process

PIC18F87K90 FAMILY
30.7 MPLAB SIM Software Simulator 30.9 MPLAB ICD 3 In-Circuit Debugger
The MPLAB SIM Software Simulator allows code
System
development in a PC-hosted environment by simulat- MPLAB ICD 3 In-Circuit Debugger System is Micro-
ing the PIC MCUs and dsPIC® DSCs on an instruction chip's most cost effective high-speed hardware
level. On any given instruction, the data areas can be debugger/programmer for Microchip Flash Digital Sig-
examined or modified and stimuli can be applied from nal Controller (DSC) and microcontroller (MCU)
a comprehensive stimulus controller. Registers can be devices. It debugs and programs PIC® Flash microcon-
logged to files for further run-time analysis. The trace trollers and dsPIC® DSCs with the powerful, yet easy-
buffer and logic analyzer display extend the power of to-use graphical user interface of MPLAB Integrated
the simulator to record and track program execution, Development Environment (IDE).
actions on I/O, most peripherals and internal registers.
The MPLAB ICD 3 In-Circuit Debugger probe is con-
The MPLAB SIM Software Simulator fully supports nected to the design engineer's PC using a high-speed
symbolic debugging using the MPLAB C Compilers, USB 2.0 interface and is connected to the target with a
and the MPASM and MPLAB Assemblers. The soft- connector compatible with the MPLAB ICD 2 or MPLAB
ware simulator offers the flexibility to develop and REAL ICE systems (RJ-11). MPLAB ICD 3 supports all
debug code outside of the hardware laboratory envi- MPLAB ICD 2 headers.
ronment, making it an excellent, economical software
development tool. 30.10 PICkit 3 In-Circuit Debugger/
Programmer and
30.8 MPLAB REAL ICE In-Circuit
PICkit 3 Debug Express
Emulator System
The MPLAB PICkit 3 allows debugging and program-
MPLAB REAL ICE In-Circuit Emulator System is ming of PIC® and dsPIC® Flash microcontrollers at a
Microchip’s next generation high-speed emulator for most affordable price point using the powerful graphical
Microchip Flash DSC and MCU devices. It debugs and user interface of the MPLAB Integrated Development
programs PIC® Flash MCUs and dsPIC® Flash DSCs Environment (IDE). The MPLAB PICkit 3 is connected
with the easy-to-use, powerful graphical user interface of to the design engineer's PC using a full speed USB
the MPLAB Integrated Development Environment (IDE), interface and can be connected to the target via an
included with each kit. Microchip debug (RJ-11) connector (compatible with
The emulator is connected to the design engineer’s PC MPLAB ICD 3 and MPLAB REAL ICE). The connector
using a high-speed USB 2.0 interface and is connected uses two device I/O pins and the reset line to imple-
to the target with either a connector compatible with in- ment in-circuit debugging and In-Circuit Serial Pro-
circuit debugger systems (RJ11) or with the new high- gramming™.
speed, noise tolerant, Low-Voltage Differential Signal The PICkit 3 Debug Express include the PICkit 3, demo
(LVDS) interconnection (CAT5). board and microcontroller, hookup cables and CDROM
The emulator is field upgradable through future firmware with user’s guide, lessons, tutorial, compiler and
downloads in MPLAB IDE. In upcoming releases of MPLAB IDE software.
MPLAB IDE, new devices will be supported, and new
features will be added. MPLAB REAL ICE offers
significant advantages over competitive emulators
including low-cost, full-speed emulation, run-time
variable watches, trace analysis, complex breakpoints, a
ruggedized probe interface and long (up to three meters)
interconnection cables.

PIC18F87K90 FAMILY
30.11 PICkit 2 Development 30.13 Demonstration/Development
Programmer/Debugger and Boards, Evaluation Kits, and
PICkit 2 Debug Express Starter Kits
The PICkit™ 2 Development Programmer/Debugger is A wide variety of demonstration, development and
a low-cost development tool with an easy to use inter- evaluation boards for various PIC MCUs and dsPIC
face for programming and debugging Microchip’s Flash DSCs allows quick application development on fully func-
families of microcontrollers. The full featured tional systems. Most boards include prototyping areas for
Windows® programming interface supports baseline adding custom circuitry and provide application firmware
(PIC10F, PIC12F5xx, PIC16F5xx), midrange and source code for examination and modification.
(PIC12F6xx, PIC16F), PIC18F, PIC24, dsPIC30, The boards support a variety of features, including LEDs,
dsPIC33, and PIC32 families of 8-bit, 16-bit, and 32-bit temperature sensors, switches, speakers, RS-232
microcontrollers, and many Microchip Serial EEPROM interfaces, LCD displays, potentiometers and additional
products. With Microchip’s powerful MPLAB Integrated EEPROM memory.
Development Environment (IDE) the PICkit™ 2
enables in-circuit debugging on most PIC® microcon- The demonstration and development boards can be
trollers. In-Circuit-Debugging runs, halts and single used in teaching environments, for prototyping custom
steps the program while the PIC microcontroller is circuits and for learning about various microcontroller
embedded in the application. When halted at a break- applications.
point, the file registers can be examined and modified. In addition to the PICDEM™ and dsPICDEM™ demon-
The PICkit 2 Debug Express include the PICkit 2, demo stration/development board series of circuits, Microchip
board and microcontroller, hookup cables and CDROM has a line of evaluation kits and demonstration software
with user’s guide, lessons, tutorial, compiler and for analog filter design, KEELOQ® security ICs, CAN,
MPLAB IDE software. IrDA®, PowerSmart battery management, SEEVAL®
evaluation system, Sigma-Delta ADC, flow rate
sensing, plus many more.
30.12 MPLAB PM3 Device Programmer
Also available are starter kits that contain everything
The MPLAB PM3 Device Programmer is a universal, needed to experience the specified device. This usually
CE compliant device programmer with programmable includes a single application and debug capability, all
voltage verification at VDDMIN and VDDMAX for on one board.
maximum reliability. It features a large LCD display
(128 x 64) for menus and error messages and a modu- Check the Microchip web page (www.microchip.com)
lar, detachable socket assembly to support various for the complete list of demonstration, development
package types. The ICSP™ cable assembly is included and evaluation kits.
as a standard item. In Stand-Alone mode, the MPLAB
PM3 Device Programmer can read, verify and program
PIC devices without a PC connection. It can also set
code protection in this mode. The MPLAB PM3
connects to the host PC via an RS-232 or USB cable.
The MPLAB PM3 has high-speed communications and
optimized algorithms for quick programming of large
memory devices and incorporates an MMC card for file
storage and data applications.

PIC18F87K90 FAMILY
31.0 ELECTRICAL CHARACTERISTICS
Absolute Maximum Ratings(†)

Ambient temperature under bias.............................................................................................................-40°C to +125°C
Storage temperature .............................................................................................................................. -65°C to +150°C
Voltage on any digital only I/O pin with respect to VSS (except VDD)........................................................... -0.3V to 7.5V
Voltage on MCLR with respect to VSS........................................................................................................ -0.3V to +9.0V
Voltage on any combined digital and analog pin with respect to VSS (except VDD and MCLR)...... -0.3V to (VDD + 0.3V)
Voltage on VDD with respect to VSS (regulator enabled) ............................................................................ -0.3V to 5.5V
Voltage on VDD with respect to VSS (regulator disabled) ........................................................................... -0.3V to 3.6V
Total power dissipation (Note 1) ..................................................................................................................................1W
Maximum current out of VSS pin ...........................................................................................................................300 mA
Maximum current into VDD pin ..............................................................................................................................250 mA
Input clamp current, IIK (VI < 0 or VI > VDD) .......................................................................................................... ±20 mA
Output clamp current, IOK (VO < 0 or VO > VDD) ...................................................................................................±20 mA
Maximum output current sunk by PORTA<7:6> and any PORTB and PORTC I/O pins.........................................25 mA
Maximum output current sunk by any PORTD, PORTE and PORTJ I/O pins ..........................................................8 mA
Maximum output current sunk by PORTA<5:0> and any PORTF, PORTG and PORTH I/O pins ............................2 mA
Maximum output current sourced by PORTA<7:6> and any PORTB and PORTC I/O pins ...................................25 mA
Maximum output current sourced by any PORTD, PORTE and PORTJ I/O pins .....................................................8 mA
Maximum output current sourced by PORTA<5:0> and any PORTF, PORTG and PORTH I/O pins .......................2 mA
Maximum current sunk byall ports combined.......................................................................................................200 mA
Note 1: Power dissipation is calculated as follows:
Pdis = VDD x {IDD –  IOH} +  {(VDD – VOH) x IOH} + (VOL x IOL)
† NOTICE: Stresses above those listed under “Absolute Maximum Ratings” may cause permanent damage to the
device. This is a stress rating only and functional operation of the device at those or any other conditions above
those indicated in the operation listings of this specification is not implied. Exposure to maximum rating conditions
for extended periods may affect device reliability.

PIC18F87K90 FAMILY
FIGURE 31-1: VOLTAGE-FREQUENCY GRAPH, REGULATOR ENABLED
(INDUSTRIAL/EXTENDED)(1)
6V
5.5V
5V
4V PIC18F87K90 Family PIC18F87K90 Family

(Extended) (Industrial Only)
Voltage (VDD)
3V
3V
1.8V
0 4 MHz 48 MHz 64 MHz(1)

Frequency
Note 1: FMAX = 64 MHz in all other modes. For VDD values, 1.8V to 3V, FMAX = (VDD – 1.72)/0.02 MHz.
FIGURE 31-2: VOLTAGE-FREQUENCY GRAPH, REGULATOR DISABLED

(INDUSTRIAL/EXTENDED)(1,2)
4V
3.75V
3.6V
PIC18F87K90 Family
3.25V PIC18F87K90 Family (Industrial Only)
(Extended) 3V
Voltage (VDD)
2.5V
1.8V
4 MHz 48 MHz 64 MHz

Frequency
Note 1: When the on-chip voltage regulator is disabled, VDD must be maintained so that VDD 3.6V.
2: For VDD values, 1.8V to 3V, FMAX = (VDD – 1.72)/0.02 MHz.

PIC18F87K90 FAMILY
31.1 DC Characteristics: Supply Voltage
PIC18F87K90 Family (Industrial/Extended)
Standard Operating Conditions (unless otherwise stated)
PIC18F87K90 Family Operating temperature -40°C  TA  +85°C for industrial
-40°C  TA  +125°C for extended
Param
Symbol Characteristic Min Typ Max Units Conditions
No.
D001 VDD Supply Voltage 1.8 — 3.6 V ENVREG tied to VSS
1.8 — 5.5 V ENVREG tied to VDD
D001C AVDD Analog Supply Voltage VDD – 0.3 — VDD + 0.3 V
D001D AVSS Analog Ground Potential VSS – 0.3 — VSS + 0.3 V
D002 VDR RAM Data Retention 1.5 — — V
Voltage(1)
D003 VPOR VDD Start Voltage — — 0.7 V See Section 5.3 “Power-on
to Ensure Internal Reset (POR)” for details
Power-on Reset Signal
D004 SVDD VDD Rise Rate 0.05 — — V/ms See Section 5.3 “Power-on
to Ensure Internal Reset (POR)” for details
Power-on Reset Signal
D005 BVDD Brown-out Reset Voltage
(High/Medium/Low-Power
mode)
BORV<1:0> = 11(2) 1.69 1.8 1.91
BORV<1:0> = 10 1.88 2.0 2.12
BORV<1:0> = 01 2.53 2.7 2.86
BORV<1:0> = 00 2.82 3.0 3.18
Note 1: This is the limit to which VDD can be lowered in Sleep mode, or during a device Reset, without losing RAM data.
2: The device will operate normally until Brown-out Reset occurs, even though VDD may be below VDDMIN.

PIC18F87K90 FAMILY
31.2 DC Characteristics: Power-Down and Supply Current
PIC18F87K90 Family (Industrial/Extended)
Param
Device Typ Max Units Conditions
No.
Power-Down Current (IPD)(1)
All devices 10 500 nA -40°C
20 500 nA +25°C VDD = 1.8V(4)
120 600 nA +60°C (Sleep mode)
630 1800 nA +85°C Regulator Disabled
4 9 A +125°C
All devices 50 700 A -40°C
60 700 nA +25°C VDD = 3.3V(4)
700 2700 nA +85°C Regulator Disabled
5 11 A +125°C
All devices 350 1300 nA -40°C
400 1400 nA +25°C VDD = 5V(5)
1350 4000 nA +85°C Regulator Enabled
6 12 A +125°C
Note 1: The power-down current in Sleep mode does not depend on the oscillator type. Power-down current is measured with
the part in Sleep mode, with all I/O pins in a high-impedance state and tied to VDD or VSS, and all features that add delta
current are disabled (such as WDT, SOSC oscillator, BOR, etc.).
2: The supply current is mainly a function of operating voltage, frequency and mode. Other factors, such as I/O pin loading
and switching rate, oscillator type and circuit, internal code execution pattern and temperature, also have an impact on
the current consumption.
The test conditions for all IDD measurements in active operation mode are:
OSC1 = external square wave, from rail-to-rail; all I/O pins tri-stated, pulled to VDD;
MCLR = VDD; WDT enabled/disabled as specified.
3: Standard, low-cost 32 kHz crystals have an operating temperature range of -10°C to +70°C. Extended temperature
crystals are available at a much higher cost.
4: Voltage regulator disabled (ENVREG = 0, tied to VSS, RETEN (CONFIG1L<0>) = 1).
5: Voltage regulator enabled (ENVREG = 1, tied to VDD, SRETEN (WDTCON<4>) = 1 and RETEN (CONFIG1L<0>) = 0).
6: LCD glass is not connected; resistor current is not included.
7: 48 MHz maximum frequency at 125°C.

PIC18F87K90 FAMILY
PIC18F87K90 Family (Industrial/Extended) (Continued)
Param
No.
Supply Current (IDD)(2,3)
All devices 5.3 10 A -40°C
5.5 10 A +25°C VDD = 1.8V(4)
5.5 10 A +85°C Regulator Disabled
12 24 A +125°C
FOSC = 31 kHz
10 16 A +25°C VDD = 3.3V(4)
(RC_RUN mode,
11 17 A +85°C Regulator Disabled
LF-INTOSC)
15 35 A +125°C
80 185 A +25°C VDD = 5V(5)
90 190 A +85°C Regulator Enabled
200 500 A +125°C
410 800 A +25°C VDD = 1.8V(4)
700 1500 A +125°C
FOSC = 1 MHz
680 960 A +25°C VDD = 3.3V(4)
(RC_RUN mode,
HF-INTOSC)
800 1700 A +125°C
780 1400 A +25°C VDD = 5V(5)
1200 2400 A +125°C

PIC18F87K90 FAMILY
Param
No.
760 1400 A +25°C VDD = 1.8V(4)
800 1700 A +125°C
All devices 1.4 2.5 mA -40°C
FOSC = 4 MHz
1.4 2.5 mA +25°C VDD = 3.3V(4)
(RC_RUN mode,
1.4 2.5 mA +85°C Regulator Disabled
HF-INTOSC)
1.5 3.0 mA +125°C
1.5 2.7 mA +25°C VDD = 5V(5)
1.5 2.7 mA +85°C Regulator Enabled
1.6 3.3 mA +125°C

PIC18F87K90 FAMILY
Param
No.
Supply Current (IDD) Cont.(2,3)
All devices 2.1 5.5 A -40°C
2.1 5.7 A +25°C VDD = 1.8V(4)
2.2 6.0 A +85°C Regulator Disabled
10 20 A +125°C
FOSC = 31 kHz
3.9 7.8 A +25°C VDD = 3.3V(4)
(RC_IDLE mode,
3.9 8.5 A +85°C Regulator Disabled
LF-INTOSC)
12 24 A +125°C
80 190 A +25°C VDD = 5V(5)
200 420 A +125°C
330 640 A +25°C VDD = 1.8V(4)
500 850 A +125°C
FOSC = 1 MHz
520 900 A +25°C VDD = 3.3V(4)
(RC_IDLE mode,
HF-INTOSC)
800 1200 A +125°C
600 960 A +25°C VDD = 5V(5)
1000 1400 A +125°C
470 770 A +25°C VDD = 1.8V(4)
700 1000 A +125°C
FOSC = 4 MHz
800 1350 A +25°C VDD = 3.3V(4)
(RC_IDLE mode,
internal HF-INTOSC)
1100 1400 A +125°C
890 1700 A +25°C VDD = 5V(5)
1200 2200 A +125°C

PIC18F87K90 FAMILY
Param
No.
130 390 A +25°C VDD = 1.8V(4)
250 500 A +125°C
FOSC = 1 MHZ
270 790 A +25°C VDD = 3.3V(4)
(PRI_RUN mode,
EC oscillator)
400 900 A +125°C
450 980 A +25°C VDD = 5V(5)
600 1300 A +125°C
530 900 A +25°C VDD = 1.8V(4)
750 1600 A +125°C
FOSC = 4 MHz
850 1700 A +25°C VDD = 3.3V(4)
(PRI_RUN mode,
EC oscillator)
1150 2400 A +125°C
1.1 2.6 mA +25°C VDD = 5V(5)
2.0 4.0 mA +125°C
All devices 12 19 mA -40°C
12 19 mA +25°C VDD = 3.3V(4)
12 19 mA +85°C Regulator Disabled
FOSC = 64 MHZ
13 22 mA +125°C(7)
(PRI_RUN mode,
EC oscillator)
13 20 mA +25°C VDD = 5V(5)
13 20 mA +85°C Regulator Enabled
14 23 mA +125°C(7)

PIC18F87K90 FAMILY
Param
No.
3.3 5.5 mA +25°C VDD = 3.3V(4)
3.3 5.5 mA +85°C Regulator Disabled
FOSC = 16 MHZ,
3.6 6.0 mA +125°C
(PRI_RUN mode, 4 MHz
EC oscillator with PLL)
3.5 5.8 mA +25°C VDD = 5V(5)
3.8 7.0 mA +125°C
12 18 mA +25°C VDD = 3.3V(4)
12 18 mA +85°C Regulator Disabled
FOSC = 64 MHZ,
13 22 mA +125°C(7)
(PRI_RUN mode, 16 MHz
EC oscillator with PLL)
13 20 mA +25°C VDD = 5V(5)
13 20 mA +85°C Regulator Enabled
14 24 mA +125°C(7)

PIC18F87K90 FAMILY
Param
No.
42 73 A +25°C VDD = 1.8V(4)
53 100 A +125°C
FOSC = 1 MHz
110 195 A +25°C VDD = 3.3V(4)
(PRI_IDLE mode,
EC oscillator)
130 250 A +125°C
290 440 A +25°C VDD = 5V(5)
330 500 A +125°C
160 360 A +25°C VDD = 1.8V(4)
200 400 A +125°C
FOSC = 4 MHz
340 660 A +25°C VDD = 3.3V(4)
(PRI_IDLE mode,
EC oscillator)
370 700 A +125°C
520 950 A +25°C VDD = 5V(5)
600 1200 A +125°C
All devices 4.7 9 mA -40°C
4.8 9 mA +25°C VDD = 3.3V(4)
4.8 10 mA +85°C Regulator Disabled
FOSC = 64 MHz
5.2 12 mA +125°C(7)
(PRI_IDLE mode,
All devices 5.1 11 mA -40°C
EC oscillator)
5.1 11 mA +25°C VDD = 5V(5)
5.2 12 mA +85°C Regulator Enabled
5.7 14 mA +125°C(7)

PIC18F87K90 FAMILY
Param
No.
All devices 3.7 8.5 µA -40°C
5.4 10 µA +25°C VDD = 1.8V(4)
6.6 13 µA +85°C Regulator Disabled
13 30 µA +125°C
All devices 8.7 18 µA -40°C
FOSC = 32 kHz(3)
10 20 µA +25°C VDD = 3.3V(4)
(SEC_RUN mode,
12 23 µA +85°C Regulator Disabled
SOSCSEL = 01)
25 60 µA +125°C
All devices 60 160 µA -40°C
90 190 µA +25°C VDD = 5V(5)
100 240 µA +85°C Regulator Enabled
200 450 µA +125°C
1.7 5 µA +25°C VDD = 1.8V(4)
9 20 µA +125°C
FOSC = 32 kHz(3)
2.8 9 µA +25°C VDD = 3.3V(4)
(SEC_IDLE mode,
SOSCSEL = 01)
17 40 µA +125°C
All devices 60 150 µA -40°C
80 180 µA +25°C VDD = 5V(5)
100 240 µA +85°C Regulator Enabled
180 440 µA +125°C

PIC18F87K90 FAMILY
Param
No.
Module Differential Currents (IWDT, IBOR, IHLVD, IOSCB, IAD)
D022 Watchdog Timer
(IWDT) All devices 0.3 1 A -40°C
0.3 1 A +25°C VDD = 1.8V(4)
0.5 2 µA +125°C
0.6 2 A +25°C VDD = 3.3V(4)
1 3 µA +125°C
0.6 2 A +25°C VDD = 5V(5)
0.7 2 A +85°C Regulator Enabled
1.5 4 µA +125°C
D022A Brown-out Reset
(IBOR) All devices 4.6 19 A -40°C
(IBOR) 4.5 20 A +25°C VDD = 3.3V(4)
High-Power BOR
18 40 µA +125°C
4.3 20 A +25°C VDD = 5V(5)
High-Power BOR
20 40 µA +125°C
D022B High/Low-Voltage Detect
(IHLVD) All devices 3.8 9 A -40°C
4.2 9 A +25°C VDD = 1.8V(4)
4.5 12 µA +125°C
4.8 12 A +25°C VDD = 3.3V(4)
5.0 14 µA +125°C
4.9 13 A +25°C VDD = 5V(5)
5.3 15 µA +125°C

PIC18F87K90 FAMILY
Param
No.
D025 Real-Time Clock/Calendar with SOSC Oscillator
(IRTCC) All devices 0.7 2.7 µA -40°C
0.7 2.7 µA +25°C
VDD = 1.8V(4)
1.1 2.8 µA +60°C
Regulator Disabled
1.1 2.9 µA +85°C
2.2 4.4 µA +125°C
All devices 1.2 2.9 µA -40°C
1.1 2.8 µA +25°C
VDD = 3.3V(4) 32.768 kHz,
2 4.6 µA +60°C
Regulator Disabled SOSCGO = 1
2 4.8 A +85°C
4 6.5 µA +125°C
1.5 4.4 A +25°C
VDD = 5V(5)
1.7 4.7 µA +60°C
Regulator Enabled
1.7 4.7 A +85°C
3.5 6.0 µA +125°C
D025B LCD Internal Biasing
(ILCD) PIC18FXXK90 0.6 2.8 A -40°C
0.6 2.8 µA +25°C VDD = 1.8V(4)
0.7 3.4 µA +85°C Regulator Disabled
0.8 5.5 µA +125°C
PIC18FXXK90 1.1 3.9 A -40°C
Internal biasing(6)
1.0 3.9 µA +25°C VDD = 3.3V(4) 1/4 Multiplex mode
1.1 4.5 µA +85°C Regulator Disabled Type-A wave form
1.2 6.6 µA +125°C LCD clock is internal RC
PIC18FXXK90 1.3 5.8 µA -40°C
1.2 5.8 µA +25°C VDD = 5V
1.6 6.5 µA +85°C Regulator Enabled
1.8 9.8 µA +125°C

PIC18F87K90 FAMILY
Param
No.
D025B LCD External Biasing
(ILCD) PIC18FXXK90 0.3 1.4 µA -40°C
0.3 1.4 µA +25°C VDD = 1.8V(4)
0.7 1.7 µA +85°C Regulator Disabled
0.8 2.3 A +125°C
PIC18FXXK90 0.7 2.9 A -40°C External biasing(6)
0.7 3.5 A +25°C VDD = 3.3V (4) 1/4 Multiplex mode
1.1 3.9 A +85°C Regulator Disabled Type-A wave form
LCD clock is internal RC
1.2 5.8 A +125°C
PIC18FXXK90 0.8 3.3 A -40°C
1.1 4.1 A +25°C VDD = 5V
1.1 4.2 A +85°C Regulator Enabled
1.3 6.2 A +125°C

PIC18F87K90 FAMILY
31.3 DC Characteristics: PIC18F87K90 Family (Industrial/Extended)
DC CHARACTERISTICS Operating temperature -40°C TA  +85°C for industrial
-40°C TA  +125°C for extended
Param
Symbol Characteristic Min Max Units Conditions
No.
VIL Input Low Voltage
All I/O Ports:
D030 with TTL Buffer VSS 0.15 VDD V VDD < 4.5V
D031 with Schmitt Trigger Buffer — 0.8 V 4.5  VDD  5.5V
RC3, RC4 VSS 0.2 VDD V VDD < 4.5
RD5, RD6 VSS 1.5 V 4.5  VDD  5.5V
D031A RC3, RC4 VSS 0.3 VDD V I2C™ enabled
D031B RD5, RD6 VSS 0.8 V SMBus enabled
D032 MCLR VSS 0.2 VDD V
D033 OSC1 VSS 0.2 VDD V LP, XT, HS, HSPLL modes
D033A OSC1 VSS 0.2 VDD V EC, ECPLL modes
D034 SOSCI VSS 0.3 VDD V
VIH Input High Voltage
I/O Ports:
D040 with TTL Buffer 0.25 VDD VDD V VDD < 4.5V
2.0 VDD 4.5  VDD  5.5V
D041 with Schmitt Trigger Buffer 0.8 VDD VDD V
RC3, RC4 0.7 VDD VDD V VDD < 4.5
RD5, RD6 3V 5.5 V 4.5  VDD  5.5V
D041A RC3, RC4 0.7 VDD VDD V I2C enabled
D041B RD5, RD6 2.1 VDD V SMBus enabled
D042 MCLR 0.8 VDD VDD V
D043 OSC1 0.7 VDD VDD V LP, XT, HS, HSPLL modes
D043A OSC1 0.8 VDD VDD V EC, ECPLL modes
D044 SOSCI 0.7 VDD VDD V
IIL Input Leakage Current(1)
D060 I/O Ports ±50 ±200 nA VSS VPIN VDD,
Pin at High-Impedance
D061 MCLR — ±5 A VSS VPIN VDD
D063 OSC1 — ±5 A VSS VPIN VDD
IPU Weak Pull-up Current
D070 IPURB PORTB Weak Pull-up Current 50 400 A VDD = 3.3V, VPIN = VSS
Note 1: Negative current is defined as current sourced by the pin.

PIC18F87K90 FAMILY
31.3 DC Characteristics: PIC18F87K90 Family (Industrial/Extended) (Continued)
DC CHARACTERISTICS Operating temperature -40°C TA  +85°C for industrial
Param
No.
VOL Output Low Voltage
D080 I/O Ports:
PORTA,PORTB,PORTC — 0.6 V IOL = 8.5 mA, VDD = 4.5V,
-40C to +125C
PORTD, PORTE, PORTF, — 0.6 V IOL = 3.5 mA, VDD = 4.5V,
PORTG, PORTH, PORTJ -40C to +125C
D083 OSC2/CLKO — 0.6 V IOL = 1.6 mA, VDD = 5.5V,
(EC modes) -40C to +125C
VOH Output High Voltage(1)
D090 I/O Ports:
PORTA,PORTB,PORTC VDD – 0.7 — V IOH = -3 mA, VDD = 4.5V,
-40C to +125C
PORTD, PORTE, PORTF, VDD – 0.7 — V IOH = -2 mA, VDD = 4.5V,
PORTG, PORTH, PORTJ -40C to +125C
D092 OSC2/CLKO VDD – 0.7 — V IOH = -1 mA, VDD = 5.5V,
(INTOSC, EC modes) -40C to +125C
Capacitive Loading Specs
on Output Pins
D100 COSC2 OSC2 Pin — 20 pF In HS mode when external
clock is used to drive OSC1
D101 CIO All I/O Pins and OSC2 — 50 pF To meet the AC Timing
Specifications
D102 CB SCLx, SDAx — 400 pF I2C™ Specification
Note 1: Negative current is defined as current sourced by the pin.
31.4 DC Characteristics: CTMU Current Source Specifications

Standard Operating Conditions: 1.8V to 3.6V (unless otherwise stated)
DC CHARACTERISTICS Operating temperature -40°C  TA  +85°C for industrial
Param
Sym Characteristic Min Typ(1) Max Units Conditions
No.
IOUT1 CTMU Current Source,
— 550 — nA CTMUICON<1:0> = 01
Base Range
— 5.5 — A CTMUICON<1:0> = 10
10x Range
— 55 — A CTMUICON<1:0> = 11
100x Range
Note 1: Nominal value at center point of current trim range (CTMUICON<7:2> = 000000).

PIC18F87K90 FAMILY
TABLE 31-1: MEMORY PROGRAMMING REQUIREMENTS
DC CHARACTERISTICS Operating temperature -40°C  TA  +85°C for industrial
Param
Sym Characteristic Min Typ† Max Units Conditions
No.
Internal Program Memory
Programming Specifications(1)
D110 VPP Voltage on MCLR/VPP/RE5 pin VDD + 1.5 — 9 V (Note 3)
D113 IDDP Supply Current during — — 10 mA
Programming
Data EEPROM Memory (Note 2)
D120 ED Byte Endurance 100K — — E/W -40C to +125C
D121 VDRW VDD for Read/Write 1.8 — 5.5 V Using EECON to read/write
ENVREG tied to VDD
1.8 — 3.6 V Using EECON to read/write
ENVREG tied to VSS
D122 TDEW Erase/Write Cycle Time — 4 — ms
D123 TRETD Characteristic Retention 40 — — Year Provided no other
specifications are violated
D124 TREF Number of Total Erase/Write 1M 10M — E/W -40°C to +125°C
Cycles before Refresh(2)
Program Flash Memory
D130 EP Cell Endurance 10K — — E/W -40C to +125C
D131 VPR VDD for Read 1.8 — 5.5 V ENVREG tied to VDD
1.8 — 3.6 V ENVREG tied to VSS
D132B VPEW Voltage for Self-Timed Erase or
Write Operations
VDD 1.8 — 5.5 V ENVREG tied to VDD
D133A TIW Self-Timed Write Cycle Time — 2 — ms
D134 TRETD Characteristic Retention 40 — — Year Provided no other
specifications are violated
D135 IDDP Supply Current during — — 10 mA
Programming
D140 TWE Writes per Erase Cycle — — 1 For each physical address
† Data in “Typ” column is at 3.3V, 25°C unless otherwise stated. These parameters are for design guidance
only and are not tested.
Note 1: These specifications are for programming the on-chip program memory through the use of table write
instructions.
2: Refer to Section 8.8 “Using the Data EEPROM” for a more detailed discussion on data EEPROM
endurance.
3: The MPLAB® ICD 2 does not support variable VPP output. Circuitry to limit the ICD 2 VPP voltage must be
placed between the ICD 2 and target system when programming or debugging with the ICD 2.

PIC18F87K90 FAMILY
TABLE 31-2: COMPARATOR SPECIFICATIONS
Operating Conditions: 1.8V  VDD  5V, -40°C  TA  +125°C (unless otherwise stated)
Param
Sym Characteristics Min Typ Max Units Comments
No.
D300 VIOFF Input Offset Voltage — ±5.0 40 mV
D301 VICM Input Common-Mode Voltage — — AVDD – 1.5 V
D302 CMRR Common-Mode Rejection Ratio 55 — — dB
D303 TRESP Response Time(1) — 150 400 ns
D304 TMC2OV Comparator Mode Change to — — 10 s
Output Valid*
Note 1: Response time measured with one comparator input at (AVDD – 1.5)/2, while the other input transitions
from VSS to VDD.
TABLE 31-3: VOLTAGE REFERENCE SPECIFICATIONS

Operating Conditions: 1.8V  VDD  5V, -40°C  TA  +125°C (unless otherwise stated)
Param
No.
D310 VRES Resolution — VDD/32 — LSb
D311 VRAA Absolute Accuracy — — 1/2 LSb
D312 VRUR Unit Resistor Value (R) — 2k — 
D313 TSET Settling Time(1) — — 10 s
Note 1: Settling time measured while CVRR = 1 and CVR<3:0> transitions from ‘0000’ to ‘1111’.
TABLE 31-4: INTERNAL VOLTAGE REGULATOR SPECIFICATIONS

Operating Conditions: -40°C  TA  +125°C (unless otherwise stated)
Param
No.
VRGOUT Regulator Output Voltage — 3.3 — V
CEFC External Filter Capacitor Value 4.7 10 — F Capacitor must be
low-ESR, a low series
resistance (< 5)

PIC18F87K90 FAMILY
31.5 AC (Timing) Characteristics
31.5.1 TIMING PARAMETER SYMBOLOGY
The timing parameter symbols have been created
following one of the following formats:
1. TppS2ppS 3. TCC:ST (I2C specifications only)

2. TppS 4. Ts (I2C specifications only)
T
F Frequency T Time
Lowercase letters (pp) and their meanings:
pp
cc CCP1 osc OSC1
ck CLKO rd RD
cs CS rw RD or WR
di SDI sc SCK
do SDO ss SS
dt Data in t0 T0CKI
io I/O port t1 T1CKI
mc MCLR wr WR
Uppercase letters and their meanings:
S
F Fall P Period
H High R Rise
I Invalid (High-Impedance) V Valid
L Low Z High-Impedance
I2C only
AA output access High High
BUF Bus free Low Low
TCC:ST (I2C specifications only)
CC
HD Hold SU Setup
ST
DAT DATA input hold STO Stop condition
STA Start condition

PIC18F87K90 FAMILY
31.5.2 TIMING CONDITIONS
The temperature and voltages specified in Table 31-5
apply to all timing specifications unless otherwise
noted. Figure 31-3 specifies the load conditions for the
timing specifications.
TABLE 31-5: TEMPERATURE AND VOLTAGE SPECIFICATIONS – AC

Operating temperature -40°C  TA +85°C for industrial
AC CHARACTERISTICS
-40°C  TA +125°C for extended
Operating voltage VDD range as described in Section 31.1 and Section 31.3.
FIGURE 31-3: LOAD CONDITIONS FOR DEVICE TIMING SPECIFICATIONS
Load Condition 1 Load Condition 2
VDD/2
RL
CL Pin CL
Pin
VSS VSS
RL = 464
CL = 50 pF for all pins except OSC2/CLKO/RA6
and including D and E outputs as ports
CL = 15 pF for OSC2/CLKO/RA6

PIC18F87K90 FAMILY
31.5.3 TIMING DIAGRAMS AND SPECIFICATIONS
FIGURE 31-4: EXTERNAL CLOCK TIMING
Q4 Q1 Q2 Q3 Q4 Q1
OSC1
1 3 3 4 4
2
CLKO
TABLE 31-6: EXTERNAL CLOCK TIMING REQUIREMENTS

Param.
No.
1A FOSC External CLKIN DC 64 MHz EC, ECIO Oscillator mode,
Frequency(1) -40°C  TA +85°C
DC 48 MHz -40°C  TA +125°C
(1)
Oscillator Frequency DC 4 MHz RC Oscillator mode
0.1 4 MHz XT Oscillator mode
4 16 MHz HS Oscillator mode
4 16 MHz HS + PLL Oscillator mode
5 33 kHz LP Oscillator mode
1 TOSC External CLKIN Period(1) 15.6 — ns EC, ECIO Oscillator mode
(1)
Oscillator Period 250 — ns RC Oscillator mode
250 10,000 ns XT Oscillator mode
40 250 ns HS Oscillator mode
62.5 250 ns HS + PLL Oscillator mode
5 200 s LP Oscillator mode
2 TCY Instruction Cycle Time(1) 62.5 — ns TCY = 4/FOSC
3 TOSL, External Clock in (OSC1) 30 — ns XT Oscillator mode
TOSH High or Low Time 2.5 — s LP Oscillator mode
10 — ns HS Oscillator mode
4 TOSR, External Clock in (OSC1) — 20 ns XT Oscillator mode
TOSF Rise or Fall Time — 50 ns LP Oscillator mode
— 7.5 ns HS Oscillator mode
Note 1: Instruction cycle period (TCY) equals four times the input oscillator time base period for all configurations
except PLL. All specified values are based on characterization data for that particular oscillator type under
standard operating conditions with the device executing code. Exceeding these specified limits may result
in an unstable oscillator operation and/or higher than expected current consumption. All devices are tested
to operate at “min.” values with an external clock applied to the OSC1/CLKIN pin. When an external clock
input is used, the “max.” cycle time limit is “DC” (no clock) for all devices.

PIC18F87K90 FAMILY
TABLE 31-7: PLL CLOCK TIMING SPECIFICATIONS (VDD = 1.8V TO 5.5V)
Param
Sym Characteristic Min Typ† Max Units Conditions
No.
F10 FOSC Oscillator Frequency Range 4 — 5 MHz VDD = 1.8-5.5V
4 — 16 MHz VDD = 3.0-5.5V,
-40°C to +85°C
4 — 12 MHz VDD = 3.0-5.5V,
-40°C to +125°C
F11 FSYS On-Chip VCO System Frequency 16 — 20 MHz VDD = 1.8-5.5V
16 — 64 MHz VDD = 3.0-5.5V,
-40°C to +85°C
16 — 48 MHz VDD = 3.0-5.5V,
-40°C to +125°C
F12 trc PLL Start-up Time (Lock Time) — — 2 ms
F13 CLK CLKOUT Stability (Jitter) -2 — +2 %
† Data in “Typ” column is at 3V, 25C unless otherwise stated. These parameters are for design guidance
only and are not tested.
TABLE 31-8: INTERNAL RC ACCURACY (INTOSC)

PIC18F87K90 Family Operating temperature -40°C  TA  +85°C industerial
-40°C  TA  +125°C extended
Param
Min Typ Max Units Conditions
No.
OA1 HF-INTOSC Accuracy @ Freq = 16 MHz, 8 MHz, 4 MHz, 2 MHz, 1 MHz, 500 kHz, 250 kHz(1)
-2 — 2 % +25°C VDD = 3.0-5.0V
-5 — 5 % -40°C to +85°C VDD = 3.0-5.0V
-5 — 5 % -40°C to +85°C VDD = 1.8-5.0V
-10 — 10 % -40°C to +125°C VDD = 1.8-5.5V
OA2 LF-INTOSC Accuracy @ Freq = 31 kHz
-15 — 15 % -40°C to +125°C VDD = 1.8-5.0V
Note 1: Frequency calibrated at 25°C. OSCTUNE register can be used to compensate for temperature drift.

PIC18F87K90 FAMILY
FIGURE 31-5: CLKO AND I/O TIMING
Q4 Q1 Q2 Q3
OSC1
10 11
CLKO
13 12
14 19 18
16
I/O Pin
(Input)
17 15
I/O Pin Old Value New Value

(Output)
20, 21
Note: Refer to Figure 31-3 for load conditions.
TABLE 31-9: CLKO AND I/O TIMING REQUIREMENTS

Param
No.
10 TOSH2CKL OSC1  to CLKO  — 75 200 ns (Note 1)
11 TOSH2CKH OSC1  to CLKO  — 75 200 ns (Note 1)
12 TCKR CLKO Rise Time — 15 30 ns (Note 1)
13 TCKF CLKO Fall Time — 15 30 ns (Note 1)
14 TCKL2IOV CLKO  to Port Out Valid — — 0.5 TCY + 20 ns
15 TIOV2CKH Port In Valid before CLKO  0.25 TCY + 25 — — ns
16 TCKH2IOI Port In Hold after CLKO  0 — — ns
17 TOSH2IOV OSC1  (Q1 cycle) to Port Out Valid — 50 150 ns
18 TOSH2IOI OSC1  (Q2 cycle) to Port Input Invalid 100 — — ns
(I/O in hold time)
19 TIOV2OSH Port Input Valid to OSC1  0 — — ns
(I/O in setup time)
20 TIOR Port Output Rise Time — 10 25 ns
21 TIOF Port Output Fall Time — 10 25 ns
22† TINP INTx pin High or Low Time 20 — — ns
23† TRBP RB<7:4> Change INTx High or Low TCY — — ns
Time
† These parameters are asynchronous events not related to any internal clock edges.
Note 1: Measurements are taken in EC mode, where CLKO output is 4 x TOSC.

PIC18F87K90 FAMILY
FIGURE 31-6: RESET, WATCHDOG TIMER, OSCILLATOR START-UP TIMER AND
POWER-UP TIMER TIMING
VDD
MCLR
30
Internal
POR
33
PWRT
Time-out 32
Oscillator
Time-out
Internal
Reset
Watchdog
Timer
Reset
31
34 34
I/O Pins
FIGURE 31-7: BROWN-OUT RESET TIMING
VDD BVDD
35
VBGAP = 1.2V
VIRVST
Enable Internal
Reference Voltage
Internal Reference
Voltage Stable 36

PIC18F87K90 FAMILY
TABLE 31-10: RESET, WATCHDOG TIMER, OSCILLATOR START-UP TIMER, POWER-UP TIMER
AND BROWN-OUT RESET REQUIREMENTS
Param.
No.
30 TmcL MCLR Pulse Width (low) 2 — — s VDD = 3.3-5.0V,

-40°C to +85°C
5 — — s VDD = 3.3-5.0V
31 TWDT Watchdog Timer Time-out Period — 4.00 — ms
(no postscaler)
32 TOST Oscillation Start-up Timer Period 1024 TOSC — 1024 TOSC — TOSC = OSC1 period
33 TPWRT Power-up Timer Period — 65.5 140 ms
34 TIOZ I/O High-Impedance from MCLR — 2 — s
Low or Watchdog Timer Reset
35 TBOR Brown-out Reset Pulse Width 200 — — s VDD  BVDD
(see D005)
36 TIRVST Time for Internal Reference — 25 — s
Voltage to become Stable
37 THLVD High/Low-Voltage Detect Pulse Width 200 — — s VDD  VHLVD
38 TCSD CPU Start-up Time 5 — 10 s
39 TIOBST Time for INTOSC to Stabilize — 1 — s

PIC18F87K90 FAMILY
FIGURE 31-8: HIGH/LOW-VOLTAGE DETECT CHARACTERISTICS
For VDIRMAG = 1: VDD
VHLVD
(HLVDIF set by hardware) (HLVDIF can be

cleared in software)
VHLVD
For VDIRMAG = 0: VDD
HLVDIF
TABLE 31-11: HIGH/LOW-VOLTAGE DETECT CHARACTERISTICS

Operating temperature -40°C  TA  +85°C for industrial
Param
Sym Characteristic Min Typ Max Units Conditions
No.
D420 HLVD Voltage on VDD HLVDL<3:0> = 0000 1.80 1.86 1.90 V
Transition High-to-Low HLVDL<3:0> = 0001 2.03 2.12 2.13 V
HLVDL<3:0> = 0010 2.24 2.33 2.35 V
HLVDL<3:0> = 0011 2.40 2.49 2.53 V
HLVDL<3:0> = 0100 2.50 2.59 2.62 V
HLVDL<3:0> = 0101 2.70 2.75 2.84 V
HLVDL<3:0> = 0110 2.82 2.93 2.97 V
HLVDL<3:0> = 0111 2.95 3.07 3.10 V
HLVDL<3:0> = 1000 3.24 3.30 3.41 V
HLVDL<3:0> = 1001 3.42 3.48 3.59 V
HLVDL<3:0> = 1010 3.61 3.67 3.79 V
HLVDL<3:0> = 1011 3.82 3.87 4.01 V
HLVDL<3:0> = 1100 4.06 4.21 4.26 V
HLVDL<3:0> = 1101 4.33 4.42 4.55 V
HLVDL<3:0> = 1110 4.64 4.77 4.87 V

PIC18F87K90 FAMILY
FIGURE 31-9: TIMER0 AND TIMER1 EXTERNAL CLOCK TIMINGS
T0CKI
40 41
42
SOSCO/T1CKI
45 46
47 48
TMR0 or
TMR1
TABLE 31-12: TIMER0 AND TIMER1 EXTERNAL CLOCK REQUIREMENTS

Param
No.
40 TT0H T0CKI High Pulse Width No prescaler 0.5 TCY + 20 — ns
With prescaler 10 — ns
41 TT0L T0CKI Low Pulse Width No prescaler 0.5 TCY + 20 — ns
With prescaler 10 — ns
42 TT0P T0CKI Period No prescaler TCY + 10 — ns
With prescaler Greater of: — ns N = prescale
20 ns or value
(TCY + 40)/N (1, 2, 4,..., 256)
45 TT1H T1CKI High Synchronous, no prescaler 0.5 TCY + 20 — ns
Time Synchronous, with prescaler 10 — ns
Asynchronous 30 — ns
46 TT1L T1CKI Low Synchronous, no prescaler 0.5 TCY + 5 — ns
Time Synchronous, with prescaler 10 — ns
47 TT1P T1CKI Input Synchronous Greater of: — ns N = prescale
Period 20 ns or value
(TCY + 40)/N (1, 2, 4, 8)
FT 1 T1CKI Oscillator Input Frequency Range DC 50 kHz
48 TCKE2TMRI Delay from External T1CKI Clock Edge to 2 TOSC 7 TOSC —
Timer Increment

PIC18F87K90 FAMILY
FIGURE 31-10: CAPTURE/COMPARE/PWM TIMINGS (ECCP1, ECCP2 MODULES)
CCPx
(Capture Mode)
50 51
52
CCPx
(Compare or PWM Mode)
53 54
TABLE 31-13: CAPTURE/COMPARE/PWM REQUIREMENTS (ECCP1, ECCP2 MODULES)

Param
No.
50 TCCL CCPx Input Low No prescaler 0.5 TCY + 20 — ns
Time With prescaler 10 — ns
51 TCCH CCPx Input No prescaler 0.5 TCY + 20 — ns
High Time With prescaler 10 — ns
52 TCCP CCPx Input Period 3 TCY + 40 — ns N = prescale
N value (1, 4 or 16)
53 TCCR CCPx Output Fall Time — 25 ns
54 TCCF CCPx Output Fall Time — 25 ns

PIC18F87K90 FAMILY
FIGURE 31-11: EXAMPLE SPI MASTER MODE TIMING (CKE = 0)
SCKx
(CKP = 0)
78 79
SCKx
(CKP = 1)
79 78
80
SDOx MSb bit 6 - - - - - - 1 LSb
75, 76
SDIx MSb In bit 6 - - - - 1 LSb In

74
73
TABLE 31-14: EXAMPLE SPI MODE REQUIREMENTS (MASTER MODE, CKE = 0)

Param
No.
73 TDIV2SCH, Setup Time of SDIx Data Input to SCKx Edge 20 — ns
TDIV2SCL
73A TB2B Last Clock Edge of Byte 1 to the 1st Clock Edge 1.5 TCY + 40 — ns
of Byte 2
74 TSCH2DIL, Hold Time of SDIx Data Input to SCKx Edge 40 — ns
TSCL2DIL
75 TDOR SDOx Data Output Rise Time — 25 ns
76 TDOF SDOx Data Output Fall Time — 25 ns
78 TSCR SCKx Output Rise Time (Master mode) — 25 ns
79 TSCF SCKx Output Fall Time (Master mode) — 25 ns
80 TSCH2DOV, SDOx Data Output Valid after SCKx Edge — 50 ns
TSCL2DOV

PIC18F87K90 FAMILY
FIGURE 31-12: EXAMPLE SPI MASTER MODE TIMING (CKE = 1)
81
SCKx
(CKP = 0)
79
73
SCKx
(CKP = 1)
80
78
75, 76
74
TABLE 31-15: EXAMPLE SPI MODE REQUIREMENTS (MASTER MODE, CKE = 1)

Param.
No.
73A TB2B Last Clock Edge of Byte 1 to the 1st Clock Edge 1.5 TCY + 40 — ns
of Byte 2
TSCL2DIL
TSCL2DOV
81 TDOV2SCH, SDOx Data Output Setup to SCKx Edge TCY — ns
TDOV2SCL

PIC18F87K90 FAMILY
FIGURE 31-13: EXAMPLE SPI SLAVE MODE TIMING (CKE = 0)
SSx
70
SCKx
(CKP = 0) 83
71 72
78 79
SCKx
(CKP = 1)
79 78
80
75, 76 77

74
73
TABLE 31-16: EXAMPLE SPI MODE REQUIREMENTS (SLAVE MODE TIMING, CKE = 0)
Param
No.
70 TSSL2SCH, SSx  to SCKx  or SCKx  Input 3 TCY — ns
TSSL2SCL
70A TSSL2WB SSx to write to SSPxBUF 3 TCY — ns
71 TSCH SCKx Input High Time Continuous 1.25 TCY + 30 — ns
71A (Slave mode) Single Byte 40 — ns (Note 1)
72 TSCL SCKx Input Low Time Continuous 1.25 TCY + 30 — ns
73 TDIV2SCH, Setup Time of SDIx Data Input to SCKx Edge 20 — ns
TDIV2SCL
73A TB2B Last Clock Edge of Byte 1 to the First Clock Edge of Byte 2 1.5 TCY + 40 — ns (Note 2)
TSCL2DIL
77 TSSH2DOZ SSx  to SDOx Output High-Impedance 10 50 ns
TSCL2DOV
83 TSCH2SSH, SSx  after SCKx Edge 1.5 TCY + 40 — ns
TSCL2SSH
Note 1: Requires the use of Parameter #73A.
2: Only if Parameter #71A and #72A are used.

PIC18F87K90 FAMILY
FIGURE 31-14: EXAMPLE SPI SLAVE MODE TIMING (CKE = 1)
82
SSx
70
SCKx 83
(CKP = 0)
71 72
SCKx
(CKP = 1)
80
75, 76 77
SDIx
MSb In bit 6 - - - - 1 LSb In
74
TABLE 31-17: EXAMPLE SPI SLAVE MODE REQUIREMENTS (CKE = 1)

Param
No.
70 TSSL2SCH, SSx  to SCKx  or SCKx  Input 3 TCY — ns
TSSL2SCL
70A TSSL2WB SSx to write to SSPxBUF 3 TCY — ns
71 TSCH SCKx Input High Time Continuous 1.25 TCY + 30 — ns
72 TSCL SCKx Input Low Time Continuous 1.25 TCY + 30 — ns
73A TB2B Last Clock Edge of Byte 1 to the First Clock Edge of Byte 2 1.5 TCY + 40 — ns (Note 2)
TSCL2DIL
77 TSSH2DOZ SSx  to SDOx Output High-Impedance 10 50 ns
TSCL2DOV
82 TSSL2DOV SDOx Data Output Valid after SSx  Edge — 50 ns
83 TSCH2SSH, SSx  after SCKx Edge 1.5 TCY + 40 — ns
TSCL2SSH
Note 1: Requires the use of Parameter #73A.
2: Only if Parameter #71A and #72A are used.

PIC18F87K90 FAMILY
FIGURE 31-15: I2C™ BUS START/STOP BITS TIMING
SCLx
91 93
90 92
SDAx
Start Stop
Condition Condition
TABLE 31-18: I2C™ BUS START/STOP BITS REQUIREMENTS (SLAVE MODE)

Param.
No.
90 TSU:STA Start Condition 100 kHz mode 4700 — ns Only relevant for Repeated
Setup Time 400 kHz mode 600 — Start condition
91 THD:STA Start Condition 100 kHz mode 4000 — ns After this period, the first
Hold Time 400 kHz mode 600 — clock pulse is generated
92 TSU:STO Stop Condition 100 kHz mode 4700 — ns
Setup Time 400 kHz mode 600 —
93 THD:STO Stop Condition 100 kHz mode 4000 — ns
Hold Time 400 kHz mode 600 —

PIC18F87K90 FAMILY
FIGURE 31-16: I2C™ BUS DATA TIMING
103 100 102

101
SCLx
90
106 107
91 92
SDAx
In
110
109 109
SDAx
Out
TABLE 31-19: I2C™ BUS DATA REQUIREMENTS (SLAVE MODE)

Param.
No.
100 THIGH Clock High Time 100 kHz mode 4.0 — s

400 kHz mode 0.6 — s
MSSP module 1.5 TCY —
101 TLOW Clock Low Time 100 kHz mode 4.7 — s
400 kHz mode 1.3 — s
MSSP module 1.5 TCY —
102 TR SDAx and SCLx Rise Time 100 kHz mode — 1000 ns
400 kHz mode 20 + 0.1 CB 300 ns CB is specified to be from
10 to 400 pF
103 TF SDAx and SCLx Fall Time 100 kHz mode — 300 ns
400 kHz mode 20 + 0.1 CB 300 ns CB is specified to be from
10 to 400 pF
90 TSU:STA Start Condition Setup Time 100 kHz mode 4.7 — s Only relevant for Repeated
400 kHz mode 0.6 — s Start condition
91 THD:STA Start Condition Hold Time 100 kHz mode 4.0 — s After this period, the first clock
400 kHz mode 0.6 — s pulse is generated
106 THD:DAT Data Input Hold Time 100 kHz mode 0 — ns
400 kHz mode 0 0.9 s
107 TSU:DAT Data Input Setup Time 100 kHz mode 250 — ns (Note 2)
400 kHz mode 100 — ns
92 TSU:STO Stop Condition Setup Time 100 kHz mode 4.7 — s
400 kHz mode 0.6 — s
109 TAA Output Valid from Clock 100 kHz mode — 3500 ns (Note 1)
400 kHz mode — — ns
110 TBUF Bus Free Time 100 kHz mode 4.7 — s Time the bus must be free before
400 kHz mode 1.3 — s 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 SCLx 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 SCLx signal. If
such a device does stretch the LOW period of the SCLx signal, it must output the next data bit to the SDAx line,
TR max. + TSU:DAT = 1000 + 250 = 1250 ns (according to the Standard mode I2C bus specification), before the SCLx line
is released.

PIC18F87K90 FAMILY
FIGURE 31-17: MSSP I2C™ BUS START/STOP BITS TIMING WAVEFORMS
SCLx
91 93
90 92
SDAx
Start Stop
Condition Condition
TABLE 31-20: MSSP I2C™ BUS START/STOP BITS REQUIREMENTS

Param.
No.
90 TSU:STA Start Condition 100 kHz mode 2(TOSC)(BRG + 1) — ns Only relevant for
Setup Time 400 kHz mode 2(TOSC)(BRG + 1) — Repeated Start
condition
1 MHz mode(1) 2(TOSC)(BRG + 1) —
91 THD:STA Start Condition 100 kHz mode 2(TOSC)(BRG + 1) — ns After this period, the
Hold Time 400 kHz mode 2(TOSC)(BRG + 1) — first clock pulse is
generated
92 TSU:STO Stop Condition 100 kHz mode 2(TOSC)(BRG + 1) — ns
Setup Time 400 kHz mode 2(TOSC)(BRG + 1) —
93 THD:STO Stop Condition 100 kHz mode 2(TOSC)(BRG + 1) — ns
Hold Time 400 kHz mode 2(TOSC)(BRG + 1) —
Note 1: Maximum pin capacitance = 10 pF for all I2C™ pins.
FIGURE 31-18: MSSP I2C™ BUS DATA TIMING
103 100 102

101
SCLx
90 106
91 107 92
SDAx
In
109 109 110
SDAx
Out

PIC18F87K90 FAMILY
TABLE 31-21: MSSP I2C™ BUS DATA REQUIREMENTS
Param.
No.
100 THIGH Clock High Time 100 kHz mode 2(TOSC)(BRG + 1) — —
400 kHz mode 2(TOSC)(BRG + 1) — —
1 MHz mode(1) 2(TOSC)(BRG + 1) — —
101 TLOW Clock Low Time 100 kHz mode 2(TOSC)(BRG + 1) — —
400 kHz mode 2(TOSC)(BRG + 1) — —
102 TR SDAx and SCLx 100 kHz mode — 1000 ns CB is specified to be from
Rise Time 400 kHz mode 20 + 0.1 CB 300 ns 10 to 400 pF
1 MHz mode(1) — 300 ns
103 TF SDAx and SCLx 100 kHz mode — 300 ns CB is specified to be from
Fall Time 400 kHz mode 20 + 0.1 CB 300 ns 10 to 400 pF
1 MHz mode(1) — 100 ns
90 TSU:STA Start Condition 100 kHz mode 2(TOSC)(BRG + 1) — — Only relevant for Repeated
Setup Time 400 kHz mode 2(TOSC)(BRG + 1) — — Start condition
91 THD:STA Start Condition 100 kHz mode 2(TOSC)(BRG + 1) — — After this period, the first
Hold Time 400 kHz mode 2(TOSC)(BRG + 1) — — clock pulse is generated
106 THD:DAT Data Input 100 kHz mode 0 — ns
Hold Time 400 kHz mode 0 0.9 s
1 MHz mode(1) — — ns
107 TSU:DAT Data Input 100 kHz mode 250 — ns (Note 2)
Setup Time 400 kHz mode 100 — ns
1 MHz mode(1) — — ns
92 TSU:STO Stop Condition 100 kHz mode 2(TOSC)(BRG + 1) — —
Setup Time 400 kHz mode 2(TOSC)(BRG + 1) — —
109 TAA Output Valid 100 kHz mode — 3500 ns
from Clock 400 kHz mode — 1000 ns
(1)
1 MHz mode — — ns
110 TBUF Bus Free Time 100 kHz mode 4.7 — s Time the bus must be free
400 kHz mode 1.3 — s before a new transmission
can start
1 MHz mode(1) — — s
D102 CB Bus Capacitive Loading — 400 pF
Note 1: Maximum pin capacitance = 10 pF for all I2C™ pins.
2: A Fast mode I2C bus device can be used in a Standard mode I2C bus system, but Parameter #107  250 ns
must then be met. This will automatically be the case if the device does not stretch the LOW period of the
SCLx signal. If such a device does stretch the LOW period of the SCLx signal, it must output the next data
bit to the SDAx line, Parameter #102 + Parameter #107 = 1000 + 250 = 1250 ns (for 100 kHz mode), before
the SCLx line is released.

PIC18F87K90 FAMILY
FIGURE 31-19: EUSART SYNCHRONOUS TRANSMISSION (MASTER/SLAVE) TIMING
TXx/CKx
Pin
121 121
RXx/DTx
Pin
120
122
TABLE 31-22: EUSART/AUSART SYNCHRONOUS TRANSMISSION REQUIREMENTS

Param
No.
120 TCKH2DTV SYNC XMIT (MASTER and SLAVE)
Clock High to Data Out Valid — 40 ns
121 TCKRF Clock Out Rise Time and Fall Time (Master mode) — 20 ns
122 TDTRF Data Out Rise Time and Fall Time — 20 ns
FIGURE 31-20: EUSART/AUSART SYNCHRONOUS RECEIVE (MASTER/SLAVE) TIMING
TXx/CKx
Pin 125
RXx/DTx
Pin
126
TABLE 31-23: EUSART/AUSART SYNCHRONOUS RECEIVE REQUIREMENTS

Param.
No.
125 TDTV2CKL SYNC RCV (MASTER and SLAVE)
Data Hold before CKx  (DTx hold time) 10 — ns
126 TCKL2DTL Data Hold after CKx  (DTx hold time) 15 — ns
TABLE 31-24: ULTRA LOW-POWER WAKE-UP SPECIFICATIONS

Standard Operating Conditions: 3.0V < VDD < 3.6V
Operating temperature -40°C  TA  +85°C (unless otherwise stated)
Param.
Symbol Characteristic Min Typ Max Units Comments
No.
Dxxx IULP Ultra Low-Power Wake-up Sink Current — 60 — nA Net of I/O leakage and current
sink at 1.6V on pin,
VDD = 3.3V

PIC18F87K90 FAMILY
TABLE 31-25: A/D CONVERTER CHARACTERISTICS: PIC18F87K90 FAMILY
(INDUSTRIAL/EXTENDED)
Param
Sym Characteristic Min Typ Max Units Conditions
No.
A01 NR Resolution — — 12 bit VREF  5.0V

A03 EIL Integral Linearity Error — ±1 ±6.0 LSB VREF = 5.0V
A04 EDL Differential Linearity Error — ±1 +3.0/-1.0 LSB VREF = 5.0V
A06 EOFF Offset Error — ±1 ±9.0 LSB VREF = 5.0V
A07 EGN Gain Error — ±1 ±8.0 LSB VREF = 5.0V
A10 — Monotonicity(1) — — — — VSS  VAIN  VREF
A20 VREF Reference Voltage Range 3 — VDD – VSS V
(VREFH – VREFL)
A21 VREFH Reference Voltage High VSS + 3.0V — VDD + 0.3V V
A22 VREFL Reference Voltage Low VSS – 0.3V — VDD – 3.0V V
A25 VAIN Analog Input Voltage VREFL — VREFH V
A30 ZAIN Recommended — — 2.5 k
Impedance of Analog
Voltage Source
A50 IREF VREF Input Current(2) — — 5 A During VAIN acquisition.
— — 150 A During A/D conversion cycle.
Note 1: The A/D conversion result never decreases with an increase in the input voltage.
2: VREFH current is from the RA3/AN3/VREF+ pin or VDD, whichever is selected as the VREFH source. VREFL current is from
the RA2/AN2/VREF-/CVREF pin or VSS, whichever is selected as the VREFL source.

PIC18F87K90 FAMILY
FIGURE 31-21: A/D CONVERSION TIMING
BSF ADCON0, GO
(Note 2)
131
Q4
130
A/D CLK 132
A/D DATA 11 10 9 ... ... 2 1 0
ADRES OLD_DATA NEW_DATA
ADIF TCY (Note 1)
GO DONE
SAMPLING STOPPED
SAMPLE
Note 1: If the A/D clock source is selected as RC, a time of TCY is added before the A/D clock starts. This allows the SLEEP instruction to
be executed.
2: This is a minimal RC delay (typically 100 ns), which also disconnects the holding capacitor from the analog input.
TABLE 31-26: A/D CONVERSION REQUIREMENTS

Param
No.
130 TAD A/D Clock Period 0.8 12.5(1) s TOSC based, VREF  3.0V
1.4 25 (1)
s VDD = 3.0V; TOSC based,
VREF full range
— 1 s A/D RC mode
— 3 s VDD = 3.0V; A/D RC mode
131 TCNV Conversion Time 14 15 TAD
(not including acquisition time)(2)
132 TACQ Acquisition Time(3) 1.4 — s -40C to +125C
135 TSWC Switching Time from Convert  Sample — (Note 4)
137 TDIS Discharge Time 0.2 — s
Note 1: The time of the A/D clock period is dependent on the device frequency and the TAD clock divider.
2: ADRES registers may be read on the following TCY cycle.
3: The time for the holding capacitor to acquire the “New” input voltage when the voltage changes full scale
after the conversion (VDD to VSS or VSS to VDD). The source impedance (RS) on the input channels is 50.
4: On the following cycle of the device clock.

PIC18F87K90 FAMILY
NOTES:

PIC18F87K90 FAMILY
32.0 PACKAGING INFORMATION
32.1 Package Marking Information
64-Lead QFN (9x9x0.9 mm) Example
XXXXXXXXXXX PIC18F67K90
XXXXXXXXXXX -I/MR e3
XXXXXXXXXXX 1110017
YYWWNNN
64-Lead TQFP Example
XXXXXXXXXX 18F67K90
XXXXXXXXXX -I/PT e3
XXXXXXXXXX 1110017
YYWWNNN
80-Lead TQFP Example
XXXXXXXXXXXX PIC18F87K90
XXXXXXXXXXXX -I/PT e3
YYWWNNN 1110017
Legend: XX...X Customer-specific information

Y Year code (last digit of calendar year)
YY Year code (last 2 digits of calendar year)
WW Week code (week of January 1 is week ‘01’)
NNN Alphanumeric traceability code
e3 Pb-free JEDEC designator for Matte Tin (Sn)
* This package is Pb-free. The Pb-free JEDEC designator ( e3)
can be found on the outer packaging for this package.
Note: In the event the full Microchip part number cannot be marked on one line, it will
be carried over to the next line, thus limiting the number of available
characters for customer-specific information.

PIC18F87K90 FAMILY
32.2 Package Details
The following sections give the technical details of the
packages.
Note: For the most current package drawings, please see the Microchip Packaging Specification located at
http://www.microchip.com/packaging

PIC18F87K90 FAMILY

PIC18F87K90 FAMILY

PIC18F87K90 FAMILY
/HDG3ODVWLF7KLQ4XDG)ODWSDFN37±[[PP%RG\PP>74)3@
1RWH )RUWKHPRVWFXUUHQWSDFNDJHGUDZLQJVSOHDVHVHHWKH0LFURFKLS3DFNDJLQJ6SHFLILFDWLRQORFDWHGDW
KWWSZZZPLFURFKLSFRPSDFNDJLQJ
D
D1
E
e
E1
N
b
NOTE 1 123 NOTE 2

α
A
c φ
A2
β A1
L L1
8QLWV 0,//,0(7(56
'LPHQVLRQ/LPLWV 0,1 120 0$;
1XPEHURI/HDGV 1
/HDG3LWFK H %6&
2YHUDOO+HLJKW $ ± ±
0ROGHG3DFNDJH7KLFNQHVV $
6WDQGRII $ ±
)RRW/HQJWK /
)RRWSULQW / 5()
)RRW$QJOH
2YHUDOO:LGWK ( %6&
2YHUDOO/HQJWK ' %6&
0ROGHG3DFNDJH:LGWK ( %6&
0ROGHG3DFNDJH/HQJWK ' %6&
/HDG7KLFNQHVV F ±
/HDG:LGWK E
0ROG'UDIW$QJOH7RS
0ROG'UDIW$QJOH%RWWRP
1RWHV
3LQYLVXDOLQGH[IHDWXUHPD\YDU\EXWPXVWEHORFDWHGZLWKLQWKHKDWFKHGDUHD
&KDPIHUVDWFRUQHUVDUHRSWLRQDOVL]HPD\YDU\
'LPHQVLRQV'DQG(GRQRWLQFOXGHPROGIODVKRUSURWUXVLRQV0ROGIODVKRUSURWUXVLRQVVKDOOQRWH[FHHGPPSHUVLGH
'LPHQVLRQLQJDQGWROHUDQFLQJSHU$60(<0
%6& %DVLF'LPHQVLRQ7KHRUHWLFDOO\H[DFWYDOXHVKRZQZLWKRXWWROHUDQFHV
5() 5HIHUHQFH'LPHQVLRQXVXDOO\ZLWKRXWWROHUDQFHIRULQIRUPDWLRQSXUSRVHVRQO\
0LFURFKLS 7HFKQRORJ\ 'UDZLQJ &%

PIC18F87K90 FAMILY

PIC18F87K90 FAMILY
D
D1
E
e
E1
b N
NOTE 1
12 3 NOTE 2 α
A
c
φ
β A2
A1
L L1
8QLWV 0,//,0(7(56
'LPHQVLRQ/LPLWV 0,1 120 0$;
1XPEHURI/HDGV 1
/HDG3LWFK H %6&
2YHUDOO+HLJKW $ ± ±
0ROGHG3DFNDJH7KLFNQHVV $
6WDQGRII $ ±
)RRW/HQJWK /
)RRWSULQW / 5()
)RRW$QJOH
2YHUDOO:LGWK ( %6&
2YHUDOO/HQJWK ' %6&
0ROGHG3DFNDJH:LGWK ( %6&
0ROGHG3DFNDJH/HQJWK ' %6&
/HDG7KLFNQHVV F ±
/HDG:LGWK E
0ROG'UDIW$QJOH7RS
0ROG'UDIW$QJOH%RWWRP
1RWHV
3LQYLVXDOLQGH[IHDWXUHPD\YDU\EXWPXVWEHORFDWHGZLWKLQWKHKDWFKHGDUHD
&KDPIHUVDWFRUQHUVDUHRSWLRQDOVL]HPD\YDU\
'LPHQVLRQV'DQG(GRQRWLQFOXGHPROGIODVKRUSURWUXVLRQV0ROGIODVKRUSURWUXVLRQVVKDOOQRWH[FHHGPPSHUVLGH
'LPHQVLRQLQJDQGWROHUDQFLQJSHU$60(<0
%6& %DVLF'LPHQVLRQ7KHRUHWLFDOO\H[DFWYDOXHVKRZQZLWKRXWWROHUDQFHV
5() 5HIHUHQFH'LPHQVLRQXVXDOO\ZLWKRXWWROHUDQFHIRULQIRUPDWLRQSXUSRVHVRQO\
0LFURFKLS 7HFKQRORJ\ 'UDZLQJ &%

PIC18F87K90 FAMILY

PIC18F87K90 FAMILY
APPENDIX A: REVISION HISTORY APPENDIX B: MIGRATION FROM
PIC18F85J90 AND
Revision A (September 2009)
PIC18F87J90 TO
Original data sheet for PIC18F87K90 family devices. PIC18F87K90
Revision B (April 2010) Devices in the PIC18F87K90, PIC18F85J90 and
PIC18F87J90 families are almost similar in their func-
Changes to Section 32.0 “Packaging Information”, tions and features. Code can be migrated from the
including new packaging diagrams. Changes to some 18F85J90 to the PIC18F87K90 without many changes.
of the values in Section 31.0 “Electrical Characteris- The differences between the two device families are
tics”. The new Section 2.0 “Guidelines for Getting listed in Table B-1.
Started with PIC18FXXKXX Microcontrollers” has
been added. Minor text edits throughout the document.
Revision C (March 2011)

Updated notes for clamping diodes, updated D080,
D090, D121, D131 and D310. Also, updated the abso-
lute maximum specification for the I/O pin and the
maximum specification for the input/output clamp cur-
rent. The 64-lead QFN packaging diagram was updated.
Revision D (July 2011)

Updated the specification values in Section 31.0
“Electrical Characteristics”. Minor text edits
throughout the document.

PIC18F87K90 FAMILY
TABLE B-1: NOTABLE DIFFERENCES BETWEEN PIC18F87K90, PIC18F87J90 AND PIC18F85J90
FAMILIES
Characteristic PIC18F87K90 Family PIC18F87J90 Family PIC18F85J90 Family
Max Operating Frequency 64 MHz 48 MHz 40 MHz
Max Program Memory 128 Kbytes 128 Kbytes 32 Kbytes
Data Memory 4 Kbytes 4 Kbytes 2 Kbytes
Program Memory Endurance 10,000 Write/Erase 10,000 Write/Erase 1,000 Write/Erase
(minimum) (minimum) (minimum)
Single-Word Write for Flash Yes Yes No
Oscillator Options PLL can be used with PLL cannot be used with
Yes
INTOSC INTOSC
CTMU Yes Yes No
RTCC Yes Yes No
SOSC Oscillator Options Low-power oscillator option Low-power oscillator option
No
for SOSC for SOSC
TICKI Clock T1CKI can be used as a
clock without enabling the No No
SOSC oscillator
INTOSC Up to 16 MHz 8 MHz 8 MHz
SPI/I2C™ 2 1 1
Timers 11 4 4
ECCP 3 No No
CCP 7 2 2
Data EEPROM Yes No No
Programmable BOR Multiple level of BOR No No
WDT Prescale Options 22 16 16
5V Operation Yes No No
nanoWatt XLP Yes No No
Regulator Yes Yes Yes
Low-Power BOR Yes No No
ADC 24-Channel 12-Channel 12-Channel
Differential (12-bit) Not differential (10-bit) Not Differential (10-bit)
Internal Temperature Sensor Yes No No
Programmable HLVD Yes No No
EUSART 2 EUSARTs 1 EUSART, 1 AUSART 1 EUSART, 1 AUSART
Comparators 3 2 2
Oscillator Options 14 options by OSC<3:0> 8 options by OSC<3:0> 8 options by OSC<3:0>
Ultra Low-Power Wake-up
Yes No No
(ULPW)
Power-up Timer Yes No No
MCLR Pin as Input Port Yes No No
LCD Charge Pump No Yes Yes
Internal Resistor Ladder for
Yes No No
Biasing

PIC18F87K90 FAMILY
INDEX
A CTMU Current Source Calibration Circuit ................ 413
CTMU Typical Connections and Internal
A/D .................................................................................... 373
A/D Converter Interrupt, Configuring ........................ 383 Configuration for Pulse Delay Generation ........ 421
Acquisition Requirements ......................................... 384 CTMU Typical Connections and Internal
Configuration for Time Measurement ............... 420
ADRESH Register..................................................... 381
Analog Port Pins, Configuring................................... 385 Device Clock............................................................... 42
Associated Registers ................................................ 388 Enhanced PWM Mode.............................................. 257
EUSART Receive ..................................................... 362
Automatic Acquisition Time....................................... 385
Configuring the Module............................................. 383 EUSART Transmit .................................................... 359
Conversion Clock (TAD) ............................................ 385 External Clock Input (EC) ........................................... 48
External Clock Input (HS) ........................................... 48
Conversion Requirements ........................................ 543
Conversion Status (GO/DONE Bit) ........................... 381 External Power-on Reset Circuit
Conversions .............................................................. 386 (Slow VDD Power-up) ......................................... 71
Fail-Safe Clock Monitor (FSCM)............................... 445
Converter Characteristics ......................................... 542
Differential................................................................. 373 Full-Bridge Application.............................................. 261
Operation in Power-Managed Modes ....................... 387 Generic I/O Port Operation....................................... 153
Half-Bridge Applications ................................... 260, 267
Use of the Special Event Triggers ............................ 387
Absolute Maximum Ratings .............................................. 505 High/Low-Voltage Detect with External Input ........... 402
AC (Timing) Characteristics .............................................. 523 Internal Reference and Contrast Control.................. 284
Interrupt Logic........................................................... 130
Load Conditions for Device Timing
Specifications.................................................... 524 INTIO1 Oscillator Mode .............................................. 49
Parameter Symbology .............................................. 523 INTIO2 Oscillator Mode .............................................. 49
LCD Bias Internal Resistor Ladder Connection........ 282
Timing Conditions ..................................................... 524
ACKSTAT ......................................................................... 339 LCD Clock Generation.............................................. 280
ACKSTAT Status Flag ...................................................... 339 LCD Driver Module ................................................... 273
MSSPx (I2C Master Mode) ....................................... 333
ADCON0 Register
GO/DONE Bit............................................................ 381 MSSPx (I2C Mode) ................................................... 313
ADDFSR ........................................................................... 494 MSSPx (SPI Mode) .................................................. 303
On-Chip Reset Circuit................................................. 69
ADDLW ............................................................................. 457
ADDULNK......................................................................... 494 Open-Drain Output (USART Example)..................... 154
ADDWF ............................................................................. 457 PIC18F6XK90............................................................. 12
PIC18F8XK90............................................................. 13
ADDWFC .......................................................................... 458
ADRESL Register ............................................................. 381 PLL ............................................................................. 49
Analog and Digital Ports ................................................... 156 PWM Operation (Simplified) ..................................... 247
RC Oscillator Mode .................................................... 48
Analog-to-Digital Converter. See A/D.
ANDLW ............................................................................. 458 RCIO Oscillator Mode................................................. 48
ANDWF ............................................................................. 459 Reads from Flash Program Memory ........................ 115
RTCC........................................................................ 217
Assembler
MPASM Assembler................................................... 502 Simplified Steering.................................................... 270
Single Comparator.................................................... 392
B SOSC External Components.................................... 191
Baud Rate Generator ........................................................ 335 SPI Master/Slave Connection................................... 307
BC ..................................................................................... 459 Table Read Operation .............................................. 111
BCF ................................................................................... 460 Table Write Operation .............................................. 112
BF ..................................................................................... 339 Table Writes to Flash Program Memory................... 117
BF Status Flag .................................................................. 339 Timer0 in 16-Bit Mode .............................................. 184
Block Diagrams Timer0 in 8-Bit Mode ................................................ 184
A/D ............................................................................ 382 Timer1 ...................................................................... 190
A/D Differential Channel Measurement .................... 373 Timer2 ...................................................................... 200
Analog Input Model ................................................... 383 Timer3/5/7 ................................................................ 205
Baud Rate Generator................................................ 335 Timer4/6/8/10/12 ...................................................... 215
Capture Mode Operation .................................. 243, 255 Ultra Low-Power Wake-up Initialization...................... 66
Clock Source Multiplexing......................................... 230 Watchdog Timer ....................................................... 440
Comparator Analog Input Model ............................... 392 BN..................................................................................... 460
Comparator Configurations....................................... 394 BNC .................................................................................. 461
Comparator Output ................................................... 389 BNN .................................................................................. 461
Comparator Voltage Reference ................................ 398 BNOV ............................................................................... 462
Comparator Voltage Reference Output BNZ .................................................................................. 462
Buffer Example ................................................. 399 BOR. See Brown-out Reset.
Compare Mode Operation ................................ 245, 256 BOV .................................................................................. 465
Connections for On-Chip Voltage Regulator............. 442 BRA .................................................................................. 463
Crystal/Ceramic Resonator (HS, HSPLL) ................... 47 BRG. See Baud Rate Generator.
CTMU........................................................................ 407

PIC18F87K90 FAMILY
Brown-out Reset (BOR) ...................................................... 71 How to Clear RAM (Bank 1) Using Indirect
Detecting ..................................................................... 71 Addressing........................................................ 104
BSF ................................................................................... 463 Initializing PORTA..................................................... 157
BTFSC .............................................................................. 464 Initializing PORTB..................................................... 160
BTFSS............................................................................... 464 Initializing PORTC .................................................... 163
BTG................................................................................... 465 Initializing PORTD .................................................... 166
BZ...................................................................................... 466 Initializing PORTE..................................................... 169
Initializing PORTF..................................................... 172
C Initializing PORTG .................................................... 175
C Compilers Initializing PORTH .................................................... 177
MPLAB C18 .............................................................. 502 Initializing PORTJ ..................................................... 180
CALL ................................................................................. 466 Loading the SSP1BUF (SSP1SR) Register.............. 306
CALLW.............................................................................. 495 Reading a Flash Program Memory Word ................. 115
Capture (CCP Module)...................................................... 242 Routine for Capacitive Touch Switch........................ 419
Associated Registers ................................................ 246 Routine for Temperature Measurement Using
CCP Pin Configuration .............................................. 242 Internal Diode ................................................... 422
CCPR4H:CCPR4L Registers.................................... 242 Saving STATUS, WREG and BSR
Prescaler ................................................................... 243 Registers in RAM.............................................. 151
Software Interrupt ..................................................... 243 Setting the RTCWREN Bit ........................................ 231
Timer1/3/5/7 Mode Selection .................................... 242 Setup for CTMU Calibration Routines ...................... 414
Capture (ECCP Module) ................................................... 255 Ultra Low-Power Wake-up Initialization ...................... 66
CCPRxH:CCPRxL Registers .................................... 255 Writing to Flash Program Memory .................... 118–119
ECCP Pin Configuration ........................................... 255 Code Protection ........................................................ 425, 447
Prescaler ................................................................... 255 Associated Registers ................................................ 448
Software Interrupt ..................................................... 255 Data EEPROM.......................................................... 450
Timer1/Timer3 Mode Selection ................................. 255 Program Memory ...................................................... 448
Capture/Compare/PWM (CCP)......................................... 237 COMF ............................................................................... 468
Capture Mode. See Capture. Comparator....................................................................... 389
CCP6/7/8/9 Pin Assignment ..................................... 242 Analog Input Connection Considerations ................. 392
CCPRxH Register ..................................................... 241 Associated Registers ................................................ 396
CCPRxL Register...................................................... 241 Control and Configuration......................................... 393
Compare Mode. See Compare. Effects of a Reset ..................................................... 396
Configuration............................................................. 241 Enable and Input Selection....................................... 393
Mode and Timer Resources...................................... 241 Enable and Output Selection .................................... 393
Open-Drain Output Option ........................................ 242 Interrupts .................................................................. 395
Charge Time Measurement Unit (CTMU) ......................... 407 Operation .................................................................. 392
Associated Registers ................................................ 423 Operation During Sleep ............................................ 396
Calibrating the Module .............................................. 412 Response Time......................................................... 392
Creating a Delay ....................................................... 421 Comparator Voltage Reference ........................................ 397
Effects of a Reset...................................................... 423 Accuracy and Error ................................................... 398
Measuring Capacitance with the CTMU ................... 418 Associated Registers ................................................ 399
Measuring Temperature with Use ............................. 422 Configuring ............................................................... 397
Measuring Time ........................................................ 420 Connection Considerations....................................... 398
Module Initialization .................................................. 412 Effects of a Reset ..................................................... 398
Operation .................................................................. 411 Operation During Sleep ............................................ 398
During Sleep and Idle Modes............................ 423 Compare (CCP Module) ................................................... 244
Clock Sources Associated Registers ................................................ 246
Default System Clock on Reset .................................. 46 CCPR4 Register ....................................................... 244
Selection Using OSCCON Register ............................ 46 Pin Configuration ...................................................... 244
CLRF................................................................................. 467 Software Interrupt ..................................................... 244
CLRWDT........................................................................... 467 Special Event Trigger ............................................... 244
Code Examples Timer1/3/5/7 Mode Selection.................................... 244
16 x 16 Signed Multiply Routine ............................... 128 Compare (ECCP Module)................................................. 256
16 x 16 Unsigned Multiply Routine ........................... 128 CCPRx Register ....................................................... 256
8 x 8 Signed Multiply Routine ................................... 127 Pin Configuration ...................................................... 256
8 x 8 Unsigned Multiply Routine ............................... 127 Software Interrupt ..................................................... 256
Capacitance Calibration Routine .............................. 417 Special Event Trigger ....................................... 210, 256
Changing Between Capture Prescalers ............ 243, 255 Timer1/Timer3 Mode Selection................................. 256
Computed GOTO Using an Offset Value .................... 89 Computed GOTO................................................................ 89
Current Calibration Routine ...................................... 415 Configuration Bits ............................................................. 425
Data EEPROM Read ................................................ 123 Configuration Mismatch (CM) Reset................................... 72
Data EEPROM Refresh Routine ............................... 124 Configuration Register Protection..................................... 450
Data EEPROM Write ................................................ 123
Erasing a Flash Program Memory Row .................... 116
Fast Register Stack..................................................... 89

PIC18F87K90 FAMILY
Core Features E
Easy Migration .............................................................. 9 Effect on Standard PIC18 Instructions.............................. 498
Extended Instruction Set............................................... 9 Effects of Power-Managed Modes on Various
Memory Options............................................................ 9
Clock Sources ............................................................ 52
nanoWatt Technology ................................................... 9 Electrical Characteristics .................................................. 505
Oscillator Options and Features ................................... 9 Enhanced Capture/Compare/PWM (ECCP)..................... 251
CPFSEQ ........................................................................... 468
Capture Mode. See Capture.
CPFSGT ........................................................................... 469 Compare Mode. See Compare.
CPFSLT ............................................................................ 469 ECCP Mode and Timer Resources .......................... 254
Crystal Oscillator/Ceramic Resonators (HS)....................... 46
Enhanced PWM Mode.............................................. 257
Customer Change Notification Service ............................. 566 Auto-Restart ..................................................... 266
Customer Notification Service........................................... 566 Auto-Shutdown ................................................. 264
Customer Support............................................................. 566
Direction Change in Full-Bridge
D Output Mode............................................. 263
Full-Bridge Application...................................... 261
Data Addressing Modes.................................................... 104
Full-Bridge Mode .............................................. 261
Comparing Addressing Modes (Bit-Oriented,
Half-Bridge Application ..................................... 260
Byte Oriented) with the Extended
Half-Bridge Application Examples .................... 267
Instruction Set Enabled..................................... 108
Half-Bridge Mode.............................................. 260
Direct......................................................................... 104
Output Relationships (Active-High and
Indexed Literal Offset
Active-Low)............................................... 258
BSR .................................................................. 109
Output Relationships Diagram.......................... 259
Mapping Access Bank ...................................... 109
Programmable Dead-Band Delay..................... 267
Indirect ...................................................................... 104
Shoot-Through Current..................................... 267
Inherent and Literal ................................................... 104
Start-up Considerations.................................... 264
Data EEPROM Memory
Outputs and Configuration........................................ 254
Associated Registers ................................................ 125
Enhanced Capture/Compare/PWM (ECCP1/2/3)
EEADR and EEADRH Registers .............................. 121
Associated Registers................................................ 272
EECON1 and EECON2 Registers ............................ 121
Enhanced Universal Synchronous Asynchronous Receiver
Operation During Code-Protect ................................ 124
Transmitter (EUSART). See EUSART.
Overview ................................................................... 121
Equations
Protection, Spurious Writes ...................................... 124
16 x 16 Signed Multiplication Algorithm.................... 128
Reading..................................................................... 123
16 x 16 Unsigned Multiplication Algorithm................ 128
Usage........................................................................ 124
A/D Acquisition Time ................................................ 384
Write Verify ............................................................... 123
A/D Minimum Charging Time ................................... 384
Writing To.................................................................. 123
Calculating the Minimum Required
Data Memory ...................................................................... 92
Acquisition Time ............................................... 384
Access Bank ............................................................... 94
Converting Error Pulses ........................................... 232
Bank Select Register (BSR)........................................ 92
EUSART
Extended Instruction Set........................................... 107
Asynchronous Mode................................................. 359
General Purpose Registers......................................... 94
12-Bit Break Transmit and Receive
Memory Maps
Sequence ................................................. 366
PIC18FX5K90/X7K90 Devices ........................... 93
Associated Registers, Receive......................... 363
Special Function Registers ................................. 95
Associated Registers, Transmit........................ 361
Special Function Registers ......................................... 95
Auto-Wake-up on Sync Break .......................... 364
Data Memory Modes
Receiver ........................................................... 362
Indexed Literal Offset................................................ 107
Setting Up 9-Bit Mode with Address Detect ..... 362
Affected Instructions ......................................... 107
Transmitter ....................................................... 359
DAW.................................................................................. 470
Baud Rate Generator
DC Characteristics
Operation in Power-Managed Modes............... 353
CTMU Current Source .............................................. 520
Baud Rate Generator (BRG) .................................... 353
PIC18F87K90 Family, Industrial ............................... 519
Associated Registers........................................ 354
Power-Down and Supply Current ............................. 508
Auto-Baud Rate Detect..................................... 357
Supply Voltage.......................................................... 507
Baud Rate Error, Calculating............................ 354
DCFSNZ ........................................................................... 471
Baud Rates, Asynchronous Modes .................. 355
DECF ................................................................................ 470
High Baud Rate Select (BRGH Bit) .................. 353
DECFSZ............................................................................ 471
Sampling .......................................................... 353
Default System Clock.......................................................... 46
Synchronous Master Mode....................................... 367
Details on Individual Family Members ................................ 10
Associated Registers, Receive......................... 370
Development Support ....................................................... 501
Associated Registers, Transmit........................ 368
Device Overview ................................................................... 9
Reception ......................................................... 369
Features (64-Pin Devices) .......................................... 11
Transmission .................................................... 367
Features (80-Pin Devices) .......................................... 11
Direct Addressing.............................................................. 105

PIC18F87K90 FAMILY
Synchronous Slave Mode ......................................... 371 Setup ........................................................................ 403
Associated Registers, Receive ......................... 372 Start-up Time ............................................................ 403
Associated Registers, Transmit ........................ 371 Typical Low-Voltage Detect Application ................... 404
Reception.......................................................... 372 HLVD. See High/Low-Voltage Detect.
Transmission..................................................... 371
Extended Instruction Set I
ADDFSR ................................................................... 494 I/O Ports............................................................................ 153
ADDULNK................................................................. 494 Open-Drain Outputs.................................................. 154
CALLW...................................................................... 495 Output Pin Drive ....................................................... 153
MOVSF ..................................................................... 495 Pin Capabilities ......................................................... 153
MOVSS ..................................................................... 496 Pull-up Configuration ................................................ 153
PUSHL ...................................................................... 496 I2C Mode (MSSP)
SUBFSR ................................................................... 497 Acknowledge Sequence Timing ............................... 342
SUBULNK ................................................................. 497 Associated Registers ................................................ 348
External Oscillator Modes Baud Rate Generator ............................................... 335
Clock Input (EC, ECPLL Modes) ................................ 48 Bus Collision
HS, HSPLL.................................................................. 46 During a Repeated Start Condition................... 346
During a Stop Condition ................................... 347
F Clock Arbitration ....................................................... 336
Fail-Safe Clock Monitor............................................. 425, 445 Clock Stretching........................................................ 328
Exiting Operation ...................................................... 445 10-Bit Slave Receive Mode (SEN = 1) ............. 328
Interrupts in Power-Managed Modes ........................ 446 10-Bit Slave Transmit Mode ............................. 328
POR or Wake from Sleep ......................................... 446 7-Bit Slave Receive Mode (SEN = 1) ............... 328
WDT During Oscillator Failure .................................. 445 7-Bit Slave Transmit Mode ............................... 328
Fast Register Stack............................................................. 89 Clock Synchronization and the CKP bit .................... 329
Firmware Instructions........................................................ 451 Effects of a Reset ..................................................... 343
Flash Program Memory..................................................... 111 General Call Address Support .................................. 332
Associated Registers ................................................ 119 I2C Clock Rate w/BRG.............................................. 335
Control Registers ...................................................... 112 Master Mode............................................................. 333
EECON1, EECON2 .......................................... 112 Operation.......................................................... 334
TABLAT (Table Latch) ...................................... 114 Reception ......................................................... 339
TBLPTR (Table Pointer) ................................... 114 Repeated Start Condition Timing ..................... 338
Erasing ...................................................................... 116 Start Condition Timing ...................................... 337
Sequence.......................................................... 116 Transmission .................................................... 339
Operation During Code-Protect ................................ 119 Multi-Master Communication, Bus Collision
Protection Against Spurious Writes .......................... 119 and Arbitration .................................................. 343
Reading..................................................................... 115 Multi-Master Mode .................................................... 343
Table Pointer Operation .................................................................. 318
Boundaries Based on Operation....................... 114 Read/Write Bit Information (R/W Bit) ................ 318, 321
Table Pointer Boundaries ......................................... 114 Registers .................................................................. 313
Table Reads, Table Writes ....................................... 111 Serial Clock (SCLx) .................................................. 321
Writing Slave Mode............................................................... 318
Unexpected Termination................................... 119 Address Masking Modes
Verify................................................................. 119 5-Bit .......................................................... 319
Writing To.................................................................. 117 7-Bit .......................................................... 320
Sequence.......................................................... 117 Addressing........................................................ 318
FSCM. See Fail-Safe Clock Monitor. Reception ......................................................... 321
Transmission .................................................... 321
G Sleep Operation........................................................ 343
GOTO................................................................................ 472 Stop Condition Timing .............................................. 342
ID Locations.............................................................. 425, 450
H INCF ................................................................................. 472
Hardware Multiplier ........................................................... 127 INCFSZ............................................................................. 473
8 x 8 Multiplication Algorithms .................................. 127 In-Circuit Debugger........................................................... 450
Operation .................................................................. 127 In-Circuit Serial Programming (ICSP)....................... 425, 450
Performance Comparison (table) .............................. 127 Indexed Literal Offset Addressing
High/Low-Voltage Detect .................................................. 401 and Standard PIC18 Instructions.............................. 498
Applications............................................................... 404 Indexed Literal Offset Mode.............................................. 498
Associated Registers ................................................ 405 Indirect Addressing ........................................................... 105
Current Consumption ................................................ 403 INFSNZ............................................................................. 473
Effects of a Reset...................................................... 405 Instruction Cycle ................................................................. 90
Operation .................................................................. 402 Clocking Scheme........................................................ 90
During Sleep ..................................................... 405 Flow/Pipelining............................................................ 90

PIC18F87K90 FAMILY
Instruction Set ................................................................... 451 RRNCF ..................................................................... 485
ADDLW ..................................................................... 457 SETF ........................................................................ 485
ADDWF..................................................................... 457 SETF (Indexed Literal Offset Mode) ......................... 499
ADDWF (Indexed Literal Offset Mode) ..................... 499 SLEEP ...................................................................... 486
ADDWFC .................................................................. 458 Standard Instructions................................................ 451
ANDLW ..................................................................... 458 SUBFWB .................................................................. 486
ANDWF..................................................................... 459 SUBLW..................................................................... 487
BC ............................................................................. 459 SUBWF..................................................................... 487
BCF........................................................................... 460 SUBWFB .................................................................. 488
BN ............................................................................. 460 SWAPF..................................................................... 488
BNC .......................................................................... 461 TBLRD...................................................................... 489
BNN .......................................................................... 461 TBLWT ..................................................................... 490
BNOV........................................................................ 462 TSTFSZ .................................................................... 491
BNZ........................................................................... 462 XORLW .................................................................... 491
BOV .......................................................................... 465 XORWF .................................................................... 492
BRA........................................................................... 463 INTCON Register
BSF ........................................................................... 463 RBIF Bit .................................................................... 160
BSF (Indexed Literal Offset Mode) ........................... 499 Inter-Integrated Circuit. See I2C.
BTFSC ...................................................................... 464 Internal Oscillator Block ...................................................... 49
BTFSS ...................................................................... 464 Adjustment.................................................................. 50
BTG........................................................................... 465 INTIO Modes .............................................................. 49
BZ ............................................................................. 466 INTOSC Frequency Drift ............................................ 50
CALL ......................................................................... 466 INTOSC Output Frequency ........................................ 50
CLRF......................................................................... 467 INTPLL Modes............................................................ 50
CLRWDT................................................................... 467 Internal RC Oscillator
COMF ....................................................................... 468 Use with WDT........................................................... 440
CPFSEQ ................................................................... 468 Internet Address ............................................................... 566
CPFSGT ................................................................... 469 Interrupt Sources .............................................................. 425
CPFSLT .................................................................... 469 A/D Conversion Complete ........................................ 383
DAW.......................................................................... 470 Capture Complete (CCP) ......................................... 243
DCFSNZ ................................................................... 471 Capture Complete (ECCP) ....................................... 255
DECF ........................................................................ 470 Compare Complete (CCP) ....................................... 244
DECFSZ.................................................................... 471 Compare Complete (ECCP) ..................................... 256
Extended Instructions ............................................... 493 Interrupt-on-Change (RB7:RB4)............................... 160
Considerations when Enabling ......................... 498 TMR0 Overflow......................................................... 185
Syntax............................................................... 493 TMR1 Overflow......................................................... 192
Use with MPLAB IDE Tools .............................. 500 TMR2 to PR2 Match (PWM)..................................... 248
General Format......................................................... 453 TMRx Overflow................................................. 201, 210
GOTO ....................................................................... 472 TMRx to PRx Match ................................................. 215
INCF.......................................................................... 472 TMRx to PRx Match (PWM) ..................................... 213
INCFSZ ..................................................................... 473 Interrupts .......................................................................... 129
INFSNZ ..................................................................... 473 Associated Registers................................................ 152
IORLW ...................................................................... 474 During, Context Saving............................................. 151
IORWF ...................................................................... 474 INTx Pin.................................................................... 151
LFSR......................................................................... 475 PORTB, Interrupt-on-Change................................... 151
MOVF........................................................................ 475 TMR0........................................................................ 151
MOVFF ..................................................................... 476 Interrupts, Flag Bits
MOVLB ..................................................................... 476 Interrupt-on-Change (RB7:RB4) Flag (RBIF Bit) ...... 160
MOVLW .................................................................... 477 TMR12 to PR12 Match Flag (TMR12IF Bit) ............. 138
MOVWF .................................................................... 477 INTOSC, INTRC. See Internal Oscillator Block.
MULLW ..................................................................... 478 IORLW .............................................................................. 474
MULWF..................................................................... 478 IORWF.............................................................................. 474
NEGF ........................................................................ 479
NOP .......................................................................... 479 L
Opcode Field Descriptions........................................ 452 LCD
POP .......................................................................... 480 Associated Registers................................................ 301
PUSH ........................................................................ 480 Bias Types................................................................ 281
RCALL ...................................................................... 481 Clock Source Selection ............................................ 280
RESET ...................................................................... 481 Configuring the Module ............................................ 300
RETFIE ..................................................................... 482 Frame Frequency ..................................................... 286
RETLW ..................................................................... 482 Interrupts .................................................................. 298
RETURN ................................................................... 483 LCDCON Register .................................................... 274
RLCF......................................................................... 483 LCDDATAx Registers............................................... 274
RLNCF ...................................................................... 484 LCDPS Register ....................................................... 274
RRCF ........................................................................ 484 LCDSEx Registers.................................................... 274
Multiplex Types......................................................... 285

PIC18F87K90 FAMILY
Operation During Sleep ............................................ 299 N
Pixel Control.............................................................. 285 NEGF................................................................................ 479
Prescaler ................................................................... 280 NOP .................................................................................. 479
Segment Enables...................................................... 285
Notable Differences Between PIC18F87K90,
Waveform Generation ............................................... 286 PIC18F87J90 and PIC18F85J90 Families ............... 554
LCD Driver .......................................................................... 10
LCDCON Register............................................................. 274 O
LCDDATAx Registers ....................................................... 274 On-Chip Regulator............................................................ 425
LCDPS Register................................................................ 274 On-Chip Voltage Regulator............................................... 442
LP3:LP0 Bits ............................................................. 280 Enable/Disable by Hardware .................................... 442
LCDSEx Registers ............................................................ 274 Oscillator Configuration ...................................................... 41
LFSR ................................................................................. 475 Clock Sources, Switching ........................................... 45
Liquid Crystal Display (LCD) Driver .................................. 273 EC............................................................................... 41
M ECIO ........................................................................... 41
HS............................................................................... 41
Master Clear (MCLR) .......................................................... 71 Internal Oscillator Block .............................................. 49
Master Synchronous Serial Port (MSSP). See MSSP.
INTIO1 ........................................................................ 41
Memory Organization.......................................................... 85 INTIO2 ........................................................................ 41
Data Memory .............................................................. 92 LP ............................................................................... 41
Program Memory
RC .............................................................................. 41
Memory Maps ..................................................... 85 RCIO........................................................................... 41
Microchip Internet Web Site .............................................. 566 Transitions .................................................................. 46
Migration From PIC18F85J90 and
XT ............................................................................... 41
PIC18F87J90 to PIC18F87K90 ................................ 553 Oscillator Selection ........................................................... 425
MOVF................................................................................ 475 Oscillator Start-up Timer (OST) .......................................... 52
MOVFF.............................................................................. 476
Oscillator, Timer1.............................................................. 187
MOVLB.............................................................................. 476
MOVLW............................................................................. 477 P
MOVSF ............................................................................. 495 P1A/P1B/P1C/P1D.See Enhanced
MOVSS ............................................................................. 496 Capture/Compare/PWM (ECCP). ............................. 257
MOVWF ............................................................................ 477 Packaging ......................................................................... 545
MPLAB ASM30 Assembler, Linker, Librarian ................... 502 Details....................................................................... 546
MPLAB Integrated Development Environment Marking ..................................................................... 545
Software .................................................................... 501 Pin Functions
MPLAB PM3 Device Programmer..................................... 504 AVDD ..................................................................... 22, 33
MPLAB REAL ICE In-Circuit Emulator System................. 503 AVSS ..................................................................... 22, 33
MPLINK Object Linker/MPLIB Object Librarian ................ 502 ENVREG .............................................................. 22, 33
MSSP MCLR/RG5 ........................................................... 14, 22
ACK Pulse......................................................... 318, 321 OSC1/CLKI/RA7 ................................................... 14, 22
I2C Mode. See I2C Mode. OSC2/CLKO/RA6 ................................................. 14, 22
Module Overview ...................................................... 303 RA0/AN0/ULPWU................................................. 15, 23
SPI Mode .................................................................. 303 RA1/AN1/SEG18 .................................................. 15, 23
Associated Registers ........................................ 312 RA2/AN2/VREF- .................................................... 15, 23
Bus Mode Compatibility .................................... 311 RA3/AN3/REF+............................................................ 15
Clock Speed, Interactions ................................. 311 RA3/AN3/VREF+ ......................................................... 23
Effects of a Reset.............................................. 311 RA4/T0CKI/SEG14 ............................................... 15, 23
Enabling SPI I/O ............................................... 307 RA5/AN4/SEG15/T1CKI/T3G/HLVDIN................. 15, 23
Master Mode ..................................................... 308 RB0/INT0/SEG30/FLT0 .............................................. 24
Operation .......................................................... 306 RB0/INT0/SEG30/FLTO ............................................. 16
Operation in Power-Managed Modes ............... 311 RB1/INT1/SEG8 ................................................... 16, 24
Serial Clock....................................................... 303 RB2/INT2/SEG9/CTED1....................................... 16, 24
Serial Data Out ................................................. 303 RB3/INT3/SEG10/CTED2/ECCP2/P2A................ 16, 24
Slave Mode ....................................................... 309 RB4/KBI0/SEG11 ................................................. 16, 24
Slave Select ...................................................... 303 RB5/KBI1/SEG29/T3CKI/T1G .............................. 16, 24
Slave Select Synchronization ........................... 309 RB6/KBI2/PGC ..................................................... 16, 25
SPI Clock .......................................................... 308 RB7/KBI3/PGD ..................................................... 16, 25
SSPxBUF Register ........................................... 308 RC0/SOSCO/SCKLI ................................................... 26
SSPxSR Register.............................................. 308 RC0/SOSCO/SCLKI ................................................... 17
Typical Connection ........................................... 307 RC1/SOSCI/ECCP2/P2A/SEG32 ............................... 17
SPI Mode Serial Data In ........................................... 303 RC1/SOSCI/ECCP2/SEG32/P2A ............................... 26
MULLW ............................................................................. 478 RC2/ECCP1/P1A/SEG13 ..................................... 17, 26
MULWF ............................................................................. 478

PIC18F87K90 FAMILY
RC3/SCK1/SCL1/SEG17...................................... 17, 26 POP .................................................................................. 480
RC4/SDI1/SDA1/SEG16....................................... 17, 26 POR. See Power-on Reset.
RC5/SDO1/SEG12 ............................................... 17, 26 PORTA
RC6/TX1/CK1/SEG27........................................... 17, 26 Associated Registers................................................ 159
RC7/RX1/DT1/SEG28 .......................................... 17, 26 LATA Register .......................................................... 157
RD0/SEG0/CTPLS................................................ 18, 27 PORTA Register....................................................... 157
RD1/SEG1/T5CKI/T7G ......................................... 18, 27 TRISA Register......................................................... 157
RD2/SEG2 ............................................................ 18, 27 PORTB
RD3/SEG3 ............................................................ 18, 27 Associated Registers................................................ 162
RD4/SEG4/SDO2 ................................................. 18, 27 LATB Register .......................................................... 160
RD5/SEG5/SDI2/SDA2......................................... 18, 27 PORTB Register....................................................... 160
RD6/SEG6/SCK2/SCL2........................................ 18, 27 RB7:RB4 Interrupt-on-Change Flag (RBIF Bit) ........ 160
RD7/SEG7/SS2 .................................................... 18, 27 TRISB Register......................................................... 160
RE0/LCDBIAS1/P2D............................................. 19, 28 PORTC
RE1/LCDBIAS2/P2C............................................. 19, 28 Associated Registers................................................ 165
RE2/LCDBIAS3/P2B/CCP10 ................................ 19, 28 LATC Register .......................................................... 163
RE3/COM0/P3C/CCP9/REFO .............................. 19, 28 PORTC Register....................................................... 163
RE4/COM1/P3B/CCP8 ......................................... 19, 28 TRISC Register ........................................................ 163
RE5/COM2/P1C/CCP7 ......................................... 19, 28 PORTD
RE6/COM3/P1B/CCP6 ......................................... 19, 28 Associated Registers................................................ 168
RE7/ECCP2/P2A/SEG31............................................ 28 LATD Register .......................................................... 166
RE7/ECCP2/SEG31/P2A............................................ 19 PORTD Register....................................................... 166
RF1/AN6/C2OUT/SEG19/CTDIN ......................... 20, 29 TRISD Register ........................................................ 166
RF2/AN7/C1OUT/SEG20 ........................................... 20 PORTE
RF2/AN7/C1OUT/SEG20/CTMUI ............................... 29 Associated Registers................................................ 171
RF3/AN8/SEG21/C2INB ............................................. 29 LATE Register .......................................................... 169
RF3/AN8/SEG21/C2INB/CTMUI................................. 20 Pins Available in Different LCD Drives ..................... 169
RF4/AN9/SEG22/C2INA ....................................... 20, 29 PORTE Register....................................................... 169
RF5/AN10 ................................................................... 20 TRISE Register......................................................... 169
RF5/AN10/CVREF/SEG23/C1INB ............................... 29 PORTF
RF6/AN11/SEG24/C1INA ..................................... 20, 29 Associated Registers................................................ 174
RF7/AN5/SS1/SEG25........................................... 20, 29 LATF Register .......................................................... 172
RG0/ECCP3/P3A.................................................. 21, 30 PORTF Register ....................................................... 172
RG1/TX2/CK2/AN19/C3OUT................................ 21, 30 TRISF Register......................................................... 172
RG2/RX2/DT2/AN18/C3INA ................................. 21, 30 PORTG
RG3/CCP4/AN17/P3D/C3INB .............................. 21, 30 Associated Registers................................................ 176
RG4/SEG26/RTCC/T7CKI/T5G/CCP5/AN16/P1D/ LATG Register.......................................................... 175
C3INC ........................................................... 21, 30 PORTG Register ...................................................... 175
RH0/SEG47/AN23 ...................................................... 31 TRISG Register ........................................................ 175
RH1/SEG46/AN22 ...................................................... 31 PORTH
RH2/SEG45/AN21 ...................................................... 31 Associated Registers................................................ 179
RH3/SEG44/AN20 ...................................................... 31 LATH Register .......................................................... 177
RH4/SEG40/CCP9/P3C/AN12/C2INC........................ 31 PORTH Register....................................................... 177
RH5/SEG41/CCP8/P3B/AN13/C2IND ........................ 31 TRISH Register ........................................................ 177
RH6/SEG42/CCP7/P1C/AN14/C1INC........................ 31 PORTJ
RH7/SEG43/CCP6/P1B/AN15.................................... 32 Associated Registers................................................ 181
RJ0.............................................................................. 33 LATJ Register........................................................... 180
RJ1/SEG33 ................................................................. 33 PORTJ Register ....................................................... 180
RJ2/SEG34 ................................................................. 33 TRISJ Register ......................................................... 180
RJ3/SEG35 ................................................................. 33 Power-Managed Modes...................................................... 53
RJ4/SEG39 ................................................................. 33 and PWM Operation ................................................. 271
RJ5/SEG38 ................................................................. 33 and SPI Operation .................................................... 311
RJ6/SEG37 ................................................................. 33 Clock Transitions and Status Indicators ..................... 54
RJ7/SEG36 ................................................................. 33 Entering ...................................................................... 53
VDD ....................................................................... 22, 33 Exiting Idle and Sleep Modes ..................................... 65
VDDCORE/VCAP ...................................................... 22, 33 by Interrupt ......................................................... 65
VSS ........................................................................ 22, 33 by Reset ............................................................. 65
Pinout I/O Descriptions by WDT Time-out ............................................... 65
PIC18F6XK90 ............................................................. 14 Without an Oscillator Start-up Delay .................. 65
PIC18F8XK90 ............................................................. 22 Idle Modes .................................................................. 58
PIR5 Register PRI_IDLE ........................................................... 59
TMR12IF Bit.............................................................. 138 RC_IDLE ............................................................ 60
PLL...................................................................................... 48 SEC_IDLE .......................................................... 59
HSPLL and ECPLL Oscillator Modes ......................... 48
Use with HF-INTOSC.................................................. 49

PIC18F87K90 FAMILY
Multiple Sleep Commands .......................................... 54 Reader Response............................................................. 567
Run Modes.................................................................. 54 Real-Time Clock and Calendar (RTCC) ........................... 217
PRI_RUN ............................................................ 54 Alarm ........................................................................ 232
RC_RUN ............................................................. 55 Configuring ....................................................... 232
SEC_RUN........................................................... 54 Interrupt ............................................................ 233
Selecting ..................................................................... 53 Mask Settings ................................................... 233
Sleep Mode ................................................................. 58 Alarm Value Registers (ALRMVAL).......................... 226
OSC1, OSC2 Pin States ..................................... 52 Clock Source ............................................................ 230
Summary (table) ......................................................... 53 Control Registers ...................................................... 219
Power-on Reset (POR) ....................................................... 71 Digit Carry Rules....................................................... 230
Power-up Delays................................................................. 52 Operation .................................................................. 229
Power-up Timer (PWRT)............................................... 52, 72 Calibration ........................................................ 232
Time-out Sequence..................................................... 72 General Functionality........................................ 231
Prescaler, Timer0.............................................................. 185 Leap Year ......................................................... 231
Prescaler, Timer2.............................................................. 248 Register Mapping ............................................. 231
PRI_IDLE Mode .................................................................. 59 Register Reads/Writes, Safety Window............ 231
PRI_RUN Mode .................................................................. 54 Write Lock......................................................... 231
Program Counter (PC) ........................................................ 87 Register Interface ..................................................... 229
PCL, PCH and PCU Registers.................................... 87 Register Maps........................................................... 235
PCLATH and PCLATU Registers ............................... 87 Alarm Value ...................................................... 235
Program Memory RTCC Control ................................................... 235
Extended Instruction Set ........................................... 106 RTCC Value ..................................................... 235
Hard Memory Vectors ................................................. 86 Registers .................................................................. 218
Instructions.................................................................. 91 Reset ........................................................................ 234
Two-Word ........................................................... 91 Device............................................................... 234
Interrupt Vector ........................................................... 86 Power-on Reset (POR)..................................... 234
Look-up Tables ........................................................... 89 RTCEN Bit, Write...................................................... 229
Memory Maps Sleep Mode .............................................................. 234
Hard Vectors and Configuration Words .............. 86 Value Registers (RTCVAL)....................................... 222
Reset Vector ............................................................... 86 Reference Clock Output ..................................................... 51
Program Verification.......................................................... 447 Register File........................................................................ 94
Programming, Device Instructions .................................... 451 Register File Summary ............................................... 97–102
Pulse Steering................................................................... 268 Registers
Pulse-Width Modulation. See PWM (CCP Module). ADCON0 (A/D Control 0).......................................... 374
PUSH ................................................................................ 480 ADCON1 (A/D Control 1).......................................... 375
PUSH and POP Instructions ............................................... 88 ADCON2 (A/D Control 2).......................................... 376
PUSHL .............................................................................. 496 ADRESH (A/D Result High Byte Left Justified,
PWM (CCP Module) ADFM = 0) ........................................................ 378
Associated Registers with PWM and Timers ............ 249 ADRESH (A/D Result High Byte Right Justified,
Duty Cycle................................................................. 248 ADFM = 1) ........................................................ 379
Example Frequencies/Resolutions, 40 MHz ............. 248 ADRESL (A/D Result Low Byte Left Justified,
Period........................................................................ 248 ADFM = 0) ........................................................ 378
Setup for Operation................................................... 249 ADRESL (A/D Result Low Byte Right Justified,
TMR12 to PR12 Match Flag (TMR12IF Bit) .............. 138 ADFM = 1) ........................................................ 379
TMR2 to PR2 Match ................................................. 248 ALRMCFG (Alarm Configuration) ............................. 221
TMRx to PRx Match .................................................. 213 ALRMDAY (Alarm Day Value) .................................. 226
PWM (ECCP Module) ALRMHR (Alarm Hours Value) ................................. 227
Effects of a Reset...................................................... 271 ALRMMIN (Alarm Minutes Value)............................. 228
Operation in Power-Managed Modes ....................... 271 ALRMMNTH (Alarm Month Value) ........................... 226
Operation with Fail-Safe Clock Monitor .................... 271 ALRMRPT (Alarm Repeat) ....................................... 222
Pulse Steering........................................................... 268 ALRMSEC (Alarm Seconds Value) .......................... 228
Steering Synchronization .......................................... 270 ALRMWD (Alarm Weekday Value)........................... 227
PWM Mode. See Enhanced Capture/Compare/PWM. ANCON0 (A/D Port Configuration 0) ........................ 380
ANCON1 (A/D Port Configuration 1) ........................ 380
Q ANCON2 (A/D Port Configuration 2) ........................ 381
Q Clock ............................................................................. 248 BAUDCONx (Baud Rate Control) ............................. 352
CCPRxH (CCPx Period High Byte) .......................... 240
R CCPRxL (CCPx Period Low Byte)............................ 240
RAM. See Data Memory. CCPTMRS0 (CCP Timer Select 0)........................... 253
RC Oscillator ....................................................................... 47 CCPTMRS1 (CCPx Timer Select 1) ......................... 238
RC_IDLE Mode ................................................................... 60 CCPTMRS2 (CCPx Timer Select 2) ......................... 239
RC_RUN Mode ................................................................... 55 CCPxCON (CCPx Control, CCP4-CCP10)............... 237
RCALL............................................................................... 481 CCPxCON (Enhanced Capture/Compare/
RCON Register PWM x Control) ................................................ 252
Bit Status During Initialization ..................................... 74

PIC18F87K90 FAMILY
CMSTAT (Comparator Status).................................. 391 PMD0 (Peripheral Module Disable 0) ......................... 64
CMxCON (Comparator Control x)............................. 390 PMD1 (Peripheral Module Disable 1) ......................... 63
CONFIG1H (Configuration 1 High) ........................... 428 PMD2 (Peripheral Module Disable 2) ......................... 62
CONFIG1L (Configuration 1 Low)............................. 427 PMD3 (Peripheral Module Disable 3) ......................... 61
CONFIG2H (Configuration 2 High) ........................... 430 PSTRxCON (Pulse Steering Control) ....................... 269
CONFIG2L (Configuration 2 Low)............................. 429 RCON (Reset Control)........................................ 70, 150
CONFIG3H (Configuration 3 High) ........................... 431 RCSTAx (Receive Status and Control) .................... 351
CONFIG3L (Configuration 3 Low)............................. 431 REFOCON (Reference Oscillator Control) ................. 51
CONFIG4L (Configuration 4 Low)............................. 432 Reserved .................................................................. 222
CONFIG5H (Configuration 5 High) ........................... 434 RTCCAL (RTCC Calibration).................................... 220
CONFIG5L (Configuration 5 Low)............................. 433 RTCCFG (RTCC Configuration)............................... 219
CONFIG6H (Configuration 6 High) ........................... 436 SECOND (Second Value) ........................................ 225
CONFIG6L (Configuration 6 Low)............................. 435 SSPxCON1 (MSSPx Control 1, I2C Mode) .............. 315
CONFIG7H (Configuration 7 High) ........................... 438 SSPxCON1 (MSSPx Control 1, SPI Mode).............. 305
CONFIG7L (Configuration 7 Low)............................. 437 SSPxCON2 (MSSPx Control 2, I2C Master Mode) .. 316
CTMUCONH (CTMU Control High) .......................... 408 SSPxCON2 (MSSPx Control 2, I2C Slave Mode) .... 317
CTMUCONL (CTMU Control Low)............................ 409 SSPxMSK (I2C Slave Address Mask, 7-Bit) ............. 317
CTMUICON (CTMU Current Control) ....................... 410 SSPxSTAT (MSSPx Status, I2C Mode).................... 314
CVRCON (Comparator Voltage Reference SSPxSTAT (MSSPx Status, SPI Mode) ................... 304
Control) ............................................................. 397 STATUS ................................................................... 103
DAY (Day Value)....................................................... 224 STKPTR (Stack Pointer)............................................. 88
DEVID1 (Device ID 1) ............................................... 439 T0CON (Timer0 Control) .......................................... 183
DEVID2 (Device ID 2) ............................................... 439 T1CON (Timer1 Control) .......................................... 187
ECCPxAS (ECCPx Auto-Shutdown Control) ............ 265 T1GCON (Timer1 Gate Control)............................... 188
ECCPxDEL (Enhanced PWM Control) ..................... 268 T2CON (Timer2 Control) .......................................... 199
EECON1 (Data EEPROM Control 1) ........................ 122 TxCON (Timer3/5/7 Control) .................................... 202
EECON1 (EEPROM Control 1)................................. 113 TxCON (Timer4/6/8/10/12 Control) .......................... 214
HLVDCON (High/Low-Voltage Detect Control)......... 401 TxGCON (Timer3/5/7 Gate Control) ......................... 203
HOUR (Hour Value) .................................................. 225 TXSTAx (Transmit Status and Control) .................... 350
INTCON (Interrupt Control)....................................... 131 WDTCON (Watchdog Timer Control) ....................... 441
INTCON2 (Interrupt Control 2).................................. 132 WEEKDAY (Weekday Value) ................................... 224
INTCON3 (Interrupt Control 3).................................. 133 YEAR (Year Value)................................................... 223
IPR1 (Peripheral Interrupt Priority 1)......................... 145 RESET.............................................................................. 481
IPR2 (Peripheral Interrupt Priority 2)......................... 146 Reset .................................................................................. 69
IPR3 (Peripheral Interrupt Priority 3)......................... 147 Brown-out Reset (BOR).............................................. 69
IPR4 (Peripheral Interrupt Priority 4)......................... 147 Configuration Mismatch (CM)..................................... 69
IPR5 (Peripheral Interrupt Priority 5)......................... 148 MCLR Reset, Normal Operation................................. 69
IPR6 (Peripheral Interrupt Priority 6)......................... 149 MCLR Reset, Power-Managed Modes....................... 69
LCDCON (LCD Control)............................................ 274 Power-on Reset (POR)............................................... 69
LCDDATAx (LCD Datax) .......................................... 279 RESET Instruction ...................................................... 69
LCDPS (LCD Phase) ................................................ 275 Stack Full.................................................................... 69
LCDREF (LCD Reference Voltage Control).............. 276 Stack Underflow ......................................................... 69
LCDRL (LCD Reference Ladder Control) ................. 277 State of Registers ....................................................... 74
LCDSEx (LCD Segmentx Enable) ............................ 278 Watchdog Timer (WDT).............................................. 69
MINUTE (Minute Value)............................................ 225 Resets .............................................................................. 425
MONTH (Month Value) ............................................. 223 Brown-out Reset (BOR)............................................ 425
ODCON1 (Peripheral Open-Drain Control 1)............ 154 Oscillator Start-up Timer (OST)................................ 425
ODCON2 (Peripheral Open-Drain Control 2)............ 155 Power-on Reset (POR)............................................. 425
ODCON3 (Peripheral Open-Drain Control 3)............ 156 Power-up Timer (PWRT) .......................................... 425
OSCCON (Oscillator Control) ..................................... 43 RETFIE ............................................................................. 482
OSCCON2 (Oscillator Control 2) ........................ 44, 204 RETLW ............................................................................. 482
OSCTUNE (Oscillator Tuning) .................................... 45 RETURN........................................................................... 483
PADCFG1 (Pad Configuration)................................. 220 Return Address Stack......................................................... 87
PIE1 (Peripheral Interrupt Enable 1)......................... 140 Return Stack Pointer (STKPTR) ......................................... 88
PIE2 (Peripheral Interrupt Enable 2)......................... 141 Revision History................................................................ 553
PIE3 (Peripheral Interrupt Enable 3)......................... 142 RLCF ................................................................................ 483
PIE4 (Peripheral Interrupt Enable 4)......................... 142 RLNCF.............................................................................. 484
PIE5 (Peripheral Interrupt Enable 5)......................... 143 RRCF................................................................................ 484
PIE6 (Peripheral Interrupt Enable 6)......................... 144 RRNCF ............................................................................. 485
PIR1 (Peripheral Interrupt Request (Flag) 1) ............ 134

PIC18F87K90 FAMILY
S T
SCKx ................................................................................. 303 Table Pointer Operations (table)....................................... 114
SDIx .................................................................................. 303 Table Reads/Table Writes .................................................. 89
SDOx................................................................................. 303 TBLRD .............................................................................. 489
SEC_IDLE Mode................................................................. 59 TBLWT.............................................................................. 490
SEC_RUN Mode ................................................................. 54 Timer0............................................................................... 183
Secondary Oscillator (SOSC) ............................................. 45 Associated Registers ................................................ 185
Selective Peripheral Module Control................................... 60 Operation .................................................................. 184
Serial Clock, SCKx............................................................ 303 Overflow Interrupt ..................................................... 185
Serial Data In (SDIx) ......................................................... 303 Prescaler .................................................................. 185
Serial Data Out (SDOx)..................................................... 303 Switching Assignment ...................................... 185
Serial Peripheral Interface. See SPI Mode. Prescaler Assignment (PSA Bit) ............................... 185
SETF ................................................................................. 485 Prescaler Select (T0PS2:T0PS0 Bits) ...................... 185
Shoot-Through Current ..................................................... 267 Reads and Writes in 16-Bit Mode ............................. 184
Slave Select (SSx) ............................................................ 303 Source Edge Select (T0SE Bit) ................................ 184
SLEEP............................................................................... 486 Source Select (T0CS Bit).......................................... 184
Software Simulator (MPLAB SIM)..................................... 503 Timer1............................................................................... 187
Special Event Trigger. See Compare (CCP Module). 16-Bit Read/Write Mode ........................................... 191
Special Event Trigger. See Compare (ECCP Mode). Associated Registers ................................................ 197
Specifications Clock Source Selection............................................. 189
AC (Timing) Characteristics Gate .......................................................................... 193
Temperature and Voltage ................................. 524 Interrupt .................................................................... 192
Capture/Compare/PWM Requirements Operation .................................................................. 189
(ECCP1, ECCP2).............................................. 532 Oscillator................................................................... 187
CLKO and I/O Requirements .................................... 527 Resetting, Using the ECCP Special
Comparator ............................................................... 522 Event Trigger .................................................... 193
EUSART/AUSART Synchronous Receive SOSC Oscillator........................................................ 191
Requirements.................................................... 541 Layout Considerations...................................... 192
EUSART/AUSART Synchronous Transmission Use as a Clock Source ..................................... 192
Requirements.................................................... 541 TMR1H Register ....................................................... 187
Example SPI Mode Requirements (Master Mode, TMR1L Register........................................................ 187
CKE = 0) ........................................................... 533 Timer2............................................................................... 199
Example SPI Mode Requirements (Master Mode, Associated Registers ................................................ 200
CKE = 1) ........................................................... 534 Interrupt .................................................................... 200
Example SPI Mode Requirements (Slave Mode, Operation .................................................................. 199
CKE = 0) ........................................................... 535 Output ....................................................................... 200
Example SPI Slave Mode Requirements PR2 Register ............................................................ 248
(CKE = 1) .......................................................... 536 TMR12 to PR12 Match Flag (TMR12IF Bit).............. 138
External Clock Requirements ................................... 525 TMR2 to PR2 Match Interrupt................................... 248
High/Low-Voltage Detect Characteristics ................. 530 Timer3/5/7......................................................................... 201
I2C Bus Data Requirements (Slave Mode) ............... 538 16-Bit Read/Write Mode ........................................... 206
I2C Bus Start/Stop Bits Requirements Associated Registers ................................................ 211
(Slave Mode)..................................................... 537 Gates ........................................................................ 206
Internal RC Accuracy (INTOSC) ............................... 526 Operation .................................................................. 205
Internal Voltage Regulator ........................................ 522 Overflow Interrupt ............................................. 201, 210
Memory Programming Requirements ....................... 521 SOSC Oscillator........................................................ 201
MSSP I2C Bus Data Requirements .......................... 540 Use as a Clock Source ..................................... 206
MSSP I2C Bus Start/Stop Bits Requirements ........... 539 Special Event Trigger (ECCP) .................................. 210
PLL Clock Timing ...................................................... 526 TMRxH Register ....................................................... 201
Timer0 and Timer1 External Clock TMRxL Register........................................................ 201
Requirements.................................................... 531 Timer4/6/8/10/12............................................................... 213
Ultra Low-Power Wake-up ........................................ 541 Associated Registers ................................................ 215
Voltage Reference .................................................... 522 Interrupt .................................................................... 215
SSPOV.............................................................................. 339 Operation .................................................................. 213
SSPOV Status Flag........................................................... 339 Output ....................................................................... 215
SSPxSTAT Register Outputs, PWM Time Base for ECCP ........................ 215
R/W Bit .............................................................. 318, 321 Postscaler. See Postscaler, Timer4/6/8/10/12.
SSx.................................................................................... 303 Prescaler. See Prescaler, Timer4/6/8/10/12.
Stack Full/Underflow Resets ............................................... 89 PRx Register............................................................. 213
SUBFSR............................................................................ 497 TMRx Register.......................................................... 213
SUBFWB........................................................................... 486 TMRx to PRx Match Interrupt ........................... 213, 215
SUBLW ............................................................................. 487
SUBULNK ......................................................................... 497
SUBWF ............................................................................. 487
SUBWFB........................................................................... 488
SWAPF ............................................................................. 488

PIC18F87K90 FAMILY
Timing Diagrams I2C Stop Condition Receive or Transmit Mode......... 342
A/D Conversion......................................................... 543 LCD Interrupt Timing in Quarter Duty Cycle Drive ... 298
Automatic Baud Rate Calculation ............................. 358 LCD Reference Ladder Power Mode Switching ....... 283
Auto-Wake-up Bit (WUE) During Normal LCD Sleep Entry/Exit When SLPEN = 1 or CS = 00 299
Operation .......................................................... 365 Low-Voltage Detect Operation (VDIRMAG = 0) ....... 403
Auto-Wake-up Bit (WUE) During Sleep .................... 365 MSSP I2C Bus Data ................................................. 539
Baud Rate Generator with Clock Arbitration ............. 336 MSSP I2C Bus Start/Stop Bits .................................. 539
BRG Overflow Sequence.......................................... 358 PWM Auto-Shutdown with Auto-Restart Enabled,
BRG Reset Due to SDAx Arbitration During PxRSEN = 1 ..................................................... 266
Start Condition .................................................. 345 PWM Auto-Shutdown with Firmware Restart,
Brown-out Reset (BOR) ............................................ 528 PxRSEN = 0 ..................................................... 266
Bus Collision During a Repeated Start PWM Direction Change ............................................ 263
Condition (Case 1) ............................................ 346 PWM Direction Change at Near 100%
Bus Collision During a Repeated Start Duty Cycle ........................................................ 264
Condition (Case 2) ............................................ 346 PWM Output ............................................................. 248
Bus Collision During a Start Condition PWM Output (Active-High) ....................................... 258
(SCLx = 0)......................................................... 345 PWM Output (Active-Low) ........................................ 259
Bus Collision During a Stop Condition Repeated Start Condition ......................................... 338
(Case 1) ............................................................ 347 Reset, Watchdog Timer (WDT), Oscillator Start-up
Bus Collision During a Stop Condition Timer (OST) and Power-up Timer (PWRT) ...... 528
(Case 2) ............................................................ 347 Send Break Character Sequence............................. 366
Bus Collision During Start Condition Slave Synchronization .............................................. 309
(SDAx Only) ...................................................... 344 Slow Rise Time (MCLR Tied to VDD,
Bus Collision for Transmit and Acknowledge............ 343 VDD Rise > TPWRT)............................................. 73
Capture/Compare/PWM (ECCP1, ECCP2) .............. 532 SPI Mode (Master Mode) ......................................... 308
CLKO and I/O ........................................................... 527 SPI Mode (Slave Mode, CKE = 0) ............................ 310
Clock Synchronization .............................................. 329 SPI Mode (Slave Mode, CKE = 1) ............................ 310
Clock/Instruction Cycle ............................................... 90 Steering Event at Beginning of Instruction
EUSART Asynchronous Reception .......................... 363 (STRSYNC = 1)................................................ 270
EUSART Asynchronous Transmission ..................... 360 Steering Event at End of Instruction
EUSART Asynchronous Transmission (STRSYNC = 0)................................................ 270
(Back-to-Back) .................................................. 360 Synchronous Master Transmission
EUSART Synchronous Master Transmission ........... 367 (Through TXEN) ............................................... 368
EUSART Synchronous Transmission Synchronous Reception (Master Mode, SREN) ....... 369
(Master/Slave) .................................................. 541 Time-out Sequence on Power-up (MCLR Not
EUSART/AUSART Synchronous Receive Tied to VDD), Case 1 .......................................... 73
(Master/Slave) .................................................. 541 Time-out Sequence on Power-up (MCLR Not
Example SPI Master Mode (CKE = 0) ...................... 533 Tied to VDD), Case 2 .......................................... 73
Example SPI Master Mode (CKE = 1) ...................... 534 Time-out Sequence on Power-up
Example SPI Slave Mode (CKE = 0) ........................ 535 (MCLR Tied to VDD, VDD Rise TPWRT) ............... 72
Example SPI Slave Mode (CKE = 1) ........................ 536 Timer Pulse Generation............................................ 234
External Clock........................................................... 525 Timer0 and Timer1 External Clock ........................... 531
Fail-Safe Clock Monitor (FSCM) ............................... 446 Timer1 Gate Count Enable Mode............................. 193
First Start Bit Timing ................................................. 337 Timer1 Gate Single Pulse Mode............................... 196
Full-Bridge PWM Output ........................................... 262 Timer1 Gate Single Pulse/Toggle
Half-Bridge PWM Output .................................. 260, 267 Combined Mode ............................................... 197
High/Low-Voltage Detect Characteristics ................. 530 Timer1 Gate Toggle Mode........................................ 195
High-Voltage Detect Operation (VDIRMAG = 1)....... 404 Timer3/5/7 Gate Count Enable Mode....................... 206
I2C Acknowledge Sequence ..................................... 342 Timer3/5/7 Gate Single Pulse Mode......................... 208
I2C Bus Data ............................................................. 538 Timer3/5/7 Gate Single Pulse/Toggle
I2C Bus Start/Stop Bits.............................................. 537 Combined Mode ............................................... 209
I2C Master Mode (7 or 10-Bit Transmission) ............ 340 Timer3/5/7 Gate Toggle Mode.................................. 207
I2C Master Mode (7-Bit Reception)........................... 341 Transition for Entry to Idle Mode ................................ 59
I2C Slave Mode (10-Bit Reception, SEN = 0, Transition for Entry to SEC_RUN Mode ..................... 55
ADMSK = 01001).............................................. 325 Transition for Entry to Sleep Mode ............................. 58
I2C Slave Mode (10-Bit Reception, SEN = 0) ........... 326 Transition for Two-Speed Start-up
I2C Slave Mode (10-Bit Reception, SEN = 1) ........... 331 (INTOSC to HSPLL) ......................................... 444
I2C Slave Mode (10-Bit Transmission)...................... 327 Transition for Wake from Idle to Run Mode................ 59
I2C Slave Mode (7-Bit Reception, SEN = 0, Transition for Wake from Sleep (HSPLL) ................... 58
ADMSK = 01011).............................................. 323 Transition from RC_RUN Mode to
I2C Slave Mode (7-Bit Reception, SEN = 0) ............. 322 PRI_RUN Mode.................................................. 57
I2C Slave Mode (7-Bit Reception, SEN = 1) ............. 330 Transition from SEC_RUN Mode to
I2C Slave Mode (7-Bit Transmission)........................ 324 PRI_RUN Mode (HSPLL) ................................... 55
I2C Slave Mode General Call Address Sequence Transition to RC_RUN Mode...................................... 57
(7 or 10-Bit Addressing Mode) .......................... 332 Type-A in 1/2 MUX, 1/2 Bias Drive........................... 288

PIC18F87K90 FAMILY
Type-A in 1/2 MUX, 1/3 Bias Drive ........................... 290 U
Type-A in 1/3 MUX, 1/2 Bias Drive ........................... 292 Ultra Low-Power
Type-A in 1/3 MUX, 1/3 Bias Drive ........................... 294 Regulator
Type-A in 1/4 MUX, 1/3 Bias Drive ........................... 296
In Sleep ............................................................ 443
Type-A/Type-B in Static Drive................................... 287 Ultra Low-Power Wake-up
Type-B in 1/2 MUX, 1/2 Bias Drive ........................... 289 Exit Delay (table) ........................................................ 67
Type-B in 1/2 MUX, 1/3 Bias Drive ........................... 291
Ultra Low-Power Wake-up (ULPWU)
Type-B in 1/3 MUX, 1/2 Bias Drive ........................... 293 Overview..................................................................... 66
Type-B in 1/3 MUX, 1/3 Bias Drive ........................... 295
Type-B in 1/4 MUX, 1/3 Bias Drive ........................... 297 W
Timing Diagrams and Specifications Watchdog Timer (WDT)............................................ 425, 440
Reset, Watchdog Timer, Oscillator Start-up Associated Registers ................................................ 441
Timer, Power-up Timer and Brown-out Control Register........................................................ 441
Reset Requirements ......................................... 529 During Oscillator Failure ........................................... 445
Top-of-Stack (TOS) Access ................................................ 87 Programming Considerations ................................... 440
TSTFSZ............................................................................. 491 WCOL ....................................................... 337, 338, 339, 342
Two-Speed Start-up .................................................. 425, 444 WCOL Status Flag.................................... 337, 338, 339, 342
IESO (CONFIG1H), Internal/External WWW Address ................................................................. 566
Oscillator Switchover Bit ................................... 428
Two-Word Instructions X
Example Cases........................................................... 91 XORLW............................................................................. 491
TXSTAx Register XORWF ............................................................................ 492
BRGH Bit .................................................................. 353

PIC18F87K90 FAMILY
THE MICROCHIP WEB SITE CUSTOMER SUPPORT
Microchip provides online support via our WWW site at Users of Microchip products can receive assistance
www.microchip.com. This web site is used as a means through several channels:
to make files and information easily available to • Distributor or Representative
customers. Accessible by using your favorite Internet
• Local Sales Office
browser, the web site contains the following
information: • Field Application Engineer (FAE)
• Technical Support
• Product Support – Data sheets and errata,
application notes and sample programs, design • Development Systems Information Line
resources, user’s guides and hardware support Customers should contact their distributor,
documents, latest software releases and archived representative or field application engineer (FAE) for
software support. Local sales offices are also available to help
• General Technical Support – Frequently Asked customers. A listing of sales offices and locations is
Questions (FAQ), technical support requests, included in the back of this document.
online discussion groups, Microchip consultant Technical support is available through the web site
program member listing at: http://microchip.com/support
• Business of Microchip – Product selector and
ordering guides, latest Microchip press releases,
listing of seminars and events, listings of
Microchip sales offices, distributors and factory
representatives
CUSTOMER CHANGE NOTIFICATION

SERVICE
Microchip’s customer notification service helps keep
customers current on Microchip products. Subscribers
will receive e-mail notification whenever there are
changes, updates, revisions or errata related to a
specified product family or development tool of interest.
To register, access the Microchip web site at
www.microchip.com. Under “Support”, click on
“Customer Change Notification” and follow the
registration instructions.

PIC18F87K90 FAMILY
READER RESPONSE
It is our intention to provide you with the best documentation possible to ensure successful use of your Microchip
product. If you wish to provide your comments on organization, clarity, subject matter, and ways in which our
documentation can better serve you, please FAX your comments to the Technical Publications Manager at
(480) 792-4150.
Please list the following information, and use this outline to provide us with your comments about this document.
TO: Technical Publications Manager Total Pages Sent ________

RE: Reader Response
From: Name
Company
Address
City / State / ZIP / Country
Telephone: (_______) _________ - _________ FAX: (______) _________ - _________
Application (optional):
Would you like a reply? Y N
Device: PIC18F87K90 Family Literature Number: DS39957D
Questions:
1. What are the best features of this document?
2. How does this document meet your hardware and software development needs?
3. Do you find the organization of this document easy to follow? If not, why?
4. What additions to the document do you think would enhance the structure and subject?
5. What deletions from the document could be made without affecting the overall usefulness?
6. Is there any incorrect or misleading information (what and where)?
7. How would you improve this document?

PIC18F87K90 FAMILY
PRODUCT IDENTIFICATION SYSTEM
To order or obtain information, e.g., on pricing or delivery, refer to the factory or the listed sales office.
PART NO. X /XX XXX
Examples:
Device Temperature Package Pattern a) PIC18F87K90-I/PT 301 = Industrial temperature,
Range TQFP package, QTP pattern #301.
b) PIC18F87K90T-I/PT = Tape and reel, Industrial
temperature, TQFP package.
c) PIC18F87K22T-E/PT = Tape and reel,
Device(1,2) PIC18F65K90, PIC18F65K90T
Extended temperature, TQFP package.
PIC18F66K90, PIC18F66K90T
Temperature Range I = -40C to +85C (Industrial)

E = -40C to +125C (Extended)
Package PT = TQFP (Plastic Thin Quad Flatpack)

MR = QFN (Plastic Quad Flat) Note 1: F = Standard Voltage Range
2: T = In tape and reel
3: RSL = Silicon revision A3
Pattern QTP, SQTP, Code or Special Requirements
(blank otherwise)

Worldwide Sales and Service
AMERICAS ASIA/PACIFIC ASIA/PACIFIC EUROPE
Corporate Office Asia Pacific Office India - Bangalore Austria - Wels
2355 West Chandler Blvd. Suites 3707-14, 37th Floor Tel: 91-80-3090-4444 Tel: 43-7242-2244-39
Chandler, AZ 85224-6199 Tower 6, The Gateway Fax: 91-80-3090-4123 Fax: 43-7242-2244-393
Tel: 480-792-7200 Harbour City, Kowloon Denmark - Copenhagen
India - New Delhi
Fax: 480-792-7277 Hong Kong Tel: 45-4450-2828
Tel: 91-11-4160-8631
Technical Support: Tel: 852-2401-1200 Fax: 45-4485-2829
Fax: 91-11-4160-8632
http://www.microchip.com/ Fax: 852-2401-3431
support India - Pune France - Paris
Australia - Sydney Tel: 91-20-2566-1512 Tel: 33-1-69-53-63-20
Web Address:
Tel: 61-2-9868-6733 Fax: 91-20-2566-1513 Fax: 33-1-69-30-90-79
www.microchip.com
Fax: 61-2-9868-6755 Germany - Munich
Atlanta Japan - Yokohama
China - Beijing Tel: 81-45-471- 6166 Tel: 49-89-627-144-0
Duluth, GA
Tel: 86-10-8569-7000 Fax: 81-45-471-6122 Fax: 49-89-627-144-44
Tel: 678-957-9614
Fax: 86-10-8528-2104 Italy - Milan
Fax: 678-957-1455 Korea - Daegu
China - Chengdu Tel: 39-0331-742611
Boston Tel: 82-53-744-4301
Tel: 86-28-8665-5511 Fax: 82-53-744-4302 Fax: 39-0331-466781
Westborough, MA
Tel: 774-760-0087 Fax: 86-28-8665-7889 Netherlands - Drunen
Korea - Seoul
Fax: 774-760-0088 China - Chongqing Tel: 82-2-554-7200 Tel: 31-416-690399
Tel: 86-23-8980-9588 Fax: 82-2-558-5932 or Fax: 31-416-690340
Chicago
Itasca, IL Fax: 86-23-8980-9500 82-2-558-5934 Spain - Madrid
Tel: 630-285-0071 China - Hangzhou Tel: 34-91-708-08-90
Malaysia - Kuala Lumpur
Fax: 630-285-0075 Tel: 86-571-2819-3180 Tel: 60-3-6201-9857 Fax: 34-91-708-08-91
Cleveland Fax: 86-571-2819-3189 Fax: 60-3-6201-9859 UK - Wokingham
Independence, OH China - Hong Kong SAR Tel: 44-118-921-5869
Malaysia - Penang
Tel: 216-447-0464 Tel: 852-2401-1200 Fax: 44-118-921-5820
Tel: 60-4-227-8870
Fax: 216-447-0643 Fax: 852-2401-3431 Fax: 60-4-227-4068
Dallas China - Nanjing Philippines - Manila
Addison, TX Tel: 86-25-8473-2460 Tel: 63-2-634-9065
Tel: 972-818-7423 Fax: 86-25-8473-2470 Fax: 63-2-634-9069
Fax: 972-818-2924
China - Qingdao Singapore
Detroit Tel: 86-532-8502-7355 Tel: 65-6334-8870
Farmington Hills, MI
Fax: 86-532-8502-7205 Fax: 65-6334-8850
Tel: 248-538-2250
Fax: 248-538-2260 China - Shanghai Taiwan - Hsin Chu
Tel: 86-21-5407-5533 Tel: 886-3-6578-300
Indianapolis Fax: 86-21-5407-5066
Noblesville, IN Fax: 886-3-6578-370
Tel: 317-773-8323 China - Shenyang Taiwan - Kaohsiung
Fax: 317-773-5453 Tel: 86-24-2334-2829 Tel: 886-7-213-7830
Fax: 86-24-2334-2393 Fax: 886-7-330-9305
Los Angeles
Mission Viejo, CA China - Shenzhen Taiwan - Taipei
Tel: 949-462-9523 Tel: 86-755-8203-2660 Tel: 886-2-2500-6610
Fax: 949-462-9608 Fax: 86-755-8203-1760 Fax: 886-2-2508-0102
Santa Clara China - Wuhan Thailand - Bangkok

Santa Clara, CA Tel: 86-27-5980-5300 Tel: 66-2-694-1351
Tel: 408-961-6444 Fax: 86-27-5980-5118 Fax: 66-2-694-1350
Fax: 408-961-6445 China - Xian
Toronto Tel: 86-29-8833-7252
Mississauga, Ontario, Fax: 86-29-8833-7256
Canada China - Xiamen
Tel: 905-673-0699 Tel: 86-592-2388138
Fax: 905-673-6509 Fax: 86-592-2388130
China - Zhuhai
Tel: 86-756-3210040
05/02/11
Fax: 86-756-3210049

Pic18f66k90t I PTRSL

Uploaded by

Copyright:

Available Formats

Pic18f66k90t I PTRSL

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

Pic18f66k90t I PTRSL

Uploaded by

Copyright:

Available Formats

PIC18F87K90 Family

 2009-2011 Microchip Technology Inc. DS39957D

Information contained in this publication regarding device Trademarks

Microchip received ISO/TS-16949:2009 certification for its worldwide

DS39957D-page 2  2009-2011 Microchip Technology Inc.

Low-Power Features: Peripheral Highlights:

Program Data EEPROM LCD CCP/

PIC18F65K90 32K 2K 1K 53 132 4/4 5/3 Yes Yes 2 16 3 Y Y

 2009-2011 Microchip Technology Inc. DS39957D-page 3

DS39957D-page 4  2009-2011 Microchip Technology Inc.

64-Pin QFN(3), TQFP

 2009-2011 Microchip Technology Inc. DS39957D-page 5

DS39957D-page 6  2009-2011 Microchip Technology Inc.

 2009-2011 Microchip Technology Inc. DS39957D-page 7

DS39957D-page 8  2009-2011 Microchip Technology Inc.

 2009-2011 Microchip Technology Inc. DS39957D-page 9

DS39957D-page 10  2009-2011 Microchip Technology Inc.

TABLE 1-2: DEVICE FEATURES FOR THE PIC18F8XK90 (80-PIN DEVICES)

 2009-2011 Microchip Technology Inc. DS39957D-page 11

ROM Latch Address

Timer Timer ADC Comparator

CCP ECCP LCD

Note 1: See Table 1-3 for I/O port pin descriptions.

DS39957D-page 12  2009-2011 Microchip Technology Inc.

ROM Latch Address

VDDCORE/VCAP VDD,VSS MCLR RJ0:RJ7(1)

Timer Timer ADC Comparator

CCP ECCP LCD

Note 1: See Table 1-3 for I/O port pin descriptions.

 2009-2011 Microchip Technology Inc. DS39957D-page 13

MCLR/RG5 7 Master Clear (input) or programming voltage (input).

DS39957D-page 14  2009-2011 Microchip Technology Inc.

PORTA is a bidirectional I/O port.

 2009-2011 Microchip Technology Inc. DS39957D-page 15

PORTB is a bidirectional I/O port. PORTB can be software

DS39957D-page 16  2009-2011 Microchip Technology Inc.

PORTC is a bidirectional I/O port.

 2009-2011 Microchip Technology Inc. DS39957D-page 17

PORTD is a bidirectional I/O port.

DS39957D-page 18  2009-2011 Microchip Technology Inc.

PORTE is a bidirectional I/O port.

 2009-2011 Microchip Technology Inc. DS39957D-page 19

PORTF is a bidirectional I/O port.

DS39957D-page 20  2009-2011 Microchip Technology Inc.

PORTG is a bidirectional I/O port.

 2009-2011 Microchip Technology Inc. DS39957D-page 21

TABLE 1-4: PIC18F8XK90 PINOUT I/O DESCRIPTIONS

MCLR/RG5 9 Master Clear (input) or programming voltage (input).

DS39957D-page 22  2009-2011 Microchip Technology Inc.

PORTA is a bidirectional I/O port.

 2009-2011 Microchip Technology Inc. DS39957D-page 23

PORTB is a bidirectional I/O port. PORTB can be software

DS39957D-page 24  2009-2011 Microchip Technology Inc.

 2009-2011 Microchip Technology Inc. DS39957D-page 25

PORTC is a bidirectional I/O port.

DS39957D-page 26  2009-2011 Microchip Technology Inc.

PORTD is a bidirectional I/O port.

 2009-2011 Microchip Technology Inc. DS39957D-page 27

PORTE is a bidirectional I/O port.