UPSD33XX [STMICROELECTRONICS]
Fast 8032 MCU with Programmable Logic; 快8032单片机的可编程逻辑型号: | UPSD33XX |
厂家: | ST |
描述: | Fast 8032 MCU with Programmable Logic |
文件: | 总231页 (文件大小:3711K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
uPSD33xx
Turbo Series
Fast 8032 MCU with Programmable Logic
PRELIMINARY DATA
FEATURES SUMMARY
■
FAST 8-BIT TURBO 8032 MCU, 40MHz
Figure 1. Packages
–
–
–
Advanced core, 4-clocks per instruction
10 MIPs peak performance at 40MHz (5V)
JTAG Debug and In-System
Programming
–
Branch Cache & 6 instruction Prefetch
Queue
–
–
Dual XDATA pointers with auto incr & decr
Compatible with 3rd party 8051 tools
TQFP52 (T)
52-lead, Thin,
Quad, Flat
■
DUAL FLASH MEMORIES WITH MEMORY
MANAGEMENT
–
Place either memory into 8032 program
address space or data address space
–
READ-while-WRITE operation for In-
Application Programming and EEPROM
emulation
–
–
Single voltage program and erase
100K guaranteed erase cycles, 15-year
retention
■
CLOCK, RESET, AND SUPPLY
MANAGEMENT
TQFP80 (U)
80-lead, Thin,
Quad, Flat
–
–
–
–
SRAM is Battery Backup capable
Flexible 8-level CPU clock divider register
Normal, Idle, and Power Down Modes
Power-on and Low Voltage reset
supervisor
Programmable Watchdog Timer
■
■
A/D CONVERTER
Eight Channels, 10-bit resolution, 6µs
TIMERS AND INTERRUPTS
–
–
■
■
PROGRAMMABLE LOGIC, GENERAL
PURPOSE
–
–
Three 8032 standard 16-bit timers
–
–
16 macrocells
Programmable Counter Array (PCA), six
16-bit modules for PWM, CAPCOM, and
timers
Create shifters, state machines, chip-
selects, glue-logic to keypads, panels,
LCDs, others
–
–
8/10/16-bit PWM operation
COMMUNICATION INTERFACES
2
11 Interrupt sources with two external
interrupt pins
–
–
–
–
–
I C Master/Slave controller, 833KHz
SPI Master controller, 10MHz
■
OPERATING VOLTAGE SOURCE (±10%)
–
Two UARTs with independent baud rate
IrDA protocol support up to 115K baud
Up to 46 I/O, 5V tolerant on 3.3V
uPSD33xxV
5V devices use both 5.0V and 3.3V
sources
3.3V devices use only 3.3V source
–
January 2005
1/231
This is preliminary information on a new product now in development or undergoing evaluation. Details are subject to change without notice.
uPSD33xx
Table 1. Device Summary
1st
2nd
Flash
SRAM
(bytes)
8032
Bus
V
CC
V
DD
Part Number
Flash
GPIO
Pkg.
Temp.
(bytes) (bytes)
uPSD3312D-40T6
uPSD3312DV-40T6
uPSD3333D-40T6
uPSD3333DV-40T6
uPSD3333D-40U6
uPSD3333DV-40U6
uPSD3334D-40U6
uPSD3334DV-40U6
uPSD3354D-40T6
uPSD3354DV-40T6
uPSD3354D-40U6
uPSD3354DV-40U6
64K
16K
16K
32K
32K
32K
32K
32K
32K
32K
32K
32K
32K
2K
2K
37
37
37
37
46
46
46
46
37
37
46
46
No
No
3.3V
3.3V
3.3V
3.3V
3.3V
3.3V
3.3V
3.3V
3.3V
3.3V
3.3V
3.3V
5.0V
3.3V
5.0V
3.3V
5.0V
3.3V
5.0V
3.3V
5.0V
3.3V
5.0V
3.3V
TQFP52 –40°C to 85°C
TQFP52 –40°C to 85°C
TQFP52 –40°C to 85°C
TQFP52 –40°C to 85°C
TQFP80 –40°C to 85°C
TQFP80 –40°C to 85°C
TQFP80 –40°C to 85°C
TQFP80 –40°C to 85°C
TQFP52 –40°C to 85°C
TQFP52 –40°C to 85°C
TQFP80 –40°C to 85°C
TQFP80 –40°C to 85°C
64K
128K
128K
128K
128K
256K
256K
256K
256K
256K
256K
8K
No
8K
No
8K
Yes
Yes
Yes
Yes
No
8K
8K
8K
32K
32K
32K
32K
No
Yes
Yes
2/231
uPSD33xx
TABLE OF CONTENTS
FEATURES SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
SUMMARY DESCRIPTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
PIN DESCRIPTIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
uPSD33xx HARDWARE DESCRIPTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
MEMORY ORGANIZATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Internal Memory (MCU Module, Standard 8032 Memory: DATA, IDATA, SFR) . . . . . . . . . . . . 16
External Memory (PSD Module: Program memory, Data memory). . . . . . . . . . . . . . . . . . . . . . 16
8032 MCU CORE PERFORMANCE ENHANCEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Pre-Fetch Queue (PFQ) and Branch Cache (BC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
PFQ Example, Multi-cycle Instructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Aggregate Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
MCU MODULE DISCRIPTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
8032 MCU REGISTERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Stack Pointer (SP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Data Pointer (DPTR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Program Counter (PC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Accumulator (ACC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
B Register (B). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
General Purpose Registers (R0 - R7). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Program Status Word (PSW) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
SPECIAL FUNCTION REGISTERS (SFR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
8032 ADDRESSING MODES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Register Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Direct Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Register Indirect Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Immediate Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
External Direct Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
External Indirect Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Indexed Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Relative Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Absolute Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Long Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Bit Addressing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
uPSD33xx INSTRUCTION SET SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3/231
uPSD33xx
DUAL DATA POINTERS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Data Pointer Control Register, DPTC (85h). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Data Pointer Mode Register, DPTM (86h) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
DEBUG UNIT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
INTERRUPT SYSTEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Individual Interrupt Sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
MCU CLOCK GENERATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
MCU_CLK. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
PERIPH_CLK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
POWER SAVING MODES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Idle Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Power-down Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Reduced Frequency Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
OSCILLATOR AND EXTERNAL COMPONENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
I/O PORTS of MCU MODULE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
MCU Port Operating Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
MCU BUS INTERFACE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Bus Read Cycles (PSEN or RD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Bus Write Cycles (WR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Controlling the PFQ and BC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
SUPERVISORY FUNCTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
External Reset Input Pin, RESET_IN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Low V Voltage Detect, LVD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
CC
Power-up Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
JTAG Debug Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Watchdog Timer, WDT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
STANDARD 8032 TIMER/COUNTERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Standard Timer SFRs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Clock Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
SFR, TCON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
SFR, TMOD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Timer 0 and Timer 1 Operating Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Timer 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
SERIAL UART INTERFACES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
UART Operation Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4/231
uPSD33xx
Serial Port Control Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
UART Baud Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
More About UART Mode 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
More About UART Mode 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
More About UART Modes 2 and 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
IrDA INTERFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Pulse Width Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
2
I C INTERFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
I2C Interface Main Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Communication Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Operating Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Bus Arbitration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Clock Synchronization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
General Call Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Serial I/O Engine (SIOE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
2
I C Interface Control Register (S1CON) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
2
I C Interface Status Register (S1STA). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
I2C Data Shift Register (S1DAT). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
2
I C Address Register (S1ADR). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
2
I C START Sample Setting (S1SETUP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
2
I C Operating Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
SPI (SYNCHRONOUS PERIPHERAL INTERFACE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
SPI Bus Features and Communication Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Full-Duplex Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Bus-Level Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
SPI SFR Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
SPI Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Dynamic Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
ANALOG-TO-DIGITAL CONVERTOR (ADC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Port 1 ADC Channel Selects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
PROGRAMMABLE COUNTER ARRAY (PCA) WITH PWM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
PCA Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
PCA Clock Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Operation of TCM Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Capture Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Timer Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Toggle Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
PWM Mode - (X8), Fixed Frequency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
PWM Mode - (X8), Programmable Frequency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
PWM Mode - Fixed Frequency, 16-bit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5/231
uPSD33xx
PWM Mode - Fixed Frequency, 10-bit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Writing to Capture/Compare Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Control Register Bit Definition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
TCM Interrupts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
PSD MODULE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
PSD Module Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
Memory Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
Runtime Control Register Definitions (csiop). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
PSD Module Detailed Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
PSD Module Reset Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
AC/DC PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
MAXIMUM RATING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
DC AND AC PARAMETERS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
PACKAGE MECHANICAL INFORMATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
PART NUMBERING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
REVISION HISTORY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
6/231
uPSD33xx
SUMMARY DESCRIPTION
The Turbo uPSD33xx Series combines a powerful
8051-based microcontroller with a flexible memory
structure, programmable logic, and a rich periph-
eral mix to form an ideal embedded controller. At
its core is a fast 4-cycle 8032 MCU with a 6-byte
instruction prefetch queue (PFQ) and a 4-entry ful-
ly associative branching cache (BC) to maximize
MCU performance, enabling loops of code in
smaller localities to execute extremely fast.
Code development is easily managed without a
hardware In-Circuit Emulator by using the serial
JTAG debug interface. JTAG is also used for In-
System Programming (ISP) in as little as 10 sec-
onds, perfect for manufacturing and lab develop-
ment. The 8032 core is coupled to Programmable
System Device (PSD) architecture to optimize the
8032 memory structure, offering two independent
banks of Flash memory that can be placed at vir-
tually any address within 8032 program or data ad-
dress space, and easily paged beyond 64K bytes
using on-chip programmable decode logic. Dual
Flash memory banks provide a robust solution for
remote product updates in the field through In-Ap-
plication Programming (IAP). Dual Flash banks
also support EEPROM emulation, eliminating the
need for external EEPROM chips. General pur-
pose programmable logic (PLD) is included to
build an endless variety of glue-logic, saving exter-
nal logic devices. The PLD is configured using the
software development tool, PSDsoft Express,
available from the web at www.st.com/psm, at no
charge. The uPSD33xx also includes supervisor
functions such as a programmable watchdog timer
and low-voltage reset.
Figure 2. Block Diagram
uPSD33xx
(3) 16-bit
Timer/
Counters
1st Flash Memory:
64K, 128K,
or 256K Bytes
Turbo
8032
Core
PFQ
&
BC
(2)
External
Interrupts
Programmable
Decode and
Page Logic
2nd Flash Memory:
I2C
16K or 32K Bytes
P3.0:7
SRAM:
2K, 8K, or 32K Bytes
UART0
(8) GPIO, Port A
PA0:7
(80-pin only)
(8) GPIO, Port 3
(8) GPIO, Port 1
(8) 10-bit ADC
General
PB0:7
PD1:2
(8) GPIO, Port B
(2) GPIO, Port D
(4) GPIO, Port C
Purpose
Programmable
Logic,
P1.0:7
16 Macrocells
PC0:7
Optional IrDA
Encoder/Decoder
UART1
JTAG ICE and ISP
MCU
Bus
8032 Address/Data/Control Bus
(80-pin device only)
SPI
16-bit PCA
Supervisor:
(6) PWM, CAPCOM, TIMER
Watchdog and Low-Voltage Reset
Dedicated
Pins
VCC, VDD, GND, Reset, Crystal In
(8) GPIO, Port 4
P4.0:7
AI08875
7/231
uPSD33xx
PIN DESCRIPTIONS
Figure 3. TQFP52 Connections
PD1/CLKIN 1
PC7 2
39 P1.5/SPIRXD(2)/ADC5
38 P1.4/SPICLK(2)/ADC4
37 P1.3/TXD1(IrDA)(2)/ADC3
36 P1.2/RXD1(IrDA)(2)/ADC2
35 P1.1/T2X(2)/ADC1
JTAG TDO 3
JTAG TDI 4
DEBUG 5
3.3V VCC
6
34 P1.0/T2(2)/ADC0
(1)
PC4/TERR 7
33 VDD
(1)
VDD
8
32 XTAL2
GND 9
PC3/TSTAT 10
PC2/VSTBY 11
JTAG TCK 12
JTAG TMS 13
31 XTAL1
30 P3.7/SCL
29 P3.6/SDA
28 P3.5/C1
27 P3.4/C0
AI07822
Note: 1. For 5V applications, V must be connected to a 5.0V source. For 3.3V applications, V must be connected to a 3.3V source.
DD
DD
2. These signals can be used on one of two different ports (Port 1 or Port 4) for flexibility. Default is Port1.
3. V and 3.3V AV are shared in the 52-pin package only. ADC channels must use AV as V for the 52-pin package.
REF
CC
CC
REF
8/231
uPSD33xx
Figure 4. TQFP80 Connections
60 P1.5/SPIRXD(2)/ADC5
PD2/CSI 1
P3.3/TG1/EXINT1 2
PD1/CLKIN 3
ALE 4
59 P1.4/SPICLK(2)/ADC4
58 P1.3/TXD1(IrDA)(2)/ADC3
57 MCU A11
56 P1.2/RXD1(IrDA)(2)/ADC2
PC7 5
55 MCU A10
JTAG TDO 6
JTAG TDI 7
DEBUG 8
54 P1.1/T2X(2)/ADC1
53 MCU A9
52 P1.0/T2(2)/ADC0
PC4/TERR 9
3.3V VCC 10
NC 11
51 MCU A8
(1)
50 VDD
(1)
49 XTAL2
VDD 12
48 XTAL1
GND 13
PC3/TSTAT 14
47 MCU AD7
46 P3.7/SCL
45 MCU AD6
44 P3.6/SDA
43 MCU AD5
42 P3.5/C1
41 MCU AD4
PC2/VSTBY 15
JTAG TCK 16
NC 17
SPISEL(2)/PCACLK1/P4.7 18
SPITXD(2)/TCM5/P4.6 19
JTAG TMS 20
AI07823
Note: NC = Not Connected
Note: 1. For 5V applications, V must be connected to a 5.0V source. For 3.3V applications, V must be connected to a 3.3V source.
DD
DD
2. These signals can be used on one of two different ports (Port 1 or Port 4) for flexibility. Default is Port1.
9/231
uPSD33xx
Table 2. Pin Definitions
Function
52-Pin
80-Pin
Signal
Name
Port Pin
In/Out
(1)
No.
No.
Basic
Alternate 1
Alternate 2
External Bus
MCUAD0
AD0
36
N/A
I/O
Multiplexed Address/
Data bus A0/D0
Multiplexed Address/
Data bus A1/D1
MCUAD1
MCUAD2
MCUAD3
MCUAD4
MCUAD5
MCUAD6
MCUAD7
MCUA8
MCUA9
MCUA10
MCUA11
P1.0
AD1
AD2
AD3
AD4
AD5
AD6
AD7
A8
37
38
39
41
43
45
47
51
53
55
57
52
54
56
58
59
60
61
64
75
77
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
34
I/O
I/O
I/O
I/O
I/O
I/O
I/O
O
Multiplexed Address/
Data bus A2/D2
Multiplexed Address/
Data bus A3/D3
Multiplexed Address/
Data bus A4/D4
Multiplexed Address/
Data bus A5/D5
Multiplexed Address/
Data bus A6/D6
Multiplexed Address/
Data bus A7/D7
External Bus, Addr
A8
External Bus, Addr
A9
A9
O
External Bus, Addr
A10
A10
A11
O
External Bus, Addr
A11
O
T2
ADC0
Timer 2 Count input ADC Channel 0
(T2) input (ADC0)
Timer 2 Trigger input ADC Channel 1
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
T2X
ADC1
P1.1
35
(T2X)
input (ADC1)
RxD1
ADC2
UART1 or IrDA
Receive (RxD1)
ADC Channel 2
input (ADC2)
P1.2
36
TXD1
ADC3
UART or IrDA
Transmit (TxD1)
ADC Channel 3
input (ADC3)
P1.3
37
SPICLK
ADC4
SPI Clock Out
(SPICLK)
ADC Channel 4
input (ADC4)
P1.4
38
SPIRxD
ADC6
SPI Receive
(SPIRxD)
ADC Channel 5
input (ADC5)
P1.5
39
SPITXD
ADC6
SPI Transmit
(SPITxD)
ADC Channel 6
input (ADC6)
P1.6
40
SPISEL
ADC7
SPI Slave Select
(SPISEL)
ADC Channel 7
input (ADC7)
P1.7
41
UART0 Receive
(RxD0)
P3.0
RxD0
TXD0
23
UART0 Transmit
(TxD0)
P3.1
24
Interrupt 0 input
General I/O port pin (EXTINT0)/Timer 0
gate control (TG0)
EXINT0
TGO
P3.2
79
25
I/O
Interrupt 1 input
General I/O port pin (EXTINT1)/Timer 1
gate control (TG1)
P3.3
P3.4
INT1
C0
2
26
27
I/O
I/O
40
General I/O port pin Counter 0 input (C0)
10/231
uPSD33xx
Function
52-Pin
Signal
Name
80-Pin
No.
Port Pin
P3.5
In/Out
I/O
(1)
No.
Basic
Alternate 1
Alternate 2
C1
42
28
29
General I/O port pin Counter 1 input (C1)
2
I C Bus serial data
P3.6
SDA
44
I/O
General I/O port pin
2
(I CSDA)
2
I C Bus clock
P3.7
SCL
46
30
I/O
General I/O port pin
2
(I CSCL)
T2
TCM0
Program Counter
Array0 PCA0-TCM0 (T2)
Timer 2 Count input
P4.0
P4.1
P4.2
P4.3
P4.4
P4.5
P4.6
P4.7
33
31
30
27
25
23
19
18
70
65
62
63
4
22
21
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I
General I/O port pin
T2X
TCM1
Timer 2 Trigger input
(T2X)
General I/O port pin PCA0-TCM1
General I/O port pin PCA0-TCM2
General I/O port pin PCACLK0
RXD1
TCM2
UART1 or IrDA
Receive (RxD1)
20
TXD1
PCACLK0
UART1 or IrDA
Transmit (TxD1)
18
SPICLK
TCM3
Program Counter
Array1 PCA1-TCM3 (SPICLK)
SPI Clock Out
17
General I/O port pin
SPIRXD
TCM4
SPI Receive
(SPIRxD)
16
General I/O port pin PCA1-TCM4
General I/O port pin PCA1-TCM5
General I/O port pin PCACLK1
SPI Transmit
(SPITxD)
SPITXD
15
SPISEL
PCACLK1
SPI Slave Select
(SPISEL)
14
Reference Voltage
input for ADC
V
N/A
N/A
N/A
N/A
N/A
44
REF
READ Signal,
external bus
RD
WR
O
WRITE Signal,
external bus
O
PSEN Signal,
external bus
PSEN
ALE
O
Address Latch
signal, external bus
O
Active low reset
input
RESET_IN
XTAL1
XTAL2
DEBUG
68
48
49
8
I
Oscillator input pin
for system clock
31
I
Oscillator output pin
for system clock
32
O
I/O to the MCU
Debug Unit
5
I/O
PA0
PA1
PA2
PA3
PA4
PA5
PA6
PA7
35
34
32
28
26
24
22
21
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
All Port A pins
support:
1. PLD Macro-cell
outputs, or
2. PLD inputs, or
3. Latched
Address Out
(A0-A7), or
4. Peripheral I/O
Mode
11/231
uPSD33xx
Function
52-Pin
Signal
Name
80-Pin
No.
Port Pin
In/Out
(1)
No.
Basic
Alternate 1
Alternate 2
PB0
PB1
80
78
76
74
73
71
67
66
20
16
52
51
50
49
48
46
43
42
13
12
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
General I/O port pin
JTAG pin (TMS)
All Port B pins
support:
1. PLD Macro-cell
outputs, or
PB2
PB3
2. PLD inputs, or
PB4
3. Latched
Address Out
(A0-A7)
PB5
PB6
PB7
JTAGTMS
JTAGTCK
TMS
TCK
I
JTAG pin (TCK)
SRAM Standby
voltage input
PLD Macrocell
output, or PLD input
V
STBY
PC2
15
11
I/O
General I/O port pin
(V
STBY
)
Optional JTAG
Status (TSTAT)
PLD, Macrocell
output, or PLD input
PC3
PC4
TSTAT
TERR
14
9
10
7
I/O
I/O
General I/O port pin
General I/O port pin
Optional JTAG
Status (TERR)
PLD, Macrocell
output, or PLD input
JTAGTDI
TDI
7
6
4
3
I
JTAG pin (TDI)
JTAG pin (TDO)
JTAGTDO
TDO
O
PLD, Macrocell
output, or PLD input
PC7
5
2
I/O
General I/O port pin
1. PLD I/O
PD1
CLKIN
CSI
3
1
I/O
General I/O port pin
2. Clock input to
PLD and APD
1. PLD I/O
PD2
1
N/A
I/O
General I/O port pin
2. Chip select ot
PSD Module
3.3V-V
V
- MCU Module
CC
10
72
6
CC
AV
Analog V Input
CC
47
CC
V
V
V
- PSD Module
- 3.3V for 3V
- 5V for 5V
DD
DD
DD
V
DD
12
50
8
3.3V or 5V
V
V
V
- PSD Module
- 3.3V for 3V
- 5V for 5V
DD
DD
DD
V
DD
33
3.3V or 5V
GND
GND
GND
NC
13
29
69
11
17
9
19
45
N/A
N/A
NC
Note: 1. N/A = Signal Not Available on 52-pin package.
12/231
uPSD33xx
uPSD33xx HARDWARE DESCRIPTION
The uPSD33xx has a modular architecture built
from a stacked die process. There are two die, one
is designated “MCU Module” in this document, and
the other is designated “PSD Module” (see Figure
5., page 14). In all cases, the MCU Module die op-
erates at 3.3V with 5V tolerant I/O. The PSD Mod-
ule is either a 3.3V die or a 5V die, depending on
the uPSD33xx device as described below.
producing a V
A, B, C, and D of the PSD Module are true 5V
ports.
For all 3.3V uPSD33xxV devices, a 3.3V MCU
Module is stacked with a 3.3V PSD Module. In this
case, a 3.3V uPSD33xx device needs to be sup-
of 2.4V min and V max). Ports
OH CC
plied with a single 3.3V voltage source at both V
CC
and V . I/O pins on Ports 3 and 4 are 5V tolerant
DD
The MCU Module consists of a fast 8032 core, that
operates with 4 clocks per instruction cycle, and
has many peripheral and system supervisor func-
tions. The PSD Module provides the 8032 with
multiple memories (two Flash and one SRAM) for
program and data, programmable logic for ad-
dress decoding and for general-purpose logic, and
additional I/O. The MCU Module communicates
with the PSD Module through internal address and
data busses (A8 – A15, AD0 – AD7) and control
signals (RD, WR, PSEN, ALE, RESET).
There are slightly different I/O characteristics for
each module. I/Os for the MCU module are desig-
nated as Ports 1, 3, and 4. I/Os for the PSD Mod-
ule are designated as Ports A, B, C, and D.
For all 5V uPSD33xx devices, a 3.3V MCU Module
is stacked with a 5V PSD Module. In this case, a
5V uPSD33xx device must be supplied with
and can be connected to external 5V peripherals
devices if desired. Ports A, B, C, and D of the PSD
Module are 3.3V ports, which are not tolerant to
external 5V devices.
Refer to Table 3 for port type and voltage source
requirements.
80-pin uPSD33xx devices provide access to 8032
address, data, and control signals on external pins
to connect external peripheral and memory devic-
es. 52-pin uPSD33xx devices do not provide ac-
cess to the 8032 system bus.
All non-volatile memory and configuration portions
of the uPSD33xx device are programmed through
the JTAG interface and no special programming
voltage is needed. This same JTAG port is also
used for debugging of the 8032 core at runtime
providing breakpoint, single-step, display, and
trace features. A non-volatile security bit may be
programmed to block all access via JTAG inter-
face for security. The security bit is defeated only
by erasing the entire device, leaving the device
blank and ready to use again.
3.3V
for the MCU Module and 5.0V
for the
CC
DD
PSD Module. Ports 3 and 4 of the MCU Module
are 3.3V ports with tolerance to 5V devices (they
can be directly driven by external 5V devices and
they can directly drive external 5V devices while
Table 3. Port Type and Voltage Source Combinations
V
for MCU
Module
V
for PSD
Module
Ports 3 and 4 on
MCU Module
Ports A, B, C, and D on
PSD Module
CC
DD
Device Type
5V:
uPSD33xx
3.3V
5.0V
3.3V
3.3V but 5V tolerant
3.3V but 5V tolerant
5V
3.3V:
uPSD33xxV
3.3V
3.3V. NOT 5V tolerant
13/231
uPSD33xx
Figure 5. uPSD33xx Functional Modules
Port 3
I2C
Port 3 - UART0,
Intr, Timers
Port 4 - PCA,
PWM, UART1
Port 1 - Timer, ADC, SPI
MCU Module
Port 3
Port 1
VCC Pins
3.3V
Turbo 8032 Core
XTAL
Clock Unit
PCA
PWM
Counters
I2C
Unit
10-bit
ADC
SPI
Dual
3 Timer /
UARTs
Counters
Interrupt
256 Byte SRAM
Ext.
Bus
8032 Internal Bus
Dedicated Memory
Interface Prefetch,
Branch Cache
Reset Input
Reset
Pin
LVD
JTAG
DEBUG
Reset Logic
Internal
Reset
WDT
8-Bit Die-to-Die Bus
PSD
Reset
Enhanced MCU Interface
PSD Page Register
Secondary
Flash
PSD Module
SRAM
Main Flash
Decode PLD
PSD Internal Bus
VDD Pins
3.3V or 5V
JTAG ISP
CPLD - 16 MACROCELLS
Port C
JTAG and
GPIO
uPSD33XX
Port D
GPIO
Port A,B,C PLD
I/O and GPIO
AI07842
14/231
uPSD33xx
MEMORY ORGANIZATION
The 8032 MCU core views memory on the MCU
module as “internal” memory and it views memory
on the PSD module as “external” memory, see
Figure 6.
Internal memory on the MCU Module consists of
DATA, IDATA, and SFRs. These standard 8032
memories reside in 384 bytes of SRAM located at
a fixed address space starting at address 0x0000.
External memory on the PSD Module consists of
four types: main Flash (64K, 128K, or 256K bytes),
a smaller secondary Flash (16K, or 32K), SRAM
(2K, 8K, or 32K bytes), and a block of PSD Module
control registers called CSIOP (256 bytes). These
external memories reside at programmable ad-
dress ranges, specified using the software tool
PSDsoft Express. See the PSD Module section of
this document for more details on these memories.
dress space is for data memory. Program memory
is accessed using the 8032 signal, PSEN. Data
memory is accessed using the 8032 signals, RD
and WR. If the 8032 needs to access more than
64K bytes of external program or data memory, it
must use paging (or banking) techniques provided
by the Page Register in the PSD Module.
Note: When referencing program and data mem-
ory spaces, it has nothing to do with 8032 internal
SRAM areas of DATA, IDATA, and SFR on the
MCU Module. Program and data memory spaces
only relate to the external memories on the PSD
Module.
External memory on the PSD Module can overlap
the internal SRAM memory on the MCU Module in
the same physical address range (starting at
0x0000) without interference because the 8032
core does not assert the RD or WR signals when
accessing internal SRAM.
External memory is accessed by the 8032 in two
separate 64K byte address spaces. One address
space is for program memory and the other ad-
Figure 6. uPSD33xx Memories
Internal SRAM on
MCU Module
External Memory on
PSD Module
Main
Flash
• External memories may be placed at virtually
any address using software tool PSDsoft Express.
Fixed
Addresses
384 Bytes SRAM
• The SRAM and Flash memories may be placed
in 8032 Program Space or Data Space using
PSDsoft Express.
FF
Indirect
128 Bytes
Direct
Addressing
• Any memory in 8032 Data Space is XDATA.
SFR
IDATA
64KB,
128KB,
or
Secondary
Flash
SRAM
Addressing
128 Bytes
80
7F
256KB
128 Bytes
2KB,
8KB,
or
16KB
or
32KB
CSIOP
DATA
32KB
256 Bytes
Direct or Indirect Addressing
0
AI07843
15/231
uPSD33xx
Internal Memory (MCU Module, Standard 8032
Memory: DATA, IDATA, SFR)
Program Memory. External program memory is
addressed by the 8032 using its 16-bit Program
Counter (PC) and is accessed with the 8032 sig-
nal, PSEN. Program memory can be present at
any address in program space between 0x0000
and 0xFFFF.
DATA Memory. The first 128 bytes of internal
SRAM ranging from address 0x0000 to 0x007F
are called DATA, which can be accessed using
8032 direct or indirect addressing schemes and
are typically used to store variables and stack.
Four register banks, each with 8 registers (R0 –
R7), occupy addresses 0x0000 to 0x001F. Only
one of these four banks may be enabled at a time.
The next 16 locations at 0x0020 to 0x002F contain
128 directly addressable bit locations that can be
used as software flags. SRAM locations 0x0030
and above may be used for variables and stack.
IDATA Memory. The next 128 bytes of internal
SRAM are named IDATA and range from address
0x0080 to 0x00FF. IDATA can be accessed only
through 8032 indirect addressing and is typically
used to hold the MCU stack as well as data vari-
ables. The stack can reside in both DATA and
IDATA memories and reach a size limited only by
the available space in the combined 256 bytes of
these two memories (since stack accesses are al-
ways done using indirect addressing, the bound-
ary between DATA and IDATA does not exist with
regard to the stack).
After a power-up or reset, the 8032 begins pro-
gram execution from location 0x0000 where the
reset vector is stored, causing a jump to an initial-
ization routine in firmware. At address 0x0003, just
following the reset vector are the interrupt service
locations. Each interrupt is assigned a fixed inter-
rupt service location in program memory. An inter-
rupt causes the 8032 to jump to that service
location, where it commences execution of the
service routine. External Interrupt 0 (EXINT0), for
example, is assigned to service location 0x0003. If
EXINT0 is going to be used, its service routine
must begin at location 0x0003. Interrupt service lo-
cations are spaced at 8-byte intervals: 0x0003 for
EXINT0, 0x000B for Timer 0, 0x0013 for EXINT1,
and so forth. If an interrupt service routine is short
enough, it can reside entirely within the 8-byte in-
terval. Longer service routines can use a jump in-
struction to somewhere else in program memory.
Data Memory. External data is referred to as
XDATA and is addressed by the 8032 using Indi-
rect Addressing via its 16-bit Data Pointer Register
(DPTR) and is accessed by the 8032 signals, RD
and WR. XDATA can be present at any address in
data space between 0x0000 and 0xFFFF.
Note: the uPSD33xx has dual data pointers
(source and destination) making XDATA transfers
much more efficient.
SFR Memory. Special Function Registers (Table
5., page 24) occupy a separate physical memory,
but they logically overlap the same 128 bytes as
IDATA, ranging from address 0x0080 to 0x00FF.
SFRs are accessed only using direct addressing.
There 86 active registers used for many functions:
changing the operating mode of the 8032 MCU
core, controlling 8032 peripherals, controlling I/O,
and managing interrupt functions. The remaining
unused SFRs are reserved and should not be ac-
cessed.
16 of the SFRs are both byte- and bit-addressable.
Bit-addressable SFRs are those whose address
ends in “0” or “8” hex.
External Memory (PSD Module: Program
memory, Data memory)
Memory Placement. PSD Module architecture
allows the placement of its external memories into
different combinations of program memory and
data memory spaces. This means the main Flash,
the secondary Flash, and the SRAM can be
viewed by the 8032 MCU in various combinations
of program memory or data memory as defined by
PSDsoft Express.
As an example of this flexibility, for applications
that require a great deal of Flash memory in data
space (large lookup tables or extended data re-
cording), the larger main Flash memory can be
placed in data space and the smaller secondary
Flash memory can be placed in program space.
The opposite can be realized for a different appli-
cation if more Flash memory is needed for code
and less Flash memory for data.
The PSD Module has four memories: main Flash,
secondary Flash, SRAM, and CSIOP. See the
PSD MODULE section for more detailed informa-
tion on these memories.
Memory mapping in the PSD Module is imple-
mented with the Decode PLD (DPLD) and option-
ally the Page Register. The user specifies decode
equations for individual segments of each of the
memories using the software tool PSDsoft Ex-
press. This is a very easy point-and-click process
allowing total flexibility in mapping memories. Ad-
ditionally, each of the memories may be placed in
various combinations of 8032 program address
space or 8032 data address space by using the
software tool PSDsoft Express.
16/231
uPSD33xx
By default, the SRAM and CSIOP memories on
the PSD Module must always reside in data mem-
ory space and they are treated by the 8032 as
XDATA. However, the SRAM may optionally re-
side in program space in addition to data space if
it is desired to execute code from SRAM. The main
Flash and secondary Flash memories may reside
in program space, data space, or both.
These memory placement choices specified by
PSDsoft Express are programmed into non-vola-
tile sections of the uPSD33xx, and are active at
power-up and after reset. It is possible to override
these initial settings during runtime for In-Applica-
tion Programming (IAP).
Standard 8032 MCU architecture cannot write to
its own program memory space to prevent acci-
dental corruption of firmware. However, this be-
comes an obstacle in typical 8032 systems when
a remote update to firmware in Flash memory is
required using IAP. The PSD module provides a
solution for remote updates by allowing 8032 firm-
ware to temporarily “reclassify” Flash memory to
reside in data space during a remote update, then
returning Flash memory back to program space
when finished. See the VM Register (Table
78., page 143) in the PSD Module section of this
document for more details.
8032 MCU CORE PERFORMANCE ENHANCEMENTS
Before describing performance features of the
uPSD33xx, let us first look at standard 8032 archi-
tecture. The clock source for the 8032 MCU cre-
ates a basic unit of timing called a machine-cycle,
which is a period of 12 clocks for standard 8032
MCUs. The instruction set for traditional 8032
MCUs consists of 1, 2, and 3 byte instructions that
execute in different combinations of 1, 2, or 4 ma-
chine-cycles. For example, there are one-byte in-
structions that execute in one machine-cycle (12
clocks), one-byte instructions that execute in four
machine-cycles (48 clocks), two-byte, two-cycle
instructions (24 clocks), and so on. In addition,
standard 8032 architecture will fetch two bytes
from program memory on almost every machine-
cycle, regardless if it needs them or not (dummy
fetch). This means for one-byte, one-cycle instruc-
tions, the second byte is ignored. These one-byte,
one-cycle instructions account for half of the
8032's instructions (126 out of 255 opcodes).
There are inefficiencies due to wasted bus cycles
and idle bus times that can be eliminated.
8032 (all opcodes, the number of bytes per in-
struction, and the native number a machine-cycles
per instruction are identical to the original 8032).
The first way performance is boosted is by reduc-
ing the machine-cycle period to just 4 MCU clocks
as compared to 12 MCU clocks in a standard
8032. This shortened machine-cycle improves the
instruction rate for one-byte, one-cycle instruc-
tions by a factor of three (Figure 7., page 18) com-
pared to standard 8051 architectures, and
significantly improves performance of multiple-cy-
cle instruction types.
The example in Figure 7 shows a continuous exe-
cution stream of one-byte, one-cycle instructions.
The 5V uPSD33xx will yield 10 MIPS peak perfor-
mance in this case while operating at 40MHz clock
rate. In a typical application however, the effective
performance will be lower since programs do not
use only one-cycle instructions, but special tech-
niques are implemented in the uPSD33xx to keep
the effective MIPS rate as close as possible to the
peak MIPS rate at all times. This is accomplished
with an instruction Pre-Fetch Queue (PFQ) and a
Branch Cache (BC) as shown in Figure
8., page 18.
The uPSD33xx 8032 MCU core offers increased
performance in a number of ways, while keeping
the exact same instruction set as the standard
17/231
uPSD33xx
Figure 7. Comparison of uPSD33xx with Standard 8032 Performance
1-byte, 1-Cycle Instructions
Instruction A
Instruction B
Instruction C
Execute Instruction and
Pre-Fetch Next Instruction
Execute Instruction and
Pre-Fetch Next Instruction
Execute Instruction and
Pre-Fetch Next Instruction
Turbo uPSD33XX
MCU Clock
4 clocks (one machine cycle)
one machine cycle
one machine cycle
12 clocks (one machine cycle)
Instruction A
Execute Instruction A
and Fetch a Second Dummy Byte
Fetch Byte for Instruction A
Standard 8032
Dummy Byte is Ignored (wasted bus access)
Turbo uPSD33XX executes instructions A, B, and C in the same
amount of time that a standard 8032 executes only instruction A.
AI08808
Figure 8. Instruction Pre-Fetch Queue and Branch Cache
Branch 4 Branch 4 Branch 4 Branch 4 Branch 4 Branch 4
Previous
Branch 4
Code
Branch 3 Branch 3 Branch 3 Branch 3 Branch 3 Branch 3
Code
Code Code Code Code Code
Branch 2
Code
Code
Code
Code
Code
Previous
Branch 3
Previous
Branch 2
Previous
Compare
Branch
Cache
(BC)
Branch 2 Branch 2 Branch 2 Branch 2 Branch 2
Code
Code Code Code Code Code
Branch 1 Branch 1 Branch 1 Branch 1 Branch 1 Branch 1
Code
Code Code Code Code Code
Branch 1
Address
Load on Branch Address Match
Current
Branch
Address
Instruction
Byte
Instruction
Byte
8032
MCU
Program
Memory on
PSD Module
8
8
6 Bytes of Instruction
Address
Address
16
16
Instruction Pre-Fetch Queue (PFQ)
Wait
Stall
AI08809
18/231
uPSD33xx
Pre-Fetch Queue (PFQ) and Branch Cache
(BC)
PFQ Example, Multi-cycle Instructions
Let us look at a string of two-byte, two-cycle in-
structions in Figure 9., page 20. There are three
instructions executed sequentially in this example,
instructions A, B, and C. Each of the time divisions
in the figure is one machine-cycle of four clocks,
and there are six phases to reference in this dis-
cussion. Each instruction is pre-fetched into the
PFQ in advance of execution by the MCU. Prior to
Phase 1, the PFQ has pre-fetched the two instruc-
tion bytes (A1 and A2) of instruction A. During
Phase one, both bytes are loaded into the MCU
execution unit. Also in Phase 1, the PFQ is pre-
fetching the first byte (B1) of instruction B from
program memory. In Phase 2, the MCU is pro-
cessing Instruction A internally while the PFQ is
pre-fetching the second byte (B2) of Instruction B.
In Phase 3, both bytes of instruction B are loaded
into the MCU execution unit and the PFQ begins
to pre-fetch bytes for the third instruction C. In
Phase 4 Instruction B is processed and the pre-
fetching continues, eliminating idle bus cycles and
feeding a continuous flow of operands and op-
codes to the MCU execution unit.
The PFQ is always working to minimize the idle
bus time inherent to 8032 MCU architecture, to
eliminate wasted memory fetches, and to maxi-
mize memory bandwidth to the MCU. The PFQ
does this by running asynchronously in relation to
the MCU, looking ahead to pre-fetch code from
program memory during any idle bus periods. Only
necessary bytes will be fetched (no dummy fetch-
es like standard 8032). The PFQ will queue up to
six code bytes in advance of execution, which sig-
nificantly optimizes sequential program perfor-
mance. However, when program execution
becomes non-sequential (program branch), a typ-
ical pre-fetch queue will empty itself and reload
new code, causing the MCU to stall. The Turbo
uPSD33xx diminishes this problem by using a
Branch Cache with the PFQ. The BC is a four-way,
fully associative cache, meaning that when a pro-
gram branch occurs, it's branch destination ad-
dress is compared simultaneously with four recent
previous branch destinations stored in the BC.
Each of the four cache entries contain up to six
bytes of code related to a branch. If there is a hit
(a match), then all six code bytes of the matching
program branch are transferred immediately and
simultaneously from the BC to the PFQ, and exe-
cution on that branch continues with minimal de-
lay. This greatly reduces the chance that the MCU
will stall from an empty PFQ, and improves perfor-
mance in embedded control systems where it is
quite common to branch and loop in relatively
small code localities.
The uPSD33xx MCU instructions are an exact 1/3
scale of all standard 8032 instructions with regard
to number of cycles per instruction. Figure
10., page 20 shows the equivalent instruction se-
quence from the example above on a standard
8032 for comparison.
Aggregate Performance
The stream of two-byte, two-cycle instructions in
Figure 9., page 20, running on a 40MHz, 5V,
uPSD33xx will yield 5 MIPs. And we saw the
stream of one-byte, one-cycle instructions in Fig-
ure 7., page 18, on the same MCU yield 10 MIPs.
Effective performance will depend on a number of
things: the MCU clock frequency; the mixture of in-
structions types (bytes and cycles) in the applica-
tion; the amount of time an empty PFQ stalls the
MCU (mix of instruction types and misses on
Branch Cache); and the operating voltage. A 5V
uPSD33xx device operates with four memory wait
states, but a 3.3V device operates with five mem-
ory wait states yielding 8 MIPS peak compared to
10 MIPs peak for 5V device. The same number of
wait states will apply to both program fetches and
to data READ/WRITEs unless otherwise specified
in the SFR named BUSCON.
By default, the PFQ and BC are enabled after
power-up or reset. The 8032 can disable the PFQ
and BC at runtime if desired by writing to a specific
SFR (BUSCON).
The memory in the PSD module operates with
variable wait states depending on the value spec-
ified in the SFR named BUSCON. For example, a
5V uPSD33xx device operating at a 40MHz crystal
frequency requires four memory wait states (equal
to four MCU clocks). In this example, once the
PFQ has one or more bytes of code, the wait
states become transparent and a full 10 MIPS is
achieved when the program stream consists of se-
quential one-byte, one machine-cycle instructions
as shown in Figure 7., page 18 (transparent be-
cause a machine-cycle is four MCU clocks which
equals the memory pre-fetch wait time that is also
four MCU clocks). But it is also important to under-
stand PFQ operation on multi-cycle instructions.
In general, a 3X aggregate performance increase
is expected over any standard 8032 application
running at the same clock frequency.
19/231
uPSD33xx
Figure 9. PFQ Operation on Multi-cycle Instructions
Three 2-byte, 2-cycle Instructions on uPSD33XX
Pre-Fetch Inst A
Pre-Fetch Inst B Pre-Fetch Inst C
Inst A, Byte 1 Inst A, Byte 2 Inst B, Byte 1 Inst B, Byte 2 Inst C, Byte 1 Inst C, Byte 2 Continue to Pre-Fetch
PFQ
4-clock
Macine Cycle
Phase 1
A1 A2
Phase 2
Phase 3
B1 B2
Phase 4
Phase 5
C1 C2
Phase 6
Previous Instruction
Process A
Process B
Process C
Next Inst
AI08810
MCU
Execution
Instruction A
Instruction B
Instruction C
Figure 10. uPSD33xx Multi-cycle Instructions Compared to Standard 8032
Three 2-byte, 2-cycle Instructions, uPSD33XX vs. Standard 8032
24 Clocks Total (4 clocks per cycle)
C1
C2 Inst C
A1 A2 Inst A
B1 B2
Inst B
uPSD33XX
Std 8032
1 Cycle
72 Clocks (12 clocks per cycle)
Byte 1 Byte 2 Process Inst B
Byte 1
Byte 1 Byte 2
Process Inst A
1 Cycle
Byte 2 Process Inst C
AI08811
20/231
uPSD33xx
MCU MODULE DISCRIPTION
This section provides a detail description of the
MCU Module system functions and peripherals, in-
cluding:
■
■
■
■
■
■
■
■
■
■
I/O Ports
MCU Bus Interface
Supervisory Functions
Standard 8032 Timer/Counters
Serial UART Interfaces
IrDA Interface
■
■
■
■
■
■
■
■
■
■
8032 MCU Registers
Special Function Registers
8032 Addressing Modes
uPSD33xx Instruction Set Summary
Dual Data Pointers
Debug Unit
Interrupt System
2
I C Interface
SPI Interface
Analog to Digital Converter
Programmable Counter Array (PCA)
MCU Clock Generation
Power Saving Modes
Note: A full description of the 8032 instruction set
may be found in the uPSD33xx Programmers
Guide.
Oscillator and External Components
8032 MCU REGISTERS
The uPSD33xx has the following 8032 MCU core
registers, also shown in Figure 11.
Very frequently, the DPTR Register is used to ac-
cess XDATA using the External Direct addressing
mode. The uPSD33xx has a special set of SFR
registers (DPTC, DPTM) to control a secondary
DPTR Register to speed memory-to-memory
XDATA transfers. Having dual DPTR Registers al-
lows rapid switching between source and destina-
tion addresses (see details in DUAL DATA
POINTERS, page 37).
Figure 11. 8032 MCU Registers
Accumulator
B Register
A
B
Stack Pointer
SP
PCL
Program Counter
PCH
Program Counter (PC)
Program Status Word
General Purpose
Register (Bank0-3)
PSW
R0-R7
The PC is a 16-bit register consisting of two 8-bit
registers, PCL and PCH. This counter indicates
the address of the next instruction in program
memory to be fetched and executed. A reset forc-
es the PC to location 0000h, which is where the re-
set jump vector is stored.
DPTR(DPH) DPTR(DPL) Data Pointer Register
AI06636
Stack Pointer (SP)
Accumulator (ACC)
The SP is an 8-bit register which holds the current
location of the top of the stack. It is incremented
before a value is pushed onto the stack, and dec-
remented after a value is popped off the stack. The
SP is initialized to 07h after reset. This causes the
stack to begin at location 08h (top of stack). To
avoid overlapping conflicts, the user must initialize
the top of the stack to 20h if all four banks of reg-
isters R0 - R7 are used, and the user must initialize
the top of stack to 30h if all of the 8032 bit memory
locations are used.
This is an 8-bit general purpose register which
holds a source operand and receives the result of
arithmetic operations. The ACC Register can also
be the source or destination of logic and data
movement operations. For MUL and DIV instruc-
tions, ACC is combined with the B Register to hold
16-bit operands. The ACC is referred to as “A” in
the MCU instruction set.
B Register (B)
The B Register is a general purpose 8-bit register
for temporary data storage and also used as a 16-
bit register when concatenated with the ACC Reg-
ister for use with MUL and DIV instructions.
Data Pointer (DPTR)
DPTR is a 16-bit register consisting of two 8-bit
registers, DPL and DPH. The DPTR Register is
used as a base register to create an address for in-
direct jumps, table look-up operations, and for ex-
ternal data transfers (XDATA). When not used for
addressing, the DPTR Register can be used as a
general purpose 16-bit data register.
21/231
uPSD33xx
General Purpose Registers (R0 - R7)
There are four banks of eight general purpose 8-
bit registers (R0 - R7), but only one bank of eight
registers is active at any given time depending on
the setting in the PSW word (described next). R0 -
R7 are generally used to assist in manipulating
values and moving data from one memory location
to another. These register banks physically reside
in the first 32 locations of 8032 internal DATA
SRAM, starting at address 00h. At reset, only the
first bank of eight registers is active (addresses
00h to 07h), and the stack begins at address 08h.
General Purpose Flag (F0). This is a bit-addres-
sable, general-purpose flag for use under software
control.
Register Bank Select Flags (RS1, RS0). These
bits select which bank of eight registers is used
during R0 - R7 register accesses (see Table 4)
Overflow Flag (OV). The OV flag is set when: an
ADD, ADDC, or SUBB instruction causes a sign
change; a MUL instruction results in an overflow
(result greater than 255); a DIV instruction causes
a divide-by-zero condition. The OV flag is cleared
by the ADD, ADDC, SUBB, MUL, and DIV instruc-
tions in all other cases. The CLRV instruction will
clear the OV flag at any time.
Program Status Word (PSW)
The PSW is an 8-bit register which stores several
important bits, or flags, that are set and cleared by
many 8032 instructions, reflecting the current
state of the MCU core. Figure 12., page 22 shows
the individual flags.
Parity Flag (P). The P flag is set if the sum of the
eight bits in the Accumulator is odd, and P is
cleared if the sum is even.
Carry Flag (CY). This flag is set when the last
arithmetic operation that was executed results in a
carry (addition) or borrow (subtraction). It is
cleared by all other arithmetic operations. The CY
flag is also affected by Shift and Rotate Instruc-
tions.
Auxiliary Carry Flag (AC). This flag is set when
the last arithmetic operation that was executed re-
sults in a carry into (addition) or borrow from (sub-
traction) the high-order nibble. It is cleared by all
other arithmetic operations.
Table 4. .Register Bank Select Addresses
Register
Bank
8032 Internal
DATA Address
RS1
RS0
0
0
1
1
0
1
0
1
0
1
2
3
00h - 07h
08h - 0Fh
10h - 17h
18h - 1Fh
Figure 12. Program Status Word (PSW) Register
MSB
LSB
CY AC FO RS1 RS0 OV
P
Reset Value 00h
Parity Flag
PSW
Carry Flag
Auxillary Carry Flag
Bit not assigned
Overflow Flag
General Purpose Flag
Register Bank Select Flags
(to select Bank0-3)
AI06639
22/231
uPSD33xx
SPECIAL FUNCTION REGISTERS (SFR)
A group of registers designated as Special Func-
tion Register (SFR) is shown in Table 5., page 24.
SFRs control the operating modes of the MCU
core and also control the peripheral interfaces and
I/O pins on the MCU Module. The SFRs can be ac-
cessed only by using the Direct Addressing meth-
od within the address range from 80h to FFh of
internal 8032 SRAM. Sixteen addresses in SFR
address space are both byte- and bit-addressable.
The bit-addressable SFRs are noted in Table 5.
86 of a possible 128 SFR addresses are occupied.
The remaining unoccupied SFR addresses (desig-
nated as “RESERVED” in Table 5) should not be
written. Reading unoccupied locations will return
an undefined value.
SCON0, SBUF0, SCON1, SBUF1
■
■
■
■
Power, clock, and bus timing registers
PCON, CCON0, BUSCON
Hardware watchdog timer registers
WDKEY, WDRST
Interrupt system registers
IP, IPA, IE, IEA
Prog. Counter Array (PCA) control
registers
PCACL0, PCACH0, PCACON0, PCASTA,
PCACL1, PCACH1, PCACON1, CCON2,
CCON3
■
PCA capture/compare and PWM registers
Note: There is a separate set of control registers
for the PSD Module, designated as csiop, and they
are described in the PSD MODULE, page 133.
The I/O pins, PLD, and other functions on the PSD
Module are NOT controlled by SFRs.
CAPCOML0, CAPCOMH0, TCMMODE0,
CAPCOML1, CAPCOMH1, TCMMODE2,
CAPCOML2, CAPCOMH2, TCMMODE2,
CAPCOML3, CAPCOMH3, TCMMODE3,
CAPCOML4, CAPCOMH4, TCMMODE4,
CAPCOML5, CAPCOMH5, TCMMODE5,
PWMF0, PMWF1
SFRs are categorized as follows:
■
■
■
■
MCU core registers:
IP, A, B, PSW, SP, DPTL, DPTH, DPTC,
DPTM
■
SPI interface registers
SPICLKD, SPISTAT, SPITDR, SPIRDR,
SPICON0, SPICON1
MCU Module I/O Port registers:
P1, P3, P4, P1SFS0, P1SFS1, P3SFS,
P4SFS0, P4SFS1
Standard 8032 Timer registers
TCON, TMOD, T2CON, TH0, TH1, TH2, TL0,
TL1, TL2, RCAP2L, RCAP2H
2
■
■
■
I C interface registers
S1SETUP, S1CON, S1STA, S1DAT, S1ADR
Analog to Digital Converter registers
ACON, ADCPS, ADAT0, ADAT1
IrDA interface register
Standard Serial Interfaces (UART)
IRDACON
23/231
uPSD33xx
Table 5. SFR Memory Map with Direct Address and Reset Value
SFR
Addr
(hex)
Bit Name and <Bit Address>
Reset
Value Descr.
(hex) with Link
Reg.
SFR
Name
7
6
5
4
3
2
1
0
80
RESERVED
Stack
Pointer
(SP), page
81
SP
SP[7:0]
07
21
82
83
84
DPL
DPH
DPL[7:0]
00
00
Data
Pointer
(DPTR), p
age 21
DPH[7:0]
RESERVED
Table
13., page
37
85
86
87
DPTC
DPTM
PCON
TCON
TMOD
–
–
AT
–
–
–
–
–
–
–
DPSEL[2:0]
00
00
00
00
00
Table
14., page
38
MD1[1:0]
MD0[1:0]
Table
24., page
50
SMOD0 SMOD1
POR
RCLK1 TCLK1
PD
IE0
IDLE
IT0
Table
39., page
70
TF1
<8Fh>
TR1
<8Eh>
TF0
<8Dh>
TR0
<8Ch>
IE1
<8Bh>
IT1
(1)
88
<8Ah> <89h> <88h>
Table
40., page
72
89
GATE
C/T
M1
M0
GATE
C/T
M1
M0
8A
8B
8C
8D
TL0
TL1
TH0
TH1
TL0[7:0]
00
00
00
00
Standard
Timer
SFRs, pag
TL1[7:0]
TH0[7:0]
TH1[7:0]
e 69
Table
29., page
60
8E
8F
P1SFS0
P1SFS1
P1
P1SFS0[7:0]
P1SFS1[7:0]
00
00
FF
00
00
00
Table
30., page
60
Table
25., page
57
P1.7
<97h>
P1.6
<96h>
P1.5
<95h>
P1.4
<94h>
P1.3
<93h>
P1.2
P1.1
P1.0
(1)
90
<92h> <91h> <90h>
Table
28., page
60
91
92
93
P3SFS
P4SFS0
P4SFS1
P3SFS[7:0]
P4SFS0[7:0]
P4SFS1[7:0]
Table
32., page
61
Table
33., page
61
24/231
uPSD33xx
SFR
Addr
(hex)
Bit Name and <Bit Address>
Reset
Reg.
Value Descr.
(hex) with Link
SFR
Name
7
6
5
4
3
2
1
0
Table
94
95
96
97
ADCPS
ADAT0
ADAT1
ACON
–
–
–
–
ADCCE
ADCPS[2:0]
00
00
00
00
00
00
64., page
122
Table
65., page
122
ADATA[7:0]
–
Table
66., page
122
–
–
–
–
–
ADATA[9:8]
Table
63., page
121
AINTF AINTEN
ADEN
ADS[2:0]
ADST ADSF
Table
45., page
82
SM0
<9Fh>
SM1
<9Eh>
SM2
<9Dh>
REN
<9Ch>
TB8
<9Bh>
RB8
TI
RI
(1)
SCON0
SBUF0
98
<9Ah> <99h> <9h8>
Figure
25., page
79
99
SBUF0[7:0]
RESERVED
9A
9B
9C
RESERVED
RESERVED
Table
35., page
63
9D
BUSCON
EPFQ
EBC
WRW1
WRW0
RDW1
RDW0
CW1
CW0
EB
9E
9F
A0
A1
RESERVED
RESERVED
RESERVED
RESERVED
Table
67., page
124
A2
A3
PCACL0
PCACH0
PCACL0[7:0]
00
00
00
00
00
00
Table
67., page
124
PCACH0[7:0]
Table
70., page
129
A4 PCACON0 EN_ALL EN_PCA EOVF1 PCA_IDL
–
–
CLK_SEL[1:0]
INTF1 INTF0
Table
72., page
131
A5
A6
A7
PCASTA
WDTRST
IEA
OVF1
INTF5
INTF4
INTF3
OVF0
INTF2
Table
38., page
68
WDTRST[7:0]
ES1
Table
18., page
44
EADC
ESPI
EPCA
–
–
EI2C
–
25/231
uPSD33xx
SFR
Addr
(hex)
Bit Name and <Bit Address>
Reset
Value Descr.
(hex) with Link
Reg.
SFR
Name
7
6
5
4
3
2
1
0
Table
17., page
EA
<AFh>
ET2
<ADh>
ES0
<ACh>
ET1
EX1
ET0
EX0
(1)
IE
–
00
A8
<ABh> <AAh> <A9h> <A8h>
43
TCMMODE
0
A9
AA
AB
AC
AD
EINTF E_COMP CAP_PE CAP_NE MATCH TOGGLE
EINTF E_COMP CAP_PE CAP_NE MATCH TOGGLE
EINTF E_COMP CAP_PE CAP_NE MATCH TOGGLE
CAPCOML0[7:0]
PWM[1:0]
PWM[1:0]
PWM[1:0]
00
00
00
00
00
Table
73., page
132
TCMMODE
1
TCMMODE
2
CAPCOML
0
Table
67., page
124
CAPCOMH
0
CAPCOMH0[7:0]
Table
37., page
68
AE WDTKEY
WDTKEY[7:0]
55
00
FF
Table
67., page
124
CAPCOML
AF
CAPCOML1[7:0]
1
Table
26., page
58
P3.7
<B7h>
P3.6
<B6h>
P3.5
<B5h>
P3.4
P3.3
P3.2
P3.1
P3.0
(1)
P3
B0
<B4h>
<B3h>
<B2h> <B1h> <B0h>
CAPCOMH
1
B1
CAPCOMH1[7:0]
CAPCOML2[7:0]
CAPCOMH2[7:0]
00
00
CAPCOML
2
Table
67., page
124
B2
B3
CAPCOMH
2
00
00
B4
B5
B6
PWMF0
PWMF0[7:0]
RESERVED
RESERVED
Table
20., page
45
B7
IPA
IP
PADC
–
PSPI
–
PPCA
PS1
–
–
PI2C
PT0
–
00
00
Table
19., page
44
PT2
<BDh>
PS0
<BCh>
PT1
PX1
PX0
(1)
B8
<BBh> <BAh> <B9h> <B8h>
B9
RESERVED
BA
PCACL1
PCACH1
PCACL1[7:0]
PCACH1[7:0]
00
00
Table
67., page
124
BB
Table
71., page
130
BC PCACON1
–
EN_PCA EOVF1 PCA_IDL
–
–
CLK_SEL[1:0]
00
26/231
uPSD33xx
SFR
Addr
(hex)
Bit Name and <Bit Address>
Reset
Reg.
Value Descr.
(hex) with Link
SFR
Name
7
6
5
4
3
2
1
0
TCMMODE
3
BD
BE
BF
EINTF E_COMP CAP_PE CAP_NE MATCH TOGGLE
EINTF E_COMP CAP_PE CAP_NE MATCH TOGGLE
EINTF E_COMP CAP_PE CAP_NE MATCH TOGGLE
PWM[1:0]
PWM[1:0]
PWM[1:0]
00
Table
73., page
132
TCMMODE
4
00
00
TCMMODE
5
Table
27., page
58
P4.7
<C7h>
P4.6
<C6h>
P4.5
<C5h>
P4.4
<C4h>
P4.3
<C3h>
P4.2
P4.1
P4.0
(1)
P4
FF
C0
<C2h> <C1h> <C0h>
CAPCOML
3
C1
C2
C3
C4
C5
CAPCOML3[7:0]
CAPCOMH3[7:0]
CAPCOML4[7:0]
CAPCOMH4[7:0]
CAPCOML5[7:0]
00
00
00
00
00
CAPCOMH
3
CAPCOML
4
Table
67., page
124
CAPCOMH
4
CAPCOML
5
CAPCOMH
5
C6
C7
CAPCOMH5[7:0]
PWMF1[7:0]
00
00
PWMF1
CP/
RL2
<C8h>
Table
41., page
75
TF2
EXF2
RCLK
<CDh>
TCLK
EXEN2
TR2
C/T2
(1)
T2CON
00
C8
<CFh> <CEh>
<CCh> <CBh> <CAh> <C9h>
C9
CA
CB
CC
CD
RESERVED
RCAP2L[7:0]
RCAP2H[7:0]
TL2[7:0]
RCAP2L
RCAP2H
TL2
00
00
00
00
Standard
Timer
SFRs, pag
e 69
TH2
TH2[7:0]
Table
48., page
93
CE IRDACON
–
IRDA_EN BIT_PULS CDIV4
CDIV3
CDIV2 CDIV1 CDIV0 0F
Program
Status
Word
(PSW), pa
ge 22
CY
<D7h>
AC
<D6h>
F0
<D5h>
RS[1:0]
<D4h, D3h>
OV
<D2h>
P
<D0>
(1)
PSW
–
–
00
D0
D1
RESERVED
Table
61., page
118
D2
D3
SPICLKD
SPISTAT
SPICLKD[5:0]
–
04
02
Table
62., page
119
–
–
–
BUSY
TEISF RORISF TISF
RISF
27/231
uPSD33xx
SFR
Addr
(hex)
Bit Name and <Bit Address>
Reset
Value Descr.
(hex) with Link
Reg.
SFR
Name
7
6
5
4
3
2
1
0
D4
D5
SPITDR
SPIRDR
SPITDR[7:0]
SPIRDR[7:0]
00
00
Table
62., page
119
Table
59., page
117
D6 SPICON0
–
–
TE
–
RE
–
SPIEN
–
SSEL
TEIE
FLSB
SPO
–
00
00
00
00
Table
60., page
118
D7 SPICON1
RORIE
TIE
TI
RIE
RI
Table
46., page
83
SM0
<DF
SM1
<DE>
SM2
<DD>
REN
<DC>
TB8
<DB>
RB8
<DA>
(1)
SCON1
SBUF1
D8
<D9> <D8>
Figure
25., page
79
D9
DA
DB S1SETUP SS_EN
SBUF1[7:0]
RESERVED
SMPL_SET[6:0]
Table
55., page
105
00
00
00
00
00
Table
50., page
100
DC
DD
DE
DF
S1CON
S1STA
S1DAT
S1ADR
CR2
GC
EN1
STA
STO
ADDR
AA
CR1
CR0
Table
52., page
103
STOP
INTR
TX_MD B_BUSY B_LOST ACK_R SLV
Table
53., page
104
S1DAT[7:0]
S1ADR[7:0]
A[7:0]
Table
54., page
104
Accumulat
or
(ACC), pa
ge 21
(1)
A
00
E0
<bit addresses: E7h, E6h, E5h, E4h, E3h, E2h, E1h, E0h>
E1
to
RESERVED
EF
B Register
(B), page
21
B[7:0]
(1)
B
00
F0
<bit addresses: F7h, F6h, F5h, F4h, F3h, F2h, F1h, F0h>
F1
F2
F3
F4
F5
F6
RESERVED
RESERVED
RESERVED
RESERVED
RESERVED
RESERVED
28/231
uPSD33xx
SFR
Addr
(hex)
Bit Name and <Bit Address>
Reset
Reg.
Value Descr.
(hex) with Link
SFR
Name
7
6
5
4
3
2
1
0
F7
F8
RESERVED
RESERVED
Table
F9
FA
FB
CCON0
–
–
–
DBGCE CPU_AR
RESERVED
CPUPS[2:0]
10
21., page
47
Table
68., page
125
CCON2
CCON3
–
–
–
–
–
–
PCA0CE
PCA0PS[3:0]
PCA1PS[3:0]
10
10
Table
69., page
125
FC
PCA1CE
FD
FE
FF
RESERVED
RESERVED
RESERVED
Note: 1. This SFR can be addressed by individual bits (Bit Address mode) or addressed by the entire byte (Direct Address mode).
29/231
uPSD33xx
8032 ADDRESSING MODES
The 8032 MCU uses 11 different addressing
modes listed below:
Immediate Addressing
This mode uses 8-bits of data (a constant) con-
tained in the second byte of the instruction, and
stores it into the memory location or register indi-
cated by the first byte of the instruction. Thus, the
data is immediately available within the instruction.
This mode is commonly used to initialize registers
and SFRs or to perform mask operations.
There is also a 16-bit version of this mode for load-
ing the DPTR Register. In this case, the two bytes
following the instruction byte contain the 16-bit val-
ue. For example:
■
■
■
■
■
■
■
■
■
■
■
Register
Direct
Register Indirect
Immediate
External Direct
External Indirect
Indexed
Relative
Absolute
Long
MOV A, 40#
; Move the constant, 40h, into
; the accumulator
Bit
Register Addressing
MOV DPTR, 1234# ; Move the constant, 1234h, into
; DPTR
This mode uses the contents of one of the regis-
ters R0 - R7 (selected by the last three bits in the
instruction opcode) as the operand source or des-
tination. This mode is very efficient since an addi-
tional instruction byte is not needed to identify the
operand. For example:
External Direct Addressing
This mode will access external memory (XDATA)
by using the 16-bit address stored in the DPTR
Register. There are only two instructions using this
mode and both use the accumulator to either re-
ceive a byte from external memory addressed by
DPTR or to send a byte from the accumulator to
the address in DPTR. The uPSD33xx has a spe-
cial feature to alternate the contents (source and
destination) of DPTR rapidly to implement very ef-
ficient memory-to-memory transfers. For example:
MOV A, R7
; Move contents of R7 to accumulator
Direct Addressing
This mode uses an 8-bit address, which is con-
tained in the second byte of the instruction, to di-
rectly address an operand which resides in either
8032 DATA SRAM (internal address range 00h-
07Fh) or resides in 8032 SFR (internal address
range 80h-FFh). This mode is quite fast since the
range limit is 256 bytes of internal 8032 SRAM.
For example:
MOVX A, @DPTR ; Move contents of accumulator to
; XDATA at address contained in
; DPTR
MOVX @DPTR, A ; Move XDATA to accumulator
Note:
POINTERS, page 37.
See
details
in
DUAL
DATA
MOV A, 40h
; Move contents of DATA SRAM
; at location 40h into the accumulator
External Indirect Addressing
Register Indirect Addressing
This mode will access external memory (XDATA)
by using the 8-bit address stored in either Register
R0 or R1. This is the fastest way to access XDATA
(least bus cycles), but because only 8-bits are
available for address, this mode limits XDATA to a
size of only 256 bytes (the traditional Port 2 of the
8032 MCU is not available in the uPSD33xx, so it
is not possible to write the upper address byte).
This mode uses an 8-bit address contained in ei-
ther Register R0 or R1 to indirectly address an op-
erand which resides in 8032 IDATA SRAM
(internal address range 80h-FFh). Although 8032
SFR registers also occupy the same physical ad-
dress range as IDATA, SFRs will not be accessed
by Register Indirect mode. SFRs may only be ac-
cesses using Direct address mode. For example:
This mode is not supported by uPSD33xx.
For example:
MOV A, @R0
; Move into the accumulator the
; contents of IDATA SRAM that is
; pointed to by the address
; contained in R0.
MOVX @R0,A
; Move into the accumulator the
; XDATA that is pointed to by
; the address contained in R0.
30/231
uPSD33xx
Indexed Addressing
Absolute Addressing
This mode is used for the MOVC instruction which
allows the 8032 to read a constant from program
memory (not data memory). MOVC is often used
to read look-up tables that are embedded in pro-
gram memory. The final address produced by this
mode is the result of adding either the 16-bit PC or
DPTR value to the contents of the accumulator.
The value in the accumulator is referred to as an
index. The data fetched from the final location in
program memory is stored into the accumulator,
overwriting the index value that was previously
stored there. For example:
This mode will append the 5 high-order bits of the
address of the next instruction to the 11 low-order
bits of an ACALL or AJUMP instruction to produce
a 16-bit jump address. The jump will be within the
same 2K byte page of program memory as the first
byte of the following instruction. For example:
AJMP 0500h
; If next instruction is located at
; address 4000h, the resulting jump
; will be made to 4500h.
Long Addressing
This mode will use the 16-bits contained in the two
bytes following the instruction byte as a jump des-
tination address for LCALL and LJMP instructions.
For example:
MOVC A, @A+DPTR; Move code byte relative to
; DPTR into accumulator
MOVC A, @A+PC ; Move code byte relative to PC
; into accumulator
LJMP 0500h
; Unconditionally jump to address
; 0500h in program memory
Relative Addressing
This mode will add the two’s-compliment number
stored in the second byte of the instruction to the
program counter for short jumps within +128 or –
127 addresses relative to the program counter.
This is commonly used for looping and is very effi-
cient since no additional bus cycle is needed to
fetch the jump destination address. For example:
Bit Addressing
This mode allows setting or clearing an individual
bit without disturbing the other bits within an 8-bit
value of internal SRAM. Bit Addressing is only
available for certain locations in 8032 DATA and
SFR memory. Valid locations are DATA address-
es 20h - 2Fh and for SFR addresses whose base
address ends with 0h or 8h. (Example: The SFR,
IE, has a base address of A8h, so each of the eight
bits in IE can be addressed individually at address
A8h, A9h, ...up to AFh.) For example:
SJMP 34h
; Jump 34h bytes ahead (in program
; memory) of the address at which
; the SJMP instruction is stored. If
; SJMP is at 1000h, program
; execution jumps to 1034h.
SETB AFh
; Set the individual EA bit (Enable All
; Interrupts) inside the SFR Register,
; IE.
31/231
uPSD33xx
uPSD33xx INSTRUCTION SET SUMMARY
Tables 6 through 11 list all of the instructions sup-
ported by the uPSD33xx, including the number of
bytes and number of machine cycles required to
implement each instruction. This is the standard
8051 instruction set.
1. a stall is imposed while loading the 8032 Pre-
Fetch Queue (PFQ); or
2. the occurrence of a cache miss in the Branch
Cache (BC) during a branch in program
execution flow.
The meaning of “machine cycles” is how many
8032 MCU core machine cycles are required to
execute the instruction. The “native” duration of all
machine cycles is set by the memory wait state
settings in the SFR, BUSCON, and the MCU clock
divider selections in the SFR, CCON0 (i.e. a ma-
chine cycle is typically set to 4 MCU clocks for a 5V
uPSD33xx). However, an individual machine cycle
may grow in duration when either of two things
happen:
See 8032 MCU CORE PERFORMANCE
ENHANCEMENTS, page 17 or more details.
But generally speaking, during typical program ex-
ecution, the PFQ is not empty and the BC has no
misses, producing very good performance without
extending the duration of any machine cycles.
The uPSD33xx Programmers Guide describes
each instruction operation in detail.
Table 6. Arithmetic Instruction Set
(1)
Mnemonic
and Use
Description
Add register to ACC
Length/Cycles
ADD
ADD
ADD
ADD
ADDC
ADDC
ADDC
ADDC
SUBB
SUBB
SUBB
SUBB
INC
A, Rn
A, Direct
A, @Ri
A, #data
A, Rn
A, direct
A, @Ri
A, #data
A, Rn
A, direct
A, @Ri
A, #data
A
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
1 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
1 byte/2 cycle
1 byte/4 cycle
1 byte/4 cycle
1 byte/1 cycle
Add direct byte to ACC
Add indirect SRAM to ACC
Add immediate data to ACC
Add register to ACC with carry
Add direct byte to ACC with carry
Add indirect SRAM to ACC with carry
Add immediate data to ACC with carry
Subtract register from ACC with borrow
Subtract direct byte from ACC with borrow
Subtract indirect SRAM from ACC with borrow
Subtract immediate data from ACC with borrow
Increment A
INC
Rn
Increment register
INC
direct
@Ri
Increment direct byte
INC
Increment indirect SRAM
Decrement ACC
DEC
DEC
DEC
DEC
INC
A
Rn
Decrement register
direct
@Ri
Decrement direct byte
Decrement indirect SRAM
Increment Data Pointer
DPTR
AB
MUL
DIV
Multiply ACC and B
AB
Divide ACC by B
DA
A
Decimal adjust ACC
Note: 1. All mnemonics copyrighted ©Intel Corporation 1980.
32/231
uPSD33xx
Table 7. Logical Instruction Set
(1)
Mnemonic
and Use
Description
AND register to ACC
Length/Cycles
ANL
ANL
ANL
ANL
ANL
ANL
ORL
ORL
ORL
ORL
ORL
ORL
SWAP
XRL
XRL
XRL
XRL
XRL
XRL
CLR
CPL
RL
A, Rn
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
2 byte/1 cycle
3 byte/2 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
2 byte/1 cycle
3 byte/2 cycle
1 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
2 byte/1 cycle
3 byte/2 cycle
1 byte/1 cycle
1 byte/1 cycle
1 byte/1 cycle
1 byte/1 cycle
1 byte/1 cycle
1 byte/1 cycle
A, direct
A, @Ri
A, #data
direct, A
direct, #data
A, Rn
AND direct byte to ACC
AND indirect SRAM to ACC
AND immediate data to ACC
AND ACC to direct byte
AND immediate data to direct byte
OR register to ACC
A, direct
A, @Ri
A, #data
direct, A
direct, #data
A
OR direct byte to ACC
OR indirect SRAM to ACC
OR immediate data to ACC
OR ACC to direct byte
OR immediate data to direct byte
Swap nibbles within the ACC
Exclusive-OR register to ACC
Exclusive-OR direct byte to ACC
Exclusive-OR indirect SRAM to ACC
Exclusive-OR immediate data to ACC
Exclusive-OR ACC to direct byte
Exclusive-OR immediate data to direct byte
Clear ACC
A, Rn
A, direct
A, @Ri
A, #data
direct, A
direct, #data
A
A
Compliment ACC
A
Rotate ACC left
RLC
RR
A
Rotate ACC left through the carry
Rotate ACC right
A
RRC
A
Rotate ACC right through the carry
Note: 1. All mnemonics copyrighted ©Intel Corporation 1980.
33/231
uPSD33xx
Table 8. Data Transfer Instruction Set
(1)
Mnemonic
and Use
Description
Move register to ACC
Length/Cycles
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOV
MOVC
MOVC
MOVX
MOVX
MOVX
MOVX
PUSH
POP
A, Rn
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/2 cycle
2 byte/1 cycle
2 byte/1 cycle
2 byte/2 cycle
3 byte/2 cycle
2 byte/2 cycle
3 byte/2 cycle
1 byte/1 cycle
2 byte/2 cycle
2 byte/1 cycle
3 byte/2 cycle
1 byte/2 cycle
1 byte/2 cycle
1 byte/2 cycle
1 byte/2 cycle
1 byte/2 cycle
1 byte/2 cycle
2 byte/2 cycle
2 byte/2 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
1 byte/1 cycle
A, direct
A, @Ri
Move direct byte to ACC
Move indirect SRAM to ACC
A, #data
Rn, A
Move immediate data to ACC
Move ACC to register
Rn, direct
Rn, #data
direct, A
direct, Rn
direct, direct
direct, @Ri
direct, #data
@Ri, A
Move direct byte to register
Move immediate data to register
Move ACC to direct byte
Move register to direct byte
Move direct byte to direct
Move indirect SRAM to direct byte
Move immediate data to direct byte
Move ACC to indirect SRAM
@Ri, direct
@Ri, #data
DPTR, #data16
A, @A+DPTR
A, @A+PC
A, @Ri
Move direct byte to indirect SRAM
Move immediate data to indirect SRAM
Load Data Pointer with 16-bit constant
Move code byte relative to DPTR to ACC
Move code byte relative to PC to ACC
Move XDATA (8-bit addr) to ACC
Move XDATA (16-bit addr) to ACC
Move ACC to XDATA (8-bit addr)
Move ACC to XDATA (16-bit addr)
Push direct byte onto stack
A, @DPTR
@Ri, A
@DPTR, A
direct
direct
Pop direct byte from stack
XCH
A, Rn
Exchange register with ACC
XCH
A, direct
A, @Ri
Exchange direct byte with ACC
Exchange indirect SRAM with ACC
Exchange low-order digit indirect SRAM with ACC
XCH
XCHD
A, @Ri
Note: 1. All mnemonics copyrighted ©Intel Corporation 1980.
34/231
uPSD33xx
Table 9. Boolean Variable Manipulation Instruction Set
(1)
Mnemonic
and Use
Description
Length/Cycles
CLR
CLR
SETB
SETB
CPL
CPL
ANL
ANL
ORL
ORL
MOV
MOV
JC
C
Clear carry
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
1 byte/1 cycle
2 byte/1 cycle
2 byte/2 cycle
2 byte/2 cycle
2 byte/2 cycle
2 byte/2 cycle
2 byte/1 cycle
2 byte/2 cycle
2 byte/2 cycle
2 byte/2 cycle
3 byte/2 cycle
3 byte/2 cycle
3 byte/2 cycle
bit
Clear direct bit
C
Set carry
bit
Set direct bit
C
Compliment carry
bit
Compliment direct bit
AND direct bit to carry
AND compliment of direct bit to carry
OR direct bit to carry
OR compliment of direct bit to carry
Move direct bit to carry
Move carry to direct bit
Jump if carry is set
C, bit
C, /bit
C, bit
C, /bit
C, bit
bit, C
rel
JNC
JB
rel
Jump if carry is not set
Jump if direct bit is set
Jump if direct bit is not set
Jump if direct bit is set and clear bit
rel
JNB
JBC
rel
bit, rel
Note: 1. All mnemonics copyrighted ©Intel Corporation 1980.
35/231
uPSD33xx
Table 10. Program Branching Instruction Set
(1)
Mnemonic
and Use
Description
Absolute subroutine call
Length/Cycles
ACALL
LCALL
RET
addr11
addr16
2 byte/2 cycle
3 byte/2 cycle
1 byte/2 cycle
1 byte/2 cycle
2 byte/2 cycle
3 byte/2 cycle
2 byte/2 cycle
1 byte/2 cycle
2 byte/2 cycle
2 byte/2 cycle
3 byte/2 cycle
3 byte/2 cycle
3 byte/2 cycle
3 byte/2 cycle
2 byte/2 cycle
3 byte/2 cycle
Long subroutine call
Return from subroutine
RETI
AJMP
LJMP
SJMP
JMP
Return from interrupt
addr11
Absolute jump
addr16
Long jump
rel
Short jump (relative addr)
@A+DPTR
rel
Jump indirect relative to the DPTR
Jump if ACC is zero
JZ
JNZ
rel
Jump if ACC is not zero
CJNE
CJNE
CJNE
CJNE
DJNZ
DJNZ
A, direct, rel
A, #data, rel
Rn, #data, rel
@Ri, #data, rel
Rn, rel
Compare direct byte to ACC, jump if not equal
Compare immediate to ACC, jump if not equal
Compare immediate to register, jump if not equal
Compare immediate to indirect, jump if not equal
Decrement register and jump if not zero
Decrement direct byte and jump if not zero
direct, rel
Note: 1. All mnemonics copyrighted ©Intel Corporation 1980.
Table 11. Miscellaneous Instruction Set
(1)
Mnemonic
and Use
Description
Length/Cycles
NOP
No Operation
1 byte/1 cycle
Note: 1. All mnemonics copyrighted ©Intel Corporation 1980.
Table 12. Notes on Instruction Set and Addressing Modes
Rn
Register R0 - R7 of the currently selected register bank.
direct
@Ri
8-bit address for internal 8032 DATA SRAM (locations 00h - 7Fh) or SFR registers (locations 80h - FFh).
8-bit internal 8032 SRAM (locations 00h - FFh) addressed indirectly through contents of R0 or R1.
8-bit constant included within the instruction.
#data
#data16 16-bit constant included within the instruction.
addr16
addr11
rel
16-bit destination address used by LCALL and LJMP.
11-bit destination address used by ACALL and AJMP.
Signed (two-s compliment) 8-bit offset byte.
Direct addressed bit in internal 8032 DATA SRAM (locations 20h to 2Fh) or in SFR registers (88h, 90h,
98h, A8h, B0, B8h, C0h, C8h, D0h, D8h, E0h, F0h).
bit
36/231
uPSD33xx
DUAL DATA POINTERS
XDATA is accessed by the External Direct ad-
dressing mode, which uses a 16-bit address
stored in the DPTR Register. Traditional 8032 ar-
chitecture has only one DPTR Register. This is a
burden when transferring data between two XDA-
TA locations because it requires heavy use of the
working registers to manipulate the source and
destination pointers.
However, the uPSD33xx has two data pointers,
one for storing a source address and the other for
storing a destination address. These pointers can
be configured to automatically increment or decre-
ment after each data transfer, further reducing the
burden on the 8032 and making this kind of data
movement very efficient.
ister at any given time. After reset, the DPSEL0 Bit
is cleared, enabling DPTR0 to function as the DP-
TR, and firmware may access DPTR0 by reading
or writing the traditional DPTR Register at SFR ad-
dresses 82h and 83h. When the DPSEL0 bit is set,
then the DPTR1 Register functions as DPTR, and
firmware may now access DPTR1 through SFR
registers at 82h and 83h. The pointer which is not
selected by the DPSEL0 bit remains in the back-
ground and is not accessible by the 8032. If the
DPSEL0 bit is never set, then the uPSD33xx will
behave like a traditional 8032 having only one
DPTR Register.
To further speed XDATA to XDATA transfers, the
SFR bit, AT, may be set to automatically toggle the
two data pointers, DPTR0 and DPTR1, each time
the standard DPTR Register is accessed by a
MOVX instruction. This eliminates the need for
firmware to manually manipulate the DPSEL0 bit
between each data transfer.
Data Pointer Control Register, DPTC (85h)
By default, the DPTR Register of the uPSD33xx
will behave no different than in a standard 8032
MCU. The DPSEL0 Bit of SFR register DPTC
shown in Table 13, selects which one of the two
“background” data pointer registers (DPTR0 or
DPTR1) will function as the traditional DPTR Reg-
Detailed description for the SFR register DPTC is
shown in Table 13.
Table 13. DPTC: Data Pointer Control Register (SFR 85h, reset value 00h)
Bit 7
–
Bit 6
AT
Bit 5
–
Bit 4
–
Bit 3
–
Bit 2
–
Bit 1
–
Bit 0
DPSEL0
Details
Bit
Symbol
R/W
Definition
7
–
–
Reserved
0 = Manually Select Data Pointer
1 = Auto Toggle between DPTR0 and DPTR1
6
5-1
0
AT
–
R,W
–
Reserved
0 = DPTR0 Selected for use as DPTR
1 = DPTR1 Selected for use as DPTR
DPSE0
R,W
37/231
uPSD33xx
Data Pointer Mode Register, DPTM (86h)
The two “background” data pointers, DPTR0 and
DPTR1, can be configured to automatically incre-
ment, decrement, or stay the same after a MOVX
instruction accesses the DPTR Register. Only the
currently selected pointer will be affected by the in-
crement or decrement. This feature is controlled
by the DPTM Register defined in Table 14.
Firmware Example. The 8051 assembly code il-
lustrated in Table 15 shows how to transfer a block
of data bytes from one XDATA address region to
another XDATA address region. Auto-address in-
crementing and auto-pointer toggling will be used.
The automatic increment or decrement function is
effective only for the MOVX instruction, and not
MOVC or any other instruction that uses the DTPR
Register.
Table 14. DPTM: Data Pointer Mode Register (SFR 86h, reset value 00h)
Bit 7
–
Bit 6
–
Bit 5
–
Bit 4
–
Bit 3
Bit 2
Bit 1
Bit 0
MD11
MD10
MD01
MD00
Details
Bit
Symbol
R/W
Definition
7-4
–
–
Reserved
DPTR1 Mode Bits
00: DPTR1 No Change
01: Reserved
10: Auto Increment
11: Auto Decrement
3-2
1-0
MD[11:10]
MD[01:00]
R,W
R,W
DPTR0 Mode Bits
00: DPTR0 No Change
01: Reserved
10: Auto Increment
11: Auto Decrement
Table 15. 8051 Assembly Code Example
MOV
MOV
MOV
MOV
MOV
MOV
R7, #COUNT
; initialize size of data block to transfer
DPTR, #SOURCE_ADDR ; load XDATA source address base into DPTR0
85h, #01h
; load DPTC to access DPTR1 pointer
DPTR, #DEST_ADDR
85h, #40h
; load XDATA destination address base into DPTR1
; load DPTC to access DPTR0 pointer and auto toggle
; load DPTM to auto-increment both pointers
86h, #0Ah
(1)
(1)
LOOP:
A, @DPTR
; load XDATA byte from source into ACC.
; after load completes, DPTR0 increments and DPTR
; switches DPTR1
MOVX
@DPTR, A
; store XDATA byte from ACC to destination.
; after store completes, DPTR1 increments and DPTR
; switches to DPTR0
MOVX
(1)
R7, LOOP
86h, #00
85h, #00
; continue until done
DJNZ
MOV
MOV
; disable auto-increment
; disable auto-toggle, now back to single DPTR mode
Note: 1. The code loop where the data transfer takes place is only 3 lines of code.
38/231
uPSD33xx
DEBUG UNIT
The 8032 MCU Module supports run-time debug-
ging through the JTAG interface. This same JTAG
interface is also used for In-System Programming
(ISP) and the physical connections are described
in the PSD Module section, JTAG ISP and JTAG
Debug, page 195.
–
There is no on-chip storage for Program Trace
data, but instead this data is scanned from the
uPSD33xx through the JTAG channel at run-
time to the PC host for proccessing. As such,
full speed program tracing is possible only
when the 8032 MCU is operating below
approximately one MIPS of performance.
Above one MIPS, the program will not run
real-time while tracing. One MIPS
Debugging with a serial interface such as JTAG is
a non-intrusive way to gain access to the internal
state of the 8032 MCU core and various memo-
ries. A traditional external hardware emulator can-
not be completely effective on the uPSD33xx
because of the Pre-Fetch Queue and Branch
Cache. The nature of the PFQ and BC hide the
visibility of actual program flow through traditional
external bus connections, thus requiring on-chip
serial debugging instead.
Debugging is supported by Windows PC based
software tools used for 8051 code development
from 3rd party vendors listed at www.st.com/psm.
Debug capabilities include:
performance is determined by the
combination of choice for MCU clock
frequency, and the bit settings in SFR
registers BUSCON and CCON0.
Breakpoints can optionally halt the MCU, and/
or assert the external Debug Event pin.
Breakpoint definitions may be qualified with
read or write operations, and may also be
qualified with an address of code, SFR, DATA,
IDATA, or XDATA memories.
Three breakpoints will compare an address,
but the fourth breakpoint can compare an
address and also data content. Additionally,
the fouth breakpoint can be logically combined
(AND/OR) with any of the other three
breakpoints.
The Debug Event pin can be configured by the
PC host to generate an output pulse for
external triggering when a break condition is
met. The pin can also be configured as an
event input to the breakpoint logic, causing a
break on the falling-edge of an external event
signal. If not used, the Debug Event pin should
–
–
–
–
■
■
■
■
■
■
■
Halt or Start MCU execution
Reset the MCU
Single Step
3 Match Breakpoints
1 Range Breakpoint (inside or outside range)
Program Tracing
Read or Modify MCU core registers, DATA,
IDATA, SFR, XDATA, and Code
■
External Debug Event Pin, Input or Output
Some key points regarding use of the JTAG De-
bugger.
be pulled up to V as described in the
CC
section, Debugging the 8032 MCU
Module., page 201.
The duration of a pulse, generated when the
Event pin configured as an output, is one MCU
clock cycle. This is an active-low signal, so the
first edge when an event occurs is high-to-low.
–
The JTAG Debugger can access MCU
registers, data memory, and code memory
while the MCU is executing at full speed by
cycle-stealing. This means “watch windows”
may be displayed and periodically updated on
the PC during full speed operation. Registers
and data content may also be modified during
full speed operation.
–
–
–
The clock to the Watchdog Timer, ADC, and
2
I C interface are not stopped by a breakpoint
halt.
The Watchdog Timer should be disabled while
debugging with JTAG, else a reset will be
generated upon a watchdog time-out.
39/231
uPSD33xx
INTERRUPT SYSTEM
The uPSD33xx has an 11-source, two priority level
interrupt structure summarized in Table 16.
The specific vector address for each of the inter-
rupt sources are listed in Table 16., page 41. How-
ever, this LCALL jump may be blocked by any of
the following conditions:
Firmware may assign each interrupt source either
high or low priority by writing to bits in the SFRs
named, IP and IPA, shown in Table 16. An inter-
rupt will be serviced as long as an interrupt of
equal or higher priority is not already being ser-
viced. If an interrupt of equal or higher priority is
being serviced, the new interrupt will wait until it is
finished before being serviced. If a lower priority
interrupt is being serviced, it will be stopped and
the new interrupt is serviced. When the new inter-
rupt is finished, the lower priority interrupt that was
stopped will be completed. If new interrupt re-
quests are of the same priority level and are re-
ceived simultaneously, an internal polling
sequence determines which request is selected
for service. Thus, within each of the two priority
levels, there is a second priority structure deter-
mined by the polling sequence.
–
–
–
–
An interrupt of equal or higher priority is
already in progress
The current machine cycle is not the final cycle
in the execution of the instruction in progress
The current instruction involves a write to any
of the SFRs: IE, IEA, IP, or IPA
The current instruction is an RETI
Note: Interrupt flags are polled based on a sample
taken in the previous MCU machine cycle. If an in-
terrupt flag is active in one cycle but is denied ser-
viced due to the conditions above, and then later it
is not active when the conditions above are finally
satisfied, the previously denied interrupt will not be
serviced. This means that active interrupts are not
remembered. Every poling cycle is new.
Firmware may individually enable or disable inter-
rupt sources by writing to bits in the SFRs named,
IE and IEA, shown in Table 16., page 41. The SFR
named IE contains a global disable bit (EA), which
can be cleared to disable all 11 interrupts at once,
as shown in Table 17., page 43. Figure
13., page 42 illustrates the interrupt priority, poll-
ing, and enabling process.
Each interrupt source has at least one interrupt
flag that indicates whether or not an interrupt is
pending. These flags reside in bits of various
SFRs shown in Table 16., page 41.
All of the interrupt flags are latched into the inter-
rupt control system at the beginning of each MCU
machine cycle, and they are polled at the begin-
ning of the following machine cycle. If polling de-
termines one of the flags was set, the interrupt
control system automatically generates an LCALL
to the user’s Interrupt Service Routine (ISR) firm-
ware stored in program memory at the appropriate
vector address.
Assuming all of the listed conditions are satisfied,
the MCU executes the hardware generated
LCALL to the appropriate ISR. This LCALL pushes
the contents of the PC onto the stack (but it does
not save the PSW) and loads the PC with the ap-
propriate interrupt vector address. Program exe-
cution then jumps to the ISR at the vector address.
Execution precedes in the ISR. It may be neces-
sary for the ISR firmware to clear the pending in-
terrupt flag for some interrupt sources, because
not all interrupt flags are automatically cleared by
hardware when the ISR is called, as shown in Ta-
ble 16., page 41. If an interrupt flag is not cleared
after servicing the interrupt, an unwanted interrupt
will occur upon exiting the ISR.
After the interrupt is serviced, the last instruction
executed by the ISR is RETI. The RETI informs
the MCU that the ISR is no longer in progress and
the MCU pops the top two bytes from the stack
and loads them into the PC. Execution of the inter-
rupted program continues where it left off.
Note: An ISR must end with a RETI instruction,
not a RET. An RET will not inform the interrupt
control system that the ISR is complete, leaving
the MCU to think the ISR is still in progress, mak-
ing future interrupts impossible.
40/231
uPSD33xx
Table 16. Interrupt Summary
Flag Bit Name
Enable Bit Name
Priority Bit Name
(SFR.bit position) Flag Bit Auto- (SFR.bit position) (SFR.bit position)
Interrupt
Source
Polling Vector
Priority Addr
Cleared
1 = Intr Pending
0 = No Interrupt
by Hardware?
1 = Intr Enabled
0 = Intr Disabled
1= High Priority
0 = Low Priority
Reserved
0 (high) 0063h
–
–
–
–
External
Interrupt INT0
Edge - Yes
Level - No
1
2
3
4
5
0003h
000Bh
0013h
001Bh
0023h
IE0 (TCON.1)
EX0 (IE.0)
PX0 (IP.0)
Timer 0
Overflow
TF0 (TCON.5)
IE1 (TCON.3
TF1 (TCON.7)
Yes
ET0 (IE.1)
EX1 (IE.2)
ET1 (IE.3)
ES0 (IE.4)
PT0 (IP.1)
PX1 (IP.2)
PT1 (IP.3)
PS0 (IP.4)
External
Interrupt INT1
Edge - Yes
Level - No
Timer 1
Overflow
Yes
No
RI (SCON0.0)
TI (SCON0.1)
UART0
Timer 2
Overflow
or TX2 Pin
TF2 (T2CON.7)
EXF2 (T2CON.6)
6
7
002Bh
0053h
No
ET2 (IE.5)
PT2 (IP.5)
TEISF, RORISF,
TISF, RISF
(SPISTAT[3:0])
SPI
Yes
ESPI (IEA.6)
–
PSPI (IPA.6)
–
Reserved
8
9
0033h
0043h
003Bh
–
–
2
2
2
INTR (S1STA.5)
AINTF (ACON.7)
Yes
No
I C
EI C (IEA.1)
PI C (IPA.1)
ADC
PCA
10
EADC (IEA.7)
EPCA (IEA.5)
PADC (IPA.7)
PPCA (IPA.5)
OFVx, INTFx
(PCASTA[0:7])
11
005Bh
No
No
RI (SCON1.0)
TI (SCON1.1)
UART1
12 (low) 004Bh
ES1 (IEA.4)
PS1 (IPA.4)
41/231
uPSD33xx
Figure 13. Enabling and Polling Interrupts
Priority
High
Interrupt
Sources
IE/IEA
IP/IPA
Reserved
Low
Ext
INT0
Timer 0
Ext
INT1
Timer 1
UART0
Timer 2
SPI
USB
I2C
ADC
PCA
UART1
Global
Enable
AI07844
42/231
uPSD33xx
Individual Interrupt Sources
External Interrupts Int0 and Int1. External in-
terrupt inputs on pins EXTINT0 and EXTINT1
(pins 3.2 and 3.3) are either edge-triggered or lev-
el-triggered, depending on bits IT0 and IT1 in the
SFR named TCON.
When an external interrupt is generated from an
edge-triggered (falling-edge) source, the appropri-
ate flag bit (IE0 or IE1) is automatically cleared by
hardware upon entering the ISR.
When an external interrupt is generated from a
level-triggered (low-level) source, the appropriate
flag bit (IE0 or IE1) is NOT automatically cleared
by hardware.
Timer 0 and 1 Overflow Interrupt. Timer 0 and
Timer 1 interrupts are generated by the flag bits
TF0 and TF1 when there is an overflow condition
in the respective Timer/Counter register (except
for Timer 0 in Mode 3).
The ISR must read flag bits in the SFR named
SCON0 for UART0, or SCON1 for UART1 to de-
termine the cause of the interrupt.
SPI Interrupt. The SPI interrupt has four interrupt
sources, which are logically ORed together when
interrupting the MCU. The ISR must read the flag
bits to determine the cause of the interrupt.
A flag bit is set for: end of data transmit (TEISF);
data receive overrun (RORISF); transmit buffer
empty (TISF); or receive buffer full (RISF).
2
I C Interrupt. The flag bit INTR is set by a variety
2
of conditions occurring on the I C interface: re-
ceived own slave address (ADDR flag); received
general call address (GC flag); received STOP
condition (STOP flag); or successful transmission
or reception of a data byte.The ISR must read the
flag bits to determine the cause of the interrupt.
ADC Interrupt. The flag bit AINTF is set when an
Timer 2 Overflow Interrupt. This interrupt is
generated to the MCU by a logical OR of flag bits,
TF2 and EXE2. The ISR must read the flag bits to
determine the cause of the interrupt.
A-to-D conversion has completed.
PCA Interrupt. The PCA has eight interrupt
sources, which are logically ORed together when
interrupting the MCU.The ISR must read the flag
bits to determine the cause of the interrupt.
–
–
TF2 is set by an overflow of Timer 2.
EXE2 is generated by the falling edge of a
signal on the external pin, T2X (pin P1.1).
–
Each of the six TCMs can generate a "match
or capture" interrupt on flag bits OFV5..0
respectively.
UART0 and UART1 Interrupt. Each
of
the
UARTs have identical interrupt structure. For each
UART, a single interrupt is generated to the MCU
by the logical OR of the flag bits, RI (byte received)
and TI (byte transmitted).
–
Each of the two 16-bit counters can generate
an overflow interrupt on flag bits INTF1 and
INTF0 respectively.
Tables 17 through Table 20., page 45 have de-
tailed bit definitions of the interrupt system SFRs.
Table 17. IE: Interrupt Enable Register (SFR A8h, reset value 00h)
Bit 7
EA
Bit 6
–
Bit 5
ET2
Bit 4
ES0
Bit 3
ET1
Bit 2
EX1
Bit 1
ET0
Bit 0
EX0
Details
Bit
Symbol
R/W
Function
Global disable bit. 0 = All interrupts are disabled. 1 = Each interrupt
source can be individually enabled or disabled by setting or clearing its
enable bit.
7
EA
R,W
Do not modify this bit. It is used by the JTAG debugger for instruction
tracing. Always read the bit and write back the same bit value when
writing this SFR.
6
–
R,W
(1)
ET2
ES0
ET1
EX1
ET0
EX0
R,W
R,W
R,W
R,W
R,W
R,W
Enable Timer 2 Interrupt
Enable UART0 Interrupt
5
(1)
4
(1)
Enable Timer 1 Interrupt
Enable External Interrupt INT1
Enable Timer 0 Interrupt
Enable External Interrupt INT0
3
(1)
2
(1)
1
(1)
0
Note: 1. 1 = Enable Interrupt, 0 = Disable Interrupt
43/231
uPSD33xx
Table 18. IEA: Interrupt Enable Addition Register (SFR A7h, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
–
2
EADC
ESPI
EPCA
ES1
–
–
EI C
Details
Bit
Symbol
EADC
ESPI
R/W
R,W
R,W
R,W
R,W
Function
(1)
Enable ADC Interrupt
Enable SPI Interrupt
7
(1)
6
(1)
EPCA
ES1
Enable Programmable Counter Array Interrupt
Enable UART1 Interrupt
5
(1)
4
3
2
–
–
–
–
Reserved, do not set to logic '1.'
Reserved, do not set to logic '1.'
(1)
2
2
R,W
–
1
EI C
Enable I C Interrupt
0
–
Reserved, do not set to logic '1.'
Note: 1. 1 = Enable Interrupt, 0 = Disable Interrupt
Table 19. IP: Interrupt Priority Register (SFR B8h, reset value 00h)
Bit 7
–
Bit 6
–
Bit 5
PT2
Bit 4
PS0
Bit 3
PT1
Bit 2
PX1
Bit 1
PT0
Bit 0
PX0
Details
Bit
7
Symbol
R/W
–
Function
–
–
Reserved
Reserved
6
–
(1)
PT2
R,W
Timer 2 Interrupt priority level
UART0 Interrupt priority level
Timer 1 Interrupt priority level
5
(1)
PS0
PT1
PX1
PT0
PX0
R,W
R,W
R,W
R,W
R,W
4
(1)
3
(1)
External Interrupt INT1 priority level
Timer 0 Interrupt priority level
2
(1)
1
(1)
External Interrupt INT0 priority level
0
Note: 1. 1 = Assigns high priority level, 0 = Assigns low priority level
44/231
uPSD33xx
Table 20. IPA: Interrupt Priority Addition register (SFR B7h, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
–
2
PADC
PSPI
PPCA
PS1
–
–
PI C
Details
Bit
Symbol
PADC
PSPI
R/W
R,W
R,W
R,W
R,W
Function
(1)
ADC Interrupt priority level
SPI Interrupt priority level
PCA Interrupt level
7
(1)
6
(1)
PPCA
PS1
5
(1)
UART1 Interrupt priority level
4
3
2
–
–
–
–
Reserved
Reserved
(1)
2
2
R,W
–
1
PI C
I C Interrupt priority level
0
–
Reserved
Note: 1. 1 = Assigns high priority level, 0 = Assigns low priority level
45/231
uPSD33xx
MCU CLOCK GENERATION
Internal system clocks generated by the clock gen-
eration unit are derived from the signal, XTAL1,
dently divide PERIPH_CLK to scale it appropriate-
ly for use.
shown in Figure 14. XTAL1 has a frequency f
,
OSC
PERIPH_CLK runs at all times except when
blocked by the PD bit in the SFR named PCON
during MCU Power-down Mode.
JTAG Interface Clock. The JTAG interface for
ISP and for Debugging uses the externally sup-
plied JTAG clock, coming in on pin TCK. This
means the JTAG ISP interface is always available,
and the JTAG Debug interface is available when
enabled, even during MCU Idle mode and Power-
down Mode.
However, since the MCU participates in the JTAG
debug process, and MCU_CLK is halted during
Idle and Power-down Modes, the majority of de-
bug functions are not available during these low
power modes. But the JTAG debug interface is ca-
pable of executing a reset command while in these
low power modes, which will exit back to normal
operating mode where all debug commands are
available again.
The CCON0 SFR contains a bit, DBGCE, which
enables the breakpoint comparators inside the
JTAG Debug Unit when set. DBGCE is set by de-
fault after reset, and firmware may clear this bit at
run-time. Disabling these comparators will reduce
current consumption on the MCU Module, and it’s
recommended to do so if the Debug Unit will not
be used (such as in the production version of an
end-product).
which comes directly from the external crystal or
oscillator device. The SFR named CCON0 (Table
21., page 47) controls the clock generation unit.
There are two clock signals produced by the clock
generation unit:
■
MCU_CLK
■
PERIPH_CLK
MCU_CLK
This clock drives the 8032 MCU core and the
Watchdog Timer (WDT). The frequency of
MCU_CLK is equal to f
by default, but it can be
OSC
divided by as much as 2048, shown in Figure 14.
The bits CPUPS[2:0] select one of eight different
divisors, ranging from 2 to 2048. The new frequen-
cy is available immediately after the CPUPS[2:0]
bits are written. The final frequency of MCU_CLK
is f
.
MCU
MCU_CLK is blocked by either bit, PD or IDL, in
the SFR named PCON during MCU Power-down
Mode or Idle Mode respectively.
MCU_CLK clock can be further divided as re-
quired for use in the WDT. See details of the WDT
in SUPERVISORY FUNCTIONS, page 65.
PERIPH_CLK
This clock drives all the uPSD33xx peripherals ex-
cept the WDT. The Frequency of PERIPH_CLK is
always f
. Each of the peripherals can indepen-
OSC
Figure 14. Clock Generation Logic
PCON[2:0]: CPUPS[2:0],
Clock Pre-Scaler Select
PCON[0]: IDL,
Idle Mode
PCON[1]: PD,
Power-Down Mode
3
XTAL1 (default)
0
XTAL1
(f
)
OSC
XTAL1 /2
Q
1
2
3
4
5
6
7
MCU_CLK (f
)
MCU
(to: 8032, WDT)
XTAL1 /4
Q
Q
Q
Q
Q
Q
M
U
X
XTAL1 /8
XTAL1 /16
XTAL1 /32
XTAL1 /1024
XTAL1 /2048
Clock Divider
PERIPH_CLK (f
)
OSC
(to: TIMER0/1/2, UART0/1, PCA0/1, SPI, I2C, ADC)
AI09197
46/231
uPSD33xx
Table 21. CCON0: Clock Control Register (SFR F9h, reset value 10h)
Bit 7
–
Bit 6
–
Bit 5
–
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
DBGCE
CPUAR
CPUPS[2:0]
Details
Bit
7
Symbol
R/W
Definition
–
–
–
–
–
–
Reserved
Reserved
Reserved
6
5
Debug Unit Breakpoint Comparator Enable
4
3
DBGCE
CPUAR
R,W
R,W
0 = JTAG Debug Unit comparators are disabled
1 = JTAG Debug Unit comparators are enabled (Default condition after
reset)
Automatic MCU Clock Recovery
0 = There is no change of CPUPS[2:0] when an interrupt occurs.
1 = Contents of CPUPS[2:0] automatically become 000b whenever any
interrupt occurs.
MCUCLK Pre-Scaler
000b: f
001b: f
010b: f
011b: f
100b: f
101b: f
110b: f
111b: f
= f
= f
= f
= f
= f
= f
= f
= f
(Default after reset)
/2
/4
/8
/16
/32
/1024
/2048
MCU
MCU
MCU
MCU
MCU
MCU
MCU
MCU
OSC
OSC
OSC
OSC
OSC
OSC
OSC
OSC
2:0
CPUPS
R,W
47/231
uPSD33xx
POWER SAVING MODES
The uPSD33xx is a combination of two die, or
modules, each module having it’s own current
consumption characteristics. This section de-
scribes reduced power modes for the MCU Mod-
Interrupt instruction (RETI), the next
instruction to be executed will be the one
which follows the instruction that set the IDL
bit in the PCON SFR.
ule.
See
the
section,
Power
–
After a reset from the supervisor, the IDL bit is
cleared, Idle Mode is terminated, and the MCU
restarts after three MCU machine cycles.
Management, page 137 for reduced power modes
of the PSD Module. Total current consumption for
the combined modules is determined in the DC
specifications at the end of this document.
Power-down Mode
Power-down Mode will halt the 8032 core and all
MCU peripherals (Power-down Mode blocks
MCU_CLK and PERIPH_CLK). This is the lowest
power state for the MCU Module. When the PSD
Module is also placed in Power-down mode, the
lowest total current consumption for the combined
die is achieved for the uPSD33xx. See Power
Management, page 137 in the PSD Module sec-
tion for details on how to also place the PSD Mod-
ule in Power-down mode. The sequence of 8032
instructions is important when placing both mod-
ules into Power-down Mode.
The instruction that sets the PD Bit in the SFR
named PCON (Table 24., page 50) is the last in-
struction executed prior to the MCU Module going
into Power-down Mode. Once in Power-down
Mode, the on-chip oscillator circuitry and all clocks
are stopped. The SFRs, DATA, IDATA,
and XDATA are preserved.
The MCU Module has three software-selectable
modes of reduced power operation.
■
■
■
Idle Mode
Power-down Mode
Reduced Frequency Mode
Idle Mode
Idle Mode will halt the 8032 MCU core while leav-
ing the MCU peripherals active (Idle Mode blocks
MCU_CLK only). For lowest current consumption
in this mode, it is recommended to disable all un-
used peripherals, before entering Idle mode (such
as the ADC and the Debug Unit breakpoint com-
parators). The following functions remain fully ac-
tive during Idle Mode (except if disabled by SFR
settings).
■
■
■
External Interrupts INT0 and INT1
Timer 0, Timer 1 and Timer 2
Power-down Mode is terminated only by a reset
from the supervisor, originating from the
RESET_IN_ pin, the Low-Voltage Detect circuit
(LVD), or a JTAG Debug reset command. Since
the clock to the WTD is not active during Power-
down mode, it is not possible for the supervisor to
generate a WDT reset.
Table 22., page 49 summarizes the status of I/O
pins and peripherals during Idle and Power-down
Modes on the MCU Module. Table 23., page 49
shows the state of 8032 MCU address, data, and
control signals during these modes.
Supervisor reset from: LVD, JTAG Debug,
External RESET_IN_, but not the WTD
■
■
■
■
■
ADC
I C Interface
UART0 and UART1 Interfaces
SPI Interface
Programmable Counter Array
2
An interrupt generated by any of these peripher-
als, or a reset generated from the supervisor, will
cause Idle Mode to exit and the 8032 MCU will re-
sume normal operation.
The output state on I/O pins of MCU ports 1, 3, and
4 remain unchanged during Idle Mode.
Reduced Frequency Mode
The 8032 MCU consumes less current when oper-
ating at a lower clock frequency. The MCU can re-
duce it’s own clock frequency at run-time by
writing to three bits, CPUPS[2:0], in the SFR
named CCON0 described in Table 21., page 47.
These bits effectively divide the clock frequency
To enter Idle Mode, the 8032 MCU executes an in-
struction to set the IDL bit in the SFR named
PCON, shown in Table 24., page 50. This is the
last instruction executed in normal operating mode
before Idle Mode is activated. Once in Idle Mode,
the MCU status is entirely preserved, and there
are no changes to: SP, PSW, PC, ACC, SFRs,
DATA, IDATA, or XDATA.
(f
OSC
) coming in from the external crystal or oscil-
lator device. The clock division range is from 1/2 to
1/2048, and the resulting frequency is f
.
MCU
This MCU clock division does not affect any of the
peripherals, except for the WTD. The clock driving
the WTD is the same clock driving the 8032 MCU
core as shown in Figure 14., page 46.
The following are factors related to Idle Mode exit:
–
Activation of any enabled interrupt will cause
the IDL bit to be cleared by hardware,
terminating Idle Mode. The interrupt is
serviced, and following the Return from
48/231
uPSD33xx
MCU firmware may reduce the MCU clock fre-
quency at run-time to consume less current when
performing tasks that are not time critical, and then
restore full clock frequency as required to perform
urgent tasks.
til an event occurs that requires full performance.
See Table 21., page 47 for details on CPUAR.
See the DC Specifications at the end of this docu-
ment to estimate current consumption based on
the MCU clock frequency.
Returning to full clock frequency is done automat-
ically upon an MCU interrupt, if the CPUAR Bit in
the SFR named CCON0 is set (the interrupt will
force CPUPS[2:0] = 000). This is an excellent way
to conserve power using a low frequency clock un-
Note: Some of the bits in the PCON SFR shown in
Table 24., page 50 are not related to power con-
trol.
Table 22. MCU Module Port and Peripheral Status during Reduced Power Modes
SUPER- UART0,
TIMER
0,1,2
EXT
INT0, 1
2
Mode
Ports 1, 3, 4
PCA
SPI
ADC
I C
VISOR
UART1
(1)
Idle
Maintain Data
Active
Active
Active
Active
Active
Active
Active
Active
Power-down Maintain Data Disabled Disabled Disabled Disabled Disabled Disabled Disabled Disabled
Note: 1. The Watchdog Timer is not active during Idle Mode. Other supervisor functions are active: LVD, external reset, JTAG Debug reset
Table 23. State of 8032 MCU Bus Signals during Power-down and Idle Modes
Mode
Idle
ALE
PSEN_
RD_
WR_
AD0-7
FFh
A8-15
FFh
0
0
1
1
1
1
1
1
Power-down
FFh
FFh
49/231
uPSD33xx
Table 24. PCON: Power Control Register (SFR 87h, reset value 00h)
Bit 7
Bit 6
Bit 5
–
Bit 4
POR
Bit 3
Bit 2
Bit 1
PD
Bit 0
IDL
SMOD0
SMOD1
RCLK1
TCLK1
Details
Bit
Symbol
R/W
Function
Baud Rate Double Bit (UART0)
7
SMOD0
R,W
0 = No Doubling
1 = Doubling
(See UART Baud Rates, page 84 for details.)
Baud Rate Double Bit for 2nd UART (UART1)
6
5
SMOD1
–
R,W
–
0 = No Doubling
1 = Doubling
(See UART Baud Rates, page 84 for details.)
Reserved
Only a power-on reset sets this bit (cold reset). Warm reset will not set
this bit.
4
POR
R,W
'0,' Cleared to zero with firmware
'1,' Is set only by a power-on reset generated by Supervisory circuit (see
Power-up Reset, page 66 for details).
Received Clock Flag (UART1)
(See Table 41., page 75 for flag description.)
3
2
RCLK1
TCLK1
R,W
R,W
Transmit Clock Flag (UART1)
(See Table 41., page 75 for flag description)
Activate Power-down Mode
1
0
PD
R,W
R,W
0 = Not in Power-down Mode
1 = Enter Power-down Mode
Activate Idle Mode
IDL
0 = Not in Idle Mode
1 = Enter Idle Mode
50/231
uPSD33xx
OSCILLATOR AND EXTERNAL COMPONENTS
The oscillator circuit of uPSD33xx devices is a sin-
gle stage, inverting amplifier in a Pierce oscillator
configuration. The internal circuitry between pins
XTAL1 and XTAL2 is basically an inverter biased
to the transfer point. Either an external quartz crys-
tal or ceramic resonator can be used as the feed-
back element to complete the oscillator circuit.
Both are operated in parallel resonance. Ceramic
resonators are lower cost, but typically have a wid-
er frequency tolerance than quartz crystals. Alter-
natively, an external clock source from an
oscillator or other active device may drive the
uPSD33xx oscillator circuit input directly, instead
of using a crystal or resonator.
The pin XTAL1 is the high gain amplifier input, and
XTAL2 is the output. To drive the uPSD33xx de-
vice externally from an oscillator or other active
device, XTAL1 is driven and XTAL2 is left open-
circuit. This external source should drive a logic
low at the voltage level of 0.3 V
or below, and
CC
logic high at 0.7V V
or above, up to 5.5V V
.
CC
CC
The XTAL1 input is 5V tolerant.
Most of the quartz crystals in the range of 25MHz
to 40MHz operate in the third overtone frequency
mode. An external LC tank circuit at the XTAL2
output of the oscillator circuit is needed to achieve
the third overtone frequency, as shown in Figure
15., page 52. Without this LC circuit, the crystal
will oscillate at a fundamental frequency mode that
is about 1/3 of the desired overtone frequency.
Note: In Figure 15., page 52 crystals which are
specified to operate in fundamental mode (not
overtone mode) do not need the LC circuit compo-
nents. Since quartz crystals and ceramic resona-
tors have their own characteristics based on their
manufacturer, it is wise to also consult the manu-
facturer’s recommended values for external com-
ponents.
The minimum frequency of the quartz crystal, ce-
ramic resonator, or external clock source is 1MHz
2
if the I C interface is not used. The minimum is
2
8MHz if I C is used. The maximum is 40MHz in all
cases. This frequency is f
, which can be divid-
OSC
ed internally as described in MCU CLOCK
GENERATION, page 46.
51/231
uPSD33xx
Figure 15. Oscillator and Clock Connections
Crystal or Resonator
Usage
XTAL1
(in)
XTAL2
(out)
L1
C1
C2
C3
XTAL
(f
)
OSC
XTAL (f
)
C3
L1
C1 = C2
OSC
40 - 50pF
15-33pF
20pF
Ceramic Resonator
None
None
10nF
None
None
2.2µH
Crystal, fundamental mode (3-40MHz)
Crystal, overtone mode (25-40MHz)
XTAL2
(out)
XTAL1
(in)
Direct Drive
No Connect
External Ocsillator or
Active Clock Source
AI09198
52/231
uPSD33xx
I/O PORTS OF MCU MODULE
The MCU Module has three 8-bit I/O ports: Port 1,
Port 3, and Port 4. The PSD Module has four other
I/O ports: Port A, B, C, and D. This section de-
scribes only the I/O ports on the MCU Module.
MCU Port Operating Modes
MCU port pins can operate as GPIO or as alter-
nate functions (see Figure 17., page 56 through
Figure 19., page 57).
I/O ports will function as bi-directional General
Purpose I/O (GPIO), but the port pins can have al-
ternate functions assigned at run-time by writing to
specific SFRs. The default operating mode (during
and after reset) for all three ports is GPIO input
mode. Port pins that have no external connection
will not float because each pin has an internal
Depending on the selected pin function, a particu-
lar pin operating mode will automatically be used:
■
■
■
■
■
■
■
■
GPIO - Quasi-bidirectional mode
UART0, UART1 - Quasi-bidirectional mode
SPI - Quasi-bidirectional mode
I2C - Open drain mode
weak pull-up (~150K ohms) to V
.
CC
ADC - Analog input mode
PCA output - Push-Pull mode
PCA input - Input only (Quasi-bidirectional)
Timer 0,1,2 - Input only (Quasi-bidirectional)
I/O ports 3 and 4 are 5V tolerant, meaning they
can be driven/pulled externally up to 5.5V without
damage. The pins on Port 4 have a higher current
capability than the pins on Ports 1 and 3.
Three additional MCU ports (only on 80-pin
uPSD33xx devices) are dedicated to bring out the
8032 MCU address, data, and control signals to
external pins. One port, named MCUA[11:8], con-
tains four MCU address signal outputs. Another
port, named MCUAD[7:0], has eight multiplexed
address/data bidirectional signals. The third port
has MCU bus control outputs: read, write, program
fetch, and address latch. These ports are typically
used to connect external parallel peripherals and
memory devices, but they may NOT be used as
GPIO. Notice that only four of the eight upper ad-
dress signals come out to pins on the port MC-
UA[11:8]. If additional high-order address signals
are required on external pins (MCU addresses
A[15:12]), then these address signals can be
brought out as needed to PLD output pins or to the
Address Out mode pins on PSD Module ports.
See PSD Module section, “Latched Address Out-
put Mode, page 177 for details.
GPIO Function. Ports in GPIO mode operate as
quasi-bidirectional pins, consistent with standard
8051 architecture. GPIO pins are individually con-
trolled by three SFRs:
■
■
■
SFR, P1 (Table 25., page 57)
SFR, P3 (Table 26., page 58)
SFR, P4 (Table 27., page 58)
These SFRs can be accessed using the Bit Ad-
dressing mode, an efficient way to control individ-
ual port pins.
GPIO Output. Simply stated, when a logic '0' is
written to a bit in any of these port SFRs while in
GPIO mode, the corresponding port pin will enable
a low-side driver, which pulls the pin to ground,
and at the same time releases the high-side driver
and pull-ups, resulting in a logic'0' output. When a
logic '1' is written to the SFR, the low-side driver is
released, the high-side driver is enabled for just
one MCU_CLK period to rapidly make the 0-to1
transition on the pin, while weak active pull-ups
Figure 16., page 55 represents the flexibility of pin
function routing controlled by the SFRs. Each of
the 24 pins on three ports, P1, P3, and P4, may be
individually routed on a pin-by-pin basis to a de-
sired function.
(total ~150K ohms) to V are enabled. This struc-
CC
ture is consistent with standard 8051 architecture.
The high side driver is momentarily enabled only
for 0-to-1 transitions, which is implemented with
the delay function at the latch output as pictured in
Figure 17., page 56 through Figure 19., page 57.
After the high-side driver is disabled, the two weak
pull-ups remain enabled resulting in a logic '1' out-
put at the pin, sourcing I
uA to an external de-
OH
vice. Optionally, an external pull-up resistor can be
added if additional source current is needed while
outputting a logic '1.'
53/231
uPSD33xx
GPIO Input. To use a GPIO port pin as an input,
the low-side driver to ground must be disabled, or
else the true logic level being driven on the pin by
an external device will be masked (always reads
logic '0'). So to make a port pin “input ready”, the
corresponding bit in the SFR must have been set
to a logic '1' prior to reading that SFR bit as an in-
put. A reset condition forces SFRs P1, P3, and P4
to FFh, thus all three ports are input ready after re-
set.
When a pin is used as an input, the stronger pull-
up “A” maintains a solid logic '1' until an external
device drives the input pin low. At this time, pull-up
“A” is automatically disabled, and only pull-up “B”
GPIO Current Capability. A GPIO pin on Port 4
can sink twice as much current than a pin on either
Port 1 or Port 3 when the low-side driver is output-
ting a logic '0' (I ). See the DC specifications at
OL
the end of this document for full details.
Reading Port Pin vs. Reading Port Latch. When
firmware reads the GPIO ports, sometimes the ac-
tual port pin is sampled in hardware, and some-
times the port SFR latch is read and not the actual
pin, depending on the type of MCU instruction
used. These two data paths are shown in Figure
17., page 56 through Figure 19., page 57. SFR
latches are read (and not the pins) only when the
read is part of a read-modify-write instruction and
the write destination is a bit or bits in a port SFR.
These instructions are: ANL, ORL, XRL, JBC,
CPL, INC, DEC, DJNZ, MOV, CLR, and SETB. All
other types of reads to port SFRs will read the ac-
tual pin logic level and not the port latch. This is
consistent with 8051 architecture.
will source the external device I uA, consistent
IH
with standard 8051 architecture.
GPIO Bi-Directional. It is possible to operate indi-
vidual port pins in bi-directional mode. For an out-
put,
firmware
would
simply
write
the
corresponding SFR bit to logic '1' or '0' as needed.
But before using the pin as an input, firmware must
first ensure that a logic '1' was the last value writ-
ten to the corresponding SFR bit prior to reading
that SFR bit as an input.
54/231
uPSD33xx
Figure 16. MCU Module Port Pin Function Routing
Ports
MCU Module
SFR
GPIO (8)
8
P3
UART0 (2)
TIMER0/1 (4)
2
I C (2)
SFR
GPIO (8)
8
SFR
P1
ADC (8)
SFR
TIMER2 (2)
UART1 (2)
SPI (4)
SFR
8
SFR
PCA (8)
P4
GPIO (8)
M
C
U
A
D
Low Addr & Data[7:0]
8
8032 MCU
CORE
On 80-pin
Devices
Only
M
C
U
A
Available on PSD
Module Pins
4
Hi Address [15:12]
Hi Address [11:8]
4
4
C
N
T
L
RD, WR, PSEN, ALE
AI09199
55/231
uPSD33xx
Figure 17. MCU I/O Cell Block Diagram for Port 1
Select_Alternate_Func
V
CC
V
V
CC
CC
DELAY,
1 MCU_CLK
WEAK
PULL-UP, B
STONGER
PULL-UP, A
Digital_Alt_Func_Data_Out
P1.X SFR Read Latch
(for R-M-W instructions)
HIGH
SIDE
P1.X Pin
SEL
IN 1
MCU_Reset
MUX
Y
LOW
SIDE
PRE
8032 Data Bus Bit
D
Q
Q
IN 0
SFR
P1.X
Latch
DELAY,
1 MCU_CLK
GPIO P1.X SFR
Write Latch
P1.X SFR Read Pin
Analog_Alt_Func_En
Digital_Pin_Data_In
Analog_Pin_In
AI09600
Figure 18. MCU I/O Cell Block Diagram for Port 3
Disables High-Side Driver
2
Enable_I C
Select_Alternate_Func
V
CC
V
V
CC
CC
DELAY,
1 MCU_CLK
STONGER
PULL-UP, A
WEAK
PULL-UP, B
Digital_Alt_Func_Data_Out
P3.X SFR Read Latch
(for R-M-W instructions)
HIGH
SIDE
P3.X Pin
SEL
IN 1
MCU_Reset
LOW
SIDE
MUX
Y
PRE
8032 Data Bus Bit
D
Q
Q
IN 0
SFR
P3.X
Latch
DELAY,
1 MCU_CLK
GPIO P3.X SFR
Write Latch
P3.X SFR Read Pin
Digital_Pin_Data_In
AI09601
56/231
uPSD33xx
Figure 19. MCU I/O Cell Block Diagram for Port 4
For PCA Alternate Function
Enable_Push_Pull
Select_Alternate_Func
V
V
V
CC
CC
CC
DELAY,
1 MCU_CLK
WEAK
PULL-UP, B
STONGER
PULL-UP, A
Digital_Alt_Func_Data_Out
P4.X SFR Read Latch
(for R-M-W instructions)
HIGH
SIDE
P4.X Pin
SEL
IN 1
MCU_Reset
LOW
SIDE
MUX
Y
PRE
8032 Data Bus Bit
D
Q
Q
IN 0
SFR
P4.X
Latch
DELAY,
1 MCU_CLK
GPIO P4.X SFR
Write Latch
P4.X SFR Read Pin
Digital_Pin_Data_In
AI09602
Table 25. P1: I/O Port 1 Register (SFR 90h, reset value FFh)
Bit 7
P1.7
Bit 6
P1.6
Bit 5
P1.5
Bit 4
P1.4
Bit 3
P1.3
Bit 2
P1.2
Bit 1
P1.1
Bit 0
P1.0
Details
(1)
Bit
7
Symbol
P1.7
P1.6
P1.5
P1.4
P1.3
P1.2
P1.1
P1.0
R/W
R,W
R,W
R,W
R,W
R,W
R,W
R,W
R,W
Function
Port pin 1.7
Port pin 1.6
Port pin 1.5
Port pin 1.4
Port pin 1.3
Port pin 1.2
Port pin 1.1
Port pin 1.0
6
5
4
3
2
1
0
Note: 1. Write '1' or '0' for pin output. Read for pin input, but prior to READ, this bit must have been set to '1' by firmware or by a reset event.
57/231
uPSD33xx
Table 26. P3: I/O Port 3 Register (SFR B0h, reset value FFh)
Bit 7
P3.7
Bit 6
P3.6
Bit 5
P3.5
Bit 4
P3.4
Bit 3
P3.3
Bit 2
P3.2
Bit 1
P3.1
Bit 0
P3.0
Details
(1)
Bit
7
Symbol
P3.7
P3.6
P3.5
P3.4
P3.3
P3.2
P3.1
P3.0
R/W
R,W
R,W
R,W
R,W
R,W
R,W
R,W
R,W
Function
Port pin 3.7
Port pin 3.6
Port pin 3.5
Port pin 3.4
Port pin 3.3
Port pin 3.2
Port pin 3.1
Port pin 3.0
6
5
4
3
2
1
0
Note: 1. Write '1' or '0' for pin output. Read for pin input, but prior to READ, this bit must have been set to '1' by firmware or by a reset event.
Table 27. P4: I/O Port 4 Register (SFR C0h, reset value FFh)
Bit 7
P4.7
Bit 6
P4.6
Bit 5
P4.5
Bit 4
P4.4
Bit 3
P4.3
Bit 2
P4.2
Bit 1
P4.1
Bit 0
P4.0
Details
(1)
Bit
7
Symbol
P4.7
P4.6
P4.5
P4.4
P4.3
P4.2
P4.1
P4.0
R/W
R,W
R,W
R,W
R,W
R,W
R,W
R,W
R,W
Function
Port pin 4.7
Port pin 4.6
Port pin 4.5
Port pin 4.4
Port pin 4.3
Port pin 4.2
Port pin 4.1
Port pin 4.0
6
5
4
3
2
1
0
Note: 1. Write '1' or '0' for pin output. Read for pin input, but prior to READ, this bit must have been set to '1' by firmware or by a reset event.
58/231
uPSD33xx
Alternate Functions. There are five SFRs used
to control the mapping of alternate functions onto
MCU port pins, and these SFRs are depicted as
switches in Figure 16., page 55.
driver are disabled. The analog input is routed di-
rectly to the ADC unit. Only Port 1 supports analog
functions (Figure 17., page 56). Port 1 is not 5V
tolerant.
2
■
Port 3 uses the SFR, P3SFS (Table
28., page 60).
If the alternate function is I C, the related pins will
be in open drain mode, which is just like quasi-bi-
directional mode but the high-side driver is not en-
abled for one cycle when outputting a 0-to-1
transition. Only the low-side driver and the internal
weak pull-ups are used. Only Port 3 supports
■
Port 1 uses SFRs, P1SFS0 (Table
29., page 60) and P1SFS1 (Table
30., page 60).
■
Port 4 uses SFRs, P4SFS0 (Table
32., page 61) and P4SFS1 (Table
33., page 61).
2
open-drain mode (Figure 18., page 56). I C re-
quires the use of an external pull-up resistor on
each bus signal, typically 4.7KΩ to V
.
CC
Since these SFRs are cleared by a reset, then by
default all port pins function as GPIO (not the alter-
nate function) until firmware initializes these SFRs.
If the alternate function is PCA output, then the re-
lated pins are in push-pull mode, meaning the pins
are actively driven and held to logic '1' by the high-
side driver, or actively driven and held to logic '0'
by the low-side driver. Only Port 4 supports push-
pull mode (Figure 19., page 57). Port 4 push-pull
Each pin on each of the three ports can be inde-
pendently assigned a different function on a pin-
by-pin basis.
2
pins can source I
current when driving logic '1,'
The peripheral functions Timer 2, UART1, and I C
OH
and sink I
current when driving logic '0.' This
may be split independently between Port 1 and
Port 4 for additional flexibility by giving a wider
choice of peripheral usage on a limited number of
device pins.
OL
current is significantly more than the capability of
pins on Port
129., page 207).
1
or Port
3
(see Table
For example, to assign these port functions:
When the selected alternate function is UART0,
UART1, or SPI, then the related pins are in quasi-
bidirectional mode, including the use of the high-
side driver for rapid 0-to-1 output transitions. The
high-side driver is enabled for just one MCU_CLK
period on 0-to-1 transitions by the delay function at
the “digital_alt_func_data_out” signal pictured in
Figure 17., page 56 through Figure 19., page 57.
If the alternate function is Timer 0, Timer 1, Timer
2, or PCA input, then the related pins are in quasi-
bidirectional mode, but input only.
If the alternate function is ADC, then for each pin
the pull-ups, the high-side driver, and the low-side
■
■
■
Port 1: UART1, ADC[1:0], P1[7:4] are GPIO
Port 3: UART0, I C, P3[5:2] are GPIO
Port 4: TCM0, SPI, P4[3:1] are GPIO
2
The following values need to be written to the
SFRs:
P1SFS0 = 00001111b, or 0Fh
P1SFS1 = 00000011b , or 03h
P3SFS = 11000011b, or C3h
P4SFS0 = 11110001b, or F1h
P4SFS1 = 11110000b, or F0h
59/231
uPSD33xx
Table 28. P3SFS: Port 3 Special Function Select Register (SFR 91h, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
P3SFS7
P3SFS6
P3SFS5
P3SFS4
P3SFS3
P3SFS2
P3SFS1
P3SFS0
Details
Default Port Function
Alternate Port Function
P3SFS[i] - 1; Port 3 Pin, i = 0..7
UART0 Receive, RXD0
Port 3 Pin
R/W
P3SFS[i] - 0; Port 3 Pin, i = 0..7
0
1
2
3
4
5
6
R,W
R,W
R,W
R,W
R,W
R,W
R,W
GPIO
GPIO
GPIO
GPIO
GPIO
GPIO
GPIO
UART0 Transmit, TXD0
Ext Intr 0/Timer 0 Gate, EXT0INT/TG0
Ext Intr 1/Timer 1 Gate, EXT1INT/TG1
Counter 0 Input, C0
Counter 0 Input, C1
2
I C Data, I2CSDA
2
7
R,W
GPIO
I C Clock, I2CCL
Table 29. P1SFS0: Port 1 Special Function Select 0 Register (SFR 8Eh, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
P1SF07
P1SF06
P1SF05
P1SF04
P1SF03
P1SF02
P1SF01
P1SF00
Details
Table 30. P1SFS1: Port 1 Special Function Select 1 Register (SFR 8Fh, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
P1SF17
P1SF16
P1SF15
P1SF14
P1SF13
P1SF12
P1SF11
P1SF10
Table 31. P1SFS0 and P1SFS1 Details
Default Port Function
Alternate 1 Port Function Alternate 2 Port Function
P1SFS0[i] = 0
P1SFS1[i] = x
P1SFS0[i] = 1
P1SFS1[i] = 0
P1SFS0[i] = 1
P1SFS1[i] = 1
Port 1 Pin
R/W
Port 1 Pin, i = 0.. 7
GPIO
Port 1 Pin, i = 0.. 7
Timer 2 Count Input, T2
Timer 2 Trigger Input, TX2
UART1 Receive, RXD1
UART1 Transmit, TXD1
SPI Clock, SPICLK
Port 1 Pin, i = 0.. 7
0
1
2
3
4
5
6
7
R,W
R,W
R,W
R,W
R,W
R,W
R,W
R,W
ADC Chn 0 Input, ADC0
ADC Chn 1 Input, ADC1
ADC Chn 2 Input, ADC2
ADC Chn 3 Input, ADC3
ADC Chn 4 Input, ADC4
ADC Chn 5 Input, ADC5
ADC Chn 6 Input, ADC6
ADC Chn 7 Input, ADC7
GPIO
GPIO
GPIO
GPIO
GPIO
SPI Receive, SPIRXD
SPI Transmit, SPITXD
SPI Select, SPISEL_
GPIO
GPIO
60/231
uPSD33xx
Table 32. P4SFS0: Port 4 Special Function Select 0 Register (SFR 92h, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
P4SF07
P4SF06
P4SF05
P4SF04
P4SF03
P4SF02
P4SF01
P4SF00
Details
Table 33. P4SFS1: Port 4 Special Function Select 1 Register (SFR 93h, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
P4SF17
P4SF16
P4SF15
P4SF14
P4SF13
P4SF12
P4SF11
P4SF10
Table 34. P4SFS0 and P4SFS1 Details
Default Port Function
Alternate 1 Port Function Alternate 2 Port Function
P4SFS0[i] = 0
P4SFS1[i] = x
P4SFS0[i] = 1
P4SFS1[i] = 0
P4SFS0[i] = 1
P4SFS1[i] = 1
Port 4 Pin
R/W
Port 4 Pin, i = 0.. 7
GPIO
Port 4 Pin, i = 0.. 7
PCA0 Module 0, TCM0
PCA0 Module 1, TCM1
PCA0 Module 2, TCM2
PCA0 Ext Clock, PCACLK0
PCA1 Module 3, TCM3
PCA1 Module 4, TCM4
PCA1 Module 5, TCM5
PCA1 Ext Clock, PCACLK1
Port 4 Pin, i = 0.. 7
Timer 2 Count Input, T2
Timer 2 Trigger Input, TX2
UART1 Receive, RXD1
UART1 Transmit, TXD1
SPI Clock, SPICLK
0
1
2
3
4
5
6
7
R,W
R,W
R,W
R,W
R,W
R,W
R,W
R,W
GPIO
GPIO
GPIO
GPIO
GPIO
SPI Receive, SPIRXD
SPI Transmit, SPITXD
SPI Select, SPISEL_
GPIO
GPIO
61/231
uPSD33xx
MCU BUS INTERFACE
The MCU Module has a programmable bus inter-
face. It is based on a standard 8032 bus, with eight
data signals multiplexed with eight low-order ad-
dress signals (AD[7:0]). It also has eight high-or-
der non-multiplexed address signals (A[15:8]).
Time multiplexing is controlled by the address
latch signal, ALE.
This bus connects the MCU Module to the PSD
Module, and also connects to external pins only on
80-pin devices. See the AC specifications section
at the end of this document for external bus timing
on 80-pin devices.
Bits in the BUSCON Register determine the num-
ber of MCU_CLK periods per bus cycle for each of
these kinds of transfers to all address ranges.
It is not possible to specify in the BUSCON Regis-
ter a different number of MCU_CLK periods for
various address ranges. For example, the user
cannot specify 4 MCU_CLK periods for RD read
cycles to one address range on the PSD Module,
and 5 MCU_CLK periods for RD read cycles to a
different address range on an external device.
However, the user can specify one number of
clock periods for PSEN read cycles and a different
number of clock periods for RD read cycles.
Note 1: A PSEN bus cycle in progress may be
aborted before completion if the PFQ and Branch
Cache (BC) determines the current code fetch cy-
cle is not needed.
Four types of data transfers are supported, each
transfer is to/from a memory location external to
the MCU Module:
–
Code Fetch cycle using the PSEN signal: fetch
a code byte for execution
Note 2: Whenever the same number of MCU_CLK
periods is specified in BUSCON for both PSEN
and RD cycles, the bus cycle timing is typically
identical for each of these types of bus cycles. In
this case, the only time PSEN read cycles are
longer than RD read cycles is when the PFQ is-
sues a stall while reloading. PFQ stalls do not af-
fect RD read cycles. By comparison, in many
traditional 8051 architectures, RD bus cycles are
always longer than PSEN bus cycles.
–
Code Read cycle using PSEN: read a code
byte using the MOVC (Move Constant)
instruction
XDATA Read cycle using the RD signal: read
a data byte using the MOVX (Move eXternal)
instruction
–
–
XDATA Write cycle using the WR signal: write
a data byte using the MOVX instruction
The number of MCU_CLK periods for these trans-
fer types can be specified at runtime by firmware
writing to the SFR register named BUSCON (Ta-
ble 35., page 63). Here, the number of MCU_CLK
clock pulses per bus cycle are specified to maxi-
mize performance.
Important: By default, the BUSCON Register is
loaded with long bus cycle times (6 MCU_CLK pe-
riods) after a reset condition. It is important that the
post-reset initialization firmware sets the bus cycle
times appropriately to get the most performance,
according to Table 36., page 64. Keep in mind that
the PSD Module has a faster Turbo Mode (default)
and a slower but less power consuming Non-Tur-
bo Mode. The bus cycle times must be pro-
grammed in BUSCON to optimize for each mode
as shown in Table 36., page 64. See PLD Non-
Turbo Mode, page 192 for more details.
Bus Write Cycles (WR)
When the WR signal is used, a byte of data is writ-
ten directly to the PSD Module or external device,
no PFQ or caching is involved. Bits in the BUS-
CON Register determine the number of
MCU_CLK periods for bus write cycles to all ad-
dresses. It is not possible to specify in BUSCON a
different number of MCU_CLK periods for writes to
various address ranges.
Controlling the PFQ and BC
The BUSCON Register allows firmware to enable
and disable the PFQ and BC at run-time. Some-
times it may be desired to disable the PFQ and BC
to ensure deterministic execution. The dynamic
action of the PFQ and BC may cause varying pro-
gram execution times depending on the events
that happen prior to a particular section of code of
interest. For this reason, it is not recommended to
implement timing loops in firmware, but instead
use one of the many hardware timers in the
uPSD33xx.
Bus Read Cycles (PSEN or RD)
When the PSEN signal is used to fetch a byte of
code, the byte is read from the PSD Module or ex-
ternal device and it enters the MCU Pre-Fetch
Queue (PFQ). When PSEN is used during a
MOVC instruction, or when the RD signal is used
to read a byte of data, the byte is routed directly to
the MCU, bypassing the PFQ.
By default, the PFQ and BC are enabled after a re-
set condition.
Important: Disabling the PFQ or BC will seriously
reduce MCU performance.
62/231
uPSD33xx
Table 35. BUSCON: Bus Control Register (SFR 9Dh, reset value EBh)
Bit 7
Bit 6
EBC
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
EPFQ
WRW[1:0]
RDW[1:0]
CW[1:0]
Details
Bit
Symbol
R/W
Definition
Enable Pre-Fetch Queue
7
EPFQ
R,W
0 = PFQ is disabled
1 = PFQ is enabled (default)
Enable Branch Cache
6
EBC
R,W
R,W
0 = BC is disabled
1 = BC is enabled (default)
WR Wait, number of MCU_CLK periods for WR write bus cycle during
any MOVX instruction
5:4
WRW[1:0]
00b: 4 clock periods
01b: 5 clock periods
10b: 6 clock periods (default)
11b: 7 clock periods
RD Wait, number of MCU_CLK periods for RD read bus cycle during any
MOVX instruction
3:2
1:0
RDW[1:0]
R,W
00b: 4 clock periods
01b: 5 clock periods
10b: 6 clock periods (default)
11b: 7 clock periods
Code Wait, number of MCU_CLK periods for PSEN read bus cycle
during any code byte fetch or during any MOVC code byte read
instruction. Periods will increase with PFQ stall
CW[1:0]
R,W
00b: 3 clock periods - exception, for MOVC instructions this setting
results 4 clock periods
01b: 4 clock periods
10b: 5 clock periods
11b: 6 clock periods (default)
63/231
uPSD33xx
Table 36. Number of MCU_CLK Periods Required to Optimize Bus Transfer Rate
RDW[1:0] Clk
WRW[1:0] Clk
Periods
CW[1:0] Clk Periods
MCU Clock Frequency,
Periods
MCU_CLK (f
)
MCU
(1)
(1)
(1)
(1)
(1)
(1)
3.3V
5V
3.3V
5V
3.3V
5V
(2)
5
4
5
4
5
4
40MHz, Turbo mode PSD
40MHz, Non-Turbo mode PSD
36MHz, Turbo mode PSD
6
5
6
5
5
4
5
4
4
3
3
5
4
4
4
4
3
4
3
3
3
3
6
5
6
5
5
4
5
4
4
4
4
5
4
4
4
4
4
4
4
4
4
4
6
5
6
5
5
4
5
4
4
4
4
5
4
4
4
4
4
4
4
4
4
4
36MHz, Non-Turbo mode PSD
32MHz, Turbo mode PSD
32MHz, Non-Turbo mode PSD
28MHz, Turbo mode PSD
28MHz, Non-Turbo mode PSD
24MHz, Turbo mode PSD
24MHz, Non-Turbo mode PSD
20MHz and below, Turbo mode PSD
20MHz and below, Non-Turbo mode PSD
Note: 1. V of the PSD Module
DD
2. “Turbo mode PSD” means that the PSD Module is in the faster, Turbo mode (default condition). A PSD Module in Non-Turbo mode
is slower, but consumes less current. See PSD Module section, titled “PLD Non-Turbo Mode” for details.
64/231
uPSD33xx
SUPERVISORY FUNCTIONS
Supervisory circuitry on the MCU Module will issue
an internal reset signal to the MCU Module and si-
multaneously to the PSD Module as a result of any
of the following four events:
“RESET_OUT” signal from a PLD output, the user
can choose to make it either active-high or active-
low logic, depending on the PLD equation.
–
–
The external RESET_IN pin is asserted
The Low Voltage Detect (LVD) circuitry has
External Reset Input Pin, RESET_IN
The RESET_IN pin can be connected directly to a
mechanical reset switch or other device which
pulls the signal to ground to invoke a reset.
detected a voltage on V below a specific
CC
threshold (power-on or voltage sags)
–
–
The JTAG Debug interface has issued a reset
command
The Watch Dog Timer (WDT) has timed out
RESET_IN is pulled up internally and enters a
Schmitt trigger input buffer with a voltage hystere-
sis of V
for immunity to the effects of slow
RST_HYS
signal rise and fall times, as shown in Figure 20.
RESET_IN is also filtered to reject a voltage spike
The resulting internal reset signal, MCU_RESET,
will force the 8032 into a known reset state while
asserted, and then 8032 program execution will
jump to the reset vector at program address 0000h
just after MCU_RESET is deasserted. The MCU
Module will also assert an active low internal reset
signal, RESET, to the PSD Module. If needed, the
signal RESET can be driven out to external sys-
tem components through any PLD output pin on
less than a duration of t
. The RESET_IN
RST_FIL
signal must be maintained at a logic '0' for at least
a duration of t while the oscillator is run-
RST_LO_IN
ning. The resulting MCU_RESET signal will last
only as long as the RESET_IN signal is active (it is
not stretched). Refer to the Supervisor AC specifi-
cations in Table 150., page 221 at the end of this
document for these parameter values.
the
PSD
Module.
When
driving
this
Figure 20. Supervisor Reset Generation
V
CC
PULL-UP
RESET_IN
MCU
Clock
Sync
MCU_RESET
to MCU and
Peripherals
PIN
Noise Filter
WDT
LVD
S
R
Q
RESET
to PSD Module
JTAG Debug
DELAY,
t
RST_ACTV
AI09603
65/231
uPSD33xx
Low V Voltage Detect, LVD
CC
An internal reset is generated by the LVD circuit
By default, the WDT is disabled after each reset.
when V
drops below the reset threshold,
CC
Note: The WDT is not active during Idle mode or
Power-down Mode.
There are two SFRs that control the WDT, they are
WDKEY (Table 37., page 68) and WDRST (Table
38., page 68).
V
. After V returns to the reset thresh-
LV_THRESH
CC
old, the MCU_RESET signal will remain asserted
for t before it is released. The LVD circuit
RST_ACTV
is always enabled (cannot be disabled by SFR),
even in Idle Mode and Power-down Mode. The
If WDKEY contains 55h, the WDT is disabled. Any
value other than 55h in WDKEY will enable the
WDT. By default, after any reset condition, WD-
KEY is automatically loaded with 55h, disabling
the WDT. It is the responsibility of initialization
firmware to write some value other than 55h to
WDKEY after each reset if the WDT is to be used.
LVD input has a voltage hysteresis of V
RST_HYS
and will reject voltage spikes less than a duration
of t
.
RST_FIL
Important: The LVD voltage threshold is
V
V
, suitable for monitoring both the 3.3V
supply on the MCU Module and the 3.3V V
LV_THRESH
CC
DD
supply on the PSD Module for 3.3V uPSD33xxV
devices, since these supplies are one in the same
on the circuit board.
The WDT consists of a 24-bit up-counter (Figure
21), whose initial count is 000000h by default after
every reset. The most significant byte of this
counter is controlled by the SFR, WDRST. After
being enabled by WDKEY, the 24-bit count is in-
creased by 1 for each MCU machine cycle. When
However, for 5V uPSD33xx devices, V
is not suitable for monitoring the 5V V
LV_THRESH
DD
voltage
supply (V
itoring the 3.3V V
is too low), but good for mon-
LV_THRESH
supply. In the case of 5V
CC
24
the count overflows beyond FFFFFh (2 MCU
uPSD33xx devices, an external means is required
machine cycles), a reset is issued and the WDT is
automatically disabled (WDKEY = 55h again).
to monitor the separate 5V V supply, if desired.
DD
Power-up Reset
To prevent the WDT from timing out and generat-
ing a reset, firmware must repeatedly write some
value to WDRST before the count reaches
FFFFFh. Whenever WDRST is written, the upper
8 bits of the 24-bit counter are loaded with the writ-
ten value, and the lower 16 bits of the counter are
cleared to 0000h.
The WDT time-out period can be adjusted by writ-
ing a value other that 00h to WDRST. For exam-
ple, if WDRST is written with 04h, then the WDT
will start counting 040000h, 040001h, 040002h,
and so on for each MCU machine cycle. In this ex-
ample, the WDT time-out period is shorter than if
WDRST was written with 00h, because the WDT
is an up-counter. A value for WDRST should never
be written that results in a WDT time-out period
shorter than the time required to complete the
longest code task in the application, else unwant-
ed WDT overflows will occur.
At power up, the internal reset generated by the
LVD circuit is latched as a logic '1' in the POR bit
of the SFR named PCON (Table 24., page 50).
Software can read this bit to determine whether
the last MCU reset was the result of a power up
(cold reset) or a reset from some other condition
(warm reset). This bit must be cleared with soft-
ware.
JTAG Debug Reset
The JTAG Debug Unit can generate a reset for de-
bugging purposes. This reset source is also avail-
able when the MCU is in Idle Mode and Power-
Down Mode (the JTAG debugger can be used to
exit these modes).
Watchdog Timer, WDT
When enabled, the WDT will generate a reset
whenever it overflows. Firmware that is behaving
correctly will periodically clear the WDT before it
overflows. Run-away firmware will not be able to
clear the WDT, and a reset will be generated.
Figure 21. Watchdog Counter
23
15
7
0
8-bits
8-bits
8-bits
SFR, WDRST
AI09604
66/231
uPSD33xx
The formula to determine WDT time-out period is:
4. Assume there are no stalls from the PFQ/BC.
In reality, there are occational stalls but their
occurance has minimal impact on WDT
timeout period.
WDT
= t
x N
PERIOD
MACH_CYC OVERFLOW
N
is the number of WDT up-counts re-
OVERFLOW
24
quired to reach FFFFFFh. This is determined by
the value written to the SFR, WDRST.
5. WDRST contains 00h, meaning a full 2 up-
counts are required to reach FFFFFh and
generate a reset.
t
is the average duration of one MCU
MACH_CYC
machine cycle. By default, an MCU machine cycle
is always 4 MCU_CLK periods for uPSD33xx, but
the following factors can sometimes add more
MCU_CLK periods per machine cycle:
In this example,
t
= 100ns (4 MCU_CLK periods x 25ns)
MACH_CYC
24
N
= 2 = 16777216 up-counts
OVERFLOW
WDT
= 100ns X 16777216 = 1.67 seconds
PERIOD
–
The number of MCU_CLK periods assigned to
MCU memory bus cycles as determined in the
SFR, BUSCON. If this setting is greater than
4, then machine cycles have additional
The actual value will be slightly longer due to PFQ/
BC.
Firmware Example: The following 8051 assem-
bly code illustrates how to operate the WDT. A
simple statement in the reset initialization firmware
enables the WDT, and then a periodic write to
clear the WDT in the main firmware is required to
keep the WDT from overflowing. This firmware is
MCU_CLK periods during memory transfers.
–
Whether or not the PFQ/BC circuitry issues a
stall during a particular MCU machine cycle. A
stall adds more MCU_CLK periods to a
machine cycle until the stall is removed.
t
is also affected by the absolute time of
MACH_CYC
based on the example above (40MHz f
,
OSC
a single MCU_CLK period. This number is fixed by
the following factors:
CCON0 = 10h, BUSCON = C1h).
For example, in the reset initialization firmware
(the function that executes after a jump to the reset
vector):
–
Frequency of the external crystal, resonator,
or oscillator: (f
)
OSC
–
Bit settings in the SFR CCON0, which can
divide f and change MCU_CLK
OSC
MOV AE, #AA
; enable WDT by writing value to
; WDKEY other than 55h
As an example, assume the following:
1. is 40MHz, thus its period is 25ns.
f
OSC
Somewhere in the flow of the main program, this
statement will execute periodically to reset the
WDT before it’s time-out period of 1.67 seconds.
For example:
2. CCON0 is 10h, meaning no clock division, so
the period of MCU_CLK is also 25ns.
3. BUSCON is C1h, meaning the PFQ and BC
are enabled, and each MCU memory bus
cycle is 4 MCU_CLK periods, adding no
additional MCU_CLK periods to MCU
MOV A6, #00
; reset WDT, loading 000000h.
; Counting will automatically
; resume as long as 55h in not in
; WDKEY
machine cycles during memory transfers.
67/231
uPSD33xx
Table 37. WDKEY: Watchdog Timer Key Register (SFR AEh, reset value 55h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
WDKEY[7:0]
Details
Bit
Symbol
R/W
Definition
55h disables the WDT from counting. 55h is automatically loaded in this
SFR after any reset condition, leaving the WDT disabled by default.
[7:0]
WDKEY
W
Any value other than 55h written to this SFR will enable the WDT, and
counting begins.
Table 38. WDRST: Watchdog Timer Reset Counter Register (SFR A6h, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
WDRST[7:0]
Details
Bit
Symbol
R/W
Definition
This SFR is the upper byte of the 24-bit WDT up-counter. Writing this
SFR sets the upper byte of the counter to the written value, and clears
the lower two bytes of the counter to 0000h.
[7:0]
WDRST
W
Counting begins when WDKEY does not contain 55h.
68/231
uPSD33xx
STANDARD 8032 TIMER/COUNTERS
There are three 8032-style 16-bit Timer/Counter
registers (Timer 0, Timer 1, Timer 2) that can be
configured to operate as timers or event counters.
Clock Sources
When enabled in the “Timer” function, the Regis-
ters THx and TLx are incremented every 1/12 of
There are two additional 16-bit Timer/Counters in
the Programmable Counter Array (PCA), seePCA
Block, page 123 for details.
the oscillator frequency (f
source is not effected by MCU clock dividers in the
CCON0, stalls from PFQ/BC, or bus transfer cy-
). This timer clock
OSC
cles. Timers are always clocked at 1/12 of f
.
OSC
Standard Timer SFRs
When enabled in the “Counter” function, the Reg-
isters THx and TLx are incremented in response to
a 1-to-0 transition sampled at their corresponding
external input pin: pin C0 for Timer 0; pin C1 for
Timer 1; or pin T2 for Timer 2. In this function, the
external clock input pin is sampled by the counter
Timer 0 and Timer 1 have very similar functions,
and they share two SFRs for control:
■
TCON (Table 39., page 70)
TMOD (Table 40., page 72).
■
Timer 0 has two SFRs that form the 16-bit counter,
or that can hold reload values, or that can scale
the clock depending on the timer/counter mode:
at a rate of 1/12 of f
. When a logic '1' is deter-
OSC
mined in one sample, and a logic '0' in the next
sample period, the count is incremented at the
very next sample period (period1: sample=1,
period2: sample=0, period3: increment count
while continuing to sample). This means the max-
■
TH0 is the high byte, address 8Ch
TL0 is the low byte, address 8Ah
■
Timer 1 has two similar SFRs:
imum count rate is 1/24 of the f
. There are no
OSC
restrictions on the duty cycle of the external input
signal, but to ensure that a given level is sampled
at least once before it changes, it should be active
■
TH1 is the high byte, address 8Dh
TL1 is the low byte, address 8Bh
■
Timer 2 has one control SFR:
T2CON (Table 41., page 75)
for at least one full sample period (12 / f
sec-
OSC,
onds). However, if MCU_CLK is divided by the
SFR CCON0, then the sample period must be cal-
culated based on the resultant, longer, MCU_CLK
frequency. In this case, an external clock signal on
pins C0, C1, or T2 should have a duration longer
■
Timer 2 has two SFRs that form the 16-bit counter,
and perform other functions:
■
TH2 is the high byte, address CDh
TL2 is the low byte, address CCh
than one MCU machine cycle, t
. The
MACH_CYC
■
section, Watchdog Timer, WDT, page 66 explains
how to estimate t
.
MACH_CYC
Timer 2 has two SFRs for capture and reload:
■
RCAP2H is the high byte, address CBh
RCAP2L is the low byte, address CAh
■
69/231
uPSD33xx
Table 39. TCON: Timer Control Register (SFR 88h, reset value 00h)
Bit 7
TF1
Bit 6
TR1
Bit 5
TF0
Bit 4
TR0
Bit 3
IE1
Bit 2
IT1
Bit 1
IE0
Bit 0
IT0
Details
Bit
Symbol
TF1
R/W
R
Definition
Timer 1 overflow interrupt flag. Set by hardware upon overflow.
Automatically cleared by hardware after firmware services the interrupt
7
for Timer 1.
6
TR1
R,W
R
Timer 1 run control. 1 = Timer/Counter 1 is on, 0 = Timer/Counter 1 is off.
Timer 0 overflow interrupt flag. Set by hardware upon overflow.
Automatically cleared by hardware after firmware services the interrupt
for Timer 0.
5
TF0
4
TR0
R,W
R
Timer 0 run control. 1 = Timer/Counter 0 is on, 0 = Timer/Counter 0 is off.
Interrupt flag for external interrupt pin, EXTINT1. Set by hardware when
edge is detected on pin. Automatically cleared by hardware after
firmware services EXTINT1 interrupt.
3
IE1
Trigger type for external interrupt pin EXTINT1. 1 = falling edge, 0 = low-
level
2
1
0
IT1
IE0
IT0
R,W
R
Interrupt flag for external interrupt pin, EXTINT0. Set by hardware when
edge is detected on pin. Automatically cleared by hardware after
firmware services EXTINT0 interrupt.
Trigger type for external interrupt pin EXTINT0. 1 = falling edge, 0 = low-
level
R,W
70/231
uPSD33xx
SFR, TCON
Timer 0 and Timer 1 share the SFR, TCON, that
controls these timers and provides information
about them. See Table 39., page 70.
Bits IE0 and IE1 are not related to Timer/Counter
functions, but they are set by hardware when a
signal is active on one of the two external interrupt
pins, EXTINT0 and EXTINT1. For system informa-
tion on all of these interrupts, see Table
16., page 41, Interrupt Summary.
Bits IT0 and IT1 are not related to Timer/Counter
functions, but they control whether or not the two
external interrupt input pins, EXTINT0 and
EXTINT1 are edge or level triggered.
Mode 0 operation is the same for the Timer 0 as
for Timer 1. Substitute TR0, TF0, C0, TL0, TH0,
and EXTINT0 for the corresponding Timer 1 sig-
nals in Figure 22. There are two different GATE
Bits, one for Timer 1 and one for Timer 0.
Mode 1. Mode 1 is the same as Mode 0, except
that the Timer Register is being run with all 16 bits.
Mode 2. Mode 2 configures the Timer Register as
an 8-bit Counter (TL1) with automatic reload, as
shown in Figure 23., page 73. Overflow from TL1
not only sets TF1, but also reloads TL1 with the
contents of TH1, which is preset with firmware.
The reload leaves TH1 unchanged. Mode 2 oper-
ation is the same for Timer/Counter 0.
SFR, TMOD
Mode 3. Timer 1 in Mode 3 simply holds its count.
The effect is the same as setting TR1 = 0.
Timer 0 and Timer 1 have four modes of operation
controlled by the SFR named TMOD (Table 40).
Timer 0 in Mode 3 establishes TL0 and TH0 as two
separate counters. The logic for Mode 3 on Timer
0 is shown in Figure 24., page 73. TL0 uses the
Timer 0 control Bits: C/T, GATE, TR0, and TF0, as
well as the pin EXTINT0. TH0 is locked into a timer
Timer 0 and Timer 1 Operating Modes
The “Timer” or “Counter” function is selected by
the C/T control bits in TMOD. The four operating
modes are selected by bit-pairs M[1:0] in TMOD.
Modes 0, 1, and 2 are the same for both Timer/
Counters. Mode 3 is different.
Mode 0. Putting either Timer/Counter into Mode 0
makes it an 8-bit Counter with a divide-by-32 pre-
scaler. Figure 22 shows Mode 0 operation as it ap-
plies to Timer 1 (same applies to Timer 0).
In this mode, the Timer Register is configured as a
13-bit register. As the count rolls over from all '1s'
to all '0s,' it sets the Timer Interrupt flag TF1. The
counted input is enabled to the Timer when
TR1 = 1 and either GATE = 0 or EXTINT1 = 1.
(Setting GATE = 1 allows the Timer to be con-
trolled by external input pin, EXTINT1, to facilitate
pulse width measurements). TR1 is a control bit in
the SFR, TCON. GATE is a bit in the SFR, TMOD.
function (counting at a rate of 1/12 f
) and takes
OSC
over the use of TR1 and TF1 from Timer 1. Thus,
TH0 now controls the “Timer 1“ interrupt flag.
Mode 3 is provided for applications requiring an
extra 8-bit timer on the counter (see Figure
24., page 73). With Timer 0 in Mode 3, a
uPSD33xx device can look like it has three Timer/
Counters (not including the PCA). When Timer 0 is
in Mode 3, Timer 1 can be turned on and off by
switching it out of and into its own Mode 3, or can
still be used by the serial port as a baud rate gen-
erator, or in fact, in any application not requiring an
interrupt.
The 13-bit register consists of all 8 bits of TH1 and
the lower 5 bits of TL1. The upper 3 bits of TL1 are
indeterminate and should be ignored. Setting the
run flag, TR1, does not clear the registers.
71/231
uPSD33xx
Table 40. TMOD: Timer Mode Register (SFR 89h, reset value 00h)
Bit 7
Bit 6
C/T
Bit 5
Bit 4
Bit 3
Bit 2
C/T
Bit 1
Bit 0
GATE
M[1:0]
GATE
M[1:0]
Details
Bit
Symbol
R/W
Timer
Definition (T/C is abbreviation for Timer/Counter)
Gate control.
7
GATE
R,W
When GATE = 1, T/C is enabled only while pin EXTINT1
is '1' and the flag TR1 is '1.' When GATE = 0, T/C is
enabled whenever the flag TR1 is '1.'
Counter or Timer function select.
6
C/T
R,W
R,W
When C/T = 0, function is timer, clocked by internal clock.
C/T = 1, function is counter, clocked by signal sampled on
external pin, C1.
Timer 1
Mode Select.
00b = 13-bit T/C. 8 bits in TH1 with TL1 as 5-bit pre-
scaler.
01b = 16-bit T/C. TH1 and TL1 are cascaded. No pre-
[5:4]
M[1:0]
scaler.
10b = 8-bit auto-reload T/C. TH1 holds a constant and
loads into TL1 upon overflow.
11b = Timer Counter 1 is stopped.
Gate control.
3
2
GATE
C/T
R,W
R,W
When GATE = 1, T/C is enabled only while pin EXTINT0
is '1' and the flag TR0 is '1.' When GATE = 0, T/C is
enabled whenever the flag TR0 is '1.'
Counter or Timer function select.
When C/T = 0, function is timer, clocked by internal clock.
C/T = 1, function is counter, clocked by signal sampled on
external pin, C0.
Timer 0
Mode Select.
00b = 13-bit T/C. 8 bits in TH0 with TL0 as 5-bit pre-
scaler.
01b = 16-bit T/C. TH0 and TL0 are cascaded. No pre-
scaler.
[1:0]
M[1:0]
R,W
10b = 8-bit auto-reload T/C. TH0 holds a constant and
loads into TL0 upon overflow.
11b = TL0 is 8-bit T/C controlled by standard Timer 0
control bits. TH0 is a separate 8-bit timer that uses Timer
1 control bits.
72/231
uPSD33xx
Figure 22. Timer/Counter Mode 0: 13-bit Counter
fOSC
÷ 12
C/T = 0
C/T = 1
TH1
(8 bits)
TL1
(5 bits)
TF1
Interrupt
C1 pin
Control
TR1
Gate
EXTINT1 pin
AI06622
Figure 23. Timer/Counter Mode 2: 8-bit Auto-reload
fOSC
÷ 12
C/T = 0
C/T = 1
TL1
(8 bits)
TF1
Interrupt
C1 pin
Control
TR1
Gate
EXTINT1 pin
TH1
(8 bits)
AI06623
Figure 24. Timer/Counter Mode 3: Two 8-bit Counters
fOSC
÷ 12
C/T = 0
C/T = 1
TL0
(8 bits)
TF0
Interrupt
C0 pin
Control
TR0
Gate
EXTINT0 pin
TH0
(8 bits)
fOSC
TF1
Interrupt
÷ 12
Control
TR1
AI06624
73/231
uPSD33xx
Timer 2
Timer 2 can operate as either an event timer or as
an event counter. This is selected by the bit C/T2
in the SFR named, T2CON (Table 41., page 75).
Timer 2 has three operating modes selected by
bits in T2CON, according to Table 42., page 76.
The three modes are:
the cause. Flags TF2 and EXF2 are not automati-
cally cleared by hardware, so the firmware servic-
ing the interrupt must clear the flag(s) upon exit of
the interrupt service routine.
Auto-reload Mode. In the Auto-reload Mode,
there are again two options, which are selected by
the bit EXEN2 in T2CON. Figure 26., page 79
shows Auto-reload mode.
If EXEN2 = 0, then when Timer 2 counts up and
rolls over from FFFFh it not only sets the interrupt
flag TF2, but also causes the Timer 2 registers to
be reloaded with the 16-bit value contained in
Registers RCAP2L and RCAP2H, which are pre-
set with firmware.
■
■
■
Capture mode
Auto re-load mode
Baud rate generator mode
Capture Mode. In Capture Mode there are two
options which are selected by the bit EXEN2 in
T2CON. Figure 25., page 79 illustrates Capture
mode.
If EXEN2 = 0, then Timer 2 is a 16-bit timer if C/T2
= 0, or it’s a 16-bit counter if C/T2 = 1, either of
which sets the interrupt flag bit TF2 upon overflow.
If EXEN2 = 1, then Timer 2 still does the above,
but with the added feature that a 1-to-0 transition
at external input T2X will also trigger the 16-bit re-
load and set the interrupt flag EXF2. Again, firm-
ware servicing the interrupt must read both TF2
and EXF2 to determine the cause, and clear the
flag(s) upon exit.
Note: The uPSD33xx does not support selectable
up/down counting in Auto-reload mode (this fea-
ture was an extension to the original 8032 archi-
tecture).
If EXEN2 = 1, then Timer 2 still does the above,
but with the added feature that a 1-to-0 transition
at external input pin T2X causes the current value
in the Timer 2 registers, TL2 and TH2, to be cap-
tured into Registers RCAP2L and RCAP2H, re-
spectively. In addition, the transition at T2X
causes interrupt flag bit EXF2 in T2CON to be set.
Either flag TF2 or EXF2 will generate an interrupt
and the MCU must read both flags to determine
74/231
uPSD33xx
Table 41. T2CON: Timer 2 Control Register (SFR C8h, reset value 00h)
Bit 7
TF2
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
TR2
Bit 1
C/T2
Bit 0
EXF2
RCLK
TCLK
EXEN2
CP/RL2
Details
Bit
Symbol
R/W
Definition
Timer 2 flag, causes interrupt if enabled.
7
TF2
R,W
TF2 is set by hardware upon overflow. Must be cleared by firmware. TF2
will not be set when either RCLK or TCLK =1.
Timer 2 flag, causes interrupt if enabled.
6
5
EXF2
R,W
R,W
EXF2 is set when a capture or reload is caused by a negative transition
on T2X pin and EXEN2 = 1. EXF2 must be cleared by firmware.
UART0 Receive Clock control.
(1)
When RCLK = 1, UART0 uses Timer 2 overflow pulses for its receive
clock in Modes 1 and 3. RCLK=0, Timer 1 overflow is used for its receive
clock
RCLK
UART0 Transmit Clock control.
(1)
4
R,W
When TCLK = 1, UART0 uses Timer 2 overflow pulses for its transmit
clock in Modes 1 and 3. TCLK=0, Timer 1 overflow is used for transmit
clock
TCLK
Timer 2 External Enable.
3
2
1
EXEN2
TR2
R,W
R,W
R,W
When EXEN2 = 1, capture or reload results when negative edge on pin
T2X occurs. EXEN2 = 0 causes Timer 2 to ignore events at pin T2X.
Timer 2 run control.
1 = Timer/Counter 2 is on, 0 = Timer Counter 2 is off.
Counter or Timer function select.
C/T2
When C/T2 = 0, function is timer, clocked by internal clock. When C/T2 =
1, function is counter, clocked by signal sampled on external pin, T2.
Capture/Reload.
When CP/RL2 = 1, capture occurs on negative transition at pin T2X if
EXEN2 = 1. When CP/RL2 = 0, auto-reload occurs when Timer 2
overflows, or on negative transition at pin T2X when EXEN2=1. When
RCLK = 1 or TCLK = 1, CP/RL2 is ignored, and Timer 2 is forced to auto-
reload upon Timer 2 overflow
0
CP/RL2
R,W
Note: 1. The RCLK1 and TCLK1 Bits in the SFR named PCON control UART1, and have the exact same function as RCLK and TCLK.
75/231
uPSD33xx
Table 42. Timer/Counter 2 Operating Modes
Bits in T2CON SFR
Input Clock
Counter,
Pin
T2X
RCLK
or
Mode
Remarks
CP/
RL2
Timer,
External
TR2
EXEN2
Internal (Pin T2,
P1.0)
TCLK
reload [RCAP2H, RCAP2L] to [TH2,
TL2] upon overflow (up counting)
0
0
1
0
x
16-bit
Auto-
reload
MAX
f
/12
OSC
f
/24
OSC
reload [RCAP2H, RCAP2L] to [TH2,
TL2] at falling edge on pin T2X
0
0
0
1
1
1
1
0
↓
x
16-bit Timer/Counter (up counting)
MAX
16-bit
Capture
Capture [TH2, TL2] and store to
[RCAP2H, RCAP2L] at falling edge on
pin T2X
f
/12
/2
OSC
f
/24
OSC
0
1
1
1
↓
1
1
x
x
x
x
1
1
0
0
1
x
x
↓
x
No overflow interrupt request (TF2)
Extra Interrupt on pin T2X, sets TF2
Timer 2 stops
Baud Rate
Generator
f
–
–
OSC
Off
–
Note: ↓ = falling edge
76/231
uPSD33xx
Baud Rate Generator Mode. The RCLK and/or
TCLK Bits in the SFR T2CON allow the transmit
and receive baud rates on serial port UART0 to be
derived from either Timer 1 or Timer 2. Figure
27., page 80 illustrates Baud Rate Generator
Mode.
The timer can be configured for either “timer” or
“counter” operation. In the most typical applica-
tions, it is configured for “timer” operation (C/T2 =
0). “Timer” operation is a little different for Timer 2
when it's being used as a baud rate generator. In
this case, the baud rate is given by the formula:
When TCLK = 0, Timer 1 is used as UART0’s
transmit baud generator. When TCLK = 1, Timer 2
will be the transmit baud generator. RCLK has the
same effect for UART0’s receive baud rate. With
these two bits, UART0 can have different receive
and transmit baud rates - one generated by Timer
1, the other by Timer 2.
Note: Bits RCLK1 and TCLK1 in the SFR named
PCON (see PCON: Power Control Register (SFR
87h, reset value 00h), page 50) have identical
functions as RCLK and TCLK but they apply to
UART1 instead. For simplicity in the following dis-
cussions about baud rate generation, no suffix will
be used when referring to SFR registers and bits
related to UART0 or UART1, since each UART in-
terface has identical operation. Example, TCLK or
TCLK1 will be referred to as just TCLK.
UART Mode 1,3 Baud Rate =
f
/(32 x [65536 – [RCAP2H, RCAP2L]))
OSC
where [RCAP2H, RCAP2L] is the content of the
SFRs RCAP2H and RCAP2L taken as a 16-bit un-
signed integer.
A roll-over in TH2 does not set TF2, and will not
generate an interrupt. Therefore, the Timer Inter-
rupt does not have to be disabled when Timer 2 is
in the Baud Rate Generator Mode.
If EXEN2 is set, a 1-to-0 transition on pin T2X will
set the Timer 2 interrupt flag EXF2, but will not
cause a reload from RCAP2H and RCAP2L to
TH2 and TL2. Thus when Timer 2 is in use as a
baud rate generator, the pin T2X can be used as
an extra external interrupt, if desired.
When Timer 2 is running (TR2 = 1) in a “timer”
function in the Baud Rate Generator Mode, firm-
ware should not read or write TH2 or TL2. Under
these conditions the results of a read or write may
not be accurate. However, SFRs RCAP2H and
RCAP2L may be read, but should not be written,
because a write might overlap a reload and cause
write and/or reload errors. Timer 2 should be
turned off (clear TR2) before accessing Timer 2 or
Registers RCAP2H and RCAP2L, in this case.
The Baud Rate Generator Mode is similar to the
Auto-reload Mode, in that a roll over in TH2 causes
the Timer 2 registers, TH2 and TL2, to be reloaded
with the 16-bit value in Registers RCAP2H and
RCAP2L, which are preset with firmware.
The baud rates in UART Modes 1 and 3 are deter-
mined by Timer 2’s overflow rate as follows:
UART Mode 1,3 Baud Rate =
Timer 2 Overflow Rate / 16
Table 43., page 78 shows commonly used baud
rates and how they can be obtained from Timer 2,
with T2CON = 34h.
77/231
uPSD33xx
Table 43. Commonly Used Baud Rates Generated from Timer2 (T2CON = 34h)
Timer 2 SFRs
Desired
Resulting
Baud Rate
Baud Rate
Deviation
f
MHz
OSC
Baud Rate
RCAP2H (hex)
RCAP2L(hex)
F5
40.0
115200
57600
28800
19200
9600
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
FF
113636
56818
29070
19231
9615
-1.36%
40.0
40.0
EA
D5
-1.36%
0.94%
40.0
BF
0.16%
40.0
7E
0.16%
36.864
36.864
36.864
36.864
36.864
36.0
115200
57600
28800
19200
9600
F6
115200
57600
28800
19200
9600
0
EC
D8
0
0
C4
0
88
0
28800
19200
9600
D9
28846
19067
9615
0.16%
36.0
C5
-0.69%
36.0
8B
0.16%
24.0
57600
28800
19200
9600
F3
57692
28846
19231
9615
0.16%
24.0
E6
0.16%
24.0
D9
0.16%
24.0
B2
0.16%
12.0
28800
9600
F3
28846
9615
0.16%
12.0
D9
0.16%
11.0592
11.0592
11.0592
11.0592
11.0592
3.6864
3.6864
3.6864
3.6864
3.6864
1.8432
1.8432
115200
57600
28800
19200
9600
FD
FA
115200
57600
28800
19200
9600
0
0
0
0
0
0
0
0
0
0
0
0
F4
EE
DC
FF
115200
57600
28800
19200
9600
115200
57600
28800
19200
9600
FE
FC
FA
F4
19200
9600
FD
FA
19200
9600
78/231
uPSD33xx
Figure 25. Timer 2 in Capture Mode
fOSC
÷ 12
C/T2 = 0
C/T2 = 1
TH2
(8 bits)
TL2
(8 bits)
TF2
T2 pin
Control
TR2
Capture
Timer 2
Interrupt
RCAP2L RCAP2H
Transition
Detector
EXP2
T2X pin
Control
EXEN2
AI06625
Figure 26. Timer 2 in Auto-Reload Mode
fOSC
÷ 12
C/T2 = 0
C/T2 = 1
TH2
(8 bits)
TL2
(8 bits)
TF2
T2 pin
Control
TR2
Reload
Timer 2
Interrupt
RCAP2H
RCAP2L
Transition
Detector
T2X pin
EXP2
Control
EXEN2
AI06626
79/231
uPSD33xx
Figure 27. Timer 2 in Baud Rate Generator Mode
Timer 1 Overflow
Note: Oscillator frequency is divided by 2,
÷ 2
'0'
not 12 like in other timer modes.
'1'
SMOD
fOSC
÷ 12
C/T2 = 0
C/T2 = 1
'1'
'1'
'0'
TH2
(8 bits)
TL2
(8 bits)
RCLK
TCLK
T2 pin
Control
TR2
RX CLK
TX CLK
÷ 16
'0'
Reload
÷ 16
RCAP2L RCAP2H
Transition
Detector
Timer 2 Interrupt
EXF2
T2X pin
Control
EXEN2
Note: Availability of additional external interrupt.
AI09605
80/231
uPSD33xx
SERIAL UART INTERFACES
uPSD33xx devices provide two standard 8032
UART serial ports.
Mode 0. Mode 0 provides asynchronous, half-du-
plex operation. Serial data is both transmitted, and
received on the RxD pin. The TxD pin outputs a
shift clock for both transmit and receive directions,
thus the MCU must be the master. Eight bits are
transmitted/received LSB first. The baud rate is
–
The first port, UART0, is connected to pins
RxD0 (P3.0) and TxD0 (P3.1)
–
The second port, UART1 is connected to pins
RxD1 (P1.2) and TxD1 (P1.3). UART1 can
optionally be routed to pins P4.2 and P4.3 as
described in Alternate Functions, page 59.
fixed at 1/12 of f
.
OSC
Mode 1. Mode 1 provides standard asynchro-
nous, full-duplex communication using a total of 10
bits per data byte. Data is transmitted through TxD
and received through RxD with: a Start Bit (logic
'0'), eight data bits (LSB first), and a Stop Bit (logic
'1'). Upon receive, the eight data bits go into the
SFR SBUF, and the Stop Bit goes into bit RB8 of
the SFR SCON. The baud rate is variable and de-
rived from overflows of Timer 1 or Timer 2.
Mode 2. Mode 2 provides asynchronous, full-du-
plex communication using a total of 11 bits per
data byte. Data is transmitted through TxD and re-
ceived through RxD with: a Start Bit (logic '0');
eight data bits (LSB first); a programmable 9th
data bit; and a Stop Bit (logic '1'). Upon Transmit,
the 9th data bit (from bit TB8 in SCON) can be as-
signed the value of '0' or '1.' Or, for example, the
Parity Bit (P, in the PSW) could be moved into
TB8. Upon receive, the 9th data bit goes into RB8
in SCON, while the Stop Bit is ignored. The baud
rate is programmable to either 1/32 or 1/64 of
The operation of the two serial ports are the same
and are controlled by two SFRs:
■
SCON0 (Table 45., page 82) for UART0
SCON1 (Table 46., page 83) for UART1
■
Each UART has its own data buffer accessed
through an SFR listed below:
■
SBUF0 for UART0, address 99h
SBUF1 for UART1, address D9h
■
When writing SBU0 or SBUF1, the data automati-
cally loads into the associated UART transmit data
register. When reading this SFR, data comes from
a different physical register, which is the receive
register of the associated UART.
Note: For simplicity in the remaining UART dis-
cussions, the suffix “0” or “1” will be dropped when
referring to SFR registers and bits related to
UART0 or UART1, since each UART interface has
identical operation. Example, SBUF0 and SBUF1
will be referred to as just SBUF.
f
.
OSC
Mode 3. Mode 3 is the same as Mode 2 in all re-
spects except the baud rate is variable like it is in
Mode 1.
In all four modes, transmission is initiated by any
instruction that uses SBUF as a destination regis-
ter. Reception is initiated in Mode 0 by the condi-
tion RI = 0 and REN = 1. Reception is initiated in
the other modes by the incoming Start Bit if
REN = 1.
Each UART serial port can be full-duplex, meaning
it can transmit and receive simultaneously. Each
UART is also receive-buffered, meaning it can
commence reception of a second byte before a
previously received byte has been read from the
SBUF Register. However, if the first byte still has
not been read by the time reception of the second
byte is complete, one of the bytes will be lost.
UART Operation Modes
Each UART can operate in one of four modes, one
mode is synchronous, and the others are asyn-
chronous as shown in Table 44.
Table 44. UART Operating Modes
Bits of SFR,
Data
Bits
SCON
Mode Synchronization
Baud Clock
Start/Stop Bits See Figure
SM0
SM1
Figure
None
f
/12
0
1
2
3
Synchronous
Asynchronous
Asynchronous
Asynchronous
0
0
1
1
0
8
8
9
9
OSC
28., page 86
Figure
1 Start, 1 Stop
1
0
1
Timer 1 or Timer 2 Overflow
/32 or f /64
30., page 88
Figure
1 Start, 1 Stop
f
OSC
OSC
32., page 90
Figure
1 Start, 1 Stop
Timer 1 or Timer 2 Overflow
34., page 91
81/231
uPSD33xx
Multiprocessor Communications. Modes 2 and
3 have a special provision for multiprocessor com-
munications. In these modes, 9 data bits are re-
ceived. The 9th one goes into bit RB8, then comes
a stop bit. The port can be programmed such that
when the stop bit is received, the UART interrupt
will be activated only if bit RB8 = 1. This feature is
enabled by setting bit SM2 in SCON. A way to use
this feature in multi-processor systems is as fol-
lows: When the master processor wants to trans-
mit a block of data to one of several slaves, it first
sends out an address byte which identifies the tar-
get slave. An address byte differs from a data byte
in that the 9th bit is 1 in an address byte and 0 in a
data byte. With SM2 = 1, no slave will be interrupt-
ed by a data byte. An address byte, however, will
interrupt all slaves, so that each slave can exam-
ine the received byte and see if it is being ad-
dressed. The addressed slave will clear its SM2 bit
and prepare to receive the data bytes that will be
coming. The slaves that were not being addressed
leave their SM2 bits set and go on about their busi-
ness, ignoring the coming data bytes.
SM2 has no effect in Mode 0, and in Mode 1, SM2
can be used to check the validity of the stop bit. In
a Mode 1 reception, if SM2 = 1, the receive inter-
rupt will not be activated unless a valid stop bit is
received.
Serial Port Control Registers
The SFR SCON0 controls UART0, and SCON1
controls UART1, shown in Table 45 and Table 46.
These registers contain not only the mode selec-
tion bits, but also the 9th data bit for transmit and
receive (bits TB8 and RB8), and the UART Inter-
rupt flags, TI and RI.
Table 45. SCON0: Serial Port UART0 Control Register (SFR 98h, reset value 00h)
Bit 7
SM0
Bit 6
SM1
Bit 5
SM2
Bit 4
REN
Bit 3
TB8
Bit 2
RB8
Bit 1
TI
Bit 0
RI
Details
Bit
Symbol
R/W
Definition
7
SM0
R,W
Serial Mode Select, See Table 44., page 81. Important, notice bit order
of SM0 and SM1.
[SM0:SM1] = 00b, Mode 0
[SM0:SM1] = 01b, Mode 1
[SM0:SM1] = 10b, Mode 2
[SM0:SM1] = 11b, Mode 3
6
5
SM1
SM2
R,W
R,W
Serial Multiprocessor Communication Enable.
Mode 0: SM2 has no effect but should remain 0.
Mode 1: If SM2 = 0 then stop bit ignored. SM2 =1 then RI active if stop
bit = 1.
Mode 2 and 3: Multiprocessor Comm Enable. If SM2=0, 9th bit is
ignored. If SM2=1, RI active when 9th bit = 1.
Receive Enable.
4
3
REN
TB8
R,W
R,W
If REN=0, UART reception disabled. If REN=1, reception is enabled
TB8 is assigned to the 9th transmission bit in Mode 2 and 3. Not used in
Mode 0 and 1.
Mode 0: RB8 is not used.
Mode 1: If SM2 = 0, the RB8 is the level of the received stop bit.
Mode 2 and 3: RB8 is the 9th data bit that was received in Mode 2 and
3.
2
1
RB8
TI
R,W
R,W
Transmit Interrupt flag.
Causes interrupt at end of 8th bit time when transmitting in Mode 0, or at
beginning of stop bit transmission in other modes. Must clear flag with
firmware.
Receive Interrupt flag.
0
RI
R,W
Causes interrupt at end of 8th bit time when receiving in Mode 0, or
halfway through stop bit reception in other modes (see SM2 for
exception). Must clear this flag with firmware.
82/231
uPSD33xx
Table 46. SCON1: Serial Port UART1 Control Register (SFR D8h, reset value 00h)
Bit 7
SM0
Bit 6
SM1
Bit 5
SM2
Bit 4
REN
Bit 3
TB8
Bit 2
RB8
Bit 1
TI
Bit 0
RI
Details
Bit
Symbol
R/W
Definition
7
SM0
R,W
Serial Mode Select, See Table 44., page 81. Important, notice bit order
of SM0 and SM1.
[SM0:SM1] = 00b, Mode 0
[SM0:SM1] = 01b, Mode 1
[SM0:SM1] = 10b, Mode 2
[SM0:SM1] = 11b, Mode 3
6
5
SM1
SM2
R,W
R,W
Serial Multiprocessor Communication Enable.
Mode 0: SM2 has no effect but should remain 0.
Mode 1: If SM2 = 0 then stop bit ignored. SM2 =1 then RI active if stop
bit = 1.
Mode 2 and 3: Multiprocessor Comm Enable. If SM2=0, 9th bit is
ignored. If SM2=1, RI active when 9th bit = 1.
Receive Enable.
4
3
REN
TB8
R,W
R,W
If REN=0, UART reception disabled. If REN=1, reception is enabled
TB8 is assigned to the 9th transmission bit in Mode 2 and 3. Not used in
Mode 0 and 1.
Mode 0: RB8 is not used.
Mode 1: If SM2 = 0, the RB8 is the level of the received stop bit.
Mode 2 and 3: RB8 is the 9th data bit that was received in Mode 2 and
3.
2
1
RB8
TI
R,W
R,W
Transmit Interrupt flag.
Causes interrupt at end of 8th bit time when transmitting in Mode 0, or at
beginning of stop bit transmission in other modes. Must clear flag with
firmware.
Receive Interrupt flag.
0
RI
R,W
Causes interrupt at end of 8th bit time when receiving in Mode 0, or
halfway through stop bit reception in other modes (see SM2 for
exception). Must clear this flag with firmware.
83/231
uPSD33xx
UART Baud Rates
The baud rate in Mode 0 is fixed:
The Timer 1 Interrupt should be disabled in this
application. The Timer itself can be configured for
either “timer” or “counter” operation, and in any of
its 3 running modes. In the most typical applica-
tions, it is configured for “timer” operation, in the
Auto-reload Mode (high nibble of the SFR TMOD
= 0010B). In that case the baud rate is given by the
formula:
Mode 0 Baud Rate = f
/ 12
OSC
The baud rate in Mode 2 depends on the value of
the bit SMOD in the SFR named PCON. If SMOD
= 0 (default value), the baud rate is 1/64 the oscil-
lator frequency, f
. If SMOD = 1, the baud rate
OSC
is 1/32 the oscillator frequency.
SMOD
Mode 2 Baud Rate = (2
/ 64) x f
Mode 1,3 Baud Rate =
OSC
SMOD
(2
/ 32) x (f
/ (12 x [256 – (TH1)]))
OSC
Baud rates in Modes 1 and 3 are determined by
the Timer 1 or Timer 2 overflow rate.
Using Timer 1 to Generate Baud Rates. When
Timer 1 is used as the baud rate generator (bits
RCLK = 0, TCLK = 0), the baud rates in Modes 1
and 3 are determined by the Timer 1 overflow rate
and the value of SMOD as follows:
Table 47 lists various commonly used baud rates
and how they can be obtained from Timer 1.
Using Timer/Counter 2 to Generate Baud
Rates. See
Baud
Rate
Generator
Mode, page 77.
Mode 1,3 Baud Rate =
SMOD
(2
/ 32) x (Timer 1 overflow rate)
Table 47. Commonly Used Baud Rates Generated from Timer 1
Timer 1
SMOD
bit in
PCON
Desired
Resultant Baud Rate
Timer
Mode in
TMOD
TH1
Reload
f
MHz
UART Mode
OSC
C/T Bit
in TMOD
Baud Rate Baud Rate Deviation
value (hex)
Mode 0 Max
Mode 2 Max
Mode 2 Max
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
Modes 1 or 3
40.0
3.33MHz
1250 k
625 k
19200
9600
3.33MHz
1250 k
625 k
18939
9470
0
X
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
X
X
X
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
X
X
X
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
X
40.0
40.0
0
X
0
-1.36%
-1.36%
-2.34%
0.47%
0.47%
0.47%
0.47%
0.16%
0.16%
0
X
40.0
F5
EA
F6
FD
FA
F7
EE
F3
F3
FF
FE
FD
FA
FF
FE
FF
FE
40.0
36.0
19200
57600
28800
19200
9600
18570
57870
28934
19290
9645
33.333
33.333
33.333
33.333
24.0
9600
9615
12.0
4800
4808
11.0592
11.0592
11.0592
11.0592
3.6864
3.6864
1.8432
1.8432
57600
28800
19200
9600
57600
28800
19200
9600
0
0
0
19200
9600
19200
9600
0
0
9600
9600
0
4800
4800
0
84/231
uPSD33xx
More About UART Mode 0
Refer to the block diagram in Figure 28., page 86,
and timing diagram in Figure 29., page 86.
Control unit to do one last shift, then deactivate
SEND, and then set the interrupt flag TI. Both of
these actions occur at S1P1.
Transmission is initiated by any instruction which
writes to the SFR named SBUF. At the end of a
write operation to SBUF, a 1 is loaded into the 9th
position of the transmit shift register and tells the
TX Control unit to begin a transmission. Transmis-
sion begins on the following MCU machine cycle,
when the “SEND” signal is active in Figure 29.
SEND enables the output of the shift register to the
alternate function on the port containing pin RxD,
and also enables the SHIFT CLOCK signal to the
alternate function on the port containing the pin,
TxD. At the end of each SHIFT CLOCK in which
SEND is active, the contents of the transmit shift
register are shifted to the right one position.
As data bits shift out to the right, zeros come in
from the left. When the MSB of the data byte is at
the output position of the shift register, then the '1'
that was initially loaded into the 9th position, is just
to the left of the MSB, and all positions to the left
of that contain zeros. This condition flags the TX
Reception is initiated by the condition REN = 1 and
RI = 0. At the end of the next MCU machine cycle,
the RX Control unit writes the bits 11111110 to the
receive shift register, and in the next clock phase
activates RECEIVE. RECEIVE enables the SHIFT
CLOCK signal to the alternate function on the port
containing the pin, TxD. Each pulse of SHIFT
CLOCK moves the contents of the receive shift
register one position to the left while RECEIVE is
active. The value that comes in from the right is the
value that was sampled at the RxD pin. As data
bits come in from the right, 1s shift out to the left.
When the 0 that was initially loaded into the right-
most position arrives at the left-most position in the
shift register, it flags the RX Control unit to do one
last shift, and then it loads SBUF. After this, RE-
CEIVE is cleared, and the receive interrupt flag RI
is set.
85/231
uPSD33xx
Figure 28. UART Mode 0, Block Diagram
Internal Bus
SBUF
Write
to
SBUF
D
S
RxD
Pin
Q
CL
Zero Detector
Shift
Start
Tx Control
T
Send
f
/12
OSC
Tx Clock
Serial
Port
Interrupt
Shift
Clock
TxD
Pin
Receive
Shift
6 5 4 3 2 1 0
R
Rx Clock
Start
REN
R1
Rx Control
7
RxD
P3.0 Alt
Input
Function
Input Shift Register
Load
SBUF
Shift
SBUF
Read
SBUF
Internal Bus
AI06824
Figure 29. UART Mode 0, Timing Diagram
Write to SBUF
Send
Shift
Transmit
Receive
D0
D1
D2
D3
D4
D5
D6
D7
RxD (Data Out)
TxD (Shift Clock)
TI
Write to SCON
Clear RI
RI
Receive
Shift
RxD (Data In)
TxD (Shift Clock)
D0
D1
D2
D3
D4
D5
D6
D7
AI06825
86/231
uPSD33xx
More About UART Mode 1
Refer to the block diagram in Figure 30., page 88,
and timing diagram in Figure 31., page 88.
with the boundaries of the incoming bit times. The
16 states of the counter divide each bit time into
16ths. At the 7th, 8th, and 9th counter states of
each bit time, the bit detector samples the value of
RxD. The value accepted is the value that was
seen in at least 2 of the 3 samples. This is done for
noise rejection. If the value accepted during the
first bit time is not '0,' the receive circuits are reset
and the unit goes back to looking for another '1'-to-
'0' transition. This is to provide rejection of false
start bits. If the start bit proves valid, it is shifted
into the input shift register, and reception of the re-
set of the rest of the frame will proceed. As data
bits come in from the right, '1s' shift out to the left.
When the start bit arrives at the left-most position
in the shift register (which in mode 1 is a 9-bit reg-
ister), it flags the RX Control unit to do one last
shift, load SBUF and RB8, and set the receive in-
terrupt flag RI. The signal to load SBUF and RB8,
and to set RI, will be generated if, and only if, the
following conditions are met at the time the final
shift pulse is generated:
Transmission is initiated by any instruction which
writes to SBUF. At the end of a write operation to
SBUF, a '1' is loaded into the 9th position of the
transmit shift register and flags the TX Control unit
that a transmission is requested. Transmission ac-
tually starts at the end of the MCU the machine cy-
cle following the next rollover in the divide-by-16
counter. Thus, the bit times are synchronized to
the divide-by-16 counter, not to the writing of
SBUF. Transmission begins with activation of
SEND which puts the start bit at pin TxD. One bit
time later, DATA is activated, which enables the
output bit of the transmit shift register to pin TxD.
The first shift pulse occurs one bit time after that.
As data bits shift out to the right, zeros are clocked
in from the left. When the MSB of the data byte is
at the output position of the shift register, then the
1 that was initially loaded into the 9th position is
just to the left of the MSB, and all positions to the
left of that contain zeros. This condition flags the
TX Control unit to do one last shift and then deac-
tivates SEND, and sets the interrupt flag, TI. This
occurs at the 10th divide-by-16 rollover after a
write to SBUF.
1. RI = 0, and
2. Either SM2 = 0, or the received stop bit = 1.
If either of these two conditions are not met, the re-
ceived frame is irretrievably lost. If both conditions
are met, the stop bit goes into RB8, the 8 data bits
go into SBUF, and RI is activated. At this time,
whether the above conditions are met or not, the
unit goes back to looking for a '1'-to-'0' transition
on pin RxD.
Reception is initiated by a detected 1-to-0 transi-
tion at the pin RxD. For this purpose RxD is sam-
pled at a rate of 16 times whatever baud rate has
been established. When a transition is detected,
the divide-by-16 counter is immediately reset, and
1FFH is written into the input shift register. Reset-
ting the divide-by-16 counter aligns its rollovers
87/231
uPSD33xx
Figure 30. UART Mode 1, Block Diagram
Timer1
Timer2
Overflow
Internal Bus
SBUF
Overflow
TB8
S
Write
to
SBUF
TxD
Pin
D
Q
÷2
CL
0
1
Zero Detector
SMOD
0
0
1
Shift
Data
Start
TCLK
Tx Control
TI
Send
÷16
Tx Clock
Serial
1
Port
Interrupt
RCLK
÷16
Sample
1-to-0
Transition
Detector
Load SBUF
Shift
RI
Rx Clock
Start
Rx Control
1FFh
Rx Detector
Input Shift Register
Load
SBUF
RxD
Pin
Shift
SBUF
Read
SBUF
Internal Bus
AI06826
Figure 31. UART Mode 1, Timing Diagram
Tx Clock
Write to SBUF
Send
Data
Transmit
Shift
Start Bit
D0
D0
D1
D1
D2
D2
D3
D4
D4
D5
D5
D6
D6
D7
D7
TxD
TI
Stop Bit
Stop Bit
Rx Clock
Start Bit
D3
RxD
Receive
Bit Detector
Sample Times
Shift
RI
AI06843
88/231
uPSD33xx
More About UART Modes 2 and 3
For Mode 2, refer to the block diagram in Figure
32., page 90, and timing diagram in Figure
33., page 90. For Mode 3, refer to the block dia-
gram in Figure 34., page 91, and timing diagram in
Figure 35., page 91.
Reception is initiated by a detected 1-to-0 transi-
tion at pin RxD. For this purpose RxD is sampled
at a rate of 16 times whatever baud rate has been
established. When a transition is detected, the di-
vide-by-16 counter is immediately reset, and 1FFH
is written to the input shift register. At the 7th, 8th,
and 9th counter states of each bit time, the bit de-
tector samples the value of RxD. The value ac-
cepted is the value that was seen in at least 2 of
the 3 samples. If the value accepted during the
first bit time is not '0,' the receive circuits are reset
and the unit goes back to looking for another '1'-to-
'0' transition. If the start bit proves valid, it is shifted
into the input shift register, and reception of the
rest of the frame will proceed. As data bits come in
from the right, '1s' shift out to the left. When the
start bit arrives at the left-most position in the shift
register (which in Modes 2 and 3 is a 9-bit regis-
ter), it flags the RX Control unit to do one last shift,
load SBUF and RB8, and set the interrupt flag RI.
The signal to load SBUF and RB8, and to set RI,
will be generated if, and only if, the following con-
ditions are met at the time the final shift pulse is
generated:
Keep in mind that the baud rate is programmable
to either 1/32 or 1/64 of f
in Mode 2, but Mode
OSC
3 uses a variable baud rate generated from Timer
1 or Timer 2 rollovers.
The receive portion is exactly the same as in Mode
1. The transmit portion differs from Mode 1 only in
the 9th bit of the transmit shift register.
Transmission is initiated by any instruction which
writes to SBUF. At the end of a write operation to
SBUF, the TB8 Bit is loaded into the 9th position of
the transmit shift register and flags the TX Control
unit that a transmission is requested. Transmis-
sion actually starts at the end of the MCU the ma-
chine cycle following the next rollover in the divide-
by-16 counter. Thus, the bit times are synchro-
nized to the divide-by-16 counter, not to the writing
of SBUF. Transmission begins with activation of
SEND which puts the start bit at pin TxD. One bit
time later, DATA is activated, which enables the
output bit of the transmit shift register to pin TxD.
The first shift pulse occurs one bit time after that.
The first shift clocks a '1' (the stop bit) into the 9th
bit position of the shift register. There-after, only
zeros are clocked in. Thus, as data bits shift out to
the right, zeros are clocked in from the left. When
bit TB8 is at the output position of the shift register,
then the stop bit is just to the left of TB8, and all po-
sitions to the left of that contain zeros. This condi-
tion flags the TX Control unit to do one last shift
and then deactivate SEND, and set the interrupt
flag, TI. This occurs at the 11th divide-by 16 roll-
over after writing to SBUF.
1. RI = 0, and
2. Either SM2 = 0, or the received 9th data bit = 1.
If either of these conditions is not met, the received
frame is irretrievably lost, and RI is not set. If both
conditions are met, the received 9th data bit goes
into RB8, and the first 8 data bits go into SBUF.
One bit time later, whether the above conditions
were met or not, the unit goes back to looking for
a '1'-to-'0' transition on pin RxD.
89/231
uPSD33xx
Figure 32. UART Mode 2, Block Diagram
Internal Bus
SBUF
f
/32
OSC
TB8
S
Write
to
SBUF
TxD
Pin
D
Q
÷2
CL
0
1
Zero Detector
SMOD
Shift
Data
Start
Tx Control
TI
Send
÷16
Tx Clock
Serial
Port
Interrupt
÷16
Sample
1-to-0
Transition
Detector
Load SBUF
Shift
RI
Rx Clock
Start
Rx Control
1FFh
Rx Detector
Input Shift Register
Load
SBUF
RxD
Pin
Shift
SBUF
Read
SBUF
Internal Bus
AI06844
Figure 33. UART Mode 2, Timing Diagram
Tx Clock
Write to SBUF
Send
Data
Transmit
Shift
Start Bit
D0
D1
D2
D3
D4
D5
D6
D7
TB8
TxD
TI
Stop Bit
Stop Bit
Generator
Rx Clock
Start Bit
D0
D1
D2
D3
D4
D5
D6
D7
RB8
RxD
Stop Bit
Receive
Bit Detector
Sample Times
Shift
RI
AI06845
90/231
uPSD33xx
Figure 34. UART Mode 3, Block Diagram
Timer1
Timer2
Overflow
Internal Bus
SBUF
Overflow
TB8
S
Write
to
SBUF
TxD
Pin
D
Q
÷2
CL
0
1
Zero Detector
SMOD
0
0
1
Shift
Data
Start
TCLK
Tx Control
TI
Send
÷16
Tx Clock
Serial
1
Port
Interrupt
RCLK
÷16
Sample
1-to-0
Transition
Detector
Load SBUF
Shift
RI
Rx Clock
Start
Rx Control
1FFh
Rx Detector
Input Shift Register
Load
SBUF
RxD
Pin
Shift
SBUF
Read
SBUF
Internal Bus
AI06846
Figure 35. UART Mode 3, Timing Diagram
Tx Clock
Write to SBUF
Send
Data
Transmit
Shift
Start Bit
D0
D1
D2
D3
D4
D5
D6
D7
TB8
TxD
TI
Stop Bit
Stop Bit
Generator
Rx Clock
Start Bit
D0
D1
D2
D3
D4
D5
D6
D7
RB8
RxD
Stop Bit
Receive
Bit Detector
Sample Times
Shift
RI
AI06847
91/231
uPSD33xx
IrDA INTERFACE
uPSD33xx devices provide an internal IrDA inter-
face that will allow the connection of the UART1
serial interface directly to an external infrared
transceiver device. The IrDA interface does this by
automatically shortening the pulses transmitted on
UART1’s TxD1 pin, and stretching the incoming
pulses received on the RxD1 pin. Reference Fig-
ures 36 and 37.
compliant with the IrDA Physical Layer Link Spec-
ification v1.4 (www.irda.org) operating from 1.2k
bps up to 115.2k bps. The pulses received on the
RxD1 pin are stretched by the IrDA interface to be
recognized by UART1’s receiver logic, also adher-
ing to the IrDA specification up to 115.2k bps.
Note: In Figure 37 a logic '0' in the serial data
stream of a UART Frame corresponds to a logic
high pulse in an IR Frame. A logic '1' in a UART
Frame corresponds to no pulse in an IR Frame.
When the IrDA interface is enabled, the output sig-
nal from UART1’s transmitter logic on pin TxD1 is
Figure 36. IrDA Interface
TxD1-IrDA
SIRClk
IrDA
IrDA
Interf ace
UART1
Transceiver
RxD1-IrDA
TxD
RxD
uPSD33XX
AI07851
Figure 37. Pulse Shaping by the IrDA Interface
UART Frame
Data Bits
Start
Bit
Stop
Bit
0
1
0
1
0
0
1
1
0
1
UART Frame
IR Frame
Start
Bit
Stop
Bit
Data Bits
0
1
0
1
0
0
1
1
0
1
IR Frame
Bit Time
Pulse Width = 3/16 Bit Time
AI09624
92/231
uPSD33xx
The UART1 serial channel can operate in one of
four different modes as shown in Table
44., page 81 in the section, SERIAL UART
INTERFACES, page 81. However, when UART1
is used for IrDA communication, UART1 must op-
erate in Mode 1 only, to be compatible with IrDA
protocol up to 115.2k bps. The IrDA interface will
support baud rates generated from Timer 1 or Tim-
er 2, just like standard UART serial communica-
tion, but with one restriction. The transmit baud
rate and receive baud rate must be the same (can-
not be different rates as is allowed by standard
UART communications).
The IrDA Interface is disabled after a reset and is
enabled by setting the IRDAEN Bit in the SFR
named IRDACON (Table 48., page 93). When
IrDA is disabled, the UART1's RxD and TxD sig-
nals will bypass the internal IrDA logic and instead
they are routed directly to the pins RxD1 and TxD1
respectively. When IrDA is enabled, the IrDA pulse
shaping logic is active and resides between
UART1 and the pins RxD1 and TxD1 as shown in
Figure 36., page 92.
Table 48. IRDACON Register Bit Definition (SFR CEh, Reset Value 0Fh)
Bit 7
–
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
IRDAEN
PULSE
CDIV4
CDIV3
CDIV2
CDIV1
CDIV0
Details
Bit
Symbol
R/W
Definition
7
–
–
Reserved
IrDA Enable
6
IRDAEN
RW
0 = IrDA Interface is disabled
1 = IrDA is enabled, UART1 outputs are disconnected from Port 1 (or
Port 4)
IrDA Pulse Modulation Select
5
PULSE
RW
RW
0 = 1.627µs
1 = 3/16 bit time pulses
4-0
CDIV[4:0]
Specify Clock Divider (see Table 49., page 94)
93/231
uPSD33xx
Pulse Width Selection
The IrDA interface has two ways to modulate the
standard UART1 serial stream:
the fastest baud rate (8.68µs bit time for 115.2k
bps rate), multiplied by the proportion, 3/16.
1. An IrDA data pulse will have a constant pulse
width for any bit time, regardless of the
selected baud rate.
2. An IrDA data pulse will have a pulse width that
is proportional to the the bit time of the
selected baud rate. In this case, an IrDA data
pulse width is 3/16 of its bit time, as shown in
Figure 37., page 92.
To produce this fixed data pulse width when the
PULSE bit = 0, a prescaler is needed to generate
an internal reference clock, SIRClk, shown in Fig-
ure 36., page 92. SIRClk is derived by dividing the
oscillator clock frequency, f
using the five bits
OSC,
CDIV[4:0] in the SFR named IRDACON. A divisor
must be chosen to produce a frequency for SIRClk
that lies between 1.34 MHz and 2.13 MHz, but it is
best to choose a divisor value that produces SIR-
Clk frequency as close to 1.83MHz as possible,
because SIRClk at 1.83MHz will produce an fixed
IrDA data pulse width of 1.63µs. Table 49 provides
recommended values for CDIV[4:0] based on sev-
The PULSE bit in the SFR named IRDACON de-
termines which method above will be used.
According to the IrDA physical layer specification,
for all baud rates at 115.2k bps and below, the
minimum data pulse width is 1.41µs. For a baud
rate of 115.2k bps, the maximum pulse width
2.23µs. If a constant pulse width is to be used for
all baud rates (PULSE bit = 0), the ideal general
pulse width is 1.63µs, derived from the bit time of
eral different values of f
.
OSC
For reference, SIRClk of 2.13MHz will generate a
fixed IrDA data pulse width of 1.41µs, and SIRClk
of 1.34MHz will generate a fixed data pulse width
of 2.23µs.
Table 49. Recommended CDIV[4:0] Values to Generate SIRClk (default CDIV[4:0] = 0Fh, 15 decimal)
f
(MHz)
Resulting f
(MHz)
SIRCLK
Value in CDIV[4:0]
16h, 22 decimal
14h, 20 decimal
0Dh, 13 decimal
06h, 6 decimal
OSC
40.00
1.82
36.864, or 36.00
24.00
1.84, or 1.80
1.84
11.059, or 12.00
1.84, or 2.00
(1)
04h, 4 decimal
1.84
7.3728
Note: 1. When PULSE bit = 0 (fixed data pulse width), this is minimum recommended f
because CDIV[4:0] must be 4 or greater.
OSC
94/231
uPSD33xx
I2C INTERFACE
2
uPSD33xx devices support one serial I C inter-
face. This is a two-wire communication channel,
having a bi-directional data signal (SDA, pin P3.6)
and a clock signal (SCL, pin P3.7) based on open-
drain line drivers, requiring external pull-up resis-
the role of Master or Slave, or a single device can
be a Slave only. Each Slave device on the bus has
a unique address, and a general broadcast ad-
dress is also available. A Master or Slave device
has the ability to suspend data transfers if the de-
vice needs more time to transmit or receive data.
tors, R , each with a typical value of 4.7kΩ (see
P
2
Figure 38).
This I C interface has the following features:
2
I C Interface Main Features
–
Serial I/O Engine (SIOE): serial/parallel
conversion; bus arbitration; clock generation
and synchronization; and handshaking are all
performed in hardware
Byte-wide data is transferred, MSB first, between
a Master device and a Slave device on two wires.
More than one bus Master is allowed, but only one
Master may control the bus at any given time. Data
is not lost when another Master requests the use
–
–
–
–
Interrupt or Polled operation
Multi-master capability
7-bit Addressing
2
of a busy bus because I C supports collision de-
tection and arbitration. The bus Master initiates all
data movement and generates the clock that per-
mits the transfer. Once a transfer is initiated by the
Master, any device addressed is considered a
Slave. Automatic clock synchronization allows I C
devices with different bit rates to communicate on
the same physical bus. A single device can play
2
Supports standard speed I C (SCL up to
2
100kHz), fast mode I C (101KHz to 400kHz),
2
and high-speed mode I C (401KHz to
2
833kHz)
2
Figure 38. Typical I C Bus Configuration
(1)
V
or V
DD
CC
2
Device with I C
Interface
R
P
R
P
SDA
SCL
2
I C BUS
SDA/P3.6 SCL/P3.7
2
2
Device with I C
Interface
Device with I C
Interface
uPSD33XX(V)
AI09623
Note: 1. For 3.3V system, connect R to 3.3V V . For 5.0V system, connect R to 5.0V V .
DD
P
CC
P
95/231
uPSD33xx
Communication Flow
2
2
I C data flow control is based on the fact that all
START conditon and begin the next transfer.
There is no limit to the number of bytes that
can be transmitted during a transfer session.
I C compatible devices will drive the bus lines with
open-drain (or open-collector) line drivers pulled
up with external resistors, creating a wired-AND
situation. This means that either bus line (SDA or
SCL) will be at a logic '1' level only when no I C de-
vice is actively driving the line to logic '0.' The logic
for handshaking, arbitration, synchronization, and
2. Data transfer from Slave Transmitter to
Master Receiver (R/W = 1). In this case, the
Master generates a START condition on the
bus and it generates a clock signal on the SCL
line. Then the Master transmits the first byte
on the SDA line containing the 7-bit Slave
address plus the R/W bit. The Slave who owns
that address will respond with an acknowledge
bit on SDA, and all other Slave devices will not
respond. Next, the addressed Slave will
transmit a data byte (or bytes) to the Master.
The Master will return an acknowledge bit
after each data byte it successfully receives,
unless it is the last byte the Master desires. If
so, the Master will not acknowledge the last
byte and from this, the Slave knows to stop
transmitting data bytes to the Master. The
Master will then generate a STOP condition on
the bus, or it will generate a RE-START
conditon and begin the next transfer. There is
no limit to the number of bytes that can be
transmitted during a transfer session.
2
2
collision detection is implemented by each I C de-
vice having:
1. The ability to hold a line low against the will of
the other devices who are trying to assert the
line high.
2. The ability of a device to detect that another
device is driving the line low against its will.
Assert high means the driver releases the line and
external pull-ups passively raise the signal to logic
'1.' Holding low means the open-drain driver is
actively pulling the signal to ground for a logic '0.'
For example, if a Slave device cannot transmit or
receive a byte because it is distracted by and inter-
rupt or it has to wait for some process to complete,
it can hold the SCL clock line low. Even though the
Master device is generating the SCL clock, the
Master will sense that the Slave is holding the SCL
line low against the will of the Master, indicating
that the Master must wait until the Slave releases
SCL before proceeding with the transfer.
Another example is when two Master devices try
to put information on the bus simultaneously, the
first one to release the SDA data line looses arbi-
tration while the winner continues to hold SDA low.
Two types of data transfers are possible with I C
depending on the R/W bit, see Figure
39., page 97.
A few things to know related to these transfers:
–
–
–
–
Either the Master or Slave device can hold the
SCL clock line low to indicate it needs more
time to handle a byte transfer. An indefinite
holding period is possible.
A START condition is generated by a Master
and recognized by a Slave when SDA has a 1-
to-0 transition while SCL is high (Figure
39., page 97).
A STOP condition is generated by a Master
and recognized by a Slave when SDA has a 0-
to1 transition while SCL is high (Figure
39., page 97).
2
1. Data transfer from Master Transmitter to
Slave Receiver (R/W = 0). In this case, the
Master generates a START condition on the
bus and it generates a clock signal on the SCL
line. Then the Master transmits the first byte
on the SDA line containing the 7-bit Slave
address plus the R/W bit. The Slave who owns
that address will respond with an acknowledge
bit on SDA, and all other Slave devices will not
respond. Next, the Master will transmit a data
byte (or bytes) that the addressed Slave must
receive. The Slave will return an acknowledge
bit after each data byte it successfully
A RE-START (repeated START) condition
generated by a Master can have the same
function as a STOP condition when starting
another data transfer immediately following
the previous data transfer (Figure
39., page 97).
When transferring data, the logic level on the
SDA line must remain stable while SCL is
high, and SDA can change only while SCL is
low. However, when not transferring data,
SDA may change state while SCL is high,
which creates the START and STOP bus
conditions.
–
receives. After the final byte is transmitted by
the Master, the Master will generate a STOP
condition on the bus, or it will generate a RE-
96/231
uPSD33xx
–
An Acknowlegde bit is generated from a
Master or a Slave by driving SDA low during
the “ninth” bit time, just following each 8-bit
byte that is transfered on the bus (Figure
39., page 97). A Non-Acknowledge occurs
when SDA is asserted high during the ninth bit
time. All byte transfers on the I C bus include
a 9th bit time reserved for an Acknowlege
(ACK) or Non-Acknowledge (NACK).
–
–
An additional Master device that desires to
control the bus should wait until the bus is not
busy before generating a START condition so
that a possible Slave operation is not
interrupted.
If two Master devices both try to generate a
START condition simultaneously, the Master
who looses arbitration will switch immediately
to Slave mode so it can recoginize it’s own
Slave address should it appear on the bus.
2
2
Figure 39. Data Transfer on an I C Bus
READ/WRITE
Indicator
Acknowledge
bits from
7-bit Slave
Address
receiver
NACK
Stop
Condition
ACK
MSB
R/W
MSB
Repeated
Start
Condition
ACK
9
1
2
3-6
7
8
9
1
2
3-8
Repeated if more
data bytes are
transferred.
Start
Condition
Clock can be held low
to stall transfer.
AI09625
97/231
uPSD33xx
Operating Modes
2
The I C interface supports four operating modes:
with the longest low period on SCL, will force
Master_Y to wait until Master_X finishes its low
period before Master_Y proceeds to assert its high
period on SCL. At this point, both Masters begin
asserting their high period on SCL simultaneously,
and the Master with the shortest high period will be
the first to drive SCL for the next low period. In this
scheme, the Master with the longest low SCL pe-
riod paces low times, and the Master with the
shortest high SCL period paces the high times,
making synchronized arbitration possible.
■
■
■
■
Master-Transmitter
Master-Receiver
Slave-Transmitter
Slave-Receiver
The interface may operate as either a Master or a
Slave within a given application, controlled by firm-
ware writing to SFRs.
By default after a reset, the I C interface is in Mas-
ter Receiver mode, and the SDA/P3.6 and SCL/
P3.7 pins default to GPIO input mode, high imped-
ance, so there is no I C bus interference. Before
using the I C interface, it must be initialized by
2
Clock Sync During Handshaking. This allows
receivers in different devices to handle various
transfer rates, either at the byte-level, or bit-level.
2
2
At the byte-level, a device may pause the transfer
between bytes by holding SCL low to have time to
store the latest received byte or fetch the next byte
to transmit.
At the bit-level, a Slave device may extend the low
period of SCL by holding it low. Thus the speed of
any Master device will adapt to the internal opera-
tion of the Slave.
firmware, and the pins must be configured. This is
discussed in I C Operating Sequences, page 108.
2
Bus Arbitration
A Master device always samples the I C bus to
2
ensure a bus line is high whenever that Master is
asserting a logic 1. If the line is low at that time, the
Master recognizes another device is overriding it’s
own transmission.
2
General Call Address
A Master may start a transfer only if the I C bus is
A General Call (GC) occurs when a Master-Trans-
mitter initiates a transfer containing a Slave ad-
dress of 0000000b, and the R/W bit is logic 0. All
Slave devices capable of responding to this broad-
cast message will acknowledge the GC simulta-
neously and then behave as a Slave-Receiver.
The next byte transmitted by the Master will be ac-
cepted and acknowledged by all Slaves capable of
handling the special data bytes. A Slave that can-
not handle one of these data bytes must ignore it
not busy. However, it’s possible that two or more
Masters may generate a START condition simulta-
neously. In this case, arbitration takes place on the
SDA line each time SCL is high. The Master that
first senses that its bus sample does not corre-
spond to what it is driving (SDA line is low while it’s
asserting a high) will immediately change from
Master-Transmitter to Slave-Receiver mode. The
arbitration process can carry on for many bit times
if both Masters are addressing the same Slave de-
vice, and will continue into the data bits if both
Masters are trying to be Master-Transmitter. It is
also possible for arbitration to carry on into the ac-
knowledge bits if both Masters are trying to be
Master-Receiver. Because address and data in-
formation on the bus is determined by the winning
Master, no information is lost during the arbitration
process.
2
by not acknowledging it. The I C specification lists
the possible meanings of the special bytes that fol-
low the first GC address byte, and the actions to
be taken by the Slave device(s) upon receiving
them. A common use of the GC by a Master is to
dynamically assign device addresses to Slave de-
vices on the bus capable of a programmable de-
vice address.
The uPSD33xx can generate a GC as a Master-
Transmitter, and it can receive a GC as a Slave.
When receiving a GC address (00h), an interrupt
will be generated so firmware may respond to the
special GC data bytes if desired.
Clock Synchronization
Clock synchronization is used to synchronize arbi-
trating Masters, or used as a handshake by a de-
vices to slow down the data transfer.
Clock Sync During Arbitration. During bus ar-
bitration between competing Masters, Master_X,
98/231
uPSD33xx
Serial I/O Engine (SIOE)
2
At the heart of the I C interface is the hardware
■
■
■
■
S1STA - Interface Status (Table
52., page 103)
S1DAT - Data Shift Register (Table
53., page 104)
S1ADR - Device Address (Table
54., page 104)
S1SETUP - Sampling Rate (Table
55., page 105)
SIOE, shown in Figure 40. The SIOE automatically
handles low-level I C bus protocol (data shifting,
handshaking, arbitration, clock generation and
synchronization) and it is controlled and monitored
by five SFRs.
2
The five SFRs shown in Figure 40 are:
■
S1CON - Interface Control (Table
50., page 100)
2
Figure 40. I C Interface SIOE Block Diagram
INTR to 8032
8
8
8
S1STA - Interface Status
S1CON - Interface Control
S1SETUP - Sample Rate
SCL / P3.7
Control (START Condition)
Open-
Drain
Output
Arbitration
and Sync
Input
Periph
Clock
Timing and
Control
(f
)
OSC
Clock
Generation
SDA / P3.6
Open-
Drain
Output
Serial DATA IN
Shift Direction
Input
8
ACK
Bit
Serial DATA OUT
S1DAT - Shift Register
b7
b0
7
Comparator
7
8
b7
b0
S1ADR - Device Address
AI09626
99/231
uPSD33xx
2
I C Interface Control Register (S1CON)
Table 50. Serial Control Register S1CON (SFR DCh, Reset Value 00h)
Bit 7
CR2
Bit 6
Bit 5
STA
Bit 4
STO
Bit 3
Bit 2
AA
Bit 1
Bit 0
ENI1
ADDR
CR[1:0]
Details
Bit
Symbol
R/W
Function
This bit, along with bits CR1 and CR0, determine the SCL clock
frequency (f ) when SIOE is in Master mode. These bits create a clock
7
CR2
R,W
SCL
divisor for f
. See Table 51.
OSC
2
I C Interface Enable
6
5
ENI1
STA
R,W
R,W
0 = SIOE disabled, 1 = SIOE enabled. When disabled, both SDA and
SCL signals are in high impedance state.
START flag.
When set, Master mode is entered and SIOE generates a START
2
condition only if the I C bus is not busy. When a START condition is
detected on the bus, the STA flag is cleared by hardware. When the STA
bit is set during an interrupt service, the START condition will be
generated after the interrupt service.
STOP flag
When STO is set in Master mode, the SIOE generates a STOP condition.
When a STOP condition is detected, the STO flag is cleared by
hardware. When the STO bit is set during an interrupt service, the STOP
condition will be generated after the interrupt service.
4
3
STO
R,W
R,W
This bit is set when an address byte received in Slave mode matches the
device address programmed into the S1ADR register. The ADDR bit
must be cleared with firmware.
ADDR
Assert Acknowledge enable
If AA = 1, an acknowledge signal (low on SDA) is automatically returned
during the acknowledge bit-time on the SCL line when any of the
following three events occur:
1. SIOE in Slave mode receives an address that matches contents of
S1ADR register
2
AA
R,W
R,W
2. A data byte has been received while SIOE is in Master Receiver
mode
3. A data byte has been received while SIOE is a selected Slave
Receiver
When AA = 0, no acknowledge is returned (high on SDA during acknowl-
edge bit-time).
These bits, along with bit CR2, determine the SCL clock frequency (f
)
SCL
1, 0
CR1, CR0
when SIOE is in Master mode. These bits create a clock divisor for f
See Table 51 for values.
.
OSC
100/231
uPSD33xx
Table 51. Selection of the SCL Frequency in Master Mode based on f
Examples
OSC
Bit Rate (kHz) @ f
OSC
f
OSC
CR2
CR1
CR0
Divided by:
12MHz f
24MHz f
36MHz f
40MHz f
OSC
OSC
OSC
OSC
(1)
(1)
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
32
48
375
250
200
100
50
750
500
400
200
100
50
X
X
750
600
300
150
75
833
666
333
166
83
60
120
240
480
960
1920
25
12.5
6.25
25
37.5
18.75
41
12.5
20
Note: 1. These values are beyond the bit rate supported by uPSD33xx.
101/231
uPSD33xx
2
I C Interface Status Register (S1STA)
The S1STA register provides status regarding im-
mediate activity and the current state of operation
on the I C bus. All bits in this register are read-only
–
When a complete data byte has been received
or transmitted by the SIOE while in Master
mode. The interrupt will occur even if the
Master looses arbitration.
When a complete data byte has been received
or transmitted by the SIOE while in selected
Slave mode.
A STOP condition on the bus has been
recognized by the SIOE while in selected
Slave mode.
2
except bit 5, INTR, which is the interrupt flag.
2
Interrupt Conditions. If the I C interrupt is en-
–
–
2
abled (EI C = 1 in SFR named IEA, and EA =1 in
SFR named IE), and the SIOE is initialized, then
an interrupt is automatically generated when any
one of the following five events occur:
–
–
When the SIOE receives an address that
matches the contents of the SFR, S1ADR.
Requirements: SIOE is in Slave Mode, and bit
AA = 1 in the SFR S1CON.
When the SIOE receives General Call
address. Requirments: SIOE is in Slave Mode,
bit AA = 1 in the SFR S1CON
Selected Slave mode means the device address
sent by the Master device at the beginning of the
current data transfer matched the address stored
in the S1ADR register.
If the I C interrupt is not enabled, the MCU may
poll the INTR flag in S1STA.
2
102/231
uPSD33xx
2
Table 52. S1STA: I C Interface Status Register (SFR DDh, reset value 00h)
Bit 7
GC
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
SLV
STOP
INTR
TX_MODE
BBUSY
BLOST
ACK_RESP
Details
Bit
Symbol
R/W
Function
General Call flag
GC = 1 if the General Call address of 00h was received when SIOE is in
Slave mode, and GC is cleared by a START or STOP condition on the
bus. If the SIOE is in Master mode when GC = 1, the Bus Lost condition
exists, and BLOST = 1.
7
GC
R
STOP flag
6
5
4
STOP
INTR
R
R,W
R
STOP = 1 while SIOE detects a STOP condition on the bus when in
Master or Slave mode.
Interrupt flag
2
INTR is set to 1 by any of the five I C interrupt conditions listed above.
INTR must be cleared by firmware.
Transmission Mode flag
TX_MODE
TX_MODE = 1 whenever the SIOE is in Master-Transmitter or Slave-
Transmitter mode. TX_MODE = 0 when SIOE is in any receiver mode.
Bus Busy flag
2
3
2
BBUSY
BLOST
R
R
BBUSY = 1 when the I C bus is in use. BBUSY is set by the SIOE when
a START condition exists on the bus and BBUSY is cleared by a STOP
condition.
Bus Lost flag
BLOST is set when the SIOE is in Master mode and it looses the
arbitration process to another Master device on the bus.
Not Acknowledge Response flag
While SIOE is in Transmitter mode:
2
–
After SIOE sends a byte, ACK_RESP = 1 whenever the external I C
device receives the byte, but that device does NOT assert an
ackowledge signal (external device asserted a high on SDA during
the acknowledge bit-time).
2
–
After SIOE sends a byte, ACK_RESP = 0 whenever the external I C
1
ACK_RESP
R
device receives the byte, and that device DOES assert an
ackowledge signal (external device drove a low on SDA during the
acknowledge bit-time)
Note: If SIOE is in Master-Transmitter mode, and ACK_RESP = 1 due to
a Slave-Transmitter not sending an Acknowledge, a STOP condition will
not automatically be generated by the SIOE. The STOP condition must
be generated with S1CON.STO = 1.
Slave Mode flag
0
SLV
R
SLV = 1 when the SIOE is in Slave mode. SLV = 0 when the SIOE is in
Master mode (default).
103/231
uPSD33xx
2
I C Data Shift Register (S1DAT)
The S1ADR register (Table 53) holds a byte of se-
rial data to be transmitted or it holds a serial byte
that has just been received. The MCU may access
S1DAT while the SIOE is not in the process of
shifting a byte (the INTR flag indicates shifting is
complete).
is set and automatically a wait condition is im-
posed on the I C bus (SCL held low by SIOE). In
Transmit mode, this wait condition is released as
soon as the MCU writes any byte to S1DAT. In Re-
ceive mode, the wait condition is released as soon
as the MCU reads the S1DAT register.
2
While transmitting, bytes are shifted out MSB first,
and when receiving, bytes are shifted in MSB first,
through the Acknowledge Bit register as shown in
Figure 40., page 99.
Bus Wait Condition. After the SIOE finishes re-
ceiving a byte in Receive mode, or transmitting a
byte in Transmit mode, the INTR flag (in S1STA)
This method allows the user to handle transmit
and receive operations within an interrupt service
routine. The SIOE will automatically stall the I C
bus at the appropriate time, giving the MCU time
to get the next byte ready to transmit or time to
read the byte that was just received.
2
2
Table 53. S1DAT: I C Data Shift register (SFR DEh, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
S1DAT[7:0]
Details
Bit
Symbol
R/W
Function
Holds the data byte to be transmitted in Transmit mode, or it holds the
data byte received in Receiver mode.
7:0
S1DAT[7:0]
R/W
2
I C Address Register (S1ADR)
The S1ADR register (Table 54) holds the 7-bit de-
vice address used when the SIOE is operating as
a Slave. When the SIOE receives an address from
a Master, it will compare this address to the con-
tents of S1ADR, as shown in Figure 40., page 99.
If the 7 bits match, the INTR Interrupt flag (in
S1STA) is set, and the ADDR Bit (in S1CON) is
set. The SIOE cannot modify the contents S1ADR,
and S1ADR is not used during Master mode.
2
Table 54. S1ADR: I C Address register (SFR DFh, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
–
SLA6
SLA5
SLA4
SLA3
SLA2
SLA1
SLA0
Details
Bit
7:1
0
Symbol
SLA[6:0]
–
R/W
R/W
–
Function
Stores desired 7-bit device address, used when SIOE is in Slave mode.
Not used
104/231
uPSD33xx
2
I C START Sample Setting (S1SETUP)
The S1SETUP register (Table 55) determines how
many times an I C bus START condition will be
sampled before the SIOE validates the START
condition, giving the SIOE the ability to reject noise
or illegal transmissions.
sample is taken 1/f
seconds after the initial 1-
OSC
2
to-0 transition was detected. However, more sam-
ples should be taken to ensure there is a valid
START condition.
To take more samples, the SIOE should be initial-
ized such that the EN_SS Bit is set, and a value is
written to the SMPL_SET[6:0] field of the
S1SETUP Register to specify how many samples
to take. The goal is to take a good number of sam-
ples during the minimum START condition hold
Because the minimum duration of an START con-
2
dition varies with I C bus speed (f
), and also
SCL
because the uPSD33xx may be operated with a
wide variety of frequencies (f ), it is necessary
OSC
to scale the number of samples per START condi-
tion based on f and f
.
time, t
, but no so many samples that the
HLDSTA
OSC
SCL
bus will be sampled after t
expires.
HLDSTA
In Slave mode, the SIOE recognizes the beginning
of a START condition when it detects a '1'-to-'0'
transition on the SDA bus line while the SCL line is
high (see Figure 39., page 97). The SIOE must
then validate the START condition by sampling the
bus lines to ensure SDA remains low and SCL re-
mains high for a minimum amount of hold time,
Table 56., page 106 describes the relationship be-
tween the contents of S1SETUP and the resulting
2
number of I C bus samples that SIOE will take af-
ter detecting the 1-to-0 transition on SDA of a
START condition.
Important: Keep in mind that the time between
t
. Once validated, the SIOE begins receiv-
HLDSTA
samples is always 1/f
.
OSC
ing the address byte that follows the START con-
dition.
If the EN_SS Bit (in the S1SETUP Register) is not
set, then the SIOE will sample only once after de-
tecting the '1'-to-'0' transition on SDA. This single
The minimum START condition hold time, t
HLDS-
2
, is different for the three common I C speed
TA
categories per Table 57., page 106.
2
Table 55. S1SETUP: I C START Condition Sample Setup register (SFR DBh, reset value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
EN_SS
SMPL_SET[6:0]
Details
Bit
Symbol
R/W
Function
Enable Sample Setup
(1)
EN_SS = 1 will force the SIOE to sample a START condition on the bus
the number of times specified in SMPL_SET[6:0].
7
EN_SS
R/W
(1)
EN_SS = 0 means the SIOE will sample a START condition only one
time, regardless of the contents of SMPL_SET[6:0].
Sample Setting
SMPL_SET
[6:0]
6:0
–
(1)
Specifies the number of bus samples taken during a START condition.
See Table 56 for values.
Note: 1. Sampling SCL and SDA lines begins after '1'-to-'0' transition on SDA occurred while SCL is high. Time between samples is 1/f
.
OSC
105/231
uPSD33xx
2
Table 56. Number of I C Bus Samples Taken after 1-to-0 Transition on SDA (START Condition)
Contents of S1SETUP
Resulting Number of Samples
Resulting value for S1SETUP
Taken After 1-to-0 on SDA Line
SS_EN bit
SMPL_SET[6:0]
XXXXXXXb
0000000b
0000001b
0000010b
...
0
1
00h (default)
1
1
80h
81h
82h
...
1
2
1
3
...
1
...
0001011b
0010111b
...
8Bh
97h
...
12
24
...
1
...
1
1111111b
FFh
128
Table 57. Start Condition Hold Time
Minimum START Condition Hold
2
2
Range of I C Clock Speed (f
)
I C Bus Speed
SCL
Time (tHLDSTA
)
Standard
Fast
Up to 100KHz
4000ns
101KHz to
400KHz
600ns
160ns
(1)
High
401KHz to 833KHz
Note: 1. 833KHz is maximum for uPSD33xx devices.
106/231
uPSD33xx
Table 58 provides recommended settings for
S1SETUP based on various combinations of f
Important: The SCL bit rate f
must first be de-
SCL
termined by bits CR[2:0] in the SFR S1CON be-
fore a value is chosen for SMPL_SET[6:0] in the
SFR S1SETUP.
OSC
and f
. Note that the “Total Sample Period”
SCL
times in Table 57., page 106 are typically slightly
less than the minimum START condition hold time,
2
t
for a given I C bus speed.
HLDSTA
2
Table 58. S1SETUP Examples for Various I C Bus Speeds and Oscillator Frequencies
2
Oscillator Frequency, f
OSC
I C Bus
Speed,
Parameter
6 MHz
12 MHz
24 MHz
33 MHz
40 MHz
f
SCL
Recommended
S1SETUP Value
93h
A7h
CFh
EEh
FFh
Number of Samples
Time Between Samples
Total Sampled Period
20
40
80
111
30ns
128
25ns
Standard
166.6ns
3332ns
83.3ns
3332ns
41.6ns
3332ns
3333ns
3200ns
Recommended
S1SETUP Value
82h
85h
8Bh
90h
93h
Number of Samples
Time Between Samples
Total Sampled Period
3
6
12
17
20
Fast
166.6ns
500ns
83.3ns
500ns
41.6ns
500ns
30ns
510ns
25ns
500ns
Recommended
S1SETUP Value
(Note 1)
80
82
83
84
Number of Samples
Time Between Samples
Total Sampled Period
-
-
-
1
3
4
5
High
83.3ns
83.3
41.6ns
125ns
30ns
120ns
25ns
125ns
2
Note: 1. Not compatible with High Speed I C.
107/231
uPSD33xx
2
I C Operating Sequences
The following pseudo-code explains hardware
control for these I C functions on the uPSD33xx:
Disable Master from returning an ACK
SFR S1CON.AA = 0
Enable I2C SIOE
SFR S1CON.INI1 = 1
Transmit Address and R/W bit = 0 to
Slave
2
–
–
–
–
–
–
–
Initialize the Interface
Function as Master-Transmitter
Function as Master-Receiver
Function as Slave-Transmitter
Function as Slave-Receiver
Interrupt Service Routine
–
–
Is bus not busy? (SFR S1STA.BBUSY
= 0?)
2
<If busy, then test until not busy>
Full C code drivers for the uPSD33xx I C inter-
face, and other interfaces are available from the
web at www.st.com\psm.
–
–
SFR S1DAT[7:0] = Load Slave Ad-
dress & FEh
SFR S1CON.STA = 1, send START on
bus
Initialization after a uPSD33xx reset
Ensure pins P3.6 and P3.7 are GPIO in-
puts
<bus transmission begins>
–
SFR P3.7 = 1 and SFR P3.6 = 1
Enable All Interrupts and go do some-
thing else
2
Configure pins P3.6 and P3.7 as I C
–
SFR IE.EA = 1
–
SFR P3SFS.6 = 1 and P3SFS.7 = 1
2
Master-Receiver
Set I C clock prescaler to determine
f
Disable all interrupts
SCL
–
SFR S1CON.CR[2:0] = desired SCL
freq.
Set bus START condition sampling
–
SFR IE.EA = 0
Set pointer to global data recv buff-
er, set count
–
SFR S1SETUP[7:0] = number of sam-
–
–
*recv_buf = *pointer to data
buf_length = number of bytes to
recv
ples
2
Enable individual I C interrupt and
set priority
–
–
Set global variables to indicate Mas-
ter-Xmitter
SFR IEA.I2C = 1
SFR IPA.I2C = 1 if high priority is
desired
–
I2C_master = 1, I2C_xmitter = 0
Disable Master from returning an ACK
SFR S1CON.AA = 0
Enable I2C SIOE
SFR S1CON.INI1 = 1
Set the Device address for Slave mode
SFR S1ADR = XXh, desired address
–
–
Enable SIOE (as Slave) to return an
ACK signal
–
Master-Transmitter
Disable all interrupts
–
Transmit Address and R/W bit = 1 to
Slave
–
SFR S1CON.AA = 1
Is bus not busy? (SFR S1STA.BBUSY
= 0?)
–
SFR IE.EA = 0
<If busy, then test until not busy>
Set pointer to global data xmit buff-
er, set count
–
–
SFR S1DAT[7:0] = Load Slave Ad-
dress # 01h
SFR S1CON.STA = 1, send START on
bus
–
–
*xmit_buf = *pointer to data
buf_length = number of bytes to
xmit
<bus transmission begins>
Set global variables to indicate Mas-
ter-Xmitter
Enable All Interrupts and go do some-
thing else
–
I2C_master = 1, I2C_xmitter = 1
–
SFR IE.EA = 1
108/231
uPSD33xx
2
Slave-Transmitter
Interrupt Service Routine (ISR). A typical I C
interrupt service routine would handle a interrupt
for any of the four combinations of Master/Slave
and Transmitter/Receiver. In the example routines
above, the firmware sets global variables,
I2C_master and I2C_xmitter, before enabling in-
terrupts. These flags tell the ISR which one of the
four cases to process. Following is pseudo-code
Disable all interrupts
–
SFR IE.EA = 0
Set pointer to global data xmit buff-
er, set count
–
–
*xmit_buf = *pointer to data
buf_length = number of bytes to
xmit
2
for high-level steps in the I C ISR:
2
2
Begin I C ISR <I C interrupt just occurred>:
Set global variables to indicate Mas-
ter-Xmitter
Clear I2C interrupt flag:
–
S1STA.INTR = 0
–
I2C_master = 0, I2C_xmitter = 1
Read status of SIOE, put in to vari-
able, status
Enable SIOE
–
SFR S1CON.INI1 = 1
–
status = S1STA
Prepare to Xmit first data byte
Read global variables that determine
the mode
–
SFR S1DAT[7:0] = xmit_buf[0]
Enable All Interrupts and go do some-
thing else
–
mode <= (I2C_master, I2C_slave)
If mode is Master-Transmitter
–
SFR IE.EA = 1
Bus Arbitration lost? (sta-
tus.BLOST=1?)
If Yes, Arbitration was lost:
Slave-Receiver
Disable all interrupts
–
SFR IE.EA = 0
–
–
S1DAT = dummy, write to release bus
Exit ISR, SIOE will switch to Slave
Recv mode
Set pointer to global data recv buff-
er, set count
–
–
*recv_buf = *pointer to data
buf_length = number of bytes to
recv
If No, Arbitration was not
lost, continue:
ACK
recvd
from
Slave?
(sta-
Set global variables to indicate Mas-
ter-Xmitter
tus.ACK_RESP=0?)
If No, an ACK was not received:
S1CON.STO = 1, set STOP bus condi-
tion
<STOP occurs after ISR exit>
S1DAT = dummy, write to release bus
Exit ISR
–
I2C_master = 0, I2C_xmitter = 0
–
Enable SIOE
–
SFR S1CON.INI1 = 1
–
–
–
Enable All Interrupts and go do some-
thing else
–
SFR IE.EA = 1
If Yes, ACK was received, then
continue:
–
S1DAT = xmit_buf[buffer_index],
transmit byte
Was that the last byte of data to
transmit?
If No, it was not the last byte,
then:
–
–
Exit ISR, transmit next byte on
next interrupt
If Yes, it was the last byte,
then:
S1CON.STO = 1, set STOP bus condi-
tion
<STOP occurs after ISR exit>
S1DAT = dummy, write to release bus
Exit ISR
–
–
109/231
uPSD33xx
Else If mode is Master-Receiver:
Is this the last data byte to receive
from Slave?
Bus
Arbitration
lost?
(sta-
tus.BLOST=1?)
If Yes, tell Slave to stop
transmitting:
If Yes, Arbitration was lost:
–
–
S1CON.STO = 1, set STOP bus condi-
tion
<STOP occurs after ISR exit>
Exit ISR, finished receiving data
from Slave
–
–
S1DAT = dummy, write to release bus
Exit ISR, SIOE will switch to Slave
Recv mode
If No, Aribitration was not
lost, continue:
If No, continue:
Is this the next to last byte to re-
ceive from Slave?
Is this Interrupt from sending an ad-
dress to Slave, or is it from receiv-
ing a data byte from Slave?
If its from sending Slave ad-
dress, goto A:
If this is the next to last
byte, do not allow Master to ACK
on next interrupt.
If its from receiving Slave da-
ta, goto B:
A: (Interrupt is from Master sending
addr to Slave)
–
–
S1CON.AA = 0, don’t let Master re-
turn ACK
Exit ISR, now ready to recv last
byte from Slv
ACK
recvd
from
Slave?
(sta-
tus.ACK_RESP=0?)
If this is not next to last
byte, let Master send ACK to
Slave
If No, an ACK was not received:
S1CON.STO = 1, set STOP condition
<STOP occurs after ISR exit>
dummy = S1DAT, read to release bus
Exit ISR
–
<S1CON.AA is already 1>
–
Exit ISR, ready to recv more bytes
from Slave
–
–
Else If mode is Slave-Transmitter:
Is this Intr from SIOE detecting a
STOP on bus?
If Yes, ACK was received, then
continue:
–
dummy = S1DAT, read to release bus
If Yes, a STOP was detected:
S1DAT = dummy, write to release bus
Exit ISR, Master needs no more data
bytes
Does Master want to receive just one
data byte?
–
–
If Yes, do not allow Master to
ACK on next interrupt:
<S1CON.AA is already 0>
If No, a STOP was not detected,
continue:
–
Exit ISR, now ready to recv one
byte from Slv
ACK
recvd
from
Master?
(sta-
tus.ACK_RESP=0?)
If No, Master can ACK next byte
from Slv
If No, an ACK was not received:
S1DAT = dummy, write to release bus
Exit ISR, Master needs no more data
bytes
If Yes, ACK was received, then
continue:
S1DAT = xmit_buf[buffer_index],
transmit byte
–
–
–
–
S1CON.AA = 1, allow Master to send
ACK
Exit ISR, now ready to recv data
from Slave
B: (Interrupt is from Master recving
data from Slv)
–
–
–
recv_buf[buffer_index] = S1DAT,
read byte
Exit ISR, transmit next byte on
next interrupt
110/231
uPSD33xx
Else If mode is Slave-Receiver:
Is this Intr from SIOE detecting a
STOP on bus?
–
S1CON.ADDR = 0, clear address
match flag
Determine if R/W bit indicates trans-
mit or receive.
Does status.TX_MODE = 1?
If Yes, a STOP was detected:
–
–
recv_buf[buffer_index] = S1DAT,
get last byte
Exit ISR, Master has sent last byte
If Yes, Master wants transmit
mode
Exit ISR, indicate Master wants
Slv-Xmit mode
–
If No, a STOP was not detected,
continue:
If No, Master wants Slave-Recv
mode
dummy = S1DAT, read to release bus
Exit ISR, ready to recv data on
next interrupt
Determine if this Interrupt is from
receiving an address or a data byte
from a Master.
–
–
Is (S1CON.ADDR = 1 and S1CON.AA =1)?
If No, intr is from receiving
data, goto C:
If Yes, intr is from an address,
continue:
C: (Interrupt is from Slv receiving
data from Mastr)
–
recv_buf[buffer_index] = S1DAT,
read byte
Exit ISR, recv next byte on next
interrupt
–
slave_is_adressed = 1, local vari-
able set true
–
<indicates Master selected this
slave>
111/231
uPSD33xx
SPI (SYNCHRONOUS PERIPHERAL INTERFACE)
uPSD33xx devices support one serial SPI inter-
face in Master Mode only. This is a three- or four-
wire synchronous communication channel, capa-
ble of full-duplex operation on 8-bit serial data
transfers. The four SPI bus signals are:
This SPI interface supports single-Master/multi-
ple-Slave connections. Multiple-Master connec-
tions are not directly supported by the uPSD33xx
(no internal logic for collision detection).
If more than one Slave device is required, the
SPISEL signal may be generated from uPSD33xx
GPIO outputs (one for each Slave) or from the
PLD outputs of the PSD Module. Figure 41. illus-
trates three examples of SPI device connections
using the uPSD33xx:
■
■
■
■
SPIRxD
Pin P1.5 or P4.5 receives data from the Slave
SPI device to the uPSD33xx
SPITxD
Pin P1.6 or P4.6 transmits data from the
uPSD33xx to the Slave SPI device
SPICLK
Pin P1.4 or P4.4 clock is generated from the
uPSD33xx to the SPI Slave device
SPISEL
■
■
■
Single-Master/Single-Slave with SPISEL
Single-Master/Single-Slave without SPISEL
Single-Master/Multiple-Slave without SPISEL
Pin P1.7 or P4.7 selects the signal from the
uPSD33xx to an individual Slave SPI device
Figure 41. SPI Device Connection Examples
SPI Bus
SPI Bus
SPIRxD
SPITxD
SPICLK
SPISEL
MISO
MOSI
SCLK
SS
SPIRxD
SPITxD
SPICLK
MISO
MOSI
SCLK
SS
uPSD33xx
SPI Master
SPI Slave
Device
uPSD33xx
SPI Master
SPI Slave
Device
Single-Master/Single-Slave, with SPISEL
Single-Master/Single-Slave, without SPISEL
SPI Bus
SPIRxD
SPITxD
MISO
MOSI
SCLK
SS
SPI Slave
Device
SPICLK
GPIO or PLD
uPSD33xx
SPI Master
MISO
MOSI
SCLK
SS
SPI Slave
Device
GPIO or PLD
Single-Master/Multiple-Slave, without SPISEL
AI07853b
112/231
uPSD33xx
SPI Bus Features and Communication Flow
The SPICLK signal is a gated clock generated
from the uPSD33xx (Master) and regulates the
flow of data bits. The Master may transmit at a va-
riety of baud rates, and the SPICLK signal will
clock one period for each bit of transmitted data.
Data is shifted on one edge of SPICLK and sam-
pled on the opposite edge.
The SPITxD signal is generated by the Master and
received by the Slave device. The SPIRxD signal
is generated by the Slave device and received by
the Master. There may be no more than one Slave
device transmitting data on SPIRxD at any given
time in a multi-Slave configuration. Slave selection
is accomplished when a Slave’s “Slave Select”
(SS) input is permanently grounded or asserted
active-low by a Master device. Slave devices that
are not selected do not interfere with SPI activities.
Slave devices ignore SPICLK and keep their
MISO output pins in high-impedance state when
not selected.
The Slave device will use this first clock edge as a
transmission start indicator, and therefore the
Slave’s Slave Select input signal may remain
grounded in a single-Master/single-Slave configu-
ration (which means the user does not have to use
the SPISEL signal from uPSD33xx in this case).
The SPI specification does not specify high-level
protocol for data exchange, only low-level bit-seri-
al transfers are defined.
Full-Duplex Operation
When an SPI transfer occurs, 8 bits of data are
shifted out on one pin while a different 8 bits of
data are simultaneously shifted in on a second pin.
Another way to view this transfer is that an 8-bit
shift register in the Master and another 8-bit shift
register in the Slave are connected as a circular
16-bit shift register. When a transfer occurs, this
distributed shift register is shifted 8 bit positions;
thus, the data in the Master and Slave devices are
effectively exchanged (see Figure 42.).
The SPI specification allows a selection of clock
polarity and clock phase with respect to data. The
uPSD33xx supports the choice of clock polarity,
but it does not support the choice of clock phase
(phase is fixed at what is typically known as
CPHA = 1). See Figure 43. and Figure
44., page 114 for SPI data and clock relationships.
Referring to these figures (43 and 44), when the
phase mode is defined as such (fixed at
CPHA =1), in a new SPI data frame, the Master
device begins driving the first data bit on SPITxD
at the very first edge of the first clock period of SPI-
CLK.
Bus-Level Activity
Figure 43. details an SPI receive operation (with
respect to bus Master) and Figure 44. details an
SPI transmit operation. Also shown are internal
flags available to firmware to manage data flow.
These flags are accessed through a number of
SFRs.
Note: The uPSD33xx SPI interface SFRs allow
the choice of transmitting the most significant bit
(MSB) of a byte first, or the least significant bit
(LSB) first. The same bit-order applies to data re-
ception. Figures 43 and 44 illustrate shifting the
LSB first.
Figure 42. SPI Full-Duplex Data Exchange
Master Device
Slave Device
SPI Bus
SPIRxD
MISO
8-Bit Shift
Register
8-Bit Shift
Register
SPITxD
MOSI
SCLK
SPICLK
Baud Rate
Generator
SS
AI10485
113/231
uPSD33xx
Figure 43. SPI Receive Operation Example
1 frame
SPICLK
(SPO=0)
SPICLK
(SPO=1)
SPIRXD
RISF
Bit7
Bit0
Bit1
Bit7
Bit0
Bit1
Bit7
RORIS
BUSY
SPIINTR
Interrupt handler
read data in SPIRDR
Transmit End
interrupt requested
SPIRDR Full
interrupt requested
SPIRDR Full
interrupt requested
AI07855
Figure 44. SPI Transmit Operation Example
1 frame
SPICLK
(SPO=0)
SPICLK
(SPO=1)
SPITXD
TISF
Bit0
Bit1
Bit7
Bit0
Bit1
Bit7
TEISF
BUSY
SPISEL
SPIINTR
Interrupt handler
write data in TDR
Transmit End
interrupt requested
SPITDR Empty
interrupt requested
SPITDR Empty
interrupt requested
AI07854
114/231
uPSD33xx
SPI SFR Registers
Six SFR registers control the SPI interface:
The SPI interface functional block diagram (Figure
45.) shows these six SFRs. Both the transmit and
receive data paths are double-buffered, meaning
that continuous transmitting or receiving (back-to-
back transfer) is possible by reading from SPIRDR
or writing data to SPITDR while shifting is taking
place. There are a number of flags in the SPISTAT
register that indicate when it is full or empty to as-
sist the 8032 MCU in data flow management.
When enabled, these status flags will cause an in-
terrupt to the MCU.
■
■
■
■
■
■
SPICON0 (Table 59., page 117) for interface
control
SPICON1 (Table 60., page 118) for interrupt
control
SPITDR (SFR D4h, Write only) holds byte to
transmit
SPIRDR (SFR D5h, Read only) holds byte
received
SPICLKD (Table 61., page 118) for clock
divider
SPISTAT (Table 62., page 119) holds
interface status
Figure 45. SPI Interface, Master Mode Only
8032 MCU DATA BUS
8
8
INTR
to
8032
SPICON0, SPICON1
- CONTROL REGISTERS
SPITDR - TRANSMIT REGISTER
8
SPIRxD /
P1.5 or P4.5
8-bit SHIFT REGISTER
TIMING AND CONTROL
8
SPIRDR - RECEIVE REGISTER
8
SPISTAT - STATUS REGISTER
8
SPITxD / P1.6 or P4.6
SPISEL / P1.7 or P4.7
PERIPH_CLK
÷1
÷4
÷8
(f
)
OSC
SPICLK / P1.4 or P4.4
CLOCK
DIVIDE
CLOCK
÷16
÷32
GENERATE
÷64
÷128
8
SPICLKD - DIVIDE SELECT
AI10486
115/231
uPSD33xx
SPI Configuration
The SPI interface is reset by the MCU reset, and
firmware needs to initialize the SFRs SPICON0,
SPICON1, and SPICLKD to define several opera-
tion parameters.
The SPICLK frequency must be set low enough to
allow the MCU time to read received data bytes
without loosing data. This is dependent upon
many things, including the crystal frequency of the
MCU and the efficiency of the SPI firmware.
The SPO Bit in SPICON0 determines the clock po-
larity. When SPO is set to '0,' a data bit is transmit-
ted on SPITxD from one rising edge of SPICLK to
the next and is guaranteed to be valid during the
falling edge of SPICLK. When SPO is set to '1,' a
data bit is transmitted on SPITxD from one falling
edge of SPICLK to the next and is guaranteed to
be valid during the rising edge of SPICLK. The
uPSD33xx will sample received data on the appro-
priate edge of SPICLK as determined by SPO.
The effect of the SPO Bit can be seen in Figure 43.
and Figure 44., page 114.
Dynamic Control
At runtime, bits in registers SPICON0, SPICON1,
and SPISTAT are managed by firmware for dy-
namic control over the SPI interface. The bits
Transmitter Enable (TE) and Receiver Enable
(RE) when set will allow transmitting and receiving
respectively. If TE is disabled, both transmitting
and receiving are disabled because SPICLK is
driven to constant output logic ‘0’ (when SPO = 0)
or logic '1' (when SPO = 1).
When the SSEL Bit is set, the SPISEL pin will drive
to logic '0' (active) to select a connected slave de-
vice at the appropriate time before the first data bit
of a byte is transmitted, and SPISEL will automat-
ically return to logic '1' (inactive) after transmitting
the eight bit of data, as shown in Figure
44., page 114. SPISEL will continue to automati-
cally toggle this way for each byte data transmis-
sion while the SSEL bit is set by firmware. When
the SSEL Bit is cleared, the SPISEL pin will drive
to constant logic '1' and stay that way (after a
transmission in progress completes).
The Interrupt Enable Bits (TEIE, RORIE,TIE, and
RIE) when set, will allow an SPI interrupt to be
generated to the MCU upon the occurrence of the
condition enabled by these bits. Firmware must
read the four corresponding flags in the SPISTAT
register to determine the specific cause of inter-
rupt. These flags are automatically cleared when
firmware reads the SPISTAT register.
The FLSB Bit in SPICON0 determines the bit order
while transmitting and receiving the 8-bit data.
When FLSB is '0,' the 8-bit data is transferred in or-
der from MSB (first) to LSB (last). When FLSB Bit
is set to '1,' the data is transferred in order from
LSB (first) to MSB (last).
The clock signal generated on SPICLK is derived
from
PERIPH_CLK always operates at the frequency,
, and runs constantly except when stopped in
the
internal
PERIPH_CLK
signal.
f
OSC
MCU Power Down mode. SPICLK is a result of di-
viding PERIPH_CLK by a sum of different divisors
selected by the value contained in the SPICLKD
register. The default value in SPICLKD after a re-
set divides PERIPH_CLK by a factor of 4. The bits
in SPICLKD can be set to provide resulting divisor
values in of sums of multiples of 4, such as 4, 8,
12, 16, 20, all the way up to 252. For example, if
SPICLKD contains 0x24, SPICLK has the fre-
quency of PERIH_CLK divided by 36 decimal.
116/231
uPSD33xx
Table 59. SPICON0: Control Register 0 (SFR D6h, Reset Value 00h)
Bit 7
–
Bit 6
TE
Bit 5
RE
Bit 4
Bit 3
Bit 2
Bit 1
SBO
Bit 0
–
SPIEN
SSEL
FLSB
Details
Bit
Symbol
R/W
Definition
7
–
–
Reserved
Transmitter Enable
6
5
4
TE
RE
RW
RW
RW
0 = Transmitter is disabled
1 = Transmitter is enabled
Receiver Enable
0 = Receiver is disabled
1 = Receiver is enabled
SPI Enable
SPIEN
0 = Entire SPI Interface is disabled
1 = Entire SPI Interface is enabled
Slave Selection
3
2
SSEL
FLSB
RW
RW
0 = SPISEL output pin is constant logic '1' (slave device not selected)
1 = SPISEL output pin is logic '0' (slave device is selected) during data
transfers
First LSB
0 = Transfer the most significant bit (MSB) first
1 = Transfer the least significant bit (LSB) first
Sampling Polarity
0 = Sample transfer data at the falling edge of clock (SPICLK is '0' when
idle)
1 = Sample transfer data at the rising edge of clock (SPICLK is '1' when
idle)
1
0
SPO
–
–
–
Reserved
117/231
uPSD33xx
Table 60. SPICON1: SPI Interface Control Register 1 (SFR D7h, Reset Value 00h)
Bit 7
–
Bit 6
–
Bit 5
–
Bit 4
–
Bit 3
Bit 2
Bit 1
TIE
Bit 0
RIE
TEIE
RORIE
Details
Bit
Symbol
R/W
Definition
7-4
–
–
Reserved
Transmission End Interrupt Enable
3
2
1
0
TEIE
RORIE
TIE
RW
RW
RW
RW
0 = Disable Interrupt for Transmission End
1 = Enable Interrupt for Transmission End
Receive Overrun Interrupt Enable
0 = Disable Interrupt for Receive Overrun
1 = Enable Interrupt for Receive Overrun
Transmission Interrupt Enable
0 = Disable Interrupt for SPITDR empty
1 = Enable Interrupt for SPITDR empty
Reception Interrupt Enable
RIE
0 = Disable Interrupt for SPIRDR full
1 = Enable Interrupt for SPIRDR full
Table 61. SPICLKD: SPI Prescaler (Clock Divider) Register (SFR D2h, Reset Value 04h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
–
Bit 0
–
DIV128
DIV64
DIV32
DIV16
DIV8
DIV4
Details
Bit
Symbol
R/W
Definition
0 = No division
7
DIV128
RW
1 = Divide f
clock by 128
clock by 64
clock by 32
clock by 16
clock by 8
OSC
0 = No division
1 = Divide f
6
5
4
3
DIV64
DIV32
DIV16
DIV8
RW
RW
RW
RW
OSC
0 = No division
1 = Divide f
OSC
0 = No division
1 = Divide f
OSC
0 = No division
1 = Divide f
OSC
0 = No division
1 = Divide f
2
DIV4
RW
–
clock by 4
OSC
1-0
Not Used
118/231
uPSD33xx
Table 62. SPISTAT: SPI Interface Status Register (SFR D3h, Reset Value 02h)
Bit 7
–
Bit 6
–
Bit 5
–
Bit 4
Bit 3
Bit 2
Bit 1
TISF
Bit 0
BUSY
TEISF
RORISF
RISF
Details
Bit
Symbol
–
R/W
–
Definition
7-5
Reserved
SPI Busy
4
3
2
BUSY
TEISF
R
R
R
0 = Transmit or Receive is completed
1 = Transmit or Receive is in process
Transmission End Interrupt Source flag
0 = Automatically resets to '0' when firmware reads this register
1 = Automatically sets to '1' when transmission end occurs
Receive Overrun Interrupt Source flag
RORISF
0 = Automatically resets to '0' when firmware reads this register
1 = Automatically sets to '1' when receive overrun occurs
Transfer Interrupt Source flag
0 = Automatically resets to '0' when SPITDR is full (just after the SPITDR
is written)
1 = Automatically sets to '1' when SPITDR is empty (just after byte loads
from SPITDR into SPI shift register)
1
0
TISF
RISF
R
R
Receive Interrupt Source flag
0 = Automatically resets to '0' when SPIRDR is empty (after the SPIRDR
is read)
1 = Automatically sets to '1' when SPIRDR is full
119/231
uPSD33xx
ANALOG-TO-DIGITAL CONVERTOR (ADC)
The ADC unit in the uPSD33xx is a SAR type ADC
Register. The ADC operates within a range of 2 to
16MHz, with typical ADCCLK frequency at 8MHz.
with an SAR register, an auto-zero comparator
and three internal DACs. The unit has 8 input
channels with 10-bit resolution. The A/D converter
The conversion time is 4µs typical at 8MHz.
The processing of conversion starts when the
Start Bit ADST is set to '1.' After one cycle, it is
cleared by hardware. The ADC is monotonic with
no missing codes. Measurement is by continuous
conversion of the analog input. The ADAT Regis-
ter contains the results of the A/D conversion.
When conversion is complete, the result is loaded
into the ADAT. The A/D Conversion Status Bit
ADSF is set to '1.' The block diagram of the A/D
module is shown in Figure 46. The A/D status bit
ADSF is set automatically when A/D conversion is
completed and cleared when A/D conversion is in
process.
has its own V
input (80-pin package only),
REF
which specifies the voltage reference for the A/D
operations. The analog to digital converter (A/D)
allows conversion of an analog input to a corre-
sponding 10-bit digital value. The A/D module has
eight analog inputs (P1.0 through P1.7) to an 8x1
multiplexor. One ADC channel is selected by the
bits in the configuration register. The converter
generates a 10-bits result via successive approxi-
mation. The analog supply voltage is connected to
the V
input, which powers the resistance lad-
REF
der in the A/D module.
The A/D module has 3 registers, the control regis-
ter ACON, the A/D result register ADAT0, and the
second A/D result register ADAT1. The ADAT0
Register stores Bits 0.. 7 of the converter output,
Bits 8.. 9 are stored in Bits 0..1 of the ADAT1 Reg-
ister. The ACON Register controls the operation of
the A/D converter module. Three of the bits in the
ACON Register select the analog channel inputs,
and the remaining bits control the converter oper-
ation.
ADC channel pin input is enabled by setting the
corresponding bit in the P1SFS0 and P1SFS1
Registers to '1' and the channel select bits in the
ACON Register.
In addition, the ADC unit sets the interrupt flag in
the ACON Register after a conversion is complete
(if AINTEN is set to '1'). The ADC interrupts the
CPU when the enable bit AINTEN is set.
Port 1 ADC Channel Selects
The P1SFS0 and P1SFS1 Registers control the
selection of the Port 1 pin functions. When the
P1SFS0 Bit is '0,' the pin functions as a GPIO.
When bits are set to '1,' the pins are configured as
alternate functions. A new P1SFS1 Register se-
lects which of the alternate functions is enabled.
The ADC channel is enabled when the bit in
P1SFS1 is set to '1.'
Note: In the 52-pin package, there is no individual
The ADC reference clock (ADCCLK) is generated
V
pin because V
is combined with AV
REF CC
REF
from f
divided by the divider in the ADCPS
OSC
pin.
Figure 46. 10-Bit ADC
AV
REF
AV
REF
P1.0
P1.1
P1.2
P1.3
P1.4
P1.5
P1.6
P1.7
ADC0
ADC1
10-BIT SAR ADC
ADC2
ADC3
ADC4
ADC5
ANALOG
MUX
CONTROL
ADC OUT - 10 BITS
ADC6
ADC7
SELECT
ADAT1
REG
ACON REG
ADAT 0 REG
AI07856
120/231
uPSD33xx
Table 63. ACON Register (SFR 97h, Reset Value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
AINTF
AINTEN
ADEN
ADS2
ADS1
ADS0
ADST
ADSF
Details
Bit
Symbol
Function
ADC Interrupt flag. This bit must be cleared with software.
7
AINTF
0 = No interrupt request
1 = The AINTF flag is set when ADSF goes from '0' to '1.' Interrupts CPU when both
AINTF and AINTEN are set to '1.'
ADC Interrupt Enable
6
5
AINTEN
ADEN
0 = ADC interrupt is disabled
1 = ADC interrupt is enabled
ADC Enable Bit
0 = ADC shut off and consumes no operating current
1 = Enable ADC. After ADC is enabled, 16ms of calibration is needed before ADST Bit is
set.
Analog channel Select
000 Select channel 0 (P1.0)
001 Select channel 0 (P1.1)
010 Select channel 0 (P1.2)
011 Select channel 0 (P1.3)
101 Select channel 0 (P1.5)
110 Select channel 0 (P1.6)
111 Select channel 0 (P1.7)
4.. 2
ADS2.. 0
ADC Start Bit
1
0
ADST
ADSF
0 = Force to zero
1 = Start ADC, then after one cycle, the bit is cleared to '0.'
ADC Status Bit
0 = ADC conversion is not completed
1 = ADC conversion is completed. The bit can also be cleared with software.
121/231
uPSD33xx
Table 64. ADCPS Register Details (SFR 94h, Reset Value 00h)
Bit
Symbol
Function
7:4
–
Reserved
ADC Conversion Reference Clock Enable
3
ADCCE
0 = ADC reference clock is disabled (default)
1 = ADC reference clock is enabled
ADC Reference Clock PreScaler
Only three Prescaler values are allowed:
2:0
ADCPS[2:0]
ADCPS[2:0] = 0, for f
ADCPS[2:0] = 1, for f
ADCPS[2:0] = 2, for f
frequency 16MHz or less. Resulting ADC clock is f
frequency 32MHz or less. Resulting ADC clock is f
frequency 32MHz > 40MHz. Resulting ADC clock is f
.
OSC
OSC
OSC
OSC
/2.
OSC
/4.
OSC
Table 65. ADAT0 Register (SFR 95H, Reset Value 00h)
Bit
Symbol
Function
Function
7:0
–
Store ADC output, Bit 7 - 0
Table 66. ADAT1 Register (SFR 96h, Reset Value 00h)
Bit
7:2
Symbol
–
–
Reserved
1.. 0
Store ADC output, Bit 9, 8
122/231
uPSD33xx
PROGRAMMABLE COUNTER ARRAY (PCA) WITH PWM
There are two Programmable Counter Array
blocks (PCA0 and PCA1) in the uPSD33xx. A PCA
block consists of a 16-bit up-counter, which is
shared by three TCM (Timer Counter Module). A
TCM can be programmed to perform one of the
following four functions:
of clock input: from an external pin, Timer 0 Over-
flow, or PCA Clock.
A PCA block has 3 Timer Counter Modules (TCM)
which share the 16-bit Counter output. The TCM
can be configured to capture or compare counter
value, generate a toggling output, or PWM func-
tions. Except for the PWM function, the other TCM
functions can generate an interrupt when an event
occurs.
1. Capture Mode: capture counter values by
external input signals
2. Timer Mode
Every TCM is connected to a port pin in Port 4; the
TCM pin can be configured as an event input, a
PWMs, a Toggle Output, or as External Clock In-
put. The pins are general I/O pins when not as-
signed to the TCM.
3. Toggle Output Mode
4. PWM Mode: fixed frequency (8-bit or 16-bit),
programmable frequency (8-bit only)
PCA Block
The 16-bit Up-Counter in the PCA block is a free-
running counter (except in PWM Mode with pro-
grammable frequency). The Counter has a choice
The TCM operation is configured by Control regis-
ters and Capture/Compare registers. Table
67., page 124 lists the SFR registers in the PCA
blocks.
Figure 47. PCA0 Block Diagram
16-bit up Timer/Counter
PCA0CLK
TIMER0
OVERFLOW
PCACH0
8-bit
PCACL0
8-bit
INT
OVF0
P4.3/ECI
EOVFI
P4.0/CEX0
P4.1/CEX1
TCM0
CLKSEL0
CLKSEL1
EN_ALL
EN_PCA
TCM1
TCM2
PCAIDLE
P4.2/CEX2
IDLE MODE
(From CPU)
PWM FREQ
COMPARE
CLEAR COUNTER
AI07857
123/231
uPSD33xx
Table 67. PCA0 and PCA1 Registers
SFR Address
Register Name
RW
Register Function
PCA0
PCA1
BA
PCA0
PCA1
A2
A3
PCACL0
PCACH0
PCACL1
PCACH1
RW
RW
The low 8 bits of PCA 16-bit counter.
The high 8 bits of PCA 16-bit counter.
Control Register
BB
–
Enable PCA, Timer Overflow flag ,
PCA Idle Mode, and Select clock
source.
A4
A5
BC
A5
PCACON0
PCASTA
PCACON1
N/A
RW
RW
RW
Status Register, Interrupt Status flags
Common for both PCA Block 0 and 1.
–
TCM Mode
–
A9,
AA,
AB
BD,
BE,
BF
TCMMODE0
TCMMODE1
TCMMODE2
TCMMODE3
TCMMODE4
TCMMODE5
Capture, Compare, and Toggle
Enable Interrupts
PWM Mode Select.
–
AC
AD
C1
C2
CAPCOML0
CAPCOMH0
CAPCOML3
CAPCOMH3
RW
RW
RW
Capture/Compare registers of TCM0
Capture/Compare registers of TCM1
Capture/Compare registers of TCM2
AF
B1
C3
C4
CAPCOML1
CAPCOMH1
CAPCOML4
CAPCOMH4
B2
B3
C5
C6
CAPCOML2
CAPCOMH2
CAPCOML5
CAPCOMH5
The 8-bit register to program the PWM
frequency. This register is used for
programmable, 8-bit PWM Mode only.
B4
FB
C7
FC
PWMF0
CCON2
PWMF1
CCON3
RW
RW
Specify the pre-scaler value of PCA0 or
PCA1 clock input
124/231
uPSD33xx
PCA Clock Selection
The clock input to the 16-bit up counter in the PCA
block is user-programmable. The three clock
sources are:
The clock source is selected in the configuration
register PCACON. The Prescaler output clock
PCACLK is the f
divided by the divisor which is
OSC
specified in the CCON2 or CCON3 Register.
When External Clock is selected, the maximum
–
–
–
PCA Prescaler Clock (PCA0CLK, PCA1CLK)
Timer 0 Overflow
External Clock, Pin P4.3 or P4.7
clock frequency should not exceed f
/4.
OSC
Table 68. CCON2 Register Bit Definition (SFR 0FBh, Reset Value 10h)
Bit 7
–
Bit 6
–
Bit 5
–
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PCA0CE
PCA0PS3
PCA0PS2
PCA0PS1
PCA0PS0
Details
Bit
Symbol
R/W
R/W
Definition
PCA0 Clock Enable
4
PCA0CE
0 = PCA0CLK is disabled
1 = PCA0CLK is enabled (default)
PCA0 Prescaler
PCA0PS
[3:0]
3:0
R/W
f
= f
OSC
/ (2 ^ PCA0PS[3:0])
PCA0CLK
Divisor range: 1, 2, 4, 8, 16... 16384, 32768
Table 69. CCON3 Register Bit Definition (SFR 0FCh, Reset Value 10h)
Bit 7
–
Bit 6
–
Bit 5
–
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PCA1CE
PCA1PS3
PCA1PS2
PCA1PS1
PCA1PS0
Details
Bit
Symbol
R/W
R/W
Definition
PCA1 Clock Enable
4
PCA1CE
0 = PCA1CLK is disabled
1 = PCA1CLK is enabled (default)
PCA1 Prescaler
PCA1PS
[3:0]
3:0
R/W
f
= f
OSC
/ (2 ^ PCA1PS[3:0])
PCA1CLK
Divisor range: 1, 2, 4, 8, 16... 16384, 32768
125/231
uPSD33xx
Operation of TCM Modes
Toggle Mode
Each of the TCM in a PCA block supports four
modes of operation. However, an exception is
when the TCM is configured in PWM Mode with
programmable frequency. In this mode, all TCM in
a PCA block must be configured in the same mode
or left to be not used.
In this mode, the user writes a value to the TCM's
CAPCOM registers and enables the comparator.
When there is a match with the Counter output, the
output of the TCM pin toggles. This mode is a sim-
ple extension of the Timer Mode.
PWM Mode - (X8), Fixed Frequency
Capture Mode
In this mode, one or all the TCM's can be config-
ured to have a fixed frequency PWM output on the
port pins. The PWM frequency depends on when
the low byte of the Counter overflows (modulo
256). The duty cycle of each TCM module can be
specified in the CAPCOMHn Register. When the
PCA_Counter_L value is equal to or greater than
the value in CAPCOMHn, the PWM output is
The CAPCOM registers in the TCM are loaded
with the counter values when an external pin input
changes state. The user can configure the counter
value to be loaded by positive edge, negative edge
or any transition of the input signal. At loading, the
TCM can generate an interrupt if it is enabled.
Timer Mode
switched to
a
high state. When the
The TCM modules can be configured as software
timers by enable the comparator. The user writes
a value to the CAPCOM registers, which is then
compared with the 16-bit counter. If there is a
match, an interrupt can be generated to CPU.
PCA_Counter_L Register overflows, the content
in CAPCOMHn is loaded to CAPCOMLn and a
new PWM pulse starts.
Figure 48. Timer Mode
MATCH_TIMER
INTR
INTFn
CAPCOMLn
CAPCOMHn
PCASTA
8
8
ENABLE
16-bit COMPARATOR
MATCH
8
8
PCACLm
PCACHm
16-bit up Timer/Counter
TOGGLE
0
PWM1
0
PWM0
0
TCMMODEn
E_COMP CAP_PE CAP_NE MATCH
EINTF
0
0
RESET
WRITE to
CAPCOMHn
1
0
C
D
EN_FLAG
WRITE to
CAPCOMLn
AI07858
Note: m = 0: n = 0, 1, or 2
m = 1: n = 3, 4, or 5
126/231
uPSD33xx
Figure 49. PWM Mode - (X8), Fixed Frequency
CAPCOMHn
8
CAPCOMLn
ENABLE
MATCH
SET
CLR
8-bit COMPARATORn
Q
Q
S
R
CEXn
8
OVERFLOW
PCACLm
TOGGLE
0
PWM1
PWM0
TCMMODEn
E_COMP CAP_PE CAP_NE MATCH
EINTF
0
0
0
0
AI07859
Note: m = 0: n = 0, 1, or 2
m = 1: n = 3, 4, or 5
127/231
uPSD33xx
PWM Mode - (X8), Programmable Frequency
In this mode, the PWM frequency is not deter-
mined by the overflow of the low byte of the
Counter. Instead, the frequency is determined by
the PWMFm Register. The user can load a value
in the PWMFm Register, which is then compared
to the low byte of the Counter. If there is a match,
the Counter is cleared and the Load registers
(PWMFm, CAPCOMHn) are re-loaded for the next
PWM pulse. There is only one PWMFm Register
which serves all 3 TCM in a PCA block.
If one of the TCM modules is operating in this
mode, the other modules in the PCA must be con-
figured to the same mode or left not to be used.
The duty cycle of the PWM can be specified in the
CAPCOMHn Register as in the PWM with fixed
frequency mode. Different TCM modules can have
their own duty cycle.
Note: The value in the Frequency Register (PWM-
Fm) must be larger than the duty cycle register
(CAPCOM).
Figure 50. PWM Mode - (X8) Programmable Frequency
PWM FREQ COMPARE
PWMFm
CAPCOMHn
8
8
PWMFm = PCACLm
CAPCOMLn
PCACHm
MATCH
SET
CLR
Q
Q
S
R
CEXn
ENABLE
ENABLE
8-bit COMPARATORm
8-bit COMPARATORn
8
PCACLm
CLR
TOGGLE
0
PWM1
PWM0
TCMMODEn
E_COMP CAP_PE CAP_NE MATCH
EINTF
0
0
0
0
AI07860
Note: m = 0: n = 0, 1, or 2
m = 1: n = 3, 4, or 5
128/231
uPSD33xx
PWM Mode - Fixed Frequency, 16-bit
The operation of the 16-bit PWM is the same as
the 8-bit PWM with fixed frequency. In this mode,
one or all the TCM can be configured to have a
fixed frequency PWM output on the port pins. The
PWM frequency is depending on the clock input
frequency to the 16-bit Counter. The duty cycle of
each TCM module can be specified in the CAP-
COMHn and CAPCOMLn Registers. When the 16-
bit PCA_Counter is equal or greater than the val-
ues in registers CAPCOMHn and CAPCOMLn, the
PWM output is switched to a high state. When the
PCA_Counter overflows, CEXn is asserted low.
output switches to a high state. When the 10-bit
PCA counter overflows, the PWM pin is switched
to a logic low and starts the next PWM pulse.
The most-significant 6 bits in the PCACHm
counter and CAPCOMH Register are “Don’t cares”
and have no effect on the PWM generation.
Writing to Capture/Compare Registers
When writing a 16-bit value to the PCA Capture/
Compare registers, the low byte should always be
written first. Writing to CAPCOMLn clears the
E_COMP Bit to '0'; writing to CAPCOMHn sets
E_COMP to '1' the largest duty cycle is 100%
(CAPCOMHn CAPCOMLn = 0x0000), and the
smallest duty cycle is 0.0015% (CAPCOMHn
CAPCOMLn = 0xFFFF). A 0% duty cycle may be
generated by clearing the E_COMP Bit to ‘0’.
PWM Mode - Fixed Frequency, 10-bit
The 10-bit PWM logic requires that all 3 TCMs in
PCA0 or PCA1 operate in the same 10-bit PWM
mode. The 10-bit PWM operates in a similar man-
ner as the 16-bit PWM, except the PCACHm and
PCACLm counters are reconfigured as 10-bit
counters. The CAPCOMHn and CAPCOMLn Reg-
isters become 10-bit registers.
PWM duty cycle of each TCM module can be
specified in the 10-bit CAPCOMHn and CAP-
COMLn Registers. When the 10-bit PCA counter
is equal or greater than the values in the 10-bit
registers CAPCOMHn and CAPCOMLn, the PWM
Control Register Bit Definition
Each PCA has its own PCA_CONFIGn, and each
module within the PCA block has its own
TCM_Mode Register which defines the operation
of that module (see Table 70., page 129 through
Table 71., page 130). There is one PCA_STATUS
Register that covers both PCA0 and PCA1 (see
Table 72., page 131).
Table 70. PCA0 Control Register PCACON0 (SFR 0A4h, Reset Value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
–
Bit 2
–
Bit 1
Bit 0
EN-ALL
EN_PCA
EOVFI
PCAIDLE
CLK_SEL[1:0]
Details
Bit
Symbol
Function
0 = No impact on TCM modules
1 = Enable both PCA counters simultaneously (override the EN_PCA Bits)
7
EN-ALL
This bit is to start the two 16-bit counters in the PCA. For customers who want 5 PWM,
for example, this bit can start all of the PWM outputs.
0 = PCA counter is disabled
1 = PCA counter is enabled
6
EN_PCA
EN_PCA Counter Run Control Bit. Set with software to turn the PCA counter on. Must
be cleared with software to turn the PCA counter off.
5
4
3
2
EOVFI
PCAIDLE
–
1 = Enable Counter Overflow Interrupt if overflow flag (OVF) is set
0 = PCA operates when CPU is in Idle Mode
1 = PCA stops running when CPU is in Idle Mode
Reserved
0 = Select 16-bit PWM
1 = Select 10-bit PWM
10B_PWM
00 Select Prescaler clock as Counter clock
01 Select Timer 0 Overflow
10 Select External Clock pin (P4.3 for PCA0) (MAX clock rate = f
CLK_SEL
[1:0]
1-0
/4)
OSC
129/231
uPSD33xx
Table 71. PCA1 Control Register PCACON1 (SFR 0BCh, Reset Value 00h)
Bit 7
–
Bit 6
Bit 5
Bit 4
Bit 3
–
Bit 2
–
Bit 1
Bit 0
EN_PCA
EOVFI
PCAIDLE
CLK_SEL[1:0]
Details
Bit
Symbol
Function
0 = PCA counter is disabled
1 = PCA counter is enabled
6
EN_PCA
EN_PCA Counter Run Control Bit. Set with software to turn the PCA counter on. Must
be cleared with software to turn the PCA counter off.
5
4
3
2
EOVFI
PCAIDLE
–
1 = Enable Counter Overflow Interrupt if overflow flag (OVF) is set
0 = PCA operates when CPU is in Idle Mode
1 = PCA stops running when CPU is in Idle Mode
Reserved
0 = Select 16-bit PWM
1 = Select 10-bit PWM
10B_PWM
00 Select Prescaler clock as Counter clock
01 Select Timer 0 Overflow
10 Select External Clock pin (P4.7 for PCA1) (MAX clock rate = f
CLK_SEL
[1:0]
1-0
/4)
OSC
130/231
uPSD33xx
Table 72. PCA Status Register PCASTA (SFR 0A5h, Reset Value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
OVF1
INTF5
INTF4
INTF3
OVF0
INTF2
INTF1
INTF0
Details
Bit
Symbol
Function
PCA1 Counter OverFlow flag
7
OFV1
Set by hardware when the counter rolls over. OVF1 flags an interrupt if Bit EOVFI in
PCACON1 is set. OVF1 may be set with either hardware or software but can only be
cleared with software.
TCM5 Interrupt flag
6
5
4
INTF5
INTF4
INTF3
Set by hardware when a match or capture event occurs.
Must be clear with software.
TCM4 Interrupt flag
Set by hardware when a match or capture event occurs.
Must be clear with software.
TCM3 Interrupt flag
Set by hardware when a match or capture event occurs.
Must be clear with software.
PCA0 Counter OverFlow flag
3
OVF0
Set by hardware when the counter rolls over. OVF0 flags an interrupt if Bit EOVFI in
PCACON0 is set. OVF1 may be set with either hardware or software but can only be
cleared with software.
TCM2 Interrupt flag
2
1
0
INTF2
INTF1
INTF0
Set by hardware when a match or capture event occurs.
Must be clear with software.
TCM1 Interrupt flag
Set by hardware when a match or capture event occurs.
Must be clear with software.
TCM0 Interrupt flag
Set by hardware when a match or capture event occurs.
Must be clear with software.
131/231
uPSD33xx
TCM Interrupts
There are 8 TCM interrupts: 6 match or capture in-
terrupts and two counter overflow interrupts. The 8
interrupts are “ORed” as one PCA interrupt to the
CPU.
By the nature of PCA application, it is unlikely that
many of the interrupts occur simultaneously. If
they do, the CPU has to read the interrupt flags
and determine which one to serve. The software
has to clear the interrupt flag in the Status Register
after serving the interrupt.
Table 73. TCMMODE0 - TCMMODE5 (6 Registers, Reset Value 00h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
EINTF
E_COMP
CAP_PE
CAP_NE
MATCH
TOGGLE
PWM[1:0]
Details
Bit
7
Symbol
EINTF
Function
1 - Enable the interrupt flags (INTF) in the Status Register to generate an interrupt.
1 - Enable the comparator when set
6
E_COMP
CAP_PE
CAP_NE
MATCH
5
1 - Enable Capture Mode, a positive edge on the CEXn pin.
1 - Enable Capture Mode, a negative edge on the CEXn pin.
1 - A match from the comparator sets the INTF bits in the Status Register.
1 - A match on the comparator results in a toggling output on CEXn pin.
4
3
2
TOGGLE
01 Enable PWM Mode (x8), fixed frequency. Enable the CEXn pin as a PWM output.
10 Enable PWM Mode (x8) with programmable frequency. Enable the CEXn pin as a
1-0
PWM[1:0]
PWM output.
11 Enable PWM Mode (x10 or x16), fixed frequency. Enable the CEXn pin as a PWM
output.
Table 74. TCMMODE Register Configurations
EINTF E_COMP CAP_PE CAP_NE MATCH TOGGLE PWM1 PWM0
TCM FUNCTION
No operation (reset value)
8-bit PWM, fixed frequency
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
8-bit PWM, programmable
frequency
0
0
1
1
0
0
0
0
0
0
0
0
1
1
0
1
10-bit or 16-bit PMW, fixed
(1)
frequency
X
X
X
X
X
1
1
0
0
0
1
1
0
0
1
0
1
1
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
16-bit toggle
16-bit Software Timer
X
X
X
16-bit capture, negative trigger
16-bit capture, positive trigger
16-bit capture, transition trigger
Note: 1. 10-bit PWM mode requires the 10B_PWM Bit in the PCACON Register set to '1.'
132/231
uPSD33xx
PSD MODULE
The PSD Module is stacked with the MCU Module
to form the uPSD33xx, see uPSD33xx HARD-
WARE DESCRIPTION, page 13. Details of the
PSD Module are shown in Figure 51. The two sep-
arate modules interface with each other at the
8032 Address, Data, and Control interface blocks
in Figure 51.
Figure 51. PSD Module Block Diagram
S U
P L D I N P U T B
PLD INPUT BUS
8032 MCU Module
AI07872B
133/231
uPSD33xx
PSD Module Functional Description
Major functional blocks are shown in Figure
51., page 133. The next sections describe each
major block.
8032 Address/Data/Control Interface. These
signals attach directly to the MCU Module to im-
plement a typical multiplexed 8051-style bus be-
tween the two stacked die. The MCU instruction
prefetch and branch cache logic resides on the
MCU Module, leaving a standard 8051-style mem-
ory interface on the PSD Module.
The active-low reset signal originating from the
MCU Module goes to the PSD Module reset input
(RST). This reset signal can then be routed as an
external output from the uPSD33xx to the system
PC board, if needed, through any one of the PLD
output pins as active-high or active-low logic by
specifying logic equations in PSDsoft Express.
The 8032 address and data busses are routed
throughout the PSD Module as shown in Figure 51
connecting many elements on the PSD Module to
the 8032 MCU. The 8032 bus is not only connect-
ed to the memories, but also to the General PLD,
making it possible for the 8032 to directly read and
write individual logic macrocells inside the General
PLD.
Dual Flash Memories and IAP. uPSD33xx de-
vices contain two independent Flash memory ar-
rays. This means that the 8032 can read
instructions from one Flash memory array while
erasing or writing the other Flash memory array.
Concurrent operation like this enables robust re-
mote updates of firmware, also known as In-Appli-
cation Programming (IAP). IAP can occur using
any uPSD33xx interface (e.g., UART, I2C, SPI).
Concurrent memory operation also enables the
designer to emulate EEPROM memory within ei-
ther of the two Flash memory arrays for small data
sets that have frequent updates.
The 8032 can erase Flash memories by individual
sectors or it can erase an entire Flash memory ar-
ray at one time. Each sector in either Flash mem-
ory may be individually write protected, blocking
any WRITEs from the 8032 (good for boot and
start-up code protection). The Flash memories au-
tomatically go to standby between 8032 READ or
WRITE accesses to conserve power. Minimum
erase cycles is 100K and minimum data retention
is 15 years. Flash memory, as well as the entire
PSD Module may be programmed with the JTAG
In-System Programming (ISP) interface with no
8032 involvement, good for manufacturing and lab
development.
Main Flash Memory. The Main Flash memory is
divided into equal sized sectors that are individual-
ly selectable by the Decode PLD output signals,
named FSx, one signal for each Main Flash mem-
ory sector. Each Flash sector can be located at
any address within 8032 program address space
(accessed with PSEN) or data address space,
also known as 8032 XDATA space (accessed with
RD or WR), as defined with the software develop-
ment tool, PSDsoft Express. The user only has to
specify an address range for each segment and
specify if Main Flash memory will reside in 8032
data or program address space, and then PSEN,
RD, or WR are automatically activated for the
specified range. 8032 firmware is easily pro-
grammed into Main Flash memory using PSDsoft
Express or other software tools. See Table
75., page 135 for Main Flash sector sizes on the
various uPSD33xx devices.
Secondary Flash Memory. The smaller Second-
ary Flash memory is also divided into equal sized
sectors that are individually selectable by the De-
code PLD signals, named CSBOOTx, one signal
for each Secondary Flash memory sector. Each
sector can be located at any address within 8032
program address space (accessed with PSEN) or
XDATA space (accessed with RD or WR) as de-
fined with PSDsoft Express. The user only has to
specify an address range for each segment, and
specify if Secondary Flash memory will reside in
8032 data or program address space, and then
PSEN, RD, or WR are automatically activated for
the specified range. 8032 firmware is easily pro-
grammed into Secondary Flash memory using PS-
Dsoft Express and others. See Table
75., page 135 for Secondary Flash sector sizes.
SRAM. The SRAM is selected by a single signal,
named RS0, from the Decode PLD. SRAM may be
located at any address within 8032 XDATA space
(accessed with RD or WR), or optionally within
8032 program address space (accessed with
PSEN) to execute code from SRAM. The default
setting places SRAM in XDATA space only. These
choices are specified using PSDSoft Express,
where the user specifies an SRAM address range.
The user would also specify (at run-time) if SRAM
will additionally reside in 8032 program address
space, and then PSEN, RD, or WR are automati-
cally activated for the specified range. See Table
75., page 135 for SRAM sizes.
The SRAM may optionally be backed up by an ex-
ternal battery (or other DC source) to make its con-
tents non-volatile (see SRAM Standby Mode
(battery backup), page 193).
134/231
uPSD33xx
Table 75. uPSD33xx Memory Configuration
Main Flash Memory
Secondary Flash Memory
Total Individual Number of
SRAM
SRAM
Total
Individual
Number of
Device
Flash Size Sector Size Sectors (Sector FlashSize Sector Size Sectors (Sector
Size
(bytes)
64K
(bytes)
16K
Select Signal)
4 (FS0-3)
(bytes)
16K
(bytes)
8K
Select Signal)
2 (CSBOOT0-1)
4 (CSBOOT0-3)
4 (CSBOOT0-3)
4 (CSBOOT0-3)
(bytes)
uPSD3312
uPSD3333
uPSD3334
uPSD3354
2K
8K
128K
256K
256K
16K
8 (FS0-7)
32K
8K
32K
8 (FS0-7)
32K
8K
8K
32K
8 (FS0-7)
32K
8K
32K
Runtime Control Registers, CSIOP. A block of
256 bytes is decoded inside the PSD Module for
module control and status (see Table
79., page 145). The base address of these 256 lo-
cations is referred to in this data sheet as csiop
(Chip Select I/O Port), and is selected by the De-
code PLD output signal, CSIOP. The csiop regis-
ters are always viewed by the 8032 as XDATA,
and are accessed with RD and WR signals. The
address range of CSIOP is specified using PSD-
soft Express where the user only has to specify an
address range of 256 bytes, and then the RD or
WR signals are automatically activated for the
specified range. Individual registers within this
block are accessed with an offset from the speci-
fied csiop base address. 39 registers are used out
of the 256 locations to control the output state of I/
O pins, to read I/O pins, to set the memory page,
to control 8032 program and data address space,
to control power management, to READ/WRITE
macrocells inside the General PLD, and other
functions during runtime. Unused locations within
csiop are reserved and should not be accessed.
Programmable Logic (PLDs) . The uPSD33xx
contains two PLDs (Figure 63., page 157) that
may optionally run in Turbo or Non-Turbo mode.
PLDs operate faster (less propagation delay)
while in Turbo mode but consume more power
than in Non-Turbo mode. Non-Turbo mode allows
the PLDs to go to standby automatically when no
PLD inputs are changing to conserve power.
The logic configuration (from equations) of both
PLDs is stored with non-volatile Flash technology
and the logic is active upon power-up. PLDs may
NOT be programmed by the 8032, PLD program-
ming only occurs through the JTAG interface.
Figure 52. Memory Page Register
Page
Register
D0
D1
D2
D3
D4
D5
D6
D7
Q0
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Chip-
Selects
and
General
Logic
8032
Data
Bus
PGR0-7
DPLD
and
GPLD
Memory Page Register. 8032 MCU architecture
has an inherent size limit of 64K bytes in either
program address space or XDATA space. Some
uPSD33xx devices have much more memory that
64K, so special logic such as this page register is
needed to access the extra memory. This 8-bit
page register (Figure 52) can be loaded and read
by the 8032 at runtime as one of the csiop regis-
ters. Page register outputs feed directly into both
PLDs creating extended address signals used to
“page” memory beyond the 64K byte limit (pro-
gram space or XDATA). Most 8051 compilers di-
rectly support memory paging, also known as
memory banking. If memory paging is not needed,
or if not all eight page register bits are needed for
memory paging, the remaining bits may be used in
the General PLD for general logic. Page Register
outputs are cleared to logic ’0’ at reset and power-
up.
Load or
Read via
csiop +
offset E0h
RST
RST
(PSD Module Reset)
AI09172
135/231
uPSD33xx
PLD #1, Decode PLD (DPLD). This programma-
ble logic implements memory mapping and is used
to select one of the individual Main Flash memory
segments, one of individual Secondary Flash
memory segments, the SRAM, or the group of
csiop registers when the 8032 presents an ad-
dress to DPLD inputs (see Figure 64., page 159).
The DPLD can also optionally drive external chip
select signals on Port D pins. The DPLD also op-
tionally produces two select signals (PSEL0 and
PSEL1) used to enable a special data bus repeat-
er function on Port A, referred to as Peripheral I/O
Mode. There are 69 DPLD input signals which in-
clude: 8032 address and control signals, Page
Register outputs, PSD Module Port pin inputs, and
GPLD logic feedback.
OMC Allocator. The OMC allocator (Figure
67., page 163) will route eight of the OMCs from
MCELLAB to pins on either Port A or Port B, and
will route eight of the OMCs from MCELLBC to
pins on either Port B or Port C, based on what is
specified in PSDsoft Express.
IMCs. Inputs from pins on Ports A, B, and C are
routed to IMCs for conditioning (clocking or latch-
ing) as they enter the chip, which is good for sam-
pling and debouncing inputs. Alternatively, IMCs
can pass port input signals directly to PLD inputs
without
clocking
or
latching
(Figure
68., page 167). The 8032 may read the IMCs
asynchronously at any time through IMC registers
in csiop.
Note: The JTAG signals TDO, TDI, TCK, and TMS
on Port C do not route through IMCs, but go direct-
ly to JTAG logic.
PLD #2, General PLD (GPLD). This
program-
mable logic is used to create both combinatorial
and sequential general purpose logic (see Figure
65., page 161). The GPLD contains 16 Output
Macrocells (OMCs) and 20 Input Macrocells
(IMCs). Output Macrocell registers are unique in
that they have direct connection to the 8032 data
bus allowing them to be loaded and read directly
by the 8032 at runtime through OMC registers in
csiop. This direct access is good for making small
peripheral devices (shifters, counters, state ma-
chines, etc.) that are accessed directly by the 8032
with little overhead. There are 69 GPLD inputs
which include: 8032 address and control signals,
Page Register outputs, PSD Module Port pin in-
puts, and GPLD feedback.
I/O Ports. For 80-pin uPSD33xx devices, the
PSD Module has 22 individually configurable I/O
pins distributed over four ports (these I/O are in
addition to I/O on MCU Module). For 52-pin
uPSD33xx devices, the PSD Module has 13 indi-
vidually configurable I/O pins distributed over
three ports. See Figure 74., page 181 for I/O port
pin availability on these two packages.
I/O port pins on the PSD Module (Ports A, B, C,
and D) are completely separate from the port pins
on the MCU Module (Ports 1, 3, and 4). They even
have different electrical characteristics. I/O port
pins on the PSD Module are accessed by csiop
registers, or they are controlled by PLD equations.
Conversely, I/O Port pins on the MCU Module are
controlled by the 8032 SFR registers.
OMCs. There are two banks of eight OMCs inside
the GPLD, MCELLAB, and MCELLBC, totalling 16
OMCs all together. Each individual OMC is a base
logic element consisting of a flip-flop and some
AND-OR logic (Figure 66., page 162). The gener-
al structure of the GPLD with OMCs is similar in
nature to a 22V10 PLD device with the familiar
sum-of-products (AND-OR) construct. True and
compliment versions of 69 input signals are avail-
able to the inputs of a large AND-OR array. AND-
OR array outputs feed into an OR gate within each
OMC, creating up to 10 product-terms for each
OMC. Logic output of the OR gate can be passed
on as combinatorial logic or combined with a flip-
flop within in each OMC to realize sequential logic.
OMC outputs can be used as a buried nodes driv-
ing internal feedback to the AND-OR array, or
OMC outputs can be routed to external pins on
Ports A, B, or C through the OMC Allocator.
Table 76. General I/O pins on PSD Module
Pkg
Port A Port B Port D Port D Total
52-pin
80-pin
0
8
8
8
4
4
1
2
13
22
Note: Four pins on Port C are dedicated to JTAG, leaving four pins
for general I/O.
136/231
uPSD33xx
Each I/O pin on the PSD Module can be individu-
ally configured for different functions on a pin-by-
pin basis (Figure 69., page 169). Following are the
available functions on PSD Module I/O pins.
the IEEE-1149.1 Boundary Scan functions, but
uses the JTAG interface for ISP an 8032 debug.
The PSD Module can reside in a standard JTAG
chain with other JTAG devices and it will remain in
BYPASS mode when other devices perform JTAG
functions.
ISP programming time can be reduced as much as
30% by using two optional JTAG signals on Port
C, TSTAT and TERR, in addition to TMS, TCK,
TDI and TDO, and this is referred to as “6-pin
JTAG”. The FlashLINK JTAG programming cable
is available from STMicroelectronics and PSDsoft
Express software is available at no charge from
www.st.com/psm. More JTAG ISP information
maybe found in the section titled “JTAG ISP and
Debug” on page 137.
–
MCU I/O: 8032 controls the output state of
each port pin or it reads input state of each
port pin, by accessing csiop registers at run-
time. The direction (in or out) of each pin is
also controlled by csiop registers at run-time.
–
PLD I/O: PSDsoft Express logic equations
and pin configuration selections determine if
pins are connected to OMC outputs or IMC
inputs. This is a static and non-volatile
configuration. Port pins connected to PLD
outputs can no longer be driven by the 8032
using MCU I/O output mode.
The MCU module is also included in the JTAG
chain within the uPSD33xx device for 8032 debug-
ging and emulation. While debugging, the PSD
Module is in BYPASS mode. Conversely, during
ISP, the MCU Module is in BYPASS mode.
–
–
Latched MCU Address Output: Port A or
Port B can output de-multiplexed 8032
address signals A0 - A7 on a pin-by-pin basis
as specified in csiop registers at run-time. In
addition, Port B can also be configured to
output de-multiplexed A8-A15 in PSDsoft
Express.
Power Management. The PSD Module has bits
in csiop registers that are configured at run-time by
the 8032 to reduce power consumption of the
GPLD. The Turbo Bit in the PMMR0 Register can
be set to logic ’1’ and both PLDs will go to Non-
Turbo mode, meaning it will latch its outputs and
go to sleep until the next transition on its inputs.
There is a slight penalty in PLD performance
(longer propagation delay), but significant power
savings are realized. Going to Non-Turbo mode
may require an additional wait state in the 8032
SFR, BUSCON, because memory decode signals
are also delayed. The default state of the Turbo Bit
is logic '0,' meaning by default, the GPLD is in fast
Turbo mode until the Turbo mode is turned off.
Data Bus Repeater: Port A can bi-
directionally buffer the 8032 data bus (de-
multiplexed) for a specified address range in
PSDsoft Express. This is referred to as
Peripheral I/O Mode in this document.
Open Drain Outputs: Some port pins can
function as open-drain as specified in csiop
registers at run-time.
–
–
Pins on Port D can be used for external chip-
select outputs originating from the DPLD,
without consuming OMC resources within the
GPLD.
JTAG Port. In-System Programming (ISP) can
be performed through the JTAG signals on Port C.
This serial interface allows programming of the en-
tire PSD Module device or subsections of the PSD
Module (for example, only Flash memory but not
the PLDs) without the participation of the 8032. A
blank uPSD33xx device soldered to a circuit board
can be completely programmed in 10 to 25 sec-
onds. The four basic JTAG signals on Port C;
TMS, TCK, TDI, and TDO form the IEEE-1149.1
interface. The PSD Module does not implement
Additionally, bits in csiop registers PMMR0 and
PMMR2 can be set by the 8032 to selectively
block signals from entering both PLDs which fur-
ther reduces power consumption. There is also an
Automatic Power Down counter that detects lack
of 8032 activity and reduces power consumption
on the PSD Module to its lowest level (see Power
Management, page 137).
137/231
uPSD33xx
Security and NVM Sector Protection. A
pro-
8032 Program Address Space. In the example
of Figure 53, six sectors of Main Flash memory
(fs2.. fs7) are paged across three memory pages
in the upper half of program address space, and
the remaining two sectors of Main Flash memory
(fs0, fs1) reside in the lower half of program ad-
dress space, and these two sectors are indepen-
dent of paging (they reside in “common” program
address space). This paged memory example is
quite common and supported by many 8051 soft-
ware compilers.
8032 Data Address Space (XDATA). Four sec-
tors of Secondary Flash memory reside in the up-
per half of 8032 XDATA space in the example of
Figure 53. SRAM and csiop registers are in the
lower half of XDATA space. The 8032 SFR regis-
ters and local SRAM inside the 8032 MCU Module
do not reside in XDATA space, so it is OK to place
PSD Module SRAM or csiop registers at an ad-
dress that overlaps the address of internal 8032
MCU Module SRAM and registers.
grammable security bit in the PSD Module pro-
tects its contents from unauthorized viewing and
copying. The security bit is specified in PSDsoft
Express and programmed into the uPSD33xx with
JTAG. Once set, the security bit will block access
of JTAG programming equipment to the PSD Mod-
ule Flash memory and PLD configuration, and also
blocks JTAG debugging access to the MCU Mod-
ule. The only way to defeat the security bit is to
erase the entire PSD Module using JTAG (the
erase command is the only JTAG command al-
lowed after the security bit has been set), after
which the device is blank and may be used again.
Additionally and independently, the contents of
each individual Flash memory sector can be write
protected (sector protection) by configuration with
PSDsoft Express. This is typically used to protect
8032 boot code from being corrupted by inadvert-
ent WRITEs to Flash memory from the 8032.
Status of sector protection bits may be read (but
not written) using two registers in csiop space.
Figure 53. Typical System Memory Map
Memory Mapping
There many different ways to place (or map) the
address range of PSD Module memory and I/O
depending on system requirements. The DPLD
provides complete mapping flexibility. Figure 53
shows one possible system memory map. In this
example, 128K bytes of Main Flash memory for a
uPSD3333 device is in 8032 program address
space, and 32K bytes of Secondary Flash memo-
ry, the SRAM, and csiop registers are all in 8032
XDATA space.
In Figure 53, the nomenclature fs0..fs7 are desig-
nators for the individual sectors of Main Flash
memory, 16K bytes each. CSBOOT0..CSBOOT3
are designators for the individual Secondary Flash
memory segments, 8K bytes each. rs0 is the des-
ignator for SRAM, and csiop designates the PSD
Module control register set.
8032 XDATA
SPACE
8032 PROGRAM SPACE
(PSEN)
(RD and WR)
Page X
Page 0
Page 1 Page 2
FFFFh
E000h
C000h
A000h
8000h
FFFFh
csboot3
8KB
fs3
fs7
fs5
16KB
16KB
16KB
csboot2
8KB
C000h
8000h
csboot1
8KB
fs2
fs6
fs4
16KB
16KB
16KB
csboot0
8KB
fs1, 16KB
Common Memory to All Pages
System
I/O
4000h
0000h
csiop
256B
2000h
0000h
fs0, 16KB
Common Memory to All Pages
The designer may easily specify memory mapping
in a point-and-click software environment using
PSDsoft Express, creating a non-volatile configu-
ration when the DPLD is programmed using
JTAG.
rs0, 8KB
AI09173
138/231
uPSD33xx
Specifying the Memory Map with PSDsoft Ex-
press. The memory map example shown in Fig-
ure 53., page 138 is implemented using PSDsoft
Express in a point-and-click environment. PSDsoft
Express will automatically generate Hardware
Definition Language (HDL) statements of the
ABEL language for the DPLD, such as those
shown in Table 77.
Specifying these equations using PSDsoft Ex-
press is very simple. For example, Figure 54, page
84 shows how to specify the chip-select equation
for the 16K byte Flash memory segment, fs4. No-
tice fs4 is on memory page 1. This specification
process is repeated for all other Flash memory
segments, the SRAM, the csiop register block, and
any external chip select signals that may be need-
ed.
Table 77. HDL Statement Example Generated from PSDsoft Express for Memory Map
rs0 = ((address ≥ ^h0000)& (address ≤ ^h1FFF));
csiop = ((address ≥ ^h2000)& (address ≤ ^h20FF));
fs0 = ((address ≥ ^h0000)& (address ≤ ^h3FFF));
fs1 = ((address ≥ ^h4000)& (address ≤ ^h7FFF));
fs2 = ((page == 0)
fs3 = ((page == 0)
fs4 = ((page == 1)
fs5 = ((page == 1)
fs6 = ((page == 2)
fs7 = ((page == 2)
& (address ≥ ^h8000) & (address ≤ ^hBFFF));
& (address ≥ ^hC000) & (address ≤ ^hFFFF));
& (address ≥ ^h8000) & (address ≤ ^hBFFF));
& (address ≥ ^hC000) & (address ≤ ^hFFFF));
& (address ≥ ^h8000) & (address ≤ ^hBFFF));
& (address ≥ ^hC000) & (address ≤ ^hFFFF));
csboot0 = ((address ≥ ^h8000)& (address ≤ ^h9FFF));
csboot1 = ((address ≥ ^hA000)& (address ≤ ^hBFFF));
csboot2 = ((address ≥ ^hC000)& (address ≤ ^hDFFF));
csboot3 = ((address ≥ ^hE000)& (address ≤ ^hFFFF));
Figure 54. PSDsoft Express Memory Mapping
139/231
uPSD33xx
EEPROM Emulation. EEPROM emulation is
needed if it is desired to repeatedly change only a
small number of bytes of data in Flash memory. In
this case EEPROM emulation is needed because
although Flash memory can be written byte-by-
byte, it must be erased sector-by-sector, it is not
erasable byte-by-byte (unlike EEPROM which is
written AND erased byte-by-byte). So changing
one or two bytes in Flash memory typically re-
quires erasing an entire sector each time only one
byte is changed within that sector.
–
Figure 56. Place both the Main and
Secondary Flash memories into program
space for maximum code storage, with no
Flash memory in XDATA space.
Figure 55. Mapping: Split Second Flash in Half
8032 XDATA SPACE
(RD and WR)
8032 PROGRAM
SPACE (PSEN)
Page Page Page Page
Page X
FFFFh
0
1
2
3
FFFFh
However, two of the 8K byte sectors of Secondary
Flash memory may be used to emulate EEPROM
by using a linked-list software technique to create
a small data set that is maintained by alternating
between the two flash sectors. For example, a
data set of 128 bytes is written and maintained by
software in a distributed fashion across one 8K
byte sector of Secondary Flash memory until it be-
comes full. Then the writing continues on the other
8K byte sector while erasing the first 8K byte sec-
tor. This process repeats continuously, bouncing
back and forth between the two 8K byte sectors.
This creates a wear-leveling effect, which increas-
es the effective number of erase cycles for a data
set of 128 bytes to many times more than the base
100K erase cycles of the Flash memory. EEPROM
emulation in Flash memory is typically faster than
writing to actual EEPROM memory, and more reli-
able because the last known value in a data set is
maintained even if a WRITE cycle is corrupted by
a power outage. The EEPROM emulation function
can be called by the firmware, making it appear
that the user is writing a single byte, or data
record, thus hiding all of the data management
that occurs within the two 8K byte flash sectors.
EEPROM emulation firmware for the uPSD33xx is
available from www.st.com/psm.
fs1
16KB
fs5
16KB
fs7
16KB
fs3
16KB
C000h
8000h
System I/O
fs0
16KB
fs4
fs6
fs2
16KB
16KB 16KB
8000h
csboot3
8KB
6000h
Nothing Mapped
csboot2
8KB
4000h
2100h
2000h
4000h
2000h
0000h
System I/O
csiop, 256B
csboot1, 8KB
Common Memory to All Pages
csboot0, 8KB
rs0, 8KB
Common Memory to All Pages
0000h
AI09174
Figure 56. Mapping: All Flash in Code Space
8032 XDATA SPACE
(RD and WR)
8032 PROGRAM
SPACE (PSEN)
Page Page Page Page
0
Page X
FFFFh
1
2
3
FFFFh
C000h
fs1
16KB
fs5
16KB
fs7
16KB
fs3
16KB
Alternative Mapping Schemes. Here are more
possible memory maps for the uPSD3333.
Note: Mapping examples would be slightly differ-
ent for uPSD3312, uPSD3334, and uPSD3354
because of the different sizes of individual Flash
memory sectors and SRAM as defined in Table
82., page 155.
fs0
16KB
fs4
fs6
fs2
16KB
16KB 16KB
System I/O
8000h
6000h
4000h
2000h
csboot3, 8KB
Common Memory to All Pages
–
Figure 55. Place the larger Main Flash
Memory into program space, but split the
Secondary Flash in half, placing two of it’s
sectors into XDATA space and remaining two
sectors into program space. This method
allows the designer to put IAP code (or boot
code) into two sectors of Secondary Flash in
program space, and use the other two
csboot2, 8KB
Common Memory to All Pages
csboot1, 8KB
2100h
Common Memory to All Pages
csiop, 256B
2000h
csboot0, 8KB
rs0, 8KB
Common Memory to All Pages
0000h
0000h
AI09175
Secondary Flash sectors for data storage,
such as EEPROM emulation in XDATA space.
140/231
uPSD33xx
–
Figure 57. Place the larger Main Flash
Memory into XDATA space and the smaller
Secondary Flash into program space for
systems that need a large amount of Flash for
data recording or large look-up tables, and not
so much Flash for 8032 firmware.
ly “reclassify” the Main Flash memory into XDATA
space to erase and rewrite it while executing IAP
code from the Secondary Flash memory in pro-
gram space. After the writing is complete, the Main
Flash can be “reclassified” back to program space,
then execution can continue from the new code in
Main Flash memory. The mapping example of Fig-
ure 57 will accommodate this operation.
Figure 57. Mapping: Small Code / Big Data
Memory Sector Select Rules. When
defining
8032 XDATA SPACE
(RD and WR)
8032 PROGRAM
SPACE (PSEN)
sector select signals (FSx, CSBOOTx, RS0,
CSIOP, PSELx) in PSDsoft Express, keep these
rules in mind:
Page Page Page Page
Page X
0
1
2
3
FFFFh
FFFFh
–
Main Flash and Secondary Flash memory
sector select signals may not be larger than
their physical sector size as defined in Table
75., page 135.
fs1
fs5
fs7
fs3
16KB
16KB
16KB
16KB
Nothing
Mapped
C000h
8000h
–
Any Main Flash memory sector select may not
be mapped in the same address range as
another Main Flash sector select (cannot
overlap segments of Main Flash on top of
each other).
fs0
16KB
fs4
fs6
fs2
16KB
16KB 16KB
8000h
6000h
4000h
csboot3
8KB
System I/O
–
Any Secondary Flash memory sector select
may not be mapped in the same address
range as another Secondary Flash sector
select (cannot overlap segments of
csboot2
8KB
2100h
2000h
csiop, 256 bytes,
Common to All Pages
csboot1
8KB
2000h
0000h
Secondary Flash on top of each other).
rs0, 8KB
csboot0
Common Memory to All Pages
8KB
–
–
A Secondary Flash memory sector may
overlap a Main Flash memory sector. In the
case of overlap, priority is given to the
Secondary Flash memory sector.
SRAM, CSIOP, or PSELx may overlap any
Flash memory sector. In the case of overlap,
priority is given to SRAM, CSIOP, or PSELx.
0000h
AI09176
It is also possible to “reclassify” the Flash memo-
ries during runtime, moving the memories be-
tween XDATA memory space and program
memory space on-the-fly. This essentially means
that the user can override the initial setting during
run-time by writing to a csiop register (the VM Reg-
ister). This is useful for IAP, because standard
8051 architecture does not allow writing to pro-
gram space. For example, if the user wants to up-
date firmware in Main Flash memory that is
residing in program space, the user can temporari-
Note: PSELx is for optional Peripheral I/O
Mode on Port A.
–
The address range for sector selects for
SRAM, PSELx, and CSIOP must not overlap
each other as they have the same priority,
causing contention if overlapped.
141/231
uPSD33xx
Figure 58 illustrates the priority scheme of the
memory elements of the PSD Module. Priority re-
fers to which memory will ultimately produce a
byte of data or code to the 8032 MCU for a given
bus cycle. Any memory on a higher level can over-
lap and has priority over any memory on a lower
level. Memories on the same level must not over-
lap.
Example: FS0 is valid when the 8032 produces an
address in the range of 8000h to BFFFh.
CSBOOT0 is valid from 8000h to 9FFFh. RS0 is
valid from 8000h to 87FFh. Any address from the
8032 in the range of RS0 always accesses the
SRAM. Any address in the range of CSBOOT0
greater than 87FFh (and less than 9FFFh) auto-
matically addresses Secondary Flash memory.
Any address greater than 9FFFh accesses Main
Flash memory. One-half of the Main Flash memo-
ry segment, and one-fourth of the Secondary
Flash memory segment cannot be accessed by
the 8032 in this example.
The VM Register. One of the csiop registers (the
VM Register) controls whether or not the 8032 bus
control signals RD, WR, and PSEN are routed to
the Main Flash memory, the Secondary Flash
memory, or the SRAM. Routing of these signals to
these PSM Module memories determines if mem-
ories reside in 8032 program address space, 8032
XDATA space, or both. The initial setting of the VM
Register is determined by a choice in PSDsoft Ex-
press and programmed into the uPSD33xx in a
non-volatile fashion using JTAG. This initial setting
is loaded into the VM Register upon power-up and
also loaded upon any reset event. However, the
8032 may override the initial VM Register setting
at run-time by writing to the VM Register, which is
useful for IAP.
Table 78., page 143 defines bit functions within
the VM Register.
Note: Bit 7, PIO_EN, is not related to the memory
manipulation functions of Bits 0, 1, 2, 3, and 4.
Also note that SRAM must at least always be in
8032 XDATA space (default condition). Bit 0 al-
lows the user to optionally place SRAM into 8032
program space in addition to XDATA space.
CSIOP registers are always in XDATA space and
cannot reside in program space.
Figure 58. PSD Module Memory Priority
Highest Priority
Figure 59., page 144 illustrates how the VM Reg-
ister affects the routing of RD, WR, and PSEN to
the memories on the PSD Module. As an example,
if we apply the value 0Ch to the VM Register to im-
plement the memory map example shown in Fig-
ure 53., page 138, then the routing of RD, WR,
and PSEN would look like that shown in Figure
60., page 145.
Level 1
SRAM,
CSIOP, and
Peripheral I/O
Mode
Level 2
Secondary
Flash Memory
In this example, the configuration is specified in
PSDsoft Express and programmed into the
uPSD33xx using JTAG. Upon power-on or any re-
set condition, the non-volatile value 0Ch is loaded
into the VM Register. At runtime, the value 0Ch in
the VM Register may be changed (overridden) by
the 8032 if desired to implement IAP or other func-
tions.
Level 3
Main Flash Memory
Lowest Priority
AI02867E
142/231
uPSD33xx
Table 78. VM Register (address = csiop + offset E2h)
Bit 1
Secondary
Flash
Program
Space
Bit 4
Main Flash
XDATA
Bit 3
Bit 2
Main Flash
Program
Space
Bit 0
SRAM
Program
Space
Bit 7
PIO_EN
Secondary
FlashXDATA
Space
Bit 6
Bit 5
Space
0 = RD or WR
cannot
access
Secondary
0 = PSEN
cannot
access
0 = RDor WR
cannot
access Main
Flash
0 = PSEN
cannot
access Main
Flash
0 = PSEN
cannot
access
SRAM
0 = disable
Peripheral I/O not used not used
Mode on Port A
Secondary
Flash
Flash
1 = RD or WR
can access
Secondary
Flash
1 = PSEN
can access
Secondary
Flash
1 = enable
Peripheral I/O not used not used
Mode on Port A
1 = RDor WR
can access
Main Flash
1 = PSEN
can access
Main Flash
1 = PSEN
can access
SRAM
Note: 1. Default value of Bits 0, 1, 2, 3, and 4 is loaded from Non-Volatile setting as specified from PSDsoft Express upon any reset or power-
up condition. The default value of these bits can be overridden by 8032 at run-time.
2. Default value of Bit 7 is zero upon any reset condition.
143/231
uPSD33xx
Figure 59. VM Register Control of Memories
CS
CS
RS0
8032 Address
Secondary
Flash
Memory
CSBOOT0 - CSBOOT3
FS0 - FS7
Main Flash
Memory
SRAM
DPLD
53 Other PLD Inputs
CS
WR
OE
WR
OE
WR
OE
WR
WR
VM REG BIT 4
VM REG BIT 3
RD
VM REG BIT 2
RD
VM REG BIT 1
VM REG BIT 0
PSEN
AI02870D
144/231
uPSD33xx
Figure 60. VM Register Example Corresponding to Memory Map Example of Figure 33
CS
CS
RS0
8032 Address
Secondary
Flash
Memory
CSBOOT0 - CSBOOT3
FS0 - FS7
Main Flash
Memory
CS
DPLD
SRAM
53 Other PLD Inputs
WR
OE
WR
OE
WR
OE
VM Register = 0Ch
PSEN
WR
RD
AI02869D
Runtime Control Register Definitions (csiop)
The 39 csiop registers are defined in Table 79.
The 8032 can access each register by the address
offset (specified in Table 79) added to the csiop
base address that was specified in PSDsoft Ex-
press. Do not write to unused locations within the
csiop block of 256 registers, they should remain
logic zero.
Table 79. CSIOP Registers and their Offsets (in hexadecimal)
Register
Name
Port A
(80-pin)
Port B Port C Port D Other
Description
Link
MCU I/O input mode. Read to obtain
Table
Data In
00h
02h
01h
03h
10h
11h
current logic level of pins on Ports A, B, 95., page
C, or D. No WRITEs.
172
Selects MCUI/O or Latched Address
Out mode. Logic 0 = MCU I/O, 1 = 8032
Addr Out. Write to select mode. Read for
status.
Table
107., page
177
Control
MCU I/O output mode. Write to set logic
level on pins of Ports A, B, C, or D. Read
to check status. This register has no
effect if a port pin is driven by an OMC
output from PLD.
Table
99., page
172
Data Out
04h
06h
08h
05h
07h
09h
12h
14h
16h
13h
15h
17h
MCU I/O mode. Configures port pin as
input or output. Write to set direction of
port pins.
Logic 1 = out, Logic 0 = in. Read to
check status.
Table
103., page
173
Direction
Write to configure port pins as either
CMOS push-pull or Open Drain on some
pins, while selecting high slew rate on
other pins. Read to check status. Default
output type is CMOS push-pull.
Table
109., page
179
Drive Select
145/231
uPSD33xx
Register
Name
Port A
(80-pin)
Port B Port C Port D Other
Description
Link
Table
90., page
167
Input
Macrocells
Read to obtain logic state of IMCs. No
WRITEs.
0Ah
0Bh
0Dh
18h
1Ah
Read state of output enable logic on
each I/O port driver. 1 = driver output is
enabled, 0 = driver is off, and it is in high
impedance state. No WRITEs.
Table
113., page
180
Enable Out
OCh
1Bh
Output
Macrocells AB
(MCELLAB)
Read logic state of MCELLAB outputs
20h (bank of eight OMCs).
Table
86., page
165
Write to load MCELLAB flip-flops.
Output
Macrocells BC
(MCELLBC)
Read logic state of MCELLBC outputs
21h (bank of eight OMCs).
Table
87., page
165
Write to load MCELLBC flip-flops.
Write to set mask for MCELLAB. Logic
'1' blocks READs/WRITEs of OMC.
Logic '0' will pass OMC value. Read to
check status.
Table
88., page
166
Mask
Macrocells AB
22h
Write to set mask for MCELLBC. Logic
'1' blocks READs/WRITEs of OMC.
Logic '0' will pass OMC value. Read to
check status.
Table
89., page
166
Mask
Macrocells BC
23h
Read to determine Main Flash Sector
Protection Setting (non-volatile) that was
specified in PSDsoft Express. No
WRITEs.
Main Flash
Sector
Protection
Table
82., page
155
C0h
Read to determine if PSD Module
device Security Bit is active (non-
volatile) Logic 1 = device secured. Also
read to determine Secondary Flash
Protection Setting (non-volatile) that was
Security Bit
andSecondary
Flash Sector
Protection
Table
83., page
155
C2h
specified in PSDsoft. No WRITEs.
Table
117., page
188
Power Management Register 0. WRITE
and READ.
PMMR0
PMMR2
PMMR3
Page
B0h
Table
118., page
188
Power Management Register 2. WRITE
and READ.
B4h
Power Management Register 3. WRITE
C7h and READ. However, Bit 1 can be
cleared only by a reset condition.
Table
119., page
188
Figure
52., page
135
Memory Page Register. WRITE and
READ.
E0h
Places PSD Module memories into 8032
Program Address Space and/or 8032
XDATA Address Space. (VM overrides
initial non-volatile setting that was
specified in PSDsoft Express. Reset
restores initial setting)
Table
78., page
143
VM (Virtual
Memory)
E2h
146/231
uPSD33xx
PSD Module Detailed Operation
Specific details are given here for the following key
functional areas on the PSD Module:
Flash Memory Instruction Sequences. An in-
struction sequence consists of a sequence of spe-
cific byte WRITE and byte READ operations. Each
byte written to either Flash memory array on the
PSD Module is received by a state machine inside
the Flash array and sequentially decoded to exe-
cute an embedded algorithm. The algorithm is ex-
ecuted when the correct number of bytes are
properly received and the time between two con-
secutive bytes is shorter than the time-out period
of 80µs. Some instruction sequences are struc-
tured to include READ operations after the initial
WRITE operations.
■
■
■
■
■
Flash Memories
PLDs (DPLD and GPLD)
I/O Ports
Power Management
JTAG ISP and Debug Interface
Flash Memory Operation. The Flash memories
are accessed through the 8032 Address, Data,
and Control Bus interfaces. Flash memories (and
SRAM) cannot be accessed by any other bus
master other than the 8032 MCU (these are not
dual-port memories).
An instruction sequence must be followed exactly.
Any invalid combination of instruction bytes or
time-out between two consecutive bytes while ad-
dressing Flash memory resets the PSD Module
Flash logic into Read Array mode (where Flash
memory is read like a ROM device). The Flash
memories support instruction sequences summa-
rized in Table 80., page 148.
The 8032 cannot write to Flash memory as it
would an SRAM (supply address, supply data,
supply WR strobe, assume the data was correctly
written to memory). Flash memory must first be
“unlocked” with a special instruction sequence of
byte WRITE operations to invoke an internal algo-
rithm inside either Flash memory array, then a sin-
gle data byte is written (programmed) to the Flash
memory array, then programming status is
checked by a byte READ operation or by checking
the Ready/Busy pin (PC3). Table 80., page 148
lists all of the special instruction sequences to pro-
gram a byte to either of the Flash memory arrays,
erase the arrays, and check for different types of
status from the arrays.
■
■
■
■
■
Program a Byte
Unlock Sequence Bypass
Erase memory by array or by sector
Suspend or resume a sector erase
Reset to Read Array mode
The first two bytes of an instruction sequence are
8032 bus WRITE operations to “unlock” the Flash
array, followed by writing a command byte. The
bus operations consist of writing the data AAh to
address X555h during the first bus cycle and data
55h to address XAAAh during the second bus cy-
cle. 8032 address signals A12-A15 are “Don’t
care” during the instruction sequence during
WRITE cycles. However, the appropriate sector
select signal (FSx or CSBOOTx) from the DPLD
must be active during the entire instruction se-
quence to complete the entire 8032 address (this
includes the page number when memory paging is
used). Ignoring A12-A15 means the user has more
flexibility in memory mapping. For example, in
many traditional Flash memories, instruction se-
quences must be written to addresses AAAAh and
5555h, not XAAAh and X555h like supported on
the PSD Module. When AAAAh and 5555h must
be written to, the memory mapping options are lim-
ited.
This unlocking sequence is typical for many Flash
memories to prevent accidental WRITEs by errant
code. However, it is possible to bypass this un-
locking sequence to save time while intentionally
programming Flash memory.
IMPORTANT: The 8032 may not read and exe-
cute code from the same Flash memory array for
which it is directing an instruction sequence. Or
more simply stated, the 8032 may not read code
from the same Flash array that is writing or eras-
ing. Instead, the 8032 must execute code from an
alternate memory (like SRAM or a different Flash
array) while sending instruction sequences to a
given Flash array. Since the two Flash memory ar-
rays inside the PSD Module device are completely
independent, the 8032 may read code from one
array while sending instructions to the other. It is
possible, however, to suspend a sector erase op-
eration in one particular Flash array in order to ac-
cess a different sector within that same Flash
array, then resume the erase later.
The Main Flash and Secondary Flash memories
each have the same instruction set shown in Table
80., page 148, but the sector select signals deter-
mine which memory array will receive and execute
the instructions.
After a Flash memory array is programmed or
erased it will go to “Read Array” mode, then the
8032 can read from Flash memory just as it would
read from any 8-bit ROM or SRAM device.
147/231
uPSD33xx
(1,2)
Table 80. Flash Memory Instruction Sequences
Instr.
Sequence
Bus
Cycle 1
Bus
Cycle 2
Bus
Cycle 3
Bus
Cycle 4
Bus
Cycle 5
Bus
Cycle 6
Bus
Cycle 7
Link
Read
Read byte
from any
valid Flash
memory
addr
Memory
Contents
(Read
Array
mode)
Read
Memory
Contents., p
age 149
Program
(write) a
Byte to
Flash
Write A0h Write
Programmin
g Flash
Memory., pa
Write AAh
to X555h
(unlock)
Write 55h
to XAAAh
(unlock)
to X555h
data byte
(command to
)
address
ge 150
Memory
Write 20h
to X555h
(command
)
Bypassed
Unlock
Sequence, p
Write AAh
to X555h
(unlock)
Write 55h
to XAAAh
(unlock)
Bypass
Unlock
age 153
Program a
Byte to
Flash
Memory
with
Bypassed
Unlock
Sequence, p
Write A0h to Write data
XXXXh byte to
(command) address
age 153
Bypassed
Unlock
Write 00h
Write 90h to
to XXXXh
(command
(command)
)
Bypassed
Unlock
Sequence, p
age 153
Reset
Bypass
Unlock
XXXXh
Write 80h Write
Write AAh
to X555h
(unlock)
Write 55h
to XAAAh
(unlock)
Write 55h Write 10h
to XAAAh to X555h
Flash Bulk
Erase., page
153
Flash Bulk
to X555h
(command X555h
(unlock)
AAh to
(3)
Erase
(unlock)
(command)
)
Write 80h Write
to X555h AAh to
(command X555h
(unlock)
Write 30h
to desired
Sector
Write 30h
to another
Sector
Flash
Sector
Erase
Write AAh
to X555h
(unlock)
Write 55h
to XAAAh
(unlock)
Write 55h
to XAAAh
(unlock)
Flash Sector
Erase., page
154
)
(command)
(command)
Write B0h to
address that
activates
FSx or
CSBOOTx
where erase
is in
Suspend
Sector
Erase., page
154
Suspend
Sector
Erase
progress
(command)
Write 30h to
address that
activates
FSx or
CSBOOTx
where
Resume
Sector
Erase., page
Resume
Sector
Erase
desired to
resume
154
erase
(command)
148/231
uPSD33xx
Instr.
Sequence
Bus
Cycle 1
Bus
Cycle 2
Bus
Cycle 3
Bus
Cycle 4
Bus
Cycle 5
Bus
Cycle 6
Bus
Cycle 7
Link
Write F0h to
address that
activates
FSx or
CSBOOTx
in desired
array.
Reset
Flash, page
154
Reset
Flash
(command)
Note: 1. All values are in hexadecimal, X = Don’t care
2. 8032 addresses A12 through A15 are “Don’t care” during the instruction sequence decoding. Only address bits A0-A11 are used
during decoding of Flash memory instruction sequences. The individual sector select signal (FS0 - FS7 or CSBOOT0-CSBOOT3)
which is active during the instruction sequence determines the complete address.
3. Directing this command to any individual sector within a Flash memory array will invoke the bulk erase of all Flash memory sectors
within that array.
Reading Flash Memory. Under typical condi-
tions, the 8032 may read the Flash memory using
READ operations (READ bus cycles) just as it
would a ROM or RAM device. Alternately, the
8032 may use READ operations to obtain status
information about a Program or Erase operation
that is currently in progress. The following sections
describe the kinds of READ operations.
operation, DQ7 is '0.' After the erase is complete
DQ7 is '1.' The correct select signal, FSx or CS-
BOOTx, must be active during the entire polling
procedure.
DQ7 is valid after the fourth instruction byte
WRITE operation (for program instruction se-
quence) or after the sixth instruction byte WRITE
operation (for erase instruction sequence).
Read Memory Contents. Flash
memory
is
If all Flash memory sectors to be erased are pro-
tected, DQ7 is reset to ’0’ for about 100µs, and
then DQ7 returns to the value of D7 of the previ-
ously addressed byte. No erasure is performed.
placed in the Read Array mode after Power-up, af-
ter a PSD Module reset event, or after receiving a
Reset Flash memory instruction sequence from
the 8032. The 8032 can read Flash memory con-
tents using standard READ bus cycles anytime the
Flash array is in Read Array mode. Flash memo-
ries will always be in Read Array mode when the
array is not actively engaged in a program or erase
operation.
Reading the Erase/Program Status Bits. The
Flash arrays provide several status bits to be used
by the 8032 to confirm the completion of an erase
or program operation on Flash memory, shown in
Table 81., page 150. The status bits can be read
as many times as needed until an operation is
complete.
The 8032 performs a READ operation to obtain
these status bits while an erase or program oper-
ation is being executed by the state machine in-
side each Flash memory array.
Data Polling Flag (DQ7). While programming ei-
ther Flash memory, the 8032 may read the Data
Polling Flag Bit (DQ7), which outputs the comple-
ment of the D7 Bit of the byte being programmed
into Flash memory. Once the program operation is
complete, DQ7 is equal to D7 of the byte just pro-
grammed into Flash memory, indicating the pro-
gram cycle has completed successfully. The
correct select signal, FSx or CSBOOTx, must be
active during the entire polling procedure.
Toggle Flag (DQ6). The Flash memories offer an
alternate way to determine when a Flash memory
program operation has completed. During the pro-
gram operation and while the correct sector select
FSx or CSBOOTx is active, the Toggle Flag Bit
(DQ6) toggles from '0' to '1' and '1' to ’0’ on subse-
quent attempts to read any byte of the same Flash
array.
When the internal program operation is complete,
the toggling stops and the data read on the data
bus D0-7 is the actual value of the addressed
memory byte. The device is now accessible for a
new READ or WRITE operation. The operation is
finished when two successive READs yield the
same value for DQ6.
DQ6 may also be used to indicate when an erase
operation has completed. During an erase opera-
tion, DQ6 will toggle from '0' to '1' and '1' to ’0’ until
the erase operation is complete, then DQ6 stops
toggling. The erase is finished when two succes-
sive READs yield the same value of DQ6. The cor-
rect sector select signal, FSx or CSBOOTx, must
be active during the entire procedure.
DQ6 is valid after the fourth instruction byte
WRITE operation (for program instruction se-
quence) or after the sixth instruction byte WRITE
operation (for erase instruction sequence).
Polling may also be used to indicate when an
erase operation has completed. During an erase
149/231
uPSD33xx
If all the Flash memory sectors selected for era-
sure are protected, DQ6 toggles to ’0’ for about
100µs, then returns value of D6 of the previously
addressed byte.
Error Flag (DQ5). During a normal program or
erase operation, the Error Flag Bit (DQ5) is to ’0’.
This bit is set to ’1’ when there is a failure during
Flash memory byte program, sector erase, or bulk
erase operations.
In the case of Flash memory programming, DQ5
Bit indicates an attempt to program a Flash mem-
ory bit from the programmed state of 0, to the
erased state of 1, which is not valid. DQ5 may also
indicate a particular Flash cell is damaged and
cannot be programmed.
In case of an error in a Flash memory sector erase
or byte program operation, the Flash memory sec-
tor in which the error occurred or to which the pro-
grammed byte belongs must no longer be used.
Other Flash memory sectors may still be used.
DQ5 is reset after a Reset Flash instruction se-
quence.
Erase Time-out Flag (DQ3). The Erase Time-
out Flag Bit (DQ3) reflects the time-out period al-
lowed between two consecutive sector erase in-
struction sequence bytes. If multiple sector erase
commands are desired, the additional sector
erase commands (30h) must be sent by the 8032
within 80us after the previous sector erase com-
mand. DQ3 is 0 before this time period has ex-
pired, indicating it is OK to issue additional sector
erase commands. DQ3 will go to logic ’1’ if the time
has been longer than 80µs since the previous sec-
tor erase command (time has expired), indication
that is not OK to send another sector erase com-
mand. In this case, the 8032 must start a new sec-
tor erase instruction sequence (unlock and
command) beginning again after the current sec-
tor erase operation has completed.
a bit in Flash memory to a logic ’1’ once it has been
programmed to a logic '0.' A bit must be erased to
logic ’1’, and programmed to logic '0.' That means
Flash memory must be erased prior to being pro-
grammed. A byte of Flash memory is erased to all
1s (FFh). The 8032 may erase the entire Flash
memory array all at once, or erase individual sec-
tor-by-sector, but not erase byte-by-byte. Howev-
er, even though the Flash memories cannot be
erased byte-by-byte, the 8032 may program Flash
memory byte-by-byte. This means the 8032 does
not need to program group of bytes (64, 128, etc.)
at one time, like some Flash memories.
Each Flash memory requires the 8032 to send an
instruction sequence to program a byte or to erase
sectors (see Table 80., page 148).
If the byte to be programmed is in a protected
Flash memory sector, the instruction sequence is
ignored.
IMPORTANT: It is mandatory that a chip-select
signal is active for the Flash sector where a pro-
gramming instruction sequence is targeted. Make
sure that the correct chip-select equation, FSx, or
CSBOOTx specified in PSDsoft Express matches
the address range that the 8032 firmware is ac-
cessing, otherwise the instruction sequence will
not be recognized by the Flash array. If memory
paging is used, be sure that the 8032 firmware
sets the page register to the correct page number
before issuing an instruction sequence to the
Flash memory segment on a particular memory
page, otherwise the correct sector select signal
will not become active.
Once the 8032 issues a Flash memory program or
erase instruction sequence, it must check the sta-
tus bits for completion. The embedded algorithms
that are invoked inside a Flash memory array pro-
vide several ways to give status to the 8032. Sta-
tus may be checked using any of three methods:
Data Polling, Data Toggle, or Ready/Busy (pin
PC3).
Programming Flash Memory. When a byte of
Flash memory is programmed, individual bits are
programmed to logic '0.' The user cannot program
Table 81. Flash Memory Status Bit Definition
Functional Block
FSx, or CSBOOTx
DQ7
DQ6
DQ5
DQ4
DQ3
DQ2
DQ1
DQ0
Erase
Time-
out
Active (the desired
segment is selected)
Data
Polling
Toggle Error
Flag Flag
Flash Memory
X
X
X
X
Note: 1. X = Not guaranteed value, can be read either '1' or '0.'
2. DQ7-DQ0 represent the 8032 Data Bus Bits, D7-D0.
150/231
uPSD33xx
Data Polling. Polling on the Data Polling Flag Bit
(DQ7) is a method of checking whether a program
or erase operation is in progress or has complet-
ed. Figure 61 shows the Data Polling algorithm.
PSDsoft Express generates ANSI C code func-
tions for implementation of these Data Polling al-
gorithms.
When the 8032 issues a program instruction se-
quence, the embedded algorithm within the Flash
memory array begins. The 8032 then reads the lo-
cation of the byte to be programmed in Flash
memory to check status. The Data Polling Flag Bit
(DQ7) of this location becomes the compliment of
Bit D7 of the original data byte to be programmed.
The 8032 continues to poll this location, compar-
ing the Data Polling Flag Bit (DQ7) and monitoring
the Error Flag Bit (DQ5). When the Data Polling
Flag Bit (DQ7) matches Bit D7 of the original data,
then the embedded algorithm is complete. If the
Error Flag Bit (DQ5) is '1,' the 8032 should test the
Data Polling Flag Bit (DQ7) again since the Data
Polling Flag Bit (DQ7) may have changed simulta-
neously with the Error Flag Bit (DQ5) (see Figure
61).
Figure 61. Data Polling Flowchart
START
READ DQ5 & DQ7
at VALID ADDRESS
DQ7
=
DATA
YES
NO
The Error Flag Bit (DQ5) is set if either an internal
time-out occurred while the embedded algorithm
attempted to program the byte (indicating a bad
Flash cell) or if the 8032 attempted to program bit
to logic ’1’ when that bit was already programmed
to logic ’0’ (must erase to achieve logic ’1’).
NO
DQ5
= 1
YES
READ DQ7
It is suggested (as with all Flash memories) to read
the location again after the embedded program-
ming algorithm has completed, to compare the
byte that was written to the Flash memory with the
byte that was intended to be written.
DQ7
=
YES
DATA
NO
When using the Data Polling method during an
erase operation, Figure 61 still applies. However,
the Data Polling Flag Bit (DQ7) is '0' until the erase
operation is complete. A ’1’ on the Error Flag Bit
(DQ5) indicates a time-out condition on the Erase
cycle, a ’0’ indicates no error. The 8032 can read
any location within the sector being erased to get
the Data Polling Flag Bit (DQ7) and the Error Flag
Bit (DQ5).
FAIL
PASS
AI01369B
151/231
uPSD33xx
Data Toggle. Checking the Toggle Flag Bit
(DQ6) is another method of determining whether a
program or erase operation is in progress or has
completed. Figure 62 shows the Data Toggle algo-
rithm.
PSDsoft Express generates ANSI C code func-
tions for implementation of these Data Toggling al-
gorithms.
Figure 62. Data Toggle Flowchart
When the 8032 issues a program instruction se-
quence, the embedded algorithm within the Flash
memory array begins. The 8032 then reads the lo-
cation of the byte to be programmed in Flash
memory to check status. The Toggle Flag Bit
(DQ6) of this location toggles each time the 8032
reads this location until the embedded algorithm is
complete. The 8032 continues to read this loca-
tion, checking the Toggle Flag Bit (DQ6) and mon-
itoring the Error Flag Bit (DQ5). When the Toggle
Flag Bit (DQ6) stops toggling (two consecutive
reads yield the same value), then the embedded
algorithm is complete. If the Error Flag Bit (DQ5) is
'1,' the 8032 should test the Toggle Flag Bit (DQ6)
again, since the Toggle Flag Bit (DQ6) may have
changed simultaneously with the Error Flag Bit
(DQ5) (see Figure 62).
START
READ
DQ5 & DQ6
DQ6
NO
=
TOGGLE
YES
NO
DQ5
= 1
The Error Flag Bit (DQ5) is set if either an internal
time-out occurred while the embedded algorithm
attempted to program the byte, or if the 8032 at-
tempted to program bit to logic ’1’ when that bit
was already programmed to logic ’0’ (must erase
to achieve logic ’1’).
YES
READ DQ6
It is suggested (as with all Flash memories) to read
the location again after the embedded program-
ming algorithm has completed, to compare the
byte that was written to Flash memory with the
byte that was intended to be written.
DQ6
=
NO
TOGGLE
YES
FAIL
PASS
When using the Data Toggle method during an
erase operation, Figure 62 still applies. the Toggle
Flag Bit (DQ6) toggles until the erase operation is
complete. A ’1’ on the Error Flag Bit (DQ5) indi-
cates a time-out condition on the Erase cycle, a ’0’
indicates no error. The 8032 can read any location
within the sector being erased to get the Toggle
Flag Bit (DQ6) and the Error Flag Bit (DQ5).
AI01370B
152/231
uPSD33xx
Ready/Busy (PC3). This signal can be used to
output the Ready/Busy status of a program or
erase operation on either Flash memory. The out-
put on the Ready/Busy pin is a ’0’ (Busy) when ei-
ther Flash memory array is being written, or when
either Flash memory array is being erased. The
output is a ’1’ (Ready) when no program or erase
operation is in progress. To activate this function
on this pin, the user must select the “Ready/Busy”
selection in PSDsoft Express when configuring pin
PC3. This pin may be polled by the 8032 or used
as a 8032 interrupt to indicate when an erase or
program operation is complete (requires routing
the signal on PC board from PC3 back into a pin
on the MCU Module). This signal is also available
internally on the PSD Module as an input to both
PLDs (without routing a signal externally on PC
board) and it’s signal name is “rd_bsy”. The
Ready/Busy output can be probed during lab de-
velopment to check the timing of Flash memory
programming in the system at run-time.
Bypassed Unlock Sequence. The Bypass Un-
lock mode allows the 8032 to program bytes in the
Flash memories faster than using the standard
Flash program instruction sequences because the
typical AAh, 55h unlock bus cycles are bypassed
for each byte that is programmed. Bypassing the
unlock sequence is typically used when the 8032
is intentionally programming a large number of
bytes (such as during IAP). After intentional pro-
gramming is complete, typically the Bypass mode
would be disabled, and full protection is back in
place to prevent unwanted WRITEs to Flash mem-
ory.
is checked using toggle, polling, or Ready/Busy
just as before. Additional data bytes are pro-
grammed the same way until this Bypass Unlock
mode is exited.
To exit Bypass Unlock mode, the system must is-
sue the Reset Bypass Unlock instruction se-
quence. The first bus cycle of this instruction must
write 90h to any valid address within the unlocked
Flash Array; the second bus cycle must write 00h
to any valid address within the unlocked Flash Ar-
ray. After this sequence the Flash returns to Read
Array mode.
During Bypass Unlock Mode, only the Bypassed
Unlock Program instruction, or the Reset Bypass
Unlock instruction is valid, other instruction will be
ignored.
Erasing Flash Memory. Flash memory may be
erased sector-by-sector, or an entire Flash memo-
ry array may be erased with one command (bulk).
Flash Bulk Erase. The Flash Bulk Erase instruc-
tion sequence uses six WRITE operations fol-
lowed by a READ operation of the status register,
as described in Table 80., page 148. If any byte of
the Bulk Erase instruction sequence is wrong, the
Bulk Erase instruction sequence aborts and the
device is reset to the Read Array mode. The ad-
dress provided by the 8032 during the Flash Bulk
Erase command sequence may select any one of
the eight Flash memory sector select signals FSx
or one of the four signals CSBOOTx. An erase of
the entire Flash memory array will occur in a par-
ticular array even though a command was sent to
just one of the individual Flash memory sectors
within that array.
The Bypass Unlock mode is entered by first initiat-
ing two Unlock bus cycles. This is followed by a
third WRITE operation containing the Bypass Un-
lock command, 20h (as shown in Table
80., page 148). The Flash memory array that re-
ceived that sequence then enters the Bypass Un-
lock mode. After this, a two bus cycle program
operation is all that is required to program a byte
in this mode. The first bus cycle in this shortened
program instruction sequence contains the By-
passed Unlocked Program command, A0h, to any
valid address within the unlocked Flash array. The
second bus cycle contains the address and data of
the byte to be programmed. Programming status
During a Bulk Erase, the memory status may be
checked by reading the Error Flag Bit (DQ5), the
Toggle Flag Bit (DQ6), and the Data Polling Flag
Bit (DQ7). The Error Flag Bit (DQ5) returns a ’1’ if
there has been an erase failure. Details of acquir-
ing the status of the Bulk Erase operation are de-
tailed in the section entitled “Programming Flash
Memory., page 150.
During a Bulk Erase operation, the Flash memory
does not accept any other Flash instruction se-
quences.
153/231
uPSD33xx
Flash Sector Erase. The Sector Erase instruc-
tion sequence uses six WRITE operations, as de-
scribed in Table 80., page 148. Additional Flash
Sector Erase commands to other sectors within
the same Flash array may be issued by the 8032
if the additional commands are sent within a limit-
ed amount of time.
The Erase Time-out Flag Bit (DQ3) reflects the
time-out period allowed between two consecutive
sector erase instruction sequence bytes. If multi-
ple sector erase commands are desired, the addi-
tional sector erase commands (30h) must be sent
by the 8032 to another sector within 80µs after the
previous sector erase command. DQ3 is 0 before
this time period has expired, indicating it is OK to
issue additional sector erase commands. DQ3 will
go to logic ’1’ if the time has been longer than 80µs
since the previous sector erase command (time
has expired), indicating that is not OK to send an-
other sector erase command. In this case, the
8032 must start a new sector erase instruction se-
quence (unlock and command), beginning again
after the current sector erase operation has com-
pleted.
There is up to 15µs delay after the Suspend Sector
Erase command is accepted and the array goes to
Read Array mode. The 8032 will monitor the Tog-
gle Flag Bit (DQ6) to determine when the erase
operation has halted and Read Array mode is ac-
tive.
If a Suspend Sector Erase instruction sequence
was executed, the following rules apply:
–
Attempting to read from a Flash memory
sector that was being erased outputs invalid
data.
–
–
Reading from a Flash memory sector that was
not being erased is valid.
The Flash memory cannot be programmed,
and only responds to Resume Sector Erase
and Reset Flash instruction sequences.
–
If a Reset Flash instruction sequence is
received, data in the Flash memory sector that
was being erased is invalid.
Resume Sector Erase. If a Suspend Sector
Erase instruction sequence was previously exe-
cuted, the erase cycle may be resumed with this
instruction sequence. The Resume Sector Erase
instruction sequence consists of writing the com-
mand 30h to any valid address within the Flash ar-
ray that was suspended as shown in Table
80., page 148.
Reset Flash. The Reset Flash instruction se-
quence resets the embedded algorithm running on
the state machine in the targeted Flash memory
(Main or Secondary) and the memory goes into
Read Array mode. The Reset Flash instruction
consists of one bus WRITE cycle as shown in Ta-
ble 80., page 148, and it must be executed after
any error condition that has occurred during a
Flash memory Program or Erase operation.
During a Sector Erase operation, the memory sta-
tus may be checked by reading the Error Flag Bit
(DQ5), the Toggle Flag Bit (DQ6), and the Data
Polling Flag Bit (DQ7), as detailed in Reading the
Erase/Program Status Bits, page 149.
During a Sector Erase operation, a Flash memory
accepts only Reset Flash and Suspend Sector
Erase instruction sequences. Erasure of one
Flash memory sector may be suspended, in order
to read data from another Flash memory sector,
and then resumed.
The address provided with the initial Flash Sector
Erase command sequence (Table 80., page 148)
must select the first desired sector (FSx or CS-
BOOTx) to erase. Subsequent sector erase com-
mands that are appended within the time-out
period must be addressed to other desired seg-
ments within the same Flash memory array.
Suspend Sector Erase. When a Sector Erase
operation is in progress, the Suspend Sector
Erase instruction sequence can be used to sus-
pend the operation by writing B0h to any valid ad-
dress within the Flash array that currently is
undergoing an erase operation. This allows read-
ing of data from a different Flash memory sector
within the same array after the Erase operation
has been suspended. Suspend Sector Erase is
accepted only during an Erase operation.
It may take the Flash memory up to 25µs to com-
plete the Reset cycle. The Reset Flash instruction
sequence is ignored when it is issued during a
Program or Bulk Erase operation. The Reset Flash
instruction sequence aborts any on-going Sector
Erase operation and returns the Flash memory to
Read Array mode within 25µs.
Reset Signal Applied to Flash Memory. When-
ever the PSD Module receives a reset signal from
the MCU Module, any operation that is occurring in
either Flash memory array will be aborted and the
array(s) will go to Read Array mode. It may take up
to 25µs to abort an operation and achieve Read
Array mode.
A reset from the MCU Module will result from any
of these events: an active signal on the uPSD33xx
RESET_IN input pin, a watchdog timer time-out,
detection of low V , or a JTAG debug channel re-
CC
set event.
154/231
uPSD33xx
Flash Memory Sector Protection. Each Flash
memory sector can be separately protected
against program and erase operations. This mode
can be activated (or deactivated) by selecting this
feature in PSDsoft Express and then programming
through the JTAG Port. Sector protection can be
selected for individual sectors, and the 8032 can-
not override the protection during run-time. The
8032 can read, but not change, sector protection.
Any attempt to program or erase a protected Flash
memory sector is ignored. The 8032 may read the
contents of a Flash sector even when a sector is
protected.
PSD Module Security Bit. A programmable se-
curity bit in the PSD Module protects its contents
from unauthorized viewing and copying. The secu-
rity bit is set using PSDsoft Express and pro-
grammed into the PSD Module with JTAG. When
set, the security bit will block access of JTAG pro-
gramming equipment from reading or modifying
the PSD Module Flash memory and PLD configu-
ration. The security bit also blocks JTAG access to
the MCU Module for debugging. The only way to
defeat the security bit is to erase the entire PSD
Module using JTAG (erase is the only JTAG oper-
ation allowed while security bit is set), after which
the device is blank and may be used again. The
8032 MCU will always have access to Flash mem-
ory contents through its 8-bit data bus even while
the security bit is set. The 8032 can read the status
of the security bit at run-time (but it cannot change
it) by reading the csiop register defined in Table
83.
Sector protection status is not read using Flash
memory instruction sequences, but instead this
status is read by the 8032 reading two registers
within csiop address space shown in Table 82 and
Table 83.
Flash Memory Protection During Power-Up.
Flash memory WRITE operations are automatical-
ly prevented while V
is ramping up until it rises
DD
above V
voltage threshold at which time Flash
LKO
memory WRITE operations are allowed.
Table 82. Main Flash Memory Protection Register Definition (address = csiop + offset C0h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
Sec7_Prot
Sec6_Prot
Sec5_Prot
Sec4_Prot
Sec3_Prot
Sec2_Prot
Sec1_Prot
Sec0_Prot
Note: Bit Definitions:
Sec<i>_Prot 1 = Flash memory sector <i> is write protected, 0 = Flash memory sector <i> is not write protected.
Table 83. Secondary Flash Memory Protection/Security Register Definition (csiop + offset C2h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
Security_Bit
not used
not used
not used
Sec3_Prot
Sec2_Prot
Sec1_Prot
Sec0_Prot
Note: Security_Bit = 1, device is secured, 0 = not secured
Note: Sec<i>_Prot 1 = Flash memory sector <i> is write protected, 0 = Flash memory sector <i> is not write protected.
155/231
uPSD33xx
PLDs. The PSD Module contains two PLDs: the
Decode PLD (DPLD), and the General PLD
(GPLD), as shown in Figure 63., page 157. Both
PLDs are fed by a common PLD input signal bus,
and additionally, the GPLD is connected to the
8032 data bus.
PLD logic is specified using PSDsoft Express and
programmed into the PSD Module using the JTAG
ISP channel. PLD logic is non-volatile and avail-
able at power-up. PLDs may not be programmed
by the 8032. The PLDs have selectable levels of
performance and power consumption.
The DPLD performs address decoding, and gen-
erates select signals for internal and external com-
ponents, such as memory, registers, and I/O ports.
The DPLD can generate External Chip-Select
(ECS1-ECS2) signals on Port D.
Turbo Bit and PLDs. The PLDs can minimize
power consumption by going to standby after ALL
the PLD inputs remain unchanged for an extended
time (about 70ns). When the Turbo Bit is set to log-
ic one (Bit 3 of the csiop PMMR0 Register), Turbo
mode is turned off and then this automatic standby
mode is achieved. Turning off Turbo mode in-
creases propagation delays while reducing power
consumption. The default state of the Turbo Bit is
logic zero, meaning Turbo mode is on. Additional-
ly, four bits are available in the csiop PMMR0 and
PMMR2 Registers to block the 8032 bus control
signals (RD, WR, PSEN, ALE) from entering the
PLDs. This reduces power consumption and can
be used only when these 8032 control signals are
not used in PLD logic equations. See Power
Management, page 187.
The GPLD can be used for logic functions, such as
loadable counters and shift registers, state ma-
chines, encoding and decoding logic. These logic
functions can be constructed from a combination
of 16 Output Macrocells (OMC), 20 Input Macro-
cells (IMC), and the AND-OR Array.
Routing of the 16 OMCs outputs can be divided
between pins on three Ports A, B, or C by the OMC
Allocator as shown in Figure 67., page 163. Eight
of the 16 OMCs that can be routed to pins on Port
A or Port B and are named MCELLAB0-
MCELLAB7. The other eight OMCs to be routed to
pins on Port B or Port C and are named
MCELLBC0-MCELLBC7. This routing depends on
the pin number assignments that are specified in
PSDsoft Express for “PLD Outputs” in the Pin Def-
inition section. OMC outputs can also be routed in-
ternally (not to pins) used as buried nodes to
create shifters, counters, etc.
The AND-OR Array is used to form product terms.
These product terms are configured from the logic
definitions entered in PSDsoft Express. A PLD In-
put Bus consisting of 69 signals is connected to
both PLDs. Input signals are shown in Table 84,
both the true and compliment versions of each of
these signals are available at inputs to each PLD.
Note: The 8032 data bus, D0 - D7, does not route
directly to PLD inputs. Instead, the 8032 data bus
has indirect access to the GPLD (not the DPLD)
when the 8032 reads and writes the OMC and IMC
registers within csiop address space.
Table 84. DPLD and GPLD Inputs
Number
Input Source
Input Name
of
Signals
8032 Address Bus
A0-A15
16
4
PSEN, RD, WR,
ALE
8032 Bus Control Signals
Reset from MCU Module RESET
1
Power-Down from Auto-
PDN
1
Power Down Counter
PortA Input Macrocells
PA0-PA7
8
8
4
(80-pin devices only)
PortB Input Macrocells
PortC Input Macrocells
Port D Inputs
PB0-PB7
PC2, PC3, PC4,
PC7
(52-pin devices have only PD1, PD2
PD1)
2
Page Register
PGR0-PGR7
8
8
Macrocell OMC bank AB MCELLAB
Feedback FB0-7
Macrocell OMC bank BC MCELLBC
Feedback FB0-7
8
1
Flash memory Status Bit Ready/Busy
156/231
uPSD33xx
Figure 63. DPLD and GPLD
S U
P L D P U T B
AI06600A
157/231
uPSD33xx
Decode PLD (DPLD). The
64., page 159) generates the following memory
decode signals:
DPLD
(Figure
A product term indicates the logical OR of two or
more inputs. For example, three product terms in
a DPLD output means the final output signal is ca-
pable of representing the logical OR of three differ-
ent input signals, each input signal representing
the logical AND of a combination of the 69 PLD in-
puts.
Using the signal FS0 for example, the user may
create a 3-product term chip select signal that is
logic true when any one of three different address
ranges are true... FS0 = address range 1 OR ad-
dress range 2 OR address range 3.
The phrase “one product term” is a bit misleading,
but commonly used in this context. One product
term is the logical AND of two or more inputs, with
no OR logic involved at all, such as the CSIOP sig-
nal in Figure 64., page 159.
■
Eight Main Flash memory sector select signals
(FS0-FS7) with three product terms each
■
Four Secondary Flash memory sector select
signals (CSBOOT0-CSBOOT3) with three
product terms each
■
■
One SRAM select signal (RS0) with two
product terms
One select signal for the base address of 256
PSD Module device control and status
registers (CSIOP) with one product term
■
■
Two external chip-select output signals for
Port D pins, each with one product term (52-
pin devices only have one pin on Port D)
Two chip-select signals (PSEL0, PSEL1) used
to enable the 8032 data bus repeater function
(Peripheral I/O mode) for Port A on 80-pin
devices. Each has one product term.
158/231
uPSD33xx
Figure 64. DPLD Logic Array
NUMBER OF
PRODUCT TERMS
PLD INPUT BUS
FS0
FS1
3
3
3
3
3
3
3
3
3
3
3
3
8032 ADDRESS (A0 - A15)
8032 CNTL (RD, WR, PSEN, ALE)
PSM MODULE RESET (RST)
16
4
FS2
1
MAIN
FLASH
MEMORY
SECTOR
SELECTS
FS3
POWER-DOWN INDICATOR (PDN)
PIN INPUT PORTS A, B, C (IMCs)
PIN INPUT PORT D
1
FS4
20
2
FS5
FS6
PAGE REGISTER (PGR0 - PGR7)
OMC FEEDBACK (MCELLAB.FB0-7)
OMC FEEDBACK (MCELLBC.FB0-7)
FLASH MEM PROG STATUS (RDYBSY)
8
FS7
8
CSBOOT0
CSBOOT1
CSBOOT2
CSBOOT3
8
SECONDARY
FLASH
MEMORY
SECTOR
SELECTS
1
RS0
SRAM
SELECT
2
1
I/O & CONTROL
REGISTERS
SELECT
CSIOP
ECS0
ECS1
1
1
EXTERNAL
CHIP-
SELECTS
(PORT D)
PSEL0
PSEL1
PERIPHERAL
I/O MODE
RANGE
1
1
SELECTS
AI06601A
159/231
uPSD33xx
General PLD (GPLD). The GPLD is used to cre-
ate general system logic. Figure 63., page 157
shows the architecture of the entire GPLD, and
Figure 65., page 161 shows the relationship be-
tween one OMC, one IMC, and one I/O port pin,
which is representative of pins on Ports A, B, and
C. It is important to understand how these ele-
ments work together. A more detailed description
will follow for the three major blocks (OMC, IMC, I/
O Port) shown in Figure 65. Figure 65 also shows
which csiop registers to access for various PLD
and I/O functions.
The GPLD contains:
■
■
■
■
■
16 Output Macrocells (OMC)
20 Input Macrocells (IMC)
OMC Allocator
Product Term Allocator inside each OMC
AND-OR Array capable of generating up to
137 product terms
■
Three I/O Ports, A, B, and C
160/231
uPSD33xx
Figure 65. GPLD: One OMC, One IMC, and One I/O Port (typical pin, Port A, B, or C)
S U
O L B
, C A O T N A T R D
8 0 3 2 A D D R E S S ,
S U
P L D I N P U T B
AI06602A
161/231
uPSD33xx
Output Macrocell. The GPLD has 16 OMCs. Ar-
chitecture of one individual OMC is shown in Fig-
ure 66. OMCs can be used for internal node
feedback (buried registers to build shift registers,
etc.), or their outputs may be routed to external
port pins. The user can choose any mixture of
OMCs used for buried functions and OMCs used
to drive port pins.
Referring to Figure 66, for each OMC there are na-
tive product terms available from the AND-OR Ar-
ray to form logic, and also borrowed product terms
are available (if unused) from other OMCs. The
polarity of the final product term output is con-
trolled by the XOR gate. Each OMC can imple-
ment sequential logic using the flip-flop element,
or combinatorial logic when bypassing the flip-flop
as selected by the output multiplexer. An OMC
output can drive a port pin through the OMC Allo-
cator, it can also drive the 8032 data bus, and also
it can drive a feedback path to the AND-OR Array
inputs, all at the same time.
The flip-flop in each OMC can be synthesized as a
D, T, JK, or SR type in PSDsoft Express. OMC flip-
flops are specified using PSDsoft Express in the
“User Defined Nodes” section of the Design Assis-
tant. Each flip-flop’s clock, preset, and clear inputs
may be driven individually from a product term of
the AND-OR Array, defined by equations in PSD-
soft Express for signals *. c, *.pr, and *.re respec-
tively. The preset and clear inputs on the flip-flops
are level activated, active-high logic signals. The
clock inputs on the flip-flops are rising-edge logic
signals.
Optionally, the signal CLKIN (pin PD1) can be
used for a common clock source to all OMC flip-
flops. Each flip-flop is clocked on the rising edge.
A common clock is specified in PSDsoft Express
by assigning the function “Common Clock Input”
for pin PD1 in the Pin Definition section, and then
choosing the signal CLKIN when specifying the
clock input (*.c) for individual flip-flops in the “User
Defined Nodes” section.
Figure 66. Detail of a Single OMC
PRODUCT TERMS
FROM OTHER
OMCs
DATA BIT FROM 8032
INDICATES MCU WRITE
TO PARTICULAR CSIO
OMC REGISTER
BORROWED
PTs
LENDED
PTs
PT ALLOCATOR,
MCU READ OF
PARTICULAR CSIOP
OMC REGISTER
DRAWS FROM LOCAL
AND GLOBAL UNUSED
PRODUCT TERMS.
MCU OVERRIDES
PSDsoft DICTATES.
PT PRESET AND
CLR DURING
MCU WRITE
DATA BIT TO 8032
PT PRESET (.PR)
FROM AND-OR ARRAY
FROM AND-OR ARRAY
ALLOCATED PTs
NATIVE PTs
MUX
O
U
T
OMC
OUTPUT
M
U
X
POLARITY
SELECT,
PSDsoft
PRE
OMC
ALLO-
CATOR
D
Q
FROM PLD INPUT BUS
FROM AND-OR ARRAY
M
U
X
GLOBAL CLOCK (CLKIN)
PT CLOCK (.C)
PSDsoft
CLR
PSDsoft
MUX
PT CLEAR (.RE)
FROM AND-OR ARRAY
TO PLD INPUT BUS
NODE FEEDBACK (.FB)
OUTPUT MACROCELL (OMC)
AI06617A
162/231
uPSD33xx
OMC Allocator. Outputs of the 16 OMCs can be
routed to a combination of pins on Port A (80-pin
devices only), Port B, or Port C as shown in Figure
67. OMCs are routed to port pins automatically af-
ter specifying pin numbers in PSDsoft Express.
Routing can occur on a bit-by-bit basis, spitting
OMC assignment between the ports. However,
one OMC can be routed to one only port pin, not
both ports.
Product Term Allocator. Each OMC has a Prod-
uct Term Allocator as shown in Figure
66., page 162. PSDsoft Express uses PT Alloca-
tors to give and take product terms to and from
other OMCs to fit a logic design into the available
silicon resources. This happens automatically in
PSDsoft Express, but understanding how PT allo-
cation works will help the user if the logic design
does not “fit,” in which case the user may try se-
lecting a different pin or different OMC for the logic
where more product terms may be available. The
following list summarizes how product terms are
allocated to each OMC, as shown in Table
85., page 164.
Product term allocation does not add any propaga-
tion delay to the logic. The fitter report generated
by PSDsoft Express will show any PT allocation
that has occurred.
If an equation requires more product terms than
are available to it through PT allocation, then “ex-
ternal” product terms are required, which con-
sumes other OMCs. This is called product term
expansion and also happens automatically in PS-
Dsoft Express as needed. PT expansion causes
additional propagation delay because an addition-
al OMC is consumed by the expansion process
and it’s output is rerouted (or fed back) into the
AND-OR array. The user can examine the fitter re-
port generated by PSDsoft Express to see result-
ing PT allocation and PT expansion (expansion
will have signal names, such as ‘*.fb_0’ or ‘*.fb_1’).
PSDsoft Express will always try to fit the logic de-
sign first by using PT allocation, and if that is not
sufficient then PSDsoft Express will use PT expan-
sion.
Product term expansion may occur in the DPLD
for complex chip select equations for Flash mem-
ory sectors and for SRAM, but this is a rare oc-
curence. If PSDsoft Express does use PT
expansion in the DPLD, it results in an approxi-
mate 15ns additional propagation delay for that
chip select signal, which gives 15ns less time for
the memory to respond. Be aware of this and con-
sider adding a wait state to the 8032 bus access
(using the SFR named, BUSCON), or lower the
8032 clock frequency to avoid problems with
memory access time.
–
–
–
MCELLAB0-MCELLAB7 each have three
native product terms and may borrow up to six
more
MCELLBC0-MCELLBC3 each have four
native product terms and may borrow up to
five more
MCELLBC4-MCELLBC7 each have four
native product terms and may borrow up to six
more.
Native product terms come from the AND-OR Ar-
ray. Each OMC may borrow product terms only
from certain other OMCs, if they are not in use.
Figure 67. OMC Allocator
PORT A PINS
PORT B PINS
2 1
PORT C PINS
(80-pin pkg only)
7
6
5
4
3
2 1
0
7
6
5
4
3
0
7
*
*
4
3
2
*
*
* = Used for JTAG,
Pin Not Available
to GPLD
7
6
5
4 3
2
0
1
2 0
1
7 6 5 4 3
OMC Bank AB (MCELLAB0-7) OMC Bank BC (MCELLBC0-7)
AI09177
163/231
uPSD33xx
Table 85. OMC Port and Data Bit Assignments
Data Bit on 8032 Data
Bus for Loading or
Reading OMC
Port
Native Product Terms
from AND-OR Array
Maximum Borrowed
Product Terms
OMC
(1,2)
Assignment
MCELLAB0
MCELLAB1
MCELLAB2
MCELLAB3
MCELLAB4
MCELLAB5
MCELLAB6
MCELLAB7
MCELLBC0
MCELLBC1
MCELLBC2
MCELLBC3
MCELLBC4
MCELLBC5
MCELLBC6
MCELLBC7
Port A0 or B0
Port A1 or B1
Port A2 or B2
Port A3 or B3
Port A4 or B4
Port A5 or B5
Port A6 or B6
Port A7 or B7
Port B0
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
6
6
6
6
6
6
6
6
5
5
5
5
6
6
6
6
D0
D1
D2
D3
D4
D5
D6
D7
D0
D1
D2
D3
D4
D5
D6
D7
Port B1
Port B or C2
Port B3 or C3
Port B4 or C4
Port B5
Port B6
Port B7 orC7
Note: 1. MCELLAB0-MCELLAB7 can be output to Port A pins only on 80-pin devices. Port A is not available on 52-pin devices
2. Port pins PC0, PC1, PC5, and PC6 are dedicated JTAG pins and are not available as outputs for MCELLBC 0, 1, 5, or 6
164/231
uPSD33xx
Loading and Reading OMCs. Each of the two
OMC groups (eight OMCs each) occupies a byte
in csiop space, named MCELLAB and MCELLBC
(see Table 86 and Table 87). When the 8032
writes or reads these two OMC registers in csiop it
is accessing each of the OMCs through it’s 8-bit
data bus, with the bit assignment shown in Table
85., page 164. Sometimes it is important to know
the bit assignment when the user builds GPLD log-
ic that is accessed by the 8032. For example, the
user may create a 4-bit counter that must be load-
ed and read by the 8032, so the user must know
which nibble in the corresponding csiop OMC reg-
ister the firmware must access. The fitter report
generated by PSDsoft Express will indicate how it
assigned the OMCs and data bus bits to the logic.
The user can optionally force PSDsoft Express to
assign logic to specific OMCs and data bus bits if
desired by using the ‘PROPERTY’ statement in
PSDsoft Express. Please see the PSDsoft Ex-
press User’s Manual for more information on OMC
assignments.
Loading the OMC flip-flops with data from the
8032 takes priority over the PLD logic functions.
As such, the preset, clear, and clock inputs to the
flip-flop can be asynchronously overridden when
the 8032 writes to the csiop registers to load the in-
dividual OMCs.
Table 86. Output Macrocell MCELLAB (address = csiop + offset 20h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
MCELLAB7 MCELLAB6 MCELLAB5 MCELLAB4 MCELLAB3 MCELLAB2 MCELLAB1 MCELLAB0
Note: All bits clear to logic ’0’ at power-on reset, but do not clear after warm reset conditions (non-power-on reset)
Table 87. Output Macrocell MCELLBC (address = csiop + offset 21h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
MCELLBC7 MCELLBC6 MCELLBC5 MCELLBC4 MCELLBC3 MCELLBC2 MCELLBC1 MCELLBC0
Note: All bits clear to logic ’0’ at power-on reset, but do not clear after warm reset conditions (non-power-on reset)
165/231
uPSD33xx
OMC Mask Registers. There is one OMC Mask
Register for each of the two groups of eight OMCs
shown in Table 88 and Table 89. The OMC mask
registers are used to block loading of data to indi-
vidual OMCs. The default value for the mask reg-
isters is 00h, which allows loading of all OMCs.
When a given bit in a mask register is set to a '1,'
the 8032 is blocked from writing to the associated
OMC flip-flop. For example, suppose that only four
of eight OMCs (MCELLAB0-3) are being used for
a state machine. The user may not want the 8032
write to all the OMCs in MCELLAB because it
would overwrite the state machine registers.
Therefore, the user would want to load the mask
register for MCELLAB with the value 0Fh before
writing OMCs.
Table 88. Output Macrocell MCELLAB Mask Register (address = csiop + offset 22h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
Mask
Mask
Mask
Mask
Mask
Mask
Mask
Mask
MCELLAB7 MCELLAB6 MCELLAB5 MCELLAB4 MCELLAB3 MCELLAB2 MCELLAB1 MCELLAB0
Note: 1. Default is 00h after any reset condition
2. 1 = block writing to individual macrocell, 0 = allow writing to individual macrocell
Table 89. Output Macrocell MCELLBC Mask Register (address = csiop + offset 23h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
Mask
Mask
Mask
Mask
Mask
Mask
Mask
Mask
MCELLBC7 MCELLBC6 MCELLBC5 MCELLBC4 MCELLBC3 MCELLBC2 MCELLBC1 MCELLBC0
Note: 1. Default is 00h after any reset condition
2. 1 = block writing to individual macrocell, 0 = allow writing to individual macrocell
Input Macrocells. The GPLD has 20 IMCs, one
for each pin on Port A (80-pin device only), one for
each pin on Port B, and for the four pins on Port C
that are not JTAG pins. The architecture of one in-
dividual IMC is shown in Figure 68., page 167.
IMCs are individually configurable, and they can
strobe a signal coming in from a port pin as a latch
(gated), or as a register (clocked), or the IMC can
pass the signal without strobing, all prior to driving
the signal onto the PLD input bus. Strobing is use-
ful for sampling and debouncing inputs (keypad in-
puts, etc.) before entering the PLD AND-OR
arrays. The outputs of IMCs can be read by the
8032 asynchronously when the 8032 reads the
csiop registers shown in Table 90, Table 91, and
Table 92., page 167. It is possible to read a PSD
Module port pin using one of two different meth-
ods, one method is by reading IMCs as described
here, the other method is using MCU I/O mode de-
scribed in a later section.
The optional IMC clocking or gating signal used to
strobe pin inputs is driven by a product term from
the AND-OR array. There is one clocking or gating
product term available for each group of four
IMCs. Port inputs 0-3 are controlled by one prod-
uct term and 4-7 by another. To specify in PSDsoft
Express the method in which a signal will be
strobed as it enters an IMC for a given input pin on
Port A, B, or C, just specify “PT Clocked Register”
to use a rising edge to clock the incoming signal,
or specify “PT Clock Latch” to use an active high
gate signal to latch the incoming signal. Then de-
fine an equation for the IMC clock (.ld) or the IMC
gate (.le) signal in the “I/O Equations” section.
If the user would like to latch an incoming signal
using the gate signal ALE from the 8032, then in
PSDsoft Express, for a given input pin on Port A,
B, or C, specify “Latched Address” as the pin func-
tion.
If it is desired to pass an incoming signal through
an IMC directly to the AND-OR array inputs with-
out clocking or gating (this is most common), in
PSDsoft Express simply specify “Logic or Ad-
dress” for the input pin function on Port A, B, or C.
166/231
uPSD33xx
Figure 68. Detail of a Single IMC
FROM I/O PORT
LOGIC
8032 READ OF PARTICULAR CSIOP IMC REGISTER
INPUT SIGNAL
FROM PIN ON
PORT A, B, or C
8032 DATA BIT
ALE
PIN INPUT
M
U
X
LATCHED INPUT
GATED INPUT
Q
D
D
(.LD)
PSDsoft
Q
PSDsoft
TO PLD INPUT BUS
(.LE)
G
ALE
M
U
X
PT CLOCK OR GATE (.LD OR .LE)
FROM AND-OR ARRAY
INPUT MACROCELL (IMC)
THIS SIGAL IS GANGED TO 3 OTHER
IMCs, GROUPING IMC 0 - 3 or IMC 4 - 7.
AI06603A
(1)
Table 90. Input Macrocell Port A
(address = csiop + offset 0Ah)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
IMC PA7
IMC PA6
IMC PA5
IMC PA4
IMC PA3
IMC PA2
IMC PA1
IMC PA0
Note: 1. Port A not available on 52-pin uPSD33xx devices
2. 1 = current state of IMC is logic '1,' 0 = current state is logic ’0’
Table 91. Input Macrocell Port B (address = csiop + offset 0Bh)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
IMC PB7
IMC PB6
IMC PB5
IMC PB4
IMC PB3
IMC PB2
IMC PB1
IMC PB0
Note: 1 = current state of IMC is logic '1,' 0 = current state is logic ’0’
Table 92. Input Macrocell Port C (address = csiop + offset 18h)
Bit 7
Bit 6
X
Bit 5
X
Bit 4
Bit 3
Bit 2
Bit 1
X
Bit 0
X
IMC PC7
IMC PC4
IMC PC3
IMC PC2
Note: 1. X = Not guaranteed value, can be read either '1' or '0.' These are JTAG pins.
2. 1 = current state of IMC is logic '1,' 0 = current state is logic ’0’
167/231
uPSD33xx
I/O Ports. There are four programmable I/O ports
on the PSD Module: Port A (80-pin device only),
Port B, Port C, and Port D. Ports A and B are eight
bits each, Port C is four bits, and Port D is two bits
for 80-pin devices or 1-bit for 52-pin devices. Each
port pin is individually configurable, thus allowing
multiple functions per port. The ports are config-
ured using PSDsoft Express then programming
with JTAG, and also by the 8032 writing to csiop
registers at run-time.
A port pin’s output enable signal is controlled by a
two input OR gate whose inputs come from: a
product term of the AND-OR array; the output of
the csiop Direction Register. If an output enable
from the AND-OR Array is not defined, and the
port pin is not defined as an OMC output, and if
Peripheral I/O mode is not used, then the csiop Di-
rection Register has sole control of the OE signal.
As shown in Figure 69., page 169, a physical port
pin is connected to the I/O Port logic and is also
separately routed to an IMC, allowing the 8032 to
read a port pin by two different methods (MCU I/O
input mode or read the IMC).
Port Operating Modes. I/O Port logic has sever-
al modes of operation. Table 88., page 166 sum-
marizes which modes are available on each port.
Each of the port operating modes are described in
following sections. Some operating modes can be
defined using PSDsoft Express, and some by the
8032 writing to the csiop registers at run-time, and
some require both. For example, PLD I/O, Latched
Address Out, and Peripheral I/O modes must be
defined in PSDsoft Express and programmed into
the device using JTAG, but an additional step
must happen at run-time to activate Latched Ad-
dress Out mode and Peripheral I/O mode, but not
needed for PLD I/O. In another example, MCU I/O
mode is controlled completely by the 8032 at run-
time and only a simple pin name declaration is
needed in PSDsoft Express for documentation.
Topics discussed in this section are:
■
■
■
General Port architecture
Port Operating Modes
Individual Port Structure
General Port Architecture. The general archi-
tecture for a single I/O Port pin is shown in Figure
69., page 169. Port structures for Ports A, B, C,
and D differ slightly and are shown in Figure
74., page 181 though Figure 77., page 186.
Figure 69., page 169 shows four csiop registers
whose outputs are determined by the value that
the 8032 writes to csiop Direction, Drive, Control,
and Data Out. The I/O Port logic contains an out-
put mux whose mux select signal is determined by
PSDsoft Express and the csiop Control register
bits at run-time. Inputs to this output mux include
the following:
1. Data from the csiop Data Out register for MCU
I/O output mode (All ports)
Table 89., page 166 summarizes what actions are
needed in PSDsoft Express and what actions are
required by the 8032 at run-time to achieve the
various port functions.
2. Latched de-multiplexed 8032 Address for
Address Output mode (Ports A and B only)
3. Peripheral I/O mode data bit (Port A only)
4. GPLD OMC output (Ports A, B, and C).
The Port Data Buffer (PDB) provides feedback to
the 8032 and allows only one source at a time to
be read when the 8032 reads various csiop regis-
ters. There is one PDB for each port pin enabling
the 8032 to read the following on a pin-by-pin ba-
sis:
1. MCU I/O signal direction setting (csiop
Direction reg)
2. Pin drive type setting (csiop Drive Select reg)
3. Latched Addr Out mode setting (csiop Control
reg)
4. MCU I/O pin output setting (csiop Data Out
reg)
5. Output Enable of pin driver (csiop Enable Out
reg)
6. MCU I/O pin input (csiop Data In reg)
168/231
uPSD33xx
Figure 69. Detail of a Single I/O Port (typical of Ports A, B, C)
I/O PORT
LOGIC
PT OUTPUT ENABLE (.OE)
PSELx
WR
RD PIO EN
FROM AND-OR ARRAY
FROM PLD INPUT BUS
PSD MODULE RESET
PERIPHERAL I/O
MODE SETS
DIRECTION
Q
DIRECTION
(PORT A ONLY)
CSIOP
REGIS-
TERS
8032
DRIVE
Q
DRIVE TYPE
DATA
BITS
D
OE
MUX
PSDsoft
CONTROL
OUTPUT
SELECT
8032
WR
Q
OUTPUT ENABLE
(MCUI/O)
DATA OUT
1
Q
O
U
T
P
U
CLR
RESET
OUTPUT
DRIVER
LATCHED ADDR BIT, PORT A or B
2
T
TYPICAL
PIN
PORT A, B, C
D BIT, PERIPH I/O MODE, Port A
3
4
M
U
X
DIRECTION
1
P
D
B
DRIVE SELECT
8032
DATA
BIT
2
CONTROL
3
4
5
6
PERIPH I/O
DATA BIT
DATA OUT (MCUI/O)
ENABLE OUT
M
U
X
DATA IN (MCUI/O)
INPUT
BUFFER
ONE of 6
CSIOP
8032 RD
REGISTERS
FROM OMC
ALLOCATOR
FROM OMC OUTPUT
TO IMC
AI07873A
169/231
uPSD33xx
Table 93. Port Operating Modes
Port Operating Mode
Port A (80-pin only)
Port B
Port C
Port D
Find it
MCU I/O
Mode., p
age 172
M CU I/O
Yes
Yes
Yes
Yes
PLD I/O
No
OMC MCELLAB Outputs
OMC MCELLBC Outputs
External Chip-Select Outputs
PLD Inputs
Yes
No
No
Yes
Yes
No
No
No
Yes
PLD I/O
Mode., p
age 174
(1)
Yes
No
Yes
Yes
Yes
Yes
Latched
Address
Output
Mode, pa
ge 177
Latched Address Output
Yes
Yes
No
No
No
Peripher
al I/O
Mode, pa
Peripheral I/O Mode
JTAG ISP
Yes
No
No
No
No
No
ge 178
JTAG
ISP
Mode., p
age 179
(2)
Yes
Note: 1. MCELLBC outputs available only on pins PC2, PC3, PC4, and PC7.
2. JTAG pins (PC0/TMS, PC1/TCK, PC5/TDI, PC6/TDO) are dedicated to JTAG pin functions (cannot be used for general I/O).
170/231
uPSD33xx
Table 94. Port Configuration Setting Requirements
Value that 8032
writes to csiop
Value that 8032
writes to csiop
Value that 8032
writes to Bit 7
Port
Operating
Mode
Required Action in
PSDsoft Express to
Configure each Pin
Control Register at Direction Register (PIO_EN) of csiop VM
run-time
at run-time
Register at run-time
Choose the MCU I/O
function and declare the
pin name
Logic 1 = Out of
uPSD
Logic 0 = Into uPSD
MCU I/O
PLD I/O
Logic '0' (default)
N/A
Choose the PLD function
type, declare pin name,
and specify logic
Direction register
has no effect on a
pin if pin is driven
from OMC output
N/A
N/A
N/A
equation(s)
Choose Latched Address
Out function, declare pin Logic '1'
name
Latched Address
Output
Logic '1' Only
N/A
Choose Peripheral I/O
mode function and
specify address range in
PIO_EN Bit = Logic 1
(default is '0')
Peripheral I/O
N/A
DPLD for PSELx
No action required in
PSDsoft to get 4-pin
JTAG. By default TDO,
TDI, TCK, TMS are
dedicated JTAG
functions.
4-PIN JTAG ISP
N/A
N/A
N/A
N/A
N/A
Choose JTAG TSTAT
function for pin PC3 and
JTAG TERR function for
pin PC4.
6-PIN JTAG ISP
(faster
programming)
N/A
171/231
uPSD33xx
MCU I/O Mode. In MCU I/O mode, the 8032 on
the MCU Module expands its own I/O by using the
I/O Ports on the PSD Module. The 8032 can read
PSD Module I/O pins, set the direction of the I/O
pins, and change the output state of I/O pins by ac-
cessing the Data In, Direction, and Data Out csiop
registers respectively at run-time.
corresponding Data In register to determine the
state of an I/O pin, or writes to a Data Out register
to set the state of a pin. The Direction of each pin
may be changed dynamically by the 8032 if de-
sired. A mixture of input and output pins within a
single port is allowed. Figure 69., page 169 shows
the Data In, Data Out, and Direction signal paths.
To implement MCU I/O mode, each desired pin is
specified in PSDsoft Express as MCU I/O function
and given a pin name. Then 8032 firmware is writ-
ten to set the Direction bit for each corresponding
pin during initialization routines (0 = In, 1 = Out of
the chip), then the 8032 firmware simply reads the
The Data In registers are defined in Table 95 to
Table 98. The Data Out registers are defined in
Table 99 to Table 102., page 173. The Direction
registers are defined in Table 103 to Table
106., page 173.
(1)
Table 95. MCU I/O Mode Port A Data In Register (address = csiop + offset 00h)
Bit 7
PA7
Bit 6
PA6
Bit 5
PA5
Bit 4
PA4
Bit 3
PA3
Bit 2
PA2
Bit 1
PA1
Bit 0
PA0
Note: 1. Port A not available on 52-pin uPSD33xx devices
2. For each bit, 1 = current state of input pin is logic '1,' 0 = current state is logic ’0’
Table 96. MCU I/O Mode Port B Data In Register (address = csiop + offset 01h)
Bit 7
PB7
Bit 6
PB6
Bit 5
PB5
Bit 4
PB4
Bit 3
PB3
Bit 2
PB2
Bit 1
PB1
Bit 0
PB0
Note: For each bit, 1 = current state of input pin is logic '1,' 0 = current state is logic ’0’
Table 97. MCU I/O Mode Port C Data In Register (address = csiop + offset 10h)
Bit 7
PC7
Bit 6
X
Bit 5
X
Bit 4
PC4
Bit 3
PC3
Bit 2
PC2
Bit 1
X
Bit 0
X
Note: 1. X = Not guaranteed value, can be read either '1' or '0.'
2. For each bit, 1 = current state of input pin is logic '1,' 0 = current state is logic ’0’
Table 98. MCU I/O Mode Port D Data In Register (address = csiop + offset 11h)
Bit 7
X
Bit 6
X
Bit 5
X
Bit 4
X
Bit 3
X
Bit 2
Bit 1
PD1
Bit 0
X
(3)
PD2
Note: 1. X = Not guaranteed value, can be read either '1' or '0.'
2. For each bit, 1 = current state of input pin is logic '1,' 0 = current state is logic ’0’
3. Not available on 52-pin uPSD33xx devices
(1)
Table 99. MCU I/O Mode Port A Data Out Register (address = csiop + offset 04h)
Bit 7
PA7
Bit 6
PA6
Bit 5
PA5
Bit 4
PA4
Bit 3
PA3
Bit 2
PA2
Bit 1
PA1
Bit 0
PA0
Note: 1. Port A not available on 52-pin uPSD33xx devices
2. For each bit, 1 = drive port pin to logic '1,' 0 = drive port pin to logic ’0’
3. Default state of register is 00h after reset or power-up
Table 100. MCU I/O Mode Port B Data Out Register (address = csiop + offset 05h)
Bit 7
PB7
Bit 6
PB6
Bit 5
PB5
Bit 4
PB4
Bit 3
PB3
Bit 2
PB2
Bit 1
PB1
Bit 0
PB0
Note: 1. For each bit, 1 = drive port pin to logic '1,' 0 = drive port pin to logic ’0’
2. Default state of register is 00h after reset or power-up
172/231
uPSD33xx
Table 101. MCU I/O Mode Port C Data Out Register (address = csiop + offset 12h)
Bit 7
PC7
Bit 6
N/A
Bit 5
N/A
Bit 4
PC4
Bit 3
PC3
Bit 2
PC2
Bit 1
N/A
Bit 0
N/A
Note: 1. For each bit, 1 = drive port pin to logic '1,' 0 = drive port pin to logic ’0’
2. Default state of register is 00h after reset or power-up
Table 102. MCU I/O Mode Port D Data Out Register (address = csiop + offset 13h)
Bit 7
N/A
Bit 6
N/A
Bit 5
N/A
Bit 4
N/A
Bit 3
N/A
Bit 2
Bit 1
PD1
Bit 0
N/A
(3)
PD2
Note: 1. For each bit, 1 = drive port pin to logic '1,' 0 = drive port pin to logic ’0’
2. Default state for register is 00h after reset or power-up
3. Not available on 52-pin uPSD33xx devices
(1)
Table 103. MCU I/O Mode Port A Direction Register (address = csiop + offset 06h)
Bit 7
PA7
Bit 6
PA6
Bit 5
PA5
Bit 4
PA4
Bit 3
PA3
Bit 2
PA2
Bit 1
PA1
Bit 0
PA0
Note: 1. Port A not available on 52-pin uPSD33xx devices
2. For each bit, 1 = out from uPSD33xx port pin1, 0 = in to PSD33xx port pin
3. Default state for register is 00h after reset or power-up
Table 104. MCU I/O Mode Port B Direction In Register (address = csiop + offset 07h)
Bit 7
PB7
Bit 6
PB6
Bit 5
PB5
Bit 4
PB4
Bit 3
PB3
Bit 2
PB2
Bit 1
PB1
Bit 0
PB0
Note: 1. For each bit, 1 = out from uPSD33xx port pin1, 0 = in to PSD33xx port pin
2. Default state for register is 00h after reset or power-up
Table 105. MCU I/O Mode Port C Direction Register (address = csiop + offset 14h)
Bit 7
PC7
Bit 6
N/A
Bit 5
N/A
Bit 4
PC4
Bit 3
PC3
Bit 2
PC2
Bit 1
N/A
Bit 0
N/A
Note: 1. For each bit, 1 = out from uPSD33xx port pin1, 0 = in to PSD33xx port pin
2. Default state for register is 00h after reset or power-up
Table 106. MCU I/O Mode Port D Direction Register (address = csiop + offset 15h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
N/A
(3)
N/A
N/A
N/A
N/A
N/A
PD1
PD2
Note: 1. For each bit, 1 = out from uPSD33xx port pin1, 0 = in to PSD33xx port pin
2. Default state for register is 00h after reset or power-up
3. Not available on 52-pin uPSD33xx devices
173/231
uPSD33xx
PLD I/O Mode. Pins on Ports A, B, C, and D can
serve as inputs to either the DPLD or the GPLD.
Inputs to these PLDs from Ports A, B, and C are
routed through IMCs before reaching the PLD in-
put bus. Inputs to the PLDs from Port D do not
pass through IMCs, but route directly to the PLD
input bus.
Pins on Ports A, B, and C can serve as outputs
from GPLD OMCs, and Port D pins can be outputs
from the DPLD (external chip-selects) which do
not consume OMCs.
Whenever a pin is specified to be a PLD output, it
cannot be used for MCU I/O mode, or other pin
modes. If a pin is specified to be a PLD input, it is
still possible to read the pin using MCU I/O input
mode with the csiop register Data In. Also, the
csiop Direction register can still affect a pin which
is used for a PLD input. The csiop Data Out regis-
ter has no effect on a PLD output pin.
Each pin on Ports A, B, C, and D have a tri-state
buffer at the final output stage. The Output Enable
signal for this buffer is driven by the logical OR of
two signals. One signal is an Output Enable signal
generated by the AND-OR array (from an .oe
equation specified in PSDsoft), and the other sig-
nal is the output of the csiop Direction register.
This logic is shown in Figure 69., page 169. At
power-on, all port pins default to high-impedance
input (Direction registers default to 00h). However,
if an equation is written for the Output Enable that
is active at power-on, then the pin will behave as
an output.
To give a general idea how PLD logic is imple-
mented using PSDsoft Express, Figure
71., page 175 illustrates the pin declaration win-
dow of PSDsoft Express, showing the PLD output
at pin PB0 declared as “Combinatorial” in the “PLD
Output” section, and a signal name, “pld_out”, is
specified. The other three signals on pins PB1,
PB2, and PB3 would be declared as “Logic or Ad-
dress” in the “PLD Input” section, and given signal
names.
In the “Design Assistant” window of PSDsoft Ex-
press shown in Figure 72., page 176, simply enter
the logic equation for the signal “pld_out” as
shown. Either type in the logic statements or enter
them using a point-and-click method, selecting
various signal names and logic operators avail-
able in the window.
After PSDsoft Express has accepted and realized
the logic from the equations, it synthesizes the log-
ic statement:
pld_out = ( pld_in_1 # pld_in_2 ) & !pld_in_3;
to be programmed into the GPLD. See the PSD-
soft User’s Manual for all the steps.
Note: If a particular OMC output is specified as an
internal node and not specified as a port pin output
in PSDsoft Express, then the port pin that is asso-
ciated with that OMC can be used for other I/O
functions.
Figure 70. Simple PLD Logic Example
PLD I/O equations are specified in PSDsoft Ex-
press and programmed into the uPSD using
JTAG. Figure 70 shows a very simple combinato-
rial logic example which is implemented on pins of
Port B.
PLDIN 3
PB3
PLDIN 2
PLDIN 1
PLD OUT
PB2
PB1
PB0
AI09178
174/231
uPSD33xx
Figure 71. Pin Declarations in PSDsoft Express for Simple PLD Example
175/231
uPSD33xx
Figure 72. Using the Design Assistant in PSDsoft Express for Simple PLD Example
176/231
uPSD33xx
Latched Address Output Mode. In the MCU
Module, the data bus Bits D0-D15 are multiplexed
with the low address Bits A0-A15, and the ALE sig-
nal is used to separate them with respect to time.
Sometimes it is necessary to send de-multiplexed
address signals to external peripherals or memory
devices. Latched Address Output mode will drive
individual demuxed address signals on pins of
Ports A or B. Port pins can be designated for this
function on a pin-by-pin basis, meaning that an en-
tire port will not be sacrificed if only a few address
signals are needed.
To activate this mode, the desired pins on Port A
or Port B are designated as “Latched Address Out”
in PSDsoft. Then in the 8032 initialization firm-
ware, a logic ’1’ is written to the csiop Control reg-
ister for Port A or Port B in each bit position that
corresponds to the pin of the port driving an ad-
dress signal. Table 107 and Table 108 define the
csiop Control register locations and bit assign-
ments.
The latched low address byte A4-A7 is available
on both Port A and Port B. The high address byte
A8-A15 is available on Port B only. Selection of
high or low address byte is specified in PSDsoft
Express.
(1)
Table 107. Latched Address Output, Port A Control Register (address = csiop + offset 02h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PA7
PA6
PA5
PA4
PA3
PA2
PA1
PA0
(addr A7)
(addr A6)
(addr A5)
(addr A4)
(addr A3)
(Addr A2)
(addr A1)
(addr A0)
Note: 1. Port A not available on 52-pin uPSD33xx devices
2. For each bit, 1 = drive demuxed 8032 address signal on pin, 0 = pin is default mode, MCU I/O
3. Default state for register is 00h after reset or power-up
Table 108. Latched Address Output, Port B Control Register (address = csiop + offset 03h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PB7
(addr A7 or
A15)
PB6
(addr A6 or
A14)
PB5
(addr A5 or
A13)
PB4
(addr A4 or
A12)
PB3
(addr A3 or
A11)
PB2
(Addr A2 or
A10)
PB1
(addr A1 or
A9)
PB0
(addr A0 or
A8)
Note: 1. For each bit, 1 = drive demuxed 8032 address signal on pin, 0 = pin is default mode, MCU I/O
2. Default state for register is 00h after reset or power-up
177/231
uPSD33xx
Peripheral I/O Mode. This mode will provide a
data bus repeater function for the 8032 to interface
with external parallel peripherals. The mode is
only available on Port A (80-pin devices only) and
the data bus signals, D0 - D7, are de-multiplexed
(no address A0-A7). When active, this mode be-
haves like a bidirectional buffer, with the direction
automatically controlled by the 8032 RD and WR
signals for a specified address range. The DPLD
signals PSEL0 and PSEL1 determine this address
range. Figure 69., page 169 shows the action of
Peripheral I/O mode on the Output Enable logic of
the tri-state output driver for a single port pin. Fig-
ure 73., page 178 illustrates data repeater the op-
eration. To activate this mode, choose the pin
function “Peripheral I/O Mode” in PSDsoft Express
on any Port A pin (all eight pins of Port A will auto-
matically change to this mode). Next in PSDsoft,
specify an address range for the PSELx signals in
the “Chip-Select” section of the “Design Assistant.”
Specify an address range for either PSEL0 or
PSEL1. Always qualify the PSELx equation with
“PSEN is logic '1'” to ensure Peripheral I/O mode
is only active during 8032 data cycles, not code cy-
cles. Only one equation is needed since PSELx
signals are OR’ed together (Figure 73). Then in
the 8032 initialization firmware, a logic ’1’ is written
to the csiop VM register, Bit 7 (PIO_EN) as shown
in Table 73., page 132. After this, Port A will auto-
matically perform this repeater function whenever
the 8032 presents an address (and memory page
number, if paging is used) that is within the range
specified by PSELx. Once Port A is designated as
Peripheral I/O mode in PSDsoft Express, it cannot
be used for other functions.
Note: The user can alternatively connect an exter-
nal parallel peripheral to the standard 8032 AD0-
AD7 pins on an 80-pin uPSD device (not Port A),
but these pins have multiplexed address and data
signals, with a weaker fanout drive capability.
Figure 73. Peripheral I/O Mode
8032 RD
PSEL0
PSEL1
PA0 - PA7
8032 DATA
8
8
BUS D0-D7
PORT
A pins
VM REGISTER BIT 7 (PIO EN)
(DE-MUXED)
8032 WR
AI02886A
178/231
uPSD33xx
JTAG ISP Mode. Four of the pins on Port C are
based on the IEEE 1149.1 JTAG specification and
are used for In-System Programming (ISP) of the
PSD Module and debugging of the 8032 MCU
Module. These pins (TDI, TDO, TMS, TCK) are
dedicated to JTAG and cannot be used for any
other I/O function. There are two optional pins on
Port C (TSTAT and TERR) that can be used to re-
duce programming time during ISP. See JTAG
ISP and JTAG Debug, page 195.
Other Port Capabilities. It is possible to change
the type of output drive on the ports at run-time. It
is also possible to read the state of the output en-
able signal of the output driver at run-time. The fol-
lowing sections provide the details.
be sized not to exceed the current sink capability
of the pin (see DC specifications). Open Drain out-
puts are diode clamped, thus the maximum volt-
age on an pin configured as Open Drain is V
0.7V.
+
DD
A pin can be configured as Open Drain if its corre-
sponding bit in the Drive Select Register is set to
logic '1.'
Note: The slew rate is a measurement of the rise
and fall times of an output. A higher slew rate
means a faster output response and may create
more electrical noise. A pin operates in a high slew
rate when the corresponding bit in the Drive Reg-
ister is set to '1.' The default rate is standard slew
rate (see AC specifications).
Port Pin Drive Options. The csiop Drive Select
registers allow reconfiguration of the output drive
type for certain pins on Ports A, B, C, and D. The
8032 can change the default drive type setting at
run-time. The is no action needed in PSDsoft Ex-
press to change or define these pin output drive
types. Figure 69., page 169 shows the csiop Drive
Select register output controlling the pin output
driver. The default setting for drive type for all pins
on Ports A, B, C, and D is a standard CMOS push-
pull output driver.
Table 109 through Table 112., page 180 show the
csiop Drive Registers for Ports A, B, C, and D. The
tables summarize which pins can be configured as
Open Drain outputs and which pins the slew rate
can be changed. The default output type is CMOS
push/pull output with normal slew rate.
Enable Out Registers. The state of the output
enable signal for the output driver at each pin on
Ports A, B, C, and D can be read at any time by the
8032 when it reads the csiop Enable Output regis-
ters. Logic '1' means the driver is in output mode,
logic ’0’ means the output driver is in high-imped-
ance mode, making the pin suitable for input mode
(read by the input buffer shown in Figure
69., page 169). Figure 69 shows the three sources
that can control the pin output enable signal: a
product term from AND-OR array; the csiop Direc-
tion register; or the Peripheral I/O Mode logic (Port
A only). The csiop Enable Out registers represent
the state of the final output enable signal for each
port pin driver, and are defined in Table
113., page 180 through Table 116., page 180.
Note: When a pin on Port A, B, C, D is not used as
an output and has no external device driving it as
an input (floating pin), excess power consumption
can be avoided by placing a weak pull-up resistor
(100KΩ) to V which keeps the CMOS input pin
DD
from floating.
Drive Select Registers. The csiop Drive Select
Registers will configure a pin output driver as
Open Drain or CMOS push/pull for some port pins,
and controls the slew rate for other port pins. An
external pull-up resistor should be used for pins
configured as Open Drain, and the resistor should
(1)
Table 109. Port A Pin Drive Select Register (address = csiop + offset 08h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PA7
PA6
PA5
PA4
PA3
PA2
PA1
PA0
Open Drain
Open Drain
Open Drain
Open Drain
Slew Rate
Slew Rate
Slew Rate
Slew Rate
Note: 1. Port A not available on 52-pin uPSD33xx devices
2. For each bit, 1 = pin drive type is selected, 0 = pin drive type is default mode, CMOS push/pull
3. Default state for register is 00h after reset or power-up
Table 110. Port B Pin Drive Select Register (address = csiop + offset 09h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PB7
PB6
PB5
PB4
PB3
PB2
PB1
PB0
Open Drain
Open Drain
Open Drain
Open Drain
Slew Rate
Slew Rate
Slew Rate
Slew Rate
Note: 1. For each bit, 1 = pin drive type is selected, 0 = pin drive type is default mode, CMOS push/pull
2. Default state for register is 00h after reset or power-up
179/231
uPSD33xx
Table 111. Port C Pin Drive Select Register (address = csiop + offset 16h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PC7
Open Drain
PC4
Open Drain
PC3
Open Drain
PC2
Open Drain
N/A (JTAG)
N/A (JTAG)
N/A (JTAG)
N/A (JTAG)
Note: 1. For each bit, 1 = pin drive type is selected, 0 = pin drive type is default mode, CMOS push/pull
2. Default state for register is 00h after reset or power-up
Table 112. Port D Pin Drive Select Register (address = csiop + offset 17h)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
N/A
(3)
PD1
Slew Rate
PD2
N/A
N/A
N/A
N/A
N/A
Slew Rate
Note: 1. For each bit, 1 = pin drive type is selected, 0 = pin drive type is default mode, CMOS push/pull
2. Default state for register is 00h after reset or power-up
3. Pin is not available on 52-pin uPSD33xx devices
(1)
Table 113. Port A Enable Out Register (address = csiop + offset 0Ch)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PA7 OE
PA6 OE
PA5 OE
PA4 OE
PA3 OE
PA2 OE
PA1 OE
PA0 OE
Note: 1. Port A not available on 52-pin uPSD33xx devices
2. For each bit, 1 = pin drive is enabled as an output, 0 = pin drive is off (high-impedance, pin used as input)
Table 114. Port B Enable Out Register (address = csiop + offset 0Dh)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PB7 OE
PB6 OE
PB5 OE
PB4 OE
PB3 OE
PB2 OE
PB1 OE
PB0 OE
Note: For each bit, 1 = pin drive is enabled as an output, 0 = pin drive is off (high-impedance, pin used as input)
Table 115. Port C Enable Out Register (address = csiop + offset 1Ah)
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
PC7 OE
N/A (JTAG)
N/A (JTAG)
PC4 OE
PC3 OE
PC2 OE
N/A (JTAG)
N/A (JTAG)
Note: 1. For each bit, 1 = pin drive is enabled as an output, 0 = pin drive is off (high-impedance, pin used as input)
Table 116. Port D Enable Out Register (address = csiop + offset 1Bh)
Bit 7
N/A
Bit 6
N/A
Bit 5
N/A
Bit 4
N/A
Bit 3
N/A
Bit 2
Bit 1
Bit 0
N/A
(2)
PD1 OE
PD2 OE
Note: 1. For each bit, 1 = pin drive is enabled as an output, 0 = pin drive is off (high-impedance, pin used as input)
2. Pin is not available on 52-pin uPSD33xx devices
180/231
uPSD33xx
Individual Port Structures. Ports A, B, C, and D
have some differences. The structure of each indi-
vidual port is described in the next sections.
Port A Structure. Port A supports the following
operating modes:
Port A also supports Open Drain/Slew Rate output
drive type options using csiop Drive Select regis-
ters. Pins PA0-PA3 can be configured to fast slew
rate, pins PA4-PA7 can be configured to Open
Drain Mode.
See Figure 74 for details.
■
MCU I/O Mode
■
GPLD Output Mode from Output Macrocells
MCELLABx
■
■
■
GPLD Input Mode to Input Macrocells IMCAx
Latched Address Output Mode
Peripheral I/O Mode
Figure 74. Port A Structure
I/O PORT A
LOGIC
PT OUTPUT ENABLE (.OE)
PSD MODULE RESET
WR
PSELx
RD PIO EN
FROM AND-
OR ARRAY
FROM PLD
INPUT BUS
PERIPHERAL I/O
MODE SETS
DIRECTION
Q
CSIOP
REGIS-
TERS
DIRECTION
DRIVE TYPE SELECT(1)
8032
DATA
BITS
DRIVE
Q
1 = FAST
SLEW RATE,
PA0 - PA3
1 = OPEN
DRAIN,
PA4 - PA7
D
OE
MUX
PSDsoft
CONTROL
OUTPUT
SELECT
8032
WR
Q
VDD
VDD
(MCUI/O)
DATA OUT
O
U
T
P
U
T
OUTPUT
ENABLE
1
Q
TYPICAL
PIN, PORT A
PIN
OUTPUT
CLR
RESET
LATCHED ADDR BIT
2
D BIT, PERIPH I/O MODE
3
4
M
U
X
DIRECTION
1
2
3
DRIVE SELECT
CONTROL
P
D
B
8032
DATA
BIT
DATA OUT
(MCUI/O)
PERIPH I/O
DATA BIT
M
U
X
4
5
ENABLE OUT
CMOS
BUFFER
DATA IN (MCUI/O)
6
PIN INPUT
8032 RD
ONE of 6
CSIOP
REGISTERS
NO
HYSTERESIS
FROM OMC OUTPUT (MCELLABx)
FROM OMC
ALLOCATOR
IMCA0 - IMCA7
TO IMCs
AI09179
Note: 1. Port pins PA0-PA3 are capable of Fast Slew Rate output drive option. Port pins PA4-PA7 are capable of Open Drain output option.
181/231
uPSD33xx
Port B Structure. Port B supports the following
operating modes:
Port B also supports Open Drain/Slew Rate output
drive type options using the csiop Drive Select reg-
isters. Pins PB0-PB3 can be configured to fast
slew rate, pins PB4-PB7 can be configured to
Open Drain Mode.
■
MCU I/O Mode
■
GPLD Output Mode from Output Macrocells
MCELLABx, or MCELLBCx (OMC allocator
routes these signals)
See Figure 75 for detail.
■
■
GPLD Input Mode to Input Macrocells IMCBx
Latched Address Output Mode
Figure 75. Port B Structure
I/O PORT B
LOGIC
PT OUTPUT ENABLE (.OE)
PSD MODULE RESET
FROM AND-
OR ARRAY
FROM PLD
INPUT BUS
DIRECTION
Q
CSIOP
REGIS-
TERS
Q
DRIVE TYPE SELECT(1)
8032
DATA
BITS
DRIVE
1 = FAST
SLEW RATE,
PB0 - PB3
1 = OPEN
DRAIN,
PB4 - PB7
D
PSDsoft
CONTROL
OUTPUT
SELECT
8032
WR
Q
VDD
VDD
(MCUI/O)
DATA OUT
O
U
T
P
U
T
OUTPUT
ENABLE
1
Q
TYPICAL
PIN, PORT B
PIN
OUTPUT
CLR
RESET
LATCHED ADDR BIT
2
3
OUTPUT
ENABLE
M
U
X
DIRECTION
1
2
3
DRIVE SELECT
CONTROL
P
D
B
8032
DATA
BIT
DATA OUT
(MCUI/O)
M
U
X
4
5
ENABLE OUT
CMOS
BUFFER
DATA IN (MCUI/O)
6
PIN INPUT
8032 RD
ONE of 6
CSIOP
REGISTERS
NO
HYSTERESIS
FROM OMC OUTPUT
(MCELLABx or MCELLBCx)
FROM OMC
ALLOCATOR
IMCB0 - IMCB7
TO IMCs
AI09180
Note: 1. Port pins PB0-PB3 are capable of Fast Slew Rate output drive option. Port pins PB4-PB7 are capable of Open Drain output option.
182/231
uPSD33xx
Port C Structure. Port C supports the following
operating modes on pins PC2, PC3, PC4, PC7:
function is specified in PSDsoft Express as
SRAM Standby Mode (battery
backup), page 193.
■
MCU I/O Mode
–
–
PC3 can be used as an output to indicate
when a Flash memory program or erase
operation has completed. This is specified in
PSDsoft Express as Ready/Busy
(PC3), page 153.
PC4 can be used as an output to indicate
when the SRAM has switched to backup
■
GPLD Output Mode from Output Macrocells
MCELLBC2, MCELLBC3, MCELLBC4,
MCELLBC7
GPLD Input Mode to Input Macrocells IMCC2,
IMCC3, IMCC4, IMCC7
■
See Figure 76., page 184 for detail.
Port C pins can also be configured in PSDsoft for
other dedicated functions:
voltage (when V is less than the battery
DD
input voltage on PC2). This is specified in
PSDsoft Express as “Standby-On Indicator”
(see SRAM Standby Mode (battery
backup), page 193).
–
Pins PC3 and PC4 support TSTAT and TERR
status indicators, to reduce the amount of time
required for JTAG ISP programming. These
two pins must be used together for this
function, adding to the four standard JTAG
signals. When TSTAT and TERR are used, it
is referred to as “6-pin JTAG”. PC3 and PC4
cannot be used for other functions if they are
used for 6-pin JTAG. See JTAG ISP and
JTAG Debug, page 195 for details.
The remaining four pins (TDI, TDO, TCK, TMS) on
Port C are dedicated to the JTAG function and
cannot be used for any other function. See JTAG
ISP and JTAG Debug, page 195.
Port C also supports the Open Drain output drive
type options on pins PC2, PC3, PC4, and PC7 us-
ing the csiop Drive Select registers.
–
PC2 can be used as a voltage input (from
battery or other DC source) to backup the
contents of SRAM when V is lost. This
DD
183/231
uPSD33xx
Figure 76. Port C Structure
I/O PORT C
LOGIC
PT OUTPUT ENABLE, .OE (JTAG STATE MACHINE
AUTOMATICALLY CONTROLS OE FOR JTAG SIGNALS)
FROM AND-
OR ARRAY
PSD MODULE RESET
FROM PLD
INPUT BUS
(1)
VDD/VBAT
DIRECTION
Q
CSIOP
REGIS-
TERS
PULL-UP
ONLY ON
JTAG TDI,
TMS, TCK
SIGNALS
50k
DRIVE TYPE SELECT(2)
8032
DATA
BITS
DRIVE
Q
D
8032
WR
(1)
PSDsoft
VDD/VBAT
VDD
(MCUI/O)
DATA OUT
O
U
T
P
U
OUTPUT
ENABLE
1
Q
CLR
TYPICAL
PIN,
PORT C
PIN
OUTPUT
RESET
2
3
T
M
U
X
DIRECTION
4
5
1
DRIVE SELECT
P
D
B
2
8032
DATA
BIT
DATA OUT
(MCUI/O)
3
4
ENABLE OUT
M
U
X
PIN
INPUT
CMOS
BUFFER
DATA IN (MCUI/O)
5
8032 RD
ONE of 6
CSIOP
REGISTERS
NO
HYSTERESIS
FROM OMC OUTPUT (MCELLBCx)
FROM OMC
ALLOCATOR
STANDBY ON(2)
RDY/BSY(2)
FROM SRAM
BACK-UP CIRCUIT
TO SRAM
BATTERY
BACK-UP
CIRCUIT(2)
FROM FLASH MEMORIES
TDO, TSTAT(2), TERR(2)
TDI, TMS, TCK
IMCC2, IMCC3,
IMCC4, IMCC7
TO/FROM JTAG
STATE MACHINE
TO IMCs
AI09181
Note: 1. Pull-up switches to V
when SRAM goes to battery back-up mode.
BAT
2. Optional function on a specific Port C pin.
184/231
uPSD33xx
Port D Structure. Port D has two I/O pins (PD1,
PD2) on 80-pin uPSD33xx devices, and just one
pin (PD1) on 52-pin devices, supporting the follow-
ing operating modes:
Port D pins can also be configured in PSDsoft as
pins for other dedicated functions:
–
PD1 can be used as a common clock input to
all 16 OMC Flip-flops (see OMCs, page 136)
and also the Automatic Power-Down
(APD), page 189.
■
MCU I/O Mode
■
DPLD Output Mode for External Chip Selects,
ECS1, ECS2. This does not consume OMCs
in the GPLD.
–
PD2 can be used as a common chip select
signal (CSI) for the Flash and SRAM
memories on the PSD Module (see Chip Se-
lect Input (CSI), page 191). If driven to logic ’1’
by an external source, CSI will force all
memories into standby mode regardless of
what other internal memory select signals are
doing on the PSD Module. This is specified in
PSDsoft as “PSD Chip Select Input, CSI”.
■
PLD Input Mode – direct input to the PLD Input
Bus available to DPLD and GPLD. Does not
use IMCs
See Figure 77., page 186 for detail.
Port D also supports the Fast Slew Rate output
drive type option using the csiop Drive Select reg-
isters.
185/231
uPSD33xx
Figure 77. Port D Structure
I/O PORT D
LOGIC
PT OUTPUT ENABLE (.OE)
PSD MODULE RESET
FROM AND-
OR ARRAY
FROM PLD
INPUT BUS
DIRECTION
Q
CSIOP
REGIS-
TERS
Q
8032
DATA
BITS
DRIVE TYPE SELECT
DRIVE
1 = FAST
SLEW RATE
D
8032
WR
PSDsoft
VDD VDD
(MCUI/O)
DATA OUT
O
U
T
P
U
OUTPUT
ENABLE
1
Q
TYPICAL
PIN, PORT D
PIN
OUTPUT
CLR
RESET
2
T
OUTPUT
ENABLE
M
U
X
DIRECTION
1
2
P
D
B
8032
DATA
BIT
DRIVE SELECT
DATA OUT
(MCUI/O)
3
4
5
M
U
X
ENABLE OUT
CMOS
BUFFER
DATA IN (MCUI/O)
PIN INPUT
8032 RD
ONE of 5
CSIOP
REGISTERS
NO
HYSTERESIS
FROM DPLD EXTERNAL CHIP (ECSx)
FROM DPLD
CLKIN(1)
CSI(1)
TO POWER MANAGEMENT AND PLD INPUT BUS
TO POWER MANAGEMENT
PD1. PIN, PD2.PIN
DIRECTLY TO PLD INPUT BUS, NO IMC
AI09182
Note: 1. Optional function on a specific Port D pin.
186/231
uPSD33xx
Power Management. The PSD Module offers
configurable power saving options, and also a way
to manage power to the SRAM (battery backup).
These options may be used individually or in com-
binations. A top level description for these func-
tions is given here, then more detailed
descriptions will follow.
–
PSD Module Chip Select Input (CSI): This
input on pin PD2 (80-pin devices only) can be
used to disable the internal memories, placing
them in standby mode even if address inputs
are changing. This feature does not block any
internal signals (the address and data buffers
are still on but signals are ignored) and CSI
does not disable the PLDs. This is a good
alternative to using the APD counter, which
requires an external clock on pin PD1.
Non-Turbo Mode: The PLDs can operate in
Turbo or non-Turbo modes. Turbo mode has
the shortest signal propagation delay, but
consumes more current than non-Turbo
mode. A csiop register can be written by the
8032 to select modes, the default mode is with
Turbo mode enabled. In non-Turbo mode, the
PLDs can achieve very low standby current (~
zero DC current) while no PLD inputs are
changing, and the PLDs will even use less AC
current when inputs do change compared to
Turbo mode.
–
Zero-Power Memory: All memory arrays
(Flash and SRAM) in the PSD Module are built
with zero-power technology, which puts the
memories into standby mode (~ zero DC
current) when 8032 address signals are not
changing. As soon as a transition occurs on
any address input, the affected memory
“wakes up”, changes and latches its outputs,
then goes back to standby. The designer does
not have to do anything special to achieve this
memory standby mode when no inputs are
changing—it happens automatically. Thus,
the slower the 8032 clock, the lower the
current consumption.
–
Both PLDs (DPLD and GPLD) are also zero-
power, but this is not the default condition. The
8032 must set a bit in one of the csiop PMMR
registers at run-time to achieve zero-power.
When the Turbo mode is enabled, there is a
significant DC current component AND the AC
current component is higher than non-Turbo
mode, as shown in Figure 85., page 202 (5V)
and Figure 86., page 202 (3.3V).
–
Automatic Power-Down (APD): The APD
feature allows the PSD Module to reach it’s
lowest current consumption levels. If enabled,
the APD counter will time-out when there is a
lack of 8032 bus activity for an extended
amount of time (8032 asleep). After time-out
occurs, all 8032 address and data buffers on
the PSD Module are shut down, preventing
the PSD Module memories and potentially the
PLDs from waking up from standby, even if
address inputs are changing state because of
noise or any external components driving the
address lines. Since the actual address and
data buffers are turned off, current
–
Blocking Bits: Significant power savings can
be achieved by blocking 8032 bus control
signals (RD, WR, PSEN, ALE) from reaching
PLD inputs, if these signals are not used in
any PLD equations. Blocking is achieved by
the 8032 writing to the “blocking bits” in csiop
PMMR registers. Current consumption of the
PLDs is directly related to the composite
frequency of all transitions on PLD inputs, so
blocking certain PLD inputs can significantly
lower PLD operating frequency and power
consumption (resulting in a lower frequency
on the graphs of Figure 85., page 202 and
Figure 86., page 202).
consumption is even further reduced.
Note: Non-address signals are still available
to PLD inputs and will wake up the PLDs if
these signals are changing state, but will not
wake up the memories.
–
SRAM Backup Voltage: Pin PC2 can be
configured in PSDsoft to accept an alternate
DC voltage source (battery) to automatically
retain the contents of SRAM when V drops
below this alternate voltage.
Note: It is recommended to prevent unused
DD
The APD counter requires a relatively slow
external clock input on pin PD1 that does stop
when the 8032 goes to sleep mode.
inputs from floating on Ports A, B, C, and D by
–
Forced Power-Down (FPD): The MCU can
put the PSD Module into Power-Down mode
with the same results as using APD described
above, but FPD does not rely on the APD
counter. Instead, FPD will force the PSD
Module into Power-Down mode when the
MCU firmware sets a bit in one of the csiop
PMMR registers. This is a good alternative to
APD because no external clock is needed for
the APD counter.
pulling them up to V with a weak external
DD
resistor (100KΩ), or by setting the csiop
Direction register to “output” at run-time for all
unused inputs. This will prevent the CMOS
input buffers of unused input pins from
drawing excessive current.
The csiop PMMR register definitions are shown in
117 through Table 119., page 188.
187/231
uPSD33xx
Table 117. Power Management Mode Register PMMR0 (address = csiop + offset B0h)
Bit 0
Bit 1
Bit 2
Bit 3
X
APD Enable
X
0
0
1
0
Not used, and should be set to zero.
Automatic Power Down (APD) counter is disabled.
APD counter is enabled
Not used, and should be set to zero.
0 = on PLD Turbo mode is on
PLD Turbo
Disable
1 = off PLD Turbo mode is off, saving power.
CLKIN (pin PD1) to the PLD Input Bus is not blocked. Every transition of CLKIN
powers-up the PLDs.
0 = on
Blocking Bit,
CLKIN to
Bit 4
Bit 5
(1)
CLKIN input to PLD Input Bus is blocked, saving power. But CLKIN still goes to APD
counter.
PLDs
1 = off
0 = on CLKIN input is not blocked from reaching all OMC’s common clock inputs.
Blocking Bit,
CLKIN to
CLKIN input to common clock of all OMCs is blocked, saving power. But CLKIN still
1 = off
(1)
OMCs Only
goes to APD counter and all PLD logic besides the common clock input on OMCs.
Bit 6
Bit 7
X
X
0
0
Not used, and should be set to zero.
Not used, and should be set to zero.
Note: All the bits of this register are cleared to zero following Power-up. Subsequent Reset (RST) pulses do not clear the registers.
1. Blocking bits should be set to logic ’1’ only if the signal is not needed in a DPLD or GPLD logic equation.
Table 118. Power Management Mode Register PMMR2 (address = csiop + offset B4h)
Bit 0
Bit 1
X
X
0
0
Not used, and should be set to zero.
Not used, and should be set to zero.
0 = on 8032 WR input to the PLD Input Bus is not blocked.
1 = off 8032 WR input to PLD Input Bus is blocked, saving power.
0 = on 8032 RD input to the PLD Input Bus is not blocked.
1 = off 8032 RD input to PLD Input Bus is blocked, saving power.
Blocking Bit,
Bit 2
Bit 3
(1)
WR to PLDs
Blocking Bit,
(1)
RD to PLDs
Blocking Bit, 0 = on 8032 PSEN input to the PLD Input Bus is not blocked.
PSEN to
Bit 4
Bit 5
(1)
1 = off 8032 PSEN input to PLD Input Bus is blocked, saving power.
PLDs
Blocking Bit, 0 = on 8032 ALE input to the PLD Input Bus is not blocked.
ALE to
(1)
1 = off 8032 ALE input to PLD Input Bus is blocked, saving power.
PLDs
Blocking Bit, 0 = on Pin PC7 input to the PLD Input Bus is not blocked.
PC7 to
Bit 5
Bit 7
(1)
1 = off Pin PC7 input to PLD Input Bus is blocked, saving power.
PLDs
X
0
Not used, and should be set to zero.
Note: The bits of this register are cleared to zero following Power-up. Subsequent Reset (RST) pulses do not clear the registers.
1. Blocking bits should be set to logic ’1’ only if the signal is not needed in a DPLD or GPLD logic equation.
Table 119. Power Management Mode Register PMMR3 (address = csiop + offset C7h)
Bit 0
Bit 1
X
0
Not used, and should be set to zero.
FORCE_PD 0 = off APD counter will cause Power-Down Mode if APD is enabled.
1 = on Power-Down mode will be entered immediately regardless of APD activity.
Bit 3-7
X
0
Not used, and should be set to zero.
Note: The bits of this register are cleared to zero following Power-up. Subsequent Reset (RST) pulses do not clear the registers.
188/231
uPSD33xx
Automatic Power-Down (APD). The APD unit
shown in Figure 63., page 157 puts the PSD Mod-
ule into power-down mode by monitoring the activ-
ity of the 8032 Address Latch Enable (ALE) signal.
If the APD unit is enabled by writing a logic ’1’ to
Bit 1 of the csiop PMMR0 register, and if ALE sig-
nal activity has stopped (8032 in sleep mode),
then the four-bit APD counter starts counting up. If
the ALE signal remains inactive for 15 clock peri-
ods of the CLKIN signal (pin PD1), then the APD
counter will reach maximum count and the power
down indicator signal (PDN) goes to logic ’1’ forc-
ing the PSD Module into power-down mode. Dur-
ing this time, all buffers on the PSD Module for
8032 address and data signals are disabled in sil-
icon, preventing the PSD Module memories from
waking up from stand-by mode, even if noise or
other devices are driving the address lines. The
PLDs will also stay in standby mode if the PLDs
are in non-Turbo mode and if all other PLD inputs
(non-address signals) are static.
“PDN” signal in the DPLD chip select equations for
FSx, CSBOOTx, RS0, and CSIOP.
The following should be kept in mind when the
PSD Module is in power-down mode:
–
8032 address and data bus signals are
blocked from all memories and both PLDs.
–
The PSD Module comes out of power-down
mode when: ALE starts pulsing again, or the
CSI input on pin PD2 transitions from logic ’1’
to logic '0,' or the PSD Module reset signal,
RST, transitions from logic ’0’ to logic '1.'
–
–
Various signals can be blocked (prior to
power-down mode) from entering the PLDs by
using “blocking bits” in csiop PMMR registers.
All memories enter standby mode, and the
state of the PLDs and I/O Ports are
unchanged (if no PLD inputs change). Table
121., page 194 shows the effects of power-
down mode on I/O pins while in various
operating modes.
However, if the ALE signal has a transition before
the APD counter reaches max count, the APD
counter is cleared to zero and the PDN signal will
not go active, preventing power-down mode. To
prevent unwanted APD time-outs during normal
8032 operation (not sleeping), it is important to
choose a clock frequency for CLKIN that will NOT
produce 15 or more pulses within the longest peri-
od between ALE transitions. A 32768 Hz clock sig-
nal is quite often an ideal frequency for CLKIN and
APD, and this frequency is often available on ex-
ternal supervisor or real-time clock devices.
The “PDN” power-down indicator signal is avail-
able to the PLD input bus to use in any PLD equa-
tions if desired. The user may want to send this
signal as a PLD output to an external device to in-
dicate the PSD Module is in power-down mode.
PSDsoft Express automatically includes the
–
–
The 8032 Ports 1,3, and 4 on the MCU Module
are not affected at all by power-down mode in
the PSD Module.
Power-down standby current given in the AC
specifications for PSD Module assume there
are no transitions on any unblocked PLD
input, and there are no output pins driving any
loads.
The APD counter will count whenever Bit 1 of
csiop PMMR0 register is set to logic '1,' and when
the ALE signal is steady at either logic ’1’ or logic
’0’ (not transitioning). Figure 79., page 191 shows
the flow leading up to power-down mode. The only
action required in PSDsoft Express to enable APD
mode is to select the pin function “Common Clock
Input, CLKIN” before programming with JTAG.
189/231
uPSD33xx
Forced Power Down (FDP). An alternative to
APD is FPD. The resulting power-savings is the
same, but the PDN signal in Figure 78., page 191
is set and Power-Down mode is entered immedi-
ately when firmware sets the FORCE_PD Bit to
logic '1' in the csiop Register PMMR3 (Bit 1). FPD
will override APD counter activity when
FORCE_PD is set. No external clock source for
the APD counter is needed. The FORCE_PD Bit is
cleared only by a reset condition.
Caution must be used when implementing FPD
because code memory goes off-line as soon as
PSD Module Power-Down mode is entered, leav-
ing the MCU with no instruction stream to execute.
The MCU Module must put itself into Power-Down
mode after it puts the PSD Module into Power-
Down Mode. How can it do this if code memory
goes off-line? The answer is the Pre-Fetch Queue
(PFQ) in the MCU Module. By using the instruction
scheme shown in the 8051 assembly code exam-
ple in Table 120, the PFQ will be loaded with the
final instructions to command the MCU Module to
Power Down mode after the PDS Module goes to
Power-Down mode. In this case, even though the
code memory goes off-line in the PSD Module, the
last few MCU instruction are sourced from the
PFQ.
Table 120. Forced Power-Down Example
PDOWN:
ANL
ORL
MOV
A8h, #7Fh
; disable all interrupts
9Dh, #C0h
DPTR, #xxC7
; ensure PFQ and BC are enabled
; load XDATA pointer to select PMMR3 register (xx = base
; address of csiop registers)
CLR
JMP
NOP
A
; clear A
LOOP
; first loop - fill PFQ/BQ with Power Down instructions
; second loop - fetch code from PFQ/BC and set Power-
; Down bits for PSD Module and then MCU Module
LOOP:
MOVX
MOV
MOV
JMP
@DPTR, A
87h, A
; set FORCE_PD Bit in PMMR3 in PSD Module in second
; loop
; set PD Bit in PCON Register in MCU Module in second
; loop
; set power-down bit in the A Register, but not in PMMR3 or
; PCON yet in first loop
A, #02h
LOOP
; uPSD enters into Power-Down mode in second loop
190/231
uPSD33xx
Figure 78. Automatic Power Down (APD) Unit
8032 ADDR
8032 ADDR FROM MCU MODULE
8032 DATA FROM MCU MODULE
8032 DATA
PMMR3, BIT 1 (FORCE_PD)
PMMR0, BIT 1 (APD EN)
PSD
MODULE
LINE
BUFFERS
ENABLE
1 = FOUND
TRANSITION
PDN
ENABLE
FULL
COUNT
ENABLE
8032 ALE
1 = POWER
DOWN MODE
CLEAR
4-BIT APD
TRANSITION
DETECTION
FSx
CSBOOTx
RS0
UP-COUNTER
PSD MODULE RST_
DPLD CHIP
PDN
CSI
CLK
1 = FOUND
EDGE
SELECT
EDGE
EQUATIONS
DETECTION
CSIOP
CSI (pin PD2)
CLKIN (pin PD1)
PDN
OMC OUTPUTS
GPLD
WHEN CSI FUNCTION IS SPECIFIED IN PSDSOFT EXPRESS,
CSI IS PART OF EQUATIONS FOR FSx, CSBOOTx, RS0, and CSIOP
AI06608B
Figure 79. Power-Down Mode Flow Chart
Chip Select Input (CSI). Pin PD2 of Port D can
optionally be configured in PSDsoft Express as the
PSD Module Chip Select Input, CSI, which is an
active-low logic input. By default, pin PD2 does not
have the CSI function.
RESET
When the CSI function is specified in PSDsoft Ex-
press, the CSI signal is automatically included in
DPLD chip select equations for FSx, CSBOOTx,
RS0, and CSIOP. When the CSI pin is driven to
logic ’0’ from an external device, all of these mem-
ories will be available for READ and WRITE oper-
ations. When CSI is driven to logic '1,' none of
these memories are available for selection, re-
gardless of the address activity from the 8032, re-
ducing power consumption. The state of the PLD
and port I/O pins are not changed when CSI goes
to logic ’1’ (disabled).
Enable APD.
Set PMMR0,
Bit 1 = 1
OPTIONAL. Disable desired inputs to
PLDs by setting PMMR0 bits 4 and 5,
and PMMR2 bits 2 through 6
ALE idle
NO
for 15 CLKIN
clocks?
YES
PDN = 1, PSD
Module in Power-
Down Mode
AI09183
191/231
uPSD33xx
PLD Non-Turbo Mode. The power consumption
and speed of the PLDs are controlled by the Turbo
Bit (Bit 3) in the csiop PMMR0 register. By setting
this bit to logic '1,' the Turbo mode is turned off and
both PLDs consume only stand-by current when
ALL PLD inputs have no transitions for an extend-
ed time (65ns for 5V devices, 100ns for 3.3 V de-
vices), significantly reducing current consumption.
The PLDs will latch their outputs and go to stand-
by, drawing very little current. When Turbo mode
is off, PLD propagation delay time is increased as
shown in the AC specifications for the PSD Mod-
ule. Since this additional propagation delay also
effects the DPLD, the response time of the memo-
ries on the PSD Module is also lengthened by that
same amount of time. If Turbo mode is off, the
user should add an additional wait state to the
8032 BUSCON SFR register if the 8032 clock fre-
quency is higher that a particular value. Please re-
fer to Table 36., page 64 in the MCU Module
section.
Turbo Mode Current Consumption. To deter-
mine the AC current component of the specific
PLD design with Turbo mode on, the user will have
to interpolate from the graph, given the number of
product terms specified in the fitter report, and the
estimated composite frequency of PLD input sig-
nal transitions. For the DC component (y-axis
crossing), the user can calculate the number by
multiplying the number of product terms used
(from fitter report) times the DC current per prod-
uct term specified in the DC specifications for the
PSD Module. The total PLD current usage is the
sum of its AC and DC components.
Non-Turbo Mode Current Consumption. No-
tice in Figure 85., page 202 and Figure
86., page 202 that when Turbo mode is off, the DC
current consumption is “zero” (just standby cur-
rent) when the composite frequency of PLD input
transitions is zero (no input transitions). Now mov-
ing up the frequency axis to consider the AC cur-
rent component, current consumption remains
considerably less than Turbo mode until PLD input
transitions happen so rapidly that the PLDs do not
have time to latch their outputs and go to standby
between the transitions anymore. This is where
the lines converge on the graphs, and current con-
sumption becomes the same for PLD input transi-
tions at this frequency and higher regardless if
Turbo mode is on or off. To determine the current
consumption of the PLDs with Turbo mode off, ex-
trapolate the AC component from the graph based
on number of product terms and input frequency.
The only DC component in non-Turbo mode is the
PSD Module standby current.
The key to reducing PLD current consumption is to
reduce the composite frequency of transitions on
the PLD input bus, moving down the frequency
scale on the graphs. One way to do this is to care-
fully select which signals are entering PLD inputs,
not selecting high frequency signals if they are not
used in PLD equations. Another way is to use PLD
“Blocking Bits” to block certain signals from enter-
ing the PLD input bus.
The default state of the Turbo Bit is logic '0,' mean-
ing Turbo mode is on by default (after power-up
and reset conditions) until it is turned off by the
8032 writing to PMMR0.
PLD Current Consumption. Figure
85., page 202 and Figure 86., page 202 (5V and
3.3V devices respectively) show the relationship
between PLD current consumption and the com-
posite frequency of all the transitions on PLD in-
puts, indicating that a higher input frequency
results in higher current consumption.
Current consumption of the PLDs have a DC com-
ponent and an AC component. Both need to be
considered when calculating current consumption
for a specific PLD design. When Turbo mode is on,
there is a linear relationship between current and
frequency, and there is a substantial DC current
component consumed by the PSD Module when
there are no transitions on PLD inputs (composite
frequency is zero). The magnitude of this DC cur-
rent component is directly proportional to how
many product terms are used in the equations of
both PLDs. PSDsoft Express generates a “fitter”
report that specifies how many product terms were
used in a design out of a total of 186 available
product terms. Figure 85., page 202 and Figure
86., page 202 both give two examples, one with
100% of the 186 product terms used, and another
with 25% of the 186 product terms used.
192/231
uPSD33xx
PLD Blocking Bits. Blocking specific signals
from entering the PLDs using bits of the csiop
PMMR registers can further reduce PLD AC cur-
rent consumption by lowering the effective com-
posite frequency of inputs to the PLDs.
CLKIN is still available to the PLD input bus and
the APD counter.
See Table 117., page 188 for details.
SRAM Standby Mode (battery backup). The
SRAM on the PSD Module may optionally be
backed up by an external battery (or other DC
source) to make its contents non-volatile. This is
achieved by connecting a battery to pin PC2 on
Port C and selecting the “SRAM Standby” function
for pin PC2 within PSDsoft Express. Automatic
voltage supply cross-over circuitry is built into the
PSD Module to switch SRAM supply to battery as
Blocking 8032 Bus Control Signals. When the
8032 is active on the MCU Module, four bus con-
trol signals (RD, WR, PSEN, and ALE) are con-
stantly transitioning to manage 8032 bus traffic.
Each time one of these signals has a transition
from logic ’1’ to '0,' or 0 to '1,' it will wake up the
PLDs if operating in non-Turbo mode, or when in
Turbo mode it will cause the affected PLD gates to
draw current. If equations in the DPLD or GPLD do
not use the signals RD, WR, PSEN, or ALE then
these signals can be blocked which will reduce the
AC current component substantially. These bus
control signals are rarely used in DPLD equations
because they are routed in silicon directly to the
memory arrays of the PSD Module, bypassing the
PLDs. For example, it is NOT necessary to qualify
a memory chip select signal with an MCU write
strobe, such as “fs0 = address range & !WR_”.
Only “fs0 = address range” is needed.
Each of the 8032 bus control signals may be
blocked individually by writing to Bits 2, 3, 4, and 5
of the PMMR2 register shown in Table
118., page 188. Blocking any of these four bus
control signals only prevents them from reaching
the PLDs, but they will always go to the memories
directly.
However, sometimes it is necessary to use these
8032 bus control signals in the GPLD when creat-
ing interface signals to external I/O peripherals.
But it is still possible to save power by dynamically
unblocking the bus signals before reading/writing
the external device, then blocking the signals after
the communication is complete.
soon as V
drops below the voltage level of the
DD
battery. SRAM contents are protected while bat-
tery voltage is greater than 2.0V. Pin PC4 on Port
C can be used as an output to indicate that a bat-
tery switch-over has occurred. This is configured
in PSDsoft Express by selecting the “Standby On
Indicator” option for pin PC4.
PSD Module Reset Conditions
The PSD Module receives a reset signal from the
MCU Module. This reset signal is referred to as the
“RST” input in PSD Module documentation, and it
is active-low when asserted. The character of the
RST signal generated from the MCU Module is de-
scribed
in
SUPERVISORY
FUNCTIONS, page 65.
Upon power-up, and while RST is asserted, the
PSD Module immediately loads its configuration
from non-volatile bits to configure the PLDs and
other items. PLD logic is operational and ready for
use well before RST is de-asserted. The state of
PLD outputs are determined by equations speci-
fied in PSDsoft Express.
The Flash memories are reset to Read Array
mode after any assertion of RST (even if a pro-
gram or erase operation is occurring).
Flash memory WRITE operations are automatical-
The user can also block an input signal coming
from pin PC7 to the PLD input bus if desired by
writing to Bit 6 of PMMR2.
ly prevented while V
is ramping up until it rises
DD
above the V
voltage threshold at which time
LKO
Flash memory WRITE operations are allowed.
Blocking Common Clock, CLKIN. The
input
Once the uPSD33xx is up and running, any subse-
quent reset operation is referred to as a warm re-
set, until power is turned off again. Some PSD
Module functions are reset in different ways de-
pending if the reset condition was caused from a
CLKIN (from pin PD1) can be blocked to reduce
current consumption. CLKIN is used as a common
clock input to all OMC flip-flips, it is a general input
to the PLD input bus, and it is used to clock the
APD counter. In PSDsoft Express, the function of
pin PD1 must be specified as “Common Clock In-
put, CLKIN” before programming the device with
JTAG to get the CLKIN function.
Bit 4 of PMMR0 can be set to logic ’1’ to block
CLKIN from reaching the PLD input bus, but
CLKIN will still reach the APD counter.
Bit 5 of PMMR0 can be set to logic ’1’ to block
CLKIN from reaching the OMC flip-flops only, but
power-up reset or
a
warm reset. Table
121., page 194 summarizes how PSD Module
functions are affected by power-up and warm re-
sets, as well as the affect of PSD Module power-
down mode (from APD).
The I/O pins of PSD Module Ports A, B, C, and D
do not have weak internal pull-ups.
193/231
uPSD33xx
In MCU I/O mode, Latched Address Out mode,
and Peripheral I/O mode, the pins of Ports A, B, C,
and D become standard CMOS inputs during a re-
set condition. If no external devices are driving
these pins during reset, then these inputs may
float and draw excessive current. If low power con-
sumption is critical during reset, then these floating
ing them, and if there is no equation specified for
the DPLD or GPLD to make them an output. In this
case, a weak external pull-up resistor (100KΩ min-
imum) should be used on floating pins to avoid ex-
cessive current draw.
The pins on Ports 1, 3, and 4 of the 8032 MCU
module do have weak internal pull-ups and the in-
puts will not float, so no external pull-ups are need-
ed.
inputs should be pulled up externally to V with a
DD
weak (100KΩ minimum) resistor.
In PLD I/O mode, pins of Ports A, B, C, and D may
also float during reset if no external device is driv-
Table 121. Function Status During Power-Up Reset, Warm Reset, Power-down Mode
Port Configuration
Power-Up Reset
Warm Reset
APD Power-down Mode
Pin logic state is
unchanged
MCU I/O
Pins are in input mode
Pins are in input mode
Pin logic is valid after
internal PSD Module
configuration bits are
loaded. Happens long
before RST is de-asserted
Pin logic depends on inputs
to PLD (8032 addresses
are blocked from reaching
PLD inputs during power-
down mode)
Pin logic is valid and is
determined by PLD logic
equations
PLD I/O
Pins logic state not defined
since 8032 address signals
are blocked
Latched Address Out Mode Pins are High Impedance
Pins are High Impedance
Pins are High Impedance
Peripheral I/O Mode
JTAG ISP and Debug
Pins are High Impedance
Pins are High Impedance
JTAG channel is active and JTAG channel is active and JTAG channel is active and
available
available
available
Register
Power-Up Reset
Warm Reset
APD Power-down Mode
PMMR0 and PMMR2
Cleared to 00h
Unchanged
Unchanged
Depends on .re and .pr
equations
Depends on .re and .pr
equations
Output of OMC Flip-flops
Cleared to ’0’
Initialized with value that
was specified in PSDsoft
Initialized with value that
was specified in PSDsoft
(1)
Unchanged
Unchanged
VM Register
All other csiop registers
Cleared to 00h
Cleared to 00h
Note: 1. VM register Bit 7 (PIO_EN) and Bit 0 (SRAM in 8032 program space) are cleared to zero at power-up and warm reset conditions.
194/231
uPSD33xx
JTAG ISP and JTAG Debug. An IEEE 1149.1
serial JTAG interface is used on uPSD33xx devic-
es for ISP (In-System Programming) of the PSD
module, and also for debugging firmware on the
MCU Module. IEEE 1149.1 Boundary Scan oper-
ations are not supported in the uPSD33xx.
The main advantage of JTAG ISP is that a blank
uPSD33xx device may be soldered to a circuit
board and programmed with no involvement of the
8032, meaning that no 8032 firmware needs to be
present for ISP. This is good for manufacturing, for
field updates, and for easy code development in
the lab. JTAG-based programmers and debug-
gers for uPSD33xx are available from STMicro-
electronics and 3rd party vendors.
JTAG signals TCK and TMS are common to both
modules as specified in IEEE 1149.1. When JTAG
devices are chained, typically one devices is in
BYPASS mode while another device is executing
a JTAG operation. For the uPSD33xx, the PSD
Module is in BYPASS mode while debugging the
MCU Module, and the MCU Module is in BYPASS
mode while performing ISP on the PSD Module.
The RESET_IN input pin on the uPSD33xx pack-
age goes to the MCU Module, and this module will
generate the RST reset signal for the PSD Mod-
ule. These reset signals are totally independent of
the JTAG TAP controllers, meaning that the JTAG
channel is operational when the modules are held
in reset. It is required to assert RESET_IN during
ISP. STMicroelectronics and 3rd party JTAG ISP
tools will automatically assert a reset signal during
ISP. However, this reset signal must be connected
to RESET_IN as shown in examples in Figure Fig-
ure 81., page 196 and Figure 82., page 198.
ISP is different than IAP (In-Application Program-
ming). IAP involves the 8032 to program Flash
memory over any interface supported by the 8032
(e.g., UART, SPI, I2C), which is good for remote
updates over
a
communication channel.
uPSD33xx devices support both ISP and IAP. The
entire PSD Module (Flash memory and PLD) may
be programmed with JTAG ISP, but only the Flash
memories may be programmed using IAP.
Figure 80. JTAG Chain in uPSD33xx Package
uPSD33XX
JTAG Chaining Inside the Package. JTAG pro-
tocol allows serial “chaining” of more than one de-
vice in a JTAG chain. The uPSD33xx is
assembled with a stacked die process combining
the PSD Module (one die) and the MCU Module
(the other die). These two die are chained together
within the uPSD33xx package. The standard
JTAG interface has four basic signals:
MCU MODULE
8032 MCU
OPTIONAL
DEBUG
RESET_IN
RESET
JTAG TAP
CONTROLLER
TDO
TMS TCK
TDI
■
■
■
■
TDI - Serial data into device
TDO - Serial data out of device
TCK - Common clock
JTAG TDO
JTAG TCK
IEEE 1149.1
JTAG TMS
TMS - Mode Selection
Every device that supports IEEE 1149.1 JTAG
communication contains a Test Access Port (TAP)
controller, which is a small state machine to man-
age JTAG protocol and serial streams of com-
mands and data. Both the PSD Module and the
MCU Module each contain a TAP controller.
JTAG TDI
TDI
TSTAT
TDO
TMS TCK
PC3 / TSTAT
OPTIONAL
JTAG TAP
CONTROLLER
TERR
PC4 / TERR
Figure 80 illustrates how these die are chained
within a package. JTAG programming/test equip-
ment will connect externally to the four IEEE
1149.1 JTAG pins on Port C. The TDI pin on the
uPSD33xx package goes directly to the PSD Mod-
ule first, then exits the PSD Module through TDO.
TDO of the PSD Module is connected to TDI of the
MCU Module. The serial path is completed when
TDO of the MCU Module exits the uPSD33xx
package through the TDO pin on Port C. The
RST
MAIN
FLASH
MEMORY
2ND
FLASH
MEMORY
PLD
PSD MODULE
AI09184
195/231
uPSD33xx
In-System Programming. The ISP function can
use two different configurations of the JTAG inter-
face:
4-pin JTAG ISP (default). The four basic JTAG
pins on Port C are enabled for JTAG operation at
all times. These pins may not be used for other I/
O functions. There is no action needed in PSDsoft
Express to configure a device to use 4-pin JTAG,
as this is the default condition. No 8032 firmware
is needed to use 4-pin ISP because all ISP func-
tions are controlled from the external JTAG pro-
■
4-pin JTAG: TDI, TDO, TCK, TMS
■
6-pin JTAG: Signals above plus TSTAT,
TERR
At power-up, the four basic JTAG signals are all in-
puts, waiting for a command to appear on the
JTAG bus from programming or test equipment.
When the enabling command is received, TDO be-
comes an output and the JTAG channel is fully
functional. The same command that enables the
JTAG channel may optionally enable the two addi-
tional signals, TSTAT and TERR.
gram/test
equipment.
Figure
81
shows
recommended connections on a circuit board to a
JTAG program/test tool using 4-pin JTAG. It is re-
quired to connect the RST output signal from the
JTAG program/test equipment to the RESET_IN
input on the uPSD33xx. The RST signal is driven
by the equipment with an Open Drain driver, allow-
ing other sources (like a push button) to drive
RESET_IN without conflict.
Note: The recommended pull-up resistors and de-
coupling capacitor are illustrated in Figure 81.
Figure 81. Recommended 4-pin JTAG Connections
CIRCUIT
BOARD
100k
typical
JTAG
CONN.
uPSD33XX
TMS - PC0
TMS
TCK - PC1
TCK
SRAM STBY or I/O - PC2
GENERAL I/O - PC3
GENERAL I/O - PC4
TDI - PC5
JTAG
Programming
or Test
Equipment
Connects Here
TDI
TDO
TDO - PC6
(1,2)
VCC
GENERAL I/O - PC7
0.01
µF
GND
GENERAL I/O
SIGNALS
10k
RESETIN
RST(3)
100k
PUSH BUTTON
or ANY OTHER
RESET SOURCE
DEBUG
OPTIONAL
TEST POINT
AI09185
Note: 1. For 5V uPSD33xx devices, pull-up resistors and V pin on the JTAG connector should be connected to 5V system V
.
DD
CC
2. For 3.3V uPSD33xx devices, pull-up resistors and V pin on the JTAG connector should be connected to 3.3V system V
.
CC
CC
3. This signal is driven by an Open-Drain output in the JTAG equipment, allowing more than one source to activate RESETIN.
196/231
uPSD33xx
6-pin JTAG ISP (optional). The optional signals
TSTAT and TERR are programming status flags
that can reduce programming time by as much as
30% compared to 4-pin JTAG because this status
information does not have to be scanned out of the
device serially. TSTAT and TERR must be used
as a pair for 6-pin JTAG operation.
TSTAT and TERR are functional only when JTAG
ISP operations are occurring, which means they
are non-functional during JTAG debugging of the
8032 on the MCU Module.
Programming times vary depending on the num-
ber of locations to be programmed and the JTAG
programming equipment, but typical JTAG ISP
programming times are 10 to 25 seconds using 6-
pin JTAG. The signals TSTAT and TERR are not
included in the IEEE 1149.1 specification.
–
TSTAT (pin PC3) indicates when
programming of a single Flash location is
complete. Logic 1 = Ready, Logic 0 = busy.
–
TERR (pin PC4) indicates if there was a Flash
programming error. Logic 1 = no error,
Logic 0 = error.
Figure 82., page 198 shows recommended con-
nections on a circuit board to a JTAG program/test
tool using 6-pin JTAG. It is required to connect the
RST output signal from the JTAG program/test
equipment to the RESET_IN input on the
uPSD33xx. The RST signal is driven by the equip-
ment with an Open Drain driver, allowing other
sources (like a push button) to drive RESET_IN
without conflict.
The pin functions for PC3 and PC4 must be select-
ed as “Dedicated JTAG - TSTAT” and “Dedicated
JTAG - TERR” in PSDsoft Express to enable 6-pin
JTAG ISP.
No 8032 firmware is needed to use 6-pin ISP be-
cause all ISP functions are controlled from the ex-
ternal JTAG program/test equipment.
Note: The recommended pull-up resistors and de-
coupling capacitor are illustrated in Figure 82.
197/231
uPSD33xx
Figure 82. Recommended 6-pin JTAG Connections
100k typical
CIRCUIT
BOARD
JTAG
CONN.
uPSD33XX
TMS - PC0
TMS
TCK - PC1
TCK
SRAM STBY or I/O - PC2
TSTAT - PC3
TSTAT
TERR
TDI
TERR - PC4
JTAG
TDI - PC5
Programming
or Test
TDO
TDO - PC6
Equipment
Connects Here
GENERAL I/O - PC7
(1,2)
VCC
0.01
µF
GND
GENERAL I/O
SIGNALS
10k
RESETIN
RST(3)
100k
PUSH BUTTON
or ANY OTHER
RESET SOURCE
DEBUG
OPTIONAL
TEST POINT
AI09186
Note: 1. For 5V uPSD33xx devices, pull-up resistors and V pin on the JTAG connector should be connected to 5V system V
.
DD
CC
2. For 3.3V uPSD33xx devices, pull-up resistors and V pin on the JTAG connector should be connected to 3.3V system V
.
CC
CC
3. This signal is driven by an Open-Drain output in the JTAG equipment, allowing more than one source to activate RESET_IN.
198/231
uPSD33xx
Recommended JTAG Connector. There is no
industry standard JTAG connector. STMicroelec-
tronics recommends a specific JTAG connector
and pinout for uPSD3xxx so programming and de-
bug equipment will easily connect to the circuit
board. The user does not have to use this connec-
tor if there is a different connection scheme.
The recommended connector scheme can accept
a standard 14-pin ribbon cable connector (2 rows
of 7 pins on 0.1” centers, 0.025” square posts,
standard keying) as shown in Figure 83. See the
STMicroelectronics “FlashLINK, FL-101 User
Manual” for more information.
Chaining uPSD33xx Devices. It is possible to
chain a uPSD33xx device with other uPSD33xx
devices on a circuit board, and also chain with
IEEE 1149.1 compliant devices from other manu-
facturers. Figure 84., page 200 shows a chaining
example. The TDO of one device connects to the
TDI of the next device, and so on. Only one device
is performing JTAG operations at any given time
while the other two devices are in BYPASS mode.
Configuration for JTAG chaining can be made in
PSDsoft Express by choosing “More than one de-
vice” when prompted about chaining devices. No-
tice in Figure 84., page 200 that the uPSD33xx
devices are chained externally, but also be aware
that the two die within each uPSD33xx device are
chained internally. This internal chaining of die is
transparent to the user and is taken care of by PS-
Dsoft Express and 3rd party JTAG tool software.
Figure 83. Recommended JTAG Connector
VIEW: Looking into face of shrouded
male connector, with 0.025"
posts on 0.1" centers.
14
TERR
13
TDO
The example in Figure 84., page 200 also shows
how to use 6-pin JTAG when chaining devices.
The signals TSTAT and TERR are configured as
open-drain type signals from PSDsoft Express.
This facilitates a wired-OR connection of TSTAT
signals from multiple uPSD33xx devices and also
a wired-OR connection of TERR signals from
those same multiple devices. PSDsoft Express
puts TSTAT and TERR signals into open-drain
mode by default, requiring external pull-up resis-
tors. Click on 'Properties' in the JTAG-ISP window
of PSDsoft Express to change to standard CMOS
push-pull outputs if desired, but wired-OR logic is
not possible in CMOS output mode.
Connector reference:
Molex 70247-1401
12
GND TCK
11
This connector accepts a 14-pin
ribbon cable such as:
9
10
GND TMS
• Samtec:
HCSD-07-D-06.00-01-S-N
KEY
WAY
8
7
VCC
RST
• Digikey:
M3CCK-14065-ND
6
5
TSTAT TDI
3
4
CNTL GND
TRST JEN
1
2
AI09187
199/231
uPSD33xx
Figure 84. Example of Chaining uPSD33xx Devices
JTAG
CONN.
CIRCUIT BOARD
Device 1
VCC
100K
100K
100K
TMS
TCK
TMS
TCK
TDI
TDO
TDI
JTAG
Programming
or Test
Equipment
Optional
Optional
TSTAT
TSTAT
TERR
TERR
Connects Here
µPSD33XX
TDO
100K
TMS
TCK
Device 2
10K
100K
TDI
TDO
RST
IEEE 1149.1
Compliant
Device
GND
TMS
TCK
Device N
100K
TDI
TDO
System
Reset
Circuitry
TSTAT
TERR
uPSD33XX
AI09188
200/231
uPSD33xx
Debugging the 8032 MCU Module. The 8032
on the MCU module may be debugged in-circuit
using the same four basic JTAG signals as used
for JTAG ISP (TDI, TDO, TCK, TMS). The signals
TSTAT and TERR are not needed for debugging,
and they will not create a problem if they exist on
the circuit board while debugging. The same con-
nector specified in Figure 83., page 199 can be
used for ISP or for 8032 debugging. There are 3rd
party suppliers of uPSD33xx JTAG debugging
equipment (check www.st.com/psm). These are
small pods which connect to a PC (or notebook
computer) using a USB interface, and they are
driven by an 8032 Integrated Development Envi-
ronment (IDE) running on the PC.
Standard debugging features are provided
through this JTAG interface such as single-step,
breakpoints, trace, memory dump and fill, and oth-
ers. There is also a dedicated Debug pin (shown
in Figure 80., page 195) which can be configured
as an output to trigger external devices upon a
programmable internal event (e.g., breakpoint
match), or the pin can be configured as an input so
an external device can initiate an internal debug
event (e.g., break execution). The Debug pin func-
tion is configured by the 8032 IDE debug software
tool. See DEBUG UNIT, page 39 for more details.
JTAG Security Setting. A programmable securi-
ty bit in the PSD Module protects its contents from
unauthorized viewing and copying. The security
bit is set by clicking on the “Additional PSD Set-
tings” box in the main flow diagram of PSDsoft Ex-
press, then choosing to set the security bit. Once
a file with this setting is programmed into a
uPSD33xx using JTAG ISP, any further attempts
to communicate with the uPSD33xx using JTAG
will be limited. Once secured, the only JTAG oper-
ation allowed is a full-chip erase. No reading or
modifying Flash memory or PLD logic is allowed.
Debugging operations to the MCU Module are
also not allowed. The only way to defeat the secu-
rity bit is to perform a JTAG ISP full-chip erase op-
eration, after which the device is blank and may be
used again. The 8032 on the MCU Module will al-
ways have access to PSM Module memory con-
tents through the 8-bit 8032 data bus connecting
the two die, even while the security bit is set.
Initial Delivery State. When delivered from ST-
Microelectronics, uPSD33xx devices are erased,
meaning all Flash memory and PLD configuration
bits are logic '1.' Firmware and PLD logic configu-
ration must be programmed at least the first time
using JTAG ISP. Subsequent programming of
Flash memory may be performed using JTAG ISP,
JTAG debugging, or the 8032 may run firmware to
program Flash memory (IAP).
The Debug signal should always be pulled up ex-
ternally with a weak pull-up (100K minimum) to
V
even if nothing is connected to it, as shown in
CC
Figure 81., page 196 and Figure 82., page 198.
201/231
uPSD33xx
AC/DC PARAMETERS
These tables describe the AD and DC parameters
of the uPSD33xx Devices:
The following are issues concerning the parame-
ters presented:
■
■
■
DC Electrical Specification
AC Timing Specification
PLD Timing
–
In the DC specification the supply current is
given for different modes of operation.
The AC power component gives the PLD,
Flash memory, and SRAM mA/MHz
specification. Figure 85 and Figure 86 show
the PLD mA/MHz as a function of the number
of Product Terms (PT) used.
–
–
–
–
–
Combinatorial Timing
Synchronous Clock Mode
Asynchronous Clock Mode
Input Macrocell Timing
–
In the PLD timing parameters, add the
required delay when Turbo Bit is '0.'
■
MCU Module Timing
–
–
–
READ Timing
WRITE Timing
Power-down and RESET Timing
Figure 85. PLD I /Frequency Consumption (5V range)
CC
110
100
90
V
= 5V
CC
80
70
60
50
40
30
20
10
0
PT 100%
PT 25%
0
5
10
15
20
25
HIGHEST COMPOSITE FREQUENCY AT PLD INPUTS (MHz)
AI02894
Figure 86. PLD I /Frequency Consumption (3V range)
CC
60
V
= 3V
CC
50
40
30
20
10
0
PT 100%
PT 25%
0
5
10
15
20
25
HIGHEST COMPOSITE FREQUENCY AT PLD INPUTS (MHz)
AI03100
202/231
uPSD33xx
Table 122. PSD Module Example, Typ. Power Calculation at V = 5.0V (Turbo Mode Off)
CC
Conditions
MCU Clock Frequency
Highest Composite PLD input frequency
(Freq PLD)
= 12MHz
= 8MHz
= 2MHz
= 80%
MCU ALE frequency (Freq ALE)
% Flash memory Access
% SRAM access
= 15%
% I/O access
= 5% (no additional power above base)
Operational Modes
% Normal
= 40%
= 60%
% Power-down Mode
Number of product terms used
(from fitter report)
= 45 PT
% of total product terms
Turbo Mode
= 45/182 = 24.7%
= Off
Calculation (using typical values)
I
total
= I (MCUactive) x %MCUactive + I (PSDactive) x %PSDactive + I (pwrdown) x %pwrdown
CC CC PD
CC
I
I
I
(MCUactive)
= 20mA
= 250uA
= I (ac) + I (dc)
CC
(pwrdown)
PD
(PSDactive)
CC
CC
CC
= %flash x 2.5mA/MHz x Freq ALE
+ %SRAM x 1.5mA/MHz x Freq ALE
+ % PLD x (from graph using Freq PLD)
= 0.8 x 2.5mA/MHz x 2MHz + 0.15 x 1.5mA/MHz x 2MHz + 24mA
= (4 + 0.45 + 24) mA
= 28.45mA
I
total
= 20mA x 40% + 28.45mA x 40% + 250uA x 60%
= 8mA + 11.38mA + 150uA
= 19.53mA
This is the operating power with no Flash memory Erase or Program cycles in progress. Calculation is based on all I/O
CC
pins being disconnected and I = 0mA.
OUT
203/231
uPSD33xx
MAXIMUM RATING
Stressing the device above the rating listed in the
Absolute Maximum Ratings” table may cause per-
manent damage to the device. These are stress
ratings only and operation of the device at these or
any other conditions above those indicated in the
Operating sections of this specification is not im-
plied. Exposure to Absolute Maximum Rating con-
ditions for extended periods may affect device
reliability. Refer also to the STMicroelectronics
SURE Program and other relevant quality docu-
ments.
Table 123. Absolute Maximum Ratings
Symbol
Parameter
Min.
Max.
125
235
6.5
Unit
°C
°C
V
T
Storage Temperature
Lead Temperature during Soldering (20 seconds max.)
–65
STG
(1)
TLEAD
VIO
Input and Output Voltage (Q = V
Supply Voltage
or Hi-Z)
–0.5
–0.5
OH
V
CC
6.5
V
V
Device Programmer Supply Voltage
Electrostatic Discharge Voltage (Human Body Model)
–0.5
14.0
2000
V
PP
(2)
VESD
–2000
V
Note: 1. IPC/JEDEC J-STD-020A
2. JEDEC Std JESD22-A114A (C1=100pF, R1=1500 Ω, R2=500 Ω)
DC AND AC PARAMETERS
This section summarizes the operating and mea-
surement conditions, and the DC and AC charac-
teristics of the device. The parameters in the DC
and AC Characteristic tables that follow are de-
rived from tests performed under the Measure-
ment Conditions summarized in the relevant
tables. Designers should check that the operating
conditions in their circuit match the measurement
conditions when relying on the quoted parame-
ters.
Table 124. Operating Conditions (5V Devices)
Symbol
Parameter
Min.
4.5
–40
0
Max.
5.5
85
Unit
V
V
CC
Supply Voltage
Ambient Operating Temperature (industrial)
Ambient Operating Temperature (commercial)
°C
°C
T
A
70
Table 125. Operating Conditions (3.3V Devices)
Symbol
Parameter
Min.
3.0
–40
0
Max.
3.6
85
Unit
V
V
CC
Supply Voltage
Ambient Operating Temperature (industrial)
Ambient Operating Temperature (commercial)
°C
°C
T
A
70
204/231
uPSD33xx
Table 126. AC Signal Letters for Timing
Table 127. AC Signal Behavior Symbols for
Timing
A
C
D
I
Address
t
Time
Clock
L
H
Logic Level Low or ALE
Logic Level High
Valid
Input Data
Instruction
ALE
V
X
Z
L
No Longer a Valid Logic Level
Float
N
P
Q
R
W
B
RESET Input or Output
PSEN signal
Output Data
RD signal
WR signal
PW
Pulse Width
Note: Example: t
= Time from Address Valid to ALE Invalid.
AVLX
V
STBY
Output
M
Output Macrocell
Note: Example: t
= Time from Address Valid to ALE Invalid.
AVLX
Figure 87. Switching Waveforms – Key
INPUTS
OUTPUTS
WAVEFORMS
STEADY INPUT
STEADY OUTPUT
MAY CHANGE FROM
HI TO LO
WILL BE CHANGING
FROM HI TO LO
MAY CHANGE FROM
LO TO HI
WILL BE CHANGING
LO TO HI
DON'T CARE
CHANGING, STATE
UNKNOWN
OUTPUTS ONLY
CENTER LINE IS
TRI-STATE
AI03102
205/231
uPSD33xx
Table 128. Major Parameters
Parameter
Test Conditions/Comments
5.0V Value
3.3V Value
Unit
V
4.5 to 5.5 (PSD);
3.0 to 3.6 (MCU)
3.0 to 3.6
(PSD and MCU)
Operating Voltage
–
–
Operating Temperature
MCU Frequency
–40 to 85
–40 to 85
°C
2
1 Min, 40 Max
1 Min, 40 Max
MHz
mA
mA
mA
mA
8MHz (min) for I C
40MHz Crystal, Turbo
40MHz Crystal, Non-Turbo
8MHz Crystal, Turbo
50
48
21
10
40
38
18
8
Active Current, Typical
(20% of PLD used; 25°C
operation)
8MHz Crystal, Non-Turbo
Idle Current, Typical
(20% of PLD used; 25°C
operation)
40MHz Crystal divided by 2048
internally.
All interfaces are disabled.
16
11
mA
Power-down Mode
needs reset to exit.
Standby Current, Typical
140
0.5
120
0.5
µA
µA
SRAM Backup Current, Typical
If external battery is attached.
V
V
= 0.45V (max);
I
= 8 (max);
= –2 (min)
I
OL
I
OH
= 4 (max);
= –1 (min)
I/O Sink/Source Current,
Ports A, B, C, and D
OL
OL
mA
= 2.4V (min)
I
OH
OH
V
V
= 0.6V (max);
I
= 10 (max);
= –10 (min)
I
OL
I
OH
= 10 (max);
= –10 (min)
I/O Sink/Source Current,
Port 4
OL
OL
mA
–
= 2.4V (min)
I
OH
OH
For registered or
combinatorial logic
PLD Macrocells
PLD Inputs
16
69
18
15
16
69
18
22
Inputs from pins, feedback,
or MCU addresses
–
Output to pins or
internal feedback
PLD Outputs
–
PLD Propagation Delay, Typical,
Turbo Mode
PLD input to output
ns
206/231
uPSD33xx
Table 129. Preliminary MCU Module DC Characteristics
Symbol
Parameter
Test Conditions
Min.
Typ.
Max.
Unit
(1)
V
3.0
3.6
V
CC
Supply Voltage
High Level Input Voltage
(Ports 0, 1, 2, 3, 4, XTAL1,
RESET)
V
3.0V < V < 3.6V
0.7V
CC
5.5
V
V
IH
CC
5V Tolerant - max voltage 5.5V
Low Level Input Voltage
(Ports 0, 1, 2, 3, 4, XTAL1,
RESET)
V
3.0V < V < 3.6V
V
– 0.5
0.3V
CC
IL
CC
SS
I
= 10mA
0.6
0.6
V
V
V
V
V
V
V
V
V
V
OL
V
Output Low Voltage (Port 4)
OL1
I
=5mA
OL
Output Low Voltage
(Other Ports)
V
OL2
OH1
OH2
OH3
I
= –10mA
= –5mA
= –20µA
2.4
OH
Output High Voltage
(Ports 4 push-pull)
V
V
V
I
2.4
2.4
OH
Output High Voltage
(Port 0 push-pull)
I
OH
Output High Voltage
(Other Ports Bi-directional mode)
XTAL Open Bias Voltage
(XTAL1, XTAL2)
V
I
= 3.2mA
= V
1.0
–10
–20
–10
–10
–10
2.0
–55
50
V
OP
OL
RESET Pin Pull-up Current
(RESET)
I
V
uA
uA
uA
uA
uA
uA
RST
IN
SS
XTAL Feedback Resistor Current
(XTAL1)
I
FR
XTAL1 = V ; XTAL2 = V
CC SS
Input High Leakage Current
(Port 0)
I
I
V
< V < 5.5V
10
IHL1
SS IN
Input High Leakage Current
(Port 1, 2, 3, 4)
V
= 2.3V
< 0.5V
= 3.6V
10
IHL2
IH
Input Low Leakage Current
(Port 1, 2, 3, 4)
I
V
10
ILL
IL
I
PD
V
Power-down Mode
65
95
CC
(Note 2)
Active - 12MHz
Idle - 12MHz
Active - 24MHz
Idle - 24MHz
Active - 40MHz
Idle - 40MHz
14
10
19
13
26
17
20
12
30
17
40
22
mA
mA
mA
mA
mA
mA
V
CC
V
CC
V
CC
= 3.6V
= 3.6V
= 3.6V
I
CC-CPU
(Note
3,4,5)
Note: 1. Power supply (V ) is always 3.0 to 3.6V for the MCU Module. V for the PSD Module may be 3V or 5V.
CC
DD
2. I (Power-down Mode) is measured with: XTAL1 = V ; XTAL2 = NC; RESET = V ; Port 0 = V ; all other pins are disconnected.
PD
SS
CC
CC
3. I
(Active Mode) is measured with: XTAL1 driven with t
, t
= 5ns, V = V + 0.5V, V = V – 0.5V, XTAL2 = NC;
CC-CPU
CLCH CHCL IL SS IH CC
RESET = V ; Port 0 = V ; all other pins are disconnected. I would be slightly higher if a crystal oscillator is used (approximately
SS
CC
CC
1mA).
4. I
(Idle Mode) is measured with: XTAL1 driven with t
, t
= 5ns, V = V + 0.5V, V = V – 0.5V, XTAL2 = NC;
CC-CPU
CLCH CHCL IL SS IH CC
RESET = V ; Port 0 = V ; all other pins are disconnected. I would be slightly higher if a crystal oscillator is used (approximately
CC
CC
CC
1mA). All IP clocks are disabled.
5. I/O current = 0mA, all I/O pins are disconnected.
207/231
uPSD33xx
Table 130. PSD Module DC Characteristics (with 5V V
)
DD
Test Condition
Symbol
Parameter
(in addition to those in
Table 129., page 207)
Min.
Typ.
Max.
+0.5
Unit
V
IH
4.5V < V < 5.5V
V
Input High Voltage
Input Low Voltage
2
V
V
DD
DD
V
IL
4.5V < V < 5.5V
–0.5
0.8
DD
V
(min) for Flash Erase and
DD
V
LKO
2.5
4.2
V
Program
I
= 20uA, V = 4.5V
0.01
0.25
4.49
3.9
0.1
V
V
OL
DD
V
Output Low Voltage
OL
I
= 8mA, V = 4.5V
0.45
OL
DD
I
= –20uA, V = 4.5V
4.4
2.4
V
OH
DD
Output High Voltage Except
V
OH
V
STBY
On
I
= –2mA, V = 4.5V
V
OH
DD
V
OH1
Output High Voltage V
On
I
= 1uA
V
– 0.8
STBY
V
STBY
OH1
V
I
V
SRAM Stand-by Voltage
SRAM Stand-by Current
2.0
V
STBY
DD
V
= 0V
0.5
1
uA
uA
V
STBY
DD
I
Idle Current (V
input)
V
> V
DD STBY
–0.1
2
0.1
IDLE
STBY
V
Only on V
V
– 0.2
DD
SRAM Data Retention Voltage
DF
STBY
CSI > V – 0.3V
DD
Stand-by Supply Current
for Power-down Mode
I
SB
120
250
uA
(Notes 1,2)
I
V
< V < V
SS IN DD
Input Leakage Current
Output Leakage Current
–1
±0.1
±5
1
uA
uA
LI
I
LO
0.45 < V
< V
OUT DD
–10
10
PLD_TURBO = Off,
f = 0MHz (Note 4)
0
uA/PT
uA/PT
mA
PLD Only
PLD_TURBO = On,
f = 0MHz
400
15
700
30
Operating
Supply
Current
I
(DC)
CC
During Flash memory
WRITE/Erase Only
(Note 4)
Flash memory
Read only, f = 0MHz
f = 0MHz
0
0
0
0
mA
mA
SRAM
PLD AC Adder
Note 3
2.5
mA/
MHz
I
(AC)
Flash memory AC Adder
SRAM AC Adder
1.5
1.5
CC
(Note 4)
mA/
MHz
3.0
Note: 1. Internal Power-down mode is active.
2. PLD is in non-Turbo mode, and none of the inputs are switching.
3. Please see Figure 85., page 202 for the PLD current calculation.
4. I
= 0mA
OUT
208/231
uPSD33xx
Table 131. PSD Module DC Characteristics (with 3.3V V
Test Condition
)
DD
Symbol
Parameter
(in addition to those in
Table 129., page 207)
Min.
0.7V
Typ.
Max.
+0.5
Unit
V
IH
3.0V < V < 3.6V
V
DD
High Level Input Voltage
Low Level Input Voltage
V
V
DD
DD
V
IL
3.0V < V < 3.6V
–0.5
1.5
0.8
DD
V
(min) for Flash Erase and
DD
V
LKO
2.2
V
Program
I
= 20uA, V = 3.0V
0.01
0.15
2.99
2.8
0.1
V
V
OL
DD
V
Output Low Voltage
OL
I
= 4mA, V = 3.0V
0.45
OL
DD
I
= –20uA, V = 3.0V
2.9
2.7
V
OH
DD
Output High Voltage Except
V
OH
V
STBY
On
I
= –1mA, V = 3.0V
V
OH
DD
V
OH1
Output High Voltage V
On
I
= 1uA
V
– 0.8
STBY
V
STBY
OH1
V
I
V
SRAM Stand-by Voltage
SRAM Stand-by Current
2.0
V
STBY
DD
V
= 0V
0.5
1
uA
uA
V
STBY
DD
I
Idle Current (V
input)
V
> V
DD STBY
–0.1
2
0.1
IDLE
STBY
V
Only on V
V
– 0.2
DD
SRAM Data Retention Voltage
DF
STBY
CSI > V – 0.3V
Stand-by Supply Current
for Power-down Mode
DD
I
SB
50
100
uA
(Notes 1,2)
I
V
< V < V
SS IN DD
Input Leakage Current
Output Leakage Current
–1
±0.1
±5
1
uA
uA
LI
I
LO
0.45 < V < V
IN DD
–10
10
PLD_TURBO = Off,
f = 0MHz (Note 2)
0
uA/PT
uA/PT
mA
PLD Only
PLD_TURBO = On,
f = 0MHz
200
10
400
25
Operating
Supply
Current
I
(DC)
CC
During Flash memory
WRITE/Erase Only
(Note 4)
Flash memory
Read only, f = 0MHz
f = 0MHz
0
0
0
0
mA
mA
SRAM
PLD AC Adder
Note 3
mA/
MHz
I
(AC)
Flash memory AC Adder
SRAM AC Adder
1.0
0.8
1.5
1.5
CC
(Note 4)
mA/
MHz
Note: 1. Internal PD is active.
2. PLD is in non-Turbo mode, and none of the inputs are switching.
3. Please see Figure 86., page 202 for the PLD current calculation.
4. I
= 0mA
OUT
209/231
uPSD33xx
Figure 88. External PSEN/READ Cycle (80-pin Device Only)
t
t
LLPL
LHLL
ALE
t
t
PLPH
AVLL
PSEN
RD
t
t
PXAV
LLAX
t
PXIZ
t
AZPL
MCU
AD0 - AD7
INSTR
IN
A0-A7
A0-A7
t
AVIV
t
PXIX
MCU
A8 - A11
A8-A11
A8-A11
AI07875
Table 132. External PSEN or READ Cycle AC Characteristics (3V or 5V Device)
Variable Oscillator
(1)
40MHz Oscillator
1/t
= 8 to 40MHz
CLCL
Symbol
Parameter
Unit
Min
17
Max
Min
Max
t
t
– 8
ALE pulse width
ns
ns
ns
ns
ns
LHLL
CLCL
t
t
– 12
CLCL
Address setup to ALE
Address hold after ALE
ALE to PSEN or RD
13
AVLL
t
0.5t
0.5t
nt
– 5
7.5
7.5
40
LLAX
CLCL
CLCL
t
– 5
LLPL
(2)
t
– 10
PLPH
PSEN or RD pulse width
CLCL
Input instruction/data hold after
PSEN or RD
t
2
2
ns
ns
PXIX
Input instruction/data float after
PSEN or RD
t
0.5t
mt
– 2
– 5
10.5
70
PHIZ
CLCL
t
0.5t
– 5
CLCL
Address hold after PSEN or RD
7.5
–2
ns
ns
ns
PXAV
(2)
t
AVIV
CLCL
Address to valid instruction/data in
Address float to PSEN or RD
t
–2
AZPL
Note: 1. BUSCON Register is configured for 4 PFQCLK.
2. Refer to Table 133 for “n” and “m” values.
Table 133. n, m, and x, y Values
PSEN (code) Cycle
# of PFQCLK in
READ Cycle
WRITE Cycle
BUSCON Reg.
n
1
2
3
4
-
m
2
3
4
5
-
n
-
m
-
x
-
y
-
3
4
5
6
7
2
3
4
5
3
4
5
6
2
3
4
5
1
2
3
4
210/231
uPSD33xx
Figure 89. External WRITE Cycle (80-pin Device Only)
ALE
tLHLL
tWHLH
PSEN
tLLWL
tWLWH
WR
tWHQX
tAVLL
tLLAX
tQVWH
DATA OUT
MCU
AD0 - AD7
A0-A7
INSTR IN
A0-A7
tAVWL
MCU
A8 - A11
A8-A11
A8-A11
AI07877
Table 134. External WRITE Cycle AC Characteristics (3V or 5V Device)
Variable Oscillator
(1)
40MHz Oscillator
1/t
= 8 to 40MHz
CLCL
Symbol
Parameter
Unit
Min
17
Max
Min
Max
t
t
– 8
ALE pulse width
ns
ns
ns
ns
ns
ns
ns
ns
ns
LHLL
CLCL
t
t
– 12
CLCL
Address Setup to ALE
Address hold after ALE
13
AVLL
t
0.5t
xt
– 5
CLCL
7.5
40
LLAX
(2)
t
– 10
– 5
WLWH
CLCL
WR pulse width
t
0.5t
1.5t
ALE to WR
7.5
27.5
6.5
20
LLWL
CLCL
t
– 10
CLCL
Address valid to WR
WR High to ALE High
AVWL
t
0.5t
yt
– 6 0.5t
+ 2
+ 2
14.5
14.5
WHLH
CLCL
CLCL
CLCL
(y)
t
– 5
QVWH
CLCL
Data setup before WR
Data hold after WR
t
0.5t
– 6 0.5t
CLCL
6.5
WHQX
Note: 1. BUSCON Register is configured for 4 PFQCLK.
2. Refer to Table 135, page 151 for “n” and “m” values.
Table 135. External Clock Drive
Variable Oscillator
40MHz Oscillator
Min Max
1/t
= 8 to 40MHz
CLCL
(1)
Symbol
Unit
Parameter
Min
Max
t
Oscillator period
25
10
10
125
ns
ns
ns
ns
ns
CLCL
t
t
t
– t
– t
10
10
High time
Low time
Rise time
Fall time
CHCX
CLCL
CLCX
t
CLCX
CLCL
CLCX
t
CLCH
t
CHCL
211/231
uPSD33xx
Table 136. A/D Analog Specification
(1)
Symbol
Parameter
Min.
Typ.
Max.
Unit
mA
uA
V
Test Conditions
Input = AV
Normal
4.0
REF
I
DD
Power-down
40
AV
AV
REF
Analog Input Voltage
GND
IN
(2)
Analog Reference Voltage
3.6
V
AV
REF
Accuracy Resolution
10
±2
bits
Input = 0 to AV
(V)
REF
INL
Integral Nonlinearity
LSB
LSB
F
≤ 32MHz
OSC
Input = 0 to AV
(V)
REF
DNL
Differential Nonlinearity
±2
F
≤ 32MHz
OSC
f
= 500ksps
SNR
SNDR
ACLK
Signal to Noise Ratio
Signal to Noise Distortion Ratio
ADC Clock
50
48
2
54
52
8
dB
dB
SAMPLE
16
8
MHz
µs
t
C
Conversion Time
8MHz
Calibration Time
1
4
t
Power-up Time
16
ms
CAL
f
Analog Input Frequency
Total Harmonic Distortion
60
kHz
dB
IN
THD
50
54
Note: 1. f 2kHz, ACLK = 8MHz, AV
= V = 3.3V
CC
IN
REF
2. AV
= V in 52-pin package.
CC
REF
212/231
uPSD33xx
Figure 90. Input to Output Disable / Enable
INPUT
tER
tEA
INPUT TO
OUTPUT
ENABLE/DISABLE
AI02863
Table 137. CPLD Combinatorial Timing (5V PSD Module)
Slew
PT Turbo
Symbol
Parameter
Conditions
Min
Max
20
Unit
(1)
Aloc
Off
rate
CPLD Input Pin/Feedback to
CPLD Combinatorial Output
(2)
+ 2
+ 10
– 2
– 2
– 2
– 2
ns
ns
ns
ns
ns
ns
t
PD
CPLD Input to CPLD Output
Enable
t
EA
21
+ 10
+ 10
+ 10
+ 10
CPLD Input to CPLD Output
Disable
t
ER
21
CPLD Register Clear or Preset
Delay
t
21
ARP
CPLD Register Clear or Preset
Pulse Width
t
10
ARPW
Any
macrocell
t
CPLD Array Delay
11
+ 2
ARD
Note: 1. Fast Slew Rate output available on PA3-PA0, PB3-PB0, and PD2-PD1. Decrement times by given amount
2. t for MCU address and control signals refers to delay from pins on Port 0, Port 2, RD WR, PSEN and ALE to CPLD combinatorial
PD
output (80-pin package only)
Table 138. CPLD Combinatorial Timing (3V PSD Module)
Slew
PT Turbo
Symbol
Parameter
Conditions
Min
Max
35
Unit
ns
(1)
Aloc
Off
rate
CPLD Input Pin/Feedback to
CPLD Combinatorial Output
(2)
+ 4
+ 20
– 6
t
PD
CPLD Input to CPLD Output
Enable
t
38
+ 20
+ 20
+ 20
+ 20
– 6
– 6
– 6
ns
EA
CPLD Input to CPLD Output
Disable
t
ER
38
ns
CPLD Register Clear or
Preset Delay
t
35
ns
ARP
CPLD Register Clear or
Preset Pulse Width
t
18
ns
ARPW
Any
macrocell
t
CPLD Array Delay
20
+ 4
ns
ARD
Note: 1. Fast Slew Rate output available on PA3-PA0, PB3-PB0, and PD2-PD1. Decrement times by given amount
2. t for MCU address and control signals refers to delay from pins on Port 0, Port 2, RD WR, PSEN and ALE to CPLD combinatorial
PD
output (80-pin package only)
213/231
uPSD33xx
Figure 91. Synchronous Clock Mode Timing – PLD
t
t
CL
CH
CLKIN
INPUT
t
S
t
H
t
CO
REGISTERED
OUTPUT
AI02860
Table 139. CPLD Macrocell Synchronous Clock Mode Timing (5V PSD Module)
Slew
PT Turbo
Symbol
Parameter
Conditions
Min
Max
40.0
66.6
83.3
Unit
MHz
MHz
MHz
(1)
Aloc
Off
rate
Maximum Frequency
External Feedback
1/(t +t
)
S
CO
Maximum Frequency
f
1/(t +t –10)
S CO
MAX
Internal Feedback (f
)
CNT
Maximum Frequency
Pipelined Data
1/(t +t
)
CH CL
t
Input Setup Time
Input Hold Time
12
0
+ 2
+ 10
ns
ns
ns
ns
ns
ns
ns
S
t
H
t
Clock High Time
Clock Low Time
Clock Input
Clock Input
Clock Input
Any macrocell
6
CH
t
CL
6
t
Clock to Output Delay
CPLD Array Delay
13
11
– 2
CO
t
+ 2
ARD
(2)
t
t +t
CH CL
12
MIN
Minimum Clock Period
Note: 1. Fast Slew Rate output available on PA3-PA0, PB3-PB0, and PD2-PD1. Decrement times by given amount.
2. CLKIN (PD1) t
3.
= t + t .105
CH CL
CLCL
214/231
uPSD33xx
Table 140. CPLD Macrocell Synchronous Clock Mode Timing (3V PSD Module)
PT
Slew
Turbo
Off
Symbol
Parameter
Conditions
Min
Max
23.2
30.3
40.0
Unit
(1)
Aloc
rate
Maximum Frequency
External Feedback
1/(t +t
)
MHz
MHz
MHz
S
CO
Maximum Frequency
f
1/(t +t –10)
S CO
MAX
Internal Feedback (f
)
CNT
Maximum Frequency
Pipelined Data
1/(t +t
)
CH CL
t
Input Setup Time
Input Hold Time
20
0
+ 4
+ 15
ns
ns
ns
ns
S
t
H
t
Clock High Time
Clock Low Time
Clock Input
Clock Input
Clock Input
Any macrocell
15
10
CH
t
CL
t
Clock to Output Delay
CPLD Array Delay
23
20
– 6
ns
ns
ns
CO
t
+ 4
ARD
(2)
t
t +t
CH CL
25
MIN
Minimum Clock Period
Note: 1. Fast Slew Rate output available on PA3-PA0, PB3-PB0, and PD2-PD1. Decrement times by given amount.
2. CLKIN (PD1) t = t + t
.
CL
CLCL
CH
215/231
uPSD33xx
Figure 92. Asynchronous RESET / Preset
tARPW
RESET/PRESET
INPUT
tARP
REGISTER
OUTPUT
AI02864
Figure 93. Asynchronous Clock Mode Timing (Product Term Clock)
tCHA
tCLA
CLOCK
INPUT
tSA
tHA
tCOA
REGISTERED
OUTPUT
AI02859
Table 141. CPLD Macrocell Asynchronous Clock Mode Timing (5V PSD Module)
PT Turbo Slew
Symbol
Parameter
Conditions
1/(t +t
Min
Max
38.4
62.5
71.4
Unit
MHz
MHz
MHz
Aloc
Off
Rate
Maximum Frequency
External Feedback
)
SA COA
Maximum Frequency
f
1/(t +t
–10)
)
MAXA
SA COA
Internal Feedback (f
)
CNTA
Maximum Frequency
Pipelined Data
1/(t
+t
CHA CLA
t
Input Setup Time
7
8
9
9
+ 2
+ 10
ns
ns
ns
ns
ns
ns
ns
SA
t
Input Hold Time
HA
t
Clock Input High Time
Clock Input Low Time
Clock to Output Delay
CPLD Array Delay
Minimum Clock Period
+ 10
+ 10
+ 10
CHA
t
CLA
t
21
11
– 2
COA
t
Any macrocell
+ 2
ARDA
t
1/f
CNTA
16
MINA
216/231
uPSD33xx
Table 142. CPLD Macrocell Asynchronous Clock Mode Timing (3V PSD Module)
PT Turbo Slew
Symbol
Parameter
Conditions
1/(t +t
Min
Max
21.7
27.8
33.3
Unit
MHz
MHz
MHz
Aloc
Off
Rate
Maximum Frequency
External Feedback
)
SA COA
Maximum Frequency
f
1/(t +t
–10)
)
MAXA
SA COA
Internal Feedback (f
)
CNTA
Maximum Frequency
Pipelined Data
1/(t
+t
CHA CLA
t
Input Setup Time
Input Hold Time
10
12
17
13
+ 4
+ 15
ns
ns
ns
ns
ns
ns
ns
SA
t
HA
t
Clock High Time
+ 15
+ 15
+ 15
CHA
t
Clock Low Time
CLA
t
Clock to Output Delay
CPLD Array Delay
Minimum Clock Period
31
20
– 6
COA
t
Any macrocell
+ 4
ARD
t
1/f
CNTA
36
MINA
217/231
uPSD33xx
Figure 94. Input Macrocell Timing (Product Term Clock)
t
t
INL
INH
PT CLOCK
INPUT
t
t
IH
IS
OUTPUT
t
INO
AI03101
Table 143. Input Macrocell Timing (5V PSD Module)
PT
Aloc
Turbo
Off
Symbol
Parameter
Input Setup Time
Conditions
Min
Max
Unit
t
IS
0
15
9
ns
ns
ns
ns
ns
(Note 1)
(Note 1)
(Note 1)
(Note 1)
(Note 1)
t
IH
Input Hold Time
+ 10
t
NIB Input High Time
NIB Input Low Time
INH
t
9
INL
t
NIB Input to Combinatorial Delay
34
+ 2
+ 10
INO
Note: 1. Inputs from Port A, B, and C relative to register/ latch clock from the PLD. ALE/AS latch timings refer to t
and t
.
LXAX
AVLX
Table 144. Input Macrocell Timing (3V PSD Module)
PT
Aloc
Turbo
Off
Symbol
Parameter
Input Setup Time
Conditions
Min
Max
Unit
t
IS
0
ns
ns
ns
ns
ns
(Note 1)
(Note 1)
(Note 1)
(Note 1)
(Note 1)
t
IH
Input Hold Time
25
12
12
+ 15
t
NIB Input High Time
NIB Input Low Time
INH
t
INL
t
NIB Input to Combinatorial Delay
43
+ 4
+ 15
INO
Note: 1. Inputs from Port A, B, and C relative to register/latch clock from the PLD. ALE latch timings refer to t
and t
.
LXAX
AVLX
218/231
uPSD33xx
Table 145. Program, WRITE and Erase Times (5V, 3V PSD Modules)
Symbol
Parameter
Min.
Typ.
Max.
10
Unit
Flash Program
8.5
s
(1)
(2)
s
Flash Bulk Erase (pre-programmed)
Flash Bulk Erase (not pre-programmed)
Sector Erase (pre-programmed)
Sector Erase (not pre-programmed)
Byte Program
3
5
1
s
s
t
10
WHQV3
t
2.2
14
s
WHQV2
t
150
µs
WHQV1
Program/Erase Cycles (per Sector)
PLD Program/Erase Cycles
100,000
1000
cycles
cycles
µs
t
Sector Erase Time-Out
100
WHWLO
(3)
t
30
ns
Q7VQV
DQ7 Valid to Output (DQ7-DQ0) Valid (Data Polling)
Note: 1. Programmed to all zero before erase.
2. Typical after 100K program/erase cycle is 5 seconds.
3. The polling status, DQ7, is valid t
time units before the data byte, DQ0-DQ7, is valid for reading.
Q7VQV
219/231
uPSD33xx
Figure 95. Peripheral I/O READ Timing
ALE
ADDRESS
DATA VALID
A/D BUS
t
(PA)
(PA)
AVQV
t
SLQV
CSI
RD
t
(PA)
RLQV
t
(PA)
RHQZ
t
(PA)
DVQV
DATA ON PORT A
AI06610
Table 146. Port A Peripheral Data Mode READ Timing (5V PSD Module)
Turbo
Symbol
Parameter
Conditions
Min
Max
Unit
Off
+ 10
+ 10
Address Valid to Data
t
37
ns
AVQV–PA
(Note 1)
Valid
t
t
t
CSI Valid to Data Valid
27
32
22
23
ns
ns
ns
ns
SLQV–PA
RD to Data Valid
RLQV–PA
(Note 2)
Data In to Data Out Valid
RD to Data High-Z
DVQV–PA
RHQZ–PA
t
Note: 1. Any input used to select Port A Data Peripheral Mode.
2. Data is already stable on Port A.
Table 147. Port A Peripheral Data Mode READ Timing (3V PSD Module)
Turbo
Off
Symbol
Parameter
Conditions
Min
Max
Unit
t
Address Valid to Data Valid
CSI Valid to Data Valid
RD to Data Valid
50
37
45
38
36
+ 20
+ 20
ns
ns
ns
ns
ns
AVQV–PA
SLQV–PA
RLQV–PA
(Note 1)
t
t
t
(Note 2)
Data In to Data Out Valid
RD to Data High-Z
DVQV–PA
RHQZ–PA
t
Note: 1. Any input used to select Port A Data Peripheral Mode.
2. Data is already stable on Port A.
220/231
uPSD33xx
Figure 96. Peripheral I/O WRITE Timing
ALE
ADDRESS
DATA OUT
A/D BUS
tWHQZ (PA)
tWLQV (PA)
WR
tDVQV (PA)
PORT A
DATA OUT
AI06611
Table 148. Port A Peripheral Data Mode WRITE Timing (5V PSD Module)
Symbol
Parameter
WR to Data Propagation Delay
Data to Port A Data Propagation Delay
WR Invalid to Port A Tri-state
Conditions
Min
Min
Typ
Max
Unit
ns
t
25
22
20
WLQV–PA
t
ns
DVQV–PA
(Note 1)
t
ns
WHQZ–PA
Note: 1. Data stable on Port 0 pins to data on Port A.
Table 149. Port A Peripheral Data Mode WRITE Timing (3V PSD Module)
Symbol
Parameter
WR to Data Propagation Delay
Data to Port A Data Propagation Delay
WR Invalid to Port A Tri-state
Conditions
Max
42
Unit
ns
t
WLQV–PA
t
38
ns
DVQV–PA
WHQZ–PA
(Note 1)
t
33
ns
Note: 1. Data stable on Port 0 pins to data on Port A.
Table 150. Supervisor Reset and LVD
Symbol
RST_LO_IN
RST_ACTV
RST_FIL
Parameter
Conditions
Min
Max
Unit
(1)
t
t
t
Reset Input Duration
µs
1
(2)
f
= 40MHz
Generated Reset Duration
Reset Input Spike Filter
Reset Input Hysteresis
LVD Trip Threshold
ms
µs
V
OSC
10
1
V
V
V
V
= 3.3V
= 3.3V
0.1
2.6
RST_HYS
CC
CC
2.4
2.8
V
RST_THRESH
Note: 1. 25µs minimum to abort a Flash memory program or erase cycle in progress.
2. As F decreases, t increases. Example: t = 50ms when F = 8MHz.
OSC
OSC
RST_ACTV
RST_ACTV
221/231
uPSD33xx
Table 151. V
Symbol
Definitions Timing (5V, 3V PSD Modules)
STBYON
Parameter
Conditions
Min
Typ
Max
Unit
t
V
V
Detection to V
Output High
STBYON
20
µs
BVBH
STBY
(Note 1)
Off Detection to V
Output
STBY
STBYON
t
20
µs
BXBL
(Note 1)
Low
Note: 1. V
timing is measured at V ramp rate of 2ms.
CC
STBYON
Figure 97. ISC Timing
tISCCH
TCK
tISCCL
tISCPSU
tISCPH
TDI/TMS
t ISCPZV
tISCPCO
ISC OUTPUTS/TDO
tISCPVZ
ISC OUTPUTS/TDO
AI02865
Table 152. ISC Timing (5V PSD Module)
Symbol
Parameter
Conditions
Min
Max
Unit
MHz
ns
t
Clock (TCK, PC1) Frequency (except for PLD)
Clock (TCK, PC1) High Time (except for PLD)
Clock (TCK, PC1) Low Time (except for PLD)
Clock (TCK, PC1) Frequency (PLD only)
Clock (TCK, PC1) High Time (PLD only)
20
ISCCF
(Note 1)
(Note 1)
(Note 1)
(Note 2)
(Note 2)
(Note 2)
t
23
23
ISCCH
t
ns
ISCCL
t
5
MHz
ns
ISCCFP
t
90
ISCCHP
t
Clock (TCK, PC1) Low Time (PLD only)
ISC Port Set Up Time
90
7
ns
ns
ns
ns
ns
ns
ISCCLP
t
ISCPSU
t
ISC Port Hold Up Time
5
ISCPH
t
ISC Port Clock to Output
21
21
21
ISCPCO
t
ISC Port High-Impedance to Valid Output
ISC Port Valid Output to High-Impedance
ISCPZV
t
ISCPVZ
Note: 1. For non-PLD Programming, Erase or in ISC By-pass Mode.
2. For Program or Erase PLD only.
222/231
uPSD33xx
Table 153. ISC Timing (3V PSD Module)
Symbol
Parameter
Conditions
(Note 1)
(Note 1)
(Note 1)
(Note 2)
(Note 2)
(Note 2)
Min
Max
Unit
MHz
ns
t
Clock (TCK, PC1) Frequency (except for PLD)
Clock (TCK, PC1) High Time (except for PLD)
Clock (TCK, PC1) Low Time (except for PLD)
Clock (TCK, PC1) Frequency (PLD only)
Clock (TCK, PC1) High Time (PLD only)
12
5
ISCCF
t
40
40
ISCCH
t
ns
ISCCL
t
MHz
ns
ISCCFP
t
90
ISCCHP
t
Clock (TCK, PC1) Low Time (PLD only)
ISC Port Set Up Time
90
12
5
ns
ns
ns
ns
ns
ns
ISCCLP
t
ISCPSU
t
ISC Port Hold Up Time
ISCPH
t
ISC Port Clock to Output
30
30
30
ISCPCO
t
ISC Port High-Impedance to Valid Output
ISC Port Valid Output to High-Impedance
ISCPZV
t
ISCPVZ
Note: 1. For non-PLD Programming, Erase or in ISC By-pass Mode.
2. For Program or Erase PLD only.
Figure 98. MCU Module AC Measurement I/O Waveform
V
– 0.5V
0.45V
CC
0.2 V
0.2 V
+ 0.9V
CC
Test Points
– 0.1V
CC
AI06650
Note: AC inputs during testing are driven at V –0.5V for a logic '1,' and 0.45V for a logic '0.'
CC
Timing measurements are made at V (min) for a logic '1,' and V (max) for a logic '0'
IH
IL
Figure 99. PSD Module AC Float I/O Waveform
V
V
– 0.1V
OH
OL
V
V
+ 0.1V
LOAD
Test Reference Points
– 0.1V
– 0.1V
+ 0.1V
LOAD
0.2 V
CC
AI06651
Note: For timing purposes, a Port pin is considered to be no longer floating when a 100mV change from load voltage occurs, and begins to
float when a 100mV change from the loaded V or V level occurs
OH
OL
I
and I ≥ 20mA
OH
OL
223/231
uPSD33xx
Figure 100. External Clock Cycle
Figure 101. PSD Module AC Measurement I/O
Waveform
Figure 102. PSD Module AC Measurement
Load Circuit
2.01 V
3.0V
195 Ω
Test Point
1.5V
Device
Under Test
0V
CL = 30 pF
(Including Scope and
AI03103b
Jig Capacitance)
AI03104b
Table 154. I/O Pin Capacitance
(1)
2
Symbol
Test Condition
Max.
Unit
pF
Parameter
Typ.
C
V
= 0V
= 0V
Input Capacitance (for input pins)
Output Capacitance (for input/
4
6
IN
IN
pF
C
V
OUT
8
12
OUT
(3)
output pins)
Note: 1. Sampled only, not 100% tested.
2. Typical values are for T = 25°C and nominal supply voltages.
A
3. Maximum for MCU Address and Data lines is 20pF each.
224/231
uPSD33xx
PACKAGE MECHANICAL INFORMATION
Figure 103. TQFP52 – 52-lead Plastic Thin, Quad, Flat Package Outline
D
D1
D2
A2
e
Ne
E2 E1 E
b
N
1
Nd
A
CP
L1
c
A1
α
L
QFP-A
Note: Drawing is not to scale.
225/231
uPSD33xx
Table 155. TQFP52 – 52-lead Plastic Thin, Quad, Flat Package Mechanical Data
mm
Min
–
inches
Min
–
Symb
Typ
1.50
0.10
1.40
–
Max
1.70
0.20
1.50
0.40
0.20
12.20
10.20
7.93
12.20
10.20
7.93
–
Typ
0.059
0.004
0.055
–
Max
0.067
0.008
0.059
0.016
0.008
0.480
0.402
0.312
0.480
0.402
0.312
–
A
A1
A2
b
0.05
1.30
0.20
0.07
11.80
9.80
7.67
11.80
9.80
7.67
–
0.002
0.039
0.008
0.003
0.465
0.386
0.302
0.465
0.386
0.302
–
c
–
–
D
12.00
10.00
7.80
12.00
10.00
7.80
0.65
–
0.472
0.394
0.307
0.472
0.394
0.307
0.026
–
D1
D2
E
E1
E2
e
L
0.45
–
0.75
–
0.018
–
0.030
–
L1
α
1.00
–
0.039
–
0°
7°
0°
7°
n
52
52
Nd
Ne
CP
13
13
13
13
–
–
0.10
–
–
0.004
226/231
uPSD33xx
Figure 104. TQFP80 – 80-lead Plastic Thin, Quad, Flat Package Outline
D
D1
D2
A2
e
Ne
E2 E1 E
b
N
1
Nd
A
CP
L1
c
A1
α
L
QFP-A
Note: Drawing is not to scale.
227/231
uPSD33xx
Table 156. TQFP80 – 80-lead Plastic Thin, Quad, Flat Package Mechanical Data
mm
Min
inches
Min
Symb
Typ
Max
1.60
0.15
1.45
0.27
0.20
Typ
Max
A
A1
A2
b
0.063
0.006
0.057
0.011
0.008
0.05
1.35
0.17
0.09
0.002
0.053
0.007
0.004
1.40
0.055
c
D
14.00
12.00
9.50
0.551
0.472
0.374
0.551
0.472
0.374
0.020
D1
D2
E
14.00
12.00
9.50
E1
E2
e
0.50
L
0.45
0.75
7°
0.018
0.030
7°
L1
α
1.00
0.039
0°
80
20
20
0°
80
20
20
n
Nd
Ne
CP
0.08
0.003
228/231
uPSD33xx
PART NUMBERING
Table 157. Ordering Information Scheme
Example:
uPSD 33
3
4
D
V
–
40
U
6
T
Device Type
uPSD = Microcontroller PSD
Family
33 = Turbo core
SRAM Size
1 = 2Kbyte
3 = 8Kbyte
5 = 32Kbyte
Main Flash Memory Size
2 = 64Kbyte
3 = 128Kbyte
4 = 256Kbyte
IP Mix
2
D = IP Mix: I C, SPI, UART (2), IrDA, ADC, Supervisor, PCA
Operating Voltage
blank = V = 4.5 to 5.5V
CC
V = V = 3.0 to 3.6V
CC
Speed
–40 = 40MHz
Package
T = 52-pin TQFP
U = 80-pin TQFP
Temperature Range
6 = –40 to 85°C
Shipping Option
Tape & Reel Packing = T
For a list of available options (e.g., Speed, Package) or for further information on any aspect of this device,
please contact the ST Sales Office nearest to you.
229/231
uPSD33xx
REVISION HISTORY
Table 158. Document Revision History
Date
Version
Revision Details
July 1, 2003
1.0
First Issue
Update register information, electrical characteristics (Table 17, 46, 132, 133, 134, 135;
Figure 68)
15-Jul-03
1.1
03-Sep-03
05-Feb-04
1.2
2.0
Update references for Product Catalog
Reformatted; corrected mechanical dimensions (Table 158)
Reformatted; update characteristics (Figure 3, 4, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60,
61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83,
84; Table 42, 64, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93,
94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112,
113, 114, 115, 116, 117, 118, 121, 129, 130, 131, 136)
07-May-04
3.0
Reformatted; updated Feature Summary; added table (Table 128); updated graphics,
mechanical dimensions (Figure 3, 4, 37, 40, 51, 76, 80; Table 2, 3, 6, 7, 8, 9, 10, 11, 37,
38, 40, 51, 77, 84, 119, 120, 121, 129, 155, 156)
14-Sep-04
4.0
29-Oct-04
21-Jan-05
5.0
6.0
Corrected TQFP80 mechanical dimensions (Table 156)
Updated characteristics, SPI section (Figure 3, 41, 42, 45; Table 59, 60, 61, 62, 128,
138, 140, 142, 144, 145, 152, 153)
230/231
uPSD33xx
Information furnished is believed to be accurate and reliable. However, STMicroelectronics assumes no responsibility for the consequences
of use of such information nor for any infringement of patents or other rights of third parties which may result from its use. No license is granted
by implication or otherwise under any patent or patent rights of STMicroelectronics. Specifications mentioned in this publication are subject
to change without notice. This publication supersedes and replaces all information previously supplied. STMicroelectronics products are not
authorized for use as critical components in life support devices or systems without express written approval of STMicroelectronics.
The ST logo is a registered trademark of STMicroelectronics.
All other names are the property of their respective owners
© 2005 STMicroelectronics - All rights reserved
STMicroelectronics group of companies
Australia - Belgium - Brazil - Canada - China - Czech Republic - Finland - France - Germany - Hong Kong - India - Israel - Italy - Japan -
Malaysia - Malta - Morocco - Singapore - Spain - Sweden - Switzerland - United Kingdom - United States of America
www.st.com
231/231
相关型号:
UPSD3412C-24T1T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412C-24T6T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412C-24U1T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412C-24U6T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412C-40T1T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412C-40T6T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412C-40U1T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412C-40U6T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412CV-24T1T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412CV-24T6T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412CV-24U1T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
UPSD3412CV-24U6T
Flash Programmable System Devices with 8032 Microcontroller Core and 16Kbit SRAM
STMICROELECTR
©2020 ICPDF网 联系我们和版权申明