TMP88CS38B/CM38B/CP38B CMOS 8-Bit Microcontroller TMP88CS38BNG/FG, TMP88CM38BNG/FG, TMP88CP38BNG/FG The TMP88CS38B/CM38B/CP38B is the high speed and high performance 8-bit single chip microcomputers. This MCU contain CPU core, ROM, RAM, input/output ports, four Multi-function timer/counters, serial bus interface, on-screen display, PWM output, 8-bit AD converter, and remote control signal preprocessor on chip. Product No. ROM RAM 2 K × 8 bits TMP88CS38BNG/FG 64 K × 8 bits TMP88CM38BNG/FG 32 K × 8 bits TMP88CP38BNG/FG 48 K × 8 bits 1.5 K × 8 bits Package SDIP42-P-600-1.78 P-QFP44-1414-0.80K OTP MCU TMP88PS38BNG/FG Features ♦ 8-bit single chip microcomputer TLCS-870/X series ♦ Instruction execution time: 0.25 μs (at 16 MHz) ♦ 842 basic instructions • Multiplication and division (8 bits × 8 bits, 16 bits × 8 bits, 16 bits/8 bits) • Bit manipulations (Set/clear/complement/move/test/exclusive or) • 16-bit data and 20-bit data operations • 1-byte jump/subroutine call (Short relative jump/vector call) RESTRICTIONS ON PRODUCT USE 20070701-EN • The information contained herein is subject to change without notice. • TOSHIBA is continually working to improve the quality and reliability of its products. Nevertheless, semiconductor devices in general can malfunction or fail due to their inherent electrical sensitivity and vulnerability to physical stress. It is the responsibility of the buyer, when utilizing TOSHIBA products, to comply with the standards of safety in making a safe design for the entire system, and to avoid situations in which a malfunction or failure of such TOSHIBA products could cause loss of human life, bodily injury or damage to property. In developing your designs, please ensure that TOSHIBA products are used within specified operating ranges as set forth in the most recent TOSHIBA products specifications. Also, please keep in mind the precautions and conditions set forth in the “Handling Guide for Semiconductor Devices,” or “TOSHIBA Semiconductor Reliability Handbook” etc. • The TOSHIBA products listed in this document are intended for usage in general electronics applications (computer, personal equipment, office equipment, measuring equipment, industrial robotics, domestic appliances, etc.).These TOSHIBA products are neither intended nor warranted for usage in equipment that requires extraordinarily high quality and/or reliability or a malfunction or failure of which may cause loss of human life or bodily injury (“Unintended Usage”). Unintended Usage include atomic energy control instruments, airplane or spaceship instruments, transportation instruments, traffic signal instruments, combustion control instruments, medical instruments, all types of safety devices, etc.. Unintended Usage of TOSHIBA products listed in his document shall be made at the customer’s own risk. • The products described in this document shall not be used or embedded to any downstream products of which manufacture, use and/or sale are prohibited under any applicable laws and regulations. • The information contained herein is presented only as a guide for the applications of our products. No responsibility is assumed by TOSHIBA for any infringements of patents or other rights of the third parties which may result from its use. No license is granted by implication or otherwise under any patents or other rights of TOSHIBA or the third parties. • Please contact your sales representative for product-by-product details in this document regarding RoHS compatibility. Please use these products in this document in compliance with all applicable laws and regulations that regulate the inclusion or use of controlled substances. Toshiba assumes no liability for damage or losses occurring as a result of noncompliance with applicable laws and regulations. 88CS38B-1 2007-09-12 TMP88CS38B/CM38B/CP38B ♦ I/O ports: Maximum 33 (High current output: 4) ♦ 17 interrupt sources: External 6, internal 11 • All sources have independent latches each, and nested interrupt control is available. • Edge-selectable external interrupts with noise reject • High-speed task switching by register bank changeover ♦ ROM corrective function ♦ Two 16-bit timer/counters: TC1, TC2 • ♦ Timer, event counter, pulse width measurement, external trigger timer, window modes Two 8-bit timer/counters: TC3, TC4 • Timer, event counter, capture (Pulse width/duty measurement) mode ♦ Time base timer (Interrupt frequency: 0.95 Hz to 31250 Hz) ♦ Watchdog timer • ♦ Serial bus interface • ♦ Interrupt source/reset output I2C bus, 8-bit SIO mode (Selectable two I/O channels) On-screen display circuit • Font ROM characters: 384 characters • Characters display: 32 columns × 12 lines • Composition: 16 × 18 dots • Size of character: 3 kinds (Line by line) • Color of character: 8 or 15 kinds (Character by character) • Variable display position: Horizontal 256 steps, Vertical 512 steps • Fringing, smoothing, slant, underline, blinking function ♦ Jitter elimination ♦ Data slicer circuit 1 channel ♦ DA conversion (Pulse width modulation) outputs • 14- or 12-bit resolution (2 channels) • 12-bit resolution (2 channels) • 7-bit resolution (6 channels) ♦ 8-bit successive approximate type AD converter with sample and hold ♦ Remote control signal preprocessor ♦ Two power saving operating modes • STOP mode: Oscillation stops. Battery/capacitor back up. Port output hold/high impedance. • IDLE mode: CPU stops, and peripherals operate using high-frequency clock. Release by interrupts. ♦ Operating Voltage: 4.5 to 5.5 V at 16 MHz ♦ Emulation POD: BM88CS38N0A-M15 88CS38B-2 2007-09-12 TMP88CS38B/CM38B/CP38B Pin Assignments Package SDIP42-P-600-1.78 TMP88CS38BNG TMP88CP38BNG TMP88CM38BNG TMP88PS38BNG SDIP42-P-600-1.78 VSS ( PWM0 ) P40 ( PWM1 ) P41 ( PWM2 ) P42 ( PWM3 ) P43 ( PWM 4 ) P44 ( PWM5 ) P45 ( PWM6 ) P46 ( PWM7 ) P47 (TC2/ INT0 / PWM8 ) P50 (SI1/SCL1/ PWM9 ) P51 (SO1/SDA1) P52 ( KWU0 / SCK1 /INT2/TC1/AIN0) P53 ( KWU1 /AIN1) P54 ( KWU2 /AIN2) P55 ( KWU3 /AIN3) P56 ( KWU 4 /Y/BLIN/AIN4) P60 ( KWU5 /BIN/AIN5) P61 (GIN/CSOUT) P62 (RIN) P63 (I) P57 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 VDD P33 (TC4/VIN0) P32 (VIN1/CSIN) VVSS P35 (SDA0) P34 (SCL0) P31 (INT4/TC3) P30 (INT3/RXIN) P20 ( INT5 / STOP ) RESET XOUT XIN TEST OSC2 OSC1 P71 ( VD ) P70 ( HD ) P67 (Y/BL) P66 (B) P65 (G) P64 (R) Package P-QFP44-1414-0.80K 34 35 36 37 38 39 40 41 42 43 44 RESET XOUT XIN TEST OSC2 OSC1 P71 (VD) 22 21 20 19 18 17 16 15 14 13 12 P70 ( HD ) P67 (Y/BL) P66 (B) P65 (G) P64 (R) N.C. P57 P63 (RIN) P62 (GIN/CSOUT) P61 (BIN//AIN5/ KWU5 ) P60 (Y/BLIN/AIN4/ KWU4 ) (PWM4) P44 (PWM5) P45 (PWM6) P46 (PWM7) P47 (TC2/INT0/PWM8) P50 (SI1/SCL1/PWM9) P51 (SO1/SDA1) P52 (KWU0/SCK1/INT2/TC1/AIN0) P53 (KWU1/AIN1) P54 (KWU2/AIN2) P55 (KWU3/AIN3) P56 (SDA0) P35 VVSS (VIN1/CSIN) P32 (TC4/VIN0) P33 N.C. VDD VSS ( PWM0 ) P40 ( PWM1 ) P41 ( PWM2 ) P42 ( PWM3 ) P43 1 2 3 4 5 6 7 8 9 10 11 TMP88CS38BFG TMP88CP38BFG TMP88CM38BFG TMP88PS38BFG 33 32 31 30 29 28 27 26 25 24 23 P34 (SCL0) P31 (INT4/TC3) P30 (INT3/RXIN) P20 (INT5/STOP) P-QFP44-1414-0.80K 88CS38B-3 2007-09-12 TMP88CS38B/CM38B/CP38B Pin Functions (1/2) Pin Name P20 ( INT5 / STOP ) I/O I/O (Input) P35 (SDA0) I/O (Input/Output) P34 (SCL0) I/O (Input/Output) P33 (TC4/VIN0) I/O (Input) P32 (VIN1/CSIN) I/O (Input) P31 (INT4/TC3) I/O (Input) P30 (INT3/RXIN) I/O (Input) P47 ( PWM7 ) I/O (Output) P46 ( PWM6 ) I/O (Output) P45 ( PWM5 ) I/O (Output) P44 ( PWM4 ) I/O (Output) P43 ( PWM3 ) I/O (Output) P42 ( PWM2 ) I/O (Output) P41 ( PWM1 ) I/O (Output) P40 ( PWM0 ) I/O (Output) P57 (I) I/O (Output) P56 ( KWU3 /AIN3) I/O (Input) P55 ( KWU2 /AIN2) I/O (Input) P54 ( KWU1 /AIN1) I/O (Input) P53 ( KWU0 /AIN0/TC1 /INT2/ SCK1 ) P52 (SDA1/SO1) I/O (Input /Input/Input /Input/Output) Function 1-bit input/output port with latch. When External interrupt input 5 or STOP used as an input port, the latch must be mode release signal input set to “1”. 6-bit programmable input/output port. Each bit of these ports can be individually configured as an input or an output under software control. During reset, all bits are configured as inputs. When used as a serial bus interface input/output, the latch must be set to “1”. 2 I C bus serial data input/output 0 2 I C bus serial clock input/output 0 Timer counter input 4 or video signal Input 0 Video signal input 1 or composite sync input External interrupt input 4 or timer counter input 3 External interrupt input 3 or remote control signal preprocessor input 8-bit programmable input/output port. Each bit of these ports can be individually configured as an input or an output under software control. During reset, all bits are configured as inputs. When used as a PWM output, the latch must be set to “1”. 7-bit DA conversion (PWM) outputs 12-bit DA conversion (PWM) outputs 14/12-bit DA conversion (PWM) outputs 8-bit programmable input/output port. Each bit of these ports can be individually configured as an input or an output under software control. During reset, all bits are configured as inputs. When used as a PWM output, a serial bus interface input/output, the latch must be set to “1”. Translucent signal output Key-on wakeup inputs or AD converter analog inputs Key-on wakeup input or AD converter analog input or timer counter input 1 or external interrupt input 2 or SIO serial clock input/output 1 2 I C bus serial data input/output 1 or SIO serial data output 1 I/O (Input/Output /Output) P51 ( PWM9 /SCL1/SI1) I/O (Output/Input/Output /Input) 7-bit DA conversion (PWM) output or 2 I C bus serial data input/output 1 or SIO serial data input 1 P50 ( PWM8 /TC2/ INT0 ) I/O (Output/Input /Input) 7-bit DA conversion (PWM) output or timer counter input 2 or external interrupt input 0 P67 (Y/BL) I/O (Output) P66 (B) I/O (Output) P65 (G) I/O (Output) P64 (R) I/O (Output) P63 (RIN) P62 (GIN/CSOUT) I/O (Input) I/O (Input/Output) P61 ( KWU5 /BIN/AIN5) I/O (Input) P60 ( KWU4 /YBLIN/AIN4) I/O (Input) 8-bit programmable input/output port. (P67 to P64: Tri-State, P63 to P60: High current output) Each bit of these ports can be individually configured as an input or an output under software control. During reset, all bits are configured as inputs. When used P64 to P67 as port, each bit of the P6 port data selection register (Bit7 to 4 in ORP6S) must be set to “1”. Y or BL output R/G/B outputs R input G input or TEST video signal output Key-on wakeup input 5 or B input or AD converter analog input 5 Key-on wakeup input 4 or Y/BL input or AD converter analog input 4 88CS38B-4 2007-09-12 TMP88CS38B/CM38B/CP38B Pin Functions (2/2) Pin Name I/O P71 ( VD ) I/O (Input) P70 ( HD ) I/O (Input) XIN, XOUT Input, Output RESET TEST I/O Input Function 2-bit programmable input/output port. Each bit of these ports can be individually configured as an input or an output under software control. During reset, all bits are configured as inputs. Vertical synchronous signal input Horizontal synchronous signal input Resonator connecting pins. For inputting external clock, XIN is used and XOUT is opened. Reset signal input or watchdog timer output/address-trap-reset output/system-clock-reset output Test pin for out-going test. Be tied to low. OSC1, OSC2 Input, Output Resonator connecting pins for on-screen display circuitry VDD, VSS, VVSS Power supply +5 V, 0 V (GND) 88CS38B-5 2007-09-12 TMP88CS38B/CM38B/CP38B Block Diagram I/O ports P64 to P67 P70, P71 P57 OSC connecting pins for on-screen display Display memory OSC1 OSC2 Power supply VDD VSS VVSS R, G, B, Jitter Y/BL elimination Character ROM VD HD On-screen display circuit P6 TLCS-870/X CPU core Data memory (RAM) Data slicer Program counter Interrupt Controller System controller P5 ROM corrective circuit Reset I/O test pin RESET TEST P7 Standby controller Resonator connecting pins XIN XOUT Time base timer Timing generator High Clock frequency generator Watchdog timer 16-bit timer TC1 TC2 Program memory (ROM) 8-bit timer/counter TC3 TC4 Inst. register Inst. decoder P2 P4 DA converter (PWM) P5 8-bit AD Key-on wakeup P6 Video signal output Remote control signal P32 P20 P40 to P47 P50 to P56 P60 to P63 Serial bus interface P3 P30 to P35 Y/BLIN RIN GIN BIN P33 Video signal I/O ports 88CS38B-6 2007-09-12 TMP88CS38B/CM38B/CP38B Operational Description 1. CPU Core Functions The CPU core consists of a CPU, a system clock controller, and an interrupt controller. This section provides a description of the CPU core, the program memory, the data memory, the external memory interface, and the reset circuit. 1.1 Memory Address Map The TMP88CS38B/CM38B/CP38B memory consists of four blocks: ROM, RAM, SFR (Special function register), and DBR (Data buffer register). They are all mapped to a 1-Mbyte address space. Figure 1.1.1 shows the TMP88CS38B/CM38B/CP38B memory address map. There are 16 banks of the general-purpose register. The register banks are also assigned to the RAM address space. 00000H SFR RAM 0003FH 00040H 000BFH 000C0H 64 bytes 128 bytes 0003FH 00040H 000BFH 000C0H 2048 bytes 64 bytes 128 bytes 000BFH 000C0H 128 bytes 128 bytes General-purpose register banks (8 registers × 16 banks) 1536 bytes 00F80H 128 bytes 128 bytes 00FFFH 00FFFH 00FFFH 04000H 04000H 04000H 32512 bytes program area 48896 bytes program area 65280 bytes program area 64 bytes 006BFH 00F80H 00F80H 0003FH 00040H 1536 bytes 006BFH 008BFH DBR 00000H 00000H 0BEFFH 0FEFFH 13EFFH 20000H ROM 24576 bytes OSD font area 25FFFH FFF7FH FFF80H 24576 bytes OSD font area FFF00H 64 bytes 64 bytes FFF3FH FFF40H FFF7FH FFF80H 128 bytes FFFFFH 24576 bytes OSD font area FFF00H 64 bytes 64 bytes FFF3FH FFF40H FFF7FH FFF80H 64 bytes Vector table for interrupts 64 bytes Vector table for vector call instruction 128 bytes 128 bytes FFFFFH TMP88CS38B 20000H 25FFFH 25FFFH FFF00H FFF3FH FFF40H 20000H FFFFFH TMP88CP38B Vector table for interrupts TMP88CM38B ROM: Read only memory includes Program memory, Character data memory for OSD RAM: Random access memory includes Data memory, Stack, General-purpose register banks SFR: Special function register includes I/O ports, Peripheral hardware control registers, Peripheral hardware status registers, System control registers, Interrupt control registers, Program status word DBR: Data buffer register includes Control resister for on-screen display(OSD) Remote-control-receive control/status resigsters, ROM correction control registers, Test video signal control registers Figure 1.1.1 Memory Address Map 88CS38B-7 2007-09-12 TMP88CS38B/CM38B/CP38B 1.2 Program Memory (ROM) The TMP88CS38B contains a 64-Kbyte program memory (Mask ROM) at addresses from 04000H to 13EFFH and FFF00H to FFFFFH. The TMP88CM38B contains a 32-Kbyte program memory (Mask ROM) at address from 04000H to 0BEFFH and FFF00H to FFFFFH. The TMP88CP38B contains a 48-Kbyte program memory (Mask ROM) at address from 04000H to 0FEFFH and FFF00H to FFFFFH. Addresses FFF00H through FFFFFH in the program memory are also used for a particular purpose. 1.3 Data Memory (RAM) The TMP88CS38B has a 2-Kbyte data memory (Static RAM) address from 0040H to 08BFH. The TMP88CM38B/CP38B has a 1.5-Kbyte data memory (Static RAM) address from 0040H to 06BFH. The first 128 bytes (Addresses 00040H through 000BFH) in the built-in RAM are also available as general-purpose register banks. The general-purpose registers are mapped in the RAM; therefore, do not clear RAM at the current bank addresses. Example: Clears RAM to “00H” except the bank0 (TMP88CS38B/CM38B/CP38B): LD HL, 0048H ; Sets start address to HL register pair LD A, H ; Sets initial data (00H) to A register LD BC, 0877H ; Sets number of byte to BC register pair SRAMCLR: LD (HL+), A DEC BC JRS F, SRAMCLR Note: 1.4 The data memory contents become unstable when the power supply is turned on; therefore, the data memory should be initialized by an initialization routine. Note that the general-purpose registers are mapped in the RAM; therefore, do not clear RAM at the current bank addresses. System Clock Controller The system clock controller consists of a clock generator, a timing generator, and a standby controller. Timing generator control register TBTCR Clock generator XIN 00036H fc High-frequency clock oscillator Timing generator Standby controller XOUT 00038H System clocks Clock generator control 00039H SYSCR1 SYSCR2 System control registers Figure 1.4.1 System Clock Controller 88CS38B-8 2007-09-12 TMP88CS38B/CM38B/CP38B 1.4.1 Clock Generator The clock generator generates the basic clock which provides the system clocks supplied to the CPU core and peripheral hardware. It contains oscillation circuit: one for the high-frequency clock. The high-frequency (fc) clock can be easily obtained by connecting a resonator between the XIN/XOUT pin, respectively. Clock input from an external oscillator is also possible. In this case, external clock is applied to the XIN/XTIN pin not connected. The TMP88CS38B/CM38B/CP38B is not provided an LC oscillation. High-frequency clock XIN XOUT XIN XOUT (Open) (a) Crystal/Ceramic resonator (b) External oscillator Figure 1.4.2 Examples of Resonator Connection Note: Accurate adjustment of the oscillation frequency: Although hardware to externally and directly monitor the basic clock pulse is not provided, the oscillation frequency can be adjusted by making the program to output fixed frequency pulses to the port while disabling all interrupts and monitoring this pulse. With a system requiring adjustment of the oscillation frequency, the adjusting program must be created beforehand. 1.4.2 Timing Generator The timing generator generates from the basic clock the various system clocks supplied to the CPU core and peripheral hardware. The timing generator provides the following functions: 1. Generation of main system clock 2. Generation of source clocks for time base timer 3. Generation of source clocks for watchdog timer 4. Generation of internal source clocks for timer/counters TC1 to TC4 5. Generation of warm-up clocks for releasing STOP mode 6. Generation of a clock for releasing reset output (1) Configuration of timing generator The timing generator consists of a 21-stage divider with a divided by 3 prescaler, a main system clock generator, and machine cycle counters. During reset and at releasing STOP mode, the prescaler and the divider are cleared to “0”, however, the prescaler is not cleared. An input clock to the 7th stage of the divider depends on the operating mode. A divided by 256 of high-frequency clock (fc/28) is input to the 7th stage of the divider. 88CS38B-9 2007-09-12 TMP88CS38B/CM38B/CP38B fm Machine cycle counters States DV1CK Prescaler High-frequency clock Machine cycles fc 0 1 2 Divider S A Y Divider 1 2 3 4 5 6 fc/2 8 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 B Reset circuit standby controller Timer/ counters Watchdog timer Time base timer fc MK8 MHz SLICER FC8OUT D1 D0 SG JITTA Figure 1.4.3 Configuration of Timing Generator CGCR (00030H) “0” “0” DV1CK DV1CK “0” “0” “0” “0” “0” Selection of input clock to 0: fc/4 the 1st stage of the divider. 1: fc/8 (Initial value: 0000 0000) R/W Note 1: fc: High-frequency clock [Hz], *: Don’t care Note 2: The all bits except DV1CK are cleared to “0”. Figure 1.4.4 Divider Control Register FC8CR (00FEEH) D1 D1 D0 FC8OUT 1 0 1/2 fc 0 0 1/1 fc D0 Read/Write (Initial value: 0000 0010) Figure 1.4.5 FC8 Control Register 88CS38B-10 2007-09-12 TMP88CS38B/CM38B/CP38B (2) Machine cycle Instruction execution and peripheral hardware operation are synchronized with the main system clock. The minimum instruction execution unit is called a “machine cycle”. There are a total of 15 different types of instructions for the TLCS-870/X series: Ranging from 1-cycle instructions which require one machine cycle for execution to 15-cycle instructions which require 15 machine cycles for execution. A machine cycle consists of 4 states (S0 to S3), and each state consists of one main system clock. 1/fc Main System clock fm S0 State S1 S2 S3 S0 S1 S2 S3 Machine cycle (0.25 μs at fc = 16 MHz) Figure 1.4.6 Machine Cycle 1.4.3 Standby Controller The standby controller starts and stops the switches the main system clock. These modes are controlled by the system control registers (SYSCR1, SYSCR2). Figure 1.4.7 shows the operating mode transition diagram and Figure 1.4.8 shows the system control registers. (1) Single-clock mode In the single-clock mode, the machine cycle time is 4/fc [s] (0.25 μs at fc = 16 MHz). 1. NORMAL mode In this mode, both the CPU core and on-chip peripherals operate using the high-frequency clock. 2. IDLE mode In this mode, the internal oscillation circuit remains active. The CPU and the watchdog timer are halted; however, on-chip peripherals remain active (Operate using the high-frequency clock). IDLE mode is started by setting IDLE bit in the system control register 2 (SYSCR2), and IDLE1 mode is released to NORMAL mode by an interrupt request from on-chip peripherals or external interrupt inputs. When IMF (Interrupt master enable flag) is “1” (Interrupt enable), the execution will resume upon acceptance of the interrupt, and the operation will return to normal after the interrupt service is completed. When IMF is “0” (Interrupt disable), the execution will resume with the instruction which follows IDLE mode start instruction. 3. STOP mode In this mode, the internal oscillation circuit is turned off, causing all system operations to be halted. The internal status immediately prior to the halt is held with the lowest power consumption during this mode. STOP mode is started by setting STOP bit in the system control register 1 (SYSCR1), and STOP mode is released by an input (Either level-sensitive or edge-sensitive can be programmably selected) to the STOP pin. After the warm-up period is completed, the execution resumes with the next instruction which follows the STOP mode start instruction. 88CS38B-11 2007-09-12 TMP88CS38B/CM38B/CP38B RESET Reset release Software IDLE mode NORMAL mode Interrupt Software STOP STOP mode pin input (a) Single-clock mode Note: NORMAL mode are generically called NORMAL; STOP mode is called STOP; and IDLE mode is called IDLE. Operating Mode Frequency CPU Core High-frequency RESET Single clock NORMAL Reset Turning on oscillation IDLE STOP On-chip Machine Peripherals Cycle Time Turning off oscillation Operate Halt Reset Operate Halt 4/fc [s] − Figure 1.4.7 Operating Mode Transition Diagram 88CS38B-12 2007-09-12 TMP88CS38B/CM38B/CP38B System Control Register 1 SYSCR1 (00038H) 7 6 5 4 STOP RELM “0” “1” 3 2 1 0 WUT (Initial value: 0000 00**) 0: CPU core and peripherals remain active STOP STOP mode start RELM Release method for STOP mode 1: CPU core and peripherals are halted (Start STOP mode) 0: STOP Edge-sensitive release (Rising edge) 1: STOP Level-sensitive release (“H” level) Return to NORMAL mode Warm-up time at releasing STOP mode WUT 00 DV1CK = 0 DV1CK = 1 3 × 2 /fc 16 3 × 2 /fc 16 2 /fc 17 17 2 /fc 01 14 10 2 /fc 11 Reserved R/W 15 2 /fc Reserved Note 1: Always set bit5 in SYSCR1 to “0”. Note 2: When STOP mode is released with RESET pin input, a return is made to NORMAL mode regardless of the RETM contents. Note 3: fc: High-frequency clock [Hz] *: Don’t care Note 4: Bits 1 and 0 in SYSCR1 are read in as undefined data when a read instruction is executed. Note 5: Always set bit4 in SYSCR1 to “1” when STOP mode is started. System Control Register 2 SYSCR2 (00039H) 7 6 5 4 “1” “0” “0” IDLE IDLE IDLE mode start 3 2 1 0 (Initial value: 1000 ****) 0: CPU and watchdog timer remain active 1: CPU and watchdog timer are stopped (Start IDLE mode) Note 1: *: Don’t care Note 2: Always set bit7, 6 and 5 in SYSCR2 to “100”. R/W Figure 1.4.8 System Control Registers 88CS38B-13 2007-09-12 TMP88CS38B/CM38B/CP38B 1.4.4 Operating Mode Control (1) STOP mode STOP mode is controlled by the system control register 1 (SYSCR1) and the STOP pin input. The STOP pin is also used both as a port P20 and an INT5 (External interrupt input 5) pin. STOP mode is started by setting STOP (Bit7 in SYSCR1 ) to “1”. During STOP mode, the following status is maintained. 1. Oscillations are turned off, and all internal operations are halted. 2. The data memory, registers and port output latches are all held in the status in effect before STOP mode was entered. 3. The prescaler and the divider of the timing generator are cleared to “0”. 4. The program counter holds the address of the instruction following the instruction which started the STOP mode. STOP mode includes a level-sensitive release mode and an edge-sensitive release mode, either of which can be selected with RELM (Bit6 in SYSCR1). a. Level-sensitive release mode (RELM = 1) In this mode, STOP mode is released by setting the STOP pin high. This mode is used for capacitor back up when the main power supply is cut off and long term battery back up. When the STOP pin input is high, executing an instruction which starts the STOP mode will not place in STOP mode but instead will immediately start the release sequence (Warm up). Thus, to start STOP mode in the level-sensitive release mode, it is necessary for the program to first confirm that the STOP pin input is low. The following method can be used for confirmation: Using an external interrupt input INT5 ( INT5 is a falling edge-sensitive input). Example: Starting STOP mode with an INT5 interrupt. PINT5: SINT5: TEST (P2). 0 JRS F, SINT5 ; To reject noise, the STOP mode does not start if port P20 is at high LD (SYSCR1), 01010000B ; Sets up the level-sensitive release mode SET (SYSCR1). 7 ; Starts STOP mode LDW (IL), 1110011101010111B ; IL12, 11, 7, 5, 3 ← 0 (Clears interrupt latches) RETI VIH STOP pin XOUT pin NORMAL operation STOP operation Confirm by program that the STOP pin input is low and start STOP mode. Warm up NORMAL operation STOP mode is released by the hardware. Always released if the STOP pin input is high. Note 1: After warm up is started, when STOP pin input is changed “L” level, STOP mode is not placed. Note 2: When changing to the level-sensitive release mode from the edge-sensitive release mode, the release mode is not switched until a rising edge of the STOP pin input is detected. Figure 1.4.9 Level-sensitive Release Mode 88CS38B-14 2007-09-12 TMP88CS38B/CM38B/CP38B b. Edge-sensitive release mode (RELM = 0) In this mode, STOP mode is released by a rising edge of the STOP pin input. This is used in applications where a relatively short program is executed repeatedly at periodic intervals. This periodic signal (for example, a clock from a low-power consumption oscillator) is input to the STOP pin. In the edge-sensitive release mode, STOP mode is started even when the STOP pin input is high. Example: Starting STOP mode from NORMAL mode LD (SYSCR1), 10010000B ; Starts after specified to the edge-sensitive mode VIH STOP pin XOUT pin STOP operation NORMAL operation Warm up STOP mode started by the program. STOP operation NORMAL operation STOP mode is released by the hardware at the rising edge of STOP pin input. Figure 1.4.10 Edge-sensitive Release Mode STOP mode is released by the following sequence: 1. When returning to NORMAL, clock oscillator is turned on. 2. A warm-up period is inserted to allow oscillation time to stabilize. During warm-up, all internal operations remain halted. Two different warm-up times can be selected with WUT (Bits 2 and 3 in SYSCR1) as determined by the resonator characteristics. 3. When the warm-up time has elapsed, normal operation resumes with the instruction following the STOP mode start instruction (e.g., [SET (SYSCR1). 7]). The start is made after the divider of the timing generator is cleared to “0”. Table 1.4.1 Warm-up Time Example Warm-up Time [s] WUT Return to NORMAL mode DV1CK = 0 00 Note: 3 × 2 /fc 16 DV1CK = 1 (12.29 m) 3 × 2 /fc 17 (24.58 m) 17 (8.20 m) 15 (2.05 m) 16 (4.10 m) 2 /fc 14 (1.02 m) 2 /fc 01 2 /fc 10 2 /fc 11 Reserved ( - ) Reserved ( - ) The warm-up time is obtained by dividing the basic clock by the divider: therefore, the warm-up time may include a certain amount of error if there is any fluctuation of the oscillation frequency when STOP mode is released. Thus, the warm-up time must be considered an approximate value. 88CS38B-15 2007-09-12 88CS38B-16 Divider 0 Instruction Halt execution Program counter Oscillator Turn circuit off Main system clock input STOP pin Divider Instruction execution Program counter Main system clock Oscillator circuit Turn on Turn on n Count up Warm up a+2 n+2 n+3 a+3 (b) STOP mode release 0 1 Instruction at address a + 2 a+4 a+5 n+4 2 Instruction at address a + 3 (a) STOP mode start (Example: Start with SET (SYSCR1). 7 instruction located at address a) n+1 SET (SYSCR1). 7 a+3 3 Instruction at address a + 4 a+6 0 Halt Turn off TMP88CS38B/CM38B/CP38B Figure 1.4.11 STOP Mode Start/Release 2007-09-12 TMP88CS38B/CM38B/CP38B STOP mode can also be released by setting the RESET pin low, which immediately performs the normal reset operation. Note: When STOP mode is released with a low hold voltage, the following cautions must be observed. The power supply voltage must be at the operating voltage level before releasing STOP mode. The RESET pin input must also be high, rising together with the power supply voltage. In this case, if an external time constant circuit has been connected, the RESET pin input voltage will increase at a slower rate than the power supply voltage. At this time, there is a danger that a reset may occur if input voltage level of the RESET pin drops below the non-inverting high-level input voltage (hysteresis input). (2) IDLE mode IDLE mode is controlled by the system control register 2 and maskable interrupts. The following status is maintained during IDLE mode. 1. Operation of the CPU and watchdog timer is halted. On-chip peripherals continue to operate. 2. The data memory, CPU registers and port output latches are all held in the status in effect before IDLE mode was entered. 3. The program counter holds the address of the instruction following the instruction which started IDLE mode. Example: Starting IDLE mode. SET (SYSCR2). 4 ; IDLE ← 1 Starting IDLE mode by instruction CPU, WDT are halted Reset input Yes Reset No (High) No Interrupt request Normal release mode No Yes IMF = 1 Yes (Interrupt release mode) Interrupt processing Execution of the instruction which follows the IDLE mode start instruction Figure 1.4.12 IDLE Mode 88CS38B-17 2007-09-12 TMP88CS38B/CM38B/CP38B IDLE mode includes a normal release mode and an interrupt release mode. Selection is made with the interrupt master enable flag (IMF). Releasing the IDLE mode returns from IDLE to NORMAL. a. Normal release mode (IMF = “0”) IDLE mode is released by any interrupt source enabled by the individual interrupt enable flag (EF) or an external interrupt 0 ( INT0 pin) request. Execution resumes with the instruction following the IDLE mode start instruction (e.g., [SET (SYSCR2).4]). Normally, IL (Interrupt latch) of interrupt source to release IDLE mode must be cleared by load instructions. b. Interrupt release mode (IMF = “1”) IDLE mode is released and interrupt processing is started by any interrupt source enabled with the individual interrupt enable flag (EF) or an external interrupt 0 ( INT0 pin) request. After the interrupt is processed, the execution resumes from the instruction following the instruction which started IDLE mode. Note: When a watchdog timer interrupt is generated immediately before the IDLE mode is started, the watchdog timer interrupt will be processed but IDLE mode will not be started. 88CS38B-18 2007-09-12 88CS38B-19 Watchdog timer Instruction execution Program counter Interrupt request Halt Halt Halt Watchdog timer Main system clock Halt Instruction execution Program counter Interrupt request Main system clock a+3 (b) IDLE mode release (II) Interrupt release mode a+3 a+3 Operate Operate Acceptance of interrupt Instruction at address a + 2 a+4 (a) IDLE mode start (Example: starting with the SET instruction located at address a) (I) Normal release mode Operate SET (SYSCR2). 4 Instruction execution Watchdog timer a+2 Program counter Interrupt request Main system clock Halt TMP88CS38B/CM38B/CP38B Figure 1.4.13 IDLE Mode Start/Release 2007-09-12 TMP88CS38B/CM38B/CP38B IDLE mode can also be released by setting the RESET pin low, which immediately performs the reset operation. After reset, the TMP88CS38B/CM38B/CP38B is placed in NORMAL mode. 88CS38B-20 2007-09-12 TMP88CS38B/CM38B/CP38B 1.5 Interrupt Controller The TMP88CS38B/CM38B/CP38B has a total of 17 interrupt sources; 6 externals and 11 internals. Multiple interrupts with priorities are also possible. Two of the internal sources are pseudo non-maskable interrupts; the remainder are all maskable interrupts. Table 1.5.1 Interrupt Sources Interrupt Source Internal/ External (Reset) Internal INTSW (Software interrupt) Enable Condition Non maskable Pseudo non maskable Interrupt Latch Vector Table Address − FFFFCH High 0 − FFFF8H 1 2 Priority Internal INTWDT (Watchdog timer interrupt) IL2 FFFF4H External INT0 (External interrupt 0) IMF ·EF3 = 1, INT0EN = 1 IL3 FFFF0H 3 Internal INTTC1 (16-bit TC1 interrupt) IMF ·EF4 = 1 IL4 FFFECH 4 External INTKWU (Key-on wakeup) IMF ·EF5 = 1 IL5 FFFE8H 5 Internal INTTBT (Time base timer interrupt) IMF ·EF6 = 1 IL6 FFFE4H 6 External INT2 (External interrupt 2) IMF ·EF7 = 1 IL7 FFFE0H 7 Internal INTTC3 (8-bit TC3 interrupt) IMF ·EF8 = 1 IL8 FFFDCH 8 Internal INTTSBI (SBI interrupt) IMF ·EF9 = 1 IL9 FFFD8H 9 Internal INTTC4 (8-bit TC4 interrupt) IMF ·EF10 = 1 IL10 FFFD4H 10 External INT3 (External interrupt 3) IMF ·EF11 = 1 IL11 FFFD0H 11 External INT4 (External interrupt 4) IMF ·EF12 = 1 IL12 FFFCCH 12 Internal INTADC (AD converter interrupt) IMF ·EF13 = 1 IL13 FFFC8H 13 Internal INTTC2 (16-bit TC2 interrupt) IMF ·EF14 = 1 IL14 FFFC4H 14 External INT5 (External interrupt 5) IMF ·EF15 = 1 IL15 FFFC0H 15 Internal INTOSD (OSD interrupt) IMF ·EF16 = 1 IL16 FFFBCH 16 Internal INTSLI (Slicer interrupt) IMF ·EF17 = 1 IL17 FFFB8H 17 Reserved IMF ·EF18 = 1 IL18 FFFB4H 18 Reserved IMF ·EF19 = 1 IL19 FFFB0H 19 Reserved IMF ·EF20 = 1 IL20 FFFACH 20 Reserved IMF ·EF21 = 1 IL21 FFFA8H 21 Reserved IMF ·EF22 = 1 IL22 FFFA4H 22 Reserved IMF ·EF23 = 1 IL23 FFFA0H 23 Reserved IMF ·EF24 = 1 IL24 FFF9CH 24 Reserved IMF ·EF25 = 1 IL25 FFF98H 25 Reserved IMF ·EF26 = 1 IL26 FFF94H 26 Reserved IMF ·EF27 = 1 IL27 FFF90H 27 Reserved IMF ·EF28 = 1 IL28 FFF8CH 28 Reserved IMF ·EF29 = 1 IL29 FFF88H 29 Reserved IMF ·EF30 = 1 IL30 FFF84H 30 Reserved IMF ·EF31 = 1 IL31 FFF80H Low 31 Note: Before you change each enable flag (EF) and/or each interrupt latch (IL), be sure to clear the interrupt master enable flag (IMF) to “0” (to disable interrupts). a. After a DI instruction is executed. b. When an interrupt is accepted, IMF is automatically cleared to “0”. However to enable nested interrupts, change EF and/or IL before setting IMF to “1” (to enable interrupts). If the individual enable flags (EF) and interrupt latches (IL) are set under conditions other than the above, the proper operation cannot be guaranteed. 88CS38B-21 2007-09-12 Digital noise reject circuit ··························· 88CS38B-22 IL31 to 3 write data Internal reset Write strobe for IL S R S R S R S R S R IL5 IL4 IL17 IL16 Q Q Q Q IL3 Q Interrupt enable flag EF31 to EF3 Vector table address Instruction which clears IMF to “0” R S IMF Q Instruction which sets IMF to “1” [RETN] instruction only when IMF was set before interrupt was accepted [EI] instruction [RETI] instruction during maskable interrupt service Interrupt master enable flag Interrupt acceptance Vector table address generator Non-maskable interrupts request Interrupt request Maskable interrupts Release IDLE mode request request and Priority encoder [DI] Instruction ························ External interrupts control register EINTCR INT0EN Interrupt latch S IL2 Q R ························ INTSLI INTOSD INTKOW INTTC1 INT0 ········ INTSW ········ INTWDT ··· IL17 to IL2 TMP88CS38B/CM38B/CP38B Figure 1.5.1 Interrupt Controller Block Diagram 2007-09-12 TMP88CS38B/CM38B/CP38B Interrupt latches (IL) that hold the interrupt requests are provided for interrupt sources. Each interrupt vector is independent. The interrupt latch is set to “1” when an interrupt request is generated, and requests the CPU to accept the interrupt. The acceptance of maskable interrupts can be selectively enabled and disabled by program using the interrupt master enable flag (IMF) and the individual interrupt enable flags (EF). When two or more interrupts are generated simultaneously, the interrupt is accepted in the highest priority order as determined by the hardware. Figure 1.5.1 shows the interrupt controller. (1) Interrupt latches (IL31 to IL2) Interrupt latches are provided for each source, except for a software interrupt. The latch is set to “1” when an interrupt request is generated, and requests the CPU to accept the interrupt. The latch is cleared to “0” just after the interrupt is accepted. All interrupt latches are initialized to “0” during reset. The interrupt latches are assigned to addresses 0003CH, 0003DH, 0002EH and 0002FH in the SFR. Except for IL2, each latch can be cleared to “0” individually by an instruction ; however, the read-modify-write instruction such as bit manipulation or operation instructions cannot be used. When interrupt occurred during order execution, the reason is because interrupt request is cleared. Thus, interrupt requests can be canceled and initialized by the program. Note that request the interrupt latches cannot be set to “1” by an instruction. For example, it may be that each latch is cleared even if an interrupt request is generated during instruction exection. The contents of interrupt latches can be read out by an instruction. Therefore, testing interrupt request by software is possible. Example 1: Clears interrupt latches DI LDW (ILL), 1110100000111111B ; Disable interrupt ; IL12, IL10 to IL6 ← 0 ; W ← ILH, A ← ILL ; if IL7 = 1 then jump Example 2: Reads interrupt latches LD WA, (ILL) Example 3: Tests an interrupt latch TEST (ILL). 7 JR F, SSET (2) Interrupt enable register (EIR) The interrupt enable register (EIR) enables and disables the acceptance of interrupts, except for the pseudo non-maskable interrupts (Software and watchdog timer interrupts). Pseudo non-maskable interrupts are accepted regardless of the contents of the EIR; however, the pseudo non-maskable interrupt cannot be nested more than once at the same time. The EIR consists of an interrupt master enable flag (IMF) and the individual interrupt enable flags (EF). These registers are assigned to addresses 0003AH, 0003BH, 0002CH and 0002DH in the SFR, and can be read and written by an instruction (including read-modify-write instruction such as bit manipulation instructions). Note: Do not use the read-modify-write instruction for the EIRL (Address 0003AH) during pseudo non-maskable interrupt service task. If the read-modify-write instruction is used, the IMF is not set to “1” after RETN. 88CS38B-23 2007-09-12 TMP88CS38B/CM38B/CP38B 1. Interrupt master enable flag (IMF) The interrupt master enable flag (IMF) enables and disables the acceptance of all maskable interrupts. Clearing this flag to “0” disables the acceptance of all maskable interrupts. Setting to “1” enables the acceptance of interrupts. When an interrupt is accepted, this flag is cleared to “0” to temporarily disable the acceptance of other maskable interrupts. After execution of the interrupt service program, this flag is set to “1” by the maskable interrupt return instruction [RETI] to again enable the acceptance of interrupts. If an interrupt request has already been occurred, interrupt service starts immediately after execution of the [RETI] instruction. Pseudo non-maskable interrupts are returned by the [RETN] instruction. In this case, the IMF is set to “1” only when pseudo non-maskable interrupt service is started with interrupt acceptance enabled (IMF = 1). Note that the IMF remains “0” when cleared by the interrupt service program. The IMF is assigned to bit0 at address 0003AH in the SFR, and can be read and written by an instruction. The IMF is normally set and cleared by the [EI] and [DI] instructions, and the IMF is initialized to “0” during reset. 2. Individual interrupt enable flags (EF17 to EF3) These flags enable and disable the acceptance of individual maskable interrupts, except for an external interrupt 0. Setting the corresponding bit of an individual interrupt enable flag to “1” enables acceptance of an interrupt, setting the bit to “0” disables acceptance. Example 1: Sets EF for individual interrupt enable, and sets IMF to “1”. DI LD (EIRE), 00000001B LDW (EIRL), 1110100010100001B ; Disable interrupt ; EF16 ← 1 EF15 to EF13, EF11, EF7, EF5, IMF ← 1 Example 2: Sets an individual interrupt enable flag to “1”. SET (EIRH). 4 ; 88CS38B-24 EF12 ← 1 2007-09-12 TMP88CS38B/CM38B/CP38B Interrupt Latches (IL) IL (0002E, 0002FH) 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 IL31 IL30 IL29 IL28 IL27 IL26 IL25 IL24 IL23 IL22 IL21 IL20 IL19 IL18 IL17 IL16 ILD (0002FH) ILE (0002EH) (Initial value: 00000000 00000000) IL (0003C, 0003DH) IL15 IL14 IL13 IL12 IL11 IL10 IL9 IL8 IL7 IL6 IL5 IL4 ILH (0003DH) IL3 IL2 INF ILL (0003CH) (Initial value: 00000000 00000000**) Interrupt Enable Registers (EIR) EIR (0002C, 0002DH) 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 EF31 EF30 EF29 EF28 EF27 EF26 EF25 EF24 EF23 EF22 EF21 EF20 EF19 EF18 EF17 EF16 EIRD (0002DH) EIRE (0002CH) (Initial value: 00000000 00000000) EIR (0003A, 0003BH) EF15 EF14 EF13 EF12 EF11 EF10 EF9 EF8 EF7 EF6 EF5 EIRH (0003BH) EF4 EF3 IMF EIRL (0003AH) (Initial value: 00000000 0000000**0) Note 1: Do not clear IL with read-modify-write instructions such as bit operations. Note 2: Do not set IMF to “1” during non-maskable interrupt service program. Note 3: Bits 1 and 0 in ILL are read in as undefined data when a read instruction is executed. Note 4: *: Don’t care Note 5: Do not clear IL2 to “0” by an instruction. Note 6: At TMP88CS38/CM38A/CP38A, IL18 to IL31 and EF18 to EF31 are not used. Note 7: After IMF is cleared, modify EF and IL. Figure 1.5.2 Interrupt Latches (IL) and Interrupt Enable Registers (EIR) 1.5.1 Interrupt Sequence An interrupt request is held until the interrupt is accepted or the interrupt latch is cleared to “0” by a reset or an instruction. Interrupt acceptance sequence requires 12 machine cycles (3 μs at fc = 16 MHz in the NORMAL mode) after the completion of the current instruction execution. The interrupt service task terminates upon execution of an interrupt return instruction [RETI] (for maskable interrupts) or [RETN] (for pseudo non-maskable interrupts). Figure 1.5.3 shows the timing chart of interrupt acceptance processing. (1) Interrupt acceptance Interrupt acceptance processing is as follows. 1. The interrupt master enable flag (IMF) is cleared to “0” to temporarily disable the acceptance of any following maskable interrupts. When a non-maskable interrupt is accepted, the acceptance of any following interrupts is temporarily disabled. 2. The interrupt latch (IL) for the interrupt source accepted is cleared to “0”. 3. The contents of the program counter (PC) and the program status word (PSW) are saved (Pushed) on the stack in sequence of PSWH, PSWL, PCE, PCH, PCL. The stack pointer (SP) is decremented five times. 4. The entry address of the interrupt service program is read from the vector table, and set to the program counter. 88CS38B-25 2007-09-12 TMP88CS38B/CM38B/CP38B 5. The RBS control code is read from the vector table. The lower 4 bits of this code is added to the RBS. 6. The instruction stored at the entry address of the interrupt service program is executed. Example: Correspondence between vector table address for INTTBT and the entry address of the interrupt service program. Entry address Vector table address FFFE4H 43H FFFE5H D2H 0CH FFFE7H 06H Vector CD244H CD245H RBS control Interrupt service program CD246H ·· FFFE6H CD243H A maskable interrupt is not accepted until the IMF is set to “1” even if the maskable interrupt higher than the level of current servicing interrupt is occurred. When nested interrupt service is necessary, the IMF is set to “1” in the interrupt service program. In this case, acceptable interrupt sources are selectively enabled by the individual interrupt enable flags. Note: Do not use the read-modify-write instruction for the EIRL (Address 0003AH) during pseudo non-maskable interrupt service task. 88CS38B-26 2007-09-12 TMP88CS38B/CM38B/CP38B Interrupt service task 1-machine cycle INT5 INTTBT IL15 IL6 IMF Execution Instruction Address bus PC Interrupt acceptance a+1 a FFFE4 FFFE5 FFFE6 FFFE7 a+1 a n Instruction n−1 n−2 n−3 n−4 a SP b n − 1 n −2 n − 3 n − 4 n RBS b b+1 b+2 b+1 b+2 b+3 n−5 k = i + (FFFE7H). 3 − 0 i INF (a) Interrupt acceptance Interrupt service task IMF RETI instruction Execution Address bus PC SP c c c+1 n−4 n−3 n−2 n−1 c+1 C+2 n−5 n−4 n−3 n−2 n−1 RBS k n a+1 a a a+1 a+2 n i INF (b) Return from interrupt instruction Note 1: a: Return address, b: Entry address, c: Address which the RETI instruction is stored Note 2: The maximum response time from when an IL is set until an interrupt acceptance processing starts is 62/fc [s] with interrupt enabled. Figure 1.5.3 Timing Chart of Interrupt Acceptance and Interrupt Return Instruction 88CS38B-27 2007-09-12 TMP88CS38B/CM38B/CP38B (2) Saving/restoring general-purpose registers During interrupt acceptance processing, the program counter (PC) and the program status word (PSW) are automatically saved on the stack, but not the accumulator and other registers. These registers are saved by the program if necessary. Also, when nesting multiple interrupt services, it is necessary to avoid using the same data memory area for saving registers. The following method is used to save/restore the general-purpose registers. 1. General-purpose register save/restore by automatic register bank changeover The general-purpose registers can be saved at high speed by switching to a register bank that is not in use. Normally, the bank0 is used for the main task and the banks 1 to 15 are assigned to interrupt service tasks. To increase the efficiency of data memory utilization, the same bank is assigned for interrupt sources which are not nested. The switched bank is automatically restored by executing an interrupt return instruction [RETI] or [RETN]. Therefore, it is not necessary for a program to save the RBS. Example: Register bank changeover PINTxx: Interrupt processing RETI ···· VINTxx: 2. DP PINTxx DB 1 ; RBS ← RBS + 1 General-purpose register save/restore by register bank changeover The general-purpose registers can be saved at high speed by switching to a register bank that is not in use. Normally, the bank0 is used for the main tank and the banks 1 to 15 are assigned to interrupt service tasks. Example: Register bank changeover PINTxx: LD RBS, n Interrupt processing RETI Restores bank and returns ; Interrupt service routine entry address ···· ; VINTxx: Main task Bank m Acceptance of interrupt Interrupt service task DP PINTxx DB 0 Switch to bank n by LD, RBS and n instruction Main task m Acceptance of interrupt Interrupt service task m n Time m Interrupt return Saving registers Switch to bank n automatically Restore to bank m automatically by [RETI]/[RETN] Restoring registers Interrupt return (a) Saving/restoring by register bank changeover (b) Saving/restoring using push/pop or data transfer instructions Figure 1.5.4 Saving/Restoring General-purpose Registers 88CS38B-28 2007-09-12 TMP88CS38B/CM38B/CP38B 3. General-purpose registers save/restore using push and pop instructions To save only a specific register, and when the same interrupt source occurs more than once, the general-purpose registers can be saved/restored using the push/pop instructions. Example: Register save/restore using push and pop instructions PINTxx: PUSH WA ; Save WA register pair ; Restore WA register pair ; Return Interrupt processing POP WA RETI Address (Example) SP 0023AH A SP 0023B W SP 0023C PCL PCL PCL 0023D PCH PCH PCH 0023E PCE PCE PCE 0023F PSWL PSWL PSWL 00240 PSWH PSWH PSWH At acceptance of an interrupt 4. At execution of a push instruction SP At execution of a pop instruction 00241 At execution of an interrupt return instruction General-purpose registers save/restore using data transfer instructions Data transfer instruction can be used to save only a specific general-purpose register during processing of single interrupt. Example: Saving/restoring a register using data transfer instructions PINTxx: LD (GSAVA), A ; Save A register Interrupt processing LD A, (GSAVA) RETI 88CS38B-29 ; Restore A register ; Return 2007-09-12 TMP88CS38B/CM38B/CP38B (3) Interrupt return The interrupt return instructions [RETI]/[RETN] perform the following operations. [RETI] Maskable Interrupt Return [RETN] Non-maskable Interrupt Return 1. The contents of the program counter and 1. The contents of the program counter and program status the program status word are restored word are restored from the stack. from the stack. 2. The stack pointer is incremented 5 times. 2. The stack pointer is incremented 5 times. 3. The interrupt master enable flag is set to “1”. 3. The interrupt master enable flag is set to “1” only when a non-maskable interrupt is accepted in interrupt enable status. However, the interrupt master enable flag remains at “0” when so clear by an interrupt service program. 4. The interrupt nesting counter is decremented, and the interrupt nesting flag is changed. 4. The interrupt nesting counter is decremented, and the interrupt nesting flag is changed. Interrupt requests are sampled during the final cycle of the instruction being executed. Thus, the next interrupt can be accepted immediately after the interrupt return instruction is executed. Note: 1.5.2 When the interrupt processing time is longer than the interrupt request generation time, the interrupt service task is performed but not the main task. Software Interrupt (INTSW) Executing the [SWI] instruction generates a software interrupt and immediately starts interrupt processing (INTSW is highest prioritized interrupt). However, if processing of a non-maskable interrupt is already underway, executing the SWI instruction will not generate a software interrupt but will result in the same operation as the [NOP] instruction. Use the [SWI] instruction only for detection of the address error or for debugging. 1. Address error detection FFH is read if for some cause such as noise the CPU attempts to fetch an instruction from a non-existent memory address. Code FFH is the SWI instruction, so a software interrupt is generated and an address error is detected. The address error detection range can be further expanded by writing FFH to unused areas of the program memory. Address-trap reset is generated in case that an instruction is fetched from RAM, SFR or DBR areas. 2. Debugging Debugging efficiency can be increased by placing the SWI instruction at the software break point setting address. 1.5.3 External Interrupts The TMP88CS38B/CM38B/CP38B each have five external interrupt inputs ( INT0 , INT2, INT3, INT4 and INT5 ). Three of these are equipped with digital noise rejection circuits (Pulse inputs of less than a certain time are eliminated as noise). Edge selection is also possible with INT2, INT3 and INT4. The INT0 /P50 pin can be configured as either an external interrupt input pin or an input/output port, and is configured as an input port during reset. Edge selection, noise rejection control except INT3 pin input and INT0 /P50 pin function selection are performed by the external interrupt control register (EINTCR). Edge selecting and noise rejection control for INT3 pin input are preformed by the remote control signal preprocessor control registers. (Refer to the section of the remote control signal preprocessor.) When INT0EN = 0, the IL3 will not be set even if the falling edge of INT0 pin input is detected. 88CS38B-30 2007-09-12 TMP88CS38B/CM38B/CP38B Table 1.5.2 External Interrupts Source INT0 Pin Secondary Function Pin Enable Conditions Edge Digital Noise Rejection IMF = 1, INT0EN = 1, EF3 = 1 Falling edge Any pulse shorter than 2/fc [s] is regarded as noise and removed. Pulses not shorter than 7/fc [s] are definitely regarded as signals. Pulses of less than 7/fc [s] are eliminated as noise. Pulses equal to or more than 25/fc [s] are regarded as signals. INT0 P50/TC2/ PWM8 INT2 INT2 P53/TC1/ SCK1 / AIN0/ KWU0 IMF·EF7 = 1 Falling edge or rising edge INT3 INT3 P30/RXIN IMF·EF11 = 1 Falling edge, rising edge or falling/rising edge Refer to the section of the remote control preprocessor INT4 INT4 P31/TC3 IMF·EF12 = 1 Falling edge or rising edge Pulses of less than 7/fc [s] are eliminated as noise. Pulses of 25/fc [s] or more are considered to be signals. Falling edge Any pulse shorter than 2/fc [s] is regarded as noise and removed. Pulses not shorter than 7/fc [s] are definitely regarded as signals. INT5 INT5 P20/ STOP IMF·EF15 = 1 Note 1: The noise rejection function is also affected for timer counter input 1 (TC1 pin). Note 2: If a noiseless signal is input to the external interrupt pin in the NORMAL or IDLE mode, the maximum time from the edge of input signal until the IL is set is as follows: (1) INT2, INT4 pin 31/fc [s] (2) INT3 pin Refer to the section of the remote control preprocessor. Note 3: If a dual-function pin is used as an output port, changing data or switching between input and output generates a pseudo interrupt request signal. To ignore this signal, it is necessary to reset the interrupt enable flag. Note 4: If INT0EN = “0”, detecting the falling edge of the INT0 pin input does not set the interrupt latch IL3. 88CS38B-31 2007-09-12 TMP88CS38B/CM38B/CP38B EINTCR (00037H) 7 6 INT0 EN “0” INT0EN INT4ES INT2ES Note 1: 5 − 4 INT4 ES P50/ INT0 pin configuration INT4 and INT2 edge select 3 − 2 INT2 ES 1 0 “0” − (Initial value: 00*0 *00*) 0: P50 input/output port 1: INT0 pin (Port P50 should be set to an input mode) Write only 0: Rising edge 1: Falling edge fc: High-frequency clock [Hz], *: Don’t care Note 2: Edge detection during switching edge selection is invalid. Note 3: Do not change EINTCR only when IMF = 1. After changing EINTCR, interrupt latches of external interrupt inputs must be cleared to “0” using load instruction. Note 4: In order to change of external interrupt input by rewriting the contents of INT2ES and INT4ES during NORMAL mode, clear interrupt latches of external interrupt inputs (INT2 and INT4) after 8 machine cycles from the time of rewriting. Note 5: In order to change an edge of timer counter input by rewritng the contents of INT2ES during NORMAL mode, rewrite the contents after timer counter is stopped (TC*s = 0) , that is, terrupt disable state. Then, clear a interrupt latch of external interrupt input (INT2) after 8 machine cycles from the time of rewriting to change to interrupt enable state. Finally, start timer counter. Example: When change TC1 pin inputs edge in external trigger timer mode from rising edge falling edge. LD (TC1CR) , 01001000B ; TC1S ← 00 (Stops TC1) DI ; IMF ← 0 (Disables interrupt service) LD (EINTCR) , 00000100B ; INT2ES ← 1 (Change edge selection) NOP to 8-machine NOP cycles LD (ILL) , 01111111B ; IL7 ← 0 (Clears interrupt latch) EI ; IMF ← 1 (Enable interrupt service) LD (TC1CR) , 01111000B ; TC1S ← 11 (Starts TC1) Figure 1.5.5 External Interrupt Control Register 88CS38B-32 2007-09-12 TMP88CS38B/CM38B/CP38B 1.6 Reset Circuit The TMP88CS38B/CM38B/CP38B has four types of reset generation procedures: An external reset input, an address trap reset output, a watchdog timer reset output and a system clock reset output. Table 1.6.1 shows on-chip hardware initialization by reset action. The malfunction reset output circuit such as watchdog timer reset, address trap reset and system clock reset is not initialized when power is turned on. The RESET pin can output level “L” at the maximum 24/fc [s] (1.5 μs at 16 MHz) when power is turned on. Table 1.6.1 Initializing Internal Status by Reset Action On-chip Hardware Initial Value Program counter (PC) (FFFFEH to FFFFCH) Stack pointer (SP) Not initialized General-purpose registers (W, A, B, C, D, E, H, L) Not initialized Register bank selector (RBS) 0 (JF) 1 Zero flag (ZF) Not initialized Carry flag (CF) Not initialized Half carry flag (HF) Not initialized Sign flag (SF) Not initialized Overflow flag (VF) Not initialized (IMF) 0 Jump status flag Interrupt master enable flag Interrupt individual enable flags 1.6.1 − Enable Refer to I/O port circuitry Control registers Refer to each of control register RAM Not initialized 0 − 0 Output latches of I/O ports 0 (IL) Initial Value Prescaler and divider of timing generator Watchdog timer (EF) Interrupt latches On-chip Hardware External Reset Input The RESET pin contains a Schmitt trigger (Hysteresis) with an internal pull-up resistor. When the RESET pin is held at “L” level for at least 3 machine cycles (12/fc [s]) with the power supply voltage within the operating voltage range and oscillation stable, a reset is applied and the internal state is initialized. When the RESET pin input goes high, the reset operation is released and the program execution starts at the vector address stored at addresses FFFFCH to FFFFEH. VDD Reset input RESET Watchdog timer reset Sink open drain Malfunction reset output circuit Address trap reset System clock reset Figure 1.6.1 Reset Circuit 88CS38B-33 2007-09-12 TMP88CS38B/CM38B/CP38B 1.6.2 Address-trap-reset If the CPU should start looping for some cause such as noise and an attempt be made to fetch an instruction from the on-chip RAM, DBR or the SFR area, address-trap-reset will be generated. Then, the RESET pin output will go low. The reset time is about 8/fc to 24/fc [s] (0.5 to 1.5 μs at 16 MHz). Instruction execution JP a Reset release Address trap is occurred (“L” output) RESET output 8/fc to 24/fc [s] Note 1: Instruction at address (High-Z) 4/fc to 12/fc [s] 20/fc [s] (No wait) 0 ≤ a ≤ 008BFH (TMP88CS38B), 0 ≤ a ≤ 006BFH (TMP88CM38B/CP38B) 0 ≤ a ≤ 002BFH (The ROM corrective function is enabled.) Note 2: During reset release, reset vector “r” is read out, and an instruction at address “r” is fetched and decoded. Figure 1.6.2 Address-trap-reset 1.6.3 Watchdog Timer Reset Refer to section 2.4 “Watchdog Timer”. 1.6.4 System-clock-reset Clearing bits 7 in SYSCR2 to “0”, system clock stops and causes the microcomputer to deadlock. This can be prevented by automatically generating a reset signal whenever bits 7, 6 and 5 in SYSCR2 = 000 is detected to continue the oscillation. The RESET pin output goes low from high-impedance. The reset time is about 8/fc to 24/fc [s] (0.5 to 1.5 μs at 16 MHz). 88CS38B-34 2007-09-12 TMP88CS38B/CM38B/CP38B 1.7 ROM Corrective Function The ROM corrective function can patch the part (s) of on-chip ROM with some bugs. The ROM corrective function have two modes. One is to replaced the instruction on a certain address in the ROM with the jump instruction to branch into the RAM area where the patched codes (Program jump mode). The other is to replace a byte or a word (2 or 3 bytes) length data in the ROM with the patched data (Data replacement mode). Four independent location can be patched. Note 1: When use ROM corrective circuit, it is necessary to contain a program which operates to load patched program and/or replacement data from external memory into an internal data RAM in an initial routine. Note 2: The address of a instruction for IDLE mode can not be specificated as start address of corrective area. Note 3: The BM88CS38N0A does not support the ROM corrective circuit. Use the TMP88PS38B to debug a program of this circuit. In this case, note the following. In program jump mode, jump target addresses that can be specified with the TMP88CM38B/CP38B (002C0H to 006BFH) are different from those that can be specified with the TMP88PS38B (002C0H to 008BFH). Therefore, if a jump target address is within a range of 006C0H to 008BFH, it is necessary to change this addresse and also addresses for loading a patch program. Example: ROM corrective circuit ROMCDR RAM ROMCDR Serial bus interface 88CS38B-35 • Correction mode • Correction code • Patch program 2007-09-12 TMP88CS38B/CM38B/CP38B Match signal Address compare circuit 23 to Register 6 selection 5 circuit 4 3 2 1 0 5 BANK3 Data bus BANK2 Address bus BANK1 Configuration BANK0 1.7.1 Instruction fetch control circuit to to the lower the middle the upper the lower Compare address register the middle the upper Data register CM CM CM CM 0 1 2 3 Corrective mode signal Write data count Register write signal ROMCDR WDC CM3-0 ROM corrective data register Write data count register ROM corrective control register Figure 1.7.1 ROM Corrective Circuit 88CS38B-36 2007-09-12 TMP88CS38B/CM38B/CP38B 1.7.2 Control The ROM corrective function is controlled by ROM corrective control register (ROMCCR) and ROM corrective data register (ROMCDR). ROM Corrective Control Register ROMCCR (00FE0H) 7 6 5 4 3 2 1 0 − − − − CM3 CM2 CM1 CM0 CM3 Corrective mode setting (BANK3) CM2 Corrective mode setting (BANK2) CM1 Corrective mode setting (BANK1) CM0 Corrective mode setting (BANK0) (Initial value: **** 0000) 0: Proguram jump mode R/W 1: Data replacement mode ROM Corrective Status Register ROMCSR (00FE1H) 7 6 5 − − − WDC Write data counter 4 3 2 WDC 1 0 (Initial value: ***0 0000) Read only Counting the number of the byte written in ROMCDR ROM Corrective Data Register ROMCDR (00FE2H) 7 6 5 4 3 2 1 0 (Initial value: 0000 0000) ROMC Write only ROM Corrective data register Figure 1.7.2 ROM Corrective Control Register, Status Register and ROM Corrective Data Register (1) Enable and disable The ROM corrective function is disabled after releasing reset. It is enabled after setting the data for one bank into ROMCDR. And the address-trap-reset is not generated when fetching an instruction from the RAM area except the address 02C0H to 08BFH. After the ROM corrective function is enabled, it is necessary to reset the microcontroller in order to disable it. (2) Data replacement mode The ROM corrective function has the program jump mode and the data replacement mode. By setting CMx (x: 0 to 3) in ROMCCR, the data replacement mode is selected. (3) The ROM corrective data register writing The ROM corrective data register has four banks corresponding to four independent locations to patch. The write data counter (WDC) points each bank set. (Figure 1.7.2) 88CS38B-37 2007-09-12 TMP88CS38B/CM38B/CP38B ROM Corrective Data Register ROMCDR (00FE2H) ROMC7 ROMC6 ROMC5 ROMC4 ROMC3 ROMC2 ROMC1 ROMC0 (Initial value: 0000 0000) The value of WDC after writing a data to ROMCDR 00000 (Initial value) BANK0 BANK1 BANK2 BANK3 The lower start address of the corrective area (8 bits) 00001 The middle start address of the corrective area (8 bits) 00010 The upper start address of the corrective area (4 bits) 00011 The lower 8 bits of the jump address/replacement data 00100 The middle 8 bits of the jump address/replacement data 00101 The upper 4 bits of the jump address/replacement data 00110 The lower start address of the corrective area (8 bits) 00111 The middle start address of the corrective area (8 bits) 01000 The upper start address of the corrective area (4 bits) 01001 The lower 8 bits of the jump address/replacement data 01010 The middle 8 bits of the jump address/replacement data 01011 The upper 4 bits of the jump address/replacement data 01100 The lower start address of the corrective area (8 bits) 01101 The middle start address of the corrective area (8 bits) 01110 The upper start address of the corrective area (4 bits) 01111 The lower 8 bits of the jump address/replacement data 10000 The middle 8 bits of the jump address/replacement data 10001 The upper 4 bits of the jump address/replacement data 10010 The lower start address of the corrective area (8 bits) 10011 The middle start address of the corrective area (8 bits) 10100 The upper start address of the corrective area (4 bits) 10101 The lower 8 bits of the jump address/replacement data 10110 The middle 8 bits of the jump address/replacement data 10111 The upper 4 bits of the jump address/replacement data 00000 Note 1: Note 2: WDC value equals to the number of the byte stored in ROMCDR. ROMCDR is set in order of the lower (8 bits), the middle (8 bits) and the upper (4 bits) start address of the corrective area, the lower (8 bits), the middle (8 bits) and the upper (4 bits) of the jump address/the replacement data. Figure 1.7.3 Banks and WDC Value of the Program Corrective Data Register Whenever ROMCDR is written, WDC is incremented to indicate what data is writen via ROMCDR. During reset, WDC is initialized to “0”. (1) The lower start address of the corrective area (8 bits) (2) The middle start address of the corrective area (8 bits) (3) The upper start address of the corrective area (4 bits) (4) The lower jump address/replacement data (8 bits) (5) The middle jump address/replacement data (8 bits) (6) The upper jump address (4 bits)/replacement data Note 1: Corrective addresses must have over five addresses each other. Note 2: The address of a instruction for IDLE mode can not be specificated as start address of corrective area. 88CS38B-38 2007-09-12 TMP88CS38B/CM38B/CP38B 1.7.3 Functions The ROM corrective function can correct maximum four ROM areas with their corresponding four banks of ROM corrective registers. Either program jump mode or data replacement mode is selected for each bank by CM0 to CM3 respectively. (1) Program jump mode In the program jump mode, the system executes a jump instruction when the program execution reaches the instruction at the corrective ROM address, skips from the instruction which would have been executed, and executes an instruction at a preset jump address. Clearing ROMCCR CMx (x: 0 to 3) to “0” puts the system in the program jump mode. Use ROMCDR to set the corrective ROM address and jump address. When the start address of an erroneous program is a corrective ROM address, and that of the patch program is a jump address, the bug in the erroneous program can be fixed. Note that the patch program should end with a jump instruction, which causes a return to the built-in ROM. Note: For program jump mode, the address to be corrected must be the start address of the instruction. Example 1: Setting the program correction circuit with the initial routine Using the initial routine program, which is executed right after reset, set the program correction circuit's register and stores the patch program into the built-in RAM as follows. 1. Read the flag, which indicates whether to use the program correction circuit, from the external memory. 2. If that circuit is not used, perform normal initial processing. 3. If it is used, clear CMx to 0 to establish the program jump mode. 4. Read the corrective ROM address and jump address from the external memory. 5. Set the corrective ROM address and jump address, which were read in step “4.”, in ROMCDR. 6. Read the number of bytes for the patch program from the external memory. 7. Read the program with a number of bytes, equal to the byte count read in step “6.”, from the external memory, and store that program into the built-in RAM. 8. Repeat steps “4.” through “7.” as many times as there are required banks. Example 2: There is bugs on the locations from 0C020H to 0C085H The corrective address, the jump vector, the program patch codes and other information to patch the ROM with the bugs must be read out from any of memory storage that holds them during initial program routine. CMn = 0 specifies the program jump mode. Subsequently, the patch program codes are loaded into RAM (00600H to 006EFH). The start address (0C020H) of the ROM necessary to patch is written to the corrective ROM address registers, and the start address (00600H) of the RAM area to patch is loaded onto the jump address registers. When the instruction at 0C020H is fetched, the instruction to jump into 00600H is unconditionally executed instead of the instruction at 0C020H, and the subsequent patch program codes are executed. The jump instruction at the end of the patch program codes returns to the ROM at 0C086H. 88CS38B-39 2007-09-12 TMP88CS38B/CM38B/CP38B 00000H SFR 0003FH 00040H 00600H Patch program RAM 0083FH JP 0C086H 006EFH 006F0H 00F80H DBR 00FFFH Return 04000H ROM 0C020H 0C085H Bug area 0C086H FFFFFH Note: Corrective address must be assigned to 1st byte of instruction codes on the program jump mode. (2) Data replacement mode In the data replacement mode, the system replaces reference data stored in the ROM area with the new instead of correcting the data reference instruction when that reference data is changed. The program jump mode reduces the complexity of correcting the processing routine. However, when this mode is used, if there is a need to replace only the fixed data in ROM, the instruction to reference this ROM data should be corrected. Thus, a large amount of ROM is required for the patch program. To avoid this, the system has the data replacement mode. With this mode, three consecutive bytes of data can be replaced for each bank. (For an instruction which accesses only one byte, only the first byte can be replaced. For an instruction which accesses only two bytes, the two consecutive bytes can be replaced.) Setting ROMCCR CMx (x: 0 to 3) to “1” puts the system in the data replacement mode. Specify the start address of ROM data to be replaced as the corrective ROM address. Then, specify the new three-byte data as the patch data. Note: For data replacement mode, the corrective address should be the address of fixed data (including a vector). (The operation code and operand cannot be changed.) Example 1: Setting the program correction circuit with the initial routine Using the initial routine program, which is executed right after reset, set the program correction circuit's register as follows. 1. Read the flag, which indicates whether to use the program correction circuit, from the external memory. 2. If that circuit is not used, perform normal initial processing. 3. If it is used, set CMx to “1” to establish the data replacement mode. 4. Read the address of the data to be replaced and the patch data from the external memory. 5. Set the address and patch data, which were read in step “4.”, in ROMCDR. 6. Repeat steps “4.” and “5.” as many times as there are required banks. 88CS38B-40 2007-09-12 TMP88CS38B/CM38B/CP38B Example 2: Replacing data 55H at 0C020H with 33H Using the initial routine program, which is executed right after reset, read the start address of the data to be replaced and the patch data from the external memory. Set CMx (x: 0 to 3) to “1” to change the correction mode to the data replacement mode. Specify the start address (0C020H) of the data to be replaced as the corrective ROM address. Then, specify the new three-byte data (33H for 0C020H, CCH for 0C021H, and C3H for 0C022H) as the patch data. 00000H SFR 0003FH 00040H RAM 0083FH 00F80H DBR 00FFFH 04000H ROM 0C020H 55H 33H 0C021H AAH CCH 0C022H A5H 3CH Replacement data FFFFFH 1. 2. 3. 4. At HL = 0C020H, Executing LD A, (HL) loads 33H in A. (Data replacement) At HL = 0C021H, Executing LD A, (HL) loads AAH in A. (No data replacement) At HL = 0C020H, Executing LD WA, (HL) loads CC33H in WA. (Data replacement) At HL = 0C020H, Executing LD IX, (HL) loads CCC33H in IX. (Data replacement) Note 1: Corrective address must be assigned to constant data area on the data replacement mode. (Ope-code and ope-rand can’t be replaced by ROM correction circuit.) Note 2: Instructions which includes “(HL+)” or “(−HL) ” operation can't be replaced by ROM corrective circuit on the data replacement mode. 88CS38B-41 2007-09-12 TMP88CS38B/CM38B/CP38B 2. On-chip Peripheral Functions 2.1 Special Function Registers (SFR) and Data Buffer Registers (DBR) The TLCS-870/X series uses the memory mapped I/O system and all peripheral control and data transfers are performed through the special function registers (SFRs) and data buffer registers (DBR). The SFR are mapped to addresses 00000H to 0003FH, and DBR are mapped to address 00F80H to 00FFFH. Figure 2.1.1 shows the list of the TMP88CS38B/CM38B/CP38B SFRs and-DBRs. Address 00000H 00001 00002 00003 00004 00005 00006 00007 00008 00009 0000A 0000B 0000C 0000D 0000E 0000F 00010 00011 00012 00013 00014 00015 00016 00017 00018 00019 0001A 0001B 0001C 0001D 0001E 0001F Read Write Reserved Reserved P2 port P3 port P4 port P5 port P6 port P7 port − P5CR1 (P5 port I/O control1) − P7CR (P7 port I/O control) Reserved Reserved − P4CR (P4 port I/O control) − P6CR (P6 port I/O control) ADCCRA (AD converter control A) ADCCRB (AD converter control B) TC1DRAL (Timer register 1A) TC1DRAH TC1DRBL − (Timer register 1B) − TC1DRBH TC1CR (TC1 control) − TC2CR (TC2 control) − TC2DRL − TC2DRH (Timer register 2) TC3DRA (Timer register 3A) − TC3DRB (Timer register 3B) − TC3CR (TC3 control) − TC4DR (Timer register 4) − TC4CR (TC4 control) ORDSN (OSD control) ORCRAL (OSD control) ORCRAH (OSD control) Address 00020H 00021 00022 00023 00024 00025 00026 00027 00028 00029 0002A 0002B 0002C 0002D 0002E 0002F 00030 00031 00032 00033 00034 00035 00036 00037 00038 00039 0003A 0003B 0003C 0003D 0003E 0003F Read Write SBISRA (SBI statusA) SBICRA (SBI control register A) SBIDBR (SBI data buffer) − I2CAR (I2C bus address) SBISRB (SBI statusB) SBICRB (SBI control register B) − ORDMAL (OSD control) − ORDMAH (OSD control) RCSR (TC3 status) RCCR (TC3 control) PMPXCR (Port control) − PWMCR1A (PWM control 1A) − PWMCR1B (PWM control 1B) − PWMDBR1 (PWMDBR1) − P3CR1 (P3 I/O control) EIRE (Interrupt enable register) EIRD ILE (Interrupt latch) ILD CGCR (Divider control) ADCDR1 (AD conversion result) ADCDR2 (AD conversion result) Reserved − WDTCR1 − WDTCR2 Watchdog timer control TBTCR (TBT/TG control) − EINTCR (External interrupt control) SYSCR1 (System control) SYSCR2 EIRL (Interrupt enable register) EIRH ILL (Interrupt latch) ILH PSWL (Program status word) PSW H (a) Special function registers Note 1: Do not access reserved areas by the program. Note 2: −: Cannot be accessed. Note 3: Write-only registers and interrupt latches cannot use the read-modify-write instructions (Bit manipulation instructions such as SET, CLR, etc. and logical operation instructions such as AND, OR, etc.). Note 4: When defining address 0003FH with assembler symbols, use GRBS. Address 0003EH must be GPSW/GFLAG. Figure 2.1.1 (a) SFR 88CS38B-42 2007-09-12 TMP88CS38B/CM38B/CP38B Address 00F80H 81 Read Write ORDON (OSD control) − OSD control register − A1 A2 OSD control register Reserved − OSD control register − B9 BA OSD control register ORIRC (OSD display counter) ORIRC (OSD interrupt control) − OSD control register − OSD control register C0 Reserved Reserved D0 D1 IDLEINV (Key-on wakeup status) IDLECR (Key-on wakeup control) Reserved Reserved D8 D9 DA DB DC DD DE DF E0 E1 E2 E3 E4 E5 E6 E7 E8 E9 EA EB EC ED EE EF F0 F1 F2 F3 F4 F5 F6 F7 F8 SINTCR (Data slicer interrupt control) − DACLCR (Sync. tip slice level setting) SLVLCR (Slice level control) SIFDR1 (Caption data 1st byte) − SIFDR2 (Caption data 2nd byte) − − SIFSR (Data slicer status) − − SIFS1R (Data slicer status2) SIFSMS1 (Data slicer mode setting) ROMCCR (ROM corrective control) − ROMCSR (ROM corrective status) − ROMCDR (ROM corrective data) Reserved JECR (Jitter elimination control) − JESR (Jitter elimination status) − TVSCR (Test video signal output) Reserved RXCR1 (Remote control recieve control 2) RXCR2 (Remote control recieve control 1) RXCTR (Remote control receive counter) − RXDBR (Remote control receive data buffer) − RXSR (Remote control status) − Reserved FC8CR (FC8 control) Reserved Reserved SCCRB (Serial clock source control) SCSR (Serial clock source status) Reserved Reserved Reserved − PWMCR2A (PWM control 2A) − PWMCR2B (PWM control 2B) − PWMDBR2 (PWM data buffer) Reserved Reserved FE FF − PSELCR (P3, P5 control 2) Reserved (b) Data buffer registers Note 1: Do not access reserved areas by the program. Note 2: −: Cannot be accessed. Note 3: Write-only registers cannot use the read-modify-write instructions (Bit manipulation instructions such as SET, CLR, etc. and logical operation instructions such as AND, OR, etc.). Figure 2.1.2 (b) DBR 88CS38B-43 2007-09-12 TMP88CS38B/CM38B/CP38B 2.2 I/O Ports The TMP88CS38B/CM38B/CP38B has 6 parallel input/output ports (33 pins) as follows: Primary Function Secondary Functions Port P2 1-bit I/O port External interrupt input, and STOP mode release signal input Port P3 6-bit I/O port External interrupt input, remote control signal input, data slicer analog input, timer/counter input, serial bus interface input/output and data slicer input Port P4 8-bit I/O port Pulse width modulation output Port P5 8-bit I/O port Pulse width modulation output external interrupt input, timer/counter input, key-on wakeup input, serial bus interface input/output, analog input and I output from OSD circuitry. Port P6 8-bit I/O port R, G, B and Y/BL output from OSD circuitry, R.G.B and Y/BL input, analog input, test video signal output and key-on wakeup input Port P7 2-bit I/O port Horizontal synchronous pulse input and vertical synchronous pulse input to OSD circuitry Each output port contains a latch, which holds the output data. All input ports do not have latches, so the external input data should either be held externally until read or reading should be performed several times before processing. Figure 2.2.1 shows input/output timing examples. External data is read from an I/O port in the S1 state of the read cycle during execution of the read instruction. This timing can not be recognized from outside, so that transient input such as chattering must be processed by the program. Output data changes in the S2 state of the write cycle during execution of the instruction which writes to an I/O port. Fetch cycle Instruction execution cycle Fetch cycle Read cycle S0 S1 S2 S3 S0 S1 S2 S3 S0 S1 S2 S3 Ex.: LD A, (x) Input strobe Data input (a) Input timing Fetch cycle Instruction execution cycle Fetch cycle Write cycle S0 S1 S2 S3 S0 S1 S2 S3 S0 S1 S2 S3 Ex.: LD (x), A Output latch pulse Data output (b) Output timing Note: The positions of the read and write cycles may vary, dispending on the instruction. Figure 2.2.1 Input/Output Timing (Example) 88CS38B-44 2007-09-12 TMP88CS38B/CM38B/CP38B When reading an I/O port except programmable I/O ports, whether the pin input data or the output latch contents are read depends on the instructions, as shown below: (1) Instructions that read the output latch contents 1. XCH r, (src) 2. SET/CLR/CPL (src).b 3. SET/CLR/CPL (pp).g 4. LD (src).b, CF 5. LD (pp).b, CF 6. ADD/ADDC/SUB/SUBB/AND/OR/XOR (src), n 7. (src) side of ADD/ADDC/SUB/SUBB/AND/OR/XOR (src), (HL) (2) Instructions that read the pin input data 1. Instructions other than the above (1) 2. (HL) side of ADD/ADDC/SUB/SUBB/AND/OR/XOR (src), (HL) 2.2.1 Port P2 (P20) Port P2 is a 1bit input/output port. It is also used as an external interrupt input, and a STOP mode release signal input. When used as an input port, or a secondary function pin, the output latch should be set to “1”. During reset, the output latch is initialized to “1”. It is recommended that pin P20 should be used as an external interrupt input, a STOP mode release signal input, or an input port. If used as an output port, the interrupt latch is set on the falling edge of the P20 output pulse. When a read instruction for port P2 is executed, bits 7 to 1 in P2 are read in as undefined data. SET/CLR/CPI/others Output latch D Data input P20 ( INT5 / STOP ) Q Data input Control input STOP OUTEN 7 P2 (00002H) 6 5 4 3 2 1 0 P20 INT5 STOP (Initial value: **** ***1) *: Don’t care Figure 2.2.2 Port P2 88CS38B-45 2007-09-12 TMP88CS38B/CM38B/CP38B 2.2.2 Port P3 (P35 to P30) Port P3 is an 6-bit input/output port which can be configured as an input or an output in one-bit unit under software control. Input/output mode is specified by the corresponding bit in the port P3 input/output control register 1 (P3CR1). Port P3 is configured as an input if its corresponding P3CR1 bit is cleared to “0”, and as an output if its corresponding P3CR1 bit is set to “1”. During reset, P3CR1 is initialized to “0”, which configures port P3 as an input. The P3 output latches are also initialized to “1”. Data is written into the output latch regardless of the P3CR1 contents. Therefore initial output data should be written into the output latch before setting P3CR1. Port P3 is also used as an external interrupt input, remote-control signal input a timer/counter input, data slicer input and serial bus interface input/output. When used as a secondary function input pin except I2C bus interface input/output, the input pins should be set to the input mode. When used as a secondary function output pin except I2C bus interface input/output, the output pins should be set to the output mode and beforehand the output latch should be set to “1”. When P34 and P35 are used as I2C bus interface input/output, P3CR2 bits should be set to the sink open-drain mode, the output latches should be set to “1”, and the output pins should be set to the output mode. Note: Input mode port is read the state of input pin. When input/output mode is used mixed, the contents of output latch setting input mode may be changed by executing bit manipulation instructions. Example 1: Outputs an immediate data 5AH to port P3 LD (P3), 5AH ; P3 ← 5AH Example 2: Inverts the output of the lower 4 bits (P33 to P30) in port P3 XOR (P3), 00001111B ; 88CS38B-46 P33 to P30 ← P33 to P30 2007-09-12 TMP88CS38B/CM38B/CP38B STOP OUTEN STOP OUTEN P3iCR1 P3jCR1 Data input Data input Control input Control input (*1) P3iCR2 Data output D Q D Data output Q P3j P3i Control output Output latch Output latch VIN (*2) (a) P35 to P34 7 6 P3 (00003H) P3CR1 7 5 P35 SDA0 4 P34 SCL0 3 P33 VIN0 TC4 2 P32 VIN1 CSIN 1 P31 INT4 TC3 0 P30 INT3 RXIN 5 4 3 2 1 0 6 (0002BH) (b) P33 to P30 P35CR1 P34CR1 P33CR1 P32CR1 P31CR1 P30CR1 P3CR1 7 6 (0FFEH) “0” “0” P3CR2 5 (Initial value: **00 0000) 0: Input mode I/O control for P3 PSELCR (Initial value: **11 1111) Write only 1: Output mode 4 P35CR2 P34CR2 I/O control for P3 3 2 “0” 1 P52CR2 P51CR2 0 “0” (Initial value: 0*00 *00*) 0: Sink open drain Write only 1: Tri-state (*1) only P33, P31, P30 (*2) only P33, P32 Note 1: *: Don’t care, i = 5 to 4, j = 3 to 0 Note 2: P3CR1 cannot used the read-modify-write instructions. (Bit manipulation instructions such as SET, CLR, etc. and logical operation such as AND, OR, etc.) Note 3: Clear bit7, 6, 3 and 0 to “0” in PSELCR. Figure 2.2.3 Port P3 and P3CR 88CS38B-47 2007-09-12 TMP88CS38B/CM38B/CP38B 2.2.3 Port P4 (P47 to P40) Port P4 is an 8-bit input/output port which can be configured as an input or an output in one-bit unit under software control. Input/output mode is specified by the corresponding bit in the port P4 input/output control register (P4CR). Port P4 is configured as an input if its corresponding P4CR bit is cleared to “0”, and as an output if its corresponding P4CR bit is set to “1”. During reset, P4CR is initialized to “0”, which configures port P4 as an input. The P4 output latches are also initialized to “1”. Data is written into the output latch regardless of the P4CR contents. Therefore initial output data should be written into the output latch before setting P4CR. Port P4 is also used as a pulse width modulation (PWM) output. When used as a PWM output pin, the output pins should be set to the output mode and beforehand the output latch should be set to “1”. Note: Input mode port is read the state of input pin. When input/output mode is used mixed, the contents of output latch setting input mode may be changed by executing bit manipulation instructions. STOP OUTEN P4iCR Data input D Data output Q P4i Output latch PWMi P4 (00004H) P4CR (0000CH) 7 P47 6 P46 5 P45 4 P44 3 P43 2 P42 1 P41 P40 0 PWM7 PWM6 PWM5 PWM 4 PWM3 PWM2 PWM1 PWM0 7 6 5 4 3 2 1 0 P47CR P46CR P45CR P44CR P43CR P42CR P41CR P40CR P4CR Note 1: Note 2: I/O Control for port P4 (Initial value: 1111 1111) (Initial value: 0000 0000) 0: Input mode Write only 1: Output mode i = 7 to 0. P4CR cannot used the read-modify-write instructions. (Bit manipulation instructions such as SET, CLR, etc. and logical operation such as AND, OR, etc.) Figure 2.2.4 Port P4 and P4CR 88CS38B-48 2007-09-12 TMP88CS38B/CM38B/CP38B 2.2.4 Port P5 (P57 to P50) Port P5 is an 8-bit input/output port which can be configured as an input or an output in one-bit unit under software control. Input/output mode is specified by the corresponding bit in the port P5 input/output control register 1 (P5CR1). Port P5 is configured as an input if its corresponding P5CR1 bit is cleared to “0”, and as an output if its corresponding P5CR1 bit is set to “1”. During reset, P5CR1 is initialized to “0”, which configures port P5 as an input. The P5 output latches are also initialized to “1”. Data is written into the output latch regardless of the P5CR1 contents. Therefore initial output data should be written into the output latch before setting P5CR1. Port P5 is also used as is also used as AD converter analog input, a pulse width modulation (PWM) output external interrupt input, timer/counter input, serial bus interface input/output, and an on screen display (OSD) output (I signal). When used as a secondary function input pin except I2C bus interface input/output, the input pins should be set to the input mode. When used as a secondary function output pin except I2C bus interface input/output, the output pins should be set to the output mode and beforehand the output latch should be set to “1”. When P52 and P51 are used as I2C bus interface input/output, P5CR2 bits should be set to the sink open-drain mode, the output latches should be set to “1”, and the output pins should be set to the output mode. When P57 is used as an OSD output pin, the output pin should be set to the output mode and beforehand the port 6 data selection register (PIDS) should be clear to “0”. When used as port P5, the port 6 data selection register (PIDS) should be set to “1”. Note: Input mode port is read the state of input pin. When input/output mode is used mixed, the contents of output latch setting input mode may be changed by executing bit manipulation instructions. 88CS38B-49 2007-09-12 TMP88CS38B/CM38B/CP38B STOP OUTEN Analog input AINDS SAIN STOP OUTEN P5jCR1 P5iCR1 Data input Data input B D Data output Q Y P5i S Output latch D Data output A Q P5j Output latch I PIDS (b) P56 to P54 (a) P57 STOP OUTEN Analog input AINDS SAIN STOP OUTEN P5kCR1 P5lCR1 Data input Control input P5lCR2 Data input Control input D Data output Control output D Data output Q P5k Output latch Q P5l Output latch Control output (d) P52 to P51 (c) P53 STOP OUTEN P5mCR1 Data input Control input D Data output Q P5m Output latch Control output (d) P50 P5 (00005H) 7 P57 I 6 P56 AIN3 5 P55 AIN2 4 P54 AIN1 3 P53 INT2 TC1 2 P52 SO1 SDA1 SCK1 1 P51 0 P50 PWM9 PWM8 SL1 SLC1 INT0 TC2 (Initial value: 1111 1111) AIN0 P5CR1 (00008H) 7 6 P5CR1 7 6 (00FFEH) “0” “0” P5CR2 (00FBAH) 4 3 2 1 0 5 4 3 P35CR2 P34CR2 2 “0” Write only 1 0 P52CR2 P51CR2 (Initial value: 0*00 *00*) 0: Skin open drain 1: Tri-state I/O Control for P5 7 6 5 4 3 2 P67S P66S P65S P64S PIDS YBLCS PIDS (Initial value: 0000 0000) 0: Input mode 1: Output mode I/O Control for P5 PSELCR ORP6S 5 P57CR1 P56CR1 P55CR1 P54CR1 P53CR1 P52CR1 P51CR1 P50CR1 Selection of the output data for port P57 Write only 1 0 MPXS (Initial value: 0000 0000) 0: The OSD output (I) 1: Port P57 output latch Note 1: *: Don’t care, i = 7, j = 6 to 4, k = 3, l = 2 to 1, m = 0. Note 2: P5CR1 cannot be used the read-modify-write instructions. Write only (Bit manipulation instructions such as SET, CLR, etc. and logical operation such as ADN, OR, etc.) Note 3: Clear bit7, 6 and 3 to “0” in PSELCR. Figure 2.2.5 Ports P5 88CS38B-50 2007-09-12 TMP88CS38B/CM38B/CP38B 2.2.5 Port P6 (P67 to P60) Port P6 is an 8-bit input/output port which can be configured as an input or an output in one-bit unit under software control. Input/output mode is selected by the corresponding bit in the port P6 input/output control register (P6CR). Port P6 is configured as an input if its corresponding P6CR bit is cleared to “0”, and as an output if its corresponding P6CR bit is set to “1” and P6nS bit is set to “1”. P63 to P60 are sink open-drain ports. During reset, P6CR is initialized to “0”, which configures port P6 as an input. The P6 output latches are also initialized to “1”. Data is written into the output latch regardless of the P6CR contents. Therefore initial output data should be written into the output latch before setting P6CR. Port P6 is used as an on screen display (OSD) output (R, G, B and Y/BL signal)/input (RIN, GIN BIN, Y/BLIN signal), a test video signal output and AD converter analog input. When used as a test video signal output pin, the output pins should be set to the output mode and beforehand the signal control register (SGEN) should be set to “1”. When used as a secondary function input, the input pins should be set to the input mode. When used as an OSD output pin, the output pins should be set to the output mode and beforehand the port P6 data selection register (P67S to P64S) should be clear to “0”. When used as port P6, the signal control register (P67 to P64) should be set to “1”. Note: Input mode port is read the state of input pin. When input/output mode is used mixed, the contents of output latch setting input mode may be changed by executing bit manipulation instructions. Example: Sets the lower 4 bits (P63 to P60) in port P6 to the output mode, and the other bit to the input mode. LD (P6CR), 0FH ; 88CS38B-51 P6CR ← 00001111B 2007-09-12 TMP88CS38B/CM38B/CP38B STOP OUTEN STOP OUTEN P6iCR P63CR Data input Data input RIN D Data output A Q Y P6i Output latch D Data output B S R, G, B, Y/BL Q P63 Output latch P6iS (a) P67 to P64 (b) P63 STOP OUTEN Analog input AINDS SAIN STOP OUTEN P6jCR P62CR Data input GIN Data input BIN, Y/BLIN D Data output A Q Y P62 Output latch D Data output B S CSOUT Q SGEN (c) P62 P6 (00006H) P6CR (0000DH) 7 P67 Y/BL 6 P66 B (00FE6H) (00FBAH) 5 P65 G 4 P64 R 3 P63 RIN 2 P62 GIN CSOUT 1 P61 BIN AIN5 0 P60 Y/BLIN AIN4 7 6 5 4 3 2 1 0 P66CR P65CR P64CR P63CR P62CR P61CR P60CR 0: Input mode 1: Output mode I/O Control for port P6 7 6 5 4 3 SGEN SGVBLK SGPAL SGIV SGCHS SGEN ORP6S (b) P61 to P60 P67CR P6CR SGCR P6j Output latch 2 7 6 5 4 3 2 P66S P65S P64S PIDS YBLCS P67S to P64S Note 1: Note 2: 1 0 SGPAT P67S Selection of the output data for port P6i (Initial value: 0000 0000) Write only (Initial value: 0000 0000) 0: Disable 1: Enable Function selection (Initial value: 1111 1111) Write only 1 0 MPXS 0: The OSD output (R, G, B, Y/BL) 1: Port P6i output latch (Initial value: 0000 0000) Write only *: Don’t care, i = 7 to 4, j = 1 to 0. P6CR and ORP6S cannot used with the read-modify-write instructions. (Bit manipulations such as SET, CLR, etc. and logical operation such as AND, OR, etc.) Note 3: Clear bit2 and 0 to “0” in TVSCR Figure 2.2.6 Ports P6, P6CR, and P67S to P64S 88CS38B-52 2007-09-12 TMP88CS38B/CM38B/CP38B 2.2.6 Port P7 (P71 to P70) Port P7 is a 2bit input/output port, and is also used as a vertical synchronous signal ( VD ) input and a horizontal synchronous signal ( HD ) input for the on screen display (OSD) circuitry. The output latches, are initialized to “1” during reset. When used as an input port or a secondary function pin, the output latch should be set to “1”. When a read instruction for port P7 is executed, bits 7 to 2 in P7 are read in as undefined data. STOP OUTEN P7iCR Data input HD , VD D Data output Q P7i Output latch 7 6 5 4 3 2 P7 (00007H) P7CR 7 6 5 4 3 I/O Control for P7 0 P70 VD HD 2 (00009H) P7CR 1 P71 1 0 P71CR P70CR 0: Input mode 1: Output mode (Initial value: **** **11) (Initial value: **** **00) Write only Note: *: Don’t care, i = 1 to 0 Figure 2.2.7 Ports P7 88CS38B-53 2007-09-12 TMP88CS38B/CM38B/CP38B 2.3 Time Base Timer (TBT) The time base timer generates time base for key scanning, dynamic displaying, etc. It also provides a time base timer interrupt (INTTBT). The time base timer is controlled by a control register (TBTCR) shown in Figure 2.3.1. An INTTBT is generated on the first falling edge of source clock (the divider output of the timing generator) after the time base timer has been enabled. The divider is not cleared by the program; therefore, only the first interrupt may be generated ahead of the set interrupt period. The interrupt frequency (TBTCK) must be selected with the time base timer disabled (When the time base timer is changed from enabling to disabling, the interrupt frequency can’t be changed.) Both frequency selection and enabling can be performed simultaneously. 16 Example: Sets the time base timer frequency to fc/2 [Hz] and enables an INTTBT interrupt. 23 24 fc/2 , fc/2 21 22 fc/2 , fc/2 16 17 fc/2 , fc/2 14 15 fc/2 , fc/2 13 14 fc/2 , fc/2 12 13 fc/2 , fc/2 11 12 fc/2 , fc/2 9 10 fc/2 , fc/2 LD (TBTCR), 00000010B ; TBTCK = “010” LD (TBTCR), 00001010B ; TBTEN = “1” SET (EIRL). 6 MPX INTTBT interrupt A request B C Source clock Source clock Rising D Y edge E detector F TBTEN G H S INTTBT 3 TBTCK TBTEN Enable TBT TBTCR Interrupt period Time base timer control register (a) Configuration (b) Time base timer interrupt Figure 2.3.1 Time Base Timer 88CS38B-54 2007-09-12 TMP88CS38B/CM38B/CP38B TBTCR (00036H) 7 6 5 4 3 “0” − − “0” TBTEN TBTEN 2 1 0 TBTCK (Initial value: 0**0 0***) 0: Disable Time base timer enable/disable 1: Enable NORMAL, IDLE mode DV1CK = 0 DV1CK = 1 23 fc/2 [Hz] fc/2 21 fc/2 22 010 fc/2 16 fc/2 17 011 fc/2 14 fc/2 15 fc/2 13 fc/2 14 fc/2 12 fc/2 13 110 fc/2 11 fc/2 12 111 fc/2 9 fc/2 10 000 001 TBTCK Time base timer interrupt frequency select 100 101 fc/2 [Hz] 24 Write only Note 1: fc: High-frequency clock [Hz], *: Don’t care Note 2: TBTCR is a write-only register and must not be used with any of read-modify-write instruction. Note 3: Set bit7 and 4 in TBTCR to “0”. Figure 2.3.2 Time Base Timer and Divider Output Control Register Table 2.3.1 Time Base Timer Interrupt Frequency (Example: at fc = 16MHz) Time Base Timer Interrupt Frequency [Hz] TBTCK NORMAL, IDLE Mode DV1CK = 0 DV1CK = 1 000 1.90 001 7.62 3.81 010 244.14 122.07 011 976.56 488.28 100 1953.12 976.56 101 3906.25 1953.12 110 111 7812.50 31250 88CS38B-55 0.95 3906.25 15625 2007-09-12 TMP88CS38B/CM38B/CP38B 2.4 Watchdog Timer (WDT) The watchdog timer is a fail-safe system to rapidly detect the CPU malfunctions such as endless looping caused by noise or the like, or deadlock and resume the CPU to the normal state. The watchdog timer signal for detecting malfunction can be selected either a reset output or a pseudo non-maskable interrupt request. However, selection is possible only once after reset. At first the reset output is selected. When the watchdog timer is not being used for malfunction detection, it can be used as a timer to generate an interrupt at fixed intervals. Note: 2.4.1 Care must be given in system design so as to protect the watchdog timer from disturbing noise. Otherwise the watchdog timer may not fully exhibit its functionality. Watchdog Timer Configuration Reset release signal from T.G. MPX 23 24 fc/2 , fc/2 21 22 fc/2 , fc/2 19 20 fc/2 , fc/2 17 18 fc/2 , fc/2 A B C D R Binary counters Y Clock Overflow 1 S Reset output WDT output 2 S Q Clear Interrupt request 2 RESET INTWDT Internal reset Q S WDTT R WDTEN Writing disable code Writing clear code WDTOUT Controller 00034H WDTCR1 00035H WDTCR2 Watchdog timer control registers MPX: Multiplexer T.G.: Timing generator Figure 2.4.1 Watchdog Timer Configuration 88CS38B-56 2007-09-12 TMP88CS38B/CM38B/CP38B 2.4.2 Watchdog Timer Control Figure 2.4.2 shows the watchdog timer control registers (WDTCR1, WDTCR2). The watchdog timer is automatically enabled after reset. (1) Malfunction detection methods using the watchdog timer The CPU malfunction is detected at follows. 1. Setting the detection time, selecting output, and clearing the binary counter. 2. Repeatedly clearing the binary counter within the setting detection time. Note: The watchdog timer consists of an internal divider and two-stage binary counter. Writing the clear code (4EH) clears the binary counter, but not the internal divider. The minimum overflow time for the binary counter might be three quarters of the WDTCR1 (WDTT) time setting depending on when the clear code (4EH) is written into the WDTCR2 register. So, write the clear code on a cycle which is shorter than that minimum overflow time. If the CPU malfunctions such as endless looping or deadlock occur for any cause, the watchdog timer output will become active at the rising of an overflow from the binary counters unless the binary counters are cleared. At this time, when WDTOUT = 1 a reset is generated, which drivers the RESET pin low to reset the internal hardware and the external circuit. When WDTOUT = 0, a watchdog timer interrupt (INTWDT) is generated. The watchdog timer temporarily stops counting in STOP mode including warm-up or IDLE mode, and automatically restarts (Continues counting) when the STOP/IDLE mode is released. 88CS38B-57 2007-09-12 TMP88CS38B/CM38B/CP38B 21 Example: Sets the watchdog timer detection time to 2 /fc [s] and resets the CPU malfunction. LD (WDTCR2), 4EH LD (WDTCR1), 00001101B ; WDTT ← 10, WDTOUT ← 1 ; Clears the binary counters LD (WDTCR2), 4EH ; Clears the binary counters Within 3/4 of WDT detection time (always clear immediately before and after changing WDTT) LD (WDTCR2), 4EH ; Clears the binary counters LD (WDTCR2), 4EH ; Clears the binary counters Within 3/4 of WDT detection time Watchdog Timer Register 1 WDTCR1 (00034H) 7 6 5 4 3 2 WDTEN WDTEN 1 0 WDTOUT WDTT (Initial value: **** 1001) 0: Disable (It is necessary to write the disable code to WDTCR2) Watchdog timer enable/disable 1: Enable NORMAL mode DV1CK = 0 WDTT Note 1: 24 2 /fc 21 2 /fc 19 2 /fc 22 2 /fc 20 2 /fc 11 Write only 26 2 /fc 23 2 /fc 01 10 WDTOUT DV1CK = 1 25 2 /fc 00 Watchdog timer detection time [s] 0: Interrupt request Watchdog timer output select 1: Reset output WDTOUT cannot be set to “1” by program after clearing WDTOUT to “0”. Note 2: fc: High-frequency clock [Hz], *: Don’t care Note 3: WDTCR1 is a write-only register and must not be used with any of read-modify-write instructions. Note 4: The watchdog timer must be disabled or the counter must be cleared immediately before entering to the STOP mode. When the counter is cleared, the counter must be cleared again immediately after releasing the STOP mode. Note 5: Just right before disabling the watchdog timer, disable the acceptance of interrupts (DI) and clear the watchdog timer. If the watchdog timer is disabled under conditions other than the above, the proper operation cannot be guaranteed. Watchdog Timer Register 2 WDTCR2 7 6 5 4 3 2 1 0 (Initial value: **** ****) (00035H) WDTCR2 Watchdog timer control code write register 4EH: Watchdog timer binary counter clear (Clear code) B1H: Watchdog timer disable (Disable code) Others: Invalid Note 1: The disable code is invalid unless written when WDTEN = 0. Note 2: *: Don’t care Note 3: The binary counter of the watchdog timer must not be cleared by the interrupt task. Note 4: Clears the binary counter does not clear the source clock. Write only It is recommended that the time to clear is set to 3/4 of the detecting time. Note 5: The watchdog timer counter must be disabled by writing the disable code (B1H) to WDRCR2 after writing WDTCR2 to. “4EH”. Figure 2.4.2 Watchdog Timer Control Registers 88CS38B-58 2007-09-12 TMP88CS38B/CM38B/CP38B (2) Watchdog timer enable The watchdog timer is enabled by setting WDTEN (Bit3 in WDTCR1) to “1”. WDTEN is initialized to “1” during reset, so the watchdog timer operates immediately after reset is released. Example: Disables watchdog timer LDW (WDTCR1), 00001000B WDTEN ← 1 ; (3) Watchdog timer disable To disable the watchdog timer, clear the interrupt mask enable flag (IMF) to “0” and write the clear code (4EH) into WDTCR2. Then, clear WDTEN (Bit3 in WDTCR1) to “0”. When WDTEN is “0”, the watchdog timer is disabled by writing the disable code (B1H) into WDTCR2. If WDTEN is cleared to “0” after the disable code has been written into WDTCR2, the watchdog timer is not disabled. While it is disabled, its binary counter is cleared. Example: DI ; Disables interrupt acceptance. LD (WDTCR2), 4EH ; Clears the watchdog timer. LDW (WDTCR1), B101H ; Disables the watchdog timer. ; Enables interrupt acceptance. EI Table 2.4.1 Watchdog Timer Detection Time (Example: fc = 16 MHz) Watchdog Timer Detection Time [s] WDTT NORMAL Mode DV1CK = 0 2.4.3 DV1CK = 1 00 2.097 4.194 01 524.288 m 1.048 10 131.072 m 262.1 m 11 32.768 m 65.5 m Watchdog Timer Interrupt (INTWDT) This is a pseudo non-maskable interrupt which can be accepted regardless of the contents of the EIR. If a watchdog timer interrupt or a software interrupt is already accepted, however, the new watchdog timer interrupt waits until the previous interrupt processing is completed (the end of the [RETN] instruction execution). The stack pointer (SP) should be initialized before using the watchdog timer output as an interrupt source with WDTOUT. Example: Watchdog timer interrupt setting up 2.4.4 LD SP, 023FH ; Sets the stack pointer LD (WDTCR1), 00001000B ; WDTOUT ← 0 Watchdog Timer Reset If the watchdog timer output becomes active, a reset is generated, which drivers the RESET pin (Sink open-drain input/output with pull-up) low to reset the internal hardware. The reset output time is about 8/fc to 24/fc [s] (0.5 to 1.5 μs at fc = 16.0 MHz). Note: If there is any fluctuation in the oscillation frequency at the start of clock oscillation, the reset time includes error. Thus, regard the reset time as an approximate value. 88CS38B-59 2007-09-12 TMP88CS38B/CM38B/CP38B 19 2 /fc [s] 17 2 /fc (WDTT = 11B) Clock Binary counter 1 2 3 0 1 2 3 0 Overflow INTWDT interrupt WDT reset output (High-Z) (“L” output) Writes 4EH to WDTCR2 Figure 2.4.3 Watchdog Timer Interrupt/Reset 88CS38B-60 2007-09-12 12 88CS38B-61 Note: 2 Clock SCAP1 Capture Window mode S AY B MPX TC1DRA CMP 16-bit timer registers 1A, 1B TC1DRB 16-bit up counter Clear INTTC1 interrupt Match Be sure to set the function of input/output pins correctly. For details, see the section on I/O port control registers. Timer/counter 1 control register TC1CR A B Y C S D Clear Set Q Command start METT1 External trigger start MPX TC1CK S A Y B 2 TC1S Decoder MPX Falling External trigger Edge detector Rising fc/2 , fc/2 7 8 fc/2 or fc/2 3 4 fc/2 or fc/2 11 TC1 pin INT2ES Pulse width measurement mode MPX B A CMP: Comparator MPX: Multiplexer Pulse width measurement mode 2.5.1 Y 2.5 S MCAP1 TMP88CS38B/CM38B/CP38B 16-Bit Timer/Counter 1 (TC1A) Configuration Figure 2.5.1 Timer/Counter 1 2007-09-12 TMP88CS38B/CM38B/CP38B 2.5.2 Control The timer/counter 1 is controlled by a timer/counter 1 control register (TC1CR) and two 16-bit timer registers (TC1DRA and TC1DRB). TC1DRA (00010, 00011H) 15 14 13 12 11 10 9 8 7 6 5 TC1DRAH (00011H) 4 3 2 1 0 TC1DRAL (00010H) Read/Write TC1DRB (00012, 00013H) TC1DRBH (00013H) TC1DRBL (00012H) Read only TC1CR (00014H) TC1M 7 6 5 “0” ACPAP1 MCAP1 METT1 MPPG1 4 3 TC1S TC1 operating mode select 2 TC1CK 00: 01: 10: 11: 1 0 Read/Write TC1M (Initial value: 0000 0000) Timer/external trigger timer/event counter mode Window mode Pulse width measurement mode Reserved NORMAL, IDLE mode DV7CK = 0, DVCK = 00 DV1CK = 0 TC1CK TC1 source clock select [Hz] fs/2 fc/2 7 fc/2 8 fc/2 3 fc/2 4 fc/2 01 10 11 TC1S 00: 01: 10: 11: TC1 start control DV1CK = 1 11 00 12 R/W External clock (TC1 pin input) Stop and counter clear Command start External trigger start at the rising edge External trigger start at the falling edge Timer Extend Event Window Pulse ○ ○ ○ ○ ○ ○ × × × × ACAP1 Auto-capture control 0: Auto-capture disable × ○ ○ × 1: Auto-capture enable MCAP1 Pulse width measurement mode control 0: Double edge capture 1: Single edge capture METT1 External trigger timer mode control 0: Trigger start 1: Trigger start and stop ○ ○ ○ ○ ○ ○ PPG ○ × ○ ○ Note 1: fc: High-frequency clock [Hz] Note 2: The timer register consists of two shift registers. A value set in the timer register is put in effect at the rising edge of the first source clock pulse that occurs after the upper data (TC1DRAH) are written. Therefore, the lower byte must be written before the upper byte (It is recommended that a 16-bit access instruction be used in writing). Writing only the lower data (TC1DRAL) does not put the setting of the timer register in effect. Note 3: Set the mode, source clock PPG control and timer F/F control when TC1 stops (TC1S = 00). Note 4: Auto capture can be used in only timer, event counter, and window modes. Note 5: Values to be loaded to timer registers must satisfy the following condition. TC1DRA > TC1DRB, TC1DRA > 1 Note 6: Always write “0” to TFF1 except PPG output mode. Note 7: On entering STOP mode, the TC1 start control (TC1S) is cleared to “00” automatically. So, the timer stops. Once the STOP mode has been released, to start using the timer counter, set TC1S again. Note 8: In the Auto-capture function, when the capture value is read after stop and clear counter or Auto-capture disable is executed by the TC1 start control (TC1S), the correct capture value might not be able to be read.When using Auto-capture function, set capture to enable. Note 9: Since the up-counter value is captured into TC1DRB by the source clock of up-counter after setting TC1CR<ACAP1> to “1”. Therefore, to read the captured value, wait at least one cycle of the internal source clock before reading TC1DRB for the first time. Figure 2.5.2 Timer registers and TC1 control register 88CS38B-62 2007-09-12 TMP88CS38B/CM38B/CP38B 2.5.3 Function Timer/counter 1 has five operating modes: Timer, external trigger timer, event counter, window, pulse width measurement. (1) Timer mode In this mode, counting up is performed using the internal clock. The contents of TC1DRA are compared with the contents of up counter. If a match is found, an INTTC1 interrupt is generated, and the counter is cleared to “0”. Counting up resumes after the counter is cleared. The current contents of up counter can be transferred to TC1DRB by setting ACAP1 (Bit6 in TC1CR) to “1” (Software capture function). (Auto-capture function) Table 2.5.1 Source Clock (Internal clock) for Timer/Counter 1 (Example: at fc = 16.0 MHz) NORMAL, IDLE Mode DV1CK = 0 TC1CK Resolution [μs] DV1CK = 1 Maximum Time Setting [s] Resolution [μs] Maximum Time Setting [s] 16.78 00 128.0 8.39 256.0 01 8.0 0.524 16.0 1.049 10 0.5 32.77 m 1.0 65.54 m Example 1: Sets the timer mode with source clock fc/2 [Hz] and generates an interrupt 1 later (at fc = 16 MHz) 11 LDW (TC1DRA), 1E84H ; Sets the timer register (1 s ÷ 2 /fc = 1E84H) (EIRL). 4 ; Enable INTTC1 11 DI SET EI LD (TC1CR), 00000000B ; Selects the source clock and mode LD (TC1CR), 00010000B ; Starts TC1 Example 2: Auto capture LD (TC1CR), 01010000B ; ACAP1 ← 1 (Capture) : : ; Wait at least one cycle of the internal source clock LD WA, (TC1DRB) ; Reads the capture value 88CS38B-63 2007-09-12 TMP88CS38B/CM38B/CP38B Count start Source clock Up counter 0 TC1DRA ? 1 2 3 n−1 n 0 4 1 2 3 5 4 6 7 n Match detect Counter clear INTTC1 interrupt (a) Timer mode Source clock m−2 Up counter TC1DRB m−1 m−1 ? m+1 m Capture m m+2 m+1 n−1 m+2 n+1 n n−1 Capture n+1 n ACAP1 (b) Auto capture Figure 2.5.3 Timer Mode Timing Chart (2) External trigger timer mode In this mode, counting up is started by an external trigger. This trigger is the edge of the TC1 pin input. Either the rising or falling edge can be selected with TC1S. Source clock is an internal clock. The contents of TC1DRA is compared with the contents of up counter. If a match is found, an INTTC1 interrupt is generated, and the counter is cleared to “0” and halted. The counter is restarted by the selected edge of the TC1 pin input. When METT1 (Bit6 in TC1CR) is “1”, inputting the edge to the reverse direction of the trigger edge to start counting clears the counter, and the counter is stopped. Inputting a constant pulse width can generate interrupts. When METT1 is “0”, the reverse directive edge input is ignored. The TC1 pin input edge before a match detection is also ignored. The TC1 pin input has the noise rejection; therefore, pulses of 7/fc [s] or less are rejected as noise. A pulse width of 13/fc [s] or more is required for edge detection in NORMAL or IDLE mode. Example 1: Detects rising edge in TC1 pin input and generates an interrupt 100 μs later. (at fc = 16.0 MHz, DV1CK = 1) LDW (TC1DRA), 0064H ; 100 μs ÷ 2 /fc = 64H (EIRL). 4 ; INTTC1 interrupt enable 4 DI SET EI Example 2: LD (TC1CR), 00001000B ; Selects the source clock and mode LD (TC1CR), 00101000B ; TC1 external trigger start, METT1 = 0 Generates an interrupt, inputting “L” level pulse (Pulse width: 4 ms or more) to the TC1 pin. (at fc = 16.0 MHz, DV1CK = 1) LDW (TC1DRA), 00FAH ; 4 ms ÷ 2 /fc = FAH (EIRL). 4 ; INTTC1 interrupt enable 8 DI SET EI LD (TC1CR), 00000100B ; Selects the source clock and mode LD (TC1CR), 01110100B ; TC1 external trigger start, METT1 = 1 88CS38B-64 2007-09-12 TMP88CS38B/CM38B/CP38B Count start Count start TC1S = 10 at the rising edge TC1 pin input Internal clock 0 Up counter 1 2 TC1DRA 3 n−1 n 4 n Match detect INTTC1 interrupt 1 0 2 3 Counter clear (a) Trigger start (METT1 = 0) Counter clear Count start TC1S = 10 at the rising edge Count start TC1 pin input Internal clock Up counter 0 1 2 m−1 m 3 TC1DRA 1 0 2 n−1 n 3 n Match detect INTTC1 interrupt 0 Counter clear m<n (b) Trigger start and stop (METT1 = 1) Figure 2.5.4 External Trigger Timer Mode Timing Chart (3) Event counter mode In this mode, events are counted at the edge of the TC1 pin input (Either the rising or falling edge can be selected with the external trigger TC1CR<TC1S>). The contents of TC1DRA are compared with the contents of up counter. If a match is found, an INTTC1 interrupt is generated, and the counter is cleared. Match detect is executed on other edge of count-up. A match can not be detected and INTTC1 is not generated when the pulse is still in same state. Setting ACAP1 to “1” transfers the current contents of up counter to TC1DRB (Auto-capture function). Count start TC1S = 10 at the falling edge TC1 pin input 0 Up counter TC1DRA ? 1 2 n−1 n 0 1 2 n Match detect Counter clear INTTC1 interrupt Figure 2.5.5 Event Counter Mode Timing Chart Table 2.5.2 Input Pulse Width for Timer/Counter 1 Minimum Pulse Width [s] NORMAL/IDLE 3 “H” Width 2 /fc “L” Width 2 /fc 3 88CS38B-65 2007-09-12 TMP88CS38B/CM38B/CP38B (4) Window mode Counting up is performed on the rising edge of the pulse that is the logical AND-ed product of the TC1 pin input (Window pulse) and an internal clock. The contents of TC1DRA are compared with the contents of up counter. If a match is found, an INTTC1 interrupt is generated, and the counter is cleared. Positive or negative logic for the TC1 pin input can be selected with bit4 or 5 in TC1CR. It is necessary that the maximum applied frequency be such that the counter value can be analyzed by the program. That is; the frequency must be considerably slower than the selected internal clock. Count start Count stop Count start Command start TC1 pin input Internal clock Up counter 0 TC1DRA ? 1 2 3 4 5 6 7 0 1 2 3 n Match detect Counter clear INTTC1 interrupt (a) Positive logic (at TC1S = 10) Command start Count start Count stop Count start TC1 pin input Internal clock Up counter TC1DRA 0 ? 1 2 3 4 5 6 7 8 9 0 1 9 Match detect INTTC1 interrupt Counter clear (a) Negative logic (at TC1S = 11) Figure 2.5.6 Window Mode Timing Chart (5) Pulse width measurement mode In this mode, counting is started by the external trigger (Set to external trigger start by TC1CR). The trigger can be selected either the rising or falling edge of the TC1 pin input. The source clock is used an internal clock. On the next falling (rising) edge, the counter contents are transferred to TC1DRB and an INTTC1 interrupt is generated. The counter is cleared when the single edge capture mode is set. When double edge capture is set, the counter continues and, at the next rising (falling) edge, the counter contents are again transferred to TC1DRB. If a falling (rising) edge capture value is required, it is necessary to read out TC1DRB contents until a rising (falling) edge is detected. Falling or rising edge is selected with the external trigger TC1S (Bit4 or 5 in TC1CR), and single edge or double edge is selected with MCAP1 (Bit6 in TC1CR). Note 1: Be sure to read the captured value from TC1DRB before the next trigger edge is detected. If fail to read it, it becomes undefined. It is recommended that a 16-bit access instruction be used to read from TC1DRB. Note 2: If either the falling or rising edge is used in capturing values, the counter stops at “1” after a value has been captured until the next edge is detected. So, the value captured next will become “1” larger than the value captured right after capturing starts. 88CS38B-66 2007-09-12 TMP88CS38B/CM38B/CP38B Note 3: In the Pulse width measurement mode, the capture value of the first time after the timer starts might not be a correct value. Thus, execute the dummy read once. Example: Duty measurement (Resolution fc/2 [Hz] DV1CK = 0) 7 CLR (INTTC1SW). 0 ; INTTC1 service switch initial setting: Clears bit0 of INTTC1SW. This bit is inverted by CPL instruction before INTTC1 is generated. LD (TC1CR), 00000110B ; Sets the TC1 mode and source clock (EIRL). 4 ; Enables INTTC1 LD . . . CPL (TC1CR), 00100110B ; Starts TC1 with an external trigger at MCAP1 = 0 (INTTC1SW). 0 ; Complements INTTC1 service switch JRS F, SINTTC1 LD WA, (TC1DRBL) DI SET EI PINTTC1: ; Reads TC1DRB (“H” level pulse width) Lower address in TC1DRBL: TC1DRB LD (HPULSE), WA RETI SINTTC1: LD WA, (TC1DRBL) (WIDTH), WA VINTTC1: LD . . . RETI . . . DW PINTTC1 ; Reads TC1DRB (Period) ; Duty calculation ; Sets INTTC1 WIDTH HPULSE TC1 pin INTTC1 INTTC1SW 88CS38B-67 2007-09-12 88CS38B-68 INTTC1 interrupt TC1DRB Up counter Internal clock TC1 pin input INTTC1 interrupt TC1DRB Up counter Internal clock TC1 pin input 0 Count start 0 Count start 1 1 2 2 3 3 4 4 n Capture n 0 1 n n+1 n Capture n+2 (b) Double edge capture (MCAP1 = 0) n−1 (a) Single edge capture (MCAP1 = 1) n−1 1 2 m Capture m−1 m 0 Count start Count start 4 1 m Capture m−1 m 0 3 [Application] 2 n’ n’ + 1 n’ + 2 n’ Capture (1) Period/frequency measurement (2) Duty measurement n’ − 1 [Application] High or low pulse width measurement 3 Trigger TMP88CS38B/CM38B/CP38B Figure 2.5.7 Pulse Width Measurement Mode Timing Chart 2007-09-12 TMP88CS38B/CM38B/CP38B 2.6 16-Bit Timer/Counter 2 (TC2A) 2.6.1 Configuration MPX Port (Note) TC2 pin TC2S H Window 23 24 fc/2 or fc/2 13 14 fc/2 or fc/2 8 9 fc/2 or fc/2 3 4 fc/2 or fc/2 A B CY D Clear B Timer/ event counter Y Source A clock S 16-bit up counter TC2M CMP S 3 TC2CK INTTC2 interrupt TC2S TC2DR TC2CR 16-bit timer register 2 MPX: Multiplexer CMP: Comparator TC2 control register Note: Propagation of control input/output requires the correct I/O port setting. For details, see the section on I/O ports. Figure 2.6.1 Timer/Counter 2 (TC2) 88CS38B-69 2007-09-12 TMP88CS38B/CM38B/CP38B 2.6.2 Control The timer/counter 2 is controlled by a timer/counter 2 control register (TC2CR) and a 16-bit timer register 2 (TC2DR). Reset does not affect TC2DR. TC2DR (00016H, 00017H) 15 14 13 12 11 10 9 8 7 6 TC2DRH (00017H) 5 4 3 2 1 0 TC2DRL (00016H) Read/Write 7 TC2CR (00015H) TC2M 6 5 TC2S TC2 operating mode select 4 3 2 1 0 TC2CK TC2M (Initial value: **00 00*0) 0: Timer/event counter mode 1: Window mode NORMAL1/2, IDLE1/2 mode DV1CK = 0 fc/2 fc/2 24 fc/2 13 fc/2 14 fc/2 8 fc/2 9 011 fc/2 3 fc/2 4 100 Reserved 101 Reserved 000 001 TC2CK TC2S TC2 source clock select [Hz] TC2 start control DV1CK = 1 23 010 Write only Reserved Reserved 110 Reserved 111 External clock (TC2 pin input) 0: Stop and counter clear 1: Start Note 1: fc: High-frequency clock [Hz], *: Don’t care. Note 2: Writing to the lower byte of timer register 2 (TC2DRL), the comparison is inhibited until the upper byte (TC2DRH) is written. After writing to the upper byte, any match during 1 machine cycle (Instruction execution cycle) is ignored. Note 3: Set the mode and source clock when the TC2 stops (TC2S = 0). Note 4: Values to be loaded to timer registers must satisfy the following condition. Note 5: TC2CR are write-only registers and must not be used with any of the read-modify-write instructions. TC2DR > 1 Note 6: When STOP mode is started, timer counter is stopped and cleared. Set TC2S to “1” after STOP mode is released for restarting timer counter. Figure 2.6.2 Timer Registers 2 and TC2 Control Register 88CS38B-70 2007-09-12 TMP88CS38B/CM38B/CP38B 2.6.3 Function The timer/counter 2 has three operating modes: Timer, event counter and window modes. (1) Timer mode In this mode, the internal clock is used for counting up. The contents of TC2DR are compared with the contents of up counter. If a match is found, a timer/counter 2 interrupt (INTTC2) is generated, and the counter is cleared. Counting up is resumed after the counter is cleared. Table 2.6.1 Source Clock (Internal clock) for Timer/Counter 2 (at fc = 16.0 MHz) NORMAL, IDLE mode DV1CK = 0 TC2CK DV1CK = 1 Maximum Time Setting Resolution Maximum Time Setting Resolution 000 524.3 [ms] 9.54 [h] 001 512.0 [μs] 33.6 [s] 1.02 [ms] 1.05 [s] 19.1 [h] 010 16.0 [μs] 1.05 [s] 32.0 [μs] 011 0.5 [μs] 32.8 [ms] 1.0 [μs] 100 Reserved Reserved Reserved Reserved 101 Reserved Reserved Reserved Reserved 1.12 [min] 2.09 [s] 65.5 [ms] Example: Sets the source clock fc/2 [Hz] and generates an interrupt event 25 ms (at fc = 16 MHz, DV1CK = 1) 4 LDW (TC2DR), 61A8H ; Sets TC2DR (25 ms ÷ 2 /fc = 61A8H) (EIRH).6 ; Enable INTTC2 interrupt 4 DI SET EI LD (TC2CR), 00001100B ; Selects TC2 source clock LD (TC2CR), 00101100B ; Starts TC2 Count start Source clock Up counter 0 1 2 3 n−1 n 0 4 Match detect Timer trigger 1 2 3 Counter clear n INTTC2 interrupt Figure 2.6.3 Timer Mode Timing Chart 88CS38B-71 2007-09-12 TMP88CS38B/CM38B/CP38B (2) Event counter mode In this mode, events are counted on the rising edge of the TC2 pin input. The contents of TC2DR are compared with the contents of the up counter. If a match is found, an INTTC2 interrupt is generated, and the counter is cleared. The minimum pulse width to the TC2 pin is shown in Table 2.6.2. Two or more machine cycles are required for both the “H” and “L” levels of the pulse width. Match detect is executed on the falling edge of the TC2 pin. A match can not be detected and INTTC2 is not generated when the pulse is still in a falling state. Example: Sets the event counter mode and generates an INTTC2 interrupt 640 counts later. LDW (TC2DR), 640 ; Sets TC2DR (EIRH). 6 ; Enables INTTC2 interrupt DI SET EI LD (TC2CR), 00011100B ; Selects TC2 source clock LD (TC2CR), 00111100B ; Starts TC2 Table 2.6.2 Timer/Counter 2 External Clock Source Minimum Pulse Width [S] NORMAL, IDLE Mode 3 “H” Width 2 /fc “L” Width 2 /fc 3 Count start TC2 pin input Up counter 0 1 2 n−1 3 Match detect Timer register n 0 1 2 3 Counter clear n INTTC2 interrupt Figure 2.6.4 Event Counter Mode Timing Chart 88CS38B-72 2007-09-12 TMP88CS38B/CM38B/CP38B (3) Window mode In this mode, counting up performed on the rising edge of an internal clock during TC2 external pin input (window pulse) is “H” level. The contents of TC2DR are compared with the contents of up counter. If a match found, an INTTC2 interrupt is generated, and the up counter is cleared. The maximum applied frequency (TC2 input) must be considerably slower than the selected internal clock. Example: Generates an interrupt, inputting “H” level pulse width of 120 ms or more. (at fc = 16.0 MHz, DV1CK = 1) LDW (TC2DR), 0075H ; Sets TC2DR (120 ms ÷ 2 /fc = 0075H) (EIRH). 6 ; Enables INTTC2 interrupt 14 DI SET EI LD (TC2CR), 00000101B ; Selects TC2 source clock LD (TC2CR), 00100101B ; Starts TC2 TC2 pin input Internal clock Up counter TC2DR 0 1 2 n−3 n−2 n−1 n 0 1 2 3 n Match detect Counter clear INTTC2 interrupt Figure 2.6.5 Window Mode Timing Chart 88CS38B-73 2007-09-12 TMP88CS38B/CM38B/CP38B 2.7 8-Bit Timer/Counter 3 (TC3B) 2.7.1 Configuration Rising Edge detector TC3S INTTC3 interrupt Falling TC3ES Clear TC3 pin A Y B H A B C D Y E F G S fc/213 or fc/214 fc/212 or fc/213 fc/211 or fc/212 fc/210 or fc/211 fc/29 or fc/210 fc/28 or fc/29 fc/27 or fc/28 Source clock 8-bit up counter Overfolw TC3S Comparator Match detect A Y B 3 Capture Capture TC3DRB S TC3DRA TC3CK ACAP 8-bit timer register TC3S TC3M TC3CR TC3 control register Note: Propagation of control input/output requires the correct I/O port setting. For details, see the section on I/O ports. Figure 2.7.1 Timer/Counter 3 (TC3) 88CS38B-74 2007-09-12 TMP88CS38B/CM38B/CP38B 2.7.2 Control The timer/counter 3 is controlled by a timer/counter 3 control register (TC3CR) and two 8-bit timer registers (TC3DRA and TC3DRB) and port multiplex control register (PMPXCR). TC3DRA (0018H) TC3DRB (0019H) 7 TC3CR (001AH) 7 6 5 4 3 2 1 0 Read/Write (Initial value: 1111 1111) Read only (Initial value: 1111 1111) 6 5 ACAP 4 3 TC3S 2 1 0 TC3K TC3M TC3 operating mode select TC3M (Initial value: *0*0 0000) 0: Timer/event counter 1: Capture NORMAL, IDLE mode DV1CK = 0 000 001 TC3CK 010 TC3 source clock select [Hz] 011 ACAP Auto-capture control fc/2 14 fc/2 12 fc/2 13 fc/2 11 fc/2 12 fc/2 10 fc/2 11 fc/2 10 100 fc/2 101 fc/2 8 fc/2 9 fc/2 7 fc/2 8 111 TC3S 13 9 110 TC3 start control DV1CK = 1 fc/2 Write only External clock (TC3 pin input) 0: Stop and clear 1: Start 0: − 1: Auto-capture enable Note 1: fc: High-frequency clock [Hz], *: Don’t care. Note 2: Set the mode and source clock when the TC3 stops (TC3S = 0). Note 3: Values to be loaded to timer register 3A must satisfy the following condition TC3DRA > 0 (in the timer and event counter mode). Note 4: Auto-capture can be used only int the timer and event counter mode. Note 5: Before setting TC3DRA or switching the operation mode, stop the TC3 (TC3S = 0). Note 6: When STOP mode is started, timer counter is stopped and TC3 start control (TC3S) is cleared to “0” automatically. Set TC3S to “1” after STOP mode is released for restarting timer counter. Note 7: TC3CR, TCESCR is a write-only register and must not be used with any of the read-modify-write instructions. PMPXCR (0027H) 7 6 “0” CHS TC3ES Note 8 5 4 3 2 1 0 TC4ES TC3ES (Initial value: 00** **00) 0: Normal 1: Invert TC3 input control Write only Always write “0” to bit7 in PMPXCR. Figure 2.7.2 Timer Registers 3 and TC3 Control Register 88CS38B-75 2007-09-12 TMP88CS38B/CM38B/CP38B 2.7.3 Function The timer/counter 3 has three operating modes: timer, event counter, and capture mode. When it is used in the capture mode, the noise rejection time of TC3 pin input can be set by remote control receive control register. (1) Timer mode In this mode, the internal clock is used for counting up. The contents of TC3DRA are compared with the contents of up counter. If a match is found, a timer/counter 3 interrupt (INTTC3) is generated, and the up counter is cleared. The current contents of up counter are loaded into TC3DRB by setting ACAP (Bit6 in TC3CR) to “1” (Auto-capture function). The contents of up counter can be easily confirmed by executing the read instruction (RD instruction) of TC3DRB. Loading the contents of up counter is not synchronized with counting up. The contents of over flow (FFH) and 00H can not be loaded correctly. It is necessary to consider the count cycle. Clock Counter FE FF FE TC3DRB 00 01 FF/00 01 Table 2.7.1 Source Clock (Internal clock) for Timer/Counter 3 (Example: at fc = 16.0 MHz) NORMAL, IDLE Mode DV1CK = 0 TC3CK Resolution [μs] DV1CK = 1 Maximum Setting Time [ms] Resolution [μs] Maximum Setting Time [ms] 000 512 130.6 1024 261.1 001 256 65.3 512 130.6 010 128 32.6 256 65.3 011 64 16.3 128 32.6 100 32 8.2 64 16.3 101 16 4.1 32 8.2 110 8 2.0 16 4.1 88CS38B-76 2007-09-12 TMP88CS38B/CM38B/CP38B Count start Source clock Up counter 0 Timer register B ? 1 2 3 n−1 n 0 4 1 2 3 4 5 6 7 n Match detect Counter clear INTTC3 interrupt (a) Timer mode Source clock m−2 Up counter Timer register B m−1 m−1 ? m m+1 Capture m n−1 m+2 m+1 m+2 n n−1 n+1 Capture n n+1 ACAP1 (b) Auto capture Figure 2.7.3 Timer Mode Timing Chart (2) Event counter mode In this mode, the TC3 pin input pulses are used for counting up Either the rising on falling edge can be selected with TC3ES (Bit0 in PMPXCR). The contents of TC3DRA are compared with the contents of the up counter. If a match is found, an INTTC3 interrupt is generated and the counter is cleared. Match detect is executed on the falling edge of the TC3 pin. A match can not be detected, and INTTC3 is not generated when the pulse is still in a falling state. The maximum applied frequency is shown in Table 2.7.2. Two or more machine cycles are required for both the high and low levels of the pulse width. The current contents of up counter are loaded into TC3DRB by setting ACAP (Bit6 in TC3CR) to “1” (Auto-capture function). The contents of up counter can be easily confirmed by executing the read instruction (RD instruction) of TC3DRB. Loading the contents of up counter is not synchronized with counting up. The contents of over flow (FFH) and 00H can not be loaded correctly. It is necessary to consider the count cycle. Example: Generates an interrupt every 0.5 s, inputting 50 Hz pulses to the TC3 pin. LD (TC3CR), 00001110B ; LD (TC3DRA), 19H ; Sets TC3 mode and source clock 0.5 s ÷ 1/50 = 25 = 19H LD (TC3CR), 00011100B ; Starts TC3 Table 2.7.2 Source Clock (External clock) for Timer/Counter Minimum Applied Frequency [Hz] NORMAL, IDLE Mode 2 “H” Width 2 /fc “L” Width 2 /fc 2 88CS38B-77 2007-09-12 TMP88CS38B/CM38B/CP38B Count start TC3 pin input Up counter 0 1 2 n−1 3 n Match detect Timer register 0 1 2 3 Counter clear n INTTC3 interrupt Figure 2.7.4 Event Counter Mode Timing Chart (3) Capture mode In this mode, the pulse width, period and duty of the TC3 pin input are measured in this mode, which can be used in decoding the remote control signals or distinguishing AC 50/60 Hz, etc. The TC3 pin input can have its polarity changed between normal and inverse by using the TC3ES Register. a. If TC3ES = “0” (Non-inverting input) Once command operation has started, the counter free-runs on an internal source clock. When the falling edge of the TC3 pin input is detected, the counter value is loaded into TC3DRB. When the rising edge is detected, the counter value is loaded into TC3DRA, and the counter is cleared, generating an INTTC3 interrupt. If the rising edge is detected right after command operation has started, no capture to TC3DRB and an INTTC3 interrupt occurs only on capture to TC3DRA. If a read instruction is executed for TC3DRB, the value that exists at the end of the previous capture (immediately after a reset, “FF”) is read. b. If TC3ES = “1” (Inverse input) Once command operation has started, the counter free-runs on an internal clock. When the rising edge of the TC3 pin input is detected, the counter value is loaded into TC3DRB. When the falling edge is detected, the counter value is loaded into TC3DRA, and the counter is cleared, generating an INTTC3 interrupt. If the falling edge is detected right after command operation has started, the counter value is not captured into TC3DRB and an INTTC3 interrupt occurs only on capture to TC3DRA. If a read instruction is executed for TC3DRB, the value that exists at end of the previous capture (immediately after a reset, “FF”) is read. The minimum acceptable input pulse width is equal to the length of one source clock period selected by TC3CR <TC3CK>. Table 2.7.3 TC3INV-based Capture Input Edges TC3ES Capture into TC3DRB “0” (Non-inverting input) Falling edge Rising edge “1” (Inverting input) Rising edge Falling edge Note: Capture into TC3DRA INTTC3 Interrupt Capture of the TC3 pin input requires at least 1 cycle of the selected source clock. 88CS38B-78 2007-09-12 88CS38B-79 Reading TC3DRA INTTC3 interrupt TC3DRB TC3DRA Internal waveform (Invert) TC3 pin input Up counter Source clock TC3S Reading TC3DRA INTTC3 interrupt TC3DRB TC3DRA Internal waveform (Normal) TC3 pin input Up counter Source clock TC3S 1 0 1 Command start 0 Command start i−1 i−1 i i 0 1 Capture i i i+1 k−1 k k−1 k 0 m−1 m m m+1 n−1 n 0 m−1 m 0 Capture m 1 Capture n 1 2 Capture n−2 n−3 n−2 n−1 n 0 b) In case of TC3ES = “1” (Invert) k k+1 a) In case of TC3ES = “0” (Normal) Capture k 1 n 1 3 1 3 2 3 2 FF (Overflow) 2 Overflow 1 When TC3DRA is not read, capture and overflow detection are stopped. FE FF FF FE FF TMP88CS38B/CM38B/CP38B Figure 2.7.5 Capture Mode Timing Chart 2007-09-12 TMP88CS38B/CM38B/CP38B The edge of TC3 pin input is detected in the remote control receive circuit with noise rejection. The remote control receive circuit is controlled by the remote control receive control register (RCCR). The remote control receive status register (RCSR) can monitor the polarity selection and noise rejection circuit. Rising Polarity select TC3IN 8 Noise reject circuit (5-bit up-down counter) MPX A 9 fc/2 or fc/2 Capture control Edge detector Falling B Y S TC3 Source clock RCSCK RPOLS 5 RCNC RCNF RCOVF RNCM RCCR/RCSR MPX: Multiplexer Remote control receive control/status register Figure 2.7.6 Remote Control Receiving Circuit 88CS38B-80 2007-09-12 TMP88CS38B/CM38B/CP38B RCCR (00026H) RCEN RPOLS RCSCK RCNC Noise reject time select RCNC (Initial value: 0001 1111) Write only (Source clock) × (RCNC − 1) [s] 02H ≤ RCNC ≤ 1FH NORMAL, IDLE mode DV1CK = 0 Noise reject cricuit RCSCK Source clock select 0 RPOLS RCEN RCSR (00026H) 0: Positive select 1: Negative Remote control receive circuit 0: Disable operation control 1: Enable R/W 9 2 /fc 2 /fc 1 Remote control signal polarity DV1CK = 1 8 TC3CK (Note 2) Write only Note 1: Set RPOLS and RCSCK when the timer/counter stops (TC3S = 0). Note 2: Source clock of timer/counter 3. Note 3: fc: High-frequency clock [Hz], *: Don’t care Note 4: RCCR includes a write-only register and must not be used with any of read-modify-write instructions. Note5: Values to be loaded to RCNC must satisfy the following condition (02 ≤ RCNC ≤ 1F). RCNF RPOLS RCSCK RCOVF RNCM RCOVF RNCM (Initial value: 0000 0***) Remote control signal monitor 0: Low level after noise rejector 1: High level Noise reject circuit overflow 0: Signal and definition by overwriting the noise reject time RCNC flag Read only 1: Overflow NORMAL, IDLE mode DV1CK = 0 Noise reject circuit RCSCK Source clock Select 0 1 RPOLS RCNF Remote control signal polarity 0: Positive select 1: Negative Remote control signal monitor 0: Without noise after noise rejctor 1: With noise Note 1: Reading out the register RCSR resets RCNF and RCOVF. Note 2: Source clock of timer/counter 3 DV1CK = 1 8 9 2 /fc 2 /fc TC3CK R/W (Note 2) Read only Note 3: When a 5-bit up-down counter counts down to “0” after counting up, the RCNF defines to be noise. Note 4: fc: High-frequency clock [Hz], *: Don’t care. Figure 2.7.7 Remote Control Rceive Control Register and Remote Control Receive Status Register Table 2.7.4 Combination between The Polarity and The Edge Selection RPOLS TC3 Pin Input Pulse (Interrupt occurrence is shown as allow.) Measurement 0 1 Note: When TC3CK is used in RCSCK, do not select an external clock to the TC3CK. 88CS38B-81 2007-09-12 88CS38B-82 RCNC Writing RCCR Reading RCSR RCNF RCOVF RNCM TC3 pin input Up-down counter Source clock Reading RCSR RCNF RCOVF RNCM TC3 pin input Up-down counter Source clock 0 0 08H 1 1 3 3 4 1 2 1 0 1 2 1 (a) Noise (RPOLS = 0, RCNC = 03H) 0 (b) Noise rejection circuit overflow flag (RPOLS = 1, RCNC = 08H to 03H) 2 2 03H 0 0 Reset Reset TMP88CS38B/CM38B/CP38B Figure 2.7.8 Remote Control Receive Circuit Timing Chart 2007-09-12 TMP88CS38B/CM38B/CP38B 2.8 8-Bit Timer/Counter 4 (TC4) 2.8.1 Configuration TC4S 11 10 fc/2 or fc/2 7 8 fc/2 or fc/2 5 6 fc/2 or fc/2 3 4 fc/2 or fc/2 A B C D AS Y B Clear 8-bit up counter Y TC4ES TC4 pin Source clock Overfolw detect Comparator H Match detect S 3 TC4S TC4CK TC4M 2 Note: TC4CR TC4DR Timer/counter 4 control register 8-bit timer register 4 INTTC4 interrupt request signal Set the input/output control correctly for the substitutive input/output pins. For details, see the description of the input/output port control register. Figure 2.8.1 Timer/Counter 4 (TC4) 88CS38B-83 2007-09-12 TMP88CS38B/CM38B/CP38B 2.8.2 Control The timer/counter 4 is controlled by a timer/counter 4 control register (TC4CR) and an 8-bit timer register 4 (TC4DR). Reset does not affect TC4DR. TC4DR (0001BH) 7 6 5 4 3 2 1 0 TC4CR (0001CH) 7 6 5 4 3 2 1 0 Write only (Initial value: 1111 1111) TC4S TC4CK TC4M Write only (Initial value: **00 0000) 00: Timer/event counter mode TC4S 01: Reserved TC4 start control 10: Reserved 11: Reserved NORMAL, IDLE mode DV1CK = 0 TC4CK TC4 source clock select [Hz] (Note 4) DV1CK = 1 11 fs/2 fc/2 7 fs/2 8 010 fc/2 5 fs/2 6 011 fc/2 3 fs/2 4 100 Reserved Reserved 101 Reserved Reserved 110 Reserved Reserved 000 fc/2 001 111 12 R/W External clock (TC4 pin input) 00: Timer/event counter mode TC4M 01: Reserved TC4 operating mode select 10: Reserved 11: Reserved Note 1: fc: High-frequency clock [Hz], *: Don’t care Note 2: Values to be loaded to the timer register must satisfy the following condition (1 ≤ TC4DR ≤ 255). Note 3: When the TC4 is started (TC4S = 0 → 1) or disabled (TC4S = 1 → 0) or while the TC4 is operating (TC4S = 1 → 1), do not write to TC4M and TC4CK in TC4CR. If these registers are selected/changed during these operations, counting up is not performed properly. Note 4: When STOP mode is started, timer counter is stopped and cleared. Set TC4S to “1” after STOP mode is released for restarting timer counter. PMPXCR (00027H) TC4ES Note 5: Undefined values are read from bits 6 and 7 of TC4CR. Note 6: Do not change TC4DR while the TC4 is operating. 7 6 “0” CHS 5 3 2 1 0 TC4ES (TC3ES) (Initial value: 00** **00) TC4 edge select Note 1: 4 0: Rising edge 1: Falling edge Write only TC4CR, TC4DR and PMPXCR are write only register and must not be used with any of the read-modify-write instructions such as SET, CLR, etc. Figure 2.8.2 Timer Register 4 and TC4 Control Register 88CS38B-84 2007-09-12 TMP88CS38B/CM38B/CP38B 2.8.3 Function The timer/counter 4 has two operating modes: timer, event counter mode. (1) Timer mode In this mode, the internal clock is used for counting up. The contents of TC4DR are compared with the contents of up counter. If a match is found, an INTTC4 interrupt is generated and the up counter is cleared to “0”. Counting up resumes after the up counter is cleared. Table 2.8.1 Source Clock (Internal clock) for Timer/Counter 4 (Example: at fc = 16.0 MHz) NORMAL, IDLE mode DV1CK = 0 TC4CK Resolution [μs] DV1CK = 1 Maximum Setting Time [ms] Resolution [μs] Maximum Setting Time [ms] 000 128.0 32.6 256.0 65.3 001 8.0 2.0 16.0 4.1 010 2.0 0.510 4.0 1.0 100 0.5 0.128 1.0 0.255 (2) Event counter mode In this mode, the TC4 pin input (External clock) pulse is used for counting up. Either the rising or falling edge can be selected with TC4ES (Bit1 PMPXCR). The contents of TC4DR are compared with the contents of the up counter. If a match is found, an INTTC4 interrupt is generated and the counter is cleared. The maximum applied frequency is shown Table 2.8.2. Two or more machine cycles are required for both the high and low level of the pulse width. Note: The event counter mode can only be used in NORMAL or IDLE mode. Table 2.8.2 Timer/Counter 4 External Clock Source Minimum Input Pulse Width [s] NORMAL1, IDLE1 Mode 3 “H” Width 2 /fc “L” Width 2 /fc 3 88CS38B-85 2007-09-12 TMP88CS38B/CM38B/CP38B 2.9 Serial Bus Interface (SBI-ver. D) The TMP88CS38B and TMP88CM38B/CP38B has a 1-channel serial bus interface which employs a clocked-synchronous 8-bit serial bus interface and an I2C bus (a bus system by Philips). The serial bus interface pins are selectively used as either channel 0 or channel 1. The serial interface is connected to external devices through P35 (SDA0)/P52 (SDA1) and P34 (SCL0)/P51 (SCL1) in the I2C bus mode; and through P53 ( SCK1 ), P52 (SO1) and P51 (SI1) in the clocked-synchronous 8-bit SIO mode. The serial bus interface pins are also used for the P3/P5 port. When used for serial bus interface pins, set the P3/P5 output latches of these pins to “1”. When not used as serial bus interface pins, the P3/P5 port is used as a normal I/O port. Note 1: When P3 and P5 is used as serial bus interface pins, P35, P34, P51 and P50 should be set as a sink open-drain output by clearing PSELCR to “0”. Note 2: The I2C of TMP88CS38B and TMP88CM38B/CP38B can be used only in the standard mode of I2C. The fast mode and the high speed mode can not be used. 2.9.1 Configuration INTSBI interrupt request SCL SCK SIO clock control fc/2 fc/4 Source clock generator P52 Input/ (SDA1/SO1) output P51 control (SCL1/SI1) Divider I C bus clock sync. + Control SBI control register B/ SBI status register I2CAR 2 I C bus address register SBIDBR SBI data buffer register SI 2 I C bus data control Shift register SBICRB/ SBISR SO SIO data control Transfer control circuit 2 Noise canceller P53 ( SCK ) Noise canceller SDA P35 (SDA0) P34 (SCL0) SBICRA SBI control register A Figure 2.9.1 Serial Bus Interface (SBI) 88CS38B-86 2007-09-12 TMP88CS38B/CM38B/CP38B 2.9.2 Control The following registers are used for control the serial bus interface and monitor the operation status. • Serial bus interface control register A (SBICRA) • Serial bus interface control register B (SBICRB) • Serial bus interface data buffer register (SBIDBR) • I2C bus address register (I2CAR) • Serial bus interface status register A (SBISRA) • Serial bus interface status register B (SBISRB) • Serial clock source control register (SCCRB) • Serial clock control status register (SCSR) The above registers differ depending on a mode to be used. Refer to section 2.9.7 “I2C Bus Mode Control” and 2.9.9 “Clocked-synchronous 8-Bit SIO Mode Control”. 2.9.3 Serial Clock Source Control A serial bus interface circuit can reduce the power consumption by stopping a serial clock generater. Serial Clock Source Control Register SCCRB (00FF1H) 7 6 5 4 3 2 1 0 SCEN (Initial value: 0*** ****) SCEN Note: Serial clock source control 0: Do not generate source clock Write only 1: Generate source clock When SCRQ and SCEN are “1”, SCEN cannot be cleared to “0”. When SCRQ is “0”, SCEN is cleared to “0”. Serial Clock Control Status Register SCSR (00FF1H) 7 6 5 4 3 2 1 0 SCRQ SCRQ (Initial value: 0*** ****) Serial clock source request 0: No source clock request from serial bus interface 1: Source clock request from serial bus interface Read only SCRQ SCEN Source clock Clock generation “1” → SCEN Write data except “00” to SBIM Write data “0” → SCEN “00” to SBIM Figure 2.9.2 Serial Clock Source 88CS38B-87 2007-09-12 TMP88CS38B/CM38B/CP38B 2.9.4 Channel Select A serial bus interface circuit can select I/O pin when a serial bus interface is used for I2C bus mode. Port Switching Register PMPXCR (00027H) 7 6 5 “0” CHS 4 3 1 0 (TC4ES) (TC3ES) (Initial value: 00** **00) 0: Channel 0 2 I C bus Channel Select CHS 2 R/W 1: Channel 1 Note 1: When SIO mode, don’t use channel 0. Therefore, set to “1” in PMPXCR at SIO mode. Note 2: Always write “0” to bit7 in PMPXCR. Note 3: *: Don’t care Figure 2.9.3 Channel Select 2.9.5 Software Reset A serial bus interface circuit has a software reset function, when a serial bus interface circuit is locked by an external noise, etc. To occur software reset, write “01”, “10” into the SWRST (Bit1, 0 in SBICRB). During software reset, the SWRMON (Bit0 in SBISRA) is clear to “0”. 2.9.6 The Data Format in The I2C bus Mode The data format when using the TMP88CS38B and TMP88CM38B/CP38B in the I2C bus mode are shown in as below. (a) Addressing format 8 bits S Slave address 1 R A / C W K 1 to 8 bits Data 1 A C K 1 to 8 bits Data 1 A C P K 1 or more 1 (b) Addressing format (with restart) 8 bits S Slave address 1 R A / C W K 1 to 8 bits Data 1 A C S K 8 bits Slave address 1 or more 1 1 R A / C W K 1 to 8 bits Data 1 A C P K 1 or more 1 (c) Free data format 8 bits S Data 1 S: 1 A C K 1 to 8 bits Data 1 A C K 1 to 8 bits Data 1 A C P K 1 or more Start condition R/ W : Direction bit ACK: Acknowledge bit P: Stop condition Figure 2.9.4 Data Format in I2C Bus Mode 88CS38B-88 2007-09-12 TMP88CS38B/CM38B/CP38B 2.9.7 I2C Bus Mode Control The following registers are used to control the serial bus interface (SBI) and monitor the operation status in the I2C bus mode. Serial Bus Interface Control Register A SBICRA (00020H) 7 6 5 BC 4 3 2 ACK Acknowledgement mode specification (Initial value: 0000 *000) ACK = 0 Number of Bits clock 8 8 1 1 2 2 3 3 4 4 5 5 6 6 7 7 Master mode ACK = 1 Number of Bits clock 9 8 2 1 3 2 4 3 5 4 6 5 7 6 8 7 Slave mode 0 Not generate a clock pulse for an acknowledgement. Not count a clock pulse for an acknowledgement. 1 Generate a clock pulse for an acknowledgement. Count a clock pulse for an acknowledgement. 000 001 010 011 100 101 110 111 ACK Number of transferred bits ACK 0 SCK BC BC 1 Write only R/W DV1CK = 0 DV1CK = 1 000: Reserved (Note 3) 000: Reserved (Note 3) 001: Reserved (Note 3) 001: Reserved (Note 3) 010: Reserved (Note 3) 010: 58.8 kHz Serial clock selection Write SCK 011: 60.6 kHz 011: 30.3 kHz (At fc = 16 MHz, output on SCL pin) only 100: 30.7 kHz 100: 15.4 kHz 101: 15.5 kHz 101: 7.7 kHz 110: 7.8 kHz 110: 3.9 kHz 111 : Reserved 111 : Reserved Note 1: Set the BC to “000” before switching to 8-bit SIO bus mode. Note 2: SBICRA cannot be used with any of read-modify-write instructions such as bit manipulation, etc. Note 3: This I2C bus circuit does not support the Fast mode. It supports the Standard mode only. Although the I2C bus circuit itself allows the setting of a baud rate over 100 kbps, the compliance with the I2C specification is not guaranteed in that case. Serial Bus Interface Data Buffer Register SBIDBR (00021H) 7 6 5 4 3 2 1 0 (Initial value: **** ****) Note 1: Note 2: R/W For writing transmitted data, start from the MSB (Bit7). The data which was written into SBIDBR cannot be read, since a write data buffer and a read buffer are independent in SBIDBR. Therefore, SBIDBR cannot be used with any of read-modify-write instructions such as bit manipulation, etc. 2 I C bus Address Register I2CAR (00022H) 7 6 SA6 SA5 5 4 3 Slave address SA4 SA3 SA2 2 1 SA1 SA0 0 ALS (Initial value: 0000 0000) SA Slave address selection Write Address recognition mode 0: Slave address recognition only ALS specification 1: Non slave address recognition Note 1: I2CAR is write-only register and cannot be used with any of read-modify-write instruction such as bit manipulation, etc. Note 2: Do not set I2CAR to “00H” to avoid the incorrect response of acknowledgment in slave mode. If “00H” is set to I2CAR as the slave address and received “01H” in slave mode, the device might transmit the acknowledgement incorrectly. Figure 2.9.5 Serial Bus Interface Control Register A, Serial Bus Interface Data Buffer Register and I2C Bus Address Register In The I2C Bus Mode 88CS38B-89 2007-09-12 TMP88CS38B/CM38B/CP38B Serial Bus Interface Control Register B SBICRB (00023H) 7 6 5 4 MST TRX BB PIN MST TRX BB PIN SBIM 3 2 SBIM 0: 1: 0: Transmitter/receiver selection 1: 0: Start/stop generation 1: 0: Cancel interrupt service request 1: 00: 01: Serial bus interface operating mode selection 10: 11: Master/slave selection SWRST1 Software reset start bit SWRST0 1 0 SWRST1SWRST0 (Initial value: 0001 0000) Slave Master Receiver Transmitter Generate a stop condition when MST, TRX and PIN are “1”. Generate a start condition when MST, TRX and PIN are “1”. Write − only Cancel interrupt service request Port mode (Serial bus interface output disable) Clocked synchronous 8-bit SIO mode 2 I C bus mode Reserved Software reset starts by first writing “10” and next writing “01”. Note 1: Switch a mode to port after confirming that the bus is free. Note 2: Switch a mode to I C bus mode or clock synchronous 8-bit SIO mode after confirming that the port is high 2 level. Note 3: SBICRB has write-only register and must not be used with any of read-modify-write instructions such as bit manipulation, etc. Note 4: When the SWRST (Bit1, 0 in SBICRB) is written to “01”, “10”, software reset (Four machine cycles) is occurred. This time, control the serial bus interface and monitor the operation status registers except the SBIM (Bit3, 2 in SBICRB) and the CHS (Bit6 in PMPXCR) are reseted. Control the serial bus interface and monitor the operation status registers are SBICRA, SBICRB, SBIDBR, I2CAR, SBISRA, SBISRB, SCCRA and SCSR. Serial Bus Interface Status Register A 7 6 5 SBISRA (00020H) 4 3 2 1 0 SWR MON ACK (Initial value: **** ***1) 0: During software reset SWRMON Software reset monitor Read only 1: − (Initial) *: Don’t care Serial Bus Interface Status Register B SBISRB (00023H) 7 6 5 4 3 2 1 0 MST TRX BB PIN AL AAS AD0 LRB MST Master/Slave selection status monitor 0: Slave TRX Transmitter/Receiver selection status monitor 0: Receiver BB Bus status monitor PIN Interrupt service requests status monitor (Initial value: 0001 0000) 1: Master 1: Transmitter 0: Bus free 1: Bus busy 0: Requesting interrupt service 1: Releasing interrupt service request Arbitration lost detection monitor 0: − AAS Slave address match detection monitor 0: Not detect slave address match or “GENERAL CALL” AD0 “GENERAL CALL” detection monitor 0: Not detect “GENERAL CALL” LRB Last Received bit monitor AL Read only 1: Arbitration lost detected 1: Detect slave address match or “GENERAL CALL” 1: Detect “GENERAL CALL” 0: Last receive bit is “0” 1: Last receive bit is “1” Figure 2.9.6 Serial Bus Interface Control Register B and Serial Bus Interface Status Register A/B in the I2C Bus Mode 88CS38B-90 2007-09-12 TMP88CS38B/CM38B/CP38B (1) Acknowledgement mode specification a. Acknowledgement mode (ACK = “1”) To set the device as an acknowledgement mode, the ACK (bit4 in SBICRA) should be set to “1”. When a serial bus interface circuit is a master mode, an additional clock pulse is generated for an acknowledge signal. In a slave mode, a clock is counted for the acknowledge signal. In the master transmitter mode, the SDA pin is released in order to receive an acknowledge signal from the receiver during additional clock pulse cycle. In the master receiver mode, the SDA pin is set to low level generation an acknowledge signal during additional clock pulse cycle. In a slave mode, when a received slave address matches to a slave address which is set to the I2CAR or when a “GENERAL CALL” is received, the SDA pin is set to low level generating an acknowledge signal. After the matching of slave address or the detection of “GENERAL CALL”, in the transmitter the SDA pin is released in order to receive an acknowledge signal from the receiver during additional clock pulse cycle. In a receiver, the SDA pin is set to low level generation an acknowledge signal during additional clock pulse cycle after the matching of slave address or the detection of “GENERAL CALL”. The Table 2.9.1 shows the SCL and SDA pins status in acknowledgement mode. Table 2.9.1 SCL and SDA Pins Status in Acknowledgement Mode Mode Pin Transmitter SCL Master An additional clock pulse is generated. Released in order to receive and acknowledge signal. SDA SCL Slave b. SDA Receiver Set to low level generating an acknowledge signal. A clock is counted for the acknowledge signal. When slave address matches or a general call is detected − Set to low level generating an acknowledge signal. After matching of slave address or general call Released in order to receive an acknowledge signal. Set to low level generating an acknowledge signal. Non-acknowledgement mode (ACK = “0”) To set the device as a non-acknowledgement mode, the ACK should be cleared to “0”. In the master mode, a clock pulse for an acknowledge signal is not generated. In the slave mode, a clock for a acknowledge signal is not counted. (2) Number of transfer bits The BC (bits 7 to 5 in SBICRA) is used to select a number of bits for next transmitting and receiving data. Since the BC is cleared to “000” as a start condition, a slave address and direction bit transmissions are always executed in 8 bits. Other than these, the BC retains a specified value. 88CS38B-91 2007-09-12 TMP88CS38B/CM38B/CP38B (3) Serial clock a. Clock source The SCK (bits 2 to 0 in SBICRA) is used to select a maximum transfer frequency output from the SCL pin in the master mode. Set a communication baud rate that meets the I2C bus specification, such as the shortest pulse width of tLOW, based on the equations shown below. Four or more machine cycles are required for both high and low levels of pulse width in the external clock which is input from SCL pin. Note: Since the I2C of TMP88CS38B AND TMP88CM38B/CP38B can not be used as the Fast mode and the High Speed mode, do not set SCK as the frequency that is over 100 kHz. This I2C bus circuit does not support high-speed mode, it supports standard mode only. Set the baud rates, which have been calculated according to the formula below, to meet the specifications of the I2C bus, such as the smallest pulse width of tLOW, tHIGH tLOW 1/fscl tLOW = 2 /fc n n SCK (Bits 2 to 0 in the SBICRA) DV1CK = 0 DV1CK = 1 000 001 010 011 100 101 110 4 5 6 7 8 9 10 5 6 7 8 9 10 11 tHIGH = 2 /fc + 8/fc n fscl = 1/(tLow + tHIGH) fc: High-frequency clock Figure 2.9.7 Clock Source b. Clock synchronization In the I2C bus mode, in order to drive a bus with a wired AND, a master device which pulls down a clock pulse to low will, in the first place, invalidate a clock pulse of another master device which generates a high-level clock pulse. The serial bus interface circuit has a clock synchronization function. This function ensures normal transfer even if there are two or more masters on the same bus. The example explains clock synchronization procedures when two masters simultaneously exist on a bus. SCL pin (Master 1) SCL pin (Master 2) wait Count start Count reset Count reset SCL (Bus) a b c Figure 2.9.8 Clock Synchronization As master 1 pulls down the SCL pin to the low level at point “a”, the SCL line of the bus becomes the low level. After detecting this situation, master 2 resets counting a clock pulse in the high level and sets the SCL pin to the low level. 88CS38B-92 2007-09-12 TMP88CS38B/CM38B/CP38B Master 1 finishes counting a clock pulse in the low level at point “b” and sets the SCL pin to the high level. Since master 2 holds the SCL line of the bus at the low level, master 1 waits for counting a clock pulse in the high level. After master 2 sets a clock pulse to the high level at point “c” and detects the SCL line of the bus at the high level, master 1 starts counting a clock pulse in the high level. Then, the master, which has finished the counting a clock pulse in the high level, pulls down the SCL pin to the low level. The clock pulse on the bus is deteminded by the master device with the shortest high-level period and the master device with the longest low-level period from among those master devices connected to the bus. (4) Slave address and address recognition mode specification When the serial bus interface circuit is used with an addressing format to recognize the slave address, clear the ALS (Bit0 in I2CAR) to “0”, and set the SA (Bits 7 to 1 in I2CAR) to the slave address. When the serial bus interfac circuit is used with a free data format not to recognize the slave address, set the ALS to “1”. With a free data format, the slave address and the direction bit are not recognized, and they are processed as data from immediately after start condition. (5) Master/slave selection To set a master device, the MST (Bit7 in SBICRB) should be set to “1”. To set a slave device, the MST should be cleared to “0”. When a stop condition on the bus or an arbitration lost is detected, the MST is cleared to “0” by the hardware. (6) Transmitter/receiver selection To set the device as a transmitter, the TRX (Bit6 in SBICRB) should be set to “1”. To set the device as a receiver, the TRX should be cleared to “0”. When data with an addressing format is transferred in the slave mode, the TRX is set to “1” by a hardware if the direction bit (R/ W ) sent from the master device is “1”, and is cleared to “0” by a hardware if the bit is “0. In the master mode, after an acknowledge signal is returned from the slave device, the TRX is cleared to “0” by a hardware if a transmitted direction bit is “1”, and is set to “1” by a hardware if it is “0”. When an acknowledge signal is not returned, the current condition is maintained. When a stop condition on the bus or an arbitration lost is detected, the TRX is cleared to “0” by the hardware. The following table show TRX changing conditions in each mode and TRX value after changing. Mode Direction Bit Conditions TRX after Changing Slave mode “0” A received slave address is the same value set to I2CAR “0” “1” Master mode “1” “0” ACK signal is returned “1” “1” “0” When a serial bus interface circuit operates in the free data format, a slave address and a direction bit are not recognized. They are handled as data just after generating a start condition. The TRX is not changed by a hardware. 88CS38B-93 2007-09-12 TMP88CS38B/CM38B/CP38B (7) Start/stop condition generation When the BB (Bit5 in SBICRB) is “0”, a slave address and a direction bit which are set to the SBIDBR are output on a bus after generating a start condition by writing “1” to the MST, TRX, BB and PIN. It is necessary to set transmitted data to the SBIDBR and set “1” to ACK beforehand. SCL pin 1 2 3 4 5 6 7 8 SDA pin A6 A5 A4 A3 A2 A1 A0 R/ W slave address and the direction bit Start condition 9 Acknowledge signal Figure 2.9.9 Start Condition Generation and Slave Address Generation When the BB is “1”, sequence of generating a stop condition is started by writeng “1” to the MST, TRX and PIN, and “0” to the BB. Do not modify the contents of MST, TRX, BB and PIN until a stop condition is generated on a bus. SCL pin SDA pin Stop condition Figure 2.9.10 Stop Condition Generation When a stop condition is generated and the SCL line on a bus is pulled down to low level by another device, a stop condition is generated after releasing the SCL line. The bus condition can be indicated by reading the contents of the BB (Bit5 in SBISRB). The BB is set to “1” when a start condition on a bus is detected and is cleared to “0” when a stop condition is detected. (8) Interrupt service request and cancel When a serial bus interface circuit is in the master mode and transferring a number of clocks set by the BC and the ACK is complete, a serial bus interface interrupt request (INTSBI) is generated. In the slave mode, the conditions of generating INTSBI are follows: • At the end of acknowledge signal when the received slave address matches to the value set by the I2CAR • At the end of acknowledge signal when a “GENERAL CALL” is received • At the end of transferring or receiving after matching of slave address or receiving of “GENRAL CALL” When a serial bus interface interrupt request occurs, the PIN (Bit4 in SBISR) is cleared to “0”. During the time that the PIN is “0”, the SCL pin is pulled down to low level. Either writing data to SBIDBR or reading data from the SBIDBR sets the PIN to “1”. The time from the PIN being set to “1” until the SCL pin is released takes tLOW. Although the PIN (Bit4 in SBICRB) can be set to “1” by the program, the PIN can not be cleared to “0” by the program. Note: If the arbitration lost occurs, when the slave address does not match, the PIN is not cleared to “0” even thought INTSBI is generated. 88CS38B-94 2007-09-12 TMP88CS38B/CM38B/CP38B (9) Serial bus interface operating mode selection The SBIM (Bit3 and 2 in SBICRB) is used to specify a serial bus interface operation mode. Set the SBIM to “10” in order to change a operation mode to I2C bus mode. Before changing operation mode, confirm serial bus interface pins in a high level. And switch a mode to port after confirming that a bus is free. (10) Arbitration lost detection monitor Since more than one master device can exist simultaneously on a bus in the I2C bus mode, a bus arbitration procedure is implemented in order to guarantee the contents of transferred data. Data on the SDA line is used for bus arbitration of the I2C bus. The following shows an example of a bus arbitration procedure when two master devices exist simultaneously on a bus. Master 1 and master 2 output the same data until point “a”. After master 1 outputs “1” and master 2, “0”, the SDA line of a bus is wired AND and the SDA line is pulled-down to the low level by master 2. When the SCL line of a bus is pulled up at point “b”, the slave device reads data on the SDA line, that is data in master 2. Data transmitted from master 1 becomes invalid. The state in master 1 is called “arbitration lost”. A master device which loses arbitration releases the SDA pin and the SCL pin in order not to effect data transmitted from other masters with arbitration. When more than one master sends the same data at the first word, arbitration occurs continuously after the second word. SCL (Bus) SDA pin (Master 1) SDA pin becomes “1” after losing arbitration. SDA pin (Master 2) SDA (Bus) a b Figure 2.9.11 Arbitration Lost 88CS38B-95 2007-09-12 TMP88CS38B/CM38B/CP38B The serial bus interface circuit compares levels of a SDA line of a bus with its those SDA pin at the rising edge of the SCL line. If the levels are unmatched, arbitration is lost and the AL (Bit3 in SBISRB) is set to “1”. When the AL is set to “1”, the MST and TRX are cleared to “0” and the mode is switched to a slave receiver mode. The AL is cleared to “0” by writing or reading data to or from the SBIDBR or writing data to the SBICRB. SCL pin 1 2 3 4 5 6 7 8 SDA pin D7A D6A D5A D4A D3A D2A D1A D0A SCL pin 1 2 3 4 5 6 7 8 9 1 2 3 D7A’ D6A’ D5A’ Master A Master B 9 Stop clock output SDA pin D7B D6B Releasing SDA pin and SCL pin to high level as losing arbitration. AL MST TRX Accessed to SBIDBR or SBICRB INTSBI Figure 2.9.12 Example of when a Serial Bus Interface Circuit is a Master B (11) Slave address match detection monitor In the slave mode, the AAS (Bit2 in SBISR) is set to “1” when the received data is “GENERAL CALL” or the received data matches the slave address setting by I2CAR with an address recognition mode (ALS = 0). When a serial bus interface circuit operates in the free data format (ALS = 1), the AAS is set to “1” after receiving the first 1-word of data. The AAS is cleared to “0” by writing data to the SBIDBR or reading data from the SBIDBR. (12) GENERAL CALL detection monitor The AD0 (Bit1 in SBISR) is set to “1” when all 8-bit received data is “0” immediately after a start condition in a slave mode. The AD0 is cleared to “0” when a start or stop condition is detected on a bus. (13) Last received bit monitor The SDA value stored at the rising edge of the SCL is set to the LRB (Bit0 in SBISRB). In the acknowledge mode, immediately after an INTSBI interrupt request is generated, an acknowledge signal is read by reading the contents of the LRB. 88CS38B-96 2007-09-12 TMP88CS38B/CM38B/CP38B 2.9.8 Data Transfer of I2C Bus (1) Device initialization For initialization of device, set the ACK in SBICRA to “1” and the BC to “000”. Specify the data length to 8 bits to count clocks for an acknowledge signal. Set a transfer frequency to the SCK in SBICRA. Next, set the slave address to the SA in I2CAR and clear the ALS to “0” to set an addressing format. After confirming that the serial bus interface pin is high level, for specifying the default setting to a slave receiver mode, clear “0” to the MST, TRX and BB in SBICRB, set “1” to the PIN, “10” to the SBIM, and “00” to bits SWRST1 and SWRST0. Note: The initialization of a serial bus interface circuit must be complete within the time from all devices which are connected to a bus have initialized to and device does not generate a start condition. If not, the data can not be received correctly because the other device starts transferring before an end of the initialization of a serial bus interface circuit. (2) Start condition and slave address generation Confirm a bus free status (when BB = 0). Set the ACK to “1” and specify a slave address and a direction bit to be transmitted to the SBIDBR. By writing “1” to the MST, TRX, BB and PIN, the start condition is generated on a bus and then, the slave address and the direction bit which are set to the SBIDBR are output. An INTSBI interrupt request occurs at the 9th falling edge of a SCL clock cycle, and the PIN is cleared to “0”. The SCL pin is pulled down to the low level while the PIN is “0”. When an interrupt request occurs the TRX changes by the hardware according to the direction bits only when an acknowledge signal is returned from the slave device. Note 1: Do not write a slave address to be output to the SBIDBR while data is transferred. If data is written to the SBIDBR, data to been outputting may be destroyed. Note 2: The bus free must be confirmed by software within 98.0 μs (the shortest transmitting time according to the I2C bus standard) after setting of the slave address to be output. Only when the bus free is confirmed, set “1” to the MST, TRX, BB, and PIN doesn’t finish within 98.0 μs, the other masters may start the transferring and the slave address data written in SBIDBR may be broken. SCL pin 1 2 3 4 5 6 7 8 SDA pin A6 A5 A4 A3 A2 A1 A0 R/ W Start condition Slave address + direction bit 9 Acknowledge signal from a slave device PIN INTSBI interrupt request Figure 2.9.13 Start Condition Generation and Slave Address Transfer 88CS38B-97 2007-09-12 TMP88CS38B/CM38B/CP38B (3) 1-word data transfer Check the MST by the INTSBI interrupt process after an 1-word data transfer is completed, and determine whether the mode is a master or slave. a. When the MST is “1” (Master mode) Check the TRX and determine whether the mode is a transmitter or receiver. 1. When the TRX is “1” (Transmitter mode) Test the LRB. When the LRB is “1”, a receiver does not request data. Implement the process to generate a stop condition (Described later) and terminate data transfer. When the LRB is “0”, the receiver requests next data. When the next transmitted data is other than 8 bits, set the BC, set the ACK to “1”, and write the transmitted data to the SBIDBR. After writing the data, the PIN becomes “1”, a serial clock pulse is generated for transferring a next 1 word of data from the SCL pin, and then the 1-word data is transmitted. After the data is transmitted, and an INTSBI interrupt request occurs. The PIN become “0” and the SCL pin is set to low level. If the data to be transferred is more than one word in length, repeat the procedure from the LRB test above. Write to SBIDBR SCL pin 1 2 3 4 5 6 7 8 SDA pin D7 D6 D5 D4 D3 D2 D1 D0 9 Acknowledge signal from a receiver PIN INTSBI interrupt request Figure 2.9.14 Example of when BC = “000”, ACK = “1” 2. When the TRX is “0” (Receiver mode) When the next transmitted data is other than of 8 bits, set the BC again. Set the ACK to “1” and read the received data from the SBIDBR (Reading data is undefined immediately after a slave address is sent). After the data is read, the PIN becomes “1”. A serial bus interface circuit outputs a serial clock pulse to the SCL to transfer next 1 word of data and sets the SDA pin to “0” at the acknowledge signal timing. An INTSBI interrupt request occurs and the PIN becomes “0”. Then a serial bus interface circuit outputs a clock pulse for 1 word of data transfer and the acknowledge signal each time that received data is read from the SBIDBR. Read to SBIDBR SCL pin 1 2 3 4 5 6 7 8 SDA pin D7 D6 D5 D4 D3 D2 D1 D0 PIN 9 New D7 Acknowledge signal to a transmitter INTSBI interrupt Figure 2.9.15 Example of when BC = “000”, ACK = “1” 88CS38B-98 2007-09-12 TMP88CS38B/CM38B/CP38B To make the transmitter terminate transmit, clear the ACK to “0” before reading data which is 1 word before the last data to be received. A serial bus interface circuit does not generate a clock pulse for the acknowledge signal by clearing ACK. In the interrupt routine of end of transmission, when the BC is set to “001” and read the data, PIN is set to “1” and generates a clock pulse for a 1-bit data transfer. In this case, since the master device is a receiver, the SDA line on a bus keeps the high level. The transmitter receives the high-level signal as an ACK signal. The receiver indicates to the transmitter that data transfer is complete. After 1-bit data is received and an interrupt request has occurred, generates the stop condition to terminate data transter. SCL pin 1 2 3 4 5 6 7 8 SDA pin D7 D6 D5 D4 D3 D2 D1 D0 1 Acknowledge signal sent to a transmitter PIN INTSBI interrupt request “001” → BC Read SBIDBR “0” → ACK Read SBIDBR Figure 2.9.16 Termination of Data Transfer in Master Receiver Mode b. When the MST is “0” (Slave mode) In the slave mode, a serial bus interface circuit operates either in normal slave mode or in slave mode after losing arbitration. In the slave mode, the conditions of generating INTSBI are follows: • When the received slave address matches to the value set by the I2CAR • When a “GENERAL CALL” is received • At the end of transferring or receiving after matching of slave address or receiving of “GENERAL CALL” A serial bus interface circuit changes to a slave mode if arbitration is lost in the master mode. And an INTSBI interrupt request occurs when word data transfer terminates after losing arbitration. The behavior of INTSBI and PIN after losing arbitration are shown in Table 2.9.2. Table 2.9.2 The Behavior of INTSBI and PIN after Losing Arbitration When the arbitration occurs during transmission of slave address as a master INTSBI PIN When the arbitration occurs during transmission of data as a master transmit mode INTSIB is generated at the terminatin of word data. When the slave address matches the value set by I2CAR, the PIN is cleared to “0” by generating of INTSBI. When the slave address doesn’t match the value set by I2CAR, the PIN keeps “1”. PIN keeps “1”. Check the AL (Bit3 in the SBISR), the TRX (Bit6 in the SBISR), the AAS (Bit2 in the SBISR), and the AD0 (Bit1 in the SBISR) and implements processes according to conditions listed in Table 2.9.3. 88CS38B-99 2007-09-12 TMP88CS38B/CM38B/CP38B Table 2.9.3 Operation in the Slave Mode TRX AL AAS AD0 Conditions Process 1 1 1 0 A serial bus interface circuit loses arbitration when transmitting a slave address. And receives a slave address of which the value of the direction bit sent from another master is “1”. Set the number of bits in 1 word to the BC and write transmitted data to the SBIDBR. 0 1 0 In the slave receiver mode, a serial bus interface circuit receives a slave address of which the value of the direction bit sent from the master is “1”. 0 0 In the slave transmitter mode, 1-word data is transmitted. Test the LRB. If the LRB is set to “1”, set the PIN to “1” since the receiver does not request next data. Then, clear the TRX to “0” release the bus. If the LRB is set to “0”, set the number of bits in 1 word to the BC and write transmitted data to the SBIDBR since the receiver requests next data. 1 1/0 A serial bus interface circuit loses arbitration when transmitting a slave address. And receives a slave address of which the value of the direction bit sent from another master is “0” or receives a “GENERAL CALL”. Read the SBIDBR for setting the PIN to “1” (Reading dummy data) or write “1” to the PIN. 0 0 A serial bus interface circuit loses arbitration when transmitting a slave address or data. And terminates transferring word data. A serial bus interface circuit is changed to slave mode. To clear AL to “0”, read the SBIDBR or write the data to SBIDBR. 1 1/0 In the slave receiver mode, a serial bus interface circuit receives a slave address of which the value of the direction bit sent from the master is “0” or receives “GENERAL CALL”. Read the SBIDBR for setting the PIN to “1” (Reading dummy data) or write “1” to the PIN. 0 1/0 In the slave receiver mode, a serial bus interface circuit terminates receiving of 1-word data. Set the number of bits in 1 word to the BC and read received data from the SBIDBR. 0 1 0 Note: In the slave mode, if the slave address set in I2CAR is “00000000B”, the TRX changes to “1” by receiving the start byte data “00000001B”. (4) Stop condition generation When the BB is “1”, a sequence of generating a stop condition is started by setting “1” to the MST, TRX, and PIN, and clear “0” to the BB. Do not modify the contents of the MST, TRX, BB, PIN until a stop condition is generated on a bus. When a SCL line on a bus is pulled down by other devices, a serial bus interface circuit generates a stop condition after they release a SCL line. “1” → MST “1” → TRX “0” → BB “1” → PIN Stop condition SCL pin SDA pin PIN BB (Read) Figure 2.9.17 Stop Condition Generation 88CS38B-100 2007-09-12 TMP88CS38B/CM38B/CP38B (5) Restart Restart is used to change the direction of data transfer between a master device and a slave device during transferring data. The following explains how to restart a serial bus interface circuit. Clear “0” to the MST, TRX and BB and set “1” to the PIN. The SDA pin retains the high level and the SCL pin is released. Since a stop condition is not generated on a bus, a bus is assumed to be in a busy state from other devices. Test the BB until it becomes “0” to check that the SCL pin a serial bus interface circuit is released. Test the LRB until it becomes “1” to check that the SCL line on a bus is not pulled down to the low level by other devices. After confirming that a bus stays in a free state, generate a start condition with procedure (2). In order to meet setup time when restarting, take at least 4.7 μs of waiting time by software from the time of restarting to confirm that a bus is free until the time to generate a start condition. Note: When restarting after receiving in master receiver mode, because the divice doesn’t send an acknowledgement as a last data, the level of SCL line can not be conrirmied by reading LRB. Therefore, confirm the status of SCL line by reading P5PRD register. “0” → MST “0” → TRX “0” → BB “1” → PIN “1” → MST “1” → TRX “1” → BB “1” → PIN 4.7 μs (Min) Start condition SCL (Bus) SCL (Pin) SDA (Pin) LRB BB PIN Figure 2.9.18 Timing Diagram when Restarting 88CS38B-101 2007-09-12 TMP88CS38B/CM38B/CP38B 2.9.9 Clocked-synchronous 8-Bit SIO Mode Control The following registers are used to control the serial bus interface (SBI) and monitor the operation in the clocked-synchronous 8-bit SIO mode. Serial Bus Interface Control Register A SBICRA 7 6 (00020H) SIOS SIOINH SIOS SIOINH SIOM SCK Note 1: Note 2: Note 3: 5 4 SIOM 3 2 “0” 0: 1: 0: 1: 00: 01: 10: 11: Indicate transfer start/stop Continue/abort transfer Transfer mode select Serial clock selection (At fc = 16 MHz, Output on SCK pin) 1 0 (Initial value: 0000 *000) SCK Stop Start Continue transfer Abort transfer (Automatically cleared after abort) 8-bit transmit mode Reserved 8-bit transmit/receive mode 8-bit receive mode DV1CK = 0 000: 1000.0 kHz 001: 500.0 kHz 010: 250.0 kHz 011: 125.0 kHz 100: 62.5 kHz 101: 31.2 kHz 110: 15.6 kHz 111: External clock (Input from SCK pin) DV1CK = 1 000: 500.0 kHz 001: 250.0 kHz 010: 125.0 kHz 011: 62.5 kHz 100: 31.2 kHz 101: 15.6 kHz 110: 7.8 kHz 111: External clock (Input from SCK pin) Write only fc: High-frequency clock [Hz], *: Don’t care Clear the SIOS to “0” and set the SIOINH to “1” when setting the transfer mode and serial clock. SBICRA is write-only register and cannot be used with any of read-modify-write instructions such as bit manipulation, etc. Serial Bus Interface Data Register SBIDBR 7 6 5 4 3 2 1 0 (00021H) (Initial value: **** ****) Note1 : R/W The data which was written into SBIDBR cannot be read, since a write buffer and a read buffer are independent in SBIDBR. Therefore, SBIDBR cannot be used with any of read-modify-write instructions such as bit manipulation, etc. Note 2: *: Don’t care Serial Bus Interface Control Register B SBICRB 7 6 5 4 (00023H) “0” “0” “0” “1” SBIM Serial bus interface operation mode selection SWRST1 Software reset start bit SWRST0 Note 1: Note 2: Note 3: Note 4: Note 5: Note 6: 3 2 SBIM 00: 01: 10: 11: 1 0 SWRST1 SWRST0 (Initial value: **** 0000) Port mode (Serial bus interface output disable) SIO mode 2 I C bus mode Reserved Write only Software reset starts by first writing “10” and next writing “01” *: Don’t care Switch a mode to port after data transfer is complete. 2 Switch a mode to I C bus mode or clock synchronous 8-bit SIO mode after confirming that the port is high level. SBICRB is a write-only register and cannot be used with any of read-modify-write instructions such as bit manipulation, etc. Clear bit7 to 5 in SBICRB to “0”, and set bit4 to “1”. When the SWRST (Bit1, 0 in SBICRB) is written to “01”, “10”, software reset is occurred. This time, control the serial bus interface and monitor the operation status registers except the SBIM (Bit3, 2 in SBICRB) and the CHS (Bit6 in PMPXCR) are reseted. Control the serial bus interface and monitor the operation status registers are SBICRA, SBICRB, SBIDBR, I2CAR, SBISRA, SBISRB, SCCRA, SCCRB and SCSR. Figure 2.9.19 Control Register/Data Buffer Register/Status Register in SIO Mode (1) 88CS38B-102 2007-09-12 TMP88CS38B/CM38B/CP38B Serial Bus Interface Status Register A SBISRA 7 6 5 4 3 2 1 0 (00020H) SWR MON SWRMON (Initial value: **** ***1) 0: During software reset Software reset monitor Read only 1: − (Initial) Serial Bus Interface Status Register B SBISRB 7 6 5 4 3 2 1 0 (00023H) “1” “1” “1” “1” SIOF SEF “1” “1” SIOF SEF Note: Serial transfer operating status monitor Shift operating status monitor 0: Transfer terminated 1: Transfer in process Read only 0: Shift operation terminated 1: Shift operation in process Set bit7 to 4, bit1 and bit0 in SBISRB to “1”. Figure 2.9.20 Control Register/Data Buffer Register/Status Register in SIO Mode (2) (1) Serial clock a. Clock source The SCK (Bits 2 to 0 in SBICRA) is used to select the following functions. 1. Internal clock In an internal clock mode, any of seven frequencies can be selected. The serial clock is output to the outside on the SCK pin. The SCK pin becomes a high level when data transfer starts. When writing (in the transmit mode) or reading (in the receive mode) data cannot follow the serial clock rate, an automatic-wait function is executed to stop the serial clock automatically and hold the next shift operation until reading or writing is complete. Automatic-wait function SCK pin output SO pin output Write transmitted data 1 2 a0 a1 3 7 a 2 a5 a6 8 a7 a 1 2 b0 b1 b 6 7 8 1 2 3 b4 b5 b 6 b 7 c 0 c 1 c2 c Figure 2.9.21 Automatic-wait Function 2. External (SCK = “111”) An external clock supplied to the SCK pin is used as the serial clock. In order to ensure shift operation, a pulse width of at least 2-machine cycles is required for both high and low levels in the serial clock. The maximum data transfer frequency is 1MHz (fc = 16.0 MHz). SCK pin tSCKL tSCKH tSCKL, tSCKH > 2 tcyc Note: tcyc = 4/fc (in NORMAL mode, IDLE mode) Figure 2.9.22 The Maximum Data Transfer Frequency in The External Clock Input 88CS38B-103 2007-09-12 TMP88CS38B/CM38B/CP38B b. Shift edge The leading edge is used to transmit data, and the trailing edge is used to receive data. 1. Leading edge Data is shifted on the leading edge of the serial clock (at a falling edge of the SCK pin input/output). 2. Trailing edge Data is shifted on the trailing edge of the serial clock (at a rising edge of the SCK pin input/output). SCK pin SO pin Bit0 Shift register Bit1 Bit2 76543210 *7654321 **765432 Bit3 Bit4 Bit5 Bit6 Bit7 ***76543 ****7654 *****765 ******76 *******7 Bit5 Bit6 Bit7 (a) Leading edge SCK pin SI pin Bit0 Shift register ******** Bit1 0******* Bit2 10****** Bit3 210***** Bit4 3210**** 43210*** 543210** 6543210* 76543210 (b) Trailing edge *: Don’t care Figure 2.9.23 Shift Edge (2) Transfer mode The SIOM (Bits 5 and 4 in SBICRA) is used to select a transmit, receive, or transmit/receive mode. a. 8-bit transmit mode Set a control register to a transmit mode and write transmit data to the SBIDBR. After the transmit data is written, set the SIOS to “1” to start data transfer. The transmitted data is transferred from the SBIDBR to the shift register and output to the SO pin in synchronous with the serial clock, starting from the least significant bit (LSB). When the transmit data is transferred to the shift register, the SBIDBR becomes empty. The INTSBI (Buffer empty) interrupt request is generated to request new data. When the internal clock is used, the serial clock will stop and automatic-wait function will be initiated if new data is not loaded to the data buffer register after the specified 8-bit data is transmitted. When transmit new data is written, automatic-wait function is canceled. When the external clock is used, data should be written to the SBIDBR before new data is shifted. The SO pin is “1” from the time transmission starts until the first data bit is sent. When SIOF becomes “0”, the shift register is cleared. So, output of an undefined value is not prevented at the start of the next transmission. The transfer speed is determined by the maximum delay time between the time when an interrupt request is generated and the time when data is written to the SBIDBR by the interrupt service program. 88CS38B-104 2007-09-12 TMP88CS38B/CM38B/CP38B Transmitting data is ended by cleaning the SIOS to “0” by the buffer empty interrupt service program or setting the SIOINH to “1”. When the SIOS is cleared, the transmitted mode ends when all data is output. In order to confirm if data is surely transmitted by the program, set the SIOF (Bit3 in the SBISRB) to be sensed. The SIOF is cleared to “0” when transmitting is complete. When the SIOINH is set, transmitting data stops. The SIOF turns “0”. When the external clock is used, it is also necessary to clear the SIOS to “0” before new data is shifted; otherwise, dummy data is transmitted and operation ends. Clear SIOS SIOS SIOF SEF SCK pin (Output) SO pin a0 a1 a2 a3 a4 a5 a6 a7 b0 b1 b2 b3 b4 b5 b6 b7 b5 b6 b7 INTSBI interrupt request SBIDBR a b Write transmitted data (a) Internal clock Clear SIOS SIOS SIOF SEF SCK pin (Input) SO pin a0 a1 a2 a3 a4 a5 a6 a7 b0 b1 b2 b3 b4 INTSBI interrupt request SBIDBR a b Write transmitted data (b) External clock Figure 2.9.24 Transfer Mode 88CS38B-105 2007-09-12 TMP88CS38B/CM38B/CP38B Example: Program to stop transmitting data. (When external clock is used.) STEST1: STEST2: TEST (SBISRB). SEF JRS F, STEST1 TEST (P5). 3 JRS T, STEST2 LD (SBICRA), 00000111B ; If SEF = 1 then loop ; If SCK = 0 then loop ; SIOS ← 0 SCK pin SIOF SO pin Bit6 Bit7 tSODH = Min 3.5/fc [s] (in NORMAL mode, IDLE mode) Figure 2.9.25 Transmitted Data Hold Time at End of Transmit b. 8-bit receive mode Set a control register to a receive mode and the SIOS to “1” for switching to a receive mode. Data is received from the SI pin to the shift register in synchronous with the serial clock, starting from the least significant bit (LSB). When the 8-bit data is received, the data is transferred from the shift register to the SBIDBR. The INTSBI (Buffer full) interrupt request is generated to request of reading the received data. The data is read from the SBIDBR by the interrupt service program. When the external clock is used, since shift operation is synchronized with the clock pulse provided externally, the received data should be read from SBIDBR before next serial clock is input. If the received data is not read, further data to be received is canceled. When the internal clock is used, the automatic wait function is executed until received data is read from SBIDBR. The maximum transfer speed when the external clock is used is determined by the delay time between the time when an interrupt request is generated and the time when received data is read. Received data disappears if this data is not completely read before reception of the next data terminates. In this case, the next data received is read. Receiving data is ended by clearing the SIOS to “0” by the buffer full interrupt service program or setting the SIOINH to “1”. When the SIOS is cleared, received data is transferred to the SBIDBR in complete blocks. The received mode ends when the transfer is complete. In order to confirm if data is surely received by the program, set the SIOF (Bit3 in SBIDBR) to be sensed. The SIOF is cleared to “0” when receiving is complete. After confirming that receiving has ended, the last data is read. When the SIOINH is set, receiving data stops. The SIOF turns “0” (the received data becomes invalid, therefore no need to read it). Note: When the transfer mode is switched, the SBIDBR contents are lost. In case that the mode needs to be switched, receiving data is concluded by clearing the SIOS to “0”, read the last data, and then switch the mode. 88CS38B-106 2007-09-12 TMP88CS38B/CM38B/CP38B Clear SIOS SIOS SIOF SEF SCK pin (Output) SI pin a0 a1 a2 a3 a4 a5 a6 a7 b0 b1 b2 b3 b4 b5 b6 b7 INTSBI interrupt request SBIDBR a Read received data b Read received data Figure 2.9.26 Receive Mode (Example: Internal clock) c. 8-bit transmit/receive mode Set a control register to a transmit/receive mode and write data to the SBIDBR. After the data is written, set the SIOS to “1” to start transmitting/receiving. When transmitting, the data is output from the SO pin on the leading edges in synchronous with the serial clock, starting from the least significant bit (LSB). When receiving, the data is input to the SI pin on the trailing edges of the serial clock. 8-bit data is transferred from the shift register to the SBIDBR, and the INTSBI interrupt request occurs. The interrupt service program reads the received data from the data buffer register and writes data to be transmitted. The SBIDBR is used for both transmitting and receiving. Transmitted data should always be written after received data is read. When the internal clock is used, automatic-wait function is initiated until received data is read and next data is written. When the external clock is used, since the shift operation is synchronized with the external clock, received data is read and transmitted data is written before new shift operation is executed. The maximum transfer speed when the external clock is used is determined by the delay time between the time when an interrupt request is generated and the time when received data is read and transmitted data is written. When transmission starts, a value which is the same as the last bit of previously transmitted data is output from the time SIOF is set to “1” until the falling edge of SCK occurs. Transmitting/receiving data is ended by cleaning the SIOS to “0” by the INTSBI interrupt service program or setting the SIONH to “1”. When the SIOS is cleared, received data is transferred to the SBIDBR in complete blocks. The transmit/receive mode ends when the transfer is complete. In order to confirm if data is surely transmitted/received by the program, set the SIOF (bit 3 in SBISRB) to be sensed. The SIOF becomes “0” after transmitting/receiving is complete. When the SIONH is set, transmitting/receiving data stops. The SIOF turns “0”. Note: When the transfer mode is switched, the SBIDBR contents are lost. In case that the mode needs to be switched, conclude transmitting/receiving data by clearing the SIOS to “0”, read the last data, and then switch the transfer mode. 88CS38B-107 2007-09-12 TMP88CS38B/CM38B/CP38B Clear SIOS SIOS SIOF SEF SCK pin (Output) SO pin a0 a1 a2 a3 a4 a5 a6 a7 b0 b1 b2 b3 b4 b5 b6 b7 SI pin c0 c1 c2 c3 c4 c5 c6 c7 d0 d1 d2 d3 d4 d5 d6 d7 INTSBI interrupt request SBIDBR a c Write transmitted data (a) b Read received Write transmitted data (c) data (b) d Read received data (d) Figure 2.9.27 Transmit/Receive Mode (Example: Internal clock) SCK pin SIOF SO pin Bit6 Bit7 in last transmitted word tSODH = Min 4/fc [s] (In NORMAL mode, IDLE mode) Figure 2.9.28 Transmitted Data Hold Time at End of Transmit/Receive 88CS38B-108 2007-09-12 TMP88CS38B/CM38B/CP38B 2.10 Remote Control Signal Preprocessor/External Interrupt 3 Input Pin The remote control signal waveform can be determined by inputting the remote control signal waveform from which the carrier wave was eliminated by the receive circuit to P30 (INT3/RXIN) pin. When the remote control signal preprocessor/external interrupt 3 pin is also used as the P30 port, set the P30 port output latch to “1”. When it is not used as the remote control signal preprocessor/external interrupt 3 input pin, it can be used for normal port. 2.10.1 Configuration 11 fc/2 10 fc/2 8 fc/2 7 fc/2 6 fc/2 5 fc/2 2 fc/2 Receive bit counter Receive bit counter value monitor (RBCTM) Selector RNC INT3/RXIN Polarity select Noise canceller to RNCM INT3 Interrupt request Interrupt select INT. EINT Measurement width select 8-bit up counter Match detect Selector SRM 6 2 3 RPOLS fc/2 8 fc/2 10 fc/2 12 fc/2 2 RCCK Remote control receive counter register (RXCTR) 2 RMM 4 RCS CREGA RXCR1 RXCR2 Remote control receive control register 1 Remote control receive control register 2 Shift register Remote control receive data buffer register (RXDBR) Figure 2.10.1 Remote Control Signal Preprocessor 2.10.2 Remote Control Signal Preprocessor Control When the remote control signal preprocessor is used, operating states are controlled and monitored by the following registers. Interrupt requests also use the remote control signal preprocessor/external interrupt 3 input pin. • Remote control receive control register 1 (RXCR1) • Remote control receive control register 2 (RXCR2) • Remote control receive counter register (RXCTR) • Remote control receive data buffer register (RXDBR) • Remote control receive status register (RXSR) When this pin is used for the external interrupt 3 input, set EINT in RXCR1 to other than “11”. 88CS38B-109 2007-09-12 TMP88CS38B/CM38B/CP38B Remote Control Receive Control Register 1 RXCR1 7 (00FE8H) 6 RCCK 5 4 RPOLS 3 2 1 EINT 0 (Initial value: 0000 0000) RNC 6 00: fc/2 (Hz) 8-bit up counter source clock select RCCK Remote control signal polarity select RPOLS 01: fc/2 8 10: fc/2 10 11: fc/2 12 0: Positive 1: Negative 00: Rising edge EINT 01: Falling edge (at RPOLS = 0) Interrupt source select 10: Rising/falling edge R/W 11: 8-bit receive end 001: 2 /fc × 7 − 1/fc (s) 2 010: 2 /fc × 7 − 1/fc 5 011: 2 /fc × 7 − 1/fc 6 100: 2 /fc × 7 − 1/fc 7 Noise canceler noise eiminating time select RNC 101: 2 /fc × 7 − 1/fc 8 110: 2 /fc × 7 − 1/fc 10 111: 2 /fc × 7 − 1/fc 12 000: Noise canceler disable Note 1: fc: High-frequency clock [Hz] Note 2: After reset, RPOLS do not change the set value in the receiving remote control signal. For setting interrupt edge and measurement data, use EINT and RMM. Remote Control Receive Control Register 2 RXCR2 7 6 (00FE9H) 5 4 CREGA 3 2 RCS RMCEN CREGA Setting of detect time for match with 8-bit up counter upper 4 bits RCS 8-bit up counter start control RMCEN Remote control signal preprocesser enable/disable 1 0 RMM (Initial value: 0000 0000) Match detect time (Tth) = 16 × CREGA/RCCK [s] CREGA = 0H to FH CREGA = 2H, RCCK = fc/2 [Hz], at fc = 16 MHz, DV1CK = 1 Tth = 128 [μs] 6 Example: 0: Stop and counter clear 1: Start 0: Disable 1: Enable R/W 00: RMM Measurement mode select (Invalid when EINT = “10”) 01: Refer to Talbe 2.10.1 10: 11: Note 1: fc: High-frequency clock [Hz] Note 2: When an interrupt source is set for rising/falling edge, low and high widths are forcibly measured separately. Note 3: Set CREGA (0H to FH) before EINT sets to 8-bit receive end. Figure 2.10.2 Remote Control Receive Control Register 1, 2 88CS38B-110 2007-09-12 TMP88CS38B/CM38B/CP38B Remote Control Receive Counter Register RXCTR 7 6 5 4 3 2 1 0 Read only (Initial value: 0000 0000) 4 3 2 1 0 Read only (Initial value: 0000 0000) 4 3 2 1 0 OVFF SRM RNCM Read only (Initial value: 0000 *000) (00FEAH) Remote Control Receive Data Buffer Register RXDBR 7 6 5 (00FEBH) Remote Control Receive Status Register RXSR 7 6 (00FECH) 5 RBCTM RBCTM Receive bit counter value monitor OVFF 8-bit up counter overflow flag SRM Data buffer register input monitor RNCM 0: No overflow 1: Overflow 0: Upper 4 bits of 8 bit up counter < CREGA 1: Upper 4 bits of 8 bit up counter ≥ CREGA Read only Remote control signal monitor after passing through noise canceler *: Don’t care Figure 2.10.3 Remote Control Receive Counter Register, Data Buffer Register, Status Register 88CS38B-111 2007-09-12 TMP88CS38B/CM38B/CP38B Table 2.10.1 Combination of Interrupt Source and Measurement Mode RPOLS EINT RMM Interrupt Source Measurement Mode 00 00 10 11 01 0 01 10 11 10 − 00 11 Receive end 10 00 00 10 11 01 1 01 10 11 10 − 00 11 Receive end 10 88CS38B-112 2007-09-12 TMP88CS38B/CM38B/CP38B 2.10.3 Noise Elimination Time Setting The remote control receive circuit has a noise canceler. By setting RNC in RXCR1, input signals shorter than the fixed time can be eliminated as noise. Table 2.10.2 Noise Elimination Time Setting (fc = 16 MHz) RNC Minimum Signal Pulse Width − 000 2.10.4 Maximum Noise Width to be Eliminated − 001 (2 + 5)/fc (2.31 μs) (2 × 7 − 1)/fc (1.69 μs) 010 (2 + 5)/fc (16.31 μs) (2 × 7 − 1)/fc (13.88 μs) 011 (2 + 5)/fc (32.31 μs) (2 × 7 − 1)/fc (27.88 μs) 100 (2 + 5)/fc (64.31 μs) (2 × 7 − 1)/fc (55.88 μs) 101 (2 + 5)/fc (128.3 μs) (2 × 7 − 1)/fc (111.9 μs) 110 (2 + 5)/fc (512.3 μs) (2 × 7 − 1)/fc (447.9 μs) 111 (2 + 5)/fc (1.024 ms) (2 × 7 − 1)/fc (895.9 μs) 5 8 9 10 11 13 14 2 5 6 7 8 10 11 Operation (1) Interrupts at rising, falling, or rising/falling edge, and measurement modes First set EINT and RMM. Next, set RCS to “1”; the 8-bit up counter is counted up by the internal clock. After measurement, the 8-bit up counter value is saved in RXCTR. Then, the 8-bit up counter is cleared, an INT3 request is generated, and the 8-bit up counter resumes counting. If the 8-bit up counter overflows (FFH) before measurement is completed, an INT3 request is generated and the overflow flag (OVFF) is set to “1”. Then, the 8-bit up counter is cleared. An overflow can be detected by reading OVFF by the interrupt processing. To restart the 8-bit up counter, set RCS to “1”. Setting RCS to “1” zero clears OVFF. 88CS38B-113 2007-09-12 88CS38B-114 RXCTR 8-bit up counter value RXCTR 8-bit up counter value RXCTR 8-bit up counter value INT3 request RCCK RNCM I−3 I−2 I−1 I−3 I−2 I−1 I−3 I−2 I−1 I I I I I 1 1 1 2 2 2 3 3 3 1 5 6 7 8 2 m m 1 (c) High width measurement m−2m−1 2 (b) Rising edge cycle measurement 4 m (a) Low width measurement m−2m−1 n−2 n−1 3 n−2 n−1 m−4m−3m−2m−1 3 n m n m n 1 1 1 2 2 2 3 3 3 TMP88CS38B/CM38B/CP38B Figure 2.10.4 Rising Edge Interrupt Timing Chart (RPOLS = 0) 2007-09-12 88CS38B-115 RXCTR 8-bit up counter value RXCTR 8-bit up counter value RXCTR 8-bit up counter value INT3 request RCCK RNCM I−3 I−2 I−1 I−3 I−2 I−1 I−3 I−2 I−1 I I I I I 1 1 1 2 2 2 3 3 3 1 5 6 7 8 2 m m 1 (c) Low width measurement m−2m−1 2 (b) Falling edge cycle measurement 4 m (a) High width measurement m−2m−1 n−2 n−1 3 n−2 n−1 m−4m−3m−2m−1 3 n m n m n 1 1 1 2 2 2 3 3 3 TMP88CS38B/CM38B/CP38B Figure 2.10.5 Falling Edge Interrupt Timing Chart (RPOLS = 0) 2007-09-12 RXCTR 8-bit up counter value INT3 request RCCK RNCM I−3 I−2 I−1 I I 1 2 3 m m 1 2 (a) High and low width measurement m−2m−1 3 n−2 n−1 n n 1 2 3 TMP88CS38B/CM38B/CP38B Figure 2.10.6 Rising/Falling Edge Interrupt Timing Chart 88CS38B-116 2007-09-12 TMP88CS38B/CM38B/CP38B (2) 8-bit receive end interrupts and measurement modes By determining one-cycle remote control signal as one-bit data set to “0” or one-pulse width remote control signal as one-bit data set to “1”, an INT3 request is generated after 8-bit data is received. When “0” is determined, this means the upper four bits in the 8-bit up counter have not reached the CREGA value. When “1” is determined, this means the upper four bits in the 8-bit up counter have reached or exceeded the CREGA value. The 8-bit up counter value is saved in RXCTR after one bit is determined. The determined data is saved, bit by bit, in RXDBR at the rising edge of the remote control signal (when RPOLS = 1, falling edge). The number of bits saved in RXDBR is counted by the receive bit counter and saved in RBCTM. RBCTM is set to “0001B” at the rising edge of the input (when RPOLS = 1, falling edge) after the INT3 request is generated. RNCM RCCK 8-bit up counter value FE FF 1 Set to “1” by command. RCS OVFF Receive bit counter value* n−1 n RBCTM* n−1 n INT3 request *: Valid only when 8 bits are received. Figure 2.10.7 Overflow Interrupt Timing Chart 88CS38B-117 2007-09-12 RXDBR SRM 8-bit up counter value CREGA INT3 request Receive bit counter value RNCM 01 02 1 03 04 05 8-bit receive end interrupt setting 06 07 01 07 0 6 01 02 03 (a) Rising Edge cycle measurement 02 1 04 05 06 07 08 7 09 0A 0B 0C 0E 0F 10 11 80H 1 01 02 [Application] Low width measurement 0C 8 TMP88CS38B/CM38B/CP38B Figure 2.10.8 8-Bit Receive End Interrupt Timing Chart (PROLS = 0) 88CS38B-118 2007-09-12 TMP88CS38B/CM38B/CP38B Table 2.10.3 Count Clock for Remote Control Preprocessor Circuit (at fc = 16 MHz) Count Clock (RCCK) Resolution [μs] Maximum Setting Time [ms] 00 4 1.024 01 16 4.096 10 64 16.38 11 256 65.53 88CS38B-119 2007-09-12 TMP88CS38B/CM38B/CP38B 2.11 8-Bit AD Converter (ADC) The TMP88CS38B/CM38B/CP38B has a 8-bit successive approximation type AD converter. 2.11.1 Configuration Figure 2.11.1 shows the circuit configuration of the AD converter. The AD converter includes control registers ADCCRA and ADCCRB, conversion result registers ADCDR1 and ADCDR2, a DA converter, a sample hold circuit, a comparator, and sequential transducer circuit. To use P5 and P6 as analog inputs, clear the output latch for P5 and P6 to “0”. Also, clear the input/output control registers (P5CR1 and P6CR) to “0”. VDD VSS DA converter A B AIN0 AIN1 Y Reference voltage Sample hold circuit Analog input multiplexer ADS 8 Analog comparator E F AIN4 AIN5 S EN SAIN Successive approximate circuit 3 Shift clock AINDS INTADC Control circuit EN 6 3 2 8 EOCF ADBF AD8TRG external trigger signal ADRS P5CR, P6CR ADCCRA P5, P6 port input/output control register AMD ACK ADCCRB AD converter control register ADCDR1, ADCDR2 AD conversion result register Figure 2.11.1 AD Converter (ADC) 2.11.2 Control Register The following register are used for AD converter. • AD converter control register 1 (ADCCRA) • AD converter control register 2 (ADCCRB) • AD conversion result register (ADCDR1/ADCDR2) (1) AD converter control register 1 (ADCCRA) ADCCRA control AD conversion start, AD operation mode select, analog input control and analog input channel select. (2) AD converter control register 2 (ADCCRB) ADCCRB control AD conversion time select. (3) AD conversion result register (ADCDR1) AD conversion result is stored after end of conversion. (4) AD conversion result register (ADCDR2) For monitoring status of conversion. Figure 2.11.2 and Figure 2.11.3 show AD converter control register. 88CS38B-120 2007-09-12 TMP88CS38B/CM38B/CP38B AD Converter Control Register 1 ADCCRA 7 (0000EH) ADRS 6 5 AMD 4 3 AINDS “0” 2 1 0 SAIN (Initial value: 0001 0000) The ADRS bit is automatically cleared after starting AD conversion. ADRS AD conversion start AMD AD operation mode select AINDS Analog input control SAIN Analog input channel selection Note 1: During AD conversion, setting ADRS to “1” initializes the ADRS bit and resets conversion. 0: − 1: AD conversion restart 00: STOP mode 01: Software start mode 00: Trigger start mode 11: Reserved 0: Analog input enable 1: Analog input disable 000: select AIN0 001: select AIN1 010: select AIN2 011: select AIN3 100: select AIN4 101: select AIN5 110: − 111: − R/W Select analog input when AD converter stops. Note 2: When the analog input is all use disabling, the AINDS should be set to “1”. Note 3: During conversion, do not perform output instruction to maintain a precision for all of the pins. And port near to analog input, do not input intense signaling of change. Note 4: The ADRS is automatically cleared to “0” after starting conversion. Note 5: Always set bit3 in ADCCRA to “0”. Note 6: Do not set ADRS (Bit7 in ADCCRA) to “1” during AD conversion. Reset it after confirming with EOCF (Bit5 in ADCDR2) that the conversion is completed or after generation an interrupt signal (INTADC) (by the interrupt processing routine or the like). Note 7 In the trigger mode, the system does not accept the second and subsequent triggers after accepting the first trigger for starting AD conversion. To restart AD conversion by a trigger, set AMD (Bits 6 and 5 in ADCCRA) to “00” and then put the system in trigger start mode again (with AMD = “10”). Note 8: When the system enters STOP mode, AD converter control register 1 (ADCCRA) is initialized. Reset this register after the system reenters NORMAL mode. AD Converter Control Register 2 ADCCRB 7 (0000FH) 6 5 4 “0” “1” 3 2 1 0 ACK ACK “0” (Initial value: **0* 000*) DV1CK = 0 DV1CK = 1 Conversion time fc = 16 MHz fc = 8 MHz fc = 16 MHz fc = 8 MHz 000 Reserved 001 010 ACK AD conversion time select 011 156/fc [s] − 19.5 − 39 100 312/fc [s] 19.5 39.0 39 78 101 624/fc [s] 39.0 78.0 78 156 110 1248/fc [s] 78.0 − 156 − 111 R/W Reserved Note 1: Do not use setting except the above liset. Note 2: Set conversion time by analog reference voltage (VDD) as follows. Note 3: Always set bit0 and bit5 in ADCCRB to “0” and set bit4 in ADCCRB to “1”. Note 4: When a read instruction for ADCCRB, bit6 to 7 in ADCCRB read in as undefined data. VDD = 4.5 to 5.5 V (15.6 μ or more) Note 5: fc: High-frequency clock [Hz] Note 6: When the system enters STOP mode, AD converter control register 2 (ADCCRA) is initialized. Reset this register after the system reenters NORMAL mode. Figure 2.11.2 AD Converter Control Register 88CS38B-121 2007-09-12 TMP88CS38B/CM38B/CP38B AD Conversion Result Register ADCDR1 7 6 (00031H) AD07 ADCDR2 7 6 (00032H) − − AD06 5 4 3 2 1 0 AD05 AD04 AD03 AD02 AD01 AD00 5 4 3 2 1 0 EOCF ADBF − − − − EOCF AD conversion end flag ADBF AD conveersion busy flag Note 1: 0: 1: 0: 1: (Initial value: 0000 0000) (Initial value: **00 ****) Under conversion or before conversion End of conversion During stop of AD conversion During AD conversion Read only The EOCF is cleared to “0” when reading the ADCDR1. Therefore, the AD conversion result should be read to ADCDR1 more first than ADCDR2. Note 2: ADBR is set to “1” by starting AD conversion and cleared to “0” by end of AD conversion. Additionally, ADBF is cleared to “0” by setting AMD = “00” in ADCCR2 or entering to the STOP mode. Figure 2.11.3 AD Converter Result Register 2.11.3 AD Converter Operation The high side of an analog reference voltage is applied to VDD, and the low side is applied to VSS pin. Dividing a reference voltage between VDD and VSS to the voltage corresponding to a bit by a rudder resistance and comparing it with the analog input voltage converts the AD. Table 2.11.1 AD Converter Operation Mode Mode AD converter disable mode Function AD converter stop mode. This mode is always used to change modes. Software start mode Single AD conversion of 1 channel which specifies input. Trigger start mode Single AD conversion of 1 channel which specifies input (AD8TRG) from Key-on wakeup circuit as a trigger. . 2.11.4 Interrupt Interrupt request signal occur at the timing when the EOCF bit is set to “1”. 88CS38B-122 2007-09-12 TMP88CS38B/CM38B/CP38B 2.11.5 AD Converter Operation Modes When the MCU places in the STOP mode during the AD conversion, the conversion is stopped and the ADCDR2 content becomes indefinite. After returning from the STOP mode, the EOCF and INTADC does not occur. Therefore, the AD conversion must be restarted after returning from the STOP mode. ADS ADCDR2 Invalid Result Invalid Invalid Result EOCF Processing Read Start Read Start Start Figure 2.11.4 AD Conversion Timing chart (1) AD conversion in STOP mode When the AD converter stop mode is specified during AD conversion, the AD conversion is stopped immediately. The AD conversion is not implemented, so the undefined value is not written to the AD conversion result register. The AD conversion start commands which occur is the AD converter stop mode are ignored. This mode is automatically selected by reset. This mode is used to change the AD converter operation mode. (2) Single mode When the AMD (Bit6, 5 to in ADCCRA) set to “01”, the AD conversion signal mode This mode does AD conversion of single channel, and conversion result is stored in ADCDR1. The EOCF (Bit5 in ADCDR2) is set to “1” at end of one conversion, and an intcrrupt request signal occurs. The EOCF is cleared to “0” by reading the AD conversion registers. But when the AD conversion is restarted before the ADCDR is read, the EOCF is cleared to “0” and the last AD conversion result is maintained till next conversion end. Do not set ADRS (Bit7 in ADCCRA) during AD conversion. Again set it after confirming with EOCF (Bit5 in ADCDR2) that the conversion is completed or after generating an interrupt signal (INTADC) (by the interrupt processing routine or the like). ADS ADCDR2 Invalid AD conversion result EOCF ADBF Conversion time (Reference to ADCCRB register) Read Start Figure 2.11.5 Single Mode 88CS38B-123 2007-09-12 TMP88CS38B/CM38B/CP38B Example: The AD conversion starts after 19.5 μs (at fc = 16 MHz) and AIN4 pin are selected as the conversion time and the analog input channel. Confirming the EOCF, the converted value is read out, and the 8 bits data is stored to address 009EH in RAM. The operation mode is a signal mode. ;AIN SELECT LD (P5), 00000000B LD (P5CR1), 00000000B LD (P6), 00000000B LD (P6CR), 00000000B LD (ADCCRA), 00100100B ; Selects AIN4, selects the software start mode LD (ADCCRB), 00011000B ; Selects the conversion time and the operation mode ; AD CONVERT START SLOOP: SET (ADCCRA). 7 ; ADRS = 1 TEST (ADCCR2). 5 ; EOCF = 1 ? JRS T, SLOOP ; RESULT DATA READ LD (9EH), (ADCDR1) (3) Trigger start mode The AD conversion of a specified single channel is executed when input (AD8TRG) from Key-on wakeup circuit is set as trigger, the conversion result is stored in the ADCDR1. The EOCF (Bit5 in ADCDR2) is set to “1” at end of one conversion, and an interrupt request signal occurs. It needs to be set the STOP mode by bit5 to 6 in ADCCRA before the AD conversion is executed again. 2.11.6 Analog Input Voltage and AD Conversion Result The analog input voltage is corresponded to the 8-bit digital value converted by the AD as shown in Figure 2.11.6. AD Conversion result FFH FEH FDH 03H 02H 01H 0 1 2 3 253 254 255 256 × VDD − VSS Analog input voltage 256 Figure 2.11.6 Analog Input Voltage and AD Conversion Result (typ.) 88CS38B-124 2007-09-12 TMP88CS38B/CM38B/CP38B 2.11.7 STOP Modes during AD Conversion When standby mode (STOP mode) is entered forcibly during AD conversion, the AD convert operation is suspended and the AD converter is initialized. (ADCCRA and ADCCRB are initialized to initial value.) Also, the conversion result is indeterminate. (Conversion results up to the previous operation are cleared, so be sure to read the conversion results before entering standby mode.) When restored from standby mode, AD conversion is not automatically restarted, so it is necessary to restart AD conversion after setting ADCCRA and ADCCRB. Note that since the analog reference voltage is automatically disconnected, there is no possibility of current flowing into the analog reference voltage. 2.11.8 Notice of AD Converter (1) Analog input voltage range Voltage range of analog input (AIN0 to AIN5) must be forced from VSS to VDD. If input voltage of which out of range is forced to analog input pin, AD conversion result to unknown. Also, this cause other analog input pin unstable. (2) I/O port with analog input Analog input pins (AIN0 to AIN5) are also I/O port. During AD conversion using any analog input pin, don’t operate other I/O port with analog input. Because, AD accuracy would be worse. Also, other electrically swinging port without analog input may cause noise to near analog input pin. (3) Reduce to noise Figure 2.11.7 is shown as internal equivalent circuit of analog input pin. Increasing output impedance of analog input supply, cause noise or other non-good condition. Therefore, output impedance of analog input supply must be less than 5 kΩ. And we recommend to connect capacitance to analog input pin. AINx Analog input supply impedance 5 kΩ (max) Internal resistance R = 5 kΩ (typ.) Analog converter Internal capacitance C = 22 pF (typ.) DA converter Figure 2.11.7 Analog Input Equivalent Circuit and Analog Input Pin 88CS38B-125 2007-09-12 TMP88CS38B/CM38B/CP38B 2.12 Key-on Wakeup In this MCU the IDLE mode is also released by low active port inputs. The low input voltage is regulated higher than the other normal ports. Therefore the ports can be enabled by analog input level. 2.12.1 Configuration Port P53 AIN0 AD converter VIL ≤ VDD × 0.65 KWU0 AD8TRG KWU1 Port P54 AIN1 KWU2 Port P55 AIN2 KWU3 Port P56 AIN3 KWU 4 Port P60 AIN4 KWU5 Port P61 AIN5 Noise reject circuit INTKWU INTAD EN * IDLE5 IDLE4 IDLE3 IDLE2 IDLE1 IDLE0 IN IN IN IN IN IN IDLECR (00FD0H) * * IDLE5 IDLE4 IDLE3 IDLE2 IDLE1 IDLE0 IN IN IN IN IN IN IDLEIN (00FD0H) Figure 2.12.1 Key-on Wakeup Control Circuit 2.12.2 Control P53 to P56 and P60, P61 ports can be controlled by IDLE control register (IDLECR). It can be configured as enable/disable in one-bit unit. When those pins are used by IDLE mode release, those pins must be set input mode (P5CR1, P5, P6CR, P6, ADCCRA). IDLE mode is controlled by system control register 2 (SYSCR2) and maskable interrupts. After the individual enable flag (EF5) is set to “1”, the IDLE mode must starts. When enabled port input generates INTKWU interrupt, the IDLE mode is released. Low level input voltage in those ports is regulated to less than VDD × 0.65 (V). IDLE port monitorring register (IDLEIN) can be used to check state of ports. INTADEN can enable to generate AD8TRG, which is used as trigger of AD converter trigger start mode. Noise reject circuit eliminate noise, which is less than 24 μs period. 88CS38B-126 2007-09-12 TMP88CS38B/CM38B/CP38B IDLE Control Register IDLECR (00FD0H) 7 INTAD EN 6 5 4 3 2 1 0 * IDLE5 EN IDLE4 EN IDLE3 EN IDLE2 EN IDLE1 EN IDLE0 EN INTADEN Generation of AD8TRG IDLE5EN Release IDLE mode by KWU5 IDLE4EN Release IDLE mode by KWU4 IDLE3EN Release IDLE mode by KWU3 IDLE2EN Release IDLE mode by KWU2 IDLE1EN Release IDLE mode by KWU1 IDLE0EN Release IDLE mode by KWU0 0: Disable 1: Enable 0: Disable 1: Enable 0: Disable 1: Enable 0: Disable 1: Enable 0: Disable 1: Enable 0: Disable 1: Enable 0: Disable 1: Enable (Initial value: 0*00 0000) Write only *: Don’t care IDLE Port Monitoring Register IDLEIN (00FD0H) 7 6 5 4 3 2 1 0 * * IDLE5 IN IDLE4 IN IDLE3 IN IDLE2 IN IDLE1 IN IDLE0 IN IDLE5IN IDLE4IN IDLE3IN Input level of KWU5 Input level of KWU4 Input level of KWU3 IDLE2IN Input level of KWU2 IDLE1IN Input level of KWU1 IDLE0IN Input level of KWU0 0: “0” detect 1: “1” detect 0: “0” detect 1: “1” detect 0: “0” detect 1: “1” detect 0: “0” detect 1: “1” detect 0: “0” detect 1: “1” detect 0: “0” detect 1: “1” detect (Initial value: **00 0000) Read only *: Don’t care Figure 2.12.2 Key-on Wakeup Control Register 88CS38B-127 2007-09-12 TMP88CS38B/CM38B/CP38B 2.13 Pulse Width Modulation Circuit Output The TMP88CS38B/CM38B/CP38B has four 12-bit resolution PWM output channels including two 14-bit resolution selectable and six 7-bit resolution PWM output channels. DA converter output can easily be obtained by connecting an external low-pass filter. PWM outputs are multiplexed with general purpose I/O ports as; P40 ( PWM0 ) to P47 ( PWM7 ), P50 ( PWM8 ), P51 ( PWM9 ). PWM output is negative logic. When these ports are used PWM outputs, the corresponding bits of P4, P5 output latches and input/output control latches should be set to “1”. In STOP mode, PWM output pin keeps high-level. When operation mode is changed from STOP mode to NORMAL mode, PWM control register (PWMCR1A, PWMCR2A, PWMCR1B, PWMCR2B) are initialized. 88CS38B-128 2007-09-12 TMP88CS38B/CM38B/CP38B 2.13.1 Configuration 12-Bit Resolution PWM Output Internal counter (2) Internal counter (1) 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Additional pulse generate circuit PWM0 Clock 2 (fc/2 or fc/2 ) 8 7 PWM3 0 PWM data latch 5 PWM2 S R Compare circuit All “0” 13 PWM1 PWM data latch 0 7 0 Transfer buffer (the upper) Transfer buffer (the lower) 7 0 PWMDBR1 2 0 6 0 PWMCR1B PWMCR1A PWM control register 1B PWM control register 1A 7-Bit Resolution PWM Output Internal counter Clock 7 6 5 4 3 2 1 2 PWM5 (fc/2 or fc/2 ) PWM 4 PWM7 PWM8 PWM9 S R Compare circuit 6 PWM6 0 PWM data latch 6 0 Transfer buffer 6 0 PWMDBR2 2 0 PWMCR2B PWM control register 2B 6 0 PWMCR2A PWM control register 2A Figure 2.13.1 PWM Output Circuit 88CS38B-129 2007-09-12 TMP88CS38B/CM38B/CP38B 2.13.2 PWM Output Wave Form (1) PWM0 to PWM1 Outputs PWM0 and PWM1 output can be selected 12-bit or 14-bit resolution PWM outputs. 1. 12-bit resolution PWM output When these are used as 12-bit PWM output, one period is TM = 213/fc [s] (When DV1CK = 0) and TM = 214/fc [s] (When DV1CK = 1) and sub period is TS = TM/16. The lower 8 bits of the PWM data latch controls the low level pulse width with a cycle of TS. The lower 8 bits of the PWM data latch is n (n = 1 to 255), the low level pulse width with a cycle becomes n × t0 [s] (t0 = 2/fc [s] when DV1CK = 0, t0 = 4/fc [s] when DV1CK = 1). The upper 4 bits of the PWM data latch controls a position to output the additional pulses. When the upper 4 bits of the PWM data latch is m, the additional pulses are generated in each of m periods out of 16 periods contained in a TM period. The relationship between the 4-bit data and the position of TS period where the additional pulses are generated is shown in Table 2.13.1. Table 2.13.1 The Addition Pulse (12-bit mode) Bit Position of the Lower 4 Bits of PWMDRxH Bit8 Relative position of TS in TM period where the additional pulse is generated. (Number of TS (I) is listed) Bit11 Bit10 Bit9 a) 0 0 0 0 No additional pulse b) 0 0 0 1 8 c) 0 0 1 0 4, 12 d) 0 1 0 0 2, 6, 10, 14 e) 1 0 0 0 1, 3, 5, 7, 9, 11, 13, 15 Note 1: The bit positions of a) to e) can be combined. Note 2: If the low order eight bits for the PWM data latch are set to “FFH”, be sure to set the high order four bits for this latch to “00H”. 2. 14-bit resolution PWM output When these are used as 14-bit PWM output, one period is TM = 215/fc [s] (When DV1CK = 0) and TM = 216/fc [s] (When DV1CK = 1) and sub period is TS = TM/64. The lower 8 bits of the PWM data latch controls the low level pulse width with a cycle of TS. The lower 8 bits of the PWM data latch is n (n = 1 to 255), the low level pulse width with a cycle becomes n × t0 [s] (t0 = 2/fc [s] when DV1CK = 0, t0 = 4/fc [s] when DV1CK = 1). The upper 6 bits of the PWM data latch controls a position to output the additional pulses. When the upper 6 bits of the PWM data latch is m, the additional pulses are generated in each of m periods out of 64 periods contained in a TM period. The relationship between the 6-bit data and the position of TS period where the additional pulses are generated is shown in Table 2.13.2. 88CS38B-130 2007-09-12 TMP88CS38B/CM38B/CP38B Table 2.13.2 The Addition Pulse (14-bit mode) Bit Position of the Lower 6 Bits of PWMDRxH Bit13 Bit12 Bit11 Bit10 Bit9 Bit8 Relative position of TS in TM period where the additional pulse is generated. (Number of TS (I) is listed) a) 0 0 0 0 0 0 No additional pulse b) 0 0 0 0 0 1 32 c) 0 0 0 0 1 0 16, 48 d) 0 0 0 1 0 0 8, 24, 40, 56 e) 0 0 1 0 0 0 4, 12, 20, 28, 36, 44, 52, 60 f) 0 1 0 0 0 0 2, 6, 10, 14, 18, 22, 26, 30, 34, 38, 42, 46, 50, 54, 58, 62 g) 1 0 0 0 0 0 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, 45, 47, 49, 51, 53, 55, 57, 59, 61, 63 Note 1: The bit positions of a) to g) can be combined. Note 2: If the low order eight bits for the PWM data latch are set to “FFH”, be sure to set the high order 6 bits for this latch to “00H”. (2) PWM2 to PWM3 Outputs PWM2 and PWM3 output are 12-bit resolution PWM outputs. One period is TM = 213/fc [s] (When DV1CK = 0) and TM = 214/fc [s] (When DV1CK = 1) and sub period is TS = TM/16. The lower 8 bits of the PWM data latch controls the low level pulse width with a cycle of TS. The lower 8 bits of the PWM data latch is n (n = 1 to 255), the low level pulse width with a cycle becomes n × t0 [s] (t0 = 2/fc [s] when DV1CK = 0, t0 = 4/fc [s] when DV1CK = 1). The upper 4 bits of the PWM data latch controls a position to output the additional pulses. When the upper 4 bits of the PWM data latch is m, the additional pulses are generated in each of m periods out of 16 periods contained in a TM period. The relationship between the 4-bit data and the position of TS period where the additional pulses are generated is shown in Table 2.13.1. (3) PWM4 to PMW9 Outputs These are 7-bit resolution PWM outputs. One period is TN = 28/fc [s] (When DV1CK = 0) and TN = 29/fc [s] (When DV1CK = 1) . The 7 bits of the PWM data latch controls the low level pulse width with a cycle of TN. The lower 7 bits of the PWM data latch is k (k = 1 to 127), the low level pulse width with a cycle becomes k × t0 [s] (t0 = 2/fc [s] when DV1CK = 0, t0 = 4/fc [s] when DV1CK = 1). 88CS38B-131 2007-09-12 TMP88CS38B/CM38B/CP38B 14-bit resolution PWM mode: The additional pulse Ts (1) and Ts (63) TM = 64 TS TS (0) TS (1) TS (63) t0 t0 n × t0 PWM0 to PWM1 Pulse width = n × t0 Pulse width = (n + 1) t0 12-bit resolution PWM mode: The additional pulse Ts (1) and Ts (15) TS (0) TS (1) TS (15) t0 t0 n × t0 PWM2 to PWM3 Pulse width = n × t0 Pulse width = (n + 1) t0 TN PWM 4 to PWM9 Pulse width = k × t0 Note 1: Note 2: If the pulse width is set to “00H”, PWM will note operate. Its output will remain high. If the pulse width is set to “FFH”, settings for additional pulses cannot be made. Be sure to set the pulse width to “00H”. Figure 2.13.2 PWM Output Waveform 88CS38B-132 2007-09-12 TMP88CS38B/CM38B/CP38B 2.13.3 Control PWM output is controlled by PWM control register (PWMCR1A, PWMCR1B, PWMCR2A, PWMCR2B) and PWM data buffer register (PWMDBR1, PWMDBR2). PWM Control Register 1A 7 PWMCR1A (00028H) 6 − 5 4 3 2 ABORT1 START3 START2 START1 START0 ABORT1 Abort PWM operation of channel 3 to 0 START3 Start channel 3 START2 Start channel 2 START1 Start channel 1 START0 Start channel 0 RESOLUTION1 Select channel 1 resolution RESOLUTION2 Select channel 0 resolution Note 1: 1 0 RESOLUTION 1 0 (Initial value: *000 0000) 0: Operation 1: PWM abort (PWM outputs are fixed to a high level.) 0: Stop PWM3 1: Start PWM3 0: Stop PWM2 1: Start PWM2 Write only 0: Stop PWM1 1: Start PWM1 0: Stop PWM0 1: Start PWM0 0: 14-bit resolution 1: 12-bit resolution 0: 14-bit resolution 1: 12-bit resolution *: Don’t care Note 2 After set the ABORT1 to “1”, the ABORT1 is cleared to “0” automatically. Note 3: PWMCR1A is write-only register and cannot be used with any of the read-modify-write instructions such as SET, CLR, etc. PWM Control Register 1B 7 6 5 4 3 PWMCR1B 2 1 PWMCHS1 (00029H) 0 PWMHL (Initial value: **** *000) 00: Channel 0 PWMCHS1 Select the PWM data latch of 12-bit PWM channel 01: Channel 1 10: Channel 2 Write only 11: Channel 3 PWMHL Note 1: Note 2: Select upper or lower data transfer buffer (PWMDBR1) 0: Lower 8 bits 1: Upper 4 bits or 6 bits *: Don’t care PWMCR1B is write-only register and cannot be used with any of the read-modify-write instructions such as SET, CLR, etc. PWM Data Buffer Register 1 7 6 5 4 3 2 1 PWMDBR1 (0002AH) Note 1: 0 Write only (Initial value: 0000 0000) PWMDBR1 is write-only register and cannot be used with any of the read-modify-write instructions such as SET, CLR, etc. Note 2: When operation mode is changed from STOP mode to NORMAL mode, PWMCR1A, PWMCR1B are initialized. Figure 2.13.3 PWM Control Register 1A/1B and PWM Data Buffer Register 1 88CS38B-133 2007-09-12 TMP88CS38B/CM38B/CP38B PWM Control Register 2A 7 PWMCR2A (00FF5H) − 6 5 4 3 1 0 ABORT2 START9 START8 START7 START6 START5 START4 ABORT2 Abort PWM operation of channel 9 to 4 START9 Start channel 9 START8 Start channel 8 START7 Start channel 7 START6 Start channel 6 START5 Start channel 5 START4 Start channel 4 Note 1: 2 0: Operation 1: PWM abort 0: Stop PWM9 1: Start PWM9 0: Stop PWM8 1: Start PWM8 0: Stop PWM7 1: Start PWM7 0: Stop PWM6 1: Start PWM6 0: Stop PWM5 1: Start PWM5 0: Stop PWM 4 1: Start PWM 4 (Initial value: *000 0000) Write only *: Don’t care Note 2 After set the ABORT2 to “1”, the ABORT2 is cleared to “0” automatically. Note 3: PWMCR2A is write-only register and cannot be used with any of the read-modify-write instructions such as SET, CLR, etc. PWM Control Register 2B 7 6 5 4 3 2 PWMCR2B 1 PWMCHS2 0 (Initial value: **** *000) (00FF6H) PWMCHS2 Select the PWM data latch of 7-bit PWM channel 000: Channel 4 001: Channel 5 010: Channel 6 011: Channel 7 100: Channel 8 101: Channel 9 110: Reserved 111: Reserved Write only Note 1: *: Don’t care Note 2: PWMCR2B is write-only register and cannot be used with any of the read-modify-write instructions such as SET, CLR, etc. PWM Data Buffer Register 2 7 6 5 4 3 2 1 PWMDBR2 (00FF7H) 0 Write only (Initial value: *000 0000) Note 1: *: Don’t care Note 2: PWMDBR2 is write-only register and cannot be used with any of the read-modify-write instructions such as SET, CLR, etc. Note 3: When operation mode is changed from STOP mode to NORMAL mode, PWMCR2A, PWMCR2B are initialized. Figure 2.13.4 PWM Control Register 2A/2B and PWM Data Buffer Register 2 88CS38B-134 2007-09-12 TMP88CS38B/CM38B/CP38B Binary Counter Control Register CGCR 7 6 (00030H) “0” “0” 5 4 3 2 1 0 DV1CK “0” “0” “0” “0” “0” Select of input clock to 1st divider DV1CK 0: fc/4 1: fc/8 Note 1: *: Don’t care Note 2: The all bits except DV1CK are cleared to “0”. (Initial value: 0000 0000) R/W Figure 2.13.5 DIVIDER Control Register (1) Internal counter The internal counter of PWM outputs is a free running counter. The all bits of counter are set to “1” and are not counted up at one of the following conditions. 1. During reset 2. The operation mode is changed to STOP mode. 3. Setting ABORTx (x: 1, 2) to “1”. 4. The START3 to 0 are “0” in 12-bit PWM outputs. The START9 to 4 are “0” in 7-bit PWM outputs. 5. The lower 8-bit of PWM data latch in 12-bit PWM outputs is “00H”. The PWM data latch in 7-bit PWM outputs is “00H”. (2) Outputs control and programming of PWM data The PWM outputs are fixed to a high-level immediately when the ABORTx (x: 1, 2) is set to “1”. The PWM outputs starts the operation when the STARTx (x: 0 to 9) is set to “1”. The data from the transfer buffer to a PWM data latch is transferred when the all bits of internal counter are set to “1”. Therefore, the data is transferred to a PWM data latch immediately when the internal counter is initialized. And the data is transferred to a PWM data latch at the beginning of the next cycle when all bits of the internal counter are not set to “1”. The sequence of writing the output data to PWM data latches is shown as follows; 1. PWM0 to PWM1 a) Write the channel number of PWM data latch to PWMCHS1 (Bit2 and 1 in PWMCR1B) and clear PWMHL (Bit0 in PWMCR1B) to “0”. b) Write the lower 8-bit PWM output data to PWMDBR1. c) Write the channel number of PWM data latch to PWMCHS1 and set PWMHL to “1”. d) Write the upper 4-bit or 6-bit PWM output data to PWMDBR1. e) Select the resolution of PWM output to RESOLUTIONx (x: 0, 1) (Bit0 and 1 in PWMCR1A) and set STARTx (x: 0, 1) (Bit2 and 3 in PWMCR1B) to “1”. Note: PWM output data must be write to PWMDBR1 in the order of the lower 8-bit PWM output data, the upper 4-bit (or 6-bit) PWM output data. If the upper 4-bit (or 6-bit) PWM output data is write to PWMDBR1, the lower 8-bit PWM output data is not changed (except when lower 8-bit PWM output data is “00H”). 88CS38B-135 2007-09-12 TMP88CS38B/CM38B/CP38B 2. PWM2 to PWM3 a) Write the channel number of PWM data latch to PWMCHS1 and clear PWMHL to “0”. b) Write the lower 8-bit PWM output data to PWMDBR1. c) Write the channel number of PWM data latch to PWMCHS1 and set PWMHL to “1”. d) Write the upper 4-bit PWM output data to PWMDBR1. e) Set STARTx (x: 2, 3) to “1”. 1) Data transfer timing and STOP/ABORT timing (X: 0 to 3) TM TS TM TS PWMx m × t0 n × t0 Writing PWMDBR1 (Data m to n) TS PWMx STARTx = 0 or The lower 8-bit of PWM data latch = 00H TS PWMx ABORT1 = 1 or STOP mode 2) Restart timing when operating for 1ch or more TM TM PWM0 PWM1 Restarting PWM1 Restarts after one cycle. 3) Restart timing after all channels stop TM TM Start command Figure 2.13.6 Waveform of PWM0 to PWM3 Note: PWM output data must be write to PWMDBR1 in the order of the lower 8-bit PWM output data, the upper 4-bit (or 6-bit) PWM output data. If the upper 4-bit (or 6-bit) PWM output data is write to PWMDBR1, the lower 8-bit PWM output data is not changed (except when lower 8-bit PWM output data is “00H”). 88CS38B-136 2007-09-12 TMP88CS38B/CM38B/CP38B 3. PWM4 to PWM9 a) Write the channel number of PWM data latch to PWMCHS2. b) Write the lower 7-bit PWM output data to PWMDBR2. c) Set STARTx (x: 4 to 9) to “1”. 1) Data transfer timing and STOP/ABORT timing (X: 4 to 9) TN TN PWMx m × t0 n × t0 Writing PWMDBR2 (Data m to n) PWMx STARTx (x: 4 to 9) = 0 or The lower 8-bit of PWM data latch = 00H PWMx ABORT2 = 1 or STOP mode 2) Restart timing when operating for 1ch or more TN TN PWM 4 PWM5 Restarting PWM5 Restarts after one cycle. 3) Restart timing after all channels stop TN TN Start command Figure 2.13.7 Waveform of PWM4 to PWM9 88CS38B-137 2007-09-12 TMP88CS38B/CM38B/CP38B Example: at fc = 16 MHz, DV1CK = 0 PWM0 pin outputs a 14-bit resolution PWM wave form with a low level of 32 μs width and no additional pulse. PWM1 pin outputs a 12-bit resolution PWM wave form with a low level of 16 μs width and no additional pulse. PWM4 pin outputs a PWM wave form with a low level of 8 μs width. LD (CGCR), 00H ; DV1CK = 0 LD (PWMCR1B), 00H ; Select the lower 8 bits of PWM0 output data latch LD (PWMDBR1), 80H ; 32 μs ÷ 4/fc = 80H Select the upper 6 bits of PWM0 output data latch LD (PWMCR1B), 01H ; LD (PWMDBR1), 00H ; No additional pulse = 00H LD (PWMCR1B), 02H ; Select the lower 8 bits of PWM0 output data latch LD (PWMDBR1), 40H ; 16 μs ÷ 4/fc = 40H LD (PWMCR1B), 03H ; Select the upper 4 bits of PWM0 output data latch LD (PWMDBR1), 01H ; Additional pulse (Ts (8)) = 01H LD (PWMCR1A), 0DH ; Start PWM0 and PWM1 , PWM0 : 14-bit resolution, PWM1 : 12-bit resolution LD (PWMCR2B), 00H ; Select PWM4 output data latch LD (PWMDBR2), 20H ; 8 μs ÷ 2/fc = 20H LD (PWMCR2A), 01H ; Start PWM4 88CS38B-138 2007-09-12 TMP88CS38B/CM38B/CP38B 2.14 Test Video Signal Output for Adjusting TV Screen The TMP88CS38B/CM38B/CP38B has a built-in video signal output circuit to output necessary signal for TV screen adjustment. 2.14.1 Picture pattern : Total eight types, monochromatic inversion possible Output format : Three states (H, L, High-Z) output Comp.sync duration time L output Black level/pedestal duration time High-Z output White level duration time H output Configuration Horizontal pattern generation circuit Pattern mixed circuit Vertical pattern generation circuit P62 (CSOUT) Display pattern generation circuit 3 SGIV SGPAT2 to 0 SGVBLK SGPAL TVSCR Test video signal output control register Figure 2.14.1 Test Video Signal Output Circuit 88CS38B-139 2007-09-12 TMP88CS38B/CM38B/CP38B 2.14.2 Control The test video signal output circuit can be controlled with the test video signal control register. TVSCR (00FE6H) 7 6 5 SGEN SGVBLK SGPAL SGEN SGVBLK SGPAL 4 3 2 1 0 SGIV SGCHS “0” SGPAT “0” SG function selection 0: Disable 1: Enable Picuture signal for VBLK duration time 0: Output 1: No output PAL/NTSC selection 0: NTSC 1: PAL No inversion SGIV Pattern monochromatic inversion 0: 1: Inversion SGCHS OSD synchronous signal selection 0: Port 1: Pseudo signal circuit SGPAT Display pattern (Initial value: 0000 0000) Write only 000: Black on the whole screen 001: White on the whole screen 010:Cross hatch 011: Cross dot pattern 100: Cross bar 101: White on the upper side/black on the lower side 110:H signal pattern 111: H resolution pattern Note 1: Test video signal output function does work correctly when fc is not 16 MHz. Note 2: Clear the bit2 and bit0 of TVSCR to “0”. Figure 2.14.2 Test Video Signal Control Register 88CS38B-140 2007-09-12 TMP88CS38B/CM38B/CP38B 2.14.3 Functions Video signal output is to generate monochromatic picture signal output to take easily the necessary tests such as TV screen white adjustment and screen distortion amplitude adjustment implemented on the final manufacturing process of a TV receiver set. Table 2.14.1 Display Pattern and TV Screen Display Pattern TV Screen 000 (Black on the whole surface) 001 (White on the whole surface) 010 (Cross hatch) 011 (Cross dot) 100 (Cross bar) 101 (White on the upper side/ black on the lower side) 110 (H signal pattern) 111 (H resolution pattern) 88CS38B-141 2007-09-12 TMP88CS38B/CM38B/CP38B There are three states of the output to generate picture signal with the external circuit of the resistance divided voltage. Example of picture output generation) 5V TMP88CS38B/ CM38B/CP38B P62 (CSOUT) to video input GND Three state of the output Picture signal output (1 Vp-p) (5 Vp-p) 88CS38B-142 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15 On-screen Display (OSD) Circuit The TMP88CS38B/CM38B/CP38B features a built-in on-screen display circuit used to display characters and symbols on the TV screen. There are 384 characters and any characters can be displayed in an area of 32 columns × 12 lines (Include 2 columns for solid space). With an OSD interrupt, additional lines can be displayed. The functions of the OSD circuit meet the requirements of on-screen display functions of closed caption decoders based on FCC standards. OSD circuit functions are as follows: (1) Number of character fonts: 384 (2) Number of display characters: 384 (32 columns × 12 lines) (3) Composition of character: Horizontal 16 × vertical 18 dots (4) Character sizes: 3 kinds for large, middle and small characters (Selectable line by line) (5) Character ornamentation function Fringing function Smoothing function Slant function (Italics) Blinking function Underline (6) Solid space (7) Area plane function: 2 planes (8) Full-raster blanking function (9) Display colors Character colors: 8 or 15 colors (Selectable character by character) Fringe color: 8 or 15 colors (Selectable page by page) Background color: 8 or 15 colors (Selectable page by page) Area plane color: 8 or 15 colors (Selectable each of 2 planes) Raster color: 8 or 15 colors (Selectable page by page) (10) Display position: 256 horizontal steps and 512 vertical steps for code plane : 512 horizontal steps and 512 vertical steps for area plane (11) Window function: 512 vertical steps (12) Half transparency output function The TMP88CS38B/CM38B/CP38B outputs OSD through 3 planes; code, area, and raster. 3 planes function independently. In addition, they are displayed simultaneously. There is the priority among these 3 planes, so they are displayed on a screen according to the priority. These 3 planes have the priority such as Code > Area > Raster. 88CS38B-143 2007-09-12 TMP88CS38B/CM38B/CP38B 1. Code plane OSD character is displayed on the code plane. The code plane consists of 32 characters × 1 row and a total of 12 planes. The 12 planes have the priority such as code 1 > code 2 > ... > code 11 > code 12. On the code plane, characters of 16 × 18 dots is displayed. These fonts are called characters, and read from character ROM and display memory through the character code on the display memory. 2. Area plane The area on a screen is displayed on the area plane. The area plane can display 2 square areas of any size by specifying coordinates. The 2 planes have the priority such as area plane 1 > area plane 2. 2.15.1 Configuration OSC1 Oscillation circuit for OSD display OSC2 Horizontal position counter OSD interrupt OSD control Horizontal position decoder Display memory 32 × 12 × 16 bits Character code Vertical position counter Character ROM P71 ( VD ) 384 × 16 × 18 bits Vertical position decoder P60 (Y/BLIN) P61 (BIN) P62 (GIN) P63 (RIN) Character data I Y/BL B G R P57 (I) Output signal selector Jitter elimination circuit Display output control P70 ( HD ) P67 (Y/BL) P66 (B) P65 (G) P64 (R) Figure 2.15.1 OSD Circuit 88CS38B-144 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.2 Character ROM and Display Memory (1) Character ROM The character ROM contains 384 character fonts. The user can set fonts as desired. The character ROM consists of 384 characters in 16 × 18 dots (Character codes 000H to 17FH). Each dot corresponds to one bit in the character ROM. When a bit in the character ROM is set to “1”, the corresponding dot is displayed; if set to “0”, the dot is not displayed. The start address in the character ROM corresponding to a character code is determined by the following expression: Start address in character ROM = CRA × 40H + 20000H Since character code 000H is used as blank character, the character font for this character code cannot be changed. Write “0” in the data of character code 000H. Write the data “FFH” to all unused address (5th bit of an address is “1” and also the lower 4 bits of an address are 2H to FH) in character ROM. Figure 2.15.2 (a) shows an example of the character font configuration for the character code 000H and 001H, together with the ROM addresses and data. Figure 2.15.2 (b) shows the character ROM dump list for these 2 character fonts. Note 1: CRA: Character code (000H to 17FH). Note 2: A data can not be read from character ROM by software. Note 3: When ordering a mask, load the data to character ROM at addresses 20000H to 25FFFH. And the data in unused are of character ROM are must be specified to FFH. Address Data (Hex) (Hex) 20000 00 20001 00 20002 00 20003 00 20004 00 20005 00 20006 00 20007 00 20008 00 20009 00 2000A 00 2000B 00 2000C 00 2000D 00 2000E 00 2000F 00 20010 00 20011 00 Bit Bit 7 6 5 4 3 2 1 07 6 5 4 3 2 1 0 Address Data (Hex) (Hex) 20020 00 20021 00 20022 00 20023 00 20024 00 20025 00 20026 00 20027 00 20028 00 20029 00 2002A 00 2002B 00 2002C 00 2002D 00 2002E 00 2002F 00 20030 00 20031 00 Address Data (Hex) (Hex) 20040 3F 20041 7F 20042 E0 20043 C0 20044 00 20045 00 20046 00 20047 01 20048 03 20049 07 2004A 0E 2004B 1C 2004C 38 2004D 70 2004E FF 2004F FF 20050 00 20051 00 Bit Bit 7 6 5 4 3 2 1 07 6 5 4 3 2 1 0 (Character code 000H) Address Data (Hex) (Hex) 20060 C0 20061 E0 20062 70 20063 30 20064 30 20065 70 20066 E0 20067 C0 20068 80 20069 00 2006A 00 2006B 00 2006C 00 2006D 00 2006E F0 2006F F0 20070 00 20071 00 (Character code 001H) (a) Character font configuration 20000/ 20010/ 20020/ 20030/ 20040/ 20050/ 20060/ 20070/ 00 00 00 00 3F 00 C0 00 00 00 00 00 7F 00 E0 00 00 FF 00 FF E0 FF 70 FF 00 FF 00 FF C0 FF 30 FF 00 FF 00 FF 00 FF 30 FF 00 FF 00 FF 00 FF 70 FF 00 FF 00 FF 00 FF E0 FF 00 FF 00 FF 01 FF C0 FF 00 FF 00 FF 03 FF 80 FF 00 FF 00 FF 07 FF 00 FF 00 FF 00 FF 0E FF 00 FF 00 FF 00 FF 1C FF 00 FF 00 FF 00 FF 38 FF 00 FF 00 FF 00 FF 70 FF 00 FF 00 FF 00 FF FF FF F0 FF 00 FF 00 FF FF FF F0 FF (b) ROM dump list Note: Shared portions indicate unused data. Figure 2.15.2 Character Font Configuration and ROM Dump List 88CS38B-145 2007-09-12 TMP88CS38B/CM38B/CP38B (2) Display memory Each character of the 384 characters displayed in 32 columns × 12 lines consists of 16 bits in the display memory. Five data items are written to the display memory: Character code, color data, blinking specification, underline enable, and slant enable. There are two modes for writing display data to the display memory. One mode is used for writing all display data (Character code, color data, blinking specification, underline enable, and slant enable) simultaneously. The other mode is used for changing either character codes or the remaining data items (Color data, blinking specification, underline enable, and slant enable). How to write display data to the display memory is described in section 2.15.5.7 (1). Note: The display memory is in an unknown state at reset. Display memory configuration SLNT EUL • Character code specification register (9 bits).... CRA8 to CRA0 • Color data specification register (4 bits) ........... IDT/RDT/GDT/BDT • Blinking specification register (1 bit) ................ BLF • Underline enable register (1 bit)........................ EUL • Slant enable register (1 bit) ............................... SLNT BLF IDT RDT GDT BDT CRA8 CRA7 CRA6 Character color specification register CRA5 CRA4 CRA3 CRA2 CRA1 CRA0 28 30 Character code specification register Blinking specification register Underline enable register Slant enable register Figure 2.15.3 Display Memory Bit Configuration Column 1 Line 1 2 3 4 5 6 7 8 9 10 11 12 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 29 31 32 000 001 002 003 004 005 006 007 008 009 00A 00B 00C 00D 00E 00F 010 011 012 013 014 015 016 017 018 019 01A 01B 01C 01D 01E 01F 020 021 022 023 024 025 026 027 028 029 02A 02B 02C 02D 02E 02F 030 031 032 033 034 035 036 037 038 039 03A 03B 03C 03D 03E 03F 040 060 080 0A0 0C0 0E0 100 120 140 160 17F Note: Numerals in the table indicate (hexadecimal) addresses in the display memory. Figure 2.15.4 Display Memory Address Configuration 88CS38B-146 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.3 OSD Circuit Control The OSD circuit performs control functions using the OSD control registers which reside in addresses 0001DH to 0001FH and 00024H to 00025H in the special function registers (SFR), and in addresses 0F80H to 0FBFH in the data buffer register (DBR). Section 2.15.5.9 shows the OSD control registers. The OSD control registers are used to set display start position, display character designs (that is, fringing, smoothing, color data, character size, and etc.), display memory addresses, and character codes. Setting the display on-off control bit, DON, (Bit0 in ORDON) to “1” enables display (Starts display). Setting DON to “0” disables display (Halts display). Note: The contents of OSD control registers except PIDS, P67S to P64S are initialized in STOP mode. 2.15.4 OSD Control Register Write There are lists of the OSD control registers on Table 2.15.5.10 and Table 2.15.5.11. When data is written into a shaded register, the data is transferred to the OSD circuit, and then the data becomes valid. After data is written into an unshaded register, the data is transferred to the OSD circuit, and then the data becomes valid. To transfer the contents of a control register to the OSD circuit, use data transfer request register RGWR (Bit2 in ORDON). Setting “1” in the RGWR register outputs the transfer request signal to the OSD circuit. Three instruction cycles later, transfer of the written data to the OSD circuit starts. While the data is being transferred, data transfer status monitoring flag RGWR (Bit2 in ORDON) is “1”. When this transfer is completed, the flag is cleared to “0”. Written data transfer register (1 bit) ....RGWR (Bit2 in ORDON) “0” ....... Initialized state Transfers written data to OSD circuit. “1” ....... (After transfer, RGWR is reset to 0.) Note: Don’t write “0” to RGWR. 88CS38B-147 2007-09-12 TMP88CS38B/CM38B/CP38B (1) RGWR system OSD circuit Q D Transfer pulse by RGWR = 1 LE Register specified by RGWR Figure 2.15.5 RGWR System (2) Transfer timing 1. No display area When having set RGWR to “1” during no display area, the timing OSD register can be transferred is at the falling edge of HD signal. HD RGWR register Set RGWR register to “1” Clear RGWR Data transfer pulse Transfer the contents of OSD registers into OSD circuit Figure 2.15.6 Data Transfer Timing in No Display Area 2. Display area (including any lines specified as display off by character size) When having set RGWR to “1” during display area, the timing OSD register can be transferred is at the falling edge of HD signal when the display line has been finished. HD Display line RGWR register Set RGWR register to “1” Clear RGWR Data transfer pulse Transfer the contents of OSD registers into OSD circuit Figure 2.15.7 Data Transfer Timing in Display Area 88CS38B-148 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.5 OSD Function 2.15.5.1 Signal Control (Port I/O) (1) P6 port output select function This function is used to select whether the contents of port P57, P67 to P64 will be output or I, R, G, B, Y/BL signals of the OSD circuit will be output on pins P57, P67 to P64. P57 port output select registers (1 bit): PIDS (Bit3 in ORP6S) PIDS = 0 PIDS = 1 I Port P57 P67 to P64 port output select registers (4 bits): P67S, P66S, P65S, P64S, (Bit7 to 4 in ORP6S) P6nS = 0 P64 R P65 G P66 B P67 Y/BL P6nS = 1 Port (2) OSD pin output polarity control function This function is used to select the polarity of the OSD outputs for RGB, I and Y/BL. Output polarity control register (4 bits) ...... BLIV, YIV, RGBIV, IIV (Bit3 to 0 in ORIV) “0” ...... Active high “1” ...... Active low (3) OSD pin input polarity control Input polarity control Input polarity control register of RIN/GIN/BIN/Y/BLIN (2 bits) For Y/BLIN .......................... YBLII (Bit5 in ORIV) For RIN, GIN, and BIN ...... RGBII (Bit4 in ORIV) Input polarity control YBLII, RGBII “0” “1” ...... ...... Active high Active low Input polarity control register of HD / VD (2 bits) For VD ..............VDPOL (Bit7 in ORIV) For HD .............HDPOL (Bit6 in ORIV) Input polarity control VDPOL, HDPOL “0” ....... Not invert input signal “1” ....... Invert input signal 88CS38B-149 2007-09-12 TMP88CS38B/CM38B/CP38B Register setting for the following waveform Input waveform to P70, P71 P71 ( VD ) VDPOL = 0 P70 ( HD ) HDPOL = 0 P71 ( VD ) VDPOL = 1 P70 ( HD ) HDPOL = 0 P71 ( VD ) VDPOL = 0 P70 ( HD ) HDPOL = 1 P71 ( VD ) VDPOL = 1 P70 ( HD ) HDPOL = 1 Figure 2.15.8 VD / HD input and VDPOL/HDPOL (4) Y/BL signal select function This function is used to select either Y or BL signal output from the Y/BL pin. Y/BL signal select register (1 bit) .........YBLCS (Bit7 in ORP6S) “0” ....... Y signal output “1” ....... BL signal output Y signal.............. Output in all OSD areas (Logical OR for R, G, B data as character data, fringing data, area data, etc.) BL signal ........... When EXBL is “0”: Output in all display character areas (except for character code 000H: Blank character) When EXBL is “1”: Output in the whole page (5) I signal function select When PISEL (Bit6 in ORETC) is set to “1” and PIDS (Bit3 in ORP6S) is set to “0”, Port 57 (I pin) can be used as half transparency/half tone through an extra circuit. At half transparency/half tone function, contents of IDT (Bit3 in ORDSN) is make no sense. Therefore character color are limited to 8 colors. Similarly background color, fringing color, raster plane color and area plane color are limited to 8 colors. When PISEL (Bit6 in ORETC) sets to “0” and, PIDS (Bit3 in ORP6S) set to “0”, 15 colors to be selectable. 88CS38B-150 2007-09-12 TMP88CS38B/CM38B/CP38B (6) R, G, B, Y/BL Internal/external signal select. Selects either R, G, B, and Y/BL signals from the internal OSD circuit, or RIN, GIN, BIN, and Y/BLIN signals from external input. R, G, B, Y/BL signal select registers (2 bits)........ MPXS1/MPXS0 (Bits 1 and 0 in ORP6S) ....... Simultaneous output (Signal from the OSD circuit “00” has higher priority.) “01” ....... Output of signal from internal OSD circuit “10” ....... Output of signal from external input ....... Simultaneous output (External input signal has “11” higher priority.) 88CS38B-151 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.5.2 OSD Data Output Format Control (1) Scan mode The double scan mode is used to handle non-interlaced scanning TV. When double scan mode is enabled, the vertical display counter increases every 2 scan lines and a vertical size of a dot is double. This function is enabled by setting VDSMD (Bit7 in ORETC) in the OSD control register to “1”. Scan mode select register (1 bit)....... VDSMD (Bit7 in ORETC) “0” “1” ....... Normal mode ....... Double scan mode Note 1: The data written to those control register is transferred to the OSD circuit and become valid when the data is written. Note 2: When OSD circuit is used on an interlace scanning TV, a jitter elimination circuit must be enabled and set AFLD to “1” in JECR. Table 2.15.5.1 The Difference of 2 Types of Scan Mode Normal Mode Double Scan Mode Specification unit of vertical display start position One scanning line Two scanning lines 1 dot height − Normal mode height × 2 Normal mode Double scan mode Normal mode Interlace scanning Double scan mode Non-interlace scanning Figure 2.15.9 Scan Mode 88CS38B-152 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.5.3 Display Position Control (1) Code display position setting 1. Horizontal display start position The horizontal display start position can be set in 256 steps by writing to OSD control registers HS17 to HS10 (Bit7 to 0 in ORHS1). The value is in common with all lines. Specification unit: 2 TOSC Specification steps: 256 Specification horizontal display start position: Line 1 to 12: HS17 to HS10 (ORHS1) HS1 = (HS17 to HS10) H × 2TOSC + 20TOSC (Line1 to 12) Note 1: TOSC: One cycle of OSD oscillation. Note 2: The data written to these control registers is transmitted to OSD circuit by setting RGWR (Bit2 in ORDON) to “1”. 2. Vertical display start position The vertical display start position can be specified for each display line using 512 steps by writing to VSn8 to VSn0 (in ORVSn (n: 1 to 12)). Specification unit: 1 scan line Specification steps: 512 Specification vertical display start position: Line1: VS18 to VS10 (ORVS 1) Line2: VS28 to VS20 (ORVS 2) .. . Line12: VS128 to VS120 (ORVS 12) Line n: VSn = (VSn8 to VSn0) H × 1THD (n: 1 to 12) Note 1: THD: One cycle of HD signal. Note 2: The data written to these control registers is transmitted to OSD circuit by setting RGWR (Bit2 in ORDON) to “1”. Note 3: If display lines are overlapped each other, previous display line is enabled and next line is disabled. If vertical display start positions of two or more lines are set on same value, high priority line is enabled. Lines of OSD (VS1 to VS12) are fixed priority levels as follows: VS1 > VS2 > VS3 > ...... > VS12 Set the vertical display start position not to overlap display lines. VS5 (Display on, small character) VS2 (Display canceled, middle character) VS3 (Display on, small character) Occasion of overlapping 88CS38B-153 2007-09-12 TMP88CS38B/CM38B/CP38B Note 4: The line which is displayed off is managed as a small size character line. Note 5: Transfer the contents of vertical display start position registers into OSD circuit before the position of the scanning line coincides with their own vertical display start position. (2) Area display position setting The planes have the priority such as Code plane > Area plane 1 > Area plane 2 > Raster plane. 1. Horizontal display start and end position The horizontal display start position can be set in 512 steps by writing to OSD control registers AHSn8 to AHSn0 (Bit8 to 0 in ORAHSn). And also display stop position is correspond to AHEn8 to AHEn0 (Bit8 to 0 in ORAHEn). (n: 1 to 2) Horizontal display start position AHSn = (AHSn8 to AHSn0)H × 2TOSC Horizontal display end position AHEn = (AHEn8 to AHEn0)H × 2TOSC Note 1: TOSC: One cycle of OSD oscillation. Note 2: If the horizontal display start position for characters is the same as that for areas, the two positions are not displayed at the same time. The horizontal display start position for characters is displayed 16 TOSC (Corresponding to a register value of 8) later than that for areas. 88CS38B-154 2007-09-12 TMP88CS38B/CM38B/CP38B 2. Vertical display start and end position The vertical display start position can be set in 512 steps by writing to OSD control registers AVSn8to AVSn0 (Bit8 to 0 ORAVSn). And also display stop position is correspond to AVEn8 to AVEn0 (Bit8 to 0 in ORAVEn). (n: 1 to 2) Vertical display start position AVSn = (AVSn8 to AVSn0) H × THD Vertical display end position AVEn = (AVEn8 to AVEn0) H × THD Note: THD: One cycle of HD signal. HD VS1 VS2 AVS2 AVE2 AVS1 AVE1 HS1 1 2 34 5 6 7 8 910 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Code plane 1 1 2 3 4 5 6 7 8 910 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Code plane 2 AHS1 VD Area plane 1 AHE1 Area plane 2 HS1 AHS2 AHE2 SS 1 2 3 4 5 6 7 8 910 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 SS Code plane 9 SS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 SS Code plane 10 SS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 SS Code plane 11 SS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 SS Code plane 12 Figure 2.15.10 TV Scan Image 88CS38B-155 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.5.4 Character Ornamentation Control (1) Character sizes Character size can be selected line by line from 3 sizes. And display on/off also can be set line by line. Small, middle and large character size and display on/off can be set with OSD control registers CSn (n: 1 to 12, ORCS4, ORCS8, ORCS12) in the OSD control registers. Character sizes: 3 sizes (Small, middle and large) Character size and display on/off specification unit: Line Character size select/display on/off register (2 bits × 12) Line 1: CS1 Line 2: CS2 : : Line 12: CS12 Table 2.15.5.2 Character Size and Display On/Off Specifications (n: 1 to 12) CSn (Upper bit) CSn (Lower bit) Character Size Display On/Off 1 1 Small On 1 0 Middle On 0 1 Large On 0 0 − Off Note 1: The display off line operates like the width of small character size line thought the character is not displayed. Note 2: The data written to these control registers is transmitted to OSD circuit by setting RGWR (Bit2 in ORDON) to “1”. Note 3: When OSD circuit is used on an interlace scanning TV, a jitter elimination circuit must be enabled and set AFLD to “1” in JECR. Note 4: When VDSMD and AFLD are “0”, only character of even display dot is displayed. (Refer to 2.16 a jitter elimination circuit.) Table 2.15.5.3 Dot and Character Sizes VDSMD = 0 (Normal mode) Dot Size EULAn = 0 (Underline off) EULAn = 1 (Underline on) Character Size EFRn = 0 EFRn = 1 (Fringe off) (Fringe on) VDSMD = 1 (Double scan mode) Dot Size Character Size EFRn = 0 EFRn = 1 (Fringe off) (Fringe on) Small 1TOSC × 0.5THD 16TOSC × 9THD 16TOSC × 11THD 1TOSC × 1THD 16TOSC × 18THD 16TOSC × 20THD Middle 2TOSC × 1THD 32TOSC × 18THD 32TOSC × 20THD 2TOSC × 2THD 32TOSC × 36THD 32TOSC × 40THD Large 4TOSC × 2THD 64TOSC × 36THD 64TOSC × 40THD 4TOSC × 4THD 64TOSC × 72THD 64TOSC × 80THD Small 1TOSC × 0.5THD 16TOSC × 12THD 16TOSC × 13THD 1TOSC × 1THD 16TOSC × 24THD 16TOSC × 25THD Middle 2TOSC × 1THD 32TOSC × 24THD 32TOSC × 25THD 2TOSC × 2THD 32TOSC × 48THD 32TOSC × 50THD Large 4TOSC × 2THD 64TOSC × 48THD 64TOSC × 50THD 4TOSC × 4THD 64TOSC × 96THD 64TOSC × 100THD TOSC: One cycle of OSD oscillation, THD: One cycle of HD signal 88CS38B-156 2007-09-12 TMP88CS38B/CM38B/CP38B Small Middle Large Figure 2.15.11 Character Size 88CS38B-157 2007-09-12 TMP88CS38B/CM38B/CP38B (2) Smoothing function The smoothing function is used to make characters look smooth. Enabling smoothing displays 1/4 dot between two dots connecting corner to corner within a character. Small size character can not be enabled smoothing. Smoothing is enabled by setting ESMZ (Bit4 in ORETC) in the OSD control register to “1”. Smoothing specification unit:Display page Smoothing specification register (1 bit)......ESMZ (Bit4 in ORETC) “0” “1” ....... Disable smoothing ....... Enable smoothing Note: Data of the register is transferred to the OSD circuit and become valid when the data is written. Before After Before After Available form for smoothing Invalid form for smoothing Figure 2.15.12 Available Form and Invalid Form for Smoothing Original character Smoothing Figure 2.15.13 Smoothing Example 88CS38B-158 2007-09-12 TMP88CS38B/CM38B/CP38B (3) Fringing function The fringing function is used to display a character with a fringe width is 1 dot in a different color from that of the character. When a character is displayed with the maximum of 18 vertical dots and 16 horizontal dots, the fringe exceeds right and left, top, and bottom of the character display area. If there is an adjacent character that outer dot is active, then this dot will overrule the fringe in the horizontal direction. Underlines are not fringed. Fringing is enabled for each line by setting EFR1 to EFR8 (OREFR8) and EFR9 to EFR12 (OREFR12) in the OSD control register to “1”. A color for fringe is specified common to all lines using OSD control registers, IFDT, RFDT, GFDT, and BFDT (Bit3 to 0 in ORBK). Fringing specification unit: Line Fringing enable register (1 bit × 12) .... EFRn (n: 1 to 8) (OREFR8), EFRn (n: 9 to 12) (OREFR12) “0” ....... Disable fringing “1” ....... Enable fringing Fringe colors: 8 or 15 Fringe color specification unit: Display page Fringe color register (4 bits).....IFDT, RFDT, GFDT, BFDT (Bit3 to 0 in ORBK) I signal function select: PISEL (Bit6 in ORETC) “0” “1” ....... 15 colors specification I pin can be used to make a half level of R, G, B signal (Dark color) through an extra circuit. ....... 8 colors specification Contents of IDT register is disregarded. I pin can be used as half transparency/half tone through an extra circuit. Note: The fringe of 1st column character does not exceed left, and the fringe of 32th character does not exceed right. 88CS38B-159 2007-09-12 TMP88CS38B/CM38B/CP38B Table 2.15.5.4 Fringe Color (15 colors) IFDT 0 1 RFDT GFDT BFDT Figure Color Black 0 0 0 0 0 1 Blue 0 1 0 Green 0 1 1 Cyan 1 0 0 Red 1 0 1 Magenta 1 1 0 Yellow 1 1 1 White 0 0 0 Black 0 0 1 Dark blue 0 1 0 Dark green 0 1 1 Dark cyan 1 0 0 Dark red 1 0 1 Dark magenta 1 1 0 Dark yellow 1 1 1 Gray R, G, B pin output R, G, B output I pin output Figure 2.15.14 Example Circuit for 15 Colors by I Pin. 88CS38B-160 2007-09-12 TMP88CS38B/CM38B/CP38B Vertical indicate area 22 dots Vertical indicate area 18 dots 2 dots Before fringing After fringing Disable underline Vertical indicate area 26 dots Vertical indicate area 24 dots 2 dots Before fringing After fringing Enable underline a) Small character, NORMAL mode Figure 2.15.15 (a) Fringing Example 88CS38B-161 2007-09-12 TMP88CS38B/CM38B/CP38B Vertical indicate area 20 dots Vertical indicate area 18 dots 1 dot Before fringing After fringing Disable underline Vertical indicate area 25 dots Vertical indicate area 24 dots 1 dot Before fringing After fringing Enable underline b) Small character, double scan mode Figure 2.15.16 (b) Fringing Example 88CS38B-162 2007-09-12 TMP88CS38B/CM38B/CP38B Vertical indicate area 20 dots Vertical indicate area 18 dots 1 dot Before fringing After fringing Disable underline Vertical indicate area 25 dots Vertical indicate area 24 dots 1 dot Before fringing After fringing Enable underline c) Middle/Large character, NORMAL mode Figure 2.15.17 (c) Fringing Example 88CS38B-163 2007-09-12 TMP88CS38B/CM38B/CP38B Vertical indicate area 20 dots Vertical indicate area 18 dots 1 dot Before fringing After fringing Disable underline Vertical indicate area 25 dots Vertical indicate area 24 dots 1 dot Before fringing After fringing Enable underline d) Middle/Large character, double scan mode Figure 2.15.18 (d) Fringing Example 88CS38B-164 2007-09-12 TMP88CS38B/CM38B/CP38B (4) Background function Background color function is used to color the entire background for the character area (Refer to Table 2.15.4). Except the character area whose character code is 000H. This function is specified for each display page by setting EBKGD (Bit7 in ORRCL) in the OSD control register to “1”. A background color is specified for each display page by setting IBDT, RBDT, GBDT, and BBDT (Bit7 to 4 in ORBK) in the OSD control registers. Background specification unit: Display page Background enable register (1 bit) ...... EBKGD (Bit7 in ORRCL) “0” ....... Disable background “1” ....... Enable background Background color specification unit: Display page Background color specification registers (4 bits) .... IBDT, RBDT, GBDT, BBDT (Bit7 to 4 in ORBK) I signal function select: PISEL (Bit6 in ORETC) “0” ....... 15 colors specification I pin can be used to make a half level of R, G, B signal (Dark color) through an extra circuit. “1” ....... 8 colors specification Contents of IBDT register is disregarded. I pin can be used as half transparency/half tone through an extra circuit. Table 2.15.5.5 Background Color (15 colors) IBDT 0 1 RBDT GBDT BBDT Background Color Black 0 0 0 0 0 1 Blue 0 1 0 Green Cyan 0 1 1 1 0 0 Red 1 0 1 Magenta 1 1 0 Yellow 1 1 1 White 0 0 0 Black 0 0 1 Dark blue 0 1 0 Dark green Dark cyan 0 1 1 1 0 0 Dark red 1 0 1 Dark magenta 1 1 0 Dark yellow 1 1 1 Gray 88CS38B-165 2007-09-12 TMP88CS38B/CM38B/CP38B R, G, B pin output R, G, B output I pin output Figure 2.15.19 Example Circuit for 15 Colors by I Pin. Character color: Cyan Background color: Yellow Scanning line Scanning line R R G G B B Y Y BL BL 1) Disable background 2) Enable background Figure 2.15.20 Background Function Note: When the background function is enabled, the line enable the fringing function should not start with a blank character. If it starts with a blank character, a fringe is displayed to the left of the blank character. 88CS38B-166 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.5.5 OSD Display Screen Control (1) Display on/off This function is used to display characters specified for on/off display. Display on/off specification unit: Display page Display on/off specification register (1 bit) ...... DON (Bit0 in ORDON) “0” ....... Disable display “1” ....... Enable display Note: Do not start STOP mode during display is enable. (2) Window function This function is used to set upper and lower limit of display page. Window upper limit is specified by WVSH (ORWVSH). Window lower limit is specified by WVSL (ORWVSL). This function is enabled by setting EWDW (Bit1 in ORDON) in the OSD control register to 1. Window specification unit: Display page Window function enable specification register (1 bit)...... EWDW (Bit1 in ORDON) “0” ....... Disable window function “1” ....... Enable window function Window upper limit specification register (9 bits) .....WVSH8 to 0 (ORWVSH) Window lower limit specification register (9 bits)......WVSL8 to 0 (ORWVSL) Window upper and lower limit position ...... When VDSMD is “0” (Normal mode): WVSH = (WVSH8 to WVSH0) H × THD WVSL = (WVSL8 to WVSL0) H × THD When VDSMD is “1” (Double scan mode): WVSH = (WVSH8 to WVSH0) H × 2THD WVSL = (WVSL8 to WVSL0) H × 2THD Note 1: THD: One cycle of HD signal Note 2: WVSL > WVSH ≥ “1” Note 3: Modify the value of window upper and lower limit register and the value of EWDW during VD signal is low. Note 4: It is recommendable that the window function is always enabled (EWDW = “1”) and set WVSH to “01H”, WVSL to “1FEH”. Note 5: Characters and symbols at scanning line specified by WVSL are not displayed. 88CS38B-167 2007-09-12 TMP88CS38B/CM38B/CP38B HD Background color AVE2 Area plane color Picture WVSH Raster color SS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 SS AHS1 SS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 SS VD SS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 SS WVSL Picture Window display: On, Area plane display: On, Background color display: On, Raster plane display: On Correspond to Closed Caption Display off WVSH Display Figure 2.15.21 If WVSH is on a Code Plane 88CS38B-168 2007-09-12 TMP88CS38B/CM38B/CP38B (3) Full-raster blanking function Full-raster blanking function is used to color the entire background for the display area (TV screen). When using the full-raster blanking function, set YBLCS (Bit2 in ORP6S) to “1”, output BL signal from Y/BL pin, because Y signal cannot delete whole display page from video signal. This function is specified for each display page by setting EXBL (Bit6 in ORRCL) in the OSD register to “1”. Full-raster blanking specification unit: Display page Full-raster blanking enable register (1 bit) ..... EXBL (Bit6 in ORRCL) “0” ....... Disable full-raster blanking “1” ....... Enable full-raster blanking Full-raster blanking color specification ........ RCLI, RCLR, RCLG, RCLB (Bit3 to 0 in ORRCL) registers (4 bits) I signal function select: PISEL (Bit6 in ORETC) “0” “1” ....... 15 colors specification I pin can be used to make a half level of R, G, B signal (Dark color) through an extra circuit. ....... 8 colors specification Contents of RCLI register is disregarded. I pin can be used as half transparency/half tone through an extra circuit. Table 2.15.5.6 Raster Plane Color (15 colors) RCLI 0 1 RCLR RCLG RCLB Raster Plane Color 0 0 0 Black 0 0 1 Blue 0 1 0 Green Cyan 0 1 1 1 0 0 Red 1 0 1 Magenta 1 1 0 Yellow 1 1 1 White 0 0 0 Black 0 0 1 Dark blue 0 1 0 Dark green Dark cyan 0 1 1 1 0 0 Dark red 1 0 1 Dark magenta 1 1 0 Dark yellow 1 1 1 Gray 88CS38B-169 2007-09-12 TMP88CS38B/CM38B/CP38B (4) Area plane function Area plane function is used to display square area to two points on a screen. Two planes operate independently. They are displayed according to the priority (Area plane 1 > Area plane 2). See area plane display position setting in section 2.15.5.3 (2) how to set display positions for each area. Each area plane is set to ON or OFF by AON2 and AON1 (Bit5 and bit4 in ORRCL). Area plane colors are set by ACLIx, ACLRx, ACLGx, ACLBx (Bit7 to bit0 in ORACL, x: 1, 2). Area plane colors: 8 or 15 Area plane specification unit: plane Area plane color specification register (8 bits) Area plane 1: ACLI1/ACLR1/ACLG1/ACLB1 (Bit3 to 0 in ORACL) Area plane 2: ACLI2/ACLR2/ACLG2/ACLB2 (Bit7 to 4 in ORACL) I signal function select: PISEL (Bit6 in ORETC) “0” ....... 15 colors specification I pin can be used to make a half level of R, G, B signal (Dark color) through an extra circuit. “1” ....... 8 colors specification Contents of ACLI1 and ACLI2 register is disregarded. I pin can be used as half transparency/half tone through an extra circuit. Table 2.15.5.7 Area Plane Color (15 colors) ACLIx 0 1 ACLRx ACLGx ACLBx Area Plane Color Black 0 0 0 0 0 1 Blue 0 1 0 Green Cyan 0 1 1 1 0 0 Red 1 0 1 Magenta 1 1 0 Yellow 1 1 1 White 0 0 0 Black 0 0 1 Dark blue 0 1 0 Dark green Dark cyan 0 1 1 1 0 0 Dark red 1 0 1 Dark magenta 1 1 0 Dark yellow 1 1 1 Gray x: 1, 2 88CS38B-170 2007-09-12 TMP88CS38B/CM38B/CP38B I signal function select 1. Using for 15 colors (PISEL = 0) Example color Area plane l=0 l=1 Red Dark red Character background color Green Dark green Area plane color Blue Dark blue Character color Scanning line Raster plane: Off Character background: On YBLCS: 0 (Y select) R 15 colors specification G B I Y Figure 2.15.22 TV Display and OSD Signals (PISEL = 0) 88CS38B-171 2007-09-12 TMP88CS38B/CM38B/CP38B Using for half transparency/half tone (PISEL = 1) 2. Area plane: Half transparency/half tone Scanning line Example color Character color Red Character background color Green Area plane color Blue Raster plane: Off Character background: On YBLCS: 0 (Y select) R 8 colors specification G B I Y Figure 2.15.23 TV Display and OSD Signals (PISEL = 1) 88CS38B-172 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.5.6 Interrupt Control (1) Display line counter The display line counter indicates number of display line (s) by OSD circuit on the TV screen. The display line counter is a 4-bit counter which is initialized to “0” by the falling edge of the VD signal and which increments when last scanning of each display line is completed (Falling edge of the HD signal). It is necessary to be read out display line counter several times, because it does not synchronize CPU clock. Display line counter register (4 bits)..... DCTR (Bit3 to 0 in ORIRC) “0000”......No display line is completed. “0001”......1st display line is completed. “0010”......2nd display line is completed. . . .. .. “1111” ..... 15th display line is completed. Display line counter 1st Display Line Display on 2nd Display Line Display off 3rd Display Line Display on 4th Display Line with all blank characters Display on ••• 10th Display Line Display on 11th Display Line Display on 12th Display Line Display on 12 11 10 9 4 3 2 1 0 m VD signal Note 1: The display line counter also increments when a line with all blank characters or a line with display off is specified. Note 2: When display lines are overlapped each other, previous display line is enabled and next line is canceled. At this time, the display line counter does not increment for canceled line. Figure 2.15.24 Display Line Counter 88CS38B-173 2007-09-12 TMP88CS38B/CM38B/CP38B (2) Interrupt generator circuit An interrupt request is generated when a falling edge of VD signal or when line counter (DCTR) is counted to the certain value specified by ISDC. Interrupt source select register (1 bit) ...SVD (Bit4 in ORIRC) “0” ....... Interrupt request generated when the display line counter (DCTR) is counted to the certain value which is specified by ISDC. ....... Interrupt request is generated when a falling edge of VD signal. “1” Interrupt generation line specification register (4 bits) ... ISDC (Bit3 to 0 in ORIRC) “0000” “0001” “0010” . . . “1111” ....... Interrupt request generated when the display line counter is cleared. ....... Interrupt request generated at end points of the last scanning line of the first display line ....... Interrupt request generated at end points of the last scanning line of the 2nd display line ....... Interrupt request generated at end points of the last scanning line of the 15th display line 2.15.5.7 Display Memory Access (1) Display memory The display memory is accessed for two purposes, one for writing data to the display memory, and one for reading data from the display memory. Display memory address specification registers .... DMA8 to MDA0 (ORDMA) (9 bits) Display memory data write registers Character code write register (9 bits) .... CRA8 to CRA0 (ORCRA) Character ornamentation data write registers .... SLNT, EUL, BLF, IDT, RDT, (7 bits) GDT, and BDT (ORDSN) Display memory bank select register MBK (bit 1 in ORETC) “0” ... When writing either character code or character ornamentation data “1” ... When writing both character code and character ornamentation data Note 1: These control registers have a characteristic that immediately when a value is written to the register, the content of the register is transferred as valid data to the OSD circuit/display memory. Note 2: The data written to the display memory takes effect at the same time it is written. When character code or character ornamentation data is written to the display memory while it is displaying some character, the character may not be displayed correctly. When writing data to the display memory, make sure no character is being displayed in the memory location where you are going to write data. Note 3: When writing data to or reading data from the display memory, do not use two-byte transfer instructions such as “LDW(HL),mn LD rr, (pp)”. Otherwise, erroneous data may be written to the display memory or data may be written to an incorrect address. Note 4: Allow for at least two instruction cycles between a display memory address write instruction and a data write or read instruction. Also, when continuous writing data to or reading data from the display memory, allow for at least two instruction cycles between one write or read instruction and the next. Otherwise, erroneous data may be written to the display memory or data may be written to an incorrect address. Note 5: When setting display memory addresses, always be sure to write all of 9 address bits sequentially in order of DMA8 and DMA7 to DMA0. 88CS38B-174 2007-09-12 TMP88CS38B/CM38B/CP38B 1. Normal mode In normal mode, the display memory addresses are automatically incremented each time data is read from or written to the memory. Because addresses are automatically incremented, this mode may be used for reading from or writing data to multiple continuous addresses simultaneously. <Display memory write sequence in normal mode> a. When writing either character code or character ornamentation data (1) Set MFYWR, MBK, and RDWRV all to 0. (2) Write the most significant address bit of the display memory to DMA8. Go on and write the 8 low-order address bits of the display memory to DMA7 to DMA0. (3) Writing character code or character ornamentation data • Writing character code Write the most significant bit of character code to CRA8. Go on and write the 8 low-order bits of character code to CRA7 through CRA0. At this point in time, the 9 bits of character code written are transferred to the display memory, and DMA8 to DMA0 are automatically incremented. • Writing character ornamentation data Write character ornamentation data to SLNT, EUL, BLF, IDT, RDT, GDT, and BDT. At this point in time, the character ornamentation data written are transferred to the display memory, and DMA8 to DMA0 are automatically incremented. (4) To write data (character code or character ornamentation data) to continuous addresses, repeat step (3). b. When writing character code and character ornamentation data at a time (1) Set MFYWR to 0, MBK to 1, and RDWRV to 0. (2) Write the most significant address bit of the display memory to DMA8. Go on and write the 8 low-order address bits of the display memory to DMA7 to DMA0. (3) Write character ornamentation data to SLNT, EUL, BLF, IDT, RDT, GDT, and BDT. At this point in time, the character ornamentation written are transferred to the display memory. (4) Write the most significant bit of character code to CRA8. Go on and write the 8 low-order bits of character code to CRA7 to CRA0. At this point in time, the 9 bits of character code written and the character ornamentation data written in step (3) are transferred to the display memory, and DMA8 to DMA0 are automatically incremented. (5) To write data to continuous addresses, repeat steps (3) and (4). 88CS38B-175 2007-09-12 TMP88CS38B/CM38B/CP38B <Display memory read sequence in normal mode> a. When reading either character code or character ornamentation data (1) Set MFYWR to 0, MBK to 0, and RDWRV to 1. (2) Write the most significant address bit of the display memory to DMA8. Go on and write the 8 low-order address bits of the display memory to DMA7 to DMA0. (3) Reading character code or character ornamentation data • Reading character code Read the most significant bit of character code to CRA8. Go on and read the 8 low-order bits of character code to CRA7 to CRA0. At this point in time, DMA8 to DMA0 are automatically incremented. • Reading character ornamentation data Read character ornamentation data SLNT, EUL, BLF, IDT, RDT, GDT, and BDT. At this point in time, DMA8 through DMA0 are automatically incremented. (4) To read data (character code or character ornamentation data) from continuous addresses, repeat step (3). b. When reading character code and character ornamentation data at a time (1) Set MFYWR to 0, MBK to 1, and RDWRV to 1. (2) Write the most significant address bit of the display memory to DMA8. Go on and write the 8 low-order address bits of the display memory to DMA7 to DMA0. (3) Read character ornamentation data SLNT, EUL, BLF, IDT, RDT, GDT, and BDT. (4) Read the most significant bit of character code to CRA8. Read the 8 low-order bits of character code to CRA7 to CRA0. At this point in time, DMA8 to DMA0 are automatically incremented. (5) To read data from continuous addresses, repeat steps (3) and (4). 2. Read-modify-write mode When writing data in read-modify-write mode, the display memory addresses are automatically incremented as in normal mode, but when reading data in this mode, the memory addresses are not automatically incremented. Therefore, immediately after executing a read from some display memory address, you can execute a write to the same display memory address. After executing a write, the display memory addresses are automatically incremented. a. Reading/writing either character code or character ornamentation data in read-modify-write mode (1) Set MFYWR to 1 and MBK to 0, and RDWRV to 1. (2) Write the most significant address bit of the display memory to DMA8. Go on and write the 8 low-order address bits of the display memory to DMA7 to DMA0. (3) Reading character code or character ornamentation data • Reading character code Read the most significant bit of character code to CRA8. Read the 8 low-order bits of character code to CRA7 to CRA0. DMA8 to DMA0 are not incremented. 88CS38B-176 2007-09-12 TMP88CS38B/CM38B/CP38B • Reading character ornamentation data Read character ornamentation data SLNT, EUL, BLF, IDT, RDT, GDT, and BDT. DMA8 to DMA0 are not incremented. (4) Writing character code or character ornamentation data • Set RDWRV to “0”. • Writing character code Write the most significant bit of character code to CRA8. Go on and write the 8 low-order bits of character code to CRA7 to CRA0. At this point in time, the 9 bits of character code written are transferred to the display memory, and DMA8 to DMA0 are automatically incremented. • Writing character ornamentation data Write character ornamentation data to SLNT, EUL, BLF, IDT, RDT, GDT, and BDT. At this point in time, the character ornamentation data written are transferred to the display memory, and DMA8 to DMA0 are automatically incremented. (5) To continue executing read-modify-write operations, repeat steps (1) to (4). To read/write data (Character code or character ornamentation data). To continue executing read-modify-write mode from continuous addresses, repeat steps (3) and (4). b. Reading/writing both character code and character ornamentation data in read-modify-write mode (1) Set MFYWR to 1, MBK to 1 and RDWRV to 1 (2) Write the most significant address bit of the display memory to DMA8. Go on and write the 8 low-order address bits of the display memory to DMA7 to DMA0. (3) Read character ornamentation data SLNT, EUL, BLF, IDT, RDT, GDT, and BDT. At this point in time, DMA8 to DMA0 are not incremented. (4) Read the most significant bit of character code to CRA8. Read the 8 low-order bits of character code to CRA7 to CRA0. At this point in time, DMA8 to DMA0 are not incremented. (5) Set RDWRV to “0”. (6) Write character ornamentation data to SLNT, EUL, BLF, IDT, RDT, GDT, and BDT. At this point in time, the character ornamentation data written is transferred to the display memory. (7) Write the most significant bit of character code to CRA8. Go on and write the 8 low-order bits of character code to CRA7 to CRA0. At this point in time, the 9 bits of character code written and the character ornamentation data written in step (6) are transferred to the display memory, and DMA8 to DMA0 are automatically incremented. (8) To continue executing read-modify-write operations, repeat steps (1) to (7). (To read/write data to and from continuous addresses in read-modify-write mode, repeat steps (3) to (7).) 88CS38B-177 2007-09-12 TMP88CS38B/CM38B/CP38B Table 2.15.5.8 Address Increment RD (RDWRV = 1) Character Character Ornamentation Code MFYWR = 0 MFYWR = 1 WR (RDWRV = 0) Character Character Ornamentation Code MBK = 0 INC INC INC MBK = 1 − INC − INC MBK = 0 − − INC INC MBK = 1 − − − INC INC INC: Automatic address increment at read or write. −: Example: 1. No address change at data read or write. Setting a character code (020H) to the display memory (Address: 120H) and setting (001H) for a character ornamentation. MBK = 0 ; Set display memory address LD (0x25), 0x01 ; ORDMA<DMA8> LD (0x24), 0x20 ; ORDMA<DMA7:0> ; Set character code LD (0x1F), 0x00 ; ORCRA<CRA8> LD (0x1E), 0x20 ; ORCRA<CRA7:0> ; Set display memory address again LD (0x25), 0x01 LD (0x24), 0x20 ; Set character ornamentation LD 2. (0x1D), 0X01 ; ORDSN<SLNT, ..... BDT> MBK = 1 ; Set display memory address LD (0x25), 0x01 LD (0x24), 0x20 ; Set character ornamentation LD (0x1D), 0X01 ; Set character code LD (0x1F), 0x00 LD (0x1E), 0x20 Note 1: To write character code into the display memory, first write into register CRA8 and then write into registers CRA7 to CRA0. When data is written into registers CRA7 to CRA0, DMA8 to DMA0 is incremented. It is impossible to write into the display memory for CRA7 to CRA0 alone. If no data is written into register CRA8 while data is written into registers CRA7 to CRA0, the value previously written into register CRA8 is written into the associated display memory. Note 2: To read character code from the display memory, first read from register CRA8, and then read from registers CRA7 to CRA0. When data is read from registers CRA7 to CRA0, DMA8 to DMA0 is incremented. Note 3: There should be a time interval of at least two machine cycles between a DMA set instruction and a data write/read instruction. There should be a time interval of at least two machine cycles between a data write instruction and a data read instruction. (2) Character Characters: 384 (including blank character) Character specification register (9 bits) ...........CRA8 to CRA0 (Bit8 to 0 in ORCRA) Character code “000H” ............................Blank character Character code “001H” to “017FH” ........User programmable by character ROM 88CS38B-178 2007-09-12 TMP88CS38B/CM38B/CP38B (3) Character color Character colors: 8 or 15 Character color specification unit: Character Character color specification register (4 bits): IDT/RDT/GDT/BDT (Bit3 to 0 in ORDSN) I signal function select: PISEL (Bit6 in ORETC) “0” ....... 15 colors specification I pin can be used to make a half level of R, G, B signal (Dark color) through an extra circuit. “1” ....... 8 colors specification Contents of IDT register is disregarded. I pin can be used as half transparency/half tone through an extra circuit. Table 2.15.5.9 Character Color (15 colors) IDT 0 1 RDT GDT BDT Character Color 0 0 0 Black 0 0 1 Blue 0 1 0 Green 0 1 1 Cyan 1 0 0 Red 1 0 1 Magenta 1 1 0 Yellow White 1 1 1 0 0 0 Black 0 0 1 Dark blue 0 1 0 Dark green 0 1 1 Dark cyan 1 0 0 Dark red 1 0 1 Dark magenta 1 1 0 Dark yellow 1 1 1 Gray R, G, B pin output R, G, B output I pin output Figure 2.15.25 Example of Circuit for 15 Color by I Pin 88CS38B-179 2007-09-12 TMP88CS38B/CM38B/CP38B (4) Blinking function Blinking function is used to blink display characters. When BKMF is “1”, characters specified for blinking by BLF are not displayed. (If the background color function is used, the background color is not disappeared.) Blinking specification unit: Character Blinking specification register (1 bit).....BLF (Bit4 in ORDSN) “0” ....... No blinking “1” ....... Blinking Blinking master specification register (1 bit).....BKMF (Bit5 in ORETC) “0” “1” Note: ....... Disable blinking ....... Enable blinking (Characters whose BLF are set to “1” are not displayed.) Regarding the extra dot of the left and/or right character by fringing function, it is not enabled as blink. (5) Underline function Underline function is used to add a line under a display character. The underline is same color as that of character. Underline specification unit: Character/line Underline enable register (Character unit) (1 bit).....EUL (Bit5 in ORDSN) “0” ....... No underline “1” ....... Underline Underline enable register (Line unit) (1 bit × 12) Underline colors: 8 or 15 Underline color specification registers (4 bits) Note: ......EULAn (n: 1 to 8)(OREULA8), EULAn (n: 9 to 12) (OREULA12) ......RDT, GDT, BDT, IDT (Bit3 to 0 in ORDSN) (Refer to Table 2.15.5.9) To use the underline function, set both the underline enable register for underlining text in characters and that for underlining text in lines. If the former register (EUL) only is set, an underline is not displayed. 16 18 Character display area 6 Underline display area 24 EUL = 0 EUL = 1 Figure 2.15.26 Underline 88CS38B-180 2007-09-12 TMP88CS38B/CM38B/CP38B (6) Solid space control Solid space control is used to display one column of solid space to the left and right of 32 columns. Solid space control is used to delete the video signal in the areas where solid spaces are located in the original display page, then add color to them. Solid space specification unit: line Solid space specification register (24 bits) For line 1 For line 2 .. . For line 12 SOL11 and SOL10 (Bits 1 and 0 in ORSOL4) SOL21 and SOL20 (Bits 3 and 2 in ORSOL4) .. . SOL121 and SOL120 (Bits 7 and 6 in ORSOL12) Solid space specification The solid space control functions as follows: SOLx1/SOLx0 (x: 1 to 12) “00” ....... “01” ....... “10” ....... “11” ....... No solid space display Solid space display left for 32 columns Solid space display right for 32 columns Solid space display left and right for 32 columns Solid space color specification registers (4 bits) ....... IBDT, RBDT, GBDT, BBDT (Bits 3 to 0 in ORBK) (Same color as that of background) 32 columns Solid space (Left) Solid space (Right) Figure 2.15.27 Solid Space 88CS38B-181 2007-09-12 TMP88CS38B/CM38B/CP38B (7) Slant function Slant function is used to slant characters for italics. Slant specification unit: Character Slant enable register (1 bit) .....SLNT (Bit6 in ORDSN) “0” “1” ....... No slant ....... Slant Note 1: SLANT function is enabled each characters, and therefore, in case of using background function, this color of the Background is enable as slant. Regarding the extra dots of the left and/or right character by fringing function, it is not enabled as slant. Note 2: When a character is slanted in an area, which overlaps with the character field, the overlap is also slanted. Note 3: If slanting a character causes part of the character to get into the character field to the immediate right of the character, then this part is not displayed. Note 4: To provide closed caption display (CCD), specify black as the background color, and set YBLCS to “1”. R, G, B and Y are all slanted. Thus, if the Y signal is selected, a video signal is displayed above and to the left of the slant character. Note 5: When a character is slanted, the dot data to the immediate left of the character is also slanted. The same color as that of the dot on the left is displayed. When an entire character field (including its background) contains dots: When the character field on the right does not contain a dots: Figure 2.15.28 Slant 88CS38B-182 2007-09-12 TMP88CS38B/CM38B/CP38B 2.15.5.8 OSD Control Registers Can not access all OSD control registers in any of read-modify-write instructions such as bit operation, etc. 0RHS1 7 6 5 4 3 2 1 0 (00F81H) HS17 HS16 HS15 HS14 HS13 HS12 HS11 HS10 (Initial value: 0000 0000) Write only Horizontal display start position specification 7 6 5 4 3 2 1 0 (00F82H) VS17 VS16 VS15 VS14 VS13 VS12 VS11 VS10 (Initial value: 0000 0000) (00F83H) − − − − − − − VS18 (Initial value: **** ***0) ORVS2 (00F84H) VS27 VS26 VS25 VS24 VS23 VS22 VS21 VS20 (Initial value: 0000 0000) (00F85H) − − − − − − − VS28 (Initial value: **** ***0) ORVS3 (00F86H) VS37 VS36 VS35 VS34 VS33 VS32 VS31 VS30 (Initial value: 0000 0000) (00F87H) − − − − − − − VS38 (Initial value: **** ***0) ORVS4 (00F88H) VS47 VS46 VS45 VS44 VS43 VS42 VS41 VS40 (Initial value: 0000 0000) (00F89H) − − − − − − − VS48 (Initial value: **** ***0) ORVS5 (00F8AH) VS57 VS56 VS55 VS54 VS53 VS52 VS51 VS50 (Initial value: 0000 0000) (00F8BH) − − − − − − − VS58 (Initial value: **** ***0) ORVS6 (00F8CH) VS67 VS66 VS65 VS64 VS63 VS62 VS61 VS60 (Initial value: 0000 0000) (00F8DH) − − − − − − − VS68 (Initial value: **** ***0) ORVS7 (00F8EH) VS77 VS76 VS75 VS74 VS73 VS72 VS71 VS70 (Initial value: 0000 0000) (00F8FH) − − − − − − − VS78 (Initial value: **** ***0) ORVS8 (00F90H) VS87 VS86 VS85 VS84 VS83 VS82 VS81 VS80 (Initial value: 0000 0000) (00F91H) − − − − − − − VS88 (Initial value: **** ***0) ORVS9 (00F92H) VS97 VS96 VS95 VS94 VS93 VS92 VS91 VS90 (Initial value: 0000 0000) (00F93H) − − − − − − − VS98 (Initial value: **** ***0) ORVS10 (00F94H) VS107 VS106 VS105 VS104 VS103 VS102 VS101 VS100 (Initial value: 0000 0000) (00F95H) − − − − − − − VS108 (Initial value: **** ***0) ORVS11 (00F96H) VS117 VS116 VS115 VS114 VS113 VS112 VS111 VS110 (Initial value: 0000 0000) (00F97H) − − − − − − − VS118 (Initial value: **** ***0) ORVS12 (00F98H) VS127 VS126 VS125 VS124 VS123 VS122 VS121 VS120 (Initial value: 0000 0000) (00F99H) − − − − − − − VS128 (Initial value: **** ***0) ORVS1 VSn8 to VSn0 Vertical display start position for line n Write only (n: 1 to 12) Note 1: If display lines are overlapped each other, previous display line is enabled and next line is disabled. Set the vertical display start position not to overlap display lines. Note 2: Transfer the contents of vertical display start position registers into OSD circuit before a position of the scanning line coincides with their own vertical display start position. 88CS38B-183 2007-09-12 TMP88CS38B/CM38B/CP38B ORCS4 (00F9AH) ORCS8 (00F9BH) ORCS12 (00F9CH) 7 6 5 4 3 2 1 0 CS4 CS3 CS2 CS1 (Initial value: 0000 0000) CS8 CS7 CS6 CS5 (Initial value: 0000 0000) CS12 CS11 CS10 CS9 (Initial value: 0000 0000) 00: 01: 10: 11: Character size and display on/off for line n CSn Display off Large size Middle size Small size Write only (n: 1 to 12) OREULA8 (00F9DH) OREULA12 (00F9EH) EULA8 EULA7 EULA6 EULA5 − − − − EULA1 (Initial value: 0000 0000) EULA12 EULA11 EULA10 EULA9 EULA4 (Initial value: **** 0000) Underline for display line for line n EULAn EULA3 0: 1: EULA2 Display off Display on (n: 1 to 12) OREFR8 (00F9FH) OREFR12 (00FA0H) EFR8 EFR7 EFR6 EFR5 EFR4 EFR3 EFR2 EFR1 (Initial value: 0000 0000) − − − − EFR12 EFR11 EFR10 EFR9 (Initial value: **** 0000) EFRn Fringing enable specification register for line n 0: 1: Write only Disable fringing Enable fringing (n: 1 to 12) Note: When a display line is enabled fringing function, its vertical size is increased by one dot (by two dots when its character size is small) independent of its character font. Therefore, when a vertical display start position is specified to no space between the lines, the display line which is overlapped with increasing dot(s) is canceled. ORSLO4 (00FA2H) ORSLO8 (00FA3H) ORSLO12 (00FA4H) SLO4 SLO3 SLO2 SLO1 (Initial value: 0000 0000) SLO8 SLO7 SLO6 SLO5 (Initial value: 0000 0000) SLO12 SLO11 SLO10 SLO9 (Initial value: 0000 0000) SLOn Solid space for line n 00: 01: 10: 11: No solid space display Solid space display left Solid space display right Solid space display left and right Write only (n: 0 to 12) 88CS38B-184 2007-09-12 TMP88CS38B/CM38B/CP38B ORBK (00FA5H) 7 6 5 4 3 2 1 0 IBDT RBDT GBDT BBDT IFDT RFDT GFDT BFDT IBDT/ RBDT/ GBDT/ BBDT IFDT/ RFDT/ GFDT/ BFDT Background color select Fringing color select 0000: 0001: 0010: 0011: 0100: 0101: 0110: 0111: 1000: 1001: 1010: 1011: 1100: 1101: 1110: 1111: 0000: 0001: 0010: 0011: 0100: 0101: 0110: 0111: 1000: 1001: 1010: 1011: 1100: 1101: 1110: 1111: Black Blue Green Cyan Red Magenta Yellow White Black Dark blue Dark green Dark cyan Dark red Dark magenta Dark yellow Gray Black Blue Green Cyan Red Magenta Yellow White Black Dark blue Dark green Dark cyan Dark red Dark magenta Dark yellow Gray (Initial value: 0000 0000) Write only Note: Set IBDT and IFDT to 1 when PISEL (Bit6 in ORETC) sets to 1. Then background color select and fringing color select are 8 variety. 88CS38B-185 2007-09-12 TMP88CS38B/CM38B/CP38B ORACL (00FA6H) 7 6 5 4 3 2 1 0 ACLI2 ACLR2 ACLG2 ACLB2 ACLI1 ACLR1 ACLG1 ACLB1 ACLI2/ ACLR2/ ACLG2/ ACLB2 ACLI1/ ACLR1/ ACLG1/ ACLB1 Area 2 plane color select Area 1 plane color select ACLI2 ACLI1 (Initial value: 0000 0000) 0000: Black 0001: Blue 0010: Green 0011: Cyan 0100: Red 0101: Magenta 0110: Yellow 0111: White 1000: Black 1001: Dark blue 1010: Dark green 1011: Dark cyan 1100: Dark red 1101: Dark magenta 1110: Dark yellow 1111: Gray 0000: Black 0001: Blue 0010: Green 0011: Cyan 0100: Red 0101: Magenta 0110: Yellow 0111: White 1000: Black 1001: Dark blue 1010: Dark green 1011: Dark cyan 1100: Dark red 1101: Dark magenta 1110: Dark yellow 1111: Gray 0: Not assign half transparency for area 2 plane 1: Assign half transparency for area 2 plane 0: Not assign half transparency for area 1 plane 1: Assign half transparency for area 1 plane Write only Note: Set ACLI2 and ACLI1 to 1 when PISEL (Bit6 in ORETC) sets to 1. Then area 2 plane color select and area 1 plane color select are 8 variety. 88CS38B-186 2007-09-12 TMP88CS38B/CM38B/CP38B ORIV (00FBBH) 7 6 VDPOL HDPOL 5 4 3 2 1 0 YBLII RGBII YIV BLIV RGBIV IIV VDPOL VD input polarity select HDPOL HD input polarity select YBLII Y/BLIN input polarity select RGBII RIN, GIN, BIN input polarity select YIV Y output polarity select BLIV BL output polarity select RGBIV 0: 1: 0: 1: 0: 1: 0: 1: 0: 1: 0: 1: 0: 1: 0: 1: R, G, B output polarity select IIV I output polarity select (Initial value: 0000 0000) Non-invert input signal Invert input signal Non-invert input signal Invert input signal Active high Active low Active high Active low Active high Active low Active high Active low Active high Active low Active high Active low Write only ORDMA 7 6 5 4 3 2 1 0 (00024H) DMA7 DMA6 DMA5 DMA4 DMA3 DMA2 DMA1 DMA0 (Initial value: 0000 0000) (00025H) − − − − − − − DMA8 (Initial value: **** ***0) DMAn Write only Display memory address (n: 0 to 8) Note: It necessary to write all bits of display memory address, writng DMA7 to DMA0 after DMA8, when writing display address. ORDSN 7 6 5 4 3 2 1 0 (0001DH) − SLNT EUL BLF IDT RDT GDT BDT Slant enable specification register 0: 1: Disable slant Enable slant EUL Underline enable specification register 0: 1: Disable underline Enable underline BLF Blinking enable specification register 0: Disable blinking 1: Enable blinking 0000: Black 0001: Blue 0010: Green 0011: Cyan 0100: Red 0101: Magenta 0110: Yellow 0111: White 1000: Black 1001: Dark blue 1010: Dark green 1011: Dark cyan 1100: Dark red 1101: Dark magenta 1110: Dark yellow 1111: Gray SLNT IDT/ RDT/ GDT/ BDT Note: Character color select (Initial value: **** ****) Read/ Write Set IDT to 1 when PISEL (Bit6 in ORETC) sets to 1. Then character color select is 8 variety. 88CS38B-187 2007-09-12 TMP88CS38B/CM38B/CP38B 7 6 5 4 3 2 1 0 (0001EH) CRA7 CRA6 CRA5 CRA4 CRA3 CRA2 CRA1 CRA0 (Initial value: **** ****) (0001FH) − − − − − − − CRA8 (Initial value: **** ****) ORCRA CRAn Read/ Write Character code (n: 0 to 8) Note: ORWVSH (00FBCH) (00FBDH) Write or read CRA7 to CRA0 after write or read CRA8. 7 6 5 4 3 2 1 0 WVSH7 WVSH6 WVSH5 WVSH4 WVSH3 WVSH2 WVSH1 WVSH0 − − WVSLn − − − − − WVSH8 (Initial value: 0000 0000) (Initial value: **** ***0) Write only Window upper limit position (n: 0 to 8) ORWVSL (00FBEH) (00FBFH) 7 6 5 4 3 2 1 0 WVSL7 WVSL6 WVSL5 WVSL4 WVSL3 WVSL2 WVSL1 WVSL0 − − WVSLn − − − − − WVSL8 (Initial value: 0000 0000) (Initial value: **** ***0) Write only Window lower limit position (n: 0 to 8) ORDON 7 6 5 4 3 2 1 0 (00F80H) − − − − − RGWR EWDW DON RGWR Written data transfer control EWDW Window enable specification register DON Display on/off select 0: 1: 0: 1: 0: 1: (Initial value: **** *000) (Initial state) Transfers written data to OSD circuit (after transfer, RGWR is reset to 0) Disable window function Enable window function Disable display Enable display 88CS38B-188 Read/ Write 2007-09-12 TMP88CS38B/CM38B/CP38B ORRCL (00FA7H) 7 6 5 4 3 2 1 0 EBKGD EXBL AON2 AON1 RCLI RCLR RCLG RCLB Background function enable specification register 0: 1: No background function Background function enable EXBL Full-raster blanking enable specification register 0: 1: No Full-raster blanking Full-raster blanking AON2 Area 2 plane display enable specification register 0: 1: No area 2 plane display Area 2 plane display enable AON1 Area 1 plane display enable specification register 0: No area 1 plane display 1: Area 1 plane display enable 0000: Black 0001: Blue 0010: Green 0011: Cyan 0100: Red 0101: Magenta 0110: Yellow 0111: White 1000: Black 1001: Dark blue 1010: Dark green 1011: Dark cyan 1100: Dark red 1101: Dark magenta 1110: Dark yellow 1111: Gray EBKGD RCLI RCLR/ RCLG/ RCLB Note: Raster plane color select (Initial value: 0000 0000) Write only Set RCLI to 1 when PISEL (Bit6 in ORETC) sets to 1. Then transfer plane select is 8 variety. 88CS38B-189 2007-09-12 TMP88CS38B/CM38B/CP38B ORAHS1 7 6 5 4 3 2 1 0 (00FA8H) AHS17 AHS16 AHS15 AHS14 AHS13 AHS12 AHS11 AHS10 (Initial value: 0000 0000) (00FA9H) − − − − − − − AHS18 (Initial value: **** ***0) ORAHE1 (00FAAH) AHE17 AHE16 AHE15 AHE14 AHE13 AHE12 AHE11 AHE10 (Initial value: 0000 0000) (00FABH) − − − − − − − AHE18 (Initial value: **** ***0) AHS1n Horizontal start point for area 1 plane AHE1n Horizontal end point for area 1 plane Write only (n: 0 to 8) ORAVS1 (00FACH) AVS17 AVS16 AVS15 AVS14 AVS13 AVS12 AVS11 AVS10 (Initial value: 0000 0000) (00FADH) − − − − − − − AVS18 (Initial value: **** ***0) ORAVE1 (00FAEH) AVE17 AVE16 AVE15 AVE14 AVE13 AVE12 AVE11 AVE10 (Initial value: 0000 0000) (00FAFH) − − − − − − − AVE18 (Initial value: **** ***0) AVS1n Vertical start point for area 1 plane AVE1n Vertical end point for area 1 plane Write only (n: 0 to 8) ORAHS2 (00FB0H) AHS27 AHS26 AHS25 AHS24 AHS23 AHS22 AHS21 AHS20 (00FB1H) − − − − − − − AHS28 ORAHE2 (00FB2H) AHE27 AHE26 AHE25 AHE24 AHE23 AHE22 AHE21 AHE20 (Initial value: 0000 0000) (00FB3H) − − − − − − − AHE28 (Initial value: **** ***0) AHS2n Horizontal start point for area 2 plane AHE2n Horizontal end point for area 2 plane Write only (n: 0 to 8) ORAVS2 (00FB4H) AVS27 AVS26 AVS25 AVS24 AVS23 AVS22 AVS21 AVS20 (Initial value: 0000 0000) (00FB5H) − − − − − − − AVS28 (Initial value: **** ***0) ORAVE2 (00FB6H) AVE27 AVE26 AVE25 AVE24 AVE23 AVE22 AVE21 AVE20 (Initial value: 0000 0000) (00FB7H) − − − − − − − AVE28 (Initial value: **** ***0) AVS2n Vertical start point for area 2 plane AVE2n Vertical end point for area 2 plane Write only (n: 0 to 8) 88CS38B-190 2007-09-12 TMP88CS38B/CM38B/CP38B ORP6S (00FBAH) 7 6 5 4 3 2 P67S P66S P65S P64S PIDS YBLCS P67S to P64S ORETC (00FB8H) I pin output select YBLCS Y/BL signal select MPXS R, G, B, Y/BL signal select 0 MPXS (Initial value: 0000 0000) 0: 1: 0: 1: 0: 1: 00: R, G, B, Y/BL signal output P67 to P64 port output I signal output P57 port output Y signal output BL signal output Simultaneous output (Signal from the OSD circuit has higher priority) 01: Output of signal from internal OSD circuit 10: Output of signal from externally input 11: Simultaneous output (Externally input signal has higher priority) P6 port output select PIDS 1 7 6 5 4 3 2 1 0 VDSMD PISEL BKMF ESMZ “0” MFYWR MBK RDWRV VDSMD Scan mode select PISEL I pin function select BKMF Blinking master ESMZ Smoothing enable specification register MFYWR Display memory bank switching MBK RDWRV Note: Display memory read mode select Read/write mode select (Initial value: 0000 0000) 0: 1: 0: 1: 0: 1: 0: 1: Normal mode Double scan mode 15 colors Half transparency/half tone Double blinking Enable blinking Disable smoothing Enable smoothing 0: 1: 0: Normal mode Read-modify-write mode Access to either character code or character display options Access both character display option and character code Data write mode for display memory Data read mode for display memory 1: 0: 1: Write only Write only Clear “0” to bit3 in ORETC. 88CS38B-191 2007-09-12 TMP88CS38B/CM38B/CP38B ORIRC 7 6 5 4 (00FB9H) − − − SDV SVD ISDC ORIRC (00FB9H) − DCTR Note 1: 3 Interrupt generation line select − − Display line counter 1 0 ISDC Interrupt source select − 2 (Initial value: ***0 0000) 0: Interrupt request by ISDC value 1: Interrupt request at falling edge of VD signal When the line display of the ISDC value ends (with the falling edge of HD signal) while SVD = 0, interrupt request is generated. 0000: Request interrupt when display of low-order 4 bits “0000” of DCTR ends. 0001: Low-order 4 bits “0001” of DCTR 0010: Low-order 4 bits “0010” of DCTR 0011: Low-order 4 bits “0011” of DCTR 0100: Low-order 4 bits “0100” of DCTR 0101: Low-order 4 bits “0101” of DCTR 0110: Low-order 4 bits “0110” of DCTR 0111: Low-order 4 bits “0111” of DCTR 1000: Low-order 4 bits “1000” of DCTR 1001: Low-order 4 bits “1001” of DCTR 1010: Low-order 4 bits “1010” of DCTR 1011: Low-order 4 bits “1011” of DCTR 1100: Low-order 4 bits “1100” of DCTR 1101: Low-order 4 bits “1101” of DCTR 1110: Low-order 4 bits “1110” of DCTR 1111: Low-order 4 bits “1111” of DCTR DCTR Write only (Initial value: **** 0000) 0000: No line display or when the display of the 16th line ends. 0001: 1st line display ends. 0010: 2nd line display ends. 0011: 3rd line display ends. 0100: 4th line display ends. 0101: 5th line display ends. 0110: 6th line display ends. 0111: 7th line display ends. 1000: 8th line display ends. 1001: 9th line display ends. 1010: 10th line display ends. 1011: 11th line display ends. 1100: 12th line display ends. 1101: 13th line display ends. 1110: 14th line display ends. 1111: 15th line display ends. Read only The display line counter also increments when a line with all blank data or a line with display off is specified. If display lines are overlapped each other, previous display line is enabled and next line is disabled. At this time, the display line counter for canceled line does not increment. Note 2: *: Don’t care. Note 3: All OSD control registers cannot use the read-modify-write instructions. (Bit manipulation instructions such as SET, CLR, etc. and logical operation such as AND, OR, etc.) 88CS38B-192 2007-09-12 TMP88CS38B/CM38B/CP38B Table 2.15.5.10 OSD Control Register List (1/2) Register Address Register Name 00F81 00F82, 00F83 to 00F98, 00F99 00F9A 00F9B 00F9C 00F9D 00F9E 00F9F 00FA0 00FA2 00FA3 00FA4 00FA5 ORHS1 ORVSn ORCS4 ORCS8 ORCS12 OREULA8 OREULA12 OREFR8 OREFR12 ORSOL4 ORSOL8 ORSOL12 ORBK 00FA6 Register Bit Configuration Bit4 Bit3 Bit2 Bit7 Bit6 Bit5 HS17 VSn7 − HS16 VSn6 − HS15 VSn5 − HS14 VSn4 − HS13 VSn3 − HS12 VSn2 − Bit0 HS10 VSn0 VSn8 CS4 CS8 CS12 EULA8 EULA7 − − EFR8 EFR7 − − SOL4 SOL8 SOL12 IBDT RBDT CS3 CS7 CS11 EULA6 EULA5 − − EFR6 EFR5 − − SOL3 SOL7 SOL11 GBDT BBDT ORACL ACLI2 ACLR2 ACLG2 ACLB2 ACLI1 ACLR1 ACLG1 ACLB1 00FA7 CRRCL EBKGD EXBL AON2 AON1 RCLI RCLR RCLG RCLB 00FA8 00FA9 00FAA 00FAB 00FAC 00FAD 00FAE 00FAF 00FB0 00FB1 00FB2 00FB3 00FB4 00FB5 00FB6 00FB7 00FB8 ORAHS1 ORETC AHS17 − AHE17 − AVS17 − AVE17 − AHS27 − AHE27 − AVS27 − AVE27 − VDSMD AHS16 − AHE16 − AVS16 − AVE16 − AHS26 − AHE26 − AVS26 − AVE26 − PISEL AHS15 − AHE15 − AVS15 − AVE15 − AHS25 − AHE25 − AVS25 − AVE25 − BKMF AHS14 − AHE14 − AVS14 − AVE14 − AHS24 − AHE24 − AVS24 − AVE24 − ESMZ AHS13 − AHE13 − AVS13 − AVE13 − AHS23 − AHE23 − AVS23 − AVE23 − “0” AHS12 − AHE12 − AVS12 − AVE12 − AHS22 − AHE22 − AVS22 − AVE22 − MFYWR AHS11 − AHE11 − AVS11 − AVE11 − AHS21 − AHE21 − AVS21 − AVE21 − MBK AHS10 AHS18 AHE10 AHE18 AVS10 AHS18 AVE10 AVE18 AHS20 AHS28 AHE20 AHE28 AVS20 AHS28 AVE20 AVE28 RDWRV 00FB9 ORIRC − − − SVD 00FB9 00FBA ORIRC ORP6S − P67S − P66S − P65S − P64S ORAHE1 ORAVS1 ORAVE1 ORAHS2 ORAHE2 ORAVS2 ORAVE2 CS2 CS6 CS10 EULA4 EULA3 EULA12 EULA11 EFR4 EFR3 EFR12 EFR11 SOL2 SOL6 SOL10 IFDT RFDT Bit1 HS11 VSn1 − CS1 CS5 CS9 EULA2 EULA1 EULA10 EULA9 EFR2 EFR1 EFR10 EFR9 SOL1 SOL5 SOL9 GFDT BFDT ISDC PIDS DCTR YBLCS 88CS38B-193 MPXS Bit Contents HS17 to 10: Code horizontal display base position setting VSn8 to 0: Code vertical display position setting (n: 0 to 12) CSn: Character size (n: 1 to 12) 00: Display off 10: Middle size 01: Large size 11: Small size EULAn: Underline display setting for line n (n: 0 to 12) EFRn: Fringing setting for line n (n: 0 to 12) SOLn: Solid space display setting for line n (n: 0 to 12) 00: No solid space 10: Right 01: Left 11: Left and right IBDT, RBDT, GBDT, BBDT: Background color setting IFDT, RFDT, GFDT, BFDT: Fringing color setting ACLI2/ACLR2/ACLG2/ACLB2: Area 2 plane color ACLI1/ACLR1/ACLG1/ACLB1: Area 1 plane color Set ACLI2 and SCLI1 to 1, when PISEL 1 EBKGD: Background function EXBL: Full-raster blanking AON2: Area 2 plane display AON1: Area 1 plane display RCLI/R/G/B: Raster plane color Set RCLI to 1, when PISEL 1 AHSx: Area 1 plane horizontal start position (n: 0 to 8) AHE1x: Area 1 plane horizontal end position (n: 0 to 8) AVS1x: Area 1 plane vertical start position (n: 0 to 8) AVE1x: Area 1 plane vertical end position (n: 0 to 8) AHS2x: Area 2 plane horizontal start position (n: 0 to 8) AHE2x: Area 2 plane horizontal end position (n: 0 to 8) AVS2x: Area 2 plane vertical start position (n: 0 to 8) AVE2x: Area 2 plane vertical end position (n: 0 to 8) VDSMD: Scan mode select PISEL: I pin function select BKMF: Blinking master ESMZ: Smoothing MFYWR: Display memory read mode select MBK: Display memory bank switching select RDWRV: Read/write mode select at normal mode SVD: Interrupt source select ISDC: Interrupt generation line select DCTR: Display line counter P6xS: P6 port output select (x: 4 to 7) PIDS: I pin output select YBLCS: Y/BL signal select MPXS: R, G, B, Y/BL single select 2007-09-12 TMP88CS38B/CM38B/CP38B Table 2.15.5.11 OSD Control Register List (2/2) Register Bit Configuration Bit4 Bit3 Bit2 Register Address Register Name Bit7 Bit6 Bit5 Bit1 Bit0 00FBB ORIV VDPOL HDPOL YBLII RGBII YIV BLIV RGBIV IIV 00024 00025 0001D ORDMA DMA7 − − DMA6 − SLNT DMA5 − EUL DMA4 − BLF DMA3 − IDT DMA2 − RDT DMA1 − GDT DMA0 DMA8 BDT 0001E 0001F 00FBC 00FBD 00FBE 00FBF 00F80 ORCRA CRA7 − WVSH7 − WVSL7 − − CRA6 − WVSH6 − WVSL6 − − CRA5 − WVSH5 − WVSL5 − − CRA4 − WVSH4 − WVSL4 − − CRA3 − WVSH3 − WVSL3 − − CRA2 − WVSH2 − WVSL2 − RGWR CRA1 − WVSH1 − WVSL1 − EWDW CRA0 CRA8 WVSH0 WVSH8 WVSL0 WVSL8 DON ORDSN ORWVSH ORWVSl ORDON Bit Contents VDPOL: VD input polarity select HDPOL: HD input polarity select YBLII: Y/BLIN input polarity select RGBII: RIN, GIN, BIN input select YIV: Y output polarity select BLIV: BL output polarity select RGBIV: R, G, B output polarity select IIV: I pin polarity select DMAx: Display memory address setting (x: 0 to 8) SLNT: Slant EUL: Underline BLF: Blinking IDT/RDT/CDT/BDT: Character color CRAx: Character code (x: 0 to 8) WVSHx: Window upper limit position (x: 0 to 8) WVSL: Window lower limit position (x: 0 to 8) RGWR: Writing data transfer control EWDW: Window enable DON: OSD display on/off Note 1: Except the meshed registers are changed by RGWR. Note 2: Only lower 2 bits of the register in address 00F80H are changed by RGWR (the register in address 00F80H must not be used with any of the read-modify-write instructions as SET, CLR, etc.). 88CS38B-194 2007-09-12 TMP88CS38B/CM38B/CP38B 2.16 Jitter Elimination Circuit The TMP88CS38B/CM38B/CP38B has a built-in jitter elimination circuit which maintains the vertical stability of the OSD even when input of the vertical signal fluctuates. And the field decision information for the OSD circuit is detected by using jitter elimination circuit. 2.16.1 Configuration Jitter removal status register Phase detect signal PDF [2:0] JRMSR Field decision circuit HD (P70) Previous field decision signal A Y B S VD (P71) HD / VD Edge detect circuit Internal VD signal output control circuit Delay value setting circuit A B Y VD (To OSD circuit) VDSEL VD signal delay value measuring circuit fc/2 AFLD JEEN JECR Jitter elimination control register Figure 2.16.1 Jitter Elimination Circuit 88CS38B-195 2007-09-12 TMP88CS38B/CM38B/CP38B 2.16.2 Control Jitter elimination circuit is controlled by the jitter elimination control register (JECR). Jitter Elimination Control Register JECR 7 6 5 4 3 2 1 0 (00FE4H) − − − VDSEL AFLD JEEN “0” “0” VDSEL 0: 1: 0: 1: 0: 1: VD select AFLD Automatic field decision JEEN Jitter elimination enable specification VD from P71 VD from jitter elimination circuit Automatic field decision disabled Automatic field decision enabled Jitter elimination disabled Jitter elimination enabled Note 1: Clear the AFLD to “0” to disable jitter elimination circuit. Note 2: Always clear “0” to bit1 and bit0 of JECR. Note 3: (Initial value: ***0 0000) Write only Clear “0” to AFLD and VDSEL if there is no phase shift in the vertical and horizontal sync. signals every other time, such as with non-interlaced TV. Note 4: *: Don’t care Note 5: Setting JEEN to “0”, OSD display is only 2nd field. Note 6: Setting AFLD to “0”, OSD display is only 2nd field. Jitter Elimination Status Register JESR (00FE5H) 7 6 5 4 3 2 1 0 FDSF PDF1 PDF0 − − − − PDF2 (Initial value: 0*** ****) 0: FDSF Field detect status flag PDF2 to PDF0 Phase detect flag between HD and VD A position of a scanning line exists in the field which has a second display dot of character on an interlace TV screen. 1: A position of a scanning line exists in the field which has a first display dot of character on an interlace TV screen. Read 000: Phase 0 only 001: Phase 1 010: Phase 2 011: Phase 3 100: Phase 4 101: Phase 5 110: Phase 6 111: Phase 7 Note 1: FDSF is different from the 1st and the 2nd field. It is a unique field decided for OSD display. Note 2: *: Don’t care. Note 3: HD VD Phase 7 Phase 0 Phase 1 Phase 2 Phase 3 Phase 4 Phase 5 Phase 6 Phase 7 Phase 0 Figure 2.16.2 Jitter Elimination Control Register and Jitter Elimination Status Register 2.16.3 Jitter Elimination Mode The jitter elimination circuit is to identify the phase of the falling edges of the external VD signal and HD signal. When VD signal is falling within HD signal falling +/−1/4HD, the jitter is automatically eliminated and internal VD signal is set to the stable location. This function is enabled by setting JEEN (Bit2 in JECR) in the jitter elimination control register to “1”. 88CS38B-196 2007-09-12 TMP88CS38B/CM38B/CP38B 2.16.4 Auto Field Line Decision The internal vertical and horizontal sync. signals corrected by the jitter elimination circuit generate the field line decision signals used in the OSD. The OSD display in normal mode Type A) When the OSD circuit is used on the TV system which has a phase shift in the vertical and horizontal sync. Signals every other filed such as the interlace TV, enable jitter elimination circuit and set “1” to AFLD and VDSEL. At this time, the field lines which have first and second display dot of character are displayed. Type B) When the OSD circuit is used on the TV system which has no phase shift in the vertical and horizontal sync. Signals every other filed such as the non-interlace TV, enable jitter elimination circuit and clear “0” to AFLD and VDSEL. At this time, the field line which has a second display dot of character is only displayed. The OSD display in double scan mode Type C) Disable jitter elimination circuit and clear “0” to AFLD and VDSEL. At this time, the field lines which have first and second display dot of character are displayed. (1) The field line which has a first display dot of character (2) The field line which has a second display dot of character Scanning System Register Display Type A VDSEL = 1, AFLD = 1 (1) and (2) Type B VDSEL = 0, AFLD = 0 (2) Type C VDSEL = 0, AFLD = 0 (1) and (2) Figure 2.16.3 Relation with Field Line and VDSEL, AFLD 88CS38B-197 2007-09-12 TMP88CS38B/CM38B/CP38B 2.17 Data Slicer The TMP88CS38B/CM38B/CP38B contains the data slicer to decode the caption data which multiplied during vertical flyback time of the composite video signal. The composite video signal inputs to the data slicer circuit through P32 (VIN1) and P33 (VIN0). The caption data is decoded from the video signal. The composite video signal including negative sync-tip inputs to VIN0 and VIN1 pins. The data slicer can comply with the copy guard signal and special signals, and receive accurately the caption data under the condition of a weak electrical field or a ghost. Note: 2.17.1 When the data slicer is used at fc = 16 MHz, set to “02H” in FC8CR. When the data slicer is used at fc = 8 MHz, set to “00H” in FC8CR. (Refer to Figure 1.4.5) Configuration A0 A1 S Z SYNCINV Slicer mode setting register 1 SIF status read register 2 (00FDFH) C.Sync external input mode SIFSMS1/SIFS1R EXSYNC C.Sync signal Synchronous separator VIN1 Sync-tip clamp circuit Comparator 1 H timing circuit Composite video signal V timing circuit LINE21 VIN0 Pedestal clamp circuit INTSLI (to interrupt) Clamping pulse Comparator 2 Slice level control circuit Sampling clock generation circuit Data separator DA converter DACLCR (00FD9H) Sync-tip slice level setting register SIFSR (00FDDH) Slicer interface circuit DA converter SLVLCR (00FDAH) Slice level control register SIFDR1 SIFDR2 (00FDBH) (00FDCH) Data register 1 Data register 2 Figure 2.17.1 Data Slicer 88CS38B-198 2007-09-12 TMP88CS38B/CM38B/CP38B 2.17.2 Functions (1) Video signal input A low pass filter, a voltage amplifier and a condenser of about 0.1 μF are connected between the video signal and the video signal input pin of VIN1 and VIN0 pins, that is shown as Figure 2.17.3 the low pass filter functions to reduce noise and color burst from the video signal, passes the amplifier and inputs the video signal to both VIN1 and VIN0 pins. (2) Synchronous separator This circuit is to separate the synchronous signal from the video signal. When DACL7 to DACL0 of DACLCR are set for the synchronous separation, the sync slice level is capable of setting. DACL7 to DACL4 set the slice level at the rising edge of the sync signal clamped data, and DACL3 to DACL0 set the slice level at the falling edge of the sync-tip clamped data. (Refer to section 2.17.5) (3) Data separator The data separator replaces the caption data piled on the video signal with the digital signal. When SLVL5 to SLVL0 of SLVLCR are set to get the digital signal, the Initial value of the caption data slice level is capable of setting. (Refer to section 2.17.5) (4) Sync-tip clamp circuit The sync-tip level is clamped to the specified value. (5) Pedestal clamp circuit The video signal is set to the specified voltage with the clamp pulse generated from the H/V timing part, which is called as a pedestal clamp. (6) DA converter This converter gets the DA changed slice level of the clamp circuit to the comparator. (7) Comparator This comparator replaces the composite video signal with the digital value while inputting to the comparator. (8) H timing circuit This circuit detects the horizontal synchronous signal from C.Sync signal separated synchornously from the video signal, and generates the clamp pulse to clamp the video signal and provides it to the pedestal clamp circuit. In addition, the circuit detects the change of H frequency and provides the data to the sampling clock generation part. (9) V timing circuit This circuit detects the horizontal synchronous signal from C.Sync signal separated synchornously from the video signal, and provides line 21 detection signal to take out caption signal to the slice level control part. (10) Slice level control circuit This circuit detects CRI (Clock run in) signal from VIDEO signal with line 21 detection signal generated at H/V timing part after slicing, and controls to the most suitable slice level and takes out the caption data. 88CS38B-199 2007-09-12 TMP88CS38B/CM38B/CP38B (11) Sampling clock generation circuit This circuit generates the sampling clock which is phase-locked to CRI signal with CRI signal detected at the slice level control part. In addition, the circuit revises the location where the sampling clock generates with H frequency variable data generated at H timing generation part. (12) Slicer interface circuit This is a 16-bit serial interface to receive the serial data. (13) Interrupt generation circuit Interrupts are generated by a rise in the caption line detection signal. Video signal Caption line 7.7 μs Caption line detection signal ( LINE21 ) 95 μs Interrupt generation Figure 2.17.2 Interrupt Generation Timing See the description of the on-screen display circuit interrupt vectors for details of interrupt vectors. (14) C.Sync external input mode The external C.Sync signal can be used internally by setting EXSYNC (SIFSMS1 bit5) to “1“. As shown in Figure 2.17.3 (b), insert a low-pass filter (fT = 503 kHz), voltage amplifier (× 2 voltage amplification), and a capacitor of approximately 0.1 μF between the video signal and the video signal input pin VIN1 and input an external C.Sync signal to CSIN. The polarity of the C.Sync signal is selected by SYNCINV (SIFSMS1 bit6). (Internally used as C.Sync .) CSIN (P32) SYNCINV C.Sync ( ) “0” C.Sync ( ) “1” 88CS38B-200 2007-09-12 TMP88CS38B/CM38B/CP38B 2.17.3 Video Signal Connection TMP88CS38B/CM38B/CP38B Outer circuit 0.1 μF (1.0 VPP) Composite video signal P32 (VIN1) (2 VPP) Low pass filter Amplifier P33 (VIN0) 0.1 μF (a) Internal sync separation mode TMP88CS38B/CM38B/CP38B Outer circuit (5.0 VPP) Ext.C.Sync signal P32 (CSIN) (1.0 VPP) Composite video signal P33 (VIN0) (2 VPP) Low pass filter 0.1 μF Amplifier (b) C.sync external input mode Figure 2.17.3 Video Signal Connection Data Slicer Control Register SINTCR 7 6 5 4 3 2 1 0 (00FD8H) − − − − SLON SLCR − − SLON Data slicer enable/disable SLCR Data slicer interrupt control 1: 0: 1: 0: (Initial value: 0000 00**) Enable Disable Enable interrupt Disable interrupt Write only Data Slicer Interrupt Satus Register SINTCR 7 6 5 4 3 2 1 0 (00FD8H) − − − SLIS − − − − SLIS 0: 1: Data slicer interrupt status − Interrupt request Note 1: For setting SCLR to “1”, write “1” after SLON is set to “1”. Note 2: SLIS is cleared to “0” after reading SINTCR. (Initial value: ***0 ****) Read only Figure 2.17.4 Data Slicer Control (I) 88CS38B-201 2007-09-12 TMP88CS38B/CM38B/CP38B SIF Data Register 1 (Caption data 1st byte read register) (Read only) SIFDR1 (00FDBH) 7 6 5 4 3 2 1 0 D1ST7 D1ST6 D1ST5 D1ST4 D1ST3 D1ST2 D1ST1 D1ST0 D1ST7-0 Read only Caption data 1st byte read register SIF Data Register 2 (Caption data 2nd byte read register) (Read only) SIFDR2 (00FDCH) 7 6 5 4 3 2 1 0 D2ST7 D2ST6 D2ST5 D2ST4 D2ST3 D2ST2 D2ST1 D2ST0 D2ST7-0 Read only Caption data 2nd byte read register SIF Status Register (Read only) SIFST (00FDDH) 7 6 5 4 3 2 1 0 STCRI CRIN3 CRIN2 CRIN1 CRIN0 STFLD STSB STDE STCRI Clock run in detection CRIN CRI number − 1 1: Clock run in detection 0: No clock run in detection Actual CRI number − 1 1: 2nd field 0: 1st field 1: From detection of start bit until fall in VD 0: Other times STFLD Field identification STSB Start bit identification flag STDE 16-bit data receive end identification flag 1: 0: Read only From end of 16-bit data reception until fall in VD Other times Figure 2.17.5 Data Slicer Control (II) Slicer Mode Setting Register 1 (Write only) SIFSMS1 (00FDFH) 7 6 5 SYNC INV EXSYNC “0” 4 “1” 3 CLINE Note: 1 0 CLINE3 CLINE2 CLINE1 CLINE0 SYNCINV Sync signal input inversion EXSYNC 2 Sync signal selection Setting lines piled on caption data (Initial value: 0001 1011) 0: No inversion 1: Inversion of C.Sync external input signal 0: Internal sync separation 1: External C.Sync input 0000: 10 lines 0001: 11 lines 0010: 12 lines 0011: 13 lines 0100: 14 lines 0101: 15 lines 0110: 16 lines 0111: 17 lines 1000: 18 lines 1001: 19 lines 1010: 20 lines 1011: 21 lines 1100: 22 lines 1101: 23 lines 1110: 24 lines 1111: 25 lines Write only Always write “0” to bit7 of SIFSMS1 and “1” to bit4 when writing to SIFSMS1. Figure 2.17.6 Data Slicer Control (III) 88CS38B-202 2007-09-12 TMP88CS38B/CM38B/CP38B SIF Status Read Register 2 SIFS1R 7 6 (00FDFH) − − GOODV FLINE 5 4 3 2 1 0 GOODV FLINE4 FLINE3 FLINE2 FLINE1 FLINE0 Monitor signal of synchronization Field scanning line (Standard 262.5 = − 1) Two’s complement 0: 1: 00000: 00001: 00010: 00011: 00100: 00101: 00110: 00111: 01000: 01001: 01010: 01011: 01100: 01101: 01110: 01111: 10000: 10001: 10010: 10011: 10100: 10101: 10110: 10111: 11000: 11001: 11010: 11011: 11100: 11101: 11110: 11111: Out of synchronization (One or more) V timing synchronizing 0 263.5 1 264.5 2 3 4 5 6 7 8 9 10 11 12 13 14 15 278.5 V synchronizing adjustment − 15 248.5 − 14 − 13 − 12 − 11 − 10 −9 −8 −7 −6 −5 −4 −3 −2 261.5 −1 262.5 Read only Figure 2.17.7 Data Slicer Control (IV) The explanation of the monitor signals (GOODV, FLINE) are as follows. 1.GOODV 0: Data slicer can not synchronize video signal. 1: Data slicer can synchronize video signal. 2.FLINE The number of field signal scanning line which the data slicer is detecting or monitor flag of detecting state. Example: FLINE = 1FH: NTSC signal FLINE = 10H: V synchronizing adjustment 88CS38B-203 2007-09-12 TMP88CS38B/CM38B/CP38B Caption Data Slice Level Control Register (Write/Read) SLVLCR 7 6 5 4 3 2 1 0 (00FDAH) − − SLVL5 SLVL4 SLVL3 SLVL2 SLVL1 SLVL0 (Initial value: **00 1010) 000000: VPCLAMP + (1/256) VDD 000001: VPCLAMP + (2/256) VDD 000010: VPCLAMP + (3/256) VDD SLVL 000011: VPCLAMP + (4/256) VDD Slice level (Initial value:) setting Sice level setting 000100: VPCLAMP + (5/256) VDD . . . . . . 111101: VPCLAMP + (62/256) VDD Write 111110: VPCLAMP + (63/256) VDD 111111: VPCLAMP + (64/256) VDD SLVL Slice level (final value) Read Note 1: VPCLAMP (Pedestal clamp) = (1/2) VDD Note 2: The SLVLCR has different write buffer and read buffer, and cannot be read write-buffer fata. The SBIDBR cannot be used with any read-modify-write instructions. (Bit manipulation instructions such as SET, CLR, etc. and logical operation such as AND, OR, etc.) Sync-tip Slice Level Setting Register (Write only) DACLCR 7 6 5 4 3 2 1 0 (00FD9H) DACL7 DACL6 DACL5 DACL4 DACL3 DACL2 DACL1 DACL0 (Initial value: 0100 0010) 0000: VSCLAMP + (3/512) VDD 0001: VSCLAMP + (6/512) VDD DACL DACL7 to DACL4: Slice level Lower limit setting DACL3 to DACL0: Slice level Upper limit setting 0010: VSCLAMP + (9/512) VDD 0011: VSCLAMP + (12/512) VDD . . . . . . 1101: VSCLAMP + (42/512) VDD Write only 1110: VSCLAMP + (45/512) VDD 1111: VSCLAMP + (48/512) VDD Note: VSCLAMP (Sync-tip clamp) = (204/512) VDD Figure 2.17.8 Data Slicer Control (V) 88CS38B-204 2007-09-12 TMP88CS38B/CM38B/CP38B 2.17.4 Clamp and Data Slicer Operation The slicer uses the following steps to obtain the caption signals: Sync signal separation The composite video signal input via VIN1 (Pin 40) is clamped by the sync tip clamp circuit and the HD and VD sync signals separated by the sync separation circuit. Caption line detection Field decision and caption line detection are effected using the HD and VD sync signals CRI detection The CRI signal of the caption line interval is detected from the pedestal clamped video signal at VIN0 (Pin 41). Slice level setting and generation of sampling clock The slice level is controlled during the CRI signal interval, detected by the slice level control block, to obtain the optimum level. To determine the timing for extracting the caption data, a sampling clock is generated that is phase locked to the CRI. Caption signal extraction The caption data is extracted at the selected slice level using sampling clock, which is locked to the caption data. The data slicer has two separation circuits: a. Sync signal (sync tip clamp + sync signal slice) separation. b. Caption data (pedestal clamp + data slice) separation. The two circuits are described briefly below. 88CS38B-205 2007-09-12 TMP88CS38B/CM38B/CP38B a. Sync signal (Sync tip clamp + sync signal slice) a-1 Sync tip clamp (Pin 40) ........The sync tip is clamped at (204/512) VDD [V] as shown in Figure 2.17.9. Video signal 1 A, B: Sync tip slice levels A: DACL7 to DACL4 Lower-limit setting B: DACL3 to DACL0 Upper-limit setting 2 B A (204/512) VDD [V] GND After sync signal separation H L Figure 2.17.9 Sync Signal Slice a-2 Method of sync signal slice The sync signal is separated as shown in Figure 2.17.9. Sync signal separation is accomplished by comparing the voltage of the sync tip-clamped video signal with the sync tip slice level. For a 1 → 2 video signal change, if the sync signal after separation is high, the slice level A is selected; if low, the slice level B is selected. (Sync tip slice level) Slice level = VSCLAMP + {(3 + 3X)/512} VDD VDD: Power supply voltage VSCLAMP: Sync tip clamp voltage = (204/512) VDD X: Setup data (4 bits) 88CS38B-206 2007-09-12 TMP88CS38B/CM38B/CP38B b. Caption data (Pedestal clamp + data slice) b-1 Pedestal clamp (Pin 41) ......Clamped at (1/2) VDD [V] as shown in Figure 2.17.10. Slice level (1/2) VDD [V] GND Figure 2.17.10 Pedestal Clamp b-2 Method of data slice The data slice level constitutes a level at which the CCD data is differentiated. The slice level’s setup value is indicated by the following: Slice level = VPCLAMP + (X/256) VDD [V] VDD: Power supply voltage VPCLAMP: Pedestal clamp voltage = (1/2) VDD X: Setup data (6 bits) b-3 Automatic slice level correction circuit The slice level is corrected to the appropriate value during the CRI period. Slice level correction always begins with the setup value of SLVL (Bit5 to bit0 of SLVLCR). If you want the last value to become the initial value of the next slice level, set it to SLVL (Bit5 to bit0 of SLVLCR). 88CS38B-207 2007-09-12 TMP88CS38B/CM38B/CP38B Input/Output Circuit (1) Control pins The input/output circuitries of the TMP88CS38B/CM38B/CP38B control pins are shown below. Control Pin I/O Input/Output Circuitry Osc. enable XIN XOUT Remarks Resonator connection pins (High frequency) fc VDD VDD Rf I/O Rf = 1.2 MΩ (typ.) RO XIN RO = 0.5 kΩ (typ.) XOUT Sink open-drain output Hysteresis input Pull-up resistor VDD RIN R RESET I/O Address-trap-reset Watchdog-timer-reset System-clock-reset RIN = 220 kΩ (typ.) R = 1 kΩ (typ.) Hysteresis input VDD R = 1 kΩ (typ.) STOP / INT5 (P20) Input R P20/ STOP / INT5 Pull-down resistor VDD RIN = 70 MΩ (typ.) R TEST R = 1 kΩ (typ.) Input RIN Osc. enable OSC1 OSC2 fc VDD Rf I/O OSC1 RO VDD Pin for connecting a resonator for on-screen display Rf = 1.2 MΩ (typ.) RO = 0.5 kΩ (typ.) OSC2 88CS38B-208 2007-09-12 TMP88CS38B/CM38B/CP38B (2) Input/output ports Port I/O Input/Output Circuitry VDD Initial “High-Z” P20 Remarks Sink open-drain output Hysteresis input R = 1 kΩ (typ.) I/O R P30 to P33 P50, P57 Tri-state I/O Hysteresis input VDD Initial “High-Z” I/O R = 1 kΩ (typ.) R Disable P70, P71 P34, P35, P51, P52 Open drain output enable I/O R Disable R = 1 kΩ (typ.) P40 I/O R Disable P47 P53 Tri-state I/O Hysteresis input Key-on wakeup input (VIL4 = 0.65 × VDD) VDD Initial “High-Z” to R = 1 kΩ (typ.) Tri-state I/O VDD Initial “High-Z” to Tri-state I/O or open-drain output programmable Hysteresis input VDD Initial “High-Z” R Disable I/O R = 1 kΩ (typ.) RA = 5 kΩ (typ.) P56 CA = 22 pF (typ.) CA RA Key-on wakeup 88CS38B-209 2007-09-12 TMP88CS38B/CM38B/CP38B Port I/O Input/Output Circuitry Remarks Sink open-drain output High current output IOL = 20 mA (typ.) VDD Initial “High-Z” P60, P61 I/O R Disable RA = 5 kΩ (typ.) CA = 22 pF (typ.) CA Initial “High-Z” P62 (at CSOUT) RA Key-on wakeup Key-on wakeup input (VIL4 = 0.65 × VDD) Tri-state I/O High current output IOL = 20 mA (typ.) VDD I/O R Disable I/O R Disable Initial “High-Z” P67 R = 1 kΩ (typ.) Tri-state I/O VDD R = 1 kΩ (typ.) P64 to R = 1 kΩ (typ.) Sink open-drain output High current output IOL = 20 mA (typ.) VDD Initial “High-Z” P62, P63 R = 1 kΩ (typ.) I/O R Disable 88CS38B-210 2007-09-12 TMP88CS38B/CM38B/CP38B Electrical Characteristics (VSS = 0 V) Absolute Maximum Ratings Parameter Symbol Pins Supply voltage VDD − −0.3 to 6.5 Input voltage VIN − −0.3 to VDD + 0.3 Output voltage VOUT1 − −0.3 to VDD + 0.3 Output current (Per 1 pin) Output current (Total) Ratings IOUT1 Ports P2, P3, P4, P5, P64 to P67, P7 IOUT2 Ports P60 to P63 Unit V 3.2 30 mA Σ IOUT1 Ports P2, P3, P4, P5, P64 to P67, P7 Σ IOUT2 Ports P60 to P63 120 120 TMP88CS38: 600 TMP88CP38A/CM38A: 400 Power dissipation [Topr = 70°C] PD − Soldering temperature (Time) Tsld − Storage temperature Tstg − −55 to 125 Operating temperature Topr − −30 to 70 mW 260 (10 s) °C Note: The absolute maximum ratings are rated values which must not be exceeded during operation, even for an instant. Any one of the ratings must not be exceeded. If any absolute maximum rating is exceeded, a device may break down or its performance may be degraded, causing it to catch fire or explode resulting in injury to the user. Thus, when designing products which include this device, ensure that no absolute maximum rating value will ever be exceeded. (VSS = 0 V, Topr = −30 to 70°C) Recommended Operating Conditions Parameter Symbol Pins Conditions Min Max 4.5 5.5 Unit Fc = 16 MHz NORMAL mode Supply voltage Fc = 16 MHz IDLE mode VDD STOP mode Input high voltage Input low voltage Clock frequency VIH1 Except hysteresis input VIH2 Hysteresis input VIL1 Except hysteresis input VIL2 Hysteresis input VIL4 Key-on wakeup input VDD = 4.5 to 5.5V XIN, XOUT VDD = 4.5 to 5.5V OSC1, OSC2 VDD × 0.75 VDD = 4.5 to 5.5V VDD V VDD × 0.30 VDD = 4.5 to 5.5V fc fOSC VDD × 0.70 VDD = 4.5 to 5.5V 0 VDD × 0.25 VDD × 0.65 8.0 16.0 fc = 8 MHz 8.0 12.0 fc = 16 MHz 16.0 24.0 MHz Note 1: The recommended operating conditions for a device are operating conditions under which it can be guaranteed that the device will operate as specified. If the device is used under operating conditions other than the recommended operating conditions (Supply voltage, operating temperature range, specified AC/DC values etc.), malfunction may occur. Thus, when designing products which include this device, ensure that the recommended operating conditions for the device are always adhered to. Note 2: Clock frequency fc: Supply voltage range is specified in NORMAL mode and IDLE mode. Note 3: Smaller value is alternatively specified as the maximum value. 88CS38B-211 2007-09-12 TMP88CS38B/CM38B/CP38B (VSS = 0 V, Topr = −30 to 70°C) DC Characteristics Parameter Symbol Hysteresis voltage Input current Pins Conditions Typ. Max Unit − 0.9 − V VHS Hysteresis inputs IIN1 TEST VDD = 5.5 V, VIN = 5.5 V/0 V − − ±2 IIN2 Open-drain ports VDD = 5.5 V, VIN = 5.5 V/0 V − − ±2 IIN3 Tri-state ports VDD = 5.5 V, VIN = 5.5 V/0 V − − ±2 IIN4 RESET , STOP VDD = 5.5 V, VIN = 5.5 V/0 V Input resistance RIN2 RESET VDD = 5.5 V, VIN = 0 V Output leakage current ILO1 Sink open-drain ports ILO2 Output high voltage VOH2 Output low voltage Output low current − − ±2 100 220 450 VDD = 5.5 V, VOUT = 5.5 V − − 2 Tri-state ports VDD = 5.5 V, VOUT = 5.5 V/0 V − − ±2 Tri-state ports VDD = 4.5 V, IOH = − 0.7 mA 4.1 − − VOL Except XOUT, OSC2 and ports P60 to P63 VDD = 4.5 V, IOL = 1.6 mA − − 0.4 IOL3 Port P60 to P63 VDD = 4.5 V, VOL = 1.0 V − 20 − VDD = 5.5 V fc = 16 MHz VIN = 5.3 V/0.2 V − 25 30 − 20 25 − 0.5 10 Supply current in NORMAL mode Supply current in IDLE mode Min − IDD kΩ μA V (Note 3) VDD = 5.5 V VIN = 5.3 V/0.2 V Supply current in STOP mode μA mA μA Note 1: Typical values show those at Topr = 25°C, VDD = 5 V. Note 2: Input current IIN3: The current through resistor is not included. Note 3: Supply current IDD: The current (Typ. 0.5 mA) through ladder resistors of ADC is included in NORMAL mode and IDLE mode. AD Conversion Characteristics Parameter Symbol (VSS = 0 V, VDD = 4.5 V to 5.5 V, Topr = −30 to 70°C) Conditions Min Typ. Max VAREF supplied from VDD pin. − VDD − VASS supplied from VSS pin. − 0 − = VDD − VSS − VDD − VSS − VDD − − ±1 − − ±2 Full scale error − − ±2 Total error − − ±3 Analog reference voltage Analog reference voltage range Analog input voltage ΔVAREF VAIN Nonlinearity error Zero point error VDD = 5.0 V Unit V LSB Note: The total error means all error except quanting error. 88CS38B-212 2007-09-12 TMP88CS38B/CM38B/CP38B (VSS = 0 V, VDD = 4.5 V to 5.5 V, Topr = −30 to 70°C) AC Characteristics Parameter Symbol Conditions Min Typ. Max Unit 0.5 − 1.0 μs 31.25 − − ns in NORMAL mode Machine cycle time tcy in IDLE mode High level clock pulse width TWCH for external clock operation Low level clock pulse width TWCL (XIN input), fc = 16 MHz Recommended Oscillating Conditions Parameter High-frequency oscillation Oscillator Ceramic resonator (VSS = 0 V, VDD = 4.5 V to 5.5 V, Topr = −30 to 70°C) Oscillation Frequency Recommended Oscillator Recommended Constant C1 C2 8 MHz Murata CSA 8.00MTZ 30 pF 30 pF 16 MHz Murata CSA 16.00MXZ040 5 pF 5 pF XIN XOUT C1 C2 High-frequency oscillation Note 1: To keep reliable operation, shield the device electrically with the metal plate on its package mold surface against the high electric field, for example, by CRT (Cathode ray tube). Note 2: The product numbers and specifications of the resonators by Murata Manufacturing Co., Ltd. are subject to change. For up-to-date information, please refer to the following URL: http://www.murata.co.jp/search/index.html 88CS38B-213 2007-09-12 TMP88CS38B/CM38B/CP38B Recommended Oscillating Conditions Item (VSS = 0 V, VDD = 4.5 V to 5.5 V, Topr = −30 to 70°C) Oscillation for OSD Recommended Parameter Value Oscillation Resonator Frequency L (μH) C1 (pF) C2 (pF) 8 MHz 33 5 to 30 10 12 MHz 15 5 to 30 10 16 MHz 10 5 to 30 10 20 MHz 6.8 5 to 25 10 24 MHz 4.7 5 to 25 10 LC resonator OSC1 OSC2 L C2 C1 Oscillation for OSD The frequency generated in LC oscillation can be obtained using the following equations. f= 1 2π LC ,C = C1・C2 C1 + C2 C1 is not fixed at a constant value. It can be changed to tune into the desired frequency. Note 1: Toshiba’s OSD circuit determines a horizontal display start position by counting clock pulses generated in LC oscillation. For this reason, the OSD circuit may fail to detect clock pulses normally, resulting in the horizontal start position becoming unstable, at the beginning of oscillation, if the oscillation amplitude is low. Changing L and C2 from the values recommended for a specific frequency may hamper a stable OSD display. If the LC oscillation frequency is the same as a high-frequency clock value, the oscillation of the high-frequency oscillator may cause the LC oscillation frequency to fluctuate, thus making OSD displays flicker. When determining these parameters, please check the oscillation frequency and the stability of oscillation on your TV sets. Also check the determined parameters on your final products, because the optimum parameter values may vary from one product to another. Note 2: When using the LSI package in a strong electric field, such as near a CRT, electrically shield the package so that its normal operation can be maintained. 88CS38B-214 2007-09-12 TMP88CS38B/CM38B/CP38B Notice of ROM Entry When you make a ROM data entry for TMP88CS38B and TMP88CM38B/CP38B, Please transfer one file including program area, vector table area and OSD font area. The ROM area must be transferred is as follows. TMP88CS38B 4000H TMP88CM38B TMP88CP38B 4000H 4000H Program area Program area Program area BEFFH FEFFH 13EFFH 20000H 20000H 20000H OSD font area OSD font area OSD font area 25FFFH 25FFFH 25FFFH FFF00H FFF00H FFF00H Vector table area Vector table area Vector table area FFFFFH FFFFFH FFFFFH Flow of ROM data entry After evaluation finished Program and vector table OSD font Program vector table OSD font Two files are merged into one file. ROM data entry 88CS38B-215 2007-09-12 TMP88CS38B/CM38B/CP38B Package P-SDIP42-600-1.78 Unit: mm 88CS38B-216 2007-09-12 TMP88CS38B/CM38B/CP38B P-QFP44-1414-0.80K Unit: mm 88CS38B-217 2007-09-12 TMP88CS38B/CM38B/CP38B 88CS38B-218 2007-09-12