These are single-chip microcomputers designed with high-performance CMOS silicon gate technology, including the internal flash
memory. These microcomputers support the 7900 Series instruction
set, which are enhanced and expanded instruction set and are upper-compatible with the 7700/7751 Series instruction set.
The CPU of these microcomputers is a 16-bit parallel processor that
can also be switched to perform 8-bit parallel processing. Also, the
bus interface unit of these microcomputers enhances the memory
access efficiency to execute instructions fast. Therefore, these microcomputers are suitable for office, business, and industrial equipment controller that require high-speed processing of large data.
For the internal flash memory, single-power-supply programming
and erasure, using a PROM programmer or the control by the central processing unit (CPU), is supported. Also, each of these microcomputers has the memory area dedicated for storing a certain
software which controls programming and erasure (reprogramming
control software). Therefore, on these microcomputers, the program
can easily be changed even after they are mounted on the board.
DISTINCTIVE FEATURES
<Microcomputer mode>
Number of basic machine instructions .................................... 203
•
Memory
•
[M37902FCCHP]
Flash memory (User ROM area) ................................. 120 Kbytes
Serial I/O
A-D converter
D-A converter
Watchdog timer
Chip-select wait control
Flash memory (User ROM area)
RAM
Flash memory (Boot ROM area)
P0–P2, P4–P8, P10, P11
P3
TA0–TA4
TB0–TB2
UART0 and UART1
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
FunctionsParameter
203
38 ns (the fastest instruction at f(fsys) = 26 MHz)
26 MHz (Max.)
26 MHz (Max.)
(Note)
(Note)
16 Kbytes
8-bit ✕ 10
4-bit ✕ 1
16-bit ✕ 5
16-bit ✕ 3
(UART or Clock synchronous serial I/O) ✕ 2
10-bit successive approximation method ✕ 1 (8 channels)
8-bit ✕ 3
12-bit ✕ 1
Chip select area ✕ 4 (CS0–CS3). A bus cycle type and bus width
can be set for each chip select area.
Real-time output
Interrupts
Clock generating circuit
PLL frequency multiplier
Power supply voltage
Power dissipation
Ports’ input/output
characteristics
Memory expansion
Operating ambient temperature range
Device structure
Package
Note:
Flash memoryM37902FCCHP120 Kbytes
(User ROM area)M37902FGCHP248 Kbytes
RAMM37902FCCHP4096 bytes
Maskable interrups
Non-maskable interrups
Input/Output withstand voltage
Output current
M37902FJCHP498 Kbytes
M37902FGCHP6144 bytes
M37902FJCHP12288 bytes
4 bits ✕ 2 channels; or 6 bits ✕ 1 channel + 2 bits ✕ 1 channel
5 external types, 13 internal types. Each interrupt can be set to a
priority level within the range of 0–7 by software.
1 external type, 3 internal types.
Built-in (externally connected to a ceramic resonator or quartz
crystal resonator).
The following multiplication methods are available: double, triple,
and quadruple.
5 V±0.5 V
150 mW (at f(fsys) = 26 MHz, Typ., PLL frequency multiplier
stopped)
5 V
5 mA
Up to 16 Mbytes. Note that bank FF16 is a reserved area.
–20 to 85 °C
CMOS high-performance silicon gate process
100-pin plastic molded QFP
4
Page 5
M37902FCCHP, M37902FGCHP, M37902FJCHP
FUNCTIONS (Flash memory mode)
Power supply voltage
Programming/Erase voltage
Flash memory mode
Block division for erasure
Programming method
Erase method
Programming/Erase control
Data protection method
Number of commands
Maximum number of reprograms
User ROM area
Boot ROM area
Flash memory parallel I/O mode
Flash memory serial I/O mode
Flash memory CPU reprogramming mode
Flash memory parallel I/O mode
Flash memory serial I/O mode
Flash memory CPU reprogramming mode
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
FunctionsParameter
5 V±0.5 V (in the flash memory parallel I/O mode, 3.3 V±0.3 V)
5 V±0.5 V (in the flash memory parallel I/O mode, 3.3 V±0.3 V)
3 modes: parallel I/O, serial I/O, and CPU reprogramming modes
(Note 1)
1 block (16 Kbytes ✕ 1) (Note 2)
Programmed per page (in a unit of 256 Kbytes)
User ROM area + Boot ROM area
User ROM area
User ROM area
Total erase/Block erase
User ROM area + Boot ROM area
User ROM area
User ROM area
Programming/Erase control by software commands
Protected per block, by using a lock bit.
8 commands
100
Notes 1:
User ROM area
2:
On shipment, our reprogramming control firmware for the flash memory serial I/O mode has been stored into the boot ROM area.
Note that the boot ROM area can be erased/programmed only in the flash memory parallel I/O mode.
Apply 5 V±0.5 V to Vcc, and 0 V to Vss.
This pin controls the processor mode. Connect this pin to VSS for the single-chip
mode or memory expansion mode, and VCC for the microprocessor mode.
Connect this pin to Vss.
The microcomputer is reset when “L” level is applied to this pin.
These are input and output pins of the internal clock generating circuit. Connect a
ceramic or quartz- crystal resonator between the XIN and XOUT pins. When an
external clock is used, the clock source should be connected to the XIN pin, and the
XOUT pin should be left open.
This pin determines whether the external data bus has an 8-bit width or 16-bit width
for the memory expansion mode or microprocessor mode. The width is 16 bits when
“L” signal is input, and 8 bits when “H” signal is input. When BYTE = Vss level, by
the register setting, the external data bus for each of areas CS1 to CS3 can have a
width of 8 bits.
When using the PLL frequency multiplier, connect this pin to the filter circuit. When
not using, this pin should be left open.
Power supply input pins for the A-D converter and the D-A converter. Connect AVcc
to Vcc, and AVss to Vss externally.
This is the reference voltage input pin for the A-D converter and the D-A converter.
■ In single-chip mode
Port P0 is an 8-bit I/O port. This port has an I/O direction register, and each pin
can be programmed for input or output. These pins enter the input mode at
reset.
■ In memory expansion and microprocessor modes
Address (A16–A23) is output. These pins also function as I/O port pins according
to the register setting.
■ In single-chip mode
These pins have the same functions as port P0.
■ In memory expansion and microprocessor modes
The low-order 8 bits of data (D0–D7) are input/output. When the external data bus
has an 8-bit width, address (LA0–LA7) output and data (D0–D7) input/output can
be performed with the time-sharing method, according to the register setting.
■ In single-chip mode or When 8-bit external data bus is used in memory expansion
mode and microprocessor mode
These pins have the same functions as port P0.
■ When the 16-bit external data bus is used in memory expansion or microprocessor mode
The high-order 8 bits of data (D8–D15) are input or output.
■ In single-chip mode
These pins have the same functions as port P0.
■ In memory expansion mode
P30 functions as an I/O port pin; and P31, P32, and P33 function as the output
pins of RD, BLW, BHW, respectively. P30 also functions as an output pin of RDY
according to the register setting. When the external data bus has a width of 8 bits,
the BHW pin functions as an I/O port pin (P33).
■ In microprocessor mode
P30 functions as an input pin of RDY; and P31,P32, P33 function as the output
pins of RD, BLW, BHW, respectively. P30 also functions as an I/O port pin according to the register setting. When the external data bus has a width of 8 bits,
the BHW pin functions as an I/O port pin (P33).
■ In single-chip mode
These pins have the same functions as port P0.
■ In memory expansion mode
P40–P47 function as I/O port pins. According to the register setting, these pins
function as output pins or input pins of ALE, φ1, HLDA, HOLD, CS0–CS3, respectively.
■ In microprocessor mode
P40–P44 function as output or input pins of ALE, φ1, HLDA, HOLD, CS0, and
P45–P47 as I/O port pins, respectively. According to the register setting, P40–P43
also function as I/O port pins, and P45–P47 as output pins of CS1–CS3.
Vcc, Vss
MD0
MD1
RESET
XIN
XOUT
BYTE
VCONT
AVcc,
AVss
VREF
P00–P07
P10–P17
P20–P27
P30–P33
P40–P47
NamePin
Power supply input
MD0
MD1
Reset input
Clock input
Clock output
External data bus width
select input
Filter circuit connection
Analog power supply input
Reference voltage input
I/O port P0
I/O port P1
I/O port P2
I/O port P3
I/O port P4
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Functions
6
Page 7
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
P50–P57
P60–P67
P70–P77
P80–P87
P100–P107
P110–P117
NMI
NamePin
I/O port P5
I/O port P6
I/O port P7
I/O port P8
I/O port P10
I/O port P11
Non-maskable interrupt
Input/
Output
I/O
I/O
I/O
I/O
I/O
I/O
Input
Functions
In addition to having the same functions as port P0 in the single-chip mode, these
pins also function as I/O pins for timers A0–A3, output pins for the real-time output,
and input pins for the key-input interrupt.
In addition to having the same functions as port P0 in the single-chip mode, these
pins also function as I/O pins for timer A4, input pins for external interrupt inputs
____ ____
INT0–INT2, and input pins for timers B0–B2.
In addition to having the same functions as port P0 in the single-chip mode, these
pins also function as input pins for the A-D converter, output pins for the D-A
converter, and input pins for INT2, INT3, and INT4.
In addition to having the same functions as port P0 in the single-chip mode, these
pins also function as I/O pins for UART0, UART1, output pins for D-A converter,
and input pins for INT3 and INT4.
■ In single-chip mode
These pins have the same functions as port P0.
■ In memory expansion and microprocessor modes
Address (A0–A7) is output.
■ In single-chip mode
These pins have the same functions as port P0.
■ In memory expansion and microprocessor modes
Address (A8–A15) is output. Also, these pins function as I/O port pins according to
the register setting.
Power supply input
MD0
MD1
Reset input
Clock input
Clock output
BYTE
Filter circuit connection
Analog supply input
Reference voltage input
Input port P0
Input port P1
Input port P2
Input port P3
Input port P4
SCLK input
SDA I/O
BUSY output
Input port P5
Input port P6
Input port P7
Input port P8
Input port P10
Input port P11
Non-maskable interrupt
Input
/Output
—
Input
Input
Input
Input
Output
Input
—
—
Input
Input
Input
Input
Input
Input
Input
I/O
Output
Input
Input
Input
Input
Input
Input
Input
Apply 5 V ± 0.5 V to Vcc, and 0 V to Vss.
Connect this pin to Vss.
Connect this pin to Vss via a resistor of 10 kΩ to 100 kΩ.
The reset input pin.
Connect a ceramic resonator between the XIN and XOUT pins, or input an external
clock from the XIN pin with the XOUT pin left open.
Connect this pin to Vcc or Vss. (This is not used in the flash memory serial I/O mode.)
Connect this pin to the filter circuit, or leave this pin open. (This is not used in the flash memory serial I/O mode.)
Connect AVcc to Vcc, and AVss to Vss.
Input an arbitrary level within the range of VSS–VCC. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
This is an input pin for a serial clock.
This is an I/O pin for serial data. Connect this pin to VCC via a resistor (about 1 kΩ).
This is an output pin for the BUSY signal.
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H” or “L”, or leave them open. (This is not used in the flash memory serial I/O mode.)
Input “H”, or leave this pin open.
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Functions
8
Page 9
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
BASIC FUNCTION BLOCKS
These microcomputers contain the following devices on the single
chip: the flash memory, RAM, CPU, bus interface unit, and peripheral devices such as the interrupt control circuit, timers, serial I/O,
A-D converter, D-A converter, I/O ports, clock generating circuit, etc.
MEMORY
Figures 1 to 3 show the memory maps. The address space is 16
Mbytes from addresses 016 to FFFFFF16. The address space is divided into 64-Kbyte units called banks. The banks are numbered
from 016 to FF16. Bank FF16 is a reserved area for the development
support tool. Therefore, do not use bank FF16.
000000
Bank 0
Bank 1
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Bank FE
Bank FF
00FFFF
010000
01FFFF
FE0000
FEFFFF
FF0000
FFFFFF
16
16
16
16
16
16
16
16
Reserved area
for development
support tool
16
16
16
16
000000
0000FF
000800
0017FF
001800
001FFF
002000
00FFC0
00FFFF
Internal flash memory and internal RAM are assigned as shown in
Figures 1 to 3.
Addresses FFC016 to FFFF16 contain the RESET and the interrupt
vector addresses, and the interrupt vectors are stored there.
For details, refer to the section on interrupts.
Assigned to addresses 016 to FF16 are peripheral devices such as
I/O ports, A-D converter, D-A converter, UART, timers, interrupt control registers, etc. Figures 7 and 8 show the location of SFRs.
For the flash memory in the boot ROM area, refer to the section on
the flash memory mode.
16
Peripheral devices
control registers
16
16
Internal RAM
4096 bytes
16
16
16
16
16
16
Internal flash memory
120 Kbytes
(User ROM area)
00FFC0
00FFFE
Interrupt vector table
16
Reserved area
Reserved area
Reserved area
Reserved area
Reserved area
The CPU has 13 registers and is shown in Figure 9. Each of these
registers is described below.
ACCUMULATOR A (A)
Accumulator A is the main register of the microcomputer. It consists
of 16 bits and the low-order 8 bits can be used separately. Data
length flag m determines whether the register is used as 16-bit register or as 8-bit register. It is used as a 16-bit register when flag m is
“0” and as an 8-bit register when flag m is “1”. Flag m is a part of the
processor status register (PS) which is described later.
Data operations such as calculations, data transfer, input/output,
etc., are executed mainly through accumulator A.
ACCUMULATOR B (B)
Accumulator B has the same functions as accumulator A, but the use
of accumulator B requires more instruction bytes and execution
cycles than accumulator A.
ACCUMULATOR E
Accumulator E is a 32-bit register and consists of accumulator A
(low-order 16 bits) and accumulator B (high-order 16 bits). It is used
for 32-bit data processing.
INDEX REGISTER X (X)
Index register X consists of 16 bits and the low-order 8 bits can be
used separately. Index register length flag x determines whether the
register is used as 16-bit register or as 8-bit register. It is used as a
16-bit register when flag x is “0” and as an 8-bit register when flag x
is “1”. Flag x is a part of the processor status register (PS) which is
described later.
In index addressing modes in which register X is used as the index
register, the contents of this address are added to obtain the real address.
Index register X functions as a pointer register which indicates an
address of data table in instructions MVP, MVN, RMPA (Repeat
MultiPly and Accumulate).
INDEX REGISTER Y (Y)
Index register Y consists of 16 bits and the low-order 8 bits can be
used separately. The index register length flag x determines whether
the register is used as 16-bit register or as 8-bit register. It is used as
a 16-bit register when flag x is “0” and as an 8-bit register when flag
x is “1”. Flag x is a part of the processor status register (PS) which is
described later.
In index addressing modes in which register Y is used as the index
register, the contents of this address are added to obtain the real address.
Index register Y functions as a pointer register which indicates an
address of data table in instructions MVP, MVN, RMPA (Repeat
MultiPly and Accumulate).
1570
31
70
PGProgram bank register PG
70
Fig. 9 Register structure
B
H
Accumulator B
Data bank register DTDT
B
L
Accumulator A
1570
Accumulator E
1570
1570
15
1570
150
150
150
15
00000
A
H
A
H
B
H
X
H
H
Y
S
PC
DPR0 to DPR3
IPL2IPL1IPL
0
A
L
A
L
B
L
7
7
NVmxD I ZC
X
L
Y
L
0
0
Index register X
Index register Y
Stack pointer S
Program counter PC
Direct page registers DPR0 to DPR3
0
Processor status register PS
Carry flag
Zero flag
Interrupt disable flag
Decimal mode flag
Index register length flag
Data length flag
Overflow flag
Negative flag
Processor interrupt priority level IPL
13
Page 14
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
STACK POINTER (S)
Stack pointer (S) is a 16-bit register. It is used during a subroutine
call or interrupts. It is also used during stack, stack pointer relative,
or stack pointer relative indirect indexed Y addressing mode.
PROGRAM COUNTER (PC)
Program counter (PC) is a 16-bit counter that indicates the low-order
16 bits of the next program memory address to be executed. There
is a bus interface unit between the program memory and the CPU,
so that the program memory is accessed through bus interface unit.
This is described later.
PROGRAM BANK REGISTER (PG)
Program bank register is an 8-bit register that indicates the high-order 8 bits of the next program memory address to be executed.
When a carry occurs by incrementing the contents of the program
counter, the contents of the program bank register (PG) is increased
by 1. Also, when a carry or borrow occurs after adding or subtracting
the offset value to or from the contents of the program counter (PC)
using the branch instruction, the contents of the program bank register (PG) is increased or decreased by 1, so that programs can be
written without worrying about bank boundaries.
DATA BANK REGISTER (DT)
Data bank register (DT) is an 8-bit register. With some addressing
modes, the data bank register (DT) is used to specify a part of the
memory address. The contents of data bank register (DT) is used as
the high-order 8 bits of a 24-bit address. Addressing modes that use
the data bank register (DT) are direct indirect, direct indexed X indirect, direct indirect indexed Y, absolute, absolute bit, absolute indexed X, absolute indexed Y, absolute bit relative, and stack pointer
relative indirect indexed Y.
DIRECT PAGE REGISTERS 0 to 3 (DPR0 to DPR3)
The direct page register is a 16-bit register. An addressing mode of
which name includes ‘direct’ generates an address of data to be accessed, regarding the contents of this register as the base address.
The 7900 Series has been expanded direct page registers up to 4
(DPR0 to DPR3), in comparison to the 7700 Series which has the
single direct page register. Accordingly, the 7900 Series’s direct ad-
dressing method which uses direct page registers differs from that of
the 7700 Series. However, the conventional direct addressing
method, using only DPR0, is still be selectable, in order to make use
of the 7700 Series software property. For more details, refer to the
section on the direct page.
PROCESSOR STATUS REGISTER (PS)
Processor status register (PS) is an 11-bit register. It consists of
flags to indicate the result of operation and CPU interrupt levels.
Branch operations can be performed by testing the flags C, Z, V , and
N.
The details of each bit of the processor status register are described
below.
1. Carry flag (C)
The carry flag contains the carry or borrow generated by the ALU after an arithmetic operation. This flag is also affected by shift and rotate instructions. This flag can be set and reset directly with the SEC
and CLC instructions or with the SEP and CLP instructions.
2. Zero flag (Z)
The zero flag is set if the result of an arithmetic operation or data
transfer is zero and reset if it is not. This flag can be set and reset
directly with the SEP and CLP instructions.
3. Interrupt disable flag (I)
When the interrupt disable flag is set to “1”, all interrupts except
watchdog timer, NMI, and software interrupt are disabled. This flag
is set to “1” automatically when an interrupt is accepted. It can be set
and reset directly with the SEI and CLI instructions or SEP and CLP
instructions.
___
4. Decimal mode flag (D)
The decimal mode flag determines whether addition and subtraction
are performed as binary or decimal. Binary arithmetic is performed
when this flag is “0”. If it is “1”, decimal arithmetic is performed with
each word treated as 2- or 4- digit decimal. Arithmetic operation is
performed using four digits when data length flag m is “0” and with
two digits when it is “1”. Decimal adjust is automatically performed.
(Decimal operation is possible only with the ADC and SBC instructions.) This flag can be set and reset with the SEP and CLP instructions.
14
Page 15
M37902FCCHP, M37902FGCHP, M37902FJCHP
5. Index register length flag (x)
The index register length flag determines whether index register X
and index register Y are used as 16-bit registers or as 8-bit registers.
The registers are used as 16-bit registers when flag x is “0” and as 8bit registers when it is “1”.
This flag can be set and reset with the SEP and CLP instructions.
6. Data length flag (m)
The data length flag determines whether the data length is 16-bit or
8-bit. The data length is 16 bits when flag m is “0” and 8 bits when it
is “1”. This flag can be set and reset with the SEM and CLM instructions or with the SEP and CLP instructions.
7. Overflow flag (V)
The overflow flag is valid when addition or subtraction is performed
with a word treated as a signed binary number. If data length flag m
is “0”, the overflow flag is set when the result of addition or subtraction is outside the range between –32768 and +32767. If data length
flag m is “1”, the overflow flag is set when the result of addition or
subtraction is outside the range between –128 and +127. It is reset
in all other cases. The overflow flag can also be set and reset directly
with the SEP, and CLV or CLP instructions.
Additionally, the overflow flag is set when a result of unsigned/signed
division exceeds the length of the register where the result is to be
stored; the flag is also set when the addition result is outside range
of –2147483648 to +2147483647 in the RMPA operation.
MITSUBISHI MICROCOMPUTERS
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
8. Negative flag (N)
The negative flag is set when the result of arithmetic operation or
data transfer is negative (If data length flag m is “0”, data’s bit 15 is
“1”. If data length flag m is “1”, data’s bit 7 is “1”.) It is reset in all other
cases. It can also be set and reset with the SEP and CLP instructions.
9. Processor interrupt priority level (IPL)
The processor interrupt priority level (IPL) consists of 3 bits and determines the priority of processor interrupts from level 0 to level 7.
Interrupt is enabled when the interrupt priority of the device requesting interrupt (set using the interrupt control register) is higher than
the processor interrupt priority . When an interrupt is enabled, the current processor interrupt priority level is saved in a stack and the processor interrupt priority level is replaced by the interrupt priority level
of the device requesting the interrupt. Refer to the section on interrupts for more details.
Note: Fix bits 11 to 15 of the processor status register (PS) to “0”.
15
Page 16
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
BANK
In order to effectively use the integrated hardware on the chip, this
CPU core uses an address generating method with a 24-bit address
split into high-order 8 bits and low-order 16 bits. In other words, the
64 Kbytes specified by the low-order 16 bits are one unit (referred to
as “bank”), and the address space is divided into 256 banks (016 to
FF16) specified by the high-order 8 bits.
In the program area on the address space, the bank is specified by
the program bank register (PG), and the address in the bank is
specified by the program counter (PC).
As for each bank boundary, when an overflow has occurred in PC,
the contents of PG are incremented by 1. When a borrow has occurred in PC, the contents of PG are decremented by 1. Under the
normal conditions, therefore, programming without concern for the
bank boundaries is possible. Furthermore, as for the data area on
the address space, the bank is specified by the data bank register
(DT), and the address in the bank is specified by the operation result
by using the various addressing modes (Note).
Note: Some addressing modes directly specify a bank.
DIRECT PAGE
The internal memory and control registers for internal peripheral devices, etc. are assigned to bank 016 (addresses 016 to FFFF16). The
direct page and direct addressing modes have been provided for the
effective access to bank 016. In the 7900 Series, two types of direct
addressing modes are available: the conventional direct addressing
mode which uses only DPR0, as in the 7700 Series, and the expanded direct addressing mode, which uses up to 4 direct page registers as selected by the user. The addressing mode is selected
according to the contents of bit 1 of the processor mode register 1.
This bit 1 is cleared to “0” at reset. (In other words, the conventional
direct addressing mode is selected.) However, once this bit 1 has
been set to “1” by software, this bit cannot be cleared to “0” again,
except by reset. That is to say , when one of these two direct addressing modes has been selected just after reset, the selected addressing mode cannot be switched to another one while the program is
running.
Refer to “7900 Series Software Manual” for details concerning the
various addressing modes which use the direct page area.
Instruction Set
The CPU core of the 7900 Series has an expanded instruction set
based on the existing 7700/7751 Series’ CPU core. In addition, its
source code (mnemonic) has the complete upper compatibility with
the 7700 Series instruction set.
For details concerning addressing modes and instruction set, refer to
“7900 Series Software Manual”.
■ Conventional direct addressing mode
The direct page area consists of 256-byte space. Its bank address is
“0016”, and the base address of its low-order 16-bit address is specified by the contents of the direct page register 0 (DPR0). In this conventional direct addressing modes, a value (1 byte) just after an
instruction code is regarded as an offset value for the DPR0 contents, and the CPU accesses each address in the direct page area.
■ Expanded direct addressing mode
The direct page area consists of four 64-byte spaces. Their bank
address is “0016”, and the four base addresses of their low-order 16bit addresses are respectively specified by the contents of four direct
page registers. In this expanded direct addressing mode, a value (1
byte) just after an instruction code is regarded as follows:
• High-order 2 bits: regarded as a selection field for DPR0 to DPR3.
• Low-order 6 bits: regarded as an offset value for the selected direct
page register.
Then, the CPU accesses each address in each direct page area:
16
Page 17
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
BUS INTERFACE UNIT
Data transfer between the central processing unit (CPU) and internal memory, internal peripheral devices, or external areas is always
performed via the bus interface unit (BIU), which is located between
the CPU and the internal buses.
Figure 10 shows the BIU and the bus structure. The CPU and BIU
are connected by a dedicated bus, and any transfer between the
CPU and BIU is controlled by this dedicated bus.
On the other hand, data transfer between the BIU and internal peripheral devices uses the following internal common buses: 32-bit
code bus, 16-bit data bus, 24-bit address bus, and control signals.
The bus control method where the code bus and the data bus separate out (hereafter, this method is referred to as the separate code/
M37902
Central
Processing
Unit
(CPU)
CPU bus
Bus
Interface
Unit
(BIU)
Internal code bus (CB0 to CB31)
Internal data bus (DB0 to DB15)
Internal address bus (AD0 to AD23)
Internal control signal
data bus method) is employed in order to improve data transfer capabilities. As a result, the internal memory is connected to both the
code bus and the data bus, and registers of all other internal peripheral devices are connected only to the data bus.
Each width of external buses are as follows: a 24-bit address bus,
16-bit data bus.
The external data bus transfers instruction codes and data. When
the code or data access occurs for the external, the external access
is performed via the bus conversion circuit.
For details of the connection with the external devices, refer to the
section on the processor modes and chip select wait controller described later.
Internal buses
Internal
memory
Hold request
SFR : Special Function Register
❈ The CPU bus, internal bus, and external bus separate out independently.
Fig. 10 BIU and bus structure
Internal
peripheral
devices
(SFR)
conversion
circuit
Bus
External bus
A0 to A
23
D0 to D7 (LA0 to LA7)
8
to D
15
D
Control signal
HOLD
HLDA
External
devices
17
Page 18
M37902FCCHP, M37902FGCHP, M37902FJCHP
BIU structure
The BIU consists of four registers shown in Figure 11. Table 1 lists
the functions of each register.
Table 1. Functions of each register
Name
Program address register
Instruction queue buffer
Data address register
Data buffer
Indicates a storage address for an instruction to be next taken into an instruction queue buffer.
Temporarily stores an instruction which has been taken from a memory. Consists of 10 bytes.
Indicates an address where data will be next read from or written to.
Temporarily stores data which has been read from internal memory, internal peripheral devices, and
external areas by the BIU; or temporarily stores data which is to be written to internal memory, internal
peripheral devices, and external areas by the CPU. Consists of 32 bits.
MITSUBISHI MICROCOMPUTERS
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Functions
Fig.11 Register structure of BIU
b23
b0
PA
b7b0
Q0
Q9
b23b0
DA
b31b0
DQ
Program address register
Instruction queue buffer
Data address register
Data buffer
18
Page 19
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
BIU Functions
(1) Instruction prefetch
The BIU has ten instruction queue buffers; each buffer consists of 1
byte. When there is an opening in the bus and the instruction queue
buffer, an instruction code is read from the program memory (in other
words, the memory where a program is stored) and prefetched into
an instruction queue buffer. The prefetched instruction code is transferred from the BIU to the CPU, in response to a request from the
CPU, via a dedicated bus.
When a branch occurs as a result of a branch instruction (JMP, BRA,
etc.), subroutine call, or interrupt, the contents of the instruction
queue buffer are initialized and the BIU reads a new instruction from
the branch destination address.
Note that the operations of the BIU instruction prefetch also differ depending on the store addresses for instructions. The store addresses
for instructions to be prefetched are categorized as listed in Table 2.
(2) Data read operation
When executing an instruction for reading data from the internal
memory, internal peripheral devices, or external areas, at first, the
CPU informs the BIU’s data address register of the address where
the data has been located.
Next, the BIU reads the above data from the specified address,
passes it to the data buffer, and then, transfers it to the CPU.
[Instruction prefetch]
• Whether the address area locates in the internal area or the external area.
• When the address area locates in the external area
➀ Whether the external bus width = 16 bits or 8 bits:
(a) When the external bus width = 16 bits:
whether the start address for access locates at a 4byte boundary or at an 8-byte boundary.
(b) When the external bus width = 8 bits:
whether the start address for access locates at an
even address, a 4-byte boundary or at the 8-byte bound
ary.
➁ Whether the prefetch operation is generated by a branch, or
not.
➂ Number of waits
➃ Whether the burst ROM access is specified or not.
Table 2. Store addresses for instructions to be prefetched
Low-order 3 bits of store address for instruction
AD1 (A1)
X
0
0
AD0 (A0)
0
0
0
Even address
4-byte boundary
8-byte boundary
X: 0 or 1
AD2 (A2)
X
X
0
(3) Data write operation
When executing an instruction for writing data into the internal
memory, internal peripheral devices, or external area, at first, the
CPU informs the BIU’s data address register of the address where
the data has been located.
Next, the BIU passes the above data to the data buffer register, and
then, writes it into the specified address.
(4) Bus cycle
In order for the BIU to execute the above operations (1) through (3),
the 24-bit address bus, 32-bit code bus, 16-bit data bus and internal
control signals must be appropriately controlled during data transfer
between the BIU and internal memory, internal peripheral devices,
external areas. This operation is called “bus cycle”. The bus cycle is
affected by the following conditions at instruction prefetch and data
access.
[Data Access]
• Whether the address area locates in the internal area or the external area.
• Length of data to be transferred: byte, word, double word
• When the address area locates in the external area:
➀ Whether the external bus width = 16 bits or 8 bits:
➁ Number of waits
The BIU controls the bus cycle depending on the above conditions.
Figures 12 to 16 show the bus cycle waveform examples for instruction prefetch and data access.
19
Page 20
MITSUBISHI MICROCOMPUTERS
Access to internal area
Access to external area
When address locates at 4-byte boundary or when branched:
double consecutive access
When branched or at instruction
prefetch
When external data bus width = 16 bitsWhen external data bus width = 8 bits
φ
BIU
Internal address bus
Internal code bus
CB
0
to CB
31
Code
φ
1
A
0
to A
23
D
0
to D
7
D
8
to D
15
ALE
RD
BLW
BHW
D
0
to D
7
D
0
to D
7
D
8
to D
15
D
8
to D
15
AddressAddress + 2
When address of instruction to be prefetched locates at 8-byte boundary:
quadruple consecutive access
φ
1
A
0
to A
23
D
0
to D
7
D
8
to D
15
ALE
RD
BLW
BHW
D
0
to D
7
D
0
to D
7
D
0
to D
7
D
0
to D
7
D
8
to D
15
D
8
to D
15
D
8
to D
15
D
8
to D
15
AddressAddress + 2
Address + 4Address + 6
When address is even address or when branched:
double consecutive access
φ
1
A
0
to A
23
D
0
to D
7
ALE
RD
BLW
BHW
D
0
to D
7
D
0
to D
7
AddressAddress + 1
When address of instruction to be prefetched locates at 4-byte boundary or
8-byte boundary: quadruple consecutive access
φ
1
A
0
to A
23
D
0
to D
7
ALE
RD
BLW
BHW
D
0
to D
7
D
0
to D
7
D
0
to D
7
D
0
to D
7
AddressAddress + 1
Address + 2Address + 3
Address
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Fig. 12 Bus cycle waveform example for instruction prefetch
20
Page 21
8-bit
data
read
8-bit
data
written
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Access starting from even addressAccess starting from odd address
φ
Internal address bus
Internal data bus
DB0 to DB7
DB8 to DB15
φ
BIU
Internal address bus
Internal data bus
DB0 to DB7
DB8 to DB15
BIU
Address
D0 to D7
Invalid
Address
D0 to D7
Internal address bus
Internal address bus
φ
BIU
Internal data bus
DB0 to DB7
DB8 to DB15
φ
BIU
Internal data bus
DB0 to DB7
DB8 to DB15
Address
Address
Invalid
D8 to D
D8 to D
15
15
φ
BIU
16-bit
data
read
Access to internal area
16-bit
data
written
32-bit
data
read
32-bit
data
written
Internal address bus
Internal data bus
DB0 to DB7
DB8 to DB15
A0 to A23
D0 to D7
D8 to D15
φ
BIU
Internal address
bus
Internal data bus
DB0 to DB7
DB8 to DB15
φ
BIU
Internal address
bus
Internal data bus
DB0 to DB7
DB8 to DB15
φ
1
Address
Address
Address
Address
D8 to D
D8 to D
D0 to D7
D8 to D
D0 to D7
D8 to D
15
15
15
15
Address + 2
Address + 2
D0 to D7D0 to D7
D8 to D
D0 to D7D0 to D7
D8 to D
φ
BIU
D8 to D
D8 to D
15
15
Invalid
15
15
Address + 1
Address + 1
D0 to D7
D8 to D
D0 to D7
D8 to D
D0 to D7
Invalid
D0 to D7
15
15
Address + 3
D0 to D7
Invalid
Address + 3
D0 to D7
Internal address bus
Internal data bus
DB0 to DB7
DB8 to DB15
φ
1
A0 to A23
D0 to D7
D8 to D15
φ
BIU
Internal address
bus
Internal data bus
15
15
DB0 to DB7
DB8 to DB15
φ
Internal address
bus
Internal data bus
DB0 to DB7
DB8 to DB15
BIU
Address
Address
AddressAddress + 1
Invalid
D8 to D
AddressAddress + 1
D8 to D
Fig. 13 Bus cycle waveform example for data access (access to internal area)
21
Page 22
8-bit
data
read
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Access starting from even addressAccess starting from odd address
φ
A0 to A
D0 to D
D8 to D
BHW
1
ALE
RD
BLW
23
7
15
Address
D0 to D
Invalid
7
φ
A0 to A
D0 to D
D8 to D
BHW
1
ALE
RD
BLW
23
7
15
Address
D8 to D
Invalid
15
φ
1
A0 to A
23
D0 to D
D8 to D
BHW
φ
A0 to A
D0 to D
D8 to D
BHW
φ
A0 to A
D0 to D
D8 to D
BHW
7
15
ALE
RD
BLW
1
23
7
15
ALE
RD
BLW
1
23
7
15
ALE
RD
BLW
8-bit
data
written
16-bit
External data bus width = 16 bits
data
read
16-bit
data
written
Address
D0 to D
Address
D7 to D
D8 to D
Address
D0 to D
D8 to D
φ
1
A0 to A
23
D0 to D
D8 to D
BHW
φ
A0 to A
D0 to D
D8 to D
BHW
φ
A0 to A
D0 to D
D8 to D
BHW
7
15
ALE
RD
BLW
1
23
7
15
ALE
RD
BLW
1
23
7
15
ALE
RD
BLW
7
0
15
7
15
Address
D8 to D
15
AddressAddress + 1
Invalid
D8 to D
15
D0 to D
Invalid
AddressAddress + 1
D0 to D
D8 to D
15
7
7
Fig. 14 Bus cycle waveform example for data access (access to external area) (1)
22
Page 23
32-bit
data
read
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Access starting from even addressAccess starting from odd address
φ
A0 to A
D0 to D
D8 to D
BHW
1
ALE
RD
BLW
23
7
15
AddressAddress + 2
D0 to D
D8 to D
7
15
D0 to D
D8 to D
7
15
φ
A0 to A
D0 to D
D8 to D
BLW
BHW
1
ALE
RD
23
7
15
AddressAddress + 1Address + 3
Invalid
D8 to D
15
D0 to D
D8 to D
7
15
D0 to D
Invalid
7
φ
A0 to A
D0 to D
D8 to D
BLW
BHW
1
23
7
15
ALE
RD
φ
1
A0 to A
23
D0 to D
32-bit
External data bus width = 16 bits
data
written
D8 to D
BHW
7
15
ALE
RD
BLW
AddressAddress + 2
D0 to D
D8 to D
D0 to D
D8 to D
7
15
7
15
Fig. 15 Bus cycle waveform example for data access (access to external area) (2)
AddressAddress + 1Address + 3
D0 to D
D8 to D
7
15
D8 to D
15
D0 to D
7
23
Page 24
32/16/
8-bit
data
read
A0 to A
D0 to D
D8 to D
BHW
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Access starting from even or odd address
φ
1
ALE
RD
BLW
23
7
15
(Note)
(Note)
AddressAddress + 1Address + 2Address + 3
D0 to D
D0 to D
8-bit data access
16-bit data access
7
7
32-bit data access
D0 to D
7
D0 to D
7
φ
1
A0 to A
23
D0 to D
7
D8 to D
15
External data bus width = 8 bits
32/16/
8-bit
data
written
ALE
RD
BLW
BHW
(Note)
(Note)
AddressAddress + 1Address + 2Address + 3
D0 to D
D0 to D
8-bit data access
16-bit data access
7
7
32-bit data access
Note: When the voltage level at pin BYTE = “L”, functions as pins D8 to D15 are valid. However, when 8-bit width is selected
as the external bus width by the chip select wait controller, the functions as pins D
8 to D15 = floating, BHW = “H” output.) When the voltage level at pin BYTE = “H”, these pins function as programmable
(D
I/O port (P2, P3
3) pins.
Fig. 16 Bus cycle waveform example for data access (access to external area) (3)
D0 to D
7
D0 to D
7
8 to D15 and BHW become invalid.
24
Page 25
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
● Number of bus cycles
Figure 17 shows the bus cycle waveform at access to the internal
area. Bit 7 of the processor mode register 1 (address 5F16) selects
the number of bus cycles for the internal ROM: 3φ or 2φ. (This bit 7 is
the internal ROM bus cycle select bit.) The internal RAM, SFRs (in-
1 bus cycle = 3φ (Note)
(Internal ROM bus cycle select bit = 0)
1 bus cycle = 3φ
φ
BIU
ROM
RAM
Internal address bus
Internal data bus
Internal code bus
Address
Data
Internal address bus
φ
BIU
ternal peripheral devices’ control registers) are always accessed with
1 bus cycle = 2φ. Figure 18 shows the bus cycle waveform at access
to the external area. The bus cycle select bits 0, 1 (See the note in
Figure 18.) select the number of the bus cycles for each CSi area
___
from 8 types of numbers.
1 bus cycle = 2φ
(Internal ROM bus cycle select bit = 1)
1 bus cycle = 2φ
φ
BIU
Internal address bus
Internal data bus
Internal code bus
1 bus cycle = 2φ
Address
Address
Data
SFR
Internal data bus
Internal code bus
Data
Note: When reprogramming the internal flash memory in the CPU reprogramming mode, select the bus cycle = 3φ.
Fig. 17 Bus cycle waveform at access to internal area
25
Page 26
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Bus cycle
select bit 0
0 0
0 1
1 0
Bus cycle select bit 1 = 0
Bus cycle 1φ + 1φ
1 bus cycle = 2φ
φ1
External address
External data bus
Bus cycle 1φ + 2φ
External address
External data bus
Bus cycle 1φ + 3φ
External address
External data bus
bus
CSi
BLW,BHW
ALE
bus
CSi
BLW,BHW
ALE
bus
CSi
BLW,BHW
ALE
RD
φ1
RD
φ1
RD
Address
Data
1 bus cycle = 3φ
Address
1 bus cycle = 4φ
Address
Data
Data
Bus cycle 2φ + 3φ
❈
External data bus
Bus cycle 2φ + 4φ
❈
External data bus
Bus cycle 3φ + 3φ
❈
External data bus
External address
External address
External address
bus
BLW,BHW
ALE
bus
BLW,BHW
ALE
bus
BLW,BHW
ALE
Bus cycle select bit 1 = 1
1 bus cycle = 5φ
2φ
φ1
Address
CSi
RD
1 bus cycle = 6φ
φ1
CSi
RD
1 bus cycle = 6φ
φ1
CSi
RD
❈
3φ
Data
❈
4φ2φ
Address
Data
3φ3φ
Address
Data
Bus cycle 2φ + 2φ
1 bus cycle = 4φ
φ1
External address
1 1
Notes 1: The bus cycle type is determined by the following bits:
• Areas out of area CS
• Area CS
External data bus
i: area CSi bus cycle select bit 0 (bits 0 and 1 at addresses 8016, 8216, 8416, 8616)
2:❈ indicates the bus cycle, where the burst ROM access specification is enabled.
bus
CSi
RD
BLW,BHW
ALE
i : external bus cycle select bit 0 (bits 2 and 3 at address 5E16)
external bus cycle select bit 1 (bit 0 at address 5F
area CS
Address
i bus cycle select bit 1 (bit 3 at addresses 8116, 8316, 8516, 8716)
Fig. 18 Bus cycle types at access to external area
Data
Bus cycle 3φ + 4φ
φ1
External address
External data bus
16)
bus
BLW,BHW
ALE
CSi
RD
3φ
1 bus cycle = 7φ
4φ
Address
Data
26
Page 27
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
● Recovery cycle
A recovery cycle which is equivalent to 1 or 2 cycles of φ1 can be inserted after each area CSi’s access cycle. Whether the recovery
cycle is inserted or not is determined by the recovery cycle insert
select bit of each CSi control register L (bit 6 at addresses 8016, 8216,
φ1
❈
A
0
to A
ALE
access
RD
At double consecutive
φ
❈❈
At instruction prefetch
A0 to A
23
ALE
access
RD
At quadruple consecutive
At data access
___
___
Recovery cycle = 1 cycle of φ1Recovery cycle = 2 cycles of φ1
Recovery
Next access
Instruction prefetch
23
AddressAddress + 2
❈ When address locates at 4-byte boundary, or when branched.
1
AddressAddress + 2Address + 4
❈❈ When address locates at 8-byte boundary.
Access cycle
φ
1
A0 to A
23
Address
ALE
RD,
BLW, BHW
cycle
Instruction prefetch
Recovery
Next access
cycle
cycle
cycle
Address + 6
Recovery
cycle
Next access
8416, 8616). Also, the number of the recovery cycles is selected by
the recovery-cycle-insert-number select bit of the processor mode
register 1 (bit 6 at address 5F16). Figure 19 shows a waveform example when a recovery cycle is inserted.
Next access
cycle
Next access
cycle
Recovery cycle
Address + 6
cycle
A0 to A
A0 to A
Instruction prefetch
φ
1
23
AddressAddress + 2
ALE
RD
φ
1
23
AddressAddress + 2Address + 4
ALE
RD
Access cycle
φ
1
A0 to A
ALE
RD,
BLW, BHW
23
Address
Recovery cycle
Instruction prefetch
Recovery cycle
Next access
cycle
Notes 1: The recovery cycle insert is specified by the recovery cycle insert select bit and the recovery-cycle-insert-number select bit (bits 4 and 6 at address 5F16).
Recovery cycle insertion is valid only at access to area CS
2: The above is applied when 1 bus cycle = 2φ.
i.
Fig. 19 Waveform example when recovery cycle is inserted
27
Page 28
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
● Burst ROM access
When ROM supporting the burst ROM access has been allocated to
___
area CSi, the burst ROM access can be specified. The burst ROM
access is specified by each burst ROM access select bit of the CSi
control register L (bit 5 at addresses 8016, 8216, 8416, 8616). The
burst ROM access is valid only when the external data bus width =
16 bits with an instruction prefetched. In the other cases, the normal
access is performed regardless of the contents of the burst ROM access select bit. The burst ROM access can be specified only in the
case of ❈ in Figure 18.
(a)
φ
1
RD
External address bus
(A
0
to A23)
External data bus
0
to D7)
(D
External data bus
(D8 to D15)
At quadruple consecutive accessAt double consecutive access
___
AddressAddressAddressAddress
Figure 20 shows a waveform example at burst ROM access.
When an instruction is prefetched from the burst ROM, 8 bytes are
fetched starting from an 8-byte boundary (the low-order 3 bits of address, A2, A1, A0 = “000”) in waveform (a). When branched, regardless of the 8-byte boundary of the branch destination address,
access starting from the 4-byte boundary (the low-order 2 bits of address, A1, A0 = “00”) is performed in waveform (b). Once the 8-byte
boundary has been selected, instructions will be prefetched in waveform (a) until a branch.
Data
(Instruction)
Data
(Instruction)
Data
(Instruction)
Data
(Instruction)
Data
(Instruction)
Data
(Instruction)
Data
(Instruction)
Data
(Instruction)
Note: The above is applied when 1 bus cycle = 2φ.
(b)
φ
1
RD
External address bus
(A0 to A23)
External data bus
0
to D7)
(D
External data bus
8
to D15)
(D
Note: The above is applied when 1 bus cycle = 2φ.
Notes 1: The burst ROM access is selected by the burst ROM access select bit (bit 5 at addresses 80
2: The burst ROM access can be selected only in the case of ❈ in Figure 18.
Fig. 20 Waveform example at burst ROM access
AddressAddress
Data
(Instruction)
Data
(Instruction)
Data
(Instruction)
Data
(Instruction)
16
, 8216, 8416, 8616).
28
Page 29
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
● Address output selection
As shown in Figure 21, the unnecessary state change of address
output pins (A0 to A23) can be avoided, without outputting an address
at access to the internal area.
When the address output select bit of the particular function select
register 1 (bit 4 at address 6316) is set to “1”, an address is output
only at access to the external area. Also, at access to the internal
Address output select bit = 0
Address output select bit = 1
(Address waveform changes only
when external access is generated.)
A0 to A
BLW,BHW
A0 to A
BLW,BHW
area, the address at the preceding access to the external area is retained. The address output start timing in this case is the half cycle
of φ1 later than that at the normal access (when the address output
select bit = “0”). For the bit structure of the particular function select
register 1, refer to the section on the standby function.
Also, at the normal access, an address is output at both of the access to the internal and external areas.
Access to
external area
Address
Address
φ
1
RD,
RD,
23
23
Access to
external area
Address
Access to
internal area
Unde-
fined
Address
Unde-
fined
Fig. 21 Waveform example depending on address output function selection
29
Page 30
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SHINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
● Area multiplication
When area CS2’s external data bus width = 8 bits with the multiplexed bus select bit of the CS2 control register H (bit 5 at address
8516) = “1”, the external bus type can be changed to the multiplexed
bus type only at access to area CS2. In this case, the low-order 8 bits
___
___
Area CS2 bus
cycle select
bit 0
___
Bus cycle
select bit 0
Multiplexed bus select bit = 1
Bus cycle 2φ + 2φ
φ1
External address bus
At write, LA0/D0 to LA7/D7
0
1 1
At read, LA0/D0 to LA7/D7
CSi
RD,
BLW
ALE
Bus cycle 3φ + 3φ
φ1
External address bus
1
1 0
At write, LA0/D0 to LA7/D7
At read, LA0/D0 to LA7/D7
CSi
RD,
BLW
ALE
of an address (LA0 to LA7) are output, and the low-order 8 bits of
data (D0 to D7) are input/output with the time-sharing method, respectively.
Figure 22 shows a waveform example of area multiplication for each
bus cycle. Do not select the area multiplication function for a bus
cycle not shown in Figure 22.
1 bus cycle = 4φ
2φ2φ
Address
LA0 to LA7
LA0 to LA7
LA0 to LA7
LA0 to LA7
D0 to D7
D0 to D7
1 bus cycle = 6φ
3φ3φ
Address
D0 to D7
D0 to D7
Bus cycle 3φ + 4φ
φ1
External address bus
1
Notes 1: The number of bus cycles is determined by the following bits:
Area multiplication is specified by the multiplexed bus select bit (bit 5 at address 8516).
2: Do not select the area multiplication function for a bus cycle not shown in Figure 22.
1 1
• Area CS2 bus cycle select bit 0 (bits 0 and 1 at address 8416)
• Area CS2 bus cycle select bit 1 (bit 3 at address 8516)
At write, LA0/D0 to LA7/D7
At read, LA0/D0 to LA7/D7
CSi
RD,
BLW
ALE
Fig. 22 Waveform example of area multiplication for each bus cycle
30
LA0 to LA7
LA0 to LA7
1 bus cycle = 7φ
4φ3φ
Address
D0 to D7
D0 to D7
Page 31
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
PROCESSOR MODES
Any of the three processor modes (single-chip mode, memory expansion mode, microprocessor mode) can be selected with the following:
• Processor mode bits of the processor mode register 0 (bits 1 and 0
at address 5E16; Figure 24)
• Voltage level applied to pin MD0
Table 3 lists the selection method of a processor mode.
The memory map which the CPU can access depends on the selected processor mode. Figure 23 shows the memory maps in three
processor modes.
Also, the functions of ports P0 to P4, P10, P11 depend on the selected processor mode. For details, see Tables 5 and 6.
Figures 24 to 26 show the bit configurations of the processor mode
registers 0, 1, and port function control register.
In the single-chip mode, ports P0 to P4, P10, P11 function as I/O
ports. (While the internal peripheral devices are used, these ports
Single-chip mode
0
16
FF
16
Unused area
Internal RAM
SFR area
area
Memory expansion mode
function as these devices’ I/O pins.) In this mode, only the internal
area (SFRs, internal RAM, internal ROM) is accessible.
In the memory expansion and microprocessor modes, external devices assigned in the external memory area can be connected via
buses. Therefore, ports P0 to P4, P10, P11 function as I/O pins for
the address bus, data bus, bus control signals. (Some port functions
are selectable.) Table 4 lists each bus control signal’s function.
In the memory expansion mode, all of the internal area (SFRs, internal RAM, internal ROM) and external area are accessible. In the microprocessor mode, the internal area except for the internal ROM (in
other words, SFRs and internal RAM) and the external area are accessible.
Note that, when the external devices are located to an area where
the internal area and external area overlap, only the internal area
can be read/written; the external area cannot be read/written.
Microprocessor mode
SFR area
Internal RAM
area
SFR area
Internal RAM
area
Unused area
Internal ROM
SFR area : Internal peripheral devices’ control registers are allocated here.
External area : Access to this area enables the access to the devices which
Note: Do not access this area (bank FF
Fig. 23 Memory maps in three processor modes
area
Internal ROM
area
FEFFFF
16
FF0000
16
Reserved area
FFFFFF
are connected with the external.
(Note)
16
16
).
Reserved area
(Note)
31
Page 32
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Table 3. Selection method of processor mode
MD0Processor mode bits
00
VSS
VCC
Notes 1: Processor mode bits = bits 0 and 1 of the processor mode register 0 (address 5E16)
2: While the Vcc level voltage is applied to pin MD0, the processor mode bits are fixed to “10”.
Table 4. Each bus control signal’s function
Signal
RD
I/O
Output
01
10
10
Read signal. Outputs “L” at read from the external area.
(Note 1)
(Note 2)
Processor mode
Single-chip mode
Memory expansion mode
Microprocessor mode
Microprocessor mode
Function
After reset is removed, the single-chip mode is selected. By changing the processor mode bits’ contents by software, the memory expansion mode or microprocessor mode can be selected.
After reset is removed, the microprocessor mode is selected.
MITSUBISHI MICROCOMPUTERS
Description
Remarks
BLW
BHW
ALE
φ
1
RDY
HOLD
HDLA
CS0–CS3
BYTE
Output
Write signal. Outputs “L” at write to the external area.
Output
Address latch enable signal. Outputs “H” level pulse in the
period just before signals RD, BLW, BHW become “L”.
This is used to latch an address in an external circuit.
Output
Internal standard clock’s output. Outputs system clock
(fsys).
Input
Ready signal. The “L” level period of the last
cess cycle for the external area (in other words, “L” level
period of RD, BLW, BHW) will be extended while “L” level
voltage is applied to this pin.
Input
Hold request signal. Appliance of “L” level voltage will generate a hold request; appliance of “H” level voltage will request to terminate the hold state.
Output
Hold acknowledge signal. Outputs “L” in the hold state.
Output
Chip select signal. Outputs “L” in access to the specified
chip select area.
Input
Input signal to select the external data bus width. When
this pin’s level = Vss, 16-bit width will be selected; and
when Vcc, 8-bit width will be selected.
φ
1 in the ac-
For operation differences between BLW and BHW depending on the external data bus width, see Table 5.
In order to latch an address with signal ALE, do as follows:
• While ALE = “H”, be sure to open a latch, so the address
will pass it.
• While ALE = “L”, be sure to hold the address.
Acceptance and termination of a hold request is performed
at completion of the bus cycle while the BIU operates.
In the hold state, A0–A23, D0–D15, RD, BLW, BHW, ALE,
CS0–CS3 enter the floating state. At termination of the hold
state, simultaneously with the timing when HLDA becomes
“H” level, the above floating state is terminated. Then, bus
access will be restarted 1 cycle of
In the hold state, also, the CPU operates with access to
the internal area. If the CPU accesses the external area, in
the hold state, the CPU stops its operation.
For details, refer to the section on the chip select wait controller.
When BYTE = Vss level, by the register setting, each chip
select area (CS1 to CS3) can have the 8-bit data bus, independently.
For details, refer to the section on the chip select wait controller.
φ
1 after.
32
Page 33
M37902FCCHP, M37902FGCHP, M37902FJCHP
Table 5. Relationship between processor modes, memory area, and port function (1)
Mode
(Note 1)
Port pins P100 to P107
Port pins P110 to P117
Port pins P00 to P07
Port pins
P10 to P17
Port pins
P20 to P27
Port pin
P32
Port pin
P33
Pin MD0
Processor mode
bits (Note 2)
SFR area
Internal RAM area
Internal ROM area
Other area
Memory area
External data bus
width = 16 bits
External data bus
width = 8 bits
External data bus
width = 16 bits
External data bus
width = 8 bits
Port pin P30
Port pin P31
External data bus
width = 16 bits
External data bus
width = 8 bits
External data bus
width = 16 bits
External data bus
width = 8 bits
Single-chip mode
VSS level voltage is applied
00
SFR area
Internal RAM area
Internal ROM area
(Do not access.)
I/O port pins P100 to P107
I/O port pins P110 to P117
I/O port pins P00 to P07
I/O port pins P10 to P17
I/O port pins P20 to P27
I/O port pin P30
I/O port pin P31
I/O port pin P32
I/O port pin P33
Memory expansion mode
VSS level voltage is applied
01
SFR area
Internal RAM area
Internal ROM area
External memory area
Low-order address (A0 to A7) is output.
Middle-order address (A8 to A15) is
output.
I/O port pins P110 to P117 (Note 3)
High-order address (A16 to A23) is out-
put.
I/O port pins P00 to P07 (Note 3)
Low-order data (D0 to D7, data at
even address) is input/output.
Low-order data (D0 to D7, data at
even/odd address) is input/output.
Low-order address (LA0 to LA7) is out-
put. Low-order data (D0 to D7, data at
even/odd address) is input/output
(Note 4).
High-order data (D8 to D15, data at
odd address) is input/output.
I/O port pins P20 to P27(Note 5)
I/O port pin P30
Ready signal RDY is input (Note 6).
Read signal RD is output.
Write signal BLW (write to even ad-
dress) is output.
Write signal BLW (write to even/odd
address) is output.
Write signal BHW (write to odd ad-
dress) is output.
I/O port pin P33 (Note 5)
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Microprocessor mode
VCC level voltage is applied
10
SFR area
Internal RAM area
External memory area
External memory area
Low-order address (A0 to A7) is output.
Middle-order address (A8 to A15) is
output.
I/O port pins P110 to P117 (Note 3)
High-order address (A16 to A23) is out-
put.
I/O port pins P00 to P07 (Note 3)
Low-order data (D0 to D7, data at
even address) is input/output.
Low-order data (D0 to D7, data at
even/odd address) is input/output.
Low-order address (LA0 to LA7) is out-
put. Low-order data (D0 to D7, data at
even/odd address) is input/output
(Note 4).
High-order data (D8 to D15, data at
odd address) is input/output.
I/O port pins P20 to P27(Note 5)
Ready signal RDY is input.
I/O port pin P30 (Note 6)
Read signal RD is output
Write signal BLW (write to even ad-
dress) is output.
Write signal BLW (write to even/odd
address) is output.
Write signal BHW (write to odd ad-
dress) is output.
I/O port pin P33 (Note 5)
33
Page 34
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Table 6. Relationship between processor modes, memory area, and port function (2)
Single-chip mode
I/O port pin P40
Port pin P40
Port pin P41
Port pin P42
Port pin P43
Port pin P44
Port pins P45 to P47
Notes 1: For details of the processor mode setting, see Table 3.
2: Processor mode bits = bits 0 and 1 of the processor mode register 0 (address 5E
3: The middle-order/high-order address output pins in the memory expansion or microprocessor mode can be switched to I/O port pins by the address/port
switch select bits of the port function control register (bits 2 to 0 at address 92
4: When the external data bus width for the chip select area, CS
of the CS2 control register H (bit 5 at address 8516), a multiplexed bus which performs the following operations with the time-sharing method is realized:
• Output of address LA
• Input/Output of data D0 to D7
5: When one of areas CS1/CS2/CS3 is accessed under the following conditions, pins D8 to D15 enter the floating state, and pin BHW outputs “H” level.
(They do not become I/O port pins.)
• Pin BYTE is at Vss level.
• One of bit 2s at addresses 82
width = 8 bits).
6: In the memory expansion mode, by the corresponding select bits of the processor mode register 0 and 1 (addresses 5E
P43 can operate as pins for RDY input, ALE output,
In the microprocessor mode, by the above select bits, the above pins (RDY, ALE,
tively.
In the single-chip mode, port pin P4
7: In the memory expansion mode, port pin P4
16).
80
8: In the memory expansion and microprocessor modes, port pins P45 to P47 can operate as the CS1/CS2/CS3 output pins by the CSi output select bits (i =
1 to 3) (bit 7s at addresses 82
I/O port pin P41
Clock
φ
1 is output (Note 6).
I/O port pin P42
I/O port pin P43
I/O port pin P44
I/O port pins P45 to P47
0 to LA7
16, 8416, 8616 (the external data bus width select bit of the CS1/CS2/CS3 control register L) is set to “1” (external data bus
1 can operate as the φ1 output pin by the above select bits.
16, 8416, 8616).
φ
1 output, HLDA output, HOLD input, respectively.
4 can operate as the CS0 output pin by the CS0 output select bit of the CS0 control register L (bit 7 at address
Memory expansion mode
I/O port pin P40
Microprocessor mode
Address latch enable signal
ALE is output.
Address latch enable signal
I/O port pin P4
0(Note 6)
ALE is output (Note 6).
I/O port pin P41
Clock
φ
1 is output (Note 6).
I/O port pin P42
Clock
φ
1 is output.
I/O port pin P41(Note 6)
Hold acknowledge signal
HLDA is output.
Hold acknowledge signal
I/O port pin P42(Note 6)
HLDA is output (Note 6).
I/O port pin P43
Hold request signal
Signal HOLD is input.
Hold request signal
I/O port pin P43(Note 6)
HOLD is input (Note 6).
I/O port pin P44
Chip select signal CS0 is output.
Chip select signal CS0 is output
(Note 7).
I/O port pins P45 to P47
Chip select signals CS1 to CS3 are
output (Note 8).
16).
2, has been set to 8 bits, only in the access to area CS2, by the multiplexed bus select bit
16).
φ
1, HLDA, HOLD) can operate as port pins P30, P40 to P43, respec-
External bus cycle select bit 0 (Note 2)
See Figure 18.
Interrupt priority detection time select bits
0 0 : 7 cycles of f
0 1 : 4 cycles of fsys
1 0 : 2 cycles of fsys
1 1 : Do not select.
Software reset bit
By a write of “1” to this bit, the microcomputer will be reset, and then, restarted.
1 output select bit (Note 3)
Clock φ
1 output is disabled. (P41 functions as an programmable I/O port pin.)
0 : φ
1 output is enabled. (P41 functions as the clock φ1 output pin.)
1 : φ
Notes 1: While VSS level voltage is applied to pin MD0, this bit’s state is cleared to “0” at reset. While VCC level voltage is applied
to pin MD0, on the other hand, this bit’s state is set to “1” at reset. (Fixed to “1”.)
2: These bits are valid to the external area except for chip select area (area CS
by the corresponding area CS
3: While V
SS level voltage is applied to pin MD0, this bit’s state is cleared to “0” at reset. While VCC level voltage is applied
to pin MD0, on the other hand, this bit’s state is set to “1” at reset.
i bus cycle select bits 0, 1.
sys
Address
16
5E
i). The bus cycle of area CSi is selected
Fig. 24 Bit configuration of processor mode register 0
35
Page 36
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
76543210
Processor mode register 1
External bus cycle select bit 1 (Note 1)
See Figure 18.
Direct page register switch bit (Note 2)
0 : Only DPR0 is used.
1 : DPR0 to DPR3 are used.
RDY input select bit (Notes 3 to 5)
0 : RDY input is disabled. (P3
1 : RDY input is enabled. (P3
ALE output select bit (Notes 3 and 4)
0 : ALE output is disabled. (P4
1 : ALE output is enabled. (P4
Recovery cycle insert select bit (Notes 3 and 4)
0 : No recovery cycle is inserted at access to the external area.
1 : Recovery cycle is inserted at access to the external area.
HOLD input, HLDA output select bit (Notes 3 to 5)
0 : HOLD input and HLDA output are disabled.
3
and P42 function as programmable I/O port pins.)
(P4
1 : HOLD input and HLDA output are enabled.
3
and P42 function as pins HOLD and HLDA, respectively.)
(P4
Recovery-cycle-insert number select bit (Note 6)
0 : 1 cycle
1 : 2 cycles
Address
16
5F
0
functions as a programmable I/O port pin.)
0
functions as pin RDY.)
0
functions as a programmable I/O port pin.)
0
functions as pin ALE.)
Notes 1: This bit is valid to the external area except for chip select areas (area CSi), and the bus cycle of area CSi is independent
of this bit’s contents.
The bus cycle of area CS
16
, 8416, 8616; bit 3 at addresses 8116, 8316, 8516, 8716).
82
2: After reset, this bit’s contents can be switched only once. During the software execution, be sure not to switch this bit’s contents.
3: In the single-chip mode, these bits’ functions are disabled regardless of these bits’ contents.
4: While V
5: In the memory expansion or microprocessor mode, if this bit’s contents is switched from “1” to “0”, this bit will be cleared to “0”.
6: The program which switches this bit’s contents must be assigned to the internal area.
7: In the microprocessor mode, this bit is invalid.
SS
to pin MD0, on the other hand, each of these bits is “1” at reset.
After this clearance, this bit cannot return to “1”. If it is necessary to set this bit to “1”, be sure to reset the microcomputer.
When the internal flash memory is reprogrammed in the CPU reprogramming mode, be sure to clear this bit to “0”.
level voltage is applied to pin MD0, each of these bits is “0” at reset. While VCC level voltage is applied
i
is selected by the corresponding area CSi bus cycle select bits 0, 1 (bits 0, 1 at addresses 8016,
Fig. 25 Bit configuration of processor mode register 1
Internal ROM bus cycle select bit (Note 7)
0 : 3φ
1 : 2φ
36
Page 37
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
76543210
0
0
Port function control register
92
Address/Port switch select bits
Address
0
to A23 (16 Mbytes)
000 : A
0
to A21, P06, P07 (4 Mbytes)
001 : A
0
to A19, P04 to P07 (1 Mbytes)
010 : A
0
to A17, P02 to P07 (256 Kbytes)
011 : A
0
to A15, P00 to P07 (64 Kbytes)
100 : A
101 : Do not select.
0
to A11, P00 to P07, P114 to P117 (4 Kbytes)
110 : A
0
to A7, P00 to P07, P110 to P117 (256 bytes)
111 : A
Port P0 input level select bit
0 : V
IH
= 0.7VCC, VIL = 0.2V
CC
1 : VIH = 0.43VCC (Note 1), VIL = 0.16V
4
–P47 pullup select bit (Notes 2 and 3)
Pins P4
0 : Pins P4
1 : Pins P4
4
–P47 are pulled up.
4
–P47 are not pulled up.
Fix these bits to “0”.
Pin NMI pullup select bit (Note 2)
0 : Pin NMI is pulled up.
1 : Pin NMI is not pulled up.
At reset
16
CC
00
16
Notes 1: For the M37902FxM (power source voltage = 3.3 V±0.3 V), VIH = 0.5VCC.
CC
2: When MD1 = V
and MD0 = VCC (flash memory parallel I/O mode), pins P44 to P47 and NMI are
not pulled up, regardless of these bits’ contents.
SS
3: When MD1 = V
and MD0 = VCC (microprocessor mode), pin CS0 (P44) is not pulled up, regardless of the bit’s contents.
Fig. 26 Bit configuration of port function control register
37
Page 38
M37902FCCHP, M37902FGCHP, M37902FJCHP
Chip select wait controller
By the control of the chip select wait controller (CSWC), the chip select function for the maximum of 4 blocks can be set at the bus access to the external area.
Also, by the setting of the CSWC, port pins P44 to P47 can operate
as chip select output pins (CS0 to CS3).
Figure 27 shows a chip select output waveform example.
This chip select function determines the following items of the chip
select area: start address, address’s block size, wait number, external data bus width, RDY control validity, burst ROM specification,
recovery cycle insertion validity, and area multiplication validity.
For the external area except for areas CS0 to CS3, the processor
mode registers 0, 1 determine the above items. After reset is removed, when the microcomputer starts it’s operation in the microprocessor mode, area CS0 is automatically selected.
Table 7 lists the function of areas CS0 to CS3.
Figure 28 shows the bit configuration of the CS0/CS1/CS2/CS3 control register Ls. These registers determine the following items of a
device to be connected: wait number, external data bus width (Note:
The external data bus width of area CS0 is determined by pin BYTE’s
level.), RDY control validity, burst ROM access specification, recovery cycle insertion validity, and output validity of CS0 to CS3.
Figure 29 shows the bit configuration of the CS0/CS1/CS2/CS3 control register Hs. These registers determine block size, etc. of an external area to be connected. For areas CS0 to CS2, by selecting
mode 1 with the area CSk setting mode select bit, an chip select area
can be set to the external area in bank 0.
Figures 30 shows the bit configuration of the area CS0/CS1/CS2/CS3
start address registers. For details of these addresses’ setting, see
Figures 31 to 33.
(Selected by bits 0, 1 at addresses
8216, 8416 and bit 3 at addresses
8316, 8516.)
External data bus
width
Determined by pin BYTE’s level.
When BYTE = VSS level, 8-bit width
or 16-bit width can be selected
arbitrary (Note 1). (Selected by bit 2
at addresses 8216, 8416.)
RDY control
Valid (Selected by bit 2 at address
5F16 and bit 3 at address 8016.)
Valid (Selected by bit 2 at address
5F16 and bit 3 at addresses 8216,
8416.)
Burst ROM access
Available.
Available.
(Notes 2, 3)
Recovery cycle
Available.
Available.
insertion
Area multiplexed bus
access (Note 3)
Address output
Not available.
Available.
CS1: Not available.
CS2: Available. (Note 4)
Available.
selection (Note 5)
Notes 1: When BYTE = Vcc level, the external data bus width is fixed to 8 bits.
2: Burst ROM access is valid only when the external data bus width is 16 bits at instruction prefetch.
3: Burst ROM access and area multiplexed bus access cannot be used at the same time.
4: Valid only when area CS
5: Selected by the address output select bit (bit 4 at address 63
2 is accessed with the 8-bit external data bus width.
16). The address output selection for each area is not available.
0 0 0 : 0 byte (Area CS
0 0 1 : Do not select.
0 1 0 : Do not select.
0 1 1 : Do not select.
1 0 0 : 4 Kbytes
1 0 1 : 8 Kbytes
1 1 0 : Do not select.
1 1 1 : Do not select.
2
is invalid.)
“0” at read.
Multiplexed bus select bit
0 : Separated bus (D
1 : Multiplexed bus (When the CS
with area CS
Area CS
2
setting mode select bit
0 : Mode 0 (A block can be set to 16-Mbyte space in a unit of 128 Kbytes.)
1 : Mode 1 (A block can be set to bank 0 in a unit of 4 Kbytes.)
Fig. 29 Bit configuration of CS0/CS1/CS2/CS3 control register Hs
41
Page 42
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
76543210
Area CS
0
start address register
Address
8A
16
“0” at read.
When mode 0 is selected, these bits determine A
When mode 1 is selected, these bits determine A
16
8
Any of the following values can be set to these bits: “10
(Bits 0 to 3 are always “0” at read.)
Note: Do not set a value other than “1016”, “2016”, “4016”, and “8016”. See Figure 31.
76543210
Area CS
1
start address register
2
start address register
Area CS
Address
16
8C
8E
16
“0” at read.
When mode 0 is selected, these bits determine A
When mode 1 is selected, these bits determine A
16
8
(Bit 0 is always “0” at read.)
Note: The start address setting depends on the block size, which has been selected by the area CS
16
(bits 0 to 2 at address 83
76543210
, bits 0 to 2 at address 8516). See Figures 32 and 33.
Address
Area CS
3
start address register
90
16
“0” at read.
These bits determine A
16
to A23 of the area CS3 start address.
(Bit 0 is always “0” at read.)
At reset
16
10
to A23 of the area CS0 start address.
to A15 of the area CS0 start address.
16
”, “2016”, “4016”, and “8016”.
At reset
16
00
00
16
to A23 of the area CS1/CS2 start address.
to A15 of the area CS1/CS2 start address.
1
/CS2 block size select bits
At reset
16
00
Note: The start address setting depends on the block size, which has been selected by the area CS3 block size select bits
16
(bits 0 to 2 at address 87
). See Figure 33.
Fig. 30 Bit configuration of area CS0/CS1/CS2/CS3 start address registers
42
Page 43
MITSUBISHI MICROCOMPUTERS
128K
bytes
512K
bytes
2M
bytes
1M
bytes
4M
bytes
8M
bytes
128K
bytes
512K
bytes
2M
bytes
256K
bytes
1M
bytes
4M
bytes
8M
bytes
0
16
1000
16
1FFFF
16
7FFFF
16
FFFFF
16
128K
bytes
512K
bytes
2M
bytes
Start address : 1000
16
Value to be set to area CS
0
start
address register = “10
16
”
Block size
2000
16
Start address : 2000
16
Value to be set to area CS
0
start
address register = “20
16
”
4000
16
Start address : 4000
16
Value to be set to area CS
0
start
address register = “40
16
”
8000
16
128K
bytes
Start address : 8000
16
Value to be set to area CS
0
start
address register = “80
16
”
Area CS
0
cannot be assigned here.
Note: When an area where area CS
0
and the internal area overlap is accessed, the internal area will be accessed. In this case,
pin CS
0
outputs “H” level.
3FFFF
16
256K
bytes
1FFFFF
16
1M
bytes
4M
bytes
8M
bytes
512K
bytes
2M
bytes
256K
bytes
1M
bytes
4M
bytes
8M
bytes
3FFFFF
16
7FFFFF
16
Block size
Block size
Block size
256K
bytes
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Fig. 31 Area CS0 (mode 1)
43
Page 44
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Block size : 4 Kbytes
Addresses which can be
start address
(Address FFFF16 is not
included; Note 1)
016
100016
200016
300016
400016
500016
600016
700016
800016
F00016
(FFFF16)
Block size : 8 Kbytes
Addresses which can be
start address
(Address FFFF16 is not
included; Note 1)
016
4 Kbytes
200016
400016
600016
800016
E00016
(FFFF16)
8 Kbytes
Notes 1: Only A8 to A15 of one of these addresses
can be set to the area CS
1/CS2 start address
register. Do not set another address not
shown here.
2: When an area where area CS
1/CS2 and the
internal area overlap is accessed, the internal
area will be accessed. In this case, pin CS
outputs “H” level.
1/CS2
Fig. 32 Area CS1/CS2 (mode 1)
44
Page 45
Notes 1: Only A
16
to A
23
of one of these addresses can be set to the area CS
0
/CS
1
/CS
2
/CS
3
start
address register. Do not set another address not shown here.
2: When an area where area CS
0
/CS
1
/CS
2
/CS
3
and the internal area overlap is accessed,
the internal area will be accessed. In this case, pin CS
0
/CS
1
/CS
2
/CS
3
outputs “H” level.
Block size : 128 Kbytes
Addresses which can
be start address
(Addresses 0
16
and
FF0000
16
to FFFFFF
16
are not included;
Note 1)
: Area CS
0
/CS
1
/CS
2
/CS
3
cannot be assigned here.
(0
16
)
20000
16
40000
16
60000
16
80000
16
A0000
16
C0000
16
E0000
16
100000
16
F60000
16
F80000
16
FA0000
16
FC 0000
16
FE0000
16
(FF0000
16
)
(FFFFFF
16
)
120000
16
Block size : 256 Kbytes
Addresses which can
be start address
(Addresses 0
16
and
FF0000
16
to FFFFFF
16
are not included;
Note 1)
(0
16
)
40000
16
80000
16
C0000
16
100000
16
F80000
16
FC 0000
16
(FF0000
16
)
(FFFFFF
16
)
Block size : 512 Kbytes
Addresses which can
be start address
(Addresses 0
16
and
FF0000
16
to FFFFFF
16
are not included;
Note 1)
(0
16
)
80000
16
100000
16
F80000
16
(FF0000
16
)
(FFFFFF
16
)
Block size : 1 Mbytes
Addresses which can
be start address
(Addresses 0
16
and
FF0000
16
to FFFFFF
16
are not included;
Note 1)
(0
16
)
100000
16
200000
16
300000
16
400000
16
500000
16
600000
16
700000
16
800000
16
B00000
16
C00000
16
D00000
16
E00000
16
F00000
16
(FF0000
16
)
(FFFFFF
16
)
900000
16
A00000
16
Block size : 2 Mbytes
Addresses which can
be start address
(Addresses 0
16
and
FF0000
16
to FFFFFF
16
are not included;
Note 1)
(0
16
)
200000
16
400000
16
600000
16
800000
16
C00000
16
E00000
16
(FF0000
16
)
(FFFFFF
16
)
A00000
16
Block size : 4 Mbytes
Addresses which can
be start address
(Addresses 0
16
and
FF0000
16
to FFFFFF
16
are not included;
Note 1)
(0
16
)
400000
16
800000
16
C00000
16
(FF0000
16
)
(FFFFFF
16
)
Block size : 8 Mbytes
Addresses which can
be start address
(Addresses 0
16
and
FF0000
16
to FFFFFF
16
are not included;
Note 1)
(0
16
)
800000
16
(FF0000
16
)
(FFFFFF
16
)
: Reserved area. Do not access this area.
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Fig. 33 Area CS0/CS1/CS2 (mode 0) and area CS
3
45
Page 46
MITSUBISHI MICROCOMPUTERS
76543210
Note: When the key input interrupt select bit (bit 0 at address 9416) = “1”,
the status of pin INT
3
cannot be read out.
INT
0
read bit
INT
1
read bit
INT
2
read bit
INT
3
read bit (Note)
INT
4
read bit
NMI read bit
Undefined at read.
External interrupt input read register
Address
95
16
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
INTERRUPTS
Table 8 shows the interrupt sources and the corresponding interrupt
vector addresses. Reset is also handled as a type of interrupt in this
section, too.
DBC and BRK instruction are interrupts used only for debugging.
Therefore, do not use these interrupts.
Interrupts other than reset, watchdog timer, zero divide, NMI, and
address matching detection all have interrupt control registers. Table
9 shows the addresses of the interrupt control registers and Figure
35 shows the bit configuration of the interrupt control register.
The interrupt request bit is automatically cleared by the hardware
during reset or when processing an interrupt. Also, interrupt request
bits other than watchdog timer and NMI can be cleared by software.
An NMI interrupt request is a non-maskable interrupt by an external
input and is accepted at the falling edge of an input to pin NMI. Also,
pin NMI has the pullup function. For more details, refer to the section
on input/output pins.
An INTi (i = 0 to 4) interrupt request is generated by an external input.
INT0 to INT2 are external interrupts; whether to cause an interrupt at
the input level (level sense) or at the edge (edge sense) can be selected with the level/edge select bit. Furthermore, the polarity of the
interrupt input can be selected with the polarity select bit.
For INT3 and INT4, the interrupt signal’s polarity can be change by
the polarity select bit. (This is valid only in the edge sense.)
By pins INT2 to INT4 select bits (bits 4 to 6 at address 9416; see Figure 40.), pin position of INT2 to INT4 can be changed.
When using the following pins as external interrupt input pins, clear
the direction registers of the corresponding multiplexed ports to “0”:
pins P62/INT0, P63/INT1, P64(P77)/INT2, P80(P74)/INT3, and
P84(P75)/INT4.
Furthermore, the INT3 interrupt can function as the key input interrupt. For details, refer to the section on the key input interrupt.
When the external interrupt input read register (address 9516) is read
out, the status of pins INT0 to INT4 and NMI can directly be read.
Timer and UART interrupts are described in the respective section.
The priority of interrupts when multiple interrupt requests are caused
simultaneously is partially fixed by hardware, but, it can also be adjusted by software as shown in Figure 36.
The hardware priority is fixed as the following:
reset > NMI > watchdog timer > other interrupts
Table 8. Interrupt sources and interrupt vector addresses
Fig. 34 Bit configuration of external interrupt input read register
Page 47
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
76543210
Interrupt priority level select bits (Note 1)
Interrupt request bit
0 : No interrupt requested
1 : Interrupt requested
Interrupt control register bit configuration for A-D converter, UART0, UART1,
timer A0 to timer A4, and timer B0 to timer B2.
76543210
Interrupt priority level select bits (Notes 1, 2)
Interrupt request bit
0 : No interrupt requested
1 : Interrupt requested
Polarity select bit
0 : Interrupt request bit is set to “1” at “H” level when level sense is selected;
this bit is set to “1” at falling edge when edge sense is selected.
1 : Interrupt request bit is set to “1” at “L” level when level sense is selected;
this bit is set to “1” at rising edge when edge sense is selected.
Level/Edge select bit
0 : Edge sense
1 : Level sense
MITSUBISHI MICROCOMPUTERS
Interrupt control register bit configuration for INT
76543210
Interrupt control register bit configuration for INT
Notes 1: Use the MOVM (MOVMB) instruction or the STA (STAB, STAD) instruction for writing to this bit.
2: Interrupt request bits of INT
Fig. 35 Bit configuration of interrupt control register
0
– INT
2
Interrupt priority level select bits
Interrupt request bit (Note 1)
0 : No interrupt requested
1 : Interrupt requested
Polarity select bit
0 : Interrupt request bit is set to “1” at falling edge.
1 : Interrupt request bit is set to “1” at rising edge.
3
and INT
4
0
to INT2 are invalid when the level sense is selected.
47
Page 48
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Table 9. Addresses of interrupt control registers
Interrupt control registers
INT3 interrupt control register
INT4 interrupt control register
A-D interrupt control register
UART0 transmit interrupt control register
UART0 receive interrupt control register
UART1 transmit interrupt control register
UART1 receive interrupt control register
Timer A0 interrupt control register
Timer A1 interrupt control register
Timer A2 interrupt control register
Timer A3 interrupt control register
Timer A4 interrupt control register
Timer B0 interrupt control register
Timer B1 interrupt control register
Timer B2 interrupt control register
INT0 interrupt control register
INT1 interrupt control register
INT2 interrupt control register
Interrupts caused by the address matching detection and when dividing by zero are software interrupts and are not included in Figure
36.
Other interrupts previously mentioned are A-D converter, UART, etc.
interrupts. The priority of these interrupts can be changed by changing the priority level in the corresponding interrupt control register by
software.
Figure 37 shows a diagram of the interrupt priority detection circuit
When an interrupt is caused, each interrupt device compares its own
priority with the priority from above and if its own priority is higher,
then it sends the priority below and requests the interrupt. If the priorities are the same, the one above has priority.
This comparison is repeated to select the interrupt with the highest
priority among the interrupts that are being requested. Finally the
selected interrupt is compared with the processor interrupt priority
level (IPL) contained in the processor status register (PS) and the
request is accepted if it is higher than IPL and the interrupt disable
flag I is “0”. The request is not accepted if flag I is “1”. The reset, NMI,
and watchdog timer interrupts are not affected by the interrupt disable flag I.
When an interrupt is accepted, the contents of the processor status
register (PS) is saved to the stack and the interrupt disable flag I is
set to “1”.
Furthermore, the interrupt request bit of the accepted interrupt is
cleared to “0” and the processor interrupt priority level (IPL) in the
processor status register (PS) is replaced by the priority level of the
accepted interrupt.
Therefore, multi-level priority interrupts are possible by resetting the
interrupt disable flag I to “0” and enable further interrupts.
For reset, watchdog timer, zero divide, NMI, and address match detection interrupts, which do not have an interrupt control register, the
processor interrupt level (IPL) is set as shown in Table 10.
The interrupt request bit and the interrupt priority level of each interrupt source are sampled and latched at each operation code fetch
cycle while fsys is “H”. However, no sampling pulse is generated until
the cycles whose number is selected by software has passed, even
if the next operation code fetch cycle is generated. The detection of
an interrupt which has the highest priority is performed during that
time.
Priority is determined by hardware
➃
A-D converter, UART, etc. interrupts
Priority can be changed by software inside ➃.
➂➁➀
Watchdog
timer
NMI
Reset
Fig. 36 Interrupt priority
Level 0
INT
4
INT
3
A-D
Interrupt request
Reset
NMI
Watchdog timer
Interrupt disable flag I
IPL
UART1 transmit
UART1 receive
UART0 transmit
UART0 receive
Timer B2
Timer B1
Timer B0
Timer A4
Timer A3
Timer A2
Timer A1
Timer A0
INT
2
INT
2
INT
1
INT
1
INT
0
Fig. 37 Interrupt priority detection
48
Page 49
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
As shown in Figure 38, there are three different interrupt priority detection time from which one is selected by software. After the selected time has elapsed, the highest priority is determined and is
processed after the currently executing instruction has been completed.
The time is selected with bits 4 and 5 of the processor mode register
0 (address 5E16) shown in Figure 24. Table 11 shows the relationship between these bits and the number of cycles. After a reset, the
processor mode register 0 is initialized to “0016.” Therefore, the longest time is automatically set, however, the shortest time must be selected by software.
f
sys
Operation code fetch cycle
Table 10.
Reset
Watchdog timer
NMI
Zero divide
Address matching detection
Value loaded in processor interrupt level (IPL) during an interrupt
Interrupt types
Setting value
0
7
7
Not change value of IPL.
Not change value of IPL.
T able 11. Relationship between interrupt priority detection time select
bit and number of cycles
Priority detection time select bit
Bit 5
0
0
1
Bit 4
0
1
0
Number of cycles (Note)
7 cycles of fsys
4 cycles of fsys
2 cycles of fsys
Note: For system clock fsys, refer to the section on the clock gener-
ating circuit.
Sampling pulse
Priority detection time
Select one between 00 to
10 with bits 4 and 5 of
processor mode register 0
Fig. 38 Interrupt priority detection time
(Note)
b5b4
0 0
0 1
1 0
Note: This pulse resides when 2 cycles of f
sys is selected.
49
Page 50
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Key Input Interrupt
The INT3 interrupt can function as the key input interrupt by setting
bits 1 to 3 of the external interrupt input control register (address
9416). The key input interrupt uses inputs KI0 to KI3. Figure 39 shows
the block diagram of the INT3/key input interrupt input circuit, and
Figure 40 shows the bit configuration of the external interrupt input
control register.
When bit 0 of the external interrupt input control register (key input
interrupt select bit)= “0”, a signal from pin INT3 is connected to the
INT3 interrupt control circuit, and INT3 external interrupt is normally
performed. When bit 0 = “1”, signals from pins KI0 to KI3, which correspond to ports P54 to P57 pins, are inverted, and then, the logical
sum of these signals is connected to the INT3 interrupt control regis-
Pin INT3
P80/INT3
P74/(INT3)
Pullup
transistor
select bit
0
1
Port P57 direction register
Key input interrupt pin
pullup select bit
KI
3 enable signal (Note)
ter. In this case, the external interrupt which uses pins KI0 to KI3 is
performed.
Bits 2 and 3 of the external interrupt input control register are the key
input interrupt pin select bits. By setting these bits, the combination
of key input interrupt pins can be selected. The interrupt vector addresses and interrupt control register of the key input interrupt are
common to those of the INT3 interrupt. Additionally, pullup resistors
(transistors) can be added to pins KI0 to KI4 by setting as follows:
• Set bit 1 of the external interrupt input control register to “1”.
• Next, select the key input interrupt pins by bits 2 and 3 of the exter-
nal interrupt input control register.
• Then, clear the contents of the port direction register which corresponds to the selected pins to “0”.
Key input interrupt
select bit
0
Interrupt control circuit
1
INT3 interrupt control register
INT3 interrupt request
P57/KI3
P57/KI3
Pullup
transistor
P56/KI2
Pullup
transistor
P55/KI1
Pullup
transistor
P54/KI0
Port P56 direction register
Key input interrupt pin
pullup select bit
KI2 enable signal (Note)
Port P5
5 direction register
Key input interrupt pin
pullup select bit
KI1 enable signal (Note)
Port P54 direction register
Key input interrupt pin
pullup select bit
KI0 enable signal (Note)
Fig. 39 Block diagram of INT3/key input interrupt input circuit
Note: KI
i enable signal (i = 0 to 3) means a signal which becomes
“1” when the key input interrupt select bit = “1” and pin KI
selected by the key input interrupt pin select bits.
• Port P5j direction register: bit j (j = 4 to 7) at address D
• INT3 interrupt control register: address 6E16
• Key input interrupt select bit: bit 0 at address 9416
0 0: Pins KI
0 1: Pins KI0 to KI
1 0: Pins KI0 and KI
1 1: Pin KI
2
select bit
Pin INT
0: Allocate pin INT
1: Allocate pin INT
Pin INT3 select bit (Note 3)
3
0: Allocate pin INT
1: Allocate pin INT
to P80.
3
to P74.
Pin INT4 select bit
4
0: Allocate pin INT
1: Allocate pin INT
to P84.
4
to P75 (Note 4).
Fix this bit to “0”.
Address
16
94
At reset
16
00
Notes 1: When using pin KIi, do not select timer A’s output pins and pulse output pins which are multiplexed with pin KIi.
2
2: When pin INT
address 96
3: When pin INT
When pin INT
4: When pin INT
is allocated to P77, do not use pin AN7/AD
16
) to “0” (output disabled).
3
is allocated to P80, clear the D-A2 output enable bit (bit 2 at address 9616) to “0” (output disabled).
3
is allocated to P74, do not use pin AN4.
4
is allocated to P75, do not use pin AN5.
Fig. 40 Bit configuration of external interrupt input control register
TRG
. Additionally, clear the D-A1 output enable bit (bit 1 at
51
Page 52
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
TIMER
There are eight 16-bit timers. They are divided by type into timer A(5)
and timer B(3).
The timer I/O pins are multiplexed with I/O pins for port P5 and P6.
To use these pins as timer input pins, the port direction register bit
corresponding to the pin must be cleared to “0” to specify input
mode.
TIMER A
Figure 41 shows a block diagram of timer A.
Timer A has four modes: timer mode, event counter mode, one-shot
pulse mode, and pulse width modulation mode. The mode is selected with bits 0 and 1 of the timer Ai mode register (i = 0 to 4). Each
of these modes is described below.
Figure 42 shows the bit configuration of the timer A clock division select register. Timers A0 to A4 use the count source which has been
Timer A clock
division select bit
f
2
f
1
f
16
f
64
f
512
f
4096
Count source
select bits
• Timer
• One-shot pulse
• Pulse width
selected by bits 0 and 1 of this register.
(1) Timer mode [00]
Figure 43 shows the bit configuration of the timer Ai mode register
during timer mode. Bits 0, 1 and 5 of the timer Ai mode register must
be “0” in timer mode. The timer A’s count source is selected by bits 6
and 7 of the timer Ai mode register and the contents of the timer A
clock division select register. (See Table 12.)
The counting of the selected clock starts when the count start bit is
“1” and stops when it is “0”.
Figure 44 shows the bit configuration of the count start bit. The
counter is decremented, an interrupt is caused and the interrupt request bit in the timer Ai interrupt control register is set when the contents becomes 000016. At the same time, the contents of the reload
register is transferred to the counter and count is continued.
Data bus (odd)
Data bus (even)
(Low-order 8 bits)(High-order 8 bits)
Reload register(16)
Polarity
IN
OUT
selection
Pulse output
TAi
(i = 0–4)
TAi
(i = 0–4)
Fig. 41 Block diagram of timer A
Timer (gate function)
Event counter
External trigger
Count start register
16
(Address 40
Up-down register
(Address 44
Toggle flip-flop
)
Countdown
16
Counter (16)
Countup/Countdown switching
“Countdown” is always
selected when not in the
event counter mode.
Note: Timers A0 to A4 use the same clock, which is selected by the
timer A clock division select bits.
When bit 2 of the timer Ai mode register is “1”, the output is generated from TAiOUT pin. The output is toggled each time the contents
of the counter reaches to 000016. When the contents of the count
start bit is “0”, “L” is output from TAiOUT pin.
When bit 2 is “0”, TAiOUT can be used as a normal port pin. When bit
4 is “0”, TAiIN can be used as a normal port pin.
When bit 4 is “1”, counting is performed only while the input signal
from the T AiIN pin is “H” or “L” as shown in Figure 45. Therefore, this
can be used to measure the pulse width of the TAiIN input signal.
Whether to count while the input signal is “H” or while it is “L” is determined by bit 3. If bit 3 is “1”, counting is performed while the TAiIN
pin input signal is “H” and if bit 3 is “0”, counting is performed while it
is “L”.
Note that, the duration of “H” or “L” on the TAiIN pin must be 2 or
more cycles of the timer count source.
When data is written to timer Ai register with timer Ai halted, the
same data is also written to the reload register and the counter.
When data is written to timer Ai which is busy, the data is written to
the reload register, but not to the counter. The new data is reloaded
from the reload register to the counter at the next reload time and
counting continues. The contents of the counter can be read at any
time.
When the value set in the timer Ai register is n, the timer frequency
division ratio is 1/(n+1).
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
76543210
Timer A clock division select register
Timer A clock division select bit
(See Table 12.)
“0” at read.
Fig. 42 Bit configuration of timer A clock division select register
Table 12. Relationship between timer A clock division select bits,
clock source select bits, and count source
Address
45
16
7
6543210
0
Note: When using pins TA2
Because they are key input interrupt pins and are multiplexed with pins TA2
00
0 0 : Always “00” in timer mode
0 : No pulse output (TAi
1 : Pulse output (TAi
0
1 0 : Count only while TAi
1 1 : Count only while TAi
Timer A0 count start bit
Timer A1 count start bit
Timer A2 count start bit
Timer A3 count start bit
Timer A4 count start bit
Timer B0 count start bit
Timer B1 count start bit
Timer B2 count start bit
Address
16
40
TAi
IN
Timer mode register
Bit 4Bit 3
10
Timer mode register
Bit 4Bit 3
11
Fig. 45 Count waveform when gate function is available
54
Page 55
M37902FCCHP, M37902FGCHP, M37902FJCHP
(2) Event counter mode [01]
Figure 46 shows the bit configuration of the timer Ai mode register
during event counter mode. In event counter mode, bit 0 of the timer
Ai mode register must be “1” and bits 1 and 5 must be “0”.
The input signal from the TAiIN pin is counted when the count start
bit shown in Figure 44 is “1” and counting is stopped when it is “0”.
Count is performed at the fall of the input signal when bit 3 is “0” and
at the rise of the signal when it is “1”.
In event counter mode, whether to increment or decrement the count
can be selected with the up-down bit or the input signal from the
TAiOUT pin.
When bit 4 of the timer Ai mode register is “0”, the up-down bit is
used to determine whether to increment or decrement the count
(decrement when the bit is “0” and increment when it is “1”). Figure
47 shows the bit configuration of the up-down register.
When bit 4 of the timer Ai mode register is “1”, the input signal from
the TAiOUT pin is used to determine whether to increment or decrement the count. However, note that bit 2 must be “0” if bit 4 is “1.” It is
because if bit 2 is “1”, TAiOUT pin becomes an output pin to output
pulses.
The count is decremented when the input signal from the T AiOUT pin
is “L” and incremented when it is “H”. Determine the level of the input
signal from the TAiOUT pin before a valid edge is input to the TAiIN
pin.
An interrupt request signal is generated and the interrupt request bit
in the timer Ai interrupt control register is set when the counter
reaches 000016 (decrement count) or FFFF16 (increment count). At
the same time, the contents of the reload register is transferred to the
counter and the count is continued.
When bit 2 is “1,” each time the counter reaches 000016 (decrement
count) or FFFF16(increment count), the waveform’s polarity is reversed and is output from TAiOUT pin.
If bit 2 is “0”, TAiOUT pin can be used as a normal port pin.
However, if bit 4 is “1” and the TAiOUT pin is used as an output pin,
the output from the pin changes the count direction. Therefore, bit 4
must be “0” unless the output from the T AiOUT pin is to be used to select the count direction.
Note: When using pins TA2OUT and TA3OUT as pulse output pins, do
not select pins KI
pins and are multiplexed with pins TA2
Timer A3 mode register
Timer A4 mode register
100
0 1 : Always “01” in event counter mode
0 : No pulse output
1 : Pulse output
0 : Count at the falling edge of input signal
1 : Count at the rising edge of input signal
0 : Increment or decrement according
to up/down bit
1 : Increment or decrement according
to TAi
0 : Always “0” in event counter mode
× × : Not used in event counter mode
0 and KI2. Because they are key input interrupt
OUT pin input signal level
Fig. 46 Bit configuration of timer Ai mode register during event counter mode
76543210
Up-down register
Timer A0 up-down bit
Timer A1 up-down bit
Timer A2 up-down bit
Timer A3 up-down bit
Timer A4 up-down bit
Timer A2 two-phase pulse signal
processing select bit
0 : Two-phase pulse signal processing
disabled
1 : Two-phase pulse signal processing
mode
Timer A3 two-phase pulse signal
processing select bit
0 : Two-phase pulse signal processing
disabled
1 : Two-phase pulse signal processing
mode
Timer A4 two-phase pulse signal
processing select bit
0 : Two-phase pulse signal processing
disabled
1 : Two-phase pulse signal processing
mode
Addresses
16
56
5716
5816
5916
5A16
OUT and TA3OUT.
Address
16
44
Fig. 47 Bit configuration of up-down register
55
Page 56
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Data write and data read are performed in the same way as for timer
mode. That is, when data is written to timer Ai halted, it is also written to the reload register and the counter. When data is written to
timer Ai which is busy, the data is written to the reload register, but
not to the counter. The counter is reloaded with new data from the
reload register at the next reload time. The counter can be read at
any time.
In event counter mode, whether to increment or decrement the
counter can also be determined by supplying two kinds of pulses of
which phases differ by 90° to timer A2, A3, or A4. There are two types
of two-phase pulse processing operations. One uses timers A2 and
A3, and the other uses timer A4. In both processing operations, two
pulses described above are input to the TAjOUT (j = 2 to 4) pin and
TAjIN pin respectively.
When timers A2 and A3 are used, as shown in Figure 48, the count
is incremented when a rising edge is input to the TAkIN pin after the
level of TAkOUT(k=2, 3) pin changes from “L” to “H”, and when the
falling edge is input, the count is decremented.
For timer A4, as shown in Figure 49, when a phase-related pulse
with a rising edge input to the TA4IN pin is input after the level of
TA4OUT pin changes from “L” to “H”, the count is incremented at the
respective rising edge and falling edge of the T A4OUT pin and TA4IN
pin.
When a phase-related pulse with a falling edge input to the TA4OUT
pin is input after the level of TA4IN pin changes from “H” to “L”, the
count is decremented at the respective rising edge and falling edge
of the TA4IN pin and TA4OUT pin. When performing this two-phase
pulse signal processing, timer Aj mode register bit 0 and bit 4 must
be set to “1” and bits 1, 2, 3, and 5 must be “0”. Bits 6 and 7 are ig-
nored. (See Figure 50.) Note that bits 5, 6, and 7 of the up-down register (address 4416) are the two-phase pulse signal processing select bits for timers A2, A3 and A4 respectively. Each timer operates
in normal event counter mode when the corresponding bit is “0” and
performs two-phase pulse signal processing when it is “1”.
Count is started by setting the count start bit to “1”. Data write and
read are performed in the same way as for normal event counter
mode. Note that the direction register of the input port must be set to
input mode because two kinds of pulse signals, described above,
are input. Also, there can be no pulse output in this mode.
0 1 : Always “01” in event counter mode
0 1 0 0 : Always “0100” when processing
two-phase pulse signal
×× : Not used in event counter mode
Fig. 50 Bit configuration of timer Aj mode register when performing
two-phase pulse signal processing in event counter mode
TAkOUT
TAkIN
(k = 2, 3)
Incrementcount
Incrementcount
Incrementcount
Decrementcount
Decrementcount
Fig. 48 Two-phase pulse processing operation of timers A2 and A3
TA4
OUT
Decrement-count at each edgeIncrement-count at each edge
TA4
IN
Decrement-count at each edgeIncrement-count at each edge
Fig. 49 Two-phase pulse processing operation of timer A4
Decrementcount
56
Page 57
M37902FCCHP, M37902FGCHP, M37902FJCHP
(3) One-shot pulse mode [10]
Figure 51 shows the bit configuration of the timer Ai mode register
during one-shot pulse mode. In one-shot pulse mode, bit 0 and bit 5
must be “0” and bit 1 and bit 2 must be “1”.
The trigger is enabled when the count start bit is “1”. The trigger can
be generated by software or it can be input from the TAiIN pin. Software trigger is selected when bit 4 is “0” and the input signal from the
TAiIN pin is used as the trigger when it is “1“.
Bit 3 is used to determine whether to trigger at the fall of the trigger
signal or at the rise. The trigger is at the fall of the trigger signal when
bit 3 is “0” and at the rise of the trigger signal when it is “1”.
Software trigger is generated by setting “1” to a bit in the one-shot
start register. Each bit corresponds to each timer.
Figure 52 shows the bit configuration of the one-shot start register.
As shown in Figure 53, when a trigger signal is received, the counter
counts the clock selected by bits 6 and 7 and the contents of the
timer A clock division select register. (Set Table 12.)
If the contents of the counter is not 000016, the TAiOUT pin goes “H”
when a trigger signal is received. The count direction is decrement.
When the counter reaches 000116, the TAiOUT pin goes “L” and
count is stopped. The contents of the reload register is transferred to
the counter. At the same time, an interrupt request signal is generated and the interrupt request bit in the timer Ai interrupt control register is set. This is repeated each time a trigger signal is received.
The output pulse width is
1
pulse frequency of the selected clock
× (counter’s value at the time of trigger).
If the count start flag is “0”, T AiOUT goes “L”. Therefore, the value corresponding to the desired pulse width must be written to timer Ai before setting the timer Ai count start bit.
As shown in Figure 54, a trigger signal can be received before the
operation for the previous trigger signal is completed. In this case,
the contents of the reload register is transferred to the counter by the
trigger and then that value is decremented.
Except when retriggering while operating, the contents of the reload
register are not transferred to the counter by triggering.
When retriggering, there must be at least one timer count source
cycle before a new trigger can be issued.
Data write is performed in the same way as for timer mode.
When data is written in timer Ai halted, it is also written to the reload
register and the counter.
When data is written to timer Ai which is busy, the data is written to
the reload register, but not to the counter. The counter is reloaded
with new data from the reload register at the next reload time.
Undefined data is read when timer Ai is read.
1 0 : Trigger at the falling edge of TAi
input
1 1 : Trigger at the rising edge of TAi
input
0 : Always “0” in one-shot pulse mode
Clock source select bits
(See Table 12.)
Fig. 51 Bit configuration of timer Ai mode register during one-shot
pulse mode
76543210
0
One-shot start register
Timer A0 one-shot start bit
Timer A1 one-shot start bit
Timer A2 one-shot start bit
Timer A3 one-shot start bit
Timer A4 one-shot start bit
Must be fixed to “0”.
Fig. 52 Bit configuration of one-shot start register
Addresses
16
56
57
16
58
16
59
16
5A
16
IN
Address
16
42
IN
57
Page 58
M37902FCCHP, M37902FGCHP, M37902FJCHP
Selected clock
source fi
TAi
IN
(rising edge)
TAi
OUT
Example when the contents of the reload register is 000316
Fig. 53 Pulse output example when external rising edge is selected
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Selected clock
source fi
IN
TAi
(rising edge)
TAiOUT
Example when the contents of the reload register is 000416
Fig. 54 Example when trigger is re-issued during pulse output
58
Page 59
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
(4) Pulse width modulation mode [11]
Figure 55 shows the bit configuration of the timer Ai mode register
during pulse width modulation mode. In pulse width modulation
mode, bits 0, 1, and 2 must be set to “1”.
Bit 5 is used to determine whether to perform 16-bit length pulse
width modulator or 8-bit length pulse width modulator. 16-bit length
pulse width modulator is selected when bit 5 is “0” and 8-bit length
pulse width modulator is selected when it is “1”. The 16-bit length
pulse width modulator is described first.
The pulse width modulator can be started with a software trigger or
with an input signal from a TAiIN pin (external trigger).
The software trigger mode is selected when bit 4 is “0”.
Pulse width modulator is started and a pulse is output from TAiOUT
when the count start bit is set to “1”.
The external trigger mode is selected when bit 4 is “1”.
Pulse width modulation starts when a trigger signal is input from the
TAiIN pin when the count start bit is “1”. Whether to trigger at the fall
or rise of the trigger signal is determined by bit 3. The trigger is at the
fall of the trigger signal when bit 3 is “0” and at the rise when it is “1”.
When data is written to timer Ai with the pulse width modulator
halted, it is written to the reload register and the counter.
Then when the count start bit is set to “1” and a software trigger or
an external trigger is issued to start modulation, the waveform shown
in Figure 56 is output continuously.
Once modulation is started, triggers are not accepted. If the value in
the reload register is m, the duration “H” of pulse is
The low-order 8 bits function as a prescaler and the high-order 8 bits
function as the 8-bit length pulse width modulator. The prescaler
counts the clock selected by bits 6, 7, and the contents of the timer A
clock division select register. (See Table 12.) A pulse is generated
when the counter reaches 000016 as shown in Figure 57. At the
same time, the contents of the reload register is transferred to the
counter and count is continued.
1 1 : Always “11” in pulse width modulation
mode
1 : Always “1” in pulse width modulation
mode
0 × : Software trigger
1 0 : Trigger at the falling of TAi
1 1 : Trigger at the rising of TAi
An interrupt request signal is generated and the interrupt request bit
in the timer Ai interrupt control register is set at each fall of the output
pulse.
The width of the output pulse is changed by updating timer data. The
update can be performed at any time. The output pulse width is
changed at the rise of the pulse after data is written to the timer.
The contents of the reload register are transferred to the counter just
before the rise of the next pulse so that the pulse width is changed
from the next output pulse.
Undefined data is read when timer Ai is read.
The 8-bit length pulse width modulator is described next.
The 8-bit length pulse width modulator is selected when the timer Ai
mode register bit 5 is “1”.
The reload register and the counter are both divided into 8-bit
halves.
Fig. 55 Bit configuration of timer Ai mode register during pulse width
modulation mode
59
Page 60
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Therefore, if the low-order 8 bits of the reload register are n, the period of the generated pulse is
1
selected clock frequency
× (n+1).
The high-order 8 bits function as an 8-bit length pulse width modulator using this pulse as input. The operation is the same as for 16-bit
length pulse width modulator except that the length is 8 bits. If the
1/fi × (2
Selected clock
source fi
IN
TAi
(rising edge)
This trigger is not accepted
1/fi × (m)
TAi
OUT
high-order 8 bits of the reload register are m, the duration “H” of
pulse is
selected clock frequency
And the output pulse period is
selected clock frequency
16
– 1)
1
1
× (n+1) × m.
× (n+1) × (28–1).
Example when the contents of the reload register is 0003
Fig. 56 16-bit length pulse width modulator output pulse example
Selected clock
source fi
IN
TAi
(falling edge)
1/fi × (n + 1)
Prescaler output
(when n = 2)
8-bit length pulse
width modulator
output
(when m = 2)
1/fi × (n + 1) × (2
1/fi × (n + 1) × (m)
8
16
– 1)
Fig. 57 8-bit length pulse width modulator output pulse example
60
Page 61
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
TIMER B
Figure 58 shows a block diagram of timer B.
Timer B has three modes: timer mode, event counter mode, and
pulse period measurement/pulse width measurement mode. The
mode is selected with bits 0 and 1 of the timer Bi mode register (i=0
to 2). Each of these modes is described below.
(1) Timer mode [00]
Figure 59 shows the bit configuration of the timer Bi mode register
during timer mode. Bits 0 and 1 of the timer Bi mode register must
always be “0” in timer mode.
Bits 6 and 7 are used to select the clock source. The counting of the
selected clock starts when the count start bit is “1” and stops when
“0”.
Count source select bits
f
2
f
16
f
64
f
512
• Timer mode
• Pulse period measurement/Pulse
width measurement mode
• Event counter
fX
mode
32
TBiIN
Polarity selection
and edge pulse
generator
Timer B2 clock source
select bit (Note 2)
As shown in Figure 44, the timer Bi count start bit is at the same address as the timer Ai count start bit. The count is decremented, an
interrupt occurs, and the interrupt request bit in the timer Bi interrupt
control register is set when the contents becomes 000016. At the
same time, the contents of the reload register is stored in the counter
and count is continued.
Timer Bi does not have a pulse output function or a gate function like
timer A.
When data is written to timer Bi halted, it is written to the reload register and the counter. When data is written to timer Bi which is busy,
the data is written to the reload register, but not to the counter. The
new data is reloaded from the reload register to the counter at the
next reload time and counting continues.
The contents of the counter can be read at any time.
Data bus (odd)
Data bus (even)
(Low-order 8 bits)(High-order 8 bits)
Reload register (16)
Counter (16)
Timer B0 51
Count start register
(Address 40
16
)
Timer B1 5316 52
Timer B2 5516 54
Addresses
16
50
16
16
16
Timer B2 clock source select bit : Bit 6 at address 63
Notes 1: Perform a write and read to/from timer Bi register in the condition of 16-bit data length : data length flag (m) = “0”.
2: Only for timer B2, a count source in the event counter mode can be selected.
Fig. 58 Block diagram of timer B
Counter reset
circuit
16
61
Page 62
M37902FCCHP, M37902FGCHP, M37902FJCHP
(2) Event counter mode [01]
Figure 60 shows the bit configuration of the timer Bi mode register
during event counter mode. In event counter mode, bit 0 in the timer
Bi mode register must be “1” and bit 1 must be “0”.
The input signal from the TBiIN pin is counted when the count start
bit is “1” and counting is stopped when it is “0”.
Count is performed at the fall of the input signal when bits 2, and 3
are “0” and at the rise of the input signal when bit 3 is “0” and bit 2 is
“1”.
When bit 3 is “1” and bit 2 is “0”, count is performed at the rise and
fall of the input signal.
Data write, data read and timer interrupt are performed in the same
way as for timer mode.
Only for timer B2, when the timer B2 clock source select bit of the
particular function select register 1 (bit 6 at address 6316) = “1” in the
event counter mode, fX32 can be selected. (When this bit is “0”, an
input signal from pin TB2IN becomes the count source as described
above.) For the bit configuration of the particular function select register 1, refer to the section on the standby function.
Note: fX32 = f(XIN)/32
(3) Pulse period measurement/Pulse width
measurement mode [10]
Figure 61 shows the bit configuration of the timer Bi mode register
during pulse period measurement/pulse width measurement mode.
In pulse period measurement/pulse width measurement mode, bit 0
must be “0” and bit 1 must be “1”. Bits 6 and 7 are used to select the
clock source. The selected clock is counted when the count start bit
is “1” and counting stops when it is “0”.
The pulse period measurement mode is selected when bit 3 is “0”. In
pulse period measurement mode, the selected clock is counted during the interval starting at the fall of the input signal from the TBiIN pin
to the next fall or at the rise of the input signal to the next rise; the
result is stored in the reload register. In this case, the reload register
acts as a buffer register.
When bit 2 is “0”, the clock is counted from the fall of the input signal
to the next fall. When bit 2 is “1“, the clock is counted from the rise of
the input signal to the next rise.
In the case of counting from the fall of the input signal to the next fall,
counting is performed as follows. As shown in Figure 62, when the
fall of the input signal from TBiIN pin is detected, the contents of the
counter is transferred to the reload register. Next, the counter is
cleared and count is started from the next clock. When the fall of the
next input signal is detected, the contents of the counter is transferred to the reload register once more, the counter is cleared, and
the count is started. The period from the fall of the input signal to the
next fall is measured in this way.
After the contents of the counter is transferred to the reload register,
an interrupt request signal is generated and the interrupt request bit
in the timer Bi interrupt control register is set. However, no interrupt
request signal is generated when the contents of the counter is transferred first to the reload register after the count start bit is set to “1”.
When bit 3 is “1”, the pulse width measurement mode is selected.
Pulse width measurement mode is the same as the pulse period
measurement mode except that the clock is counted from the fall of
the TBiIN pin input signal to the next rise or from the rise of the input
signal to the next fall as shown in Figure 63.
0 1 : Always “01” in event counter
mode
0 0 : Count at the falling edge of
input signal
0 1 : Count at the rising edge of
input signal
1 0 : Count at the both falling edge
and rising edge of input signal
× × × : Not used in event counter mode
Fig. 60 Bit configuration of timer Bi mode register during event
1 0 : Always “10” in pulse period
measurement/pulse width
measurement mode
0 0 : Count from the falling edge of
input signal to the next falling one
0 1 : Count from the rising edge of
input signal to the next rising one
1 0 : Count from the falling edge of
input signal to the next rising one
and from the rising edge to the
next falling one
Timer Bi overflow flag
Clock source select bits
0 0 : Select f
0 1 : Select f
1 0 : Select f
1 1 : Select f
2
16
64
512
Fig. 61 Bit configuration of timer Bi mode register during pulse period
measurement/pulse width measurement mode
Addresses
5B
16
5C
16
5D
16
Addresses
16
5B
5C
16
5D
16
Addresses
5B
16
5C
16
5D
16
62
Page 63
M37902FCCHP, M37902FGCHP, M37902FJCHP
When timer Bi is read, the contents of the reload register is read.
Note that in this mode, the interval between the fall of the TBiIN pin
input signal to the next rise or from the rise to the next fall must be at
least two cycles of the timer count source.
Timer Bi overflow flag which is bit 5 of timer Bi mode register is set to
“1” when the timer Bi counter reaches 000016, which indicates that a
pulse width or pulse period is longer than that which can be measured by a 16-bit length.
This flag is cleared by writing data to the corresponding timer Bi
mode register. This flag is set to “1”at reset.
Selected clock
source fi
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
TBi
IN
Reload register ← Counter
Counter ← 0
Count start bit
Interrupt request signal
Fig. 62 Pulse period measurement mode operation (example of measuring the interval between the falling edge to next falling one)
Selected clock
source fi
TBi
IN
Reload register ← Counter
Counter ← 0
Count start bit
Interrupt request signal
Fig. 63 Pulse width measurement mode operation
63
Page 64
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
SERIAL I/O PORTS
Two independent serial I/O ports are provided. Figure 64 shows a
block diagram of the serial I/O ports.
Bits 0 to 2 of the UARTi(i = 0,1) transmit/receive mode register
shown in Figure 65 are used to determine whether to use port P8 as
a programmable I/O port, clock synchronous serial I/O port, or asyn-
RXD
i
CTSi/CLK
CTS
i
/RTS
BRG count source select bits
2
f
f
16
f
64
f
512
i
i
n = a value set into the UARTi baud rate register (BRGi)
BRGi
1/(n + 1) divider
Clock synchronous (when internal clock selected)
CLK
i
CTS
i
1/16 divider
Clock synchronous
1/16 divider
Clock synchronous
Clock synchronous
1/2 divider
chronous (UART) serial I/O port which uses start and stop bits.
Figures 66 and 67 show the block diagrams of the receiver/transmitter .
Figure 68 shows the bit configuration of the UARTi transmit/receive
control register.
Each communication method is described below.
Data bus (odd)
Data bus (even)
UART
UART
(Internal clock)
Clock synchronous
(External clock)
000000
Receive
control
circuit
Transmit
control
circuit
7D6D5D4D3D2D1
0D
8
D
UARTi receive register
Transfer clock
Transfer clock
UARTi transmit register
D7D8D6D5D4D3D2D1D
Data bus (odd)
Data bus (even)
Bit converter
UART0 (Addresses 37
UART1 (Addresses 3F
Bit converter
UARTi
0
D
receive buffer register
16, 3616
16
T
XDi
UARTi
0
UART0 (Addresses 33
UART1 (Addresses 3B
transmit buffer register
, 3E16)
16
, 3216)
16
)
, 3A16)
Fig. 64 Block diagram of serial I/O port
76543210
Note: In the clock synchronous serial I/O mode, bits 4 to 6 are invalid. (Each of them may be “0” or “1”.) Furthermore, fix bit 7 to “0”.
Serial I/O mode select bits
0 0 0 : Serial I/O is invalid. (Port P8 functions as a programmable I/O port.)
0 0 1 : Clock synchronous
1 0 0 : 7-bit UART
1 0 1 : 8-bit UART
1 1 0 : 9-bit UART
Internal/External clock select bit
0 : Internal clock
1 : External clock
Stop bit length select bit (Valid in UART mode.)
0 : 1 stop bit
1 : 2 stop bits
Odd/Even parity select bit (Valid in UART mode with the parity enable bit = “1”.) (Note)
0 : Odd parity
1 : Even parity
Parity enable bit (Valid in UART mode) (Note)
0 : No parity
1 : With parity
Sleep select bit (Valid in UART mode) (Note)
0 : No sleep
1 : Sleep
Fig. 65 Bit configuration of UARTi transmit/receive mode register
Addresses
16
30
38
16
64
Page 65
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Data bus (odd)
Data bus (even)
UARTi receive
buffer register
0D1D2D3D4D5D6D7D8
0000000
8-bit UART
2SP
R
XDi
SP
SPPAR
1SP
Parity
No
parity
UART
9-bit UART
7-bit UART
8-bit UART
Synchronous
9-bit UART
Synchronous
7-bit UART
UARTi receive register
Synchronous
D
SP : Stop bit
PAR : Parity bit
Fig. 66 Block diagram of receiver
2SP
SP
SPPAR
1SP
Parity
No
parity
0
UART
Synchronous
Data bus (odd)
Data bus (even)
D
8
7-bit UART
9-bit UART
Synchronous
8-bit UART
D
7
8-bit UART
9-bit UART
Synchronous
7-bit UART
UARTi receive transmit register
D
D
D
D
D
D
D
6
4
5
2
3
0
1
T
XDi
UARTi transmit register
SP : Stop bit
PAR : Parity bit
Fig. 67 Block diagram of transmitter
65
Page 66
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
76543210
MSB
/LSB
CPL
EPTY
T
X
R/C CS1CS
76543210
RERIOERFERPERSUMTI TE
Address
UART0 transmit/receive control register 0
0
UART1 transmit/receive control register 0
34
3C
16
16
BRG count source select bits
2
00 : f
01 : f
16
10 : f
64
11 : f
512
CTS/RTS function select bit (Note 1)
0 : CTS function is selected.
1 : RTS function is selected.
Transmit register empty flag
0 : Data is present in the transmit register. (Transmission is in progress.)
1 : No data is present in the transmit register. (Transmission is completed.)
CTS/RTS enable bit
0 : CTS, RTS function is enabled.
1 : CTS, RTS function is disabled.
UARTi receive interrupt mode select bit
0 : Reception interrupt
1 : Reception error interrupt
CLK polarity select bit (This bit is used in the clock synchronous serial I/O mode.) (Note 2)
0 : At the falling edge of a transfer clock, transmit data is output;
at the rising edge, receive data is input.
When not in transfer, pin CLK’s level is “H”.
1 : At the rising edge of a transfer clock, transmit data is output;
at the falling edge, receive data is input.
When not in transfer, pin CLK’s level is “L”.
Transfer format select bit (This bit is used in the clock synchronous serial I/O mode.) (Note 2)
0 : LSB (Least Significant Bit) first
1 : MSB (Most Significant Bit) first
Address
UART0 transmit/receive control register 1
UART1 transmit/receive control register 1
35
3D
16
16
Transmit enable bit
Transmit buffer empty flag
Receive enable bit
Receive complete flag
Overrun error flag
Framing error flag (Note 3)
Parity error flag (Note 3)
Error sum flag (Note 3)
Notes 1: Valid when the CTS/RTS enable bit (bit 4) = “0”.
2: Fix these bits to “0” in UART mode or when serial I/O is invalid.
3: Valid in UART mode.
Fig. 68 Bit configuration of UARTi transmit/receive control register
66
Page 67
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
CLOCK SYNCHRONOUS SERIAL COMMUNICATION
A case where communication is performed between two clock synchronous serial I/O ports as shown in Figure 69 will be described.
(The transmission side will be denoted by subscript j and the receiving side will be denoted by subscript k.)
Bit 0 of the UARTj transmit/receive mode register and UARTk transmit/receive mode register must be set to “1” and bits 1 and 2 must be
“0”. The length of the transmission data is fixed at 8 bits.
Bit 3 of the UARTj transmit/receive mode register of the clock sending side is cleared to “0” to select the internal clock. Bit 3 of the
UARTk transmit/receive mode register of the clock receiving side is
set to “1” to select the external clock. Bits 4, 5 and 6 are ignored in
clock synchronous mode. Bit 7 must always be “0”.
The clock source is selected by bit 0 (CS0) and bit 1 (CS1) of the
clock-sending-side UARTj transmit/receive control register 0. As
shown in Figure 64, the selected clock is divided by (n+1), then by 2,
is passed through a transmission control circuit, and is output as
transmission clock CLKj. Therefore, when the selected clock is fi,
Bit Rate = fi/ {(n + 1) × 2}
On the clock receiving side, the CS0 and CS1 bits of the UARTk
transmit/receive control register 0 are ignored because an external
clock is selected.
Both of UART0 and UART1 can use CTS and RTS functions.
Bit 4 of the UARTi transmit/receive control register 0 is used to determine whether to use CTS or RTS signal. Bit 4 must be “0” when
____________________________
CTS or RTS signal is used. Bit 4 must be “1” when CTS and RTS signals are not used. When CTS and RTS signals are not used, CTS/
_______
RTS pin can be used as a normal port pin.
When using pin CTS/RTS, :
• If bit 2 of the UARTi transmit/receive control register 0 is cleared to
_______
“0”, CTS input is selected.
• If bit 2 is set to “1”, RTS output is selected.
The case using CTS and RTS signals are explained below. As
______________
_____________________
_______ _______
_______
______________
______________
shown in Figure 76, bits 2 and 3 of the serial I/O pin control register
can determine whether port pins P83 and P87 are used as pins TxDi
or as port pins. When bits 2 and 3 are “0”, P83 and P87 function as
pins TxDi; when bits 2 and 3 are “1”, P83 and P87 function as port
pins. Therefore, in the input-only system where pins TxDi are not
used, pins TxDi can function as port pins.
UARTj transmit register
UARTj transmit buffer register
UARTj receive buffer register
UARTj receive register
UARTj Transmit/Receive mode register
000
UARTj Transmit/Receive control
MSB
CPLCPL
/LSB
UARTj Transmit/Receive control
SUMTITE
register 0
T
X
EPTY
register 1
0
0
1 CS0
CS
RERIOERFERPER
TxDj
RxDj
CLKj
CTSj
TxDk
RxDk
CLKk
RTSk
UARTk transmit register
UARTk transmit buffer register
UARTk receive buffer register
UARTk receive register
UARTk Transmit/Receive mode register
0110
UARTk Transmit/Receive control
MSB
/LSB
UARTk Transmit/Receive control
register 0
T
X
EPTY
register 1
RERIOERFERPERSUMTITE
01
1
Fig. 69 Clock synchronous serial communication
67
Page 68
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Transmission
Transmission is started when bit 0 (TEj flag: transmit enable bit) of
UARTj transmit/receive control register 1 is “1”, bit 1 (TIj flag) of one
is “0”, and CTSj input is “L”. The TIj flag indicates whether the transmit buffer register is empty or not. It is cleared to “0” when data is
written in the transmit buffer register ; it is set to “1” when the contents of the transmit buffer register is transferred to the transmit register and the transmit buffer register becomes empty.
When all of the transmit conditions are satisfied, the transmit data in
the transmit buffer register are transferred to the transmit register,
and transmission starts. As shown in Figure 70, data is output from
TXDj pin each time when transmission clock CLKj changes from “H”
to “L”. (In the clock synchronous serial I/O mode, the polarity of a
transfer clock can be changed. For details, refer to the section on the
selection of the transfer clock polarity.) The data is output from the
least significant bit.
When the transmit register becomes empty after the contents has
been transmitted, data is transferred automatically from the transmit
buffer register to the transmit register if the next transmission start
condition is satisfied. The next transmission is performed
succeedingly. Once transmission has started, the TEj flag, TIj flag,
and CTSj signals are ignored until data transmission completes.
Therefore, transmission is not interrupt when CTSj input is changed
to “H” during transmission.
The transmission start condition indicated by TEj flag, TIj flag, and
________
CTSj is checked while the TENDj signal (shown in Figure 70) is “H”.
Therefore, data can be transmitted continuously if the next transmission data is written in the transmit buffer register and TIj flag is
cleared to “0” before theTENDj signal goes “H”.
Bit 3 (TXEPTYj flag) of UARTj transmit/receive control register 0
changes to “1” at the next cycle just after the TENDj signal goes “H”
and changes to “0” when transmission starts. Therefore, this flag can
be used to determine whether data transmission has completed.
When the TIj flag changes from “0” to “1”, the interrupt request bit in
the UARTj transmit interrupt control register is set to “1”.
________
________
next data reception becomes enabled. Bit 4 (OERk flag) of UARTk
transmit/receive control register 1 is set to “1” when the next data is
transferred from the receive register to the receive buffer register
while RIk flag is “1”, and indicates that the next data was transferred
to the receive register before the contents of the receive buffer register was read. (In other words, this indicates that an overrun error has
occurred.) RIk flag is automatically cleared to “0” when the low-order
byte of the receive buffer register is read or when the REk flag is
cleared to “0”. The OERk flag is cleared when the REk flag is
cleared. Bit 5 (FERk flag), bit 6 (PERk flag), and bit 7 (SUMk flag) are
ignored in clock synchronous mode.
As shown in Figure 64, with clock synchronous serial communication, data cannot be received unless the transmitter is operating because the receive clock is created from the transmission clock.
Therefore, the transmitter must be operating even when there is no
need to sent data from UARTk to UARTj.
Receive
When bit 2 of the UARTk transmit/receive control register 1 is set to
“1”, reception becomes enabled. In this case, when the CLKk signal
is input, the receive operation starts simultaneously with this signal.
__________
The RTSk output is “H” when the REK flag is “0”. When the REK flag
is set to “1”, the RTSk output becomes “L”. This informs the transmitter side that reception becomes enabled. When the receive operation starts, the RTSk output automatically becomes “H”.
When the receive operation starts, the receiver takes data from pin
RxDk each time when the transmit clock (CLKj) turns from “L” to “H”.
Simultaneously with reception, the contents of the receiver register
is shifted bit by bit.
(Note that, in the clock synchronous serial communication, the polarity of a transfer clock can be inverted. For details, refer to the section
on the polarity of the transfer clock.) When an 8-bit data is received,
the contents of the receive register is transferred to the receive buffer
register and bit 3 (RIk flag) of UARTk transmit/receive control register 1 is set to “1”. In other words, the setting “1” to the RIk flag indicates that the receive buffer register contains the received data. At
this time, if the low-order byte of the UARTk receive buffer register is
read out, the RTSk output turns back to “L”. This indicates that the
68
__________
__________
_____
Page 69
Transmission
clock
TE
j
j
TI
CTS
j
j
CLK
T
ENDj
M37902FCCHP, M37902FGCHP, M37902FJCHP
Write in transmit buffer register
i
× (n + 1) × 2
1/f
1/fi × (n + 1) × 2
Transmit register ←Transmit buffer register
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Stopped because TE
j
= “0”
TXD
j
TXEPTY
0D1D2D3D4D5D6D7
D
j
D
0
Fig. 70 Clock synchronous serial I/O timing
Interrupt request at completion of reception
When the RIk flag changes from “0” to “1”, in other words, when the
receive operation is completed, the interrupt request bit of the
UARTk receive interrupt control register can be set to “1”.
The timing when this interrupt request bit is to be set to “1” can be
selected from the following:
• Each reception
• When an error occurs at reception
If bit 5 of the UARTk transmit/receive control register 0 (UART receive interrupt mode select bit) is cleared to “0”, the interrupt request
bit is set to “1” at each reception. If bit 5 is set to “1”, the interrupt request bit is set to “1” only when an error occurs. (In the clock synchronous serial communication, only when an overrun error occurs,
the interrupt request bit is set to “1”.)
Polarity of transfer clock
In the clock synchronous serial communication, by bit 6 of the UARTj
transmit/receive control register 0 (CPL), the polarity of a transfer
clock can be selected.
As shown in Figure 71, when bit 6 = “0”, the polarity is as follows:
• In transmission, transmit data is output at the falling edge of CLKj.
• In reception, receive data is input at the rising edge of CLKk.
• When not in transfer, CLKi is at “H” level.When bit 6 = “1”, the polarity is as follows:
• In transmission, transmit data is output at the rising edge of CLKj.
• In reception, receive data is input at the rising edge of CLKk.
• When not in transfer, CLKi is at “L” level.
D1D2D3D4D5D6D
7
D0D1D2D3D4D5D6D
7
69
Page 70
■ CLK polarity select bit = 0
CLKi
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
TxDi
RxDi
❇ Transmit data is output to pin TxDi at the falling edge of transfer clock, and receive data is input from pin
RxDi at the rising edge of transfer clock.
When not in transfer, pin CLKi’s level is “H”.
D0D1D2D3D4D5D6D7
D0D1D2D3D4D5D6D7
■ CLK polarity select bit = 1
CLKi
TxDi
D0D1D2D3D4D5D6D7
RxDi
❇ Transmit data is output to pin TxDi at the rising edge of transfer clock, and receive data is input from pin
RxDi at the falling edge of transfer clock.
When not in transfer, pin CLKi’s level is “L”.
Fig. 71 Polarity of transfer clock
70
D0D1D2D3D4D5D6D7
Page 71
M37902FCCHP, M37902FGCHP, M37902FJCHP
Selection of transfer format
In clock synchronous serial communication, transfer format can be
selected by bit 7 of the transmit/receive control register 0. When bit 7
is “0”, transfer format is LSB first. When bit 7 is “1”, transfer format is
MSB first.
This function is realized by changing connection relation between
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
the transmit buffer register and the receive buffer register when writing transmit data to the transmit buffer register or reading receive
data from the receive buffer register. Accordingly, the transmitter’s
operation is the same in both transfer formats.
Figure 72 shows the connection relation.
Bit 7 in transmit/receive
control register 0
0
(LSB first)
1
(MSB first)
Data bus
DB
DB
DB
DB
DB
DB
DB
DB
Data bus
DB
DB
DB
DB
DB
DB
DB
DB
Write to transmit
buffer register
Transmit buffer
7
6
5
4
3
2
1
0
Transmit buffer
7
6
5
4
3
2
1
0
register
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
register
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
Read from receive
buffer register
Data bus
DB
7
DB
6
DB
5
DB
4
DB
3
DB
2
DB
1
DB
0
Data bus
DB
7
DB
6
DB
5
DB
4
DB
3
DB
2
DB
1
DB
0
Receive buffer
register
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
Receive buffer
register
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
Fig. 72 Connection relation between transmit buffer register, receive buffer register, and data bus
Precautions for clock synchronous serial
communication
When using pin CTS0/RTS0, be sure to clear the D-A2 output enable
bit (bit 2 at address 9616) to “0” (output disabled). Also, in the clock
synchronous serial communication, the separate function for CTSi/
_______
RTSi cannot be selected. Furthermore, when an internal clock is se-
______________
lected, RTS output is undefined. Therefore, do not use the RTS function.
Before transmit operation is performed, be sure to clear bits 2 and 3
of the serial I/O pin control register (address AC16) to “00”.
________ ________
_______
71
Page 72
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
ASYNCHRONOUS
SERIAL COMMUNICATION
Asynchronous serial communication can be performed using 7-, 8-, or
9-bit length data. The operation is the same for all data lengths. The
following is the description for 8-bit asynchronous communication.
With 8-bit asynchronous communication, bit 0 of UARTi transmit/receive mode register is “1”, bit 1 is “0”, and bit 2 is “1”.
Bit 3 is used to select an internal clock or an external clock. If bit 3 is
“0”, an internal clock is selected and if bit 3 is “1”, then external clock
is selected. If an internal clock is selected, bit 0 (CS0) and bit 1 (CS1)
of UARTi transmit/receive control register 0 are used to select the
clock source. When an internal clock is selected for asynchronous
serial communication, the CLKi pin can be used as a normal I/O pin.
The selected internal or external clock is divided by (n+1), then by
16, and is passed through a control circuit to create the UART transmission clock or UART receive clock.
Therefore, the transmission speed can be changed by changing the
contents (n) of the bit rate generator. If the selected clock is an internal clock Pfi or an external clock fEXT,
Bit Rate = (fi or fEXT) / {(n+1)×16}
(1/fi or 1/f
EXT
) × (n + 1) × 16
Transmission clock
Bit 4 is the stop bit length select bit to select 1 stop bit or 2 stop bits.
Bit 5 is a select bit of odd parity or even parity.
In the odd parity mode, the parity bit is adjusted so that the sum of 1s
in the data and parity bit is always odd.
In the even parity mode, the parity bit is adjusted so that the sum of
the 1s in the data and parity bit is always even.
Bit 6 is the parity bit select bit which indicates whether to add parity
bit or not.
Bits 4 to 6 must be set or reset according to the data format used in
the communicating devices.
Bit 7 is the sleep select bit. The sleep mode is described later.
Figure 76 shows the bit configuration of the serial I/O pin control register. By bits 0 and 1 of the serial I/O pin control register (CTSi/RTSi
separate select bits), the function of the CTS/RTS pin can be sepa-
_______ _______
________ ________
rated into two functions, and each function can be assigned to two
different pins. When bits 0 and 1 = “11”, the above separation is performed. When bits 0 and 1 = “00”, no separation is performed.
Table 13 lists the selection methods of the CTS/RTS function.
_______ _______
TE
i
TI
i
CTS
i
T
ENDi
TXD
i
TXEPTY
Written in transmit buffer register
Start bitParity bit Stop bit
D0D
1
ST
i
D2D3D4D5D6D7PSPSTD0D1D2D3D4D5D6D7PSPSTD0D
Transmit register ← Transmit
buffer register
Fig. 73 Transmit timing example when 8-bit asynchronous communication with parity and 1 stop bit selected
Fig. 74 Transmit timing example when 9-bit asynchronous communication with no parity and 2 stop bits selected
72
Page 73
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Transmission
Transmission is started when bit 0 (TEi flag transmit enable flag) of
UARTi transmit/receive control register 1 is “1”, bit 1 (TIi flag) is “0”,
________
and CTSi input (in other words, transmit enable signal input from receiver) is “L.” The TIi flag indicates whether the transmit buffer is
empty or not. It is cleared to “0” when data is written in the transmit
buffer; it is set to “1” when the contents of the transmit buffer register
is transferred to the transmit register.
When all of the transmission conditions are satisfied, transmit data
is transferred to the transmit register, and transmit operation starts.
As shown in Figures 73 and 74, data is output from the TXDi pin with
the stop bit or parity bit specified by bits 4 to 6 of UARTi transmit/receive mode register. The data is output from the least significant bit.
When the transmit register becomes empty after the contents has
been transmitted, data is transferred automatically from the transmit
buffer register to the transmit register if the next transmit start condition is satisfied. Then, the next transmission is performed
succeedingly.
f
i
or f
EXT
RE
i
RXD
i
Receive
clock
i
RI
Start bit
Check to be “L” level
Starting at the falling
edge of start bit
0
D
Once transmission has started, the TEi flag, TIi flag, and CTSi signal
are ignored until data transmission is completed.
Therefore, transmission does not stop until it completes event if, during transmission, the TEi flag is cleared to “0” or CTSi input is set to
“1”.
The transmission start condition indicated by TEi flag, TIi flag, and
________
CTSi is checked while the TENDi signal shown in Figure 73 is “H”.
Therefore, data can be transmitted continuously if the next transmission data is written in the transmit buffer register and TIi flag is
cleared to “0” before the TENDi signal goes “H”.
Bit 3 (TXEPTYi flag) of UARTi transmit/receive control register 0
changes to “1” at the next cycle just after the TENDi signal goes “H”
and changes to “0” when transmission starts. Therefore, this flag can
be used to determine whether data transmission is completed.
When the TIi flag changes from “0” to “1”, the interrupt request bit of
the UARTi transmit interrupt control register is set to “1”.
Stop bitStart bit
Data fetched
D
1
D
7
________
RTS
i
Fig. 75 Receive timing example when 8-bit asynchronous communication with no parity and 1 stop bit selected
73
Page 74
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Table 13. Selection methods of CTS/RTS function
CTS/RTS
enable bit
0
1
✕: It may be “0” or “1”.
Notes 1: When using the CTS0/RTS0 pin, be sure to clear the D-A2 output enable bit (bit 2 at address 9616) to “0”.
2: When using the CTS function, be sure to clear the corresponding bit of the port P8 direction register to “0”.
3: When CTSi and RTSi has been separated, the CLKi pin cannot be used. Therefore, in the clock synchronous serial communication, CTSi and RTSi
cannot be separated. Also, when CTSi and RTSi are separated in UART mode, be sure to select an internal clock.
CTSi/RTS
i
separate select bit
0
1
✕
CTS/RTS
function select bit
0
1
✕
✕
Pin P80/CTS0/RTS0 (Note 1)
0
CTS
RTS
0
RTS
0
P8
0
Pin P81/CTS0/CLK
P81 or CLK
P81 or CLK
CTS0 (Notes 2 and 3)
P81 or CLK
Receive
76543210
Serial I/O pin control register
CTS0/RTS0 separate select bit
0 : CTS
1 : CTS
CTS
0 : CTS
1 : CTS
TxD
0 : Functions as TxD
1 : Functions as P8
TxD1/P87 switch bit
0 : Functions as TxD
1 : Functions as P8
0
/RTS0 are used together.
0
/RTS0 are separated.
1
/RTS1 separate select bit
1
/RTS1 are used together.
1
/RTS1 are separated.
0
/P83 switch bit
Address
AC
16
0.
3.
1.
7.
Fig. 76 Bit configuration of serial I/O pin control register
At reset
16
X0
Receive is enabled when bit 2 (REi flag) of UARTi transmit/receive
control register 1 is set to “1.” As shown in Figure 75, the frequency
divider circuit (1/16) at the receiving side begin to work when a start
bit arrives and the data is received.
________
If RTSi output is selected by setting bit 2 of UARTi transmit/receive
control register 0 to “1”, the RTSi output is “H” when the REi flag is
“0”. When the REi flag changes to “1”, the RTSi output goes “L” to
inform the receiver that reception has become enabled. When the
receive operation starts, the RTSi output automatically becomes “H”.
The entire transmission data bits are received when the start bit
passes the final bit of the receive block shown in Figure 66. At this
point, the contents of the receive register is transferred to the receive
buffer register and bit 3 (Rli flag) of UARTi transmit/receive control
register 1 is set to “1.” In other words, the RIi flag indicates that the
receive buffer register contains data when it is set to “1.” At this time,
when the low-order byte of the UARTk receive buffer register is read
________
out, RTSi output goes back to “L” to indicate that the register is ready
to receive the next data.
Bit 4 (OERi flag) of UARTi transmit/receive control register 1 is set to
“1” when the next data is transferred from the receive register to the
receive buffer register while the RIi flag is “1”, in other words, when
an overrun error occurs. If the OERi flag is “1”, it indicates that the
next data has been transferred to the receive buffer register before
the contents of the receive buffer register has been read.
Bit 5 (FERi flag) is set to “1” when the number of stop bits is less than
required (framing error).
Bit 6 (PERi flag) is set to “1” when a parity error occurs.
Bit 7 (SUMi flag) is set to “1” when either the OERi flag, FERi flag, or
the PERi flag is set to “1.” Therefore, the SUMi flag can be used to
determine whether there is an error.
The setting of the RIi flag, OERi flag, FERi flag, and the PERi flag is
performed while transferring the contents of the receive register to
the receive buffer register.
Functions
0
0
0
0
Pin P84/CTS1/RTS
CTS
1
RTS
1
RTS
1
P8
4
________
________
Pin P85/CTS1/CLK
1
CTS1 (Notes 2 and 3)
________
P85 or CLK
P85 or CLK
P85 or CLK
1
1
1
1
74
Page 75
M37902FCCHP, M37902FGCHP, M37902FJCHP
The FERi, PERi, and SUMi flags are cleared to “0” when reading the
low-order byte of the receive buffer register or when writing “0” to the
REi flag.
The OERi flag is cleared to “0” when writing “0” to the REi flag.
Interrupt request at completion of reception
When the RIk flag changes from “0” to “1”, in other words, when the
receive operation is completed, the interrupt request bit of the
UARTk receive interrupt control register can be set to “1”.
The timing when this interrupt request bit is to be set to “1” can be
selected from the following:
• Each reception
• When an error occurs at reception
If bit 5 of the UARTk transmit/receive control register 0 (UART receive interrupt mode select bit) is cleared to “0”, the interrupt request
bit is set to “1” at each reception. If bit 5 is set to “1”, the interrupt
request bit is set to “1” only when an error occurs. (In the clock asynchronous serial communication, when an overrun error, framing error, or parity error occurs, the interrupt request bit is set to “1”.)
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Sleep mode
The sleep mode is used to communicate only between certain microcomputers when multiple microcomputers are connected through
serial I/O.
The microcomputer enters the sleep mode when bit 7 of UARTi
transmit/receive mode register is set to “1.”
The operation of the sleep mode for an 8-bit asynchronous communication is described below.
When sleep mode is selected, the contents of the receive register is
not transferred to the receive buffer register if bit 7 (bit 6 if 7-bit asynchronous communication and bit 8 if 9-bit asynchronous communication) of the received data is “0”. Also the RIi, OERi, FERi, PERi,
and the SUMi flags are unchanged. Therefore, the interrupt request
bit of the UARTi receive interrupt control register is also unchanged.
Normal receive operation takes place when bit 7 of the received data
is “1”.
The following is an example of how the sleep mode can be used.
The main microcomputer first sends data: bit 7 is “1” and bits 0 to 6
are set to the address of the subordinate microcomputer to be communicated with. Then all subordinate microcomputers receive this
data. Each subordinate microcomputer checks the received data,
clears the sleep bit to “0” if bits 0 to 6 are its own address and sets
the sleep bit to “1” if not. Next, the main microcomputer sends data
with bit 7 cleared. Then the microcomputer which cleared the sleep
bit will receive the data, but the microcomputers which set the sleep
bit to “1” will not. In this way , the main microcomputer is able to communicate only with the designated microcomputer.
Precautions for clock asynchronous (UART)
serial communication
When using pin CTS0/RTS0, be sure to clear the D-A2 output enable
bit (bit 2 at address 9616) to “0” (output disabled). Also, when CTSi
_______
and RTSi are separated, pin CLKi cannot be used. Therefore, when
______________
CTSi and RTSi are separated in UART mode, be sure to select an
internal clock.
Before transmit operation is performed, be sure to clear bits 2 and 3
of the serial I/O pin control register (address AC16) to “00”.
________ ________
_______
75
Page 76
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
A-D CONVERTER
The A-D converter is a 10-bit successive approximation converter.
Figure 77 shows the block diagram of the A-D converter, Figure 78
shows the bit configuration of the A-D control register 0 (address
16), and Figure 79 shows the bit configuration of the A-D control
1E
register 1 (address 1F
16).
A-D conversion accuracy
Bit 3 of A-D control register 1 is used to select whether to regard the
conversion result as 10-bit or as 8-bit data. The conversion result is
regarded as 10-bit data when bit 3 is “1” and as 8-bit data when bit 3
is “0”.
When the conversion result is used as 10-bit data, the low-order 8
bits of the conversion result is stored in the even address of the corresponding A-D register and the high-order 2 bits are stored in bits 0
and 1 at the odd address of the corresponding A-D register. Bits 2 to
7 of the A-D register odd address are “0000002” when read.
1
f
f2
V
REF
connection select bit
0
V
AV
REF
1
SS
Resistor ladder network
V
ref
When the conversion result is used as 8-bit data, the conversion result are stored in even address of the corresponding A-D register. In
this case, the value at the A-D register’s odd address is “0016” when
read.
A-D conversion frequency
An operation clock (
7 of the A-D control register 0 and bit 4 of the A-D control register 1.
When bit 4 of the A-D control register 1 is “0”,
when bit 7 of the A-D control register 0 is “0”,
bit 7 of the A-D control register 0 is “1”.
When bit 4 of the A-D control register 1 is “1”,
bit 7 of the A-D control register 0 is “0”,
the A-D control register 0 is “1”. Note that
the fastest speed) can be selected only in the 8-bit mode.
φ
AD during A-D conversion must be 250 kHz or more because the
A-D conversion can be started by an internal trigger or by an external trigger.
An internal trigger is selected when bit 5 of A-D control register 0 is
“0” and an external trigger is selected when it is “1”. When trigger is
selected, A-D conversion is started when bit 6 (A-D conversion start
bit) is set to “1.”
When an external trigger is selected, the polarity of a trigger input
can be selected by bit 5 of the A-D control register 1. When bit 5 =
“0”, a falling edge is selected, and when bit 5 = “1”, a rising edge is
selected.
A-D conversion starts when the A-D conversion start bit is “1” and the
______
ADTRG input changes from “H” to “L” (or “L” to “H.”) In this case, the
pins that can be used for A-D conversion are AN0 to AN6 because the
______
ADTRG pin is multiplexed with an analog voltage input pin, AN7. If an
76543210
A-D control register 0
Analog input select bits (Note 1)
(Valid in the one-shot and repeat modes.)
0 0 0 : Select AN
1 1 1 : Select AN7 (Note 5)
A-D operation mode select bits
0 0 : One-shot mode
0 1 : Repeat mode
1 0 : Single sweep mode
1 1 : Repeat sweep mode
Trigger select bit
0 : Internal trigger
1 : External trigger due to AD
A-D conversion start bit (Note 7)
0 : Stop A-D conversion
1 : Start A-D conversion
A-D conversion frequency (φ
external trigger is selected, even when the A-D conversion is completed, the A-D conversion start bit keeps “1”. Also, a retrigger can be
available even when A-D conversion is in progress.
VREF connection
Whether to connect the reference voltage input (VREF) with the resistor ladder network or not depends on bit 6 of the A-D control register
1. The VREF pin is connected when bit 6 is “0” and is disconnected
when bit 6 is “1” (High impedance state).
When A-D conversion is not performed, current from the VREF pin to
the resistor ladder network can be cut off by disconnecting resistor
ladder network from the VREF pin.
Before starting A-D conversion, wait for 1 µs or more after clearing
bit 6 to “0”.
Address
1E
16
0
1
2
3
6
(Note 4)
TRG
input (Note 6)
AD
) select bit 0
Notes 1: Ignored in the single sweep and repeat sweep modes. (Each of these bits may be “0” or “1”.)
2: When using the AN
3: When using the AN
4: When using the AN
5: When using the AN
output enable bit (bit 1 at address 96
6: When using an external trigger, be sure to clear the INT
enable bit (bit 1 at address 96
7: For writing to this bit, use the MOVM (MOVMB) instruction, or the STA (STAB, STAD) instruction.
8: Rewriting to each bit of the A-D control register 0 (except for bit 6) must be performed while A-D conversion is stopped.
4
pin, be sure to clear the INT3 pin select bit (bit 5 at address 9416) to “0”.
5
pin, be sure to clear the INT4 pin select bit (bit 6 at address 9416) to “0”.
6
pin, be sure to clear the D-A0 output enable bit (bit 0 at address 9616) to “0” (output disabled).
7
pin, be sure to clear both of the INT2 pin select bit (bit 4 at address 9416) and the D-A1
Fig. 78 Bit configuration of A-D control register 0
16
) to “0”.
16
) to “0”.
2
pin select bit (bit 4 at address 9416) and D-A1 output
(Valid in the single sweep mode and repeat sweep mode.)
0 0 : AN
0
, AN
0 1 : AN0–AN
1 0 : AN0–AN
1 1 : AN0–AN
Must be “0”.
Resolution select bit
0: 8-bit mode
1: 10-bit mode
A-D conversion frequency (φ
External trigger polarity select bit
(Valid when external trigger is selected.)
0: Falling edge of input signal to the AD
1: Rising edge of input signal to the AD
Notes 1: Ignored in the one-shot and repeat modes. (Each of these bits may be “0” or “1”.)
2: When using the AN
3: When using the AN
4: When using the AN
“0” (output disabled).
5: When using the AN
and the D-A
the AN
6: Once this bit is cleared from “1” to “0”, it is necessary to wait for 1 µs or more before the A-D or
D-A conversion starts.
7: Rewriting to each bit of the A-D control register 1 must be performed while A-D conversion is stopped.
1 output enable bit (bit 1 at address 9616) to “0”. When an external trigger is selected,
7 pin cannot be used as an analog input pin.
Fig. 79 Bit configuration of A-D control register 1
4 pin, be sure to clear the INT3 pin select bit (bit 5 at address 9416) to “0”.
5 pin, be sure to clear the INT4 pin select bit (bit 6 at address 9416) to “0”.
6 pin, be sure to clear the D-A0 output enable bit (bit 0 at address 9616) to
7 pin, be sure to clear both of the INT2 pin select bit (bit 4 at address 9416)
78
Page 79
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Operation mode
The operation mode is selected by bits 3 and 4 of A-D control register 0. The available operation modes are one-shot, repeat, single
sweep, and repeat sweep. Analog input port pins are multiplexed
with port P7 pins. Therefore, bits which correspond to pins for A-D
conversion must be “0” (input mode).
(1) One-shot mode
One-shot mode is selected when bits 3 and 4 of A-D control register
0 are “0” is “0”. The A-D conversion pins are selected with bits 0 to 2
of A-D control register 0. A-D conversion can be started by a software trigger or by an external trigger.
When an internal trigger is selected, A-D conversion is started when
bit 6 (A-D conversion start bit) is set to “1.”
When bit 3 of the A-D control register 1 is “1”, A-D conversion ends
after 59 φAD cycles, and the interrupt request bit of the A-D interrupt
control register is set to “1.” At the same time, A-D control register 0
bit 6 (A-D conversion start bit) is cleared to “0” and A-D conversion
stops. The result of A-D conversion is stored in the A-D register corresponding to the selected pin.
If an external trigger is selected, A-D conversion starts when the A-D
conversion start bit is “1” and a valid edge is input to the ADTRG pin,
This operation is the same as that for internal trigger except that
the A-D conversion start bit is not cleared after A-D conversion and
a retrigger can be available during A-D conversion.
(2) Repeat mode
Repeat mode is selected when bit 3 of A-D control register 0 is “1”
and bit 4 is “0”.
The operation of this mode is the same as the operation of one-shot
mode except that when A-D conversion of the selected pin is complete and the result is stored in the A-D register, conversion does not
stop, but is repeated.
No interrupt request is generated in this mode. Furthermore, if an
external trigger is selected, the A-D conversion start bit is not
cleared.
The contents of the A-D register can be read at any time.
When A-D conversion of all selected pins end, the interrupt request
bit of the A-D conversion interrupt control register is set to “1.” At the
same time, A-D conversion start bit is cleared to “0” and A-D conversion stops.
When an external trigger is selected, A-D conversion starts when the
A-D conversion start bit is “1” and a valid edge is input to the ADTRG
pin. In this case, the A-D conversion result which is stored in the A-D
register 7 becomes invalid.
The operation by external trigger is the same as that by an internal
trigger except that the A-D conversion start bit is not cleared to “0”
after A-D conversion and a retrigger can be available during A-D
conversion.
(4) Repeat sweep mode
Repeat sweep mode is selected when bit 3 of A-D control register 0
is “1” and bit 4 is “1”.
The difference from the single sweep mode is that A-D conversion
does not stop after conversion for all selected pins, but repeats again
from the AN0 pin. The repeat is performed among the selected pins.
Also, no interrupt request is generated. Furthermore, if an internal
trigger is selected, the A-D convension start bit is not cleared. The
A-D register can be read at any time.
Precautions for A-D conversion interrupt
function
Clear the interrupt request bit of the A-D interrupt control register (bit
3 at address 7016) before using the A-D interrupt. It is because the
interrupt request bit is undefined just after reset.
(3) Single sweep mode
Single sweep mode is selected when bit 3 of A-D control register 0 is
“0” and bit 4 is “1”.
In the single sweep mode, the number of analog input pins to be
swept can be selected. Analog input pins are selected by bits 1 and
0 of the A-D control register 1 (address 1F16). Two pins, four pins, six
pins, or eight pins can be selected as analog input pins, depending
on the contents of these bits.
A-D conversion is performed only for selected input pins. After A-D
conversion is performed for input of AN0 pin, the conversion result is
stored in A-D register 0, and in the same way, A-D conversion is performed for selected pins one after another. After A-D conversion is
performed for all selected pins, the sweep is stopped.
A-D conversion can be started with an internal trigger or with an external trigger input. An internal trigger is selected when bit 5 of the AD control register 0 (address 1E16) is “0” and an external trigger is
selected when it is “1”.
When an internal trigger is selected, A-D conversion is started when
A-D control register 0 bit 6. (A-D conversion start bit) is set to “1.”
79
Page 80
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
D-A CONVERTER
Three independent D-A converters are included in this microcomputer, and each D-A converter adopts an 8-bit R-2R method. Figure
80 shows the block diagram of the D-A converter, Figure 81 shows
the bit configuration of the A-D control register 1, and Figure 82
shows the bit configuration of the D-A control register.
D-A conversion is performed by writing a value to the corresponding
D-A register i. Whether to output the analog voltage or not is determined by bits 0 to 2 of the D-A control register. When any of bits 0 to
2 = “1”, the corresponding pin (D-A0 to D-A2) outputs the analog voltage.
This analog voltage (V) is determined according to value n. (“n” =
decimal number. This has been set in the D-A register.)
V = VREF✕ n/256 (n = 0 to 255)
VREF : Reference voltage
The contents of the corresponding D-A output enable bit and D-A
register are cleared to “0” at reset. Whether to connect the reference
voltage input (VREF) with the ladder network or not depends on bit 6
of the A-D control register 1. Pin VREF is connected with the ladder
network when bit 6 = “0” and is disconnected when bit 6 = “1” (high
impedance state). When not performing the A-D or D-A conversion,
current from pin VREF to the ladder network can be cut off by disconnecting ladder network from pin VREF.
Before starting A-D or D-A conversion, be sure to clear bit 6 to “0”,
and then, insert a waiting time of 1 µs or more.
An external buffer is necessary when connecting a low impedance
load with the D-A converter. It is because that a D-A output pin
doesn’t include a buffer.
Pin D-Ai is multiplexed with I/O port pins, analog input pins, serial
I/O pins, and external interrupt input pins. When a D-Ai output enable
bit = “1” (in other words, output is enabled.), however, the corresponding pin cannot function as another I/O pin, which is multiplexed
with pin D-Ai.
Also, when not using the D-A converter , be sure to clear the contents
of the corresponding D-A output enable bit and D-A register to “0”.
76543210
✕✕✕✕✕
✕✕
Note: When bit 6 has been cleared to “0” from “1”, insert a waiting time
of 1 µs or more, and then, start the D-A or A-D conversion.
A-D control register 1
Not used for D-A converter.
V
REF
connection select bit (Note)
0: Connected.
1: Disconnected.
Address
1F
16
Fig. 81 Bit configuration of A-D control register 1
76543210
Note: Pin D-Ai is multiplexed with I/O port pins, analog input pins, serial
I/O pins, and external interrupt input pins. When a D-Ai output
enable bit = “1” (in other words, output is enabled.), however, the
corresponding pin cannot function as another I/O pin, which is
multiplexed with pin D-Ai.
D-A control register
D-A0 output enable bit (Note)
0: Output is disabled.
1: Output is enabled.
D-A1 output enable bit (Note)
0: Output is disabled.
1: Output is enabled.
D-A2 output enable bit (Note)
0: Output is disabled.
1: Output is enabled.
Address
96
16
Fig. 82 Bit configuration of D-A control register
V
REF
AV
SS
Fig. 80 Block diagram of D-A converter
80
V
REF
connection
select bit
0
1
Data bus
R-2R ladder
network
Pin D-A
i
D-A register i (i = 0 to 2)
(Addresses 9816 to 9A16)
D-Ai output enable bit
Page 81
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
REAL-TIME OUTPUT
The real-time output function enables to change the output level of
several pins simultaneously with a specified timer’s counting.
Whether to use the real-time output function is decided by the waveform output select bits of the 8-bit real-time output control register
(bits 0 and 1 at address A016). (See Figure 83.) Also, the real-time
output controlled by the pulse output mode select bit of the real-time
output control register (bit 2 at address A016) and is used in one of
the following ways:
• 4 bits ✕ 2 channels
• 6 bits ✕ 1 channel + 2 bits ✕ 1 channels
(1) Pulse mode 0
When the pulse output mode select bit is cleared to “0”, the microcomputer enters pulse output port is controlled by 2 groups of 4 bits.
Figures 84 and 85 show the bit configuration of the pulse output data
register 0/1 (address A216/A416) and real-time output structure in
pulse mode 0, respectively.
When the waveform output select bits are set to “01” (bit 1 = “0” and
bit 0 = “1”), RTP03 to RTP00 become pulse output port pins, in other
words, RTP0 is selected.
When the waveform output select bits are set to “10” (bit 1 = “1” and
bit 0 = “0”), RTP13 to RTP10 become pulse output port pins, in other
words, RTP1 is selected.
When the waveform output select bits are set to “11” (bit 1 = “1” and
bit 0 = “1”), two groups consisting of RTP13 to RTP10 and RTP03 to
RTP00 become pulse output port pins, in other words, RTP1 and
RTP0 are selected.
When the waveform output select bits are set to “00” (bit 1 = bit 0 =
“0”), port P5 pins become normal programmable I/O port pins.
The contents of the pulse output data register 1 (high-order 4 bits at
address A416), which corresponds to RTP13 to RTP10, is output to
these ports each time when the contents of timer A1 counter becomes “000016”. The contents of the pulse output data register 0
(low-order 4 bits at address A216), which corresponds to RTP03 to
RTP00, is output to these ports each time when the contents of timer
A0 counter becomes “000016”.
When “0” is written to a specified bit of the pulse output data register,
a low-level signal is output to a pulse output port if the counter contents of the timer which corresponds to the bit becomes “000016”:
when “1” is written to the bit, a high-level signal is output to a pulse
output port which corresponds to the bit at the same timing.
00 : Programmable I/O port
01 : RTP0 selected
When pulse mode 0 is selected:
RTP0
When pulse mode 1 is selected:
RTP0
10 : RTP1 selected
When pulse mode 0 is selected:
RTP1
When pulse mode 1 is selected:
RTP1, RTP0
11 : RTP1 and RTP0 selected
When pulse mode 0 is selected:
RTP1 and RTP0
When pulse mode 1 is selected:
RTP1, RTP0
RTP0
Fig. 83 Bit configuration real-time output control register
76543210
Pulse output data register 0
0 pulse output data bit
RTP0
1 pulse output data bit
RTP0
2 pulse output data bit (Note 1)
RTP0
3 pulse output data bit (Note 1)
RTP0
Note 1: Used only in pulse mode 0
2: Used only in pulse mode 1
Fig. 84 Bit configuration of pulse output data register
Address
A2
16
76543210
Pulse output data register 1
RTP0
2 pulse output data bit (Note 2)
3 pulse output data bit (Note 2)
RTP0
0 pulse output data bit
RTP1
1 pulse output data bit
RTP1
2 pulse output data bit
RTP1
3 pulse output data bit
RTP1
Address
A4
16
81
Page 82
M37902FCCHP, M37902FGCHP, M37902FJCHP
Timer A2
Pulse output data register 1
(Address A4
16
)
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Pulse output mode select bit
(Address A0
16
)
MITSUBISHI MICROCOMPUTERS
0
Data bus (Even)
Pulse output data register 0
(Address A216)
Timer A0
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
T
DQ
T
DQ
T
DQ
T
DQ
DQ
T
DQ
T
DQ
T
DQ
T
a
a
a
a
a
a
a
a
a
Port P5i latch
(i = 7 to 0)
(Address B
P57/RTP1
P56/RTP1
P55/RTP1
P54/RTP1
Waveform output select bit (Address A016) bit 1
P53/RTP0
P52/RTP0
P51/RTP0
P50/RTP0
Waveform output select bit (Address A016) bit 0
Port P5i direction register
(Address D16)
“1”
16
“0”
)
3
2
1
0
3
2
1
0
Fig. 85 Real-time output structure in pulse mode 0
(2) Pulse mode 1
When the pulse output mode select bit is set to “1”, the microcomputer enters pulse mode 1, and a pulse output port pins are separately controlled (6 bits and 2 bits).
Figures 86 shows the real-time output structure in pulse mode 1.
When the waveform output select bits are set to “01” (bit 1 = “0” and
bit 0 = “1”), RTP13 to RTP10, RTP03, and RTP02 become programmable I/O port pins. Simultaneously, RTP01 and RTP00 become
pulse output port pins.
When the waveform output select bits are set to “10” (bit 1 = “1” and
bit 0 = “0”), RTP13 to RTP10, RTP03, and R TP02 become pulse output port pins. At this time, RTP01 and R TP00 become programmable
I/O port pins.
When the waveform output select bits are set to “11” (bit 1 = bit 0 =
82
“1”), pulse output port pins are divided into two groups; one consists
of RTP13 to RTP10, RTP03, RTP02 and the other consists of R TP01
and RTP00.
When the waveform output select bits are set to “00” (bit 1 = bit 0 =
“0”), port P5 pins become normal programmable I/O port pins.
RTP13 to RTP10, RTP03, and RTP02 are controlled by timer A2.
Also, RTP01 and RTP00 are controlled by timer A0.
The contents of the pulse output data register 1 (high-order 6 bits at
address A416), which corresponds to RTP13 to RTP10, RTP03, and
RTP02, are output to this port each time when the contents of timer
A2 counter becomes “000016”. The contents of the pulse output data
register 0 (low-order 2 bits at address A216), which corresponds to
RTP01 and RTP00, are output to this port each time when the contents of timer A0 counter become “000016”.
Page 83
M37902FCCHP, M37902FGCHP, M37902FJCHP
Timer A2
Pulse output data register 1
(Address A4
16
)
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Pulse output mode select bit
(Address A016)
MITSUBISHI MICROCOMPUTERS
1
Data bus (Even)
Pulse output data register 0
(Address A2
16
)
Timer A0
D
7
D
6
D
5
D
4
D
3
D
2
D
1
D
0
T
DQ
T
DQ
T
DQ
T
DQ
T
DQ
T
DQ
DQ
T
DQ
T
a
a
a
a
a
a
a
a
a
(i = 7 to 0)
Port P5i latch
(Address B
P57/RTP1
P56/RTP1
P55/RTP1
P54/RTP1
P53/RTP0
P52/RTP0
Waveform output select bit (Address A0
P51/RTP0
P50/RTP0
Waveform output select bit (Address A016) bit 0
Port P5i direction register
(Address D
16
“0”
)
“1”
3
2
1
0
3
2
1
0
16
)
16
) bit 1
Fig. 86 Real-time output structure in pulse mode 1
Table 14 lists the port P5/RTP pin output when all of the port P5 direction registers are set to the output mode.
Precautions for real-time output function
After reset, the port P5 direction register is set to the input mode, and
port P5i (i = 0 to 7) pins function as normal I/O port pins. When using
these pins as real-time output port pins, set the corresponding bits of
the port P5 direction register to the output mode. Additionally, by
reading the real-time output port’s value from the port P5 register,
output level of pins can be read out.
Table 14 Port P5/RTP pin output
Real-time output
control register
(Address A0
bit
bit
2
1
0
0
0
1
1
0
0
1
1
1
16)
bit
Store address for port P5/RTP pin output data
bit
bit
7
0B
0B
A4
A4
0B
0B
A4
A4
6
0B
0B
A4
A4
0B
0B
A4
A4
0
0
1
0
1
0
1
0
1
bit
0B
0B
A4
A4
0B
0B
A4
A4
bit
4
5
0B
0B
A4
A4
0B
0B
A4
A4
Address 0B16: Port P5
Address A216: Pulse output data register 0
Address A416: Pulse output data register 1
bit
0B
A2
0B
A2
0B
0B
A4
A4
3
bit2bit
0B
A2
0B
A2
0B
0B
A4
A4
0B
A2
0B
A2
0B
A2
0B
A2
bit
1
0
0B
A2
0B
A2
0B
A2
0B
A2
83
Page 84
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
WATCHDOG TIMER
The watchdog timer is used to detect unexpected execution sequence caused by software runaway and others. Figure 87 shows
the block diagram of the watchdog timer.
The watchdog timer consists of a 12-bit binary counter.
The watchdog timer counts clock Wf32, which is obtained by dividing
the peripheral devices’ clock f2 by 16; or clock Wf512, which is obtained by doing it by 256. Bit 0 of the watchdog timer frequency select register (watchdog timer frequency select bit) shown in Figure 88
selects which clock is to be counted.
Wf512 is selected when this bit 0 is “0”, and Wf32 is selected when bit
0 is “1”. Bit 0 is cleared to “0” after reset.
FFF16 is set in the watchdog timer when “L” level voltage is applied
to pin RESET, STP instruction is executed, data is written to the
watchdog timer register (address 6016), or the most significant bit of
the watchdog timer becomes “0”.
After FFF16 is set in the watchdog timer, when the watchdog timer
counts Wf32 or Wf512 by 2048 counts, the most significant bit of
watchdog timer becomes “0”, the watchdog timer interrupt request
bit is set to “1”, and FFF16 is set again in the watchdog timer.
In program coding, make sure that data is written in the watchdog
timer before the most significant bit of the watchdog timer becomes
“0”. If this routine is not executed owing to unexpected program execution or others, the most significant bit of the watchdog timer be-
comes “0” and an interrupt is generated.
The microcomputer can generate a reset pulse by writing “1” to bit 6
(software reset bit) of processor mode register 0 in an interrupt routine and can be restarted.
The watchdog timer can also be used to return from the STP state,
where a clock has stopped its operation owing to the STP instruction
execution. For details, refer to the sections on the clock generating
circuit and standby function.
The watchdog timer stops its operation in the following cases, and at
this time, input to the watchdog timer is disabled:
• When the external area is accessed in the hold state
• In the wait mode
• In the stop mode
76543210
76543210
Fig. 88
Bit configuration of watchdog timer frequency select register
Watchdog timer frequency select register
Watchdog timer frequency select bit
512
0 : Wf
1 : Wf
32
Watchdog timer clock source select bits at STP
state termination
32
0 0 : fX
0 1 : fX
16
1 0 : fX
128
1 1 : fX
64
Address
16
61
Access to
external area
HLDA
f2
Wait mode
Divided f(XIN)
fX16
fX32
fX64
fX128
Watchdog timer clock source select
bits at STP state termination
• Watchdog timer register: address 60
• Watchdog timer frequency select register: bit 0 at address 61
• Watchdog timer clock source select bits at STP state termination: bits 6, 7 at address 61❈ When the most significant bit of the watchdog timer becomes “0”, this signal will be generated.
Note: During the stop mode and until the stop mode is terminated, setting for disabling the
1/16
1/16
Disables watchdog
timer (Note).
Writing to watchdog
timer register
RESET
STP instruction
watchdog timer is ignored.
Wf32
Wf512
Watchdog timer
frequency select bit
1
0
16
Stop mode
16
Watchdog timer
“FFF
16
Watchdog timer
interrupt request
❈
” is set.
16
Fig. 87 Block diagram of watchdog timer
84
Page 85
M37902FCCHP, M37902FGCHP, M37902FJCHP
How to disable watchdog timer
When not using the watchdog timer, it can be disabled. When the
watchdog timer is disabled, it’s operation stops and no watchdog
timer interrupt has been generated.
Setting for disabling the watchdog timer is possible by writing “7916”
and “5016” to the particular function select register 2 (address 6416)
sequentially with the following instructions:
• MOVMB/STAB instruction, or
• MOVM/STA instruction (m = 1)
If any method other than above has been adopted in order to access
(in other words, read/write) the particular function select register 2,
the watchdog timer will not be disabled until reset operation is performed. (Also, reset is the only one method to remove the setting for
disabling the watchdog timer.)
Moreover, this setting for disabling the watchdog timer is ignored at
return from the STP mode, and the watchdog timer operates. (For
details, refer to the section on the standby function.)
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
85
Page 86
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
INPUT/OUTPUT PINS
Ports P0 to P8, P10, P11 all have the direction register, and each bit
can be programmed for input or output. A pin becomes an output pin
when the corresponding bit of direction register is “1”, and an input
pin when it is “0”.
When a pin is programmed for output, the data is written to its port
latch and it is output to the output pin. When a pin is programmed for
output, the contents of the port latch is read instead of the value of
the pin. Accordingly, a previously output value can be read correctly
even when the output “H” voltage is lowered or the output “L” voltage
is raised owing to an external load, etc.
A pin programmed as an input pin is in the flooting state, and the
value input to the pin can be read. When a pin is programmed as an
input pin, the data is written only in the port latch and the pin remains
floating.
Each of Figures 89 and 90 shows the block diagram for each port pin
and pin NMI. Figure 91 shows the bit configuration of the port function control register.
Bit 3 of the port function control register serves as the port P0 input
level select bit, which selects the VIH/VIL level under the condition
that port P0 is used as an input port.
Bit 4 of the port function control register serves as the P44–P47
76543210
0
0
Port function control register
pullup connection select bit. This bit determines whether port pins
P44–P47,which are multiplexed with chip select pins, are to be
pulled up or not. At reset, this bit 4 = “0” and P4–P47 are pulled up.
The pullup function is valid only when the corresponding port is used
an input port.
Bit 7 of the port function control register serves as the NMI pullup
connection select bit. At reset, this bit 7 = “0” and pin NMI is pulled
up. The pullup function is valid only when the corresponding port is
used as an input port.
When using port pins P54–P57 as the key input interrupt input pins
(KI0 to KI3), the pullup function can be selected, also. For details,
refer to the section on interrupts.
When using a port pin as an internal peripheral device’s input pin,
clear the corresponding port direction register’s bit to “0”. When using a port pin as an internal peripheral device’s output pin, the port
direction register’s bit may be “0” or “1”.
In the memory expansion or microprocessor mode, port pins of P0 to
P4, P10, P11 become I/O pins, and the their functions as I/O port
pins are invalid. Note that, however, some port pins can function as
port pins by the special setting. For details, refer to the section on the
processor modes.
Address
92
16
At reset
00
16
Address/Port switch bits
0
to A23 (16 Mbytes)
000 : A
0
to A21, P06, P07 (4 Mbytes)
001 : A
0
to A19, P04 to P07 (1 Mbytes)
010 : A
0
to A17, P02 to P07 (256 Kbytes)
011 : A
0
to A15, P00 to P07 (64 Kbytes)
100 : A
101 : Do not select.
0
to A11, P00 to P07, P114 to P117 (4 Kbytes)
110 : A
0
to A7, P00 to P07, P110 to P117 (256 bytes)
111 : A
Port P0 input level select bit
0 : V
IH
= 0.7VCC, VIL = 0.2V
1 : VIH = 0.43VCC (Note 1), VIL = 0.16V
4
–P47 pullup connection select bit (Notes 2 and 3)
Pins P4
0 : Pins P4
1 : Pins P4
Fix these bits to “0”.
Pin NMI pullup connection select bit (Note 2)
0 : Pin NMI is pulled up.
1 : Pin NMI is not pulled up.
Notes 1: For the M37902FxM (power source voltage = 3.3 V±0.3 V), VIH = 0.5VCC.
2: When MD1 = V
not pulled up, regardless of these bits’ contents.
3: When MD1 = V
CC
and MD0 = VCC (flash memory parallel I/O mode), pins P44 to P47 and NMI are
SS
and MD0 = VCC (microprocessor mode), pin CS0 (P44) is not pulled up, regardless of the bit’s contents.
4
–P47 are pulled up.
4
–P47 are not pulled up.
CC
CC
Fig. 91 Bit configuration of port function control register
86
Page 87
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
[Inside dotted-line not included]
P0
0
to P07, P10 to P17, P20 to P27,
P3
1
to P33, P100 to P107,
P11
0
to P11
7
[Inside dotted-line included]
P3
0
/RDY, P43/HOLD,
P6
1
/TA4IN,
P6
2
/INT0, P63/INT1, P64/INT2,
P6
5
/TB0IN, P66/TB1IN, P67/TB2IN,
P8
2/RXD0
, P86/RXD
1
[Inside dotted-line not included]
P4
0
/ALE, P41/φ1, P42/HLDA,
P8
3/TXD0
, P87/TXD
1
[Inside dotted-line included]
P6
0
/TA4
OUT
[Shaded area included]
P4
4
/CS0, P45/CS1,
P4
6
/CS2, P47/CS
3
Data bus
Data bus
Direction register
Port latch
Direction register
Output(Internal peripheral devices)
Port latch
Pullup selection
1
Pullup
transistor
[Shaded area not included]
P5
1
/TA0IN/RTP01,
P5
3
/TA1IN/RTP0
3
[Shaded area included]
P5
5
/TA2IN/RTP11/KI1,
P5
7
/TA3IN/RTP13/KI
3
[Shaded area not included]
P5
0
/TA0
OUT
/RTP00,
P5
2
/TA1
OUT
/RTP0
2
[Shaded area included]
P5
4
/TA2
OUT
/RTP10/KI0,
P5
6
/TA3
OUT
/RTP12/KI
2
Data bus
Data bus
Direction register
Port latch
LatchT Q
Timer
underflow signal
Direction register
Output
Port latch
LatchT Q
Timer
underflow signal
Pullup selection
CK
(Internal peripheral devices)
CK
Pullup selection
1
Pullup
transistor
Pullup
transistor
Fig. 89 Block diagram for each port pin and pin NMI (1)
87
Page 88
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
[Inside dotted-line not included]
0
/AN0, P71/AN1,
P7
P7
2
/AN2, P73/AN
3
[Inside dotted-line included]
P7
4
/AN4/(INT3),
P7
5
/AN5/(INT4)
P81/CTS0/CLK0,
P8
4
/CTS1/RTS1/INT4,
5
/CTS1/CLK
P8
1
[Inside dotted-line not included]
P7
6
/AN6/DA
0
[Inside dotted-line included]
P7
7
/AN7/AD
TRG
/DA1/(INT2)
Data bus
Data bus
Data bus
Direction register
Port latch
Direction register
Output
(Internal peripheral devices)
Port latch
Direction register
Port latch
Analog input
1
0
P80/CTS0/RTS0/DA2/INT
NMI
Analog input
Analog output
3
Direction register
Output
(Internal peripheral devices)
Data bus
Port latch
Pullup selection
1
0
Analog output
Pullup
transistor
Enable D-A output
Enable D-A output
Fig. 90 Block diagram for each port pin and pin NMI (2)
88
Page 89
M37902FCCHP, M37902FGCHP, M37902FJCHP
RESET CIRCUIT
While the power source voltage satisfies the recommended operating condition, reset state is removed if pin RESET’s level returns
from the stabilized “L” level to the “H” level. As a result, program execution starts from the reset vector address. This reset vector address is expressed as shown below:
• A23 to A16 = 0016
• A15 to A8 = Contents at address FFFF16
• A7 to A0 = Contents at address FFFE16
Figures 92 and 93 show the microcomputer internal register’s status
at reset, and Figure 94 shows an operation example of the reset circuit. Apply “L” level voltage to pin RESET for a period (2 µs or more)
under the following conditions:
• Pin Vcc’s level satisfies the recommended operating condition.
• Oscillator’s operation has been stabilized.
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
VCC level
V
CC
0V
RESET
0V
X
IN
0V
Power on
Fig. 94 Operation example of reset circuit (Note that proper evalua-
tion is necessary in the system development stage.)
0.2V
CC
level
2 µs
Oscillation stabilized
CS
0
control register L
0
control register H
1
control register L
1
control register H
2
control register L
2
control register H
3
control register L
3
control register H
0
start address register
1
start address register
2
start address register
3
start address register
External interrupt input control register
Address
(
)
80
16
(
)
81
16
(
)
82
16
)
(
16
83
)
(
16
84
(
)
85
16
(
)
86
16
)
(
87
16
(
)
8A
16
(
8C
16
(
)
8E
16
(
)
90
16
(
)
92
16
(
)
94
16
(
)
96
16
(Note 2)
···
···CS
···CS
···CS
···CS
···CS
···CS
···CS
···Area CS
)
···Area CS
···Area CS
···Area CS
···Port function control register
0000
···
···D-A control register
(Note 3)
010
10
01001
(Note 3)
010010
000000
(Note 3)
010010
000000
(Note 3)
010010
0000
00010000
00000000
00000000
00000000
000
00000
0000
Processor status register PS
Program bank register PG
Program counter PC
Program counter PC
Direct page registers DPR0 to DPR3
Data bank register DT
Stack pointer
000
H
L
Address
(
)
98
···D-A register 0
16
(
)
99
···D-A register 1
16
)
(
···D-A register 2
9A
16
(
)
9E
···Flash memory control register
16
(
)
A0
···Real-time output control register
16
(
)
AC
···Serial I/O pin control register
16
)
(
···Clock control register
BC
16
000
Contents at address FFFF
Contents at address FFFE
0000
FFF
00
16
00
16
00
16
000001
0000
00000111
000??
00
16
16
00
16
16
000
1??
Notes 1: The contents of the other registers and RAM are undefined at reset and must be initialized by software.
2: While Vss level voltage is applied to pin MD0, this bit is “0”. While Vcc level voltage is applied to pin MD0, on the other hand, this bit is “1”.
3: While Vss level voltage is applied to pin BYTE, these bits are “0”. While Vcc level voltage is applied to pin BYTE, on the other hand, these bits are “1”.
16
16
Fig. 93 Microcomputer internal register’s status at reset (2)
89
Page 90
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Address
Port P0 direction register
Port P1 direction register
Port P2 direction register
Port P3 direction register
Port P4 direction register
Port P5 direction register
Port P6 direction register
Port P7 direction register
Port P8 direction register
Port P10 direction register
Port P10 direction register
A-D control register 0
A-D control register 1
UART 0 Transmit/Receive mode register
UART 1 Transmit/Receive mode register
UART 0 Transmit/Receive control register 0
UART 1 Transmit/Receive control register 0
UART 0 Transmit/Receive control register 1
UART 1 Transmit/Receive control register 1
Count start register
Notes 1: The contents of the other registers and RAM are undefined at reset and must be initialized by software.
2: While Vss level voltage is applied to pin MD0, these bits are “0”. While Vcc level voltage is applied to pin MD0, on the other hand, these bits are “1”.
3: At power-on reset, these bits are clear to “0”. At hardware or software reset, on the other hand, these bits retain the value just before reset.
Fig. 92 Microcomputer internal register’s status at reset (1)
90
Page 91
M37902FCCHP, M37902FGCHP, M37902FJCHP
X
IN
R
f
X
OUT
R
d
C
IN
C
OUT
M37902
OSCILLATION CIRCUIT
An oscillation circuit locates between pins XIN and XOUT, and Figure
95 shows a circuit example with a oscillator (an external ceramic
resonator or quartz crystal oscillator). The constants such as capacitance etc. depend on a oscillator. Therefore, for these constants,
adopt the oscillator manufacturer’s recommended values.
Figure 96 shows a circuit example with an external clock source.
When an external clock is input, be sure to leave pin XOUT open.
Also, in this case, when the external clock input select bit (bit 1 of the
particular function select register 0; See Figure 100.) is set to “1”, the
oscillation circuit stops it’s operation, and the current dissipation is
reduced. Moreover, this bit has another function, which selects the
return condition from the stop mode. For details, refer to the section
on the standby function.
On the other hand, the PLL (Phase Locked Loop) frequency multiplier (hereafter, referred as PLL circuit.) is included, also. This PLL
circuit uses an clock input from pin XIN and generates a multiplied
clock. When using the PLL circuit, be sure to connect pin VCONT with
an external filter circuit. (See Figure 97.) When not using the PLL circuit, be sure to leave pin VCONT open.
When not using the PLL circuit, be sure to clear the PLL circuit operation enable bit (bit 1 of the clock control register; See Figure 99.),
so that the PLL circuit will stop it’s operation.
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Fig. 95 Circuit example with external ceramic resonator or quartz crystal oscillator
M37902
X
IN
X
OUT
Left open.
External clock source
Vcc
Vss
Fig. 96 Circuit example with external clock source
M37902
VCONT
1 KΩ
220 pF
0.1 µF
Note: Make the wiring length as short as possible,
and shield it with the GND line which
Fig. 97 Circuit example with pin V
surrounds this circuit. Also, for the clock
supply to pin X
IN, see Figures 95 and 96.
and PLL circuit
CONT
91
Page 92
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
CLOCK GENERATING CIRCUIT
Figure 98 shows the block diagram of the clock generating circuit.
The clock generating circuit consists of the clock oscillation circuit,
PLL frequency multiplier (PLL circuit), system clock switch circuit,
peripheral devices’ clock switch circuit, clock divider , standby control
circuit, etc. As control registers for the clock generating circuit, also,
the clock control register (address BC16), particular function select
register 0 (address 6216) are provided. (See Figures 99 and 100.)
As shown in Figure 98, clocks used in the CPU, BIU, peripheral devices, watchdog timer (in other words, clocks
Wf32, Wf512) are made from system clock fsys. System clock fsys can
be selected between fXIN (in other words, a clock input from pin XIN)
and fPLL (in other words, an output clock generated by the PLL circuit). By setting the clock
φ
1 output select bit (bit 7 of the processor
mode register 0) to “1”, also, system clock fsys can be output from
port pin P41, as clock
φ
1.
The PLL circuit’s operation, system clock (fsys) selection, and divide
ratio selection for peripheral devices’ clocks (f1 to f4096) are controlled by the clock control register. The following describes about
these control.
Bit 1 of the clock control register (the PLL circuit operation enable bit)
selects the PLL circuit’s operation (stopped/active). When this bit is
set to “1”, pin VCONT will becomes valid, and the PLL circuit will operate. At reset, the PLL circuit operation enable bit becomes “1”. (In this
case, the PLL circuit operates.) When not using the PLL circuit, be
sure to clear the PLL circuit operation enable bit to “0” (stopped). At
the STP instruction execution or while the flash memory parallel I/O
mode is set, the PLL circuit stops its operation, and pin VCONT is in-
φ
CPU, φBIU, f1 to f4096,
valid, regardless of this bit 1’s status.
Bits 2 and 3 of the clock control register (the PLL multiplication ratio
select bits) select the ratio of fPLL/fXIN. The PLL multiplication ratio
must be set so that the frequency of the PLL output clock (fPLL) must
be in the range from 10 MHz to 26 MHz. At reset, the PLL multiplication ratio select bits become “0,1” (✕ 2). The change of the multiplication ratio must be performed while input clock fXIN is set as system
clock. (In this case, bit 5 of the clock control register = “0”.) After that,
be sure to wait that the operation-stabilizing time of the PLL circuit
has passed, and switch the system clock to the PLL output clock
(fPLL). (In other words, set bit 5 to “1”.) Note that, after reset, the PLL
multiplication ratio select bits are allowed to be changed only once.
Bit 5 of the clock control register is the system clock select bit, and
fXIN is selected as the system clock when bit 5 = “0”. On the other
hand, when bit 5 = “1”, the PLL output clock (fPLL) is selected. At reset, the system clock select bit becomes “0”. When selecting fPLL, be
sure that the PLL circuit’s operation has been stabilized properly , and
then, set the system clock select bit to “1”. Also, when the PLL circuit
operation enable bit is cleared to “0” (the PLL circuit is stopped.), the
system clock select bit will automatically be cleared to “0”. Note that
a value of “1” cannot be written to the system clock select bit while
the PLL circuit operation enable bit =“0”.
Table 15 lists the fsys selection.
Bits 6 and 7 of the clock control register are the peripheral devices’
clock select bits 0, 1, and these bits select the multiplication ratio of
(f1 to f4096)/(fsys).
Table 16 lists the internal peripheral devices’ operation clock frequency. At reset, these bits become “0, 0”.
Table 15. f
System clock select bit
sys
(Bit 5)
selection
PLL circuit operation enable bit
(Bit 1)
PLL multiplication ratio select bits
(Bits 3, 2) (Note)
0
01 (✕ 2)
1
1
10 (✕ 3)
11 (✕ 4)
Note: The PLL multiplication ratio must be set so that the frequency of the PLL output clock (f
IN
) means the frequency of the input clock from pin XIN (fXIN). After reset, the PLL multiplication ratio select bits are allowed to be
f(X
changed only once.
Table 16. Internal peripheral devices’ operation clock frequency
Internal peripheral devices’
operation clock
1
f
f
2
f
16
f
64
f
512
f
4096
Note: When selecting the peripheral devices’ clock select bits 1, 0 = “01
• Watchdog timer frequency select bit: bit 0 at address 61
16
•
Watchdog timer clock source select bit at stop state termination :
bits 6, 7 at address 61
16
• External clock input select bit: bit 1 at address 62
16
• System clock stop select bit at WIT: bit 3 at address 63
16
• PLL circuit operation enable bit: bit 1 at address BC
16
• PLL multiplication ratio select bits: bits 2, 3 at address BC
16
• System clock select bit: bit 5 at address BC
16
• Peripheral device
’s clock select bit 0, 1
: bits 6, 7 at address BC
16
1/8
1/2
1/16
Watchdog
timer
Wf
32
Wf
512
f
16
f
1
Peripheral device
’s clocks
0
1
Watchdog timer
frequency select bit
X
IN
X
OUT
System clock stop select bi at WIT
1/16
Access to
external area
HLDA
0
1
Watchdog timer clock source select
bit at stop state termination
φ
1
Wait mode
1
0
1
0
1/2
1
0
1
Wait mode
System clock
frequency select bit
PLL frequency
multiplier
f
PLL
V
CONT
Wait mode
External clock
input select bit
Q
R
S
STP
instruction
Interrupt
request
Q
R
S
WIT
instruction
Interrupt
request
Wait mode
PLL circuit operation enable bit
PLL multiplication ratio select bits
fX
IN
f/n
0
fX
16
fX
32
fX
64
fX
128
fX
16
fX
32
fX
64
fX
128
Peripheral
device’s clock
select bit 0
Peripheral
device’s clock
select bit 1
BIU : Bus interface Unit
CPU : Central Processing Unit
❈ : Signal generated when the watchdog timer
’s most significant bit becomes
“0”.
f
sys
System clock frequency select bit
❈
Operating clock for
serial I/O, timer B
A-D conversion frequency
(φ
AD
) clock source
Operating clock for timer A
External clock input select bit
Interrupt
request
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
Fig. 98 Block diagram of clock generating circuit
93
Page 94
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
76543210
10
Notes 1: When not using the PLL frequency multiplier, clear this bit to “0”. In the stop mode or in
the flash memory parallel I/O mode, the PLL circuit stops it’s operation regardless of this
bit’s contents; at this time, pin V
2: When rewriting this bit, be sure to clear bit 5 to “0” simultaneously. Also, after this bit is
rewritten, insert a waiting time of 2 ms, and then set bit 5 to “1”.
3: When the PLL circuit operation enable bit (bit 1) has been cleared to “0”, this bit will also
be cleared to “0”. Also, bit 1 = “0”, nothing can be written to this bit. (Fixed to be “0”.)
Clock control register
Fix this bit to “1”.
PLL circuit operation enable bit (Note 1)
0: PLL frequency multiplier is stopped, and pin V
1: PLL frequency multiplier is operating, and pin V
PLL multiplication ratio select bits (Note 2)
00: Do not select.
01: Double
10: Triple
11: Quadruple
Fix this bit to “0”.
System clock select bit (Note 3)
0: fX
IN
1: f
PLL
Peripheral device’s clock select bits 1, 0
See Table 16.
CONT
is invalid.
Address
BC
At reset
16
16
07
CONT
is invalid (floating state).
CONT
is valid.
Fig. 99 Bit configuration of clock control register
76543210
00
Note:
0
Writing to these bits requires the following procedure:
16
• Write “55
• Succeedingly, write “0” or “1” to each bit.
Also, use the MOVM (MOVMB) instruction or STA (STAB, STAD) instruction
” to this register. (The bit status does not change only by this writing.)
Particular function select register 0
STP instruction invalidity select bit (Note)
0: STP instruction is valid.
1: STP instruction is invalid.
External clock input select bit (Note)
0: Oscillation circuit is active. (Oscillator is connected.)
Watchdog timer is used at stop mode termination.
1: Oscillation circuit is inactive. (External clock is input.)
When the system clock select bit = “0”,
watchdog timer is not used at stop mode termination.
When the system clock select bit = “1”,
watchdog timer is used at stop mode termination.
Fix this bit to “0”.
Fig. 100 Bit configuration of particular function select register 0
Address
62
16
94
Page 95
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
STANDBY FUNCTION
The standby function provides the stop (hereafter called STP) and
the wait (hereafter called WIT) mode. These modes are used to save
the power dissipation of the system, by stopping oscillation or system clock in the case that the CPU needs not be operating.
The microcomputer enters the STP or WIT mode by executing the
STP or WIT instruction, and either mode is terminated by acceptance
of an interrupt request or reset.
To terminate the STP or WIT mode by an interrupt request, the interrupt to be used for termination of the STP or WIT mode must be enabled in advance to execution of the STP or WIT instruction. The
interrupt priority level of this interrupt is required to be higher than the
processor interrupt priority level (IPL) of the routine where the STP
or WIT instruction will be executed.
Figures 100 to 102 show the bit configurations of the particular function select registers 0, 1, and watchdog timer frequency select register respectively. Setting the STP instruction invalidity select bit (bit 0
of the particular function select register 0) to “1” invalidates the STP
instruction, and the STP instruction will be ignored. Since the above
bit is cleared to “0” after reset is removed, however, the STP instruction is valid.
The STP- or the WIT-instruction-execution status bit (bit 0 or 1 of the
particular function select register 1) is set to “1” by the execution of
the STP or the WIT instruction, and so, after the STP or WIT mode
has been terminated, each bit will indicate that the STP or WIT instruction has been executed. Accordingly, each of these bits must be
cleared to “0” by software at termination of the STP or the WIT mode.
T able 17 explains the microcomputer’s operation in the STP and WIT
modes.
The external bus fixation function can also be provided. This function
enables the user to specify the states of the external bus and the bus
control signals in the memory expansion and the microprocessor
mode in the STP or WIT mode. For more information, refer to the
section on the power saving function.
STP mode
The execution of the STP instruction stops the oscillation circuit and
PLL circuit. It also stops input clock fXIN, system clock fsys,
φ
CPU, and peripheral devices’ clocks f1 to f4096, Wf32 and Wf512 in
the “L” state, and divide clocks fX16 to fX128 in the “H” state. In the
watchdog timer, “FFF16” is automatically set. As shown in Figure 98,
any one of divide clocks fX16 to fX128, which is selected by the
watchdog timer clock source select bits at STP termination (bits 6
and 7 of the watchdog timer frequency select register), becomes the
watchdog timer’s clock source.
In the STP mode, the A-D converter and watchdog timer, which uses
peripheral devices’ clocks f1 to f4096, Wf32 and Wf512, are stopped.
At this time, timers A and B operate only in the event counter mode,
and serial I/O communication is active while an external clock is selected.
The STP mode is terminated by acceptance of an interrupt request
or reset, and the oscillation circuit and PLL circuit restart their operations. Input clock fXIN, system clock fsys, and peripheral devices’
clocks f1 to f4096, Wf32 and Wf512 are also supplied.
When the STP mode is terminated by reset, supply of
φ
starts immediately after the oscillation circuit and PLL circuit restart
their operations. Therefore, the reset input must be raised “H” after
the operation-stabilizing time for these circuits has passed.
The following two modes are available in order to terminate the STP
mode by an interrupt:
(1) The watchdog timer is used in order to measure the period from
the operation restart of the oscillation circuit and PLL circuit until
the supply start of
(2) The supply of
φ
BIU and φCPU.
φ
BIU and φCPU is started immediately after the op-
eration restart of the oscillation circuit and PLL circuit.
When the external clock input select bit (bit 1 of the particular function select register 0) = “0” or the system clock select bit (bit 5 of the
clock control register) = “1”, the watchdog timer will start counting
φ
BIU,
BIU and φCPU
Table 17. Microcomputer’s operation in STP and WIT modes
Mode
System clock
stop select bit
at WIT
Oscillation
circuit
PLL circuit
Operations of function while WIT, STP modes
f
sys
, φ1,
1
to f
4096
f
Wf
32
, Wf
512
φ
BIU
, φ
CPU
Peripheral devices using f
Timers A, B: Operation is enabled only in the event
counter mode.
STP
—
Stopped
Stopped
Stopped
(“L”)
Stopped
(“L”)
Stopped
(“L”)
Serial I/O: Operation is enabled only while an external
clock is selected.
A-D converter: Stopped.
(Watchdog timer: Stopped.)
“0”
Active
(Note 1)
Active
(Note 2)
Active
Stopped
(“L”)
Stopped
(“L”)
Timers A, B, Serial I/O, A-D converter: Operation is enabled.
(Watchdog timer: Stopped.)
Timers A, B: Operation is enabled only in the event
WIT
“1”
Active
(Note 1)
Active
(Note 2)
Stopped
(“L”)
Stopped
(“L”)
Stopped
(“L”)
Serial I/O: Operation is enabled only while an external
counter mode.
clock is selected.
A-D converter: Stopped.
(Watchdog timer: Stopped.)
Notes 1: When the external clock input select bit = “1”, the oscillation circuit stops. Also, clock input from pin X
2: When the PLL operation enable bit = “0”, the PLL circuit stops.
IN
is available.
1
to f
4096
, Wf32, Wf
512
95
Page 96
M37902FCCHP, M37902FGCHP, M37902FJCHP
down with one of the above divide clocks, fX16 to fX128, after the oscillation circuit and PLL circuit have been restarted their operations
owing to an interrupt. The most significant bit of the watchdog timer
reaching “0”, supply of
On the other hand, when the external clock input select bit = “1 ” and
the system clock select bit = “0”, supply of
immediately after the oscillation circuit has been restarted their operations owing to an interrupt. (In actual fact, after the selected one
of the above divide clocks, fX16 to fX128, has been changed from “H”
to “L”, this supply will restart.)
φ
BIU and φCPU restarts.
φ
BIU and φCPU will restart
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
76543210
Particular function select register 1
STP-instruction-execution status bit (Note 1)
0: Normal operation.
1: STP instruction has been executed.
WIT-instruction-execution status bit (Note 1)
0: Normal operation.
1: WIT instruction has been executed.
Standby state select bit
0: External bus
1: Programmable I/O port
System clock stop select bit at WIT (Note 2)
0: In wait mode, system clock f
1: In wait mode, system clock f
Address output select bit
0: Address changes depending on bus access.
1: Address changes only at access to external address.
Timer B2 clock source select bit
In event counter mode:
0: Clock input from pin TB2
32 (f(XIN)/32) is counted.
1: fX
Notes 1: At power-on reset, this bit becomes “0”. At hardware reset or software reset, this bit
retains the value just before reset. Even when “1” is written, the bit status will not change.
2: Setting this bit to “1” must be performed just before execution of the WIT instruction.
Also, after the wait state is terminated, this bit must be cleared to “0” immediately.
Address
63
sys is active.
sys is stopped.
IN is counted.
16
Fig. 101 Bit configuration of particular function select register 1
76543210
76543210
Watchdog timer frequency select register
Watchdog timer frequency select bit
Watchdog timer clock source select bits at STP termination
Fig. 102 Bit configuration of watchdog timer frequency select register
96
0 : Select Wf
1 : Select Wf32
32
0 0 : fX
0 1 : fX16
1 0 : fX128
1 1 : fX64
512
Address
61
16
Page 97
M37902FCCHP, M37902FGCHP, M37902FJCHP
WIT mode
When the WIT instruction is executed with the system clock stop select bit at WIT (bit 3 of the particular function select register 1 in Figure 101) being “0”,
stopped in the “L“ state. However, the oscillation circuit, PLL circuit,
input clock fXIN, system clock fsys,
f1 to f4096 remain operating. Therefore, BIU and CPU are stopped,
whereas timers A and B, serial I/O, and the A-D converter, which use
the peripheral devices’ clocks f1 to f4096, are still operating. Note that
the watchdog timer is stopped.
On the other hand, when the WIT instruction is executed with the
system clock stop select bit at WIT being “1”, the oscillation circuit,
PLL circuit, and input clock fXIN are operating, while system clock
fsys,
φ
BIU, φCPU, and peripheral devices’ clocks stop operating. As a
result, the A-D converter and watchdog timer, which use peripheral
devices’ clocks f1 to f4096, Wf32 and Wf512, are stopped. At this time,
timers A and B operate only in the event counter mode, and serial
I/O communication is active only while an external clock is selected.
If the internal peripheral devices are not used in the WIT mode, the
latter is better because the current dissipation is more saved. Note
that the system clock stop select bit at WIT is to be set to “1” immediately before execution of the WIT instruction and cleared to “0” immediately after the WIT mode is terminated.
The WIT state is terminated by acceptance of an interrupt request,
and then, supply of
circuit, PLL circuit, and clock input fXIN are operating in the WIT
mode, an interrupt processing can be executed just after the WIT
mode termination.
φ
BIU, φCPU, and divide clocks Wf32 and Wf512 are
φ
1, and peripheral devices’ clock
φ
BIU and φCPU will restart. Since the oscillation
MITSUBISHI MICROCOMPUTERS
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
97
Page 98
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
POWER SAVING FUNCTION
The following functions can save the power dissipation of the whole
system.
(1) External bus fixation in standby state
By setting the standby state select bit (bit 2 of the particular function
select register 1) to “1”, in the stop or wait mode, the I/O pins of the
external buses and bus control signals can be switched to programmable I/O port pins. By setting these pins’ state with the corresponding port registers and port direction registers, unnecessary current
will not flow between the microcomputer and external devices. As a
result, in the stop or wait mode, the power dissipation of the whole
system can be lowered. Table 18 lists the correspondence between
the external buses, bus control signals, and programmable I/O port
pins.
This function is valid only in the stop or wait mode. At termination of
the stop or wait mode, the original function of external buses and bus
control signals become valid.
Table 18. Correspondence between external buses, bus control sig-
nals, and programmable I/O port pins
External buses,
Bus control signals
A0 to A7,
8
to A15,
A
16
to A
0
to D7,
8
to D
23
15
A
D
D
RD, BLW,
BHW
CS
0
Note: When the external data bus width = 8 bits (BYTE = VCC level),
this becomes a programmable I/O port pin, regardless of the
standby state select bit’s contents.
Standby state select bit
0
0
to A7,
A
8
to A15,
A
A
16
to A
23
D0 to D7,
8
to D15 (Note)
D
RD, BLW,
BHW (Note)
CS
0
1
P100 to P107,
0
to P117,
P11
0
to P0
P0
7
P10 to P17,
0
to P2
P2
7
P31, P32, P3
P9
0
3
mode, owing to an interrupt request occurrence. Therefore, an instruction can be executed just after the termination of the stop mode.
For details, refer to the section on the clock generating circuit and
standby function.
(4) Disconnection from pin VREF
When not using the A-D converter and D-A converter, by setting the
VREF connection select bit (bit 6 of the A-D control register 1) to “1”,
the resistor ladder network of the A-D converter will be disconnected
from the reference voltage input pin (VREF). In this case, no current
flows from pin VREF to the resistor ladder network, and the power dissipation can be saved. Note that, after the VREF connection select bit
changes from “1” (VREF disconnected) to “0” (VREF connected), be
sure that a waiting time of 1 µs of more has passed before the A-D
conversion starts. For details, refer to the sections on the A-D converter and D-A converter.
(5) Address output selection
In the memory expansion mode or microprocessor mode, when the
address output select bit (bit 4 of the particular function select register 1) becomes “1”, the unnecessary change of address pins’ state
will be avoided, without output of an address at access to the internal area.
For details, refer to the section on the BIU.
(2) Stop of system clock in wait mode
In the wait mode, if the internal peripheral devices need not to operate, the system clock stop select bit at WIT (bit 3 of the particular
function select register 1) = “1”, both of system clock fsys and peripheral devices’ clock stop their operations, and the power dissipation
can be saved.
For details, refer to the section on the standby function.
(3) Stop of oscillation circuit
When an externally-generated-stable clock is input to pin XIN, the
power dissipation can be saved if both of the following conditions are
met:
• the external clock input select bit (bit 1 of the particular function
select register 0) = “1”.
• the oscillation driver between pins XIN and XOUT stops its operation.
At this time, the output level at pin XOUT is fixed to “H”. When not using a PLL output clock, also, the supply of
their operations just after the microcomputers returns from the stop
98
φ
BIU and φCPU restarts
Page 99
MITSUBISHI MICROCOMPUTERS
M37902FCCHP, M37902FGCHP, M37902FJCHP
SINGLE-CHIP 16-BIT CMOS MICROCOMPUTER
DEBUG FUNCTION
When the CPU fetches an instruction code, an interrupt request will
be generated if a selected condition is satisfied, as a resultant of
comparison between a specified address and the start address
where the instruction code is stored (the contents of PG and PC).
The decision whether this condition is satisfied or not is called address matching detection, and the interrupt generated by this detection is called an address matching detection interrupt. (For interrupt
vector addresses, refer to the section on interrupts.)
In the address matching detection, a non-maskable interrupt routine
is proceeded without execution of the original instruction which has
been allocated to the target address.
The debug function provides the following two modes:
• the address matching detection mode, which is used to avoid the
area where program exists or modify a program.
• the out-of-address-area detection mode, which is used to detect a
program runaway.
Figures 103 shows the block diagram of the debug function. Figures
104 and 105 show the bit configurations of the debug control registers 0, 1, and address compare registers 0,1, respectively.
The detect condition select bits of the debug control register 0 can
select one condition between the following 4 conditions. When the
selected address condition is satisfied, an address matching detection interrupt request will be generated:
(1) Address matching detection 0
The contents of PG and PC match with the address which has
been set in the address compare register 0.
(2) Address matching detection 1
The contents of PG and PC match with the address which has
been set in the address compare register 1.
(3) Address matching detection 2
The contents of PG and PC match with the address which has
been set in either of the address compare register 0 or address
The contents of PG and PC are less than the address which has
been set in the address compare register 0 or larger than the address which has been set in the address compare register 1.
By setting the detect enable bit of the debug control register 0 to “1”,
an address matching detection interrupt request will be generated if
any one of the above address conditions is satisfied. Clearing the
detect enable bit to “0” generates no interrupt request even if any of
the above address conditions is satisfied.
The address compare register access enable bit of the debug control register 1 must be set to “1” by the instruction just before the access operation (read/write). Then, this bit must be cleared to “0”
(disabled) by the next instruction. While this bit = “0”, the address
compare registers 0, 1 cannot be accessed.
The address-matching-detection 2 decision bit of the debug control
register 1 decides, whether the address which has been set in the
address compare register 0 or 1 matches with the contents of PG,
PC, when the address matching detection 2 is selected. The contents of this bit is invalid when address matching detection 0 or 1 is
selected.
In order to use the debug function to avoid the area where program
exists or modify a program, perform the necessary processing within
an address matching interrupt routine. As a result, the contents of
PG, PC, PS at acceptance of an address matching detection interrupt request (i.e. the address at which an address matching detection condition is satisfied) have been pushed on to the stack. If a
return destination address after the interrupt processing is to be altered, rewrite the contents of the stack, and then return by the RTI
instruction.
To use the debug function to detect a program runaway, set an address area where no program exists into the address compare registers 0 and 1 by using the out-of-address-area detection. When the
CPU fetches instruction codes from this address area and executes
them, an address matching detection interrupt request will be generated.
The above debug function cannot be evaluated by a debugger, so
that the debug function must not be used while a debugger is running.
000: Do not select.
001: Address matching detection 0
010: Address matching detection 1
011: Address matching detection 2
100: Do not select.
101: Out-of-address-area detection
110: Do not select.
111: Do not select
Fix this bit to “0” (Note 1).
Detect enable bit (Note 1)
0: Detection disabled.
1: Detection enabled.
Fix this bit to “0”(Note 1).
“1” at read.
Debug control register 1
Fix this bit to “0”(Note 1).
“0” at read (Note 1).
Address compare register access enable bit (Note 2)
0: Disabled
1: Enabled
Fix this bit to “1” when using the debug function.
Fix this bit to “0”(Note 1).
While debugger is not used, “0” at read.
While debugger is used, “1” at read.
Address-matching-detection 2 decision bit
❈ Valid when address matching detection 2 is selected.
0: Matches with the contents of the address compare register 0.
1: Matches with the contents of the address compare register 1.
“0” at read.
Notes 1: At power-on reset, these bits = “0”; at hardware reset or software reset, these bits retain
the value just before reset.
2: Set this bit to “1” with the instruction just before the address compare register 0, 1
(addresses 68
just after the access.
16 to 6D16) is accessed. And then, clear this bit to “0” with the instruction
Fig. 104 Bit configuration of debug control register 0, 1
(23)
7
(16)
(15)
7
0
The address to be detected (in other words, the start address of instruction) is set here.
(8)
0
Fig. 105 Bit configuration of address compare register 0, 1