• AVR – High-performance and Low-power RISC Architecture
– 121 Powerful Instructions – Most Single Clock Cycle Execution
– 32 x 8 General-purpose Working Registers + Peripheral Control Registers
– Up to 6 MIPS Throughput at 6 MHz
Endurance: 100,000 Write/Erase Cycles
– Programming Lock for Flash Program and EEPROM Data Security
– SPI Interface for In-System Programming
• Peripheral Features
– On-chip Analog Comparator
– Programmable Watchdog Timer with On-chip Oscillator
– Programmable Serial UART
– Master/Slave SPI Serial Interface
– Real-time Counter (RTC) with Separate Oscillator
– Two 8-bit Timer/Counters with Separate Prescaler and PWM
– Expanded 16-bit Timer/Counter System with Separate Prescaler, Compare,
Capture Modes and Dual 8-, 9-, or 10-bit PWM
– Programmable Watchdog Timer with On-chip Oscillator
– 8-channel, 10-bit ADC
• Special Microcontroller Features
– Low-power Idle, Power-save and Power-down Modes
– Software Selectable Clock Frequency
– External and Internal Interrupt Sources
• Specifications
– Low-power, High-speed CMOS Process Technology
– Fully Static Operation
• Power Consumption at 4 MHz, 3V, 25°C
– Active: 5.5 mA
– Idle Mode: 1.6 mA
– Power-down Mode: < 1 µA
– 2.7 - 3.6V for ATmega103L
– 4.0 - 5.5V for ATmega103
• Speed Grades
– 0 - 4 MHz for ATmega103L
– 0 - 6 MHz for ATmega103
®
RISC Architecture
8-bit
Microcontroller
with 128K Bytes
In-System
Programmable
Flash
ATmega103(L)
Rev. 0945G–09/01
1
Pin ConfigurationTQFP
PC6(A14)
PC5(A13)
PC7(A15)
PA7(AD7)
ALE
PA6(AD6)
PA5(AD5)
PA4(AD4)
(AD2)PA2
(AD1)PA1
(AD0)PA0
VCC
GND
(ADC7)PF7
(ADC6)PF6
(ADC5)PF5
(ADC4)PF4
(ADC3)PF3
(ADC2)PF2
(ADC1)PF1
(ADC0)PF0
AREF
AGND
AVCC
PA3(AD3)
52
53
54
55
56
57
58
59
60
612361
622262
63
63
6464
INDEXCORNER
1512503494485476467458449
PEN
(PDI/RXD)PE0
(PDO/TXD)PE1
431042114112401339143815371636173518341933
(AC-)PE3
(AC+)PE2
(INT4)PE4
(INT5)PE5
(INT6)PE6
(INT7)PE7
PC4(A12)
(SS)PB0
PC3(A11)
(SCK)PB1
PC2(A10)
(MOSI)PB2
PC1(A9)
(MISO)PB3
PC0(A8)
(OC0/PWM0)PB4
(OC1A/PWM1A)PB5
RD
WR
32
31
30
29
28
27
26
25
24
21
20
(OC1B/PWM1B)PB6
PD7(T2)
PD6(T1)
PD5
PD4(IC1)
PD3(INT3)
PD2(INT2)
PD1(INT1)
PD0(INT0)
XTAL1
XTAL2
GND
VCC
RESET
TOSC1
TOSC2
PB7(OC2/PWM2)
2
ATmega103(L)
0945G–09/01
ATmega103(L)
DescriptionThe ATmega103(L) is a low-power, CMOS, 8-bit microcontroller based on the AVR
RISC architecture. By executing powerful instructions in a single clock cycle, the
ATmega103(L) achieves throughputs approaching 1 MIPS per MHz, allowing the system designer to optimize power consumption versus processing speed.
The AVR core is based on an enhanced RISC architecture that combines a rich instruction set with 32 general-purpose working registers. All the 32 registers are directly
connected to the Arithmetic Logic Unit (ALU), allowing two independent registers to be
accessed in one single instruction executed in one clock cycle. The resulting architecture is more code efficient while achieving throughputs up to ten times faster than
conventional CISC microcontrollers.
The ATmega103(L) provides the following features: 128K bytes of In-System Programmable Flash, 4K bytes EEPROM, 4K bytes SRAM, 32 general-purpose I/O lines, 8 input
lines, 8 output lines, 32 general-purpose working registers, real-time counter (RTC), 4
flexible timer/counters with compare modes and PWM, UART, programmable watchdog
timer with internal oscillator, an SPI serial port and 3 software-selectable power saving
modes. The Idle mode stops the CPU while allowing the SRAM, timer/counters, SPI port
and interrupt system to continue functioning. The Power-down mode saves the register
contents but freezes the oscillator, disabling all other chip functions until the next interrupt or hardware reset. In Power-save mode, the timer oscillator continues to run,
allowing the user to maintain a timer base while the rest of the device is sleeping.
The device is manufactured using Atmel’s high-density nonvolatile memory technology.
The On-chip ISP Flash allows the program memory to be reprogrammed in-system
through a serial interface or by a conventional nonvolatile memory programmer. By
combining an 8-bit RISC CPU with a large array of ISP Flash on a monolithic chip, the
Atmel ATmega103(L) is a powerful microcontroller that provides a highly flexible and
cost-effective solution to many embedded control applications.
The ATmega103(L) AVR is supported with a full suite of program and system development tools including: C compilers, macro assemblers, program debugger/simulators, incircuit emulators and evaluation kits.
0945G–09/01
3
Block DiagramFigure 1. The ATmega103(L) Block Diagram
PA0 - PA7PF0 - PF7
VCC
GND
AVCC
AGND
AREF
PORTF BUFFERS
ANALOG MUXADC
PROGRAM
COUNTER
PROGRAM
FLASH
INSTRUCTION
REGISTER
INSTRUCTION
DECODER
PORTA DRIVER/BUFFERS
DATAREGISTER
PORTA
STACK
POINTER
SRAM
GENERAL
PURPOSE
REGISTERS
X
Y
Z
DATADIR.
REG. PORTA
INTERNAL
OSCILLATOR
WATCHDOG
MCU CONTROL
REGISTER
COUNTERS
INTERRUPT
8-BIT DATA BUS
TIMER
TIMER/
UNIT
PC0 - PC7
PORTC DRIVERS
DATAREGISTER
PORTC
OSCILLATOR
OSCILLATOR
TIMING AND
CONTROL
XTAL1
XTAL1
TOSC2
TOSC1
RESET
ALE
WR
RD
ANALOG
COMPARATOR
DATAREGISTER
+
-
PORTE
CONTROL
LINES
DATADIR.
REG. PORTE
ALU
STATUS
REGISTER
SPI
DATAREGISTER
PORTB
PORTB DRIVER/BUFFERSPORTE DRIVER/BUFFERS
EEPROM
PROGRAMMING
LOGIC
UART
DATADIR.
REG. PORTB
PB0 - PB7PE0 - PE7
DATAREGISTER
PORTD
PORTD DRIVER/BUFFERS
PD0 - PD7
DATADIR.
REG. PORTD
PEN
VCC
GND
4
ATmega103(L)
0945G–09/01
ATmega103(L)
Pin Descriptions
VCCSupply voltage.
GNDGround.
Port A (PA7..PA0)Port A is an 8-bit bi-directional I/O port. Port pins can provide internal pull-up resistors
(selected for each bit). The Port A output buffers can sink 20 mA and can drive LED displays directly. When pins PA0 to PA7 are used as inputs and are externally pulled low,
they will source current if the internal pull-up resistors are activated.
Port A serves as Multiplexed Address/Data bus when using external SRAM.
The Port A pins are tri-stated when a reset condition becomes active, even if the clock is
not running.
Port B (PB7..PB0)Port B is an 8-bit bi-directional I/O port with internal pull-up resistors. The Port B output
buffers can sink 20 mA. As inputs, Port B pins that are externally pulled low, will source
current if the pull-up resistors are activated.
Port B also serves the functions of various special features.
The Port B pins are tri-stated when a reset condition becomes active, even if the clock is
not running.
Port C (PC7..PC0)Port C is an 8-bit output port. The Port C output buffers can sink 20 mA.
Port C also serves as Address output when using external SRAM.
Since Port C is an output only port, the Port C pins are not tri-stated when a reset condition becomes active.
Port D (PD7..PD0)Port D is an 8-bit bi-directional I/O port with internal pull-up resistors. The Port D output
buffers can sink 20 mA. As inputs, Port D pins that are externally pulled low will source
current if the pull-up resistors are activated.
Port D also serves the functions of various special features.
The Port D pins are tri-stated when a reset condition becomes active, even if the clock is
not running.
Port E (PE7..PE0)Port E is an 8-bit bi-directional I/O port with internal pull-up resistors. The Port E output
buffers can sink 20 mA. As inputs, Port E pins that are externally pulled low will source
current if the pull-up resistors are activated.
Port E also serves the functions of various special features.
The Port E pins are tri-stated when a reset condition becomes active, even if the clock is
not running
Port F (PF7..PF0)Port F is an 8-bit input port. Port F also serves as the analog inputs for the ADC.
RESET
Reset input. An external reset is generated by a low level on the RESET pin. Reset
pulses longer than 50 ns will generate a reset, even if the clock is not running. Shorter
pulses are not guaranteed to generate a reset.
XTAL1Input to the inverting oscillator amplifier and input to the internal clock operating circuit.
XTAL2Output from the inverting oscillator amplifier.
0945G–09/01
5
TOSC1Input to the inverting Timer/Counter oscillator amplifier.
TOSC2Output from the inverting Timer/Counter oscillator amplifier.
WR
RD
External SRAM write strobe
External SRAM read strobe
ALEALE is the Address Latch Enable used when the External Memory is enabled. The ALE
strobe is used to latch the low-order address (8 bits) into an address latch during the first
access cycle, and the AD0-7 pins are used for data during the second access cycle.
AVCCSupply voltage for Port F, including ADC. The pin must be connected to VCC when not
used for the ADC. See “ADC Noise Canceling Techniques” on page 77 for details when
using the ADC.
AREFAREF is the analog reference input for the ADC converter. For ADC operations, a volt-
age in the range AGND to AVCC must be applied to this pin.
AGNDIf the board has a separate analog ground plane, this pin should be connected to this
ground plane. Otherwise, connect to GND.
PEN
PEN is a programming enable pin for the serial programming mode. By holding this pin
low during a power-on reset, the device will enter the serial programming mode. PEN
has no function during normal operation.
Clock Options
Crystal OscillatorXTAL1 and XTAL2 are input and output, respectively, of an inverting amplifier, which
can be configured for use as an on-chip oscillator, as shown in Figure 2. Either a quartz
crystal or a ceramic resonator may be used.
Figure 2. Oscillator Connections
MAX 1 HC BUFFER
HC
C2
C1
Note:When using the MCU oscillator as a clock for an external device, an HC buffer should be
connected as indicated in the figure.
XTAL2
XTAL1
GND
6
ATmega103(L)
0945G–09/01
ATmega103(L)
External ClockTo drive the device from an external clock source, XTAL2 should be left unconnected
while XTAL1 is driven as shown in Figure 3.
Figure 3. External Clock Drive Configuration
NC
EXTERNAL
OSCILLATOR
SIGNAL
XTAL2
XTAL1
GND
Timer OscillatorFor the Timer Oscillator pins, TOSC1 and TOSC2, the crystal is connected directly
between the pins. No external capacitors are needed. The oscillator is optimized for use
with a 32,768 Hz watch crystal. Applying an external clock source to TOSC1 is not
recommended.
0945G–09/01
7
Architectural
Overview
Figure 4. The ATmega103(L) AVR RISC Architecture
AVR ATmega103(L) Architecture
Data Bus 8-bit
64K x 16
Program
Memory
Instruction
Register
Instruction
Decoder
Control Lines
Program
Counter
and Test
Purpose
Registers
Direct Addressing
Indirect Addressing
Status
32 x 8
General
Peripherals
ALU
4K x 8
Data
SRAM
4K x 8
EEPROM
The AVR uses a Harvard architecture concept – with separate memories and buses for
program and data. The program memory is accessed with a single-level pipeline. While
one instruction is being executed, the next instruction is pre-fetched from the program
memory. This concept enables instructions to be executed in every clock cycle. The program memory is In-System Programmable Flash memory. With a few exceptions, AVR
instructions have a single 16-bit word format, meaning that every program memory
address contains a single 16-bit instruction.
During interrupts and subroutine calls, the return address program counter (PC) is
stored on the stack. The stack is effectively allocated in the general data SRAM and,
consequently, the stack size is only limited by the total SRAM size and the usage of the
SRAM. All user programs must initialize the SP in the reset routine (before subroutines
or interrupts are executed). The 16-bit stack pointer (SP) is read/write accessible in the
I/O space.
The 4000 bytes data SRAM can be easily accessed through the five different addressing modes supported in the AVR architecture.
A flexible interrupt module has its control registers in the I/O space with an additional
global interrupt enable bit in the status register. All the different interrupts have a sepa-
8
ATmega103(L)
0945G–09/01
ATmega103(L)
rate interrupt vector in the interrupt vector table at the beginning of the
program memory. The different interrupts have priority in accordance with their interrupt
vector position. The lower the interrupt vector address, the higher the priority.
The memory spaces in the AVR architecture are all linear and regular memory maps.
General-purpose
Register File
Figure 5 shows the structure of the 32 general-purpose working registers in the CPU.
Figure 5. AVR CPU General-purpose Working Registers
70Addr.
R0 $00
R1$01
R2$02
.=.=.
R13$0D
GeneralR14$0E
PurposeR15$0F
WorkingR16$10
RegistersR17$11
. . .
R26$1AX-register low byte
R27$1BX-register high byte
R28$1CY-register low byte
R29$1DY-register high byte
R30$1EZ-register low byte
R31$1FZ-register high byte
All the register operating instructions in the instruction set have direct and single-cycle
access to all registers. The only exception are the five constant arithmetic and logic
instructions SBCI, SUBI, CPI, ANDI and ORI between a constant and a register and the
LDI instruction for load immediate constant data. These instructions apply to the second
half of the registers in the register file – R16..R31. The general SBC, SUB, CP, AND and
OR and all other operations between two registers or on a single register apply to the
entire register file.
0945G–09/01
As shown in Figure 5, each register is also assigned a data memory address, mapping
them directly into the first 32 locations of the user Data Space. Although not being physically implemented as SRAM locations, this memory organization provides great
flexibility in access of the registers, as the X-, Y-, and Z-registers can be set to index any
register in the file.
The 4K bytes of SRAM available for general data are implemented as addresses $0060
to $0FFF.
9
X-register, Y-register and Zregister
The registers R26..R31 have some added functions to their general-purpose usage.
These registers are address pointers for indirect addressing of the SRAM. The three
indirect address registers X, Y, and Z are defined as:
Figure 6. X-, Y-, and Z-registers
150
X-register7 07 0
R27 ($1B)R26 ($1A)
150
Y-register7 07 0
R29 ($1D)R28 ($1C)
150
Z-register7 07 0
R31 ($1F)R30 ($1E)
In the different addressing modes these address registers have functions as fixed displacement, automatic increment and decrement (see the descriptions for the different
instructions).
ALU – Arithmetic Logic
Unit
ISP Flash Program
Memory
The high-performance AVR ALU operates in direct connection with all the 32 generalpurpose working registers. Within a single clock cycle, ALU operations between registers in the register file are executed. The ALU operations are divided into three main
categories: arithmetic, logical and bit functions.
The ATmega103(L) contains 128K bytes of On-chip In-System Programmable Flash
memory for program storage. Since all instructions are single or double 16-bit words, the
Flash is organized as 64K x 16. The Flash memory has an endurance of at least 1000
write/erase cycles.
Constant tables can be allocated in the entire program memory space (see the LPM –
Load Program Memory and ELPM – Extended Load Program Memory instruction
descriptions).
SRAM Data MemoryThe ATmega103(L) supports two different configurations for the SRAM data memory as
listed in Table 1.
Table 1. Memory Configurations
ConfigurationInternal SRAM Data MemoryExternal SRAM Data Memory
A4000None
B4000up to 64K
Note:When using 64K of external SRAM, 60K will be available.
10
ATmega103(L)
0945G–09/01
Figure 7. Memory Configurations
Memory Configuration A
Program Flash
(32K/64K x 16)
$0000
ATmega103(L)
Data MemoryProgram Memory
32 Registers
64 I/O Registers
Internal SRAM
(4000 x 8)
$0000 - $001F
$0020 - $005F
$0060
$0FFF
Memory Configuration B
Program Memory
Program Flash
(32K/64K x 16)
$7FFF/$FFFF
$0000
Data Memory
32 Registers
64 I/O Registers
Internal SRAM
(4000 x 8)
External SRAM
(0 - 64K x 8)
$0000 - $001F
$0020 - $005F
$0060
$0FFF
$1000
0945G–09/01
$7FFF/
$FFFF
$FFFF
11
The 4096 first data memory locations address both the register file, the I/O memory and
the internal data SRAM. The first 96 locations address the register file and I/O memory,
and the next 4000 locations address the internal data SRAM.
An optional external data SRAM can be used with the ATmega103(L). This SRAM will
occupy an area in the remaining address locations in the 64K address space. This area
starts at the address following the internal SRAM. If a 64K external SRAM is used, 4K of
the external memory is lost as the addresses are occupied by internal memory.
When the addresses accessing the SRAM memory space exceeds the internal data
memory locations, the external data SRAM is accessed using the same instructions as
for the internal data memory access. When the internal data memories are accessed,
the read and write strobe pins (RD
External SRAM operation is enabled by setting the SRE bit in the MCUCR register.
Accessing external SRAM takes one additional clock cycle per byte compared to access
of the internal SRAM. This means that the commands LD, ST, LDS, STS, PUSH and
POP take one additional clock cycle. If the stack is placed in external SRAM, interrupts,
subroutine calls and returns take two clock cycles extra because the 2-byte program
counter is pushed and popped. When external SRAM interface is used with wait state,
two additional clock cycles are used per byte. This has the following effect: Data transfer
instructions take two extra clock cycles, whereas interrupt, subroutine calls and returns
will need four clock cycles more than specified in the “Instruction Set Summary” on page
130.
The five different addressing modes for the data memory cover: Direct, Indirect with Displacement, Indirect, Indirect with Pre-decrement and Indirect with Post-increment. In the
register file, registers R26 to R31 feature the indirect addressing pointer registers.
and WR) are inactive during the whole access cycle.
Program and Data
Addressing Modes
The Indirect with Displacement mode features 63 address locations reached from the
base address given by the Y- or Z-register.
When using register indirect addressing modes with automatic pre-decrement and postincrement, the address registers X, Y, and Z are decremented and incremented.
The entire data address space including the 32 general-purpose working registers and
the 64 I/O registers are all accessible through all these addressing modes. See the next
section for a detailed description of the different addressing modes.
The ATmega103(L) AVR RISC microcontroller supports powerful and efficient addressing modes for access to the program memory (Flash) and data memory (SRAM, register
file and I/O memory). This section describes the different addressing modes supported
by the AVR architecture. In the figures, OP means the operation code part of the instruction word. To simplify, not all figures show the exact location of the addressing bits.
12
ATmega103(L)
0945G–09/01
ATmega103(L)
Register Direct, Single
Register Rd
Register Direct, Two Registers
Rd and Rr
Figure 8. Direct Single Register Addressing
15
OPd
04
The operand is contained in register d (Rd).
Figure 9. Direct Register Addressing, Two Registers
15
OPdr
0459
REGISTER FILE
0
d
31
REGISTER FILE
0
Operands are contained in registers r (Rr) and d (Rd). The result is stored in register d
(Rd).
I/O DirectFigure 10. I/O Direct Addressing
15
OPP
n
d
r
31
I/O MEMORY
0
05
63
0945G–09/01
Operand address is contained in six bits of the instruction word. n is the destination or
source register address.
13
Data DirectFigure 11. Direct Data Addressing
31
OPRr/Rd
150
20 19
16 LSBs
A 16-bit data address is contained in the 16 LSBs of a 2-word instruction. Rd/Rr specify
the destination or source register.
16
Data Space
$0000
$FFFF
Data Indirect with
Figure 12. Data Indirect with Displacement
Displacement
15
Y- OR Z-REGISTER
15
OPan
Operand address is the result of the Y- or Z-register contents added to the address contained in six bits of the instruction word.
Data IndirectFigure 13. Data Indirect Addressing
X-, Y- OR Z-REGISTER
Data Space
$0000
0
05610
$FFFF
Data Space
$0000
015
14
$FFFF
Operand address is the contents of the X-, Y,- or the Z-register.
ATmega103(L)
0945G–09/01
ATmega103(L)
Data Indirect with Predecrement
Data Indirect with Postincrement
Figure 14. Data Indirect Addressing with Pre-decrement
Data Space
$0000
015
X-, Y- OR Z-REGISTER
-1
$FFFF
The X-, Y-, or the Z-register is decremented before the operation. Operand address is
the decremented contents of the X-, Y-, or the Z-register.
Figure 15. Data Indirect Addressing with Post-increment
Data Space
$0000
015
X-, Y- OR Z-REGISTER
Constant Addressing Using
the LPM and ELPM
Instructions
1
$FFFF
The X-, Y-, or the Z-register is incremented after the operation. Operand address is the
contents of the X-, Y-, or the Z-register prior to incrementing.
Figure 16. Code Memory Constant Addressing
PROGRAM MEMORY
$0000
0115
Z-REGISTER
$7FFF/$FFFF
Constant byte address is specified by the Z-register contents. The 15 MSBs select word
address (0 - 32K), LSB selects low byte if cleared (LSB = 0) or high byte if set (LSB = 1).
0945G–09/01
15
If ELPM is used, LSB of the RAM Page Z register (RAMPZ) is used to select low or high
memory page (RAMPZ0 = 0: Low Page, RAMPZ0 = 1: High Page).
Direct Program Address, JMP
and CALL
Indirect Program Addressing,
IJMP and ICALL
Figure 17. Direct Program Memory Addressing
31
OP
150
21 20
16 LSBs
16
PROGRAM MEMORY
$0000
$7FFF/$FFFF
Program execution continues at the address immediate in the instruction words.
Figure 18. Indirect Program Memory Addressing
PROGRAM MEMORY
015
Z-REGISTER
$0000
Relative Program Addressing,
RJMP and RCALL
16
ATmega103(L)
$7FFF/$FFFF
Program execution continues at address contained by the Z-register (i.e., the PC is
loaded with the contents of the Z-register).
Figure 19. Relative Program Memory Addressing
15
15
1112
OPk
PROGRAM MEMORY
0
PC
1
0
$0000
$7FFF/$FFFF
0945G–09/01
ATmega103(L)
Program execution continues at address PC + k + 1. The relative address k is -2048 to
2047.
EEPROM Data MemoryThe EEPROM memory is organized as a separate data space in which single bytes can
be read and written. The EEPROM has an endurance of at least 100,000 write/erase
cycles. The access between the EEPROM and the CPU is described on page 54 specifying the EEPROM address register, the EEPROM data register and the EEPROM
control register.
Memory Access Times
and Instruction
Execution Timing
This section describes the general access timing concepts for instruction execution and
internal memory access.
The AVR CPU is driven by the System Clock Ø, directly generated from the external
clock crystal for the chip. No internal clock division is used.
Figure 20 shows the parallel instruction fetches and instruction executions enabled by
the Harvard architecture and the fast-access register file concept. This is the basic pipelining concept to obtain up to 1 MIPS per MHz with the corresponding unique results for
functions per cost, functions per clocks and functions per power unit.
Figure 20. The Parallel Instruction Fetches and Instruction Executions
T1T2T3T4
System Clock Ø
1st Instruction Fetch
1st Instruction Execute
2nd Instruction Fetch
2nd Instruction Execute
3rd Instruction Fetch
3rd Instruction Execute
4th Instruction Fetch
0945G–09/01
Figure 21 shows the internal timing concept for the register file. In a single clock cycle,
an ALU operation using two register operands is executed and the result is stored back
to the destination register.
Figure 21. Single Cycle ALU Operation
T1T2T3T4
System Clock Ø
Total Execution Time
Register Operands Fetch
ALU Operation Execute
Result Write Back
17
The internal data SRAM access is performed in two System Clock cycles as described
in Figure 22.
Figure 22. On-chip Data SRAM Access Cycles
T1T2T3T4
System Clock Ø
Address
Prev. Address
Address
Data
WR
Data
RD
See “Interface to External SRAM” on page 79. for a description of the access to the
external SRAM.
I/O MemoryThe I/O space definition of the ATmega103(L) is shown in Table 2.
$08 ($28) ACSRAnalog Comparator Control and Status Register
$07 ($27) ADMUXADC Multiplexer Select Register
19
Table 2. ATmega103(L) I/O Space (Continued)
I/O Address (SRAM
Address) NameFunction
$06 ($26) ADCSRADC Control and Status Register
$05 ($25) ADCHADC Data Register High
$04 ($24) ADCLADC Data Register Low
$03 ($23) PORTEData Register, Port E
$02 ($22) DDREData Direction Register, Port E
$01 ($21) PINEInput Pins, Port E
$00 ($20) PINFInput Pins, Port F
Note:Reserved and unused locations are not shown in the table.
All the different ATmega103(L) I/Os and peripherals are placed in the I/O space. The different I/O locations are directly accessed by the IN and OUT instructions transferring
data between the 32 general-purpose working registers and the I/O space. I/O registers
within the address range $00 - $1F are directly bit-accessible using the SBI and CBI
instructions. In these registers, the value of single bits can be checked by using the
SBIS and SBIC instructions. Refer to the “Instruction Set Summary” on page 130 for
more details. When using the I/O specific instructions IN and OUT, the I/O register
address $00 - $3F are used. When addressing I/O registers as SRAM, $20 must be
added to this address. All I/O register addresses throughout this document are shown
with the SRAM address in parentheses.
For compatibility with future devices, reserved bits should be written to zero if accessed.
Reserved I/O memory addresses should never be written.
Some of the status flags are cleared by writing a logical “1” to them. Note that the CBI
and SBI instructions will operate on all bits in the I/O register, writing a one back into any
flag read as set, thus clearing the flag. The CBI and SBI instructions work with registers
$00 to $1F only.
The different I/O and peripherals control registers are explained in the following
sections.
Status Register – SREGThe AVR status register (SREG) at I/O space location $3F ($5F) is defined as:
Bit76543210
$3F ($5F)ITHSVNZCSREG
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
• Bit 7 – I: Global Interrupt Enable
The global interrupt enable bit must be set (one) for the interrupts to be enabled. The
individual interrupt enable control is then performed in separate control registers. If the
global interrupt enable register is cleared (zero), none of the interrupts are enabled independent of the individual interrupt enable settings. The I-bit is cleared by hardware after
an interrupt has occurred and is set by the RETI instruction to enable subsequent
interrupts.
• Bit 6 – T: Bit Copy Storage
20
The bit copy instructions BLD (Bit LoaD) and BST (Bit STore) use the T-bit as source
and destination for the operated bit. A bit from a register in the register file can be copied
ATmega103(L)
0945G–09/01
ATmega103(L)
into T by the BST instruction and a bit in T can be copied into a bit in a register in the
register file by the BLD instruction.
• Bit 5 – H: Half-carry Flag
The half-carry flag H indicates a half-carry in some arithmetic operations. See the
instruction set description on page 130 for detailed information.
• Bit 4 – S: Sign Bit, S = N
The S-bit is always an exclusive or between the negative flag N and the two’s complement overflow flag V. See the instruction set description on page 130 for detailed
information.
• Bit 3 – V: Two’s Complement Overflow Flag
The two’s complement overflow flag V supports two’s complement arithmetics. See the
instruction set description on page 130 for detailed information.
• Bit 2 – N: Negative Flag
The negative flag N indicates a negative result from an arithmetical or logical operation.
See the Instruction set description on page 130 for detailed information.
• Bit 1 – Z: Zero Flag
The zero flag Z indicates a zero result from an arithmetical or logical operation. See the
instruction set description on page 130 for detailed information.
• Bit 0 – C: Carry Flag
⊕ V
The carry flag C indicates a carry in an arithmetical or logical operation. See the instruction set description on page 130 for detailed information.
Note that the status register is not automatically stored when entering an interrupt routine or restored when returning from an interrupt routine. This must be handled by
software.
Stack Pointer – SPThe general AVR 16-bit Stack Pointer is effectively built up of two 8-bit registers in the
I/O space locations $3E ($5E) and $3D ($5D). As the ATmega103(L) supports up to
64K bytes memory, all 16 bits are used.
Bit151413121110 9 8
$3E ($5E)SP15SP14SP13SP12SP11SP10SP9SP8SPH
$3D ($5D)SP7SP6SP5SP4SP3SP2SP1SP0SPL
76543210
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
R/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
00000000
The Stack Pointer points to the data SRAM stack area where the Subroutine and Interrupt Stacks are located. This stack space in the data SRAM must be defined by the
program before any subroutine calls are executed or interrupts are enabled. The Stack
Pointer must be set to point above $60. The Stack Pointer is decremented by one when
data is pushed onto the stack with the PUSH instruction and it is decremented by 2
when an address is pushed onto the stack with subroutine calls and interrupts. The
Stack Pointer is incremented by 1 when data is popped from the stack with the POP
instruction and it is incremented by 2 when an address is popped from the stack with
return from subroutine RET or return from interrupt RETI.
0945G–09/01
21
RAM Page Z Select Register –
RAMPZ
Bit7654321 0
$3B ($5B)–––––––RAMPZ0RAMPZ
Read/WriteRRRRRRRR/W
Initial Value0000000 0
The RAMPZ register is normally used to select which 64K RAM page is accessed by the
Z pointer. As the ATmega103(L) does not support more than 64K of SRAM memory,
this register is used only to select which page in the program memory is accessed when
the ELPM instruction is used. The different settings of the RAMPZ0 bit have the following effects:
Note that LPM is not affected by the RAMPZ setting.
MCU Control Register –
MCUCR
The MCU Control Register contains control bits for general MCU functions.
Bit7654321 0
$35 ($55)SRESRWSESM1SM0–––MCUCR
Read/WriteR/WR/WR/WR/WR/WRRR
Initial Value0000000 0
• Bit 7 – SRE: External SRAM Enable
When the SRE bit is set (one), the external data SRAM is enabled, and the pin functions
AD0 - 7 (Port A), and A8 - 15 (Port C) are activated as the alternate pin functions. Then
the SRE bit overrides any pin direction settings in the respective data direction registers.
When the SRE bit is cleared (zero), the external data SRAM is disabled and the normal
pin and data direction settings are used.
• Bit 6 – SRW: External SRAM Wait State
When the SRW bit is set (one), a one-cycle wait state is inserted in the external data
SRAM access cycle. When the SRW bit is cleared (zero), the external data SRAM
access is executed with a three-cycle scheme. See Figure 51 on page 80 and Figure 52
on page 80.
• Bit 5 – SE: Sleep Enable
The SE bit must be set (one) to make the MCU enter the Sleep Mode when the SLEEP
instruction is executed. To avoid the MCU entering the Sleep Mode unless it is the programmer’s purpose, it is recommended to set the Sleep Enable (SE) bit just before the
execution of the SLEEP instruction.
This bit selects between the three available sleep modes as shown in Table 3.
Table 3. Sleep Mode Select
SM1SM0Sleep Mode
00Idle Mode
01Reserved
10Power-down
11Power-save
ATmega103(L)
0945G–09/01
ATmega103(L)
• Bits 2..0 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and always read as zero.
XTAL Divide Control Register
– XDIV
The XTAL Divide Control Register is used to divide the XTAL clock frequency by a number in the range 1 - 129. This feature can be used to decrease power consumption when
the requirement for processing power is low.
When the XDIVEN bit is set (one), the clock frequency of the CPU and all peripherals is
divided by the factor defined by the setting of XDIV6 - XDIV0. This bit can be set and
cleared run-time to vary the clock frequency as suitable to the application.
These bits define the division factor that applies when the XDIVEN bit is set (one). If the
value of these bits is denoted d, the following formula defines the resulting CPU clock
frequency f
clk
:
f
CLK
XTAL
-------------------=
129
d–
The value of these bits can only be changed when XDIVEN is zero. When XDIVEN is
set to one, the value written simultaneously into XDIV6..XDIV0 is taken as the division
factor. When XDIVEN is cleared to zero, the value written simultaneously into
XDIV6..XDIV0 is rejected. As the divider divides the master clock input to the MCU, the
speed of all peripherals is reduced when a division factor is used.
Reset and Interrupt
Handling
The ATmega103(L) provides 23 different interrupt sources. These interrupts and the
separate reset vector each have a separate program vector in the program memory
space. All interrupts are assigned individual enable bits that must be set (one) together
with the I-bit in the Status Register in order to enable the interrupt.
The lowest addresses in the program memory space are automatically defined as the
Reset and Interrupt vectors. The complete list of vectors is shown in Table 4. The list
also determines the priority levels of the different interrupts. The lower the address, the
higher the priority level. RESET has the highest priority and next is INT0 (the External
Interrupt Request 0), etc.
Table 4. Reset and Interrupt Vectors
Program
Vecto r No .
1$0000RESET
2$0002INT0External Interrupt Request 0
3$0004INT1External Interrupt Request 1
4$0006INT2External Interrupt Request 2
5$0008INT3External Interrupt Request 3
6$000AINT4External Interrupt Request 4
AddressSourceInterrupt Definition
Hardware Pin, Power-on Reset and Watchdog
Reset
0945G–09/01
23
Table 4. Reset and Interrupt Vectors (Continued)
Program
Vecto r No .
7$000CINT5External Interrupt Request 5
8$000EINT6External Interrupt Request 6
9$0010INT7External Interrupt Request 7
10$0012TIMER2 COMPTimer/Counter2 Compare Match
11$0014TIMER2 OVFTimer/Counter2 Overflow
12$0016TIMER1 CAPTTimer/Counter1 Capture Event
13$0018TIMER1 COMPATimer/Counter1 Compare Match A
14$001ATIMER1 COMPBTimer/Counter1 Compare Match B
15$001CTIMER1 OVFTimer/Counter1 Overflow
16$001ETIMER0 COMPTimer/Counter0 Compare Match
17$0020TIMER0 OVFTimer/Counter0 Overflow
18$0022SPI, STCSPI Serial Transfer Complete
19$0024UART, RXUART, Rx Complete
20$0026UART, UDREUART Data Register Empty
21$0028UART, TXUART, Tx Complete
22$002AADCADC Conversion Complete
AddressSourceInterrupt Definition
23$002CEE READYEEPROM Ready
24$002EANALOG COMPAnalog Comparator
The most typical program setup for the Reset and Interrupt vector addresses are:
Address Labels CodeComments
$0000jmpRESET; Reset Handler
$0002jmpEXT_INT0; IRQ0 Handler
$0004jmpEXT_INT1; IRQ1 Handler
$0006jmpEXT_INT2; IRQ2 Handler
$0008jmpEXT_INT3; IRQ3 Handler
$000AjmpEXT_INT4; IRQ4 Handler
$000CjmpEXT_INT5; IRQ5 Handler
$000EjmpEXT_INT6; IRQ6 Handler
$0010jmpEXT_INT7; IRQ7 Handler
$0012jmpTIM2_COMP; Timer2 Compare Handler
$0014jmpTIM2_OVF; Timer2 Overflow Handler
$0016jmpTIM1_CAPT; Timer1 Capture Handler
$0018jmpTIM1_COMPA; Timer1 CompareA Handler
$001AjmpTIM1_COMPB; Timer1 CompareB Handler
$001CjmpTIM1_OVF; Timer1 Overflow Handler
$001EjmpTIM0_COMP; Timer0 Compare Handler
$0020jmpTIM0_OVF; Timer0 Overflow Handler
$0022jmpSPI_STC; SPI Transfer Complete Handler
$0024jmpUART_RXC; UART RX Complete Handler
$0026jmpUART_DRE; UDR Empty Handler
24
ATmega103(L)
0945G–09/01
$0028jmpUART_TXC; UART TX Complete Handler
$002AjmpADC; ADC Conversion Complete Handler
$002CjmpEE_RDY; EEPROM Ready Handler
$002EjmpANA_COMP; Analog Comparator Handler
;
$0030MAIN:ldir16, high(RAMEND); Main program start
$0031outSPH,r16
$0032ldir16, low(RAMEND)
$0033outSPL,r16
$0034<instr> xxx
............
Reset SourcesThe ATmega103(L) has three sources of reset:
•Power-on Reset. The MCU is reset when the supply voltage is below the Power-on
Reset threshold (V
POT
).
•External Reset. The MCU is reset when a low level is present on the RESET pin for
more than 50 ns.
•Watchdog Reset. The MCU is reset when the Watchdog timer period expires and
the Watchdog is enabled.
ATmega103(L)
During reset, all I/O registers except the MCU Status Register are then set to their initial
values and the program starts execution from address $0000. The instruction placed in
address $0000 must be a JMP (absolute jump) instruction to the reset handling routine.
If the program never enables an interrupt source, the interrupt vectors are not used and
regular program code can be placed at these locations. The circuit diagram in Figure 23
shows the reset logic. Table 5 defines the timing and electrical parameters of the reset
circuitry.
Figure 23. Reset Logic
VCC
RESET
PEND Q
E
XTAL1
10-50K
100-500K
Power-on Reset
Circuit
Reset Circuit
Watchdog
Timer
On-chip
RC Oscillator
POR
14-stage Ripple Counter
Delay Unit
COUNTER RESET
Q8Q11 Q13
SUT0
SUT1
QS
Q
R
INTERNAL
RESET
0945G–09/01
25
Table 5. Reset Characteristics (V
SymbolParameterConditionMinTypMaxUnits
Power-on Reset Threshold
(rising)
(1)
V
POT
Power-on Reset Threshold
(falling)
= 5.0V)
CC
1.01.41.8V
0.40.60.8V
V
RST
T
TOUT
Note:1. The Power-on Reset will not work unless the supply voltage has been below V
RESET Pin Threshold VoltageVCC/2V
SUT = 005CPU cycles
Reset Delay Time-out Period
(falling).
SUT = 01
SUT = 10
SUT = 11
0.4
3.2
12.8
0.5
4.0
16.0
0.6
4.8
19.2
ms
POT
Power-on ResetA Power-on Reset (POR) circuit ensures that the device is reset from power-on. As
shown in Figure 23, an internal timer clocked from the Watchdog timer oscillator prevents the MCU from starting until after a certain period after V
on Threshold voltage (V
), regardless of the VCC rise time (see Figure 24). The Fuse
POT
has reached the Power-
CC
bits SUT1 and SUT0 are used to select start-up time as indicated in Table 5. A “0” in the
table indicates that the fuse is programmed.
The user can select the start-up time according to typical oscillator start-up time. The
number of WDT oscillator cycles used for each time-out except for SUT = 00 is shown in
Table 6. The frequency of the Watchdog oscillator is voltage-dependent as shown in
“Typical Characteristics” on page 118.
Table 6. Number of Watchdog Oscillator Cycles
SUT 1/0Time-out at VCC = 5VNumber of WDT Cycles
010.5 ms512
26
104.0 ms4K
1116.0 ms16K
The setting SUT 1/0 = 00 starts the MCU after 5 CPU clock cycles, and can be used
when an external clock signal is applied to the XTAL1 pin. This setting does not use the
WDT oscillator and enables very fast start-up from the sleep modes Power-down or
Power-save if the clock signal is present during sleep. For details, refer to the programming specification starting on page 99.
If the built-in start-up delay is sufficient, RESET
an external pull-up resistor. By holding the pin low for a period after V
applied, the Power-on Reset period can be extended. Refer to Figure 25 for a timing
example of this.
ATmega103(L)
can be connected to VCC directly or via
has been
CC
0945G–09/01
Figure 24. MCU Start-up, RESET Tied to VCC.
V
VCC
RESET
TIME-OUT
INTERNAL
RESET
POT
V
RST
t
TOUT
ATmega103(L)
Figure 25. MCU Start-up, RESET
V
VCC
RESET
TIME-OUT
INTERNAL
RESET
POT
Controlled Externally
V
RST
t
TOUT
External ResetAn external reset is generated by a low level on the RESET
than 50 ns will generate a reset even if the clock is not running. Shorter pulses are not
guaranteed to generate a reset. When the applied signal reaches the Reset Threshold
Voltage (V
period t
) on its positive edge, the delay timer starts the MCU after the Time-out
RST
has expired.
TOUT
Figure 26. External Reset during Operation
VCC
pin. Reset pulses longer
0945G–09/01
RESET
TIME-OUT
INTERNAL
RESET
V
RST
t
TOUT
27
Watchdog ResetWhen the Watchdog times out, it will generate a short reset pulse of 1 XTAL cycle dura-
tion. On the falling edge of this pulse, the delay timer starts counting the Time-out period
. Refer to page 52 for details on operation of the Watchdog.
t
TOUT
Figure 27. Watchdog Reset during Operation
VCC
RESET
MCU Status Register –
MCUSR
WDT
TIME-OUT
RESET
TIME-OUT
INTERNAL
RESET
1 XTAL Cycle
t
TOUT
The MCU Status Register provides information on which reset source caused an MCU
reset.
Bit7 654321 0
$34 ($54)– –––––EXTRFPORFMCUSR
Read/Write R RRRRRR/WR/W
Initial Value0 00000See bit description
• Bits 7..2 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and always read as zero.
• Bit 1 – EXTRF: External Reset Flag
After a Power-on Reset, this bit is undefined (X). It will be set by an external reset. A
Watchdog reset will leave this bit unchanged.
• Bit 0 – PORF: Power-on Reset Flag
28
This bit is set by a Power-on Reset. A Watchdog Reset or an External Reset will leave
this bit unchanged.
To summarize, Table 7 shows the value of these two bits after the three modes of reset:
Table 7. PORF and EXTRF Values after Reset
Reset SourceEXTRFPORF
Power-on Resetundefined1
External Reset1unchanged
Watchdog Resetunchangedunchanged
To make use of these bits to identify a reset condition, the user software should clear
both the PORF and EXTRF bits as early as possible in the program. Checking the
PORF and EXTRF values is done before the bits are cleared. If the bit is cleared before
an external or Watchdog reset occurs, the source of reset can be found by using the following truth table, Table 8.
ATmega103(L)
0945G–09/01
ATmega103(L)
Table 8. Reset Source Identification
Reset SourceEXTRFPORF
Watchdog Reset00
Power-on Reset01
External Reset10
Power-on Reset11
Interrupt HandlingThe ATmega103(L) has two dedicated 8-bit Interrupt Mask control registers; EIMSK
(External Interrupt Mask register) and TIMSK (Timer/Counter Interrupt Mask register). In
addition, other enable and mask bits can be found in the peripheral control registers.
When an interrupt occurs, the Global Interrupt Enable I-bit is cleared (zero) and all interrupts are disabled. The user software can set (one) the I-bit to enable nested interrupts.
The I-bit is set (one) when a Return from Interrupt instruction (RETI) is executed.
When the Program Counter is vectored to the actual interrupt vector in order to execute
the interrupt handling routine, hardware clears the corresponding flag that generated the
interrupt. Some of the interrupt flags can also be cleared by writing a logical “1” to the
flag bit position(s) to be cleared.
External Interrupt Mask
Register – EIMSK
If an interrupt condition occurs when the corresponding interrupt enable bit is cleared
(zero), the interrupt flag will be set and remembered until the interrupt is enabled or the
flag is cleared by software.
If one or more interrupt conditions occur when the global interrupt enable bit is cleared
(zero), the corresponding interrupt flag(s) will be set and remembered until the global
interrupt enable bit is set (one), and will be executed by order of priority.
Note that external level interrupt does not have a flag and will only be remembered for
as long as the interrupt condition is active.
Note that the Status Register is not automatically stored when entering an interrupt routine or restored when returning from an interrupt routine. This must be handled by
software.
When an INT7 - INT4 bit is set (one) and the I-bit in the Status Register (SREG) is set
(one), the corresponding external pin interrupt is enabled. The Interrupt Sense Control
bits in the External Interrupt Control Register (EICR) define whether the external interrupt is activated on rising or falling edge or is level-sensed. Activity on any of these pins
will trigger an interrupt request even if the pin is enabled as an output. This provides a
way of generating a software interrupt.
When an INT3 - INT0 bit is set (one) and the I-bit in the Status Register (SREG) is set
(one), the corresponding external pin interrupt is enabled. The external interrupts are
always low-level triggered interrupts. Activity on any of these pins will trigger an interrupt
29
External Interrupt Flag
Register – EIFR
External Interrupt Control
Register – EICR
request even if the pin is enabled as an output. This provides a way of generating a software interrupt. When enabled, a level-triggered interrupt will generate an interrupt
request as long as the pin is held low.
When an edge on the INT7 - INT4 pins triggers an interrupt request, the corresponding
interrupt flag, INTF7 - INTF4, becomes set (one). If the I-bit in SREG and the corresponding interrupt enable bit, INT7 - INT4 in EIMSK, is set (one), the MCU will jump to
the interrupt vector. The flag is cleared when the corresponding interrupt routine is executed. Alternatively, the flag is cleared by writing a logical “1” to it. These flags are
always cleared when INTF7 - INFT4 are configured as level interrupts.
• Bits 3..0 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and always read as zero.
• Bits 7..0 – ISCX1, ISCX0: External Interrupt 7 - 4 Sense Control Bits
The External Interrupts 7 - 4 are activated by the external pins INT7 - INT4 if the SREG
I-flag and the corresponding interrupt mask in the EIMSK are set. The level and edges
on the external pins that activate the interrupts are defined in Table 9.
Table 9. Interrupt Sense Control
ISCX1ISCX0Description
00The low level of INTX generates an interrupt request.
01Reserved
10The falling edge of INTX generates an interrupt request.
11The rising edge of INTX generates an interrupt request.
The value on the INTX pin is sampled before detecting edges. If edge interrupt is
selected, pulses that last longer than one CPU clock period will generate an interrupt.
Shorter pulses are not guaranteed to generate an interrupt. Observe that CPU clock frequency can be lower than the XTAL frequency if the XTAL divider is enabled. If low-level
interrupt is selected, the low level must be held until the completion of the currently executing instruction to generate an interrupt. If enabled, a level-triggered interrupt will
generate an interrupt request as long as the pin is held low.
30
ATmega103(L)
0945G–09/01
Timer/Counter Interrupt Mask
Register – TIMSK
ATmega103(L)
Bit76543210
$37 ($57)
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
• Bit 7 – OCIE2: Timer/Counter2 Output Compare Interrupt Enable
When the OCIE2 bit is set (one) and the I-bit in the Status Register is set (one), the
Timer/Counter2 Compare Match interrupt is enabled. The corresponding interrupt (at
vector $0012) is executed if a compare match in Timer/Counter2 occurs, i.e., when the
OCF2 bit is set in the Timer/Counter Interrupt Flag Register (TIFR).
• Bit 6 – TOIE2: Timer/Counter2 Overflow Interrupt Enable
When the TOIE2 bit is set (one) and the I-bit in the Status Register is set (one), the
Timer/Counter2 Overflow interrupt is enabled. The corresponding interrupt (at vector
$0014) is executed if an overflow in Timer/Counter2 occurs, i.e., when the TOV2 bit is
set in the Timer/Counter Interrupt Flag Register (TIFR).
• Bit 5 – TICIE1: Timer/Counter1 Input Capture Interrupt Enable
When the TICIE1 bit is set (one) and the I-bit in the Status Register is set (one), the
Timer/Counter1 Input Capture Event interrupt is enabled. The corresponding interrupt
(at vector $0016) is executed if a capture-triggering event occurs on pin 29, PD4(IC1),
i.e., when the ICF1 bit is set in the Timer/Counter Interrupt Flag Register.
• Bit 4 – OCE1A: Timer/Counter1 Output CompareA Match Interrupt Enable
OCIE2TOIE2TICIE1OCIE1AOCIE1BTOIE1OCIE0TOIE0TIMSK
When the OCIE1A bit is set (one) and the I-bit in the Status Register is set (one), the
Timer/Counter1 CompareA Match interrupt is enabled. The corresponding interrupt (at
vector $0018) is executed if a CompareA match in Timer/Counter1 occurs, i.e., when
the OCF1A bit is set in the Timer/Counter Interrupt Flag Register.
• Bit 3 – OCIE1B: Timer/Counter1 Output CompareB Match Interrupt Enable
When the OCIE1B bit is set (one) and the I-bit in the Status Register is set (one), the
Timer/Counter1 CompareB Match interrupt is enabled. The corresponding interrupt (at
vector $001A) is executed if a CompareB match in Timer/Counter1 occurs, i.e., when
the OCF1B bit is set in the Timer/Counter Interrupt Flag Register.
• Bit 2 – TOIE1: Timer/Counter1 Overflow Interrupt Enable
When the TOIE1 bit is set (one) and the I-bit in the Status Register is set (one), the
Timer/Counter1 Overflow interrupt is enabled. The corresponding interrupt (at vector
$001C) is executed if an overflow in Timer/Counter1 occurs, i.e., when the TOV1 bit is
set in the Timer/Counter Interrupt Flag Register.
• Bit 1 – OCIE0: Timer/Counter0 Output Compare Interrupt Enable
When the OCIE0 bit is set (one) and the I-bit in the Status Register is set (one), the
Timer/Counter0 Compare Match interrupt is enabled. The corresponding interrupt (at
vector $001E) is executed if a compare match in Timer/Counter0 occurs, i.e., when the
OCF0 bit is set in the Timer/Counter Interrupt Flag Register.
• Bit 0 – TOIE0: Timer/Counter0 Overflow Interrupt Enable
When the TOIE0 bit is set (one) and the I-bit in the Status Register is set (one), the
Timer/Counter0 Overflow interrupt is enabled. The corresponding interrupt (at vector
$0020) is executed if an overflow in Timer/Counter0 occurs, i.e., when the TOV0 bit is
set in the Timer/Counter Interrupt Flag Register.
0945G–09/01
31
Timer/Counter Interrupt Flag
Register – TIFR
Bit76543210
$36 ($56)OCF2TOV2ICF1OCF1AOCF1BTOV1OCF0TOV0TIFR
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
• Bit 7 – OCF2: Output Compare Flag 2:
The OCF2 bit is set (one) when compare match occurs between Timer/Counter2 and
the data in OCR2 – Output Compare Register 2. OCF2 is cleared by hardware when
executing the corresponding interrupt handling vector. Alternatively, OCF2 is cleared by
writing a logical “1” to the flag. When the I-bit in SREG, and OCIE2 (Timer/Counter2
Compare Interrupt Enable) and the OCF2 are set (one), the Timer/Counter2 Output
Compare interrupt is executed.
• Bit 6 – TOV2: Timer/Counter2 Overflow Flag
The TOV2 bit is set (one) when an overflow occurs in Timer/Counter2. TOV2 is cleared
by hardware when executing the corresponding interrupt handling vector. Alternatively,
TOV2 is cleared by writing a logical “1” to the flag. When the I-bit in SREG, and TOIE2
(Timer/Counter1 Overflow Interrupt Enable) and TOV2 are set (one), the
Timer/Counter2 Overflow interrupt is executed. In PWM mode, this bit is set when
Timer/Counter2 advances from $00.
• Bit 5 – ICF1: Input Capture Flag 1
The ICF1 bit is set (one) to flag an input capture event, indicating that the
Timer/Counter1 value has been transferred to the input capture register (ICR1). ICF1 is
cleared by hardware when executing the corresponding interrupt handling vector. Alternatively, ICF1 is cleared by writing a logical “1” to the flag. When the SREG I-bit, TICIE1
(Timer/Counter1 Input Capture Interrupt Enable) and ICF1 are set (one), the
Timer/Counter1 Capture interrupt is executed.
• Bit 4 – OCF1A: Output Compare Flag 1A
The OCF1A bit is set (one) when compare match occurs between the Timer/Counter1
and the data in OCR1A – Output Compare Register 1A. OCF1A is cleared by hardware
when executing the corresponding interrupt handling vector. Alternatively, OCF1A is
cleared by writing a logical “1” to the flag. When the I-bit in SREG and OCIE1A
(Timer/Counter1 Compare Interrupt Enable) and the OCF1A are set (one), the
Timer/Counter1 CompareA Match interrupt is executed.
• Bit 3 – OCF1B: Output Compare Flag 1B
The OCF1B bit is set (one) when compare match occurs between the Timer/Counter1
and the data in OCR1B – Output Compare Register 1B. OCF1B is cleared by hardware
when executing the corresponding interrupt handling vector. Alternatively, OCF1B is
cleared by writing a logical “1” to the flag. When the I-bit in SREG and OCIE1B
(Timer/Counter1 Compare Match Interrupt Enable) and the OCF1B are set (one), the
Timer/Counter1 CompareB Match interrupt is executed.
• Bit 2 – TOV1: Timer/Counter1 Overflow Flag
The TOV1 is set (one) when an overflow occurs in Timer/Counter1. TOV1 is cleared by
hardware when executing the corresponding interrupt handling vector. Alternatively,
TOV1 is cleared by writing a logical “1” to the flag. When the I-bit in SREG and TOIE1
(Timer/Counter1 Overflow Interrupt Enable) and TOV1 are set (one), the
Timer/Counter1 Overflow interrupt is executed. In PWM mode, this bit is set when
Timer/Counter1 advances from $0000.
32
ATmega103(L)
0945G–09/01
ATmega103(L)
• Bit 1 – OCF0: Output Compare Flag 0
The OCF0 bit is set (one) when compare match occurs between Timer/Counter0 and
the data in OCR0 – Output Compare Register 0. OCF0 is cleared by hardware when
executing the corresponding interrupt handling vector. Alternatively, OCF0 is cleared by
writing a logical “1” to the flag. When the I-bit in SREG and OCIE0 (Timer/Counter2
Compare Interrupt Enable) and the OCF0 are set (one), the Timer/Counter0 Output
Compare interrupt is executed.
• Bit 0 – TOV0: Timer/Counter0 Overflow Flag
The bit TOV0 is set (one) when an overflow occurs in Timer/Counter0. TOV0 is cleared
by hardware when executing the corresponding interrupt handling vector. Alternatively,
TOV0 is cleared by writing a logical “1” to the flag. When the SREG I-bit and TOIE0
(Timer/Counter0 Overflow Interrupt Enable) and TOV0 are set (one), the
Timer/Counter0 Overflow interrupt is executed. In PWM mode, this bit is set when
Timer/Counter0 advances from $00.
Interrupt Response TimeThe interrupt execution response for all the enabled AVR interrupts is four clock cycles
minimum. Four clock cycles after the interrupt flag has been set, the program vector
address for the actual interrupt handling routine is executed. During this four-clock-cycle
period, the Program Counter (2 bytes) is pushed onto the Stack, and the Stack Pointer
is decremented by 2. The vector is normally a jump to the interrupt routine, and this
jump takes three clock cycles. If an interrupt occurs during execution of a multi-cycle
instruction, this instruction is completed before the interrupt is served.
A return from an interrupt handling routine (same as for a subroutine call routine) takes
four clock cycles. During these four clock cycles, the Program Counter (2 bytes) is
popped back from the stack, and the Stack Pointer is incremented by 2. When the AVR
exits from an interrupt, it will always return to the main program and execute one more
instruction before any pending interrupt is served.
Sleep ModesTo enter any of the three sleep modes, the SE bit in MCUCR must be set (one) and a
SLEEP instruction must be executed. The SM1 and SM0 bits in the MCUCR register
select which sleep mode (Idle, Power-down, or Power-save) will be activated by the
SLEEP instruction, see Table 3 on page 22.
If an enabled interrupt occurs while the MCU is in a sleep mode, the MCU awakes, executes the interrupt routine and resumes execution from the instruction following SLEEP.
The contents of the register file, SRAM, and I/O memory are unaltered. If a reset occurs
during Sleep Mode, the MCU wakes up and executes from the Reset vector.
Idle ModeWhen the SM1/SM0 bits are set to 00, the SLEEP instruction makes the MCU enter the
Idle mode, stopping the CPU but allowing SPI, UART, Analog Comparator, ADC,
Timer/Counters, Watchdog and the interrupt system to continue operating. This enables
the MCU to wake up from external triggered interrupts as well as internal ones like the
Timer Overflow and UART Receive Complete interrupts. If wake-up from the Analog
Comparator interrupt is not required, the Analog Comparator can be powered down by
setting the ACD-bit in the Analog Comparator Control and Status Register (ACSR). This
will reduce power consumption in Idle mode. When the MCU wakes up from Idle mode,
the CPU starts program execution immediately.
Power-down ModeWhen the SM1/SM0 bits are set to 10, the SLEEP instruction makes the MCU enter the
Power-down mode. In this mode, the external oscillator is stopped while the external
interrupts and the Watchdog (if enabled) continue operating. Only an external reset, a
Watchdog reset (if enabled), or an external level interrupt can wake up the MCU.
33
0945G–09/01
Note that if a level-triggered interrupt is used for wake-up from Power-down mode, the
changed level must be held for some time to wake up the MCU. This makes the MCU
less sensitive to noise. The changed level is sampled twice by the Watchdog oscillator
clock and if the input has the required level during this time, the MCU will wake up. The
period of the Watchdog oscillator is 1 µs (nominal) at 5.0V and 25
the Watchdog oscillator is voltage-dependent, as shown in “Typical Characteristics” on
page 118.
When waking up from Power-down mode, a delay from the wake-up condition occurs
until the wake-up becomes effective. This allows the clock to restart and become stable
after having been stopped. The wake-up period is defined by the same SUT fuses that
define the reset time-out period. The wake-up period is equal to the clock reset period,
as shown in Table 5 on page 26.
If the wake-up condition disappears before the MCU wakes up and starts to execute,
e.g., a low-level on is not held long enough, the interrupt causing the wake-up will not be
executed.
Power-save ModeWhen the SM1/SM0 bits are 11, the SLEEP instruction makes the MCU enter the
Power-save mode. This mode is identical to Power-down, with one exception:
If Timer/Counter0 is clocked asynchronously, i.e., the AS0 bit in ASSR is set,
Timer/Counter0 will run during sleep. In addition to the Power-down wake-up sources,
the device can also wake up from either Timer Overflow or Output Compare event from
Timer/Counter0 if the corresponding Timer/Counter0 interrupt enable bits are set in
TIMSK. To ensure that the part executes the interrupt routine when waking up, also set
the global interrupt enable bit i SREG.
°C. The frequency of
When waking up from Power-save mode by an external interrupt, two instruction cycles
are executed before the interrupt flags are updated. When waking up by the asynchronous timer, three instruction cycles are executed before the flags are updated. During
these cycles, the processor executes instructions, but the interrupt condition is not readable and the interrupt routine has not started yet. If the asynchronous timer is not
clocked asynchronously, Power-down mode is recommended instead of Power-save
mode because the contents of the registers in the asynchronous timer should be considered undefined after wake-up in Power-save mode if AS0 is 0.
34
ATmega103(L)
0945G–09/01
ATmega103(L)
Timer/CountersThe ATmega103(L) provides three general-purpose Timer/Counters – two 8-bit T/Cs
and one 16-bit T/C. Timer/Counter0 optionally can be asynchronously clocked from an
external oscillator. This oscillator is optimized for use with a 32.768 kHz crystal, enabling
use of Timer/Counter0 as a Real-time Clock (RTC). Timer/Counter0 has its own prescaler. Timer/Counters 1 and 2 have individual prescaling selection from the same 10-bit
prescaling timer. These Timer/Counters can either be used as a timer with an internal
clock time base or as a counter with an external pin connection that triggers the
counting.
Timer/Counter
Prescalers
Figure 28. Prescaler for Timer/Counter1 and Timer/Counter2
CK
T1
T2
CS20
CS21
CS22
0
TIMER/COUNTER2 CLOCK SOURCE
TCK2
10-BIT T/C PRESCALER
CK/8
CS10
CS11
CS12
CK/64
CK/256
0
TIMER/COUNTER1 CLOCK SOURCE
CK/1024
TCK1
For Timer/Counters 1 and 2, the four different prescaled selections are: CK/8, CK/64,
CK/256 and CK/1024, where CK is the CPU clock. Observe that CPU clock frequency
can be lower than the XTAL frequency if the XTAL divider is enabled. For
Timer/Counters 1 and 2, added selections as CK, external source and stop can be
selected as clock sources.
0945G–09/01
Figure 29. The Timer/Counter0 Prescaler
CK
TOSC1
AS0
CS00
CS01
CS02
PCK0
10-BIT T/C PRESCALER
PCK0/8
TIMER/COUNTER0 CLOCK SOURCE
PCK0
PCK0/32
PCK0/64
PCK0/128
PCK0/256
PCK0/1024
35
The clock source for Timer/Counter0 prescaler is named PCK0. PCK0 is by default connected to the main system clock CK. Observe that CPU clock frequency can be lower
than the XTAL frequency if the XTAL divider is enabled. By setting the AS0 bit in ASSR,
Timer/Counter0 prescaler is asynchronously clocked from the TOSC1 pin. This enables
use of Timer/Counter0 as a Real-time Clock (RTC). A crystal can be connected between
the TOSC1 and TOSC2 pins to serve as an independent clock source for
Timer/Counter0. This oscillator is optimized for use with a 32.768 kHz crystal.
8-bit Timer/Counters
T/C0 and T/C2
Figure 30 shows the block diagram for Timer/Counter0. Figure 31 shows the block diagram for Timer/Counter2.
Figure 30. Timer/Counter0 Block Diagram
8-BIT DATA BUS
8-BIT ASYNCH T/C0 DATA BUS
OCIE1B
OCIE2
7
TIMER/COUNTER0
7
8-BIT COMPARATOR
7
OUTPUT COMPARE
REGISTER0 (OCR0)
OCIE1A
TOIE2
TICIE1
TIMER INT. MASK
REGISTER (TIMSK)
(TCNT0)
TOIE1
OCIE0
TOIE0
0
0
0
TCK0
CK
T/C0 OVER-
FLOW IRQ
T/C0 COMPARE
MATCH IRQ
TIMER INT. FLAG
REGISTER (TIFR)
ICF1
TOV2
OCF2
OCF2B
OCF2A
T/C CLEAR
T/C CLK SOURCE
UP/DOWN
SYNCH UNIT
OCF0
TOV0
TOV0
TOV1
OCF0
T/C0 CONTROL
REGISTER (TCCR0)
PWM0
COM01
COM00
CONTROL
LOGIC
ASYNCH. STATUS
REGISTER (ASSR)
CTC0
AS0
CS02
TC0UB
CS01
CS00
ICR0UB
OCR0UB
PCK0
36
ATmega103(L)
0945G–09/01
Figure 31. Timer/Counter2 Block Diagram
T/C2 OVER-
FLOW IRQ
OCF2
OCIE1A
OCIE1B
TICIE1
TOIE2
TA BUS
8-BIT DA
OCIE2
TIMER INT. MASK
REGISTER (TIMSK)
7
TIMER/COUNTER2
(TCNT2)
7
8-BIT COMPARATOR
TOIE0
OCIE0
TOIE1
0
0
TOV2
TIMER INT. FLAG
REGISTER (TIFR)
TOV2
OCF2
T/C CLEAR
T/C CLK SOURCE
UP/DOWN
T/C2 COMPARE
MATCH IRQ
ICF1
TOV1
OCF2A
OCF2B
OCF0
ATmega103(L)
T/C2 CONTROL
REGISTER (TCCR2)
CS22
CS21
CS20
CTC2
PWM2
COM21
TOV0
CONTROL
LOGIC
COM20
CK
T2
7
OUTPUT COMPARE
REGISTER2 (OCR2)
0
The 8-bit Timer/Counter0 can select clock source from PCK0 or prescaled PCK0. The 8bit Timer/Counter2 can select clock source from CK, prescaled CK or an external pin.
Both Timer/Counters can be stopped as described in the specification for the
Timer/Counter Control Registers – TCCR0 and TCCR2.
The different status flags (overflow, compare match and capture event) are found in the
Timer/Counter Interrupt Flag Register (TIFR). Control signals are found in the
Timer/Counter Control Registers – TCCR0 and TCCR2. The interrupt enable/disable
settings are found in the Timer/Counter Interrupt Mask Register (TIMSK).
When Timer/Counter2 is externally clocked, the external signal is synchronized with the
oscillator frequency of the CPU. To assure proper sampling of the external clock, the
minimum time between two external clock transitions must be at least one internal CPU
clock period. The external clock signal is sampled on the rising edge of the internal CPU
clock.
The 8-bit Timer/Counters feature a high-resolution and a high-accuracy usage with the
lower prescaling opportunities. Similarly, the high prescaling opportunities make these
units useful for lower speed functions or exact timing functions with infrequent actions.
Both Timer/Counters support two Output Compare functions using the output compare
registers (OCR0 and OCR2) as the data source to be compared to the Timer/Counter
contents. The Output Compare functions include optional clearing of the counter on
compare match and action on the output compare pins – PB4(OC0/PWM0) and
PB7(OC2/PWM2) – on compare match.
0945G–09/01
Timer/Counters 0 and 2 can also be used as 8-bit Pulse Width Modulators. In this mode
the Timer/Counter and the output compare register serve as a glitch-free, stand-alone
PWM with centered pulses. Refer to page 40 for a detailed description of this function.
37
Timer/Counter0 Control
Register – TCCR0
Timer/Counter2 Control
Register – TCCR2
Bit76543210
33 ($53)–PWM0COM01COM00CTC0CS02CS01CS00TCCR0
Read/WriteRR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
Bit76543210
$25 ($45)–PWM2COM21COM20CTC2CS22CS21CS20TCCR2
Read/WriteRR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
• Bit 7 – Res: Reserved Bit
This bit is a reserved bit in the ATmega103(L) and always reads as zero.
• Bit 6 – PWM0/PWM2: Pulse Width Modulator Enable
When set (one), this bit enables PWM mode for Timer/Counter0 or Timer/Counter2.
This mode is described on page 40.
The COMn1 and COMn0 control bits determine any output pin action following a compare match in Timer/Counter2. Any output pin actions affect pins PB4 (OC0/PWM0) or
PB7 (OC2/PWM2). Since this is an alternative function to an I/O port, the corresponding
direction control bit must be set (one) to control an output pin. The control configuration
is shown in Table 10.
Table 10. Compare Mode Select
COMn1COMn0Description
00Timer/Counter disconnected from output pin OCn/PWMn
01Toggle the OCn/PWMn output line.
10Clear the OCn/PWMn output line (to zero).
11Set the OCn/PWMn output line (to one).
Note:n = 0 or 2
In PWM mode, these bits have a different function. Refer to Table 13 for a detailed
description.
• Bit 3 – CTC0/CTC2: Clear Timer/Counter on Compare Match
When the CTC0 or CTC2 control bit is set (one), the Timer/Counter is reset to $00 in the
CPU clock cycle after a compare match. If the control bit is cleared, the timer continues
counting and is unaffected by a compare match. Since the compare match is detected in
the CPU clock cycle following the match, this function will behave differently when a
prescaling higher than 1 is used for the timer. When a prescaling of 1 is used and the
compare register is set to C, the timer will count as follows if CTC0/2 is set:
... | C-2 | C-1 | C | 0 | 1 | ...
When the prescaler is set to divide by 8, the timer will count like this:
The Clock Select2 bits 2, 1 and 0 define the prescaling source of the Timer/Counter.
Table 11. Timer/Counter0 Prescale Select
CS02CS01CS00Description
000Timer/Counter0 is stopped.
001PCK0
010PCK0/8
011PCK0/32
100PCK0/64
101PCK0/128
110PCK0/256
111PCK0/1024
Table 12. Timer/Counter2 Prescale Select
Timer/Counter0 – TCNT0
CS22CS21CS20Description
000Timer/Counter2 is stopped.
001CK
010CK/8
011CK/64
100CK/256
101CK/1024
110External Pin PD7(T2), falling edge
111External Pin PD7(T2), rising edge
The Stop condition provides a Timer Enable/Disable function. The CK down divided
modes are scaled directly from the CK CPU clock. If the external pin modes are used for
Timer/Counter2, transitions on PD7/(T2) will clock the counter even if the pin is configured as an output. This feature can give the user software control of the counting.
Bit76543210
$32 ($42)MSBLSBTCNT0
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
Timer/Counter2 – TCNT2
0945G–09/01
Bit76543210
$24 ($44)MSBLSBTCNT2
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
These 8-bit registers contain the value of the Timer/Counters.
39
Timer/Counter0 Output
Compare Register – OCR0
Timer/Counter2 Output
Compare Register – OCR2
Both Timer/Counters are realized as up or up/down (in PWM mode) counters with read
and write access. If the Timer/Counter is written to and a clock source is selected, it continues counting in the timer clock cycle after it is preset with the written value.
Bit76543210
$31 ($51)MSBLSBOCR0
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
Bit76543210
$23 ($43)MSBLSBOCR2
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
The Output Compare registers are 8-bit read/write registers.
The Timer/Counter Output Compare Registers contain the data to be continuously compared with the Timer/Counter. Actions on compare matches are specified in TCCR0 and
TCCR2. A compare match does only occur if the Timer/Counter counts to the OCR
value. A software write that sets the Timer/Counter and Output Compare Register to the
same value does not generate a compare match.
A compare match will set the compare interrupt flag in the CPU clock cycle following the
compare event.
Timer/Counters 0 and 2 in
PWM Mode
When the PWM mode is selected, the Timer/Counter and the Output Compare Register
(OCR0 or OCR2) form an 8-bit, free-running, glitch-free and phase correct PWM with
outputs on the PB4(OC0/PWM0) or PB7(OC2/PWM2) pin. The Timer/Counter acts as
an up/down counter, counting up from $00 to $FF, where it turns and counts down again
to zero before the cycle is repeated. When the counter value matches the contents of
the Output Compare register, the PB4(OC0/PWM0) or PB7(OC2/PWM2) pin is set or
cleared according to the settings of the COM01/COM00 or COM21/COM20 bits in the
Timer/Counter Control registers TCCR0 and TCCR2. Refer to Table 13 for details.
Table 13. Compare Mode Select in PWM Mode
COMn1COMn0Effect on Compare/PWM Pin
00Not connected
01Not connected
10
11
Note:n = 0 or 2
Cleared on compare match, up-counting. Set on compare match, downcounting (non-inverted PWM).
Cleared on compare match, down-counting. Set on compare match, upcounting (inverted PWM).
Note that in PWM mode, the Output Compare register is transferred to a temporary
location when written. The value is latched when the Timer/Counter reaches $FF. This
prevents the occurrence of odd-length PWM pulses (glitches) in the event of an unsynchronized OCR0 or OCR2 write. See Figure 32 for an example.
40
ATmega103(L)
0945G–09/01
ATmega103(L)
e
e
Figure 32. Effects on Unsynchronized OCR Latching
Compare Value changes
Counter Value
Compare Valu
PWM Output
Synchronized OCR Latch
Compare Value changes
Counter Value
Compare Valu
PWM Output
Unsynchronized OCR Latch
During the time between the write and the latch operation, a read from OCR0 or OCR2
will read the contents of the temporary location. This means that the most recently written value always will read out of OCR0/2.
When the OCR register (not the temporary register) is updated to $00 or $FF, the PWM
output changes to low or high immediately according to the settings of COM21/COM20
or COM11/COM10. This is shown in Table 14.
Glitch
Asynchronous Status
Register – ASSR
Table 14. PWM Outputs OCRn = $00 or $FF
COMn1COMn0OCRnOutput PWMn
10$00L
10$FFH
11$00H
11$FFL
Note:n = 0 or 2
In PWM mode, the Timer Overflow flag, TOV0 or TOV2, is set when the counter
advances from $00. Timer Overflow Interrupts 0 and 2 operate exactly as in normal
Timer/Counter mode, i.e., it is executed when TOV0 or TOV2 is set, provided that Timer
Overflow interrupt and global interrupts are enabled. This also applies to the Timer Output Compare flags and interrupts.
The frequency of the PWM will be Timer Clock Frequency divided by 510.
Bit76543 2 1 0
$30 ($50)
Read/WriteRRRRR/WRRR
Initial Value00000000
––––AS0TCN0UBOCR0UBTCR0UBASSR
• Bits 7..4 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and always read as zero.
0945G–09/01
41
• Bit 3 – AS0: Asynchronous Timer/Counter0
When set (one), Timer/Counter0 is clocked from the TOSC1 pin. When cleared (zero),
Timer/Counter0 is clocked from the internal system clock, CK. When the value of this bit
is changed, the contents of TCNT0 might get corrupted.
• Bit 2 – TCN0UB: Timer/Counter0 Update Busy
When Timer/Counter0 operates asynchronously and TCNT0 is written, this bit becomes
set (one). When TCNT0 has been updated from the temporary storage register, this bit
is cleared (zero) by hardware. A logical “0” in this bit indicates that TCNT0 is ready to be
updated with a new value.
• Bit 1 – OCR0UB: Output Compare Register0 Update Busy
When Timer/Counter0 operates asynchronously and OCR0 is written, this bit becomes
set (one). When OCR0 has been updated from the temporary storage register, this bit is
cleared (zero) by hardware. A logical “0” in this bit indicates that OCR0 is ready to be
updated with a new value.
• Bit 0 – TCR0UB: Timer/Counter Control Register0 Update Busy
When Timer/Counter0 operates asynchronously and TCCR0 is written, this bit becomes
set (one). When TCCR0 has been updated from the temporary storage register, this bit
is cleared (zero) by hardware. A logical “0” in this bit indicates that TCCR0 is ready to be
updated with a new value.
If a write is performed to any of the three Timer/Counter0 registers while its update busy
flag is set (one), the updated value might get corrupted and cause an unintentional interrupt to occur.
Asynchronous Operation of
Timer/Counter0
When reading TCNT0, OCR0 and TCCR0, there is a difference in result. When reading
TCNT0, the actual timer value is read. When reading OCR0 or TCCR0, the value in the
temporary storage register is read.
When Timer/Counter0 operates synchronously, all operations and timing are identical to
Timer/Counter2. During asynchronous operation, however, some considerations must
be taken.
•WARNING: When switching between asynchronous and synchronous clocking of
Timer/Counter0, the timer registers TCNT0, OCR0 and TCCR0 might get corrupted.
The following is the safe procedure for switching clock source:
1. Disable the timer 0 interrupts OCIE0 and TOIE0.
2. Select clock source by setting ASO as appropriate.
3. Write new values to TCNT0, OCR0 and TCCR0.
4. If switching to asynchronous operation, wait for TCNT0UB, OCR0UB and
TCR0UB to be cleared.
5. Clear the Timer/Counter0 interrupt flags.
6. Enable interrupts if needed.
•The oscillator is optimized for use with a 32,768 Hz watch crystal. An external clock
signal applied to this pin goes through the same amplifier having a bandwidth of
256 kHz. The external clock signal should therefore be in the interval 0 Hz 256 kHz. The frequency of the clock signal applied to the TOSC1 pin must be lower
than one fourth of the CPU main clock frequency. Observe that CPU clock
frequency can be lower than the XTAL frequency if the XTAL divider is enabled.
•When writing to one of the registers TCNT0, OCR0 or TCCR0, the value is
transferred to a temporary register and latched after two positive edges on TOSC1.
The user should not write a new value before the contents of the temporary register
42
ATmega103(L)
0945G–09/01
ATmega103(L)
have been transferred to its destination. Each of the three mentioned registers have
their individual temporary register, which means that e.g., writing to TCNT0 does not
disturb an OCR0 write in progress. To detect that a transfer to the destination
register has taken place, an Asynchronous Status Register (ASSR) has been
implemented.
•When entering Power-save mode after having written to TCNT0, OCR0 or TCCR0,
the user must wait until the written register has been updated if Timer/Counter0 is
used to wake up the device. Otherwise, the MCU will go to sleep before the
changes have had any effect. This is extremely important if the Output Compare0
interrupt is used to wake up the device; Output Compare is disabled during write to
OCR0 or TCNT0. If the write cycle is not finished (i.e., the user goes to sleep before
the OCR0UB bit returns to zero), the device will never get a compare match and the
MCU will not wake up.
•If Timer/Counter0 is used to wake up the device from Power-save mode,
precautions must be taken if the user wants to reenter Power-save mode: The
interrupt logic needs one TOSC1 cycle to get reset. If the time between wake-up
and reentering Power-save mode is less than one TOSC1 cycle, the interrupt will
not occur and the device will fail to wake up. If the user is in doubt whether the time
before re-entering Power-save is sufficient, the following algorithm can be used to
ensure that one TOSC1 cycle has elapsed:
1. Write a value to TCCR0, TCNT0 or OCR0.
2. Wait until the corresponding Update Busy flag in ASSR returns to zero.
3. Enter Power-save mode.
•When asynchronous operation is selected, the 32 kHz oscillator for Timer/Counter0
is always running, except in Power-down mode. After a power-up reset or wake-up
from power-down, the user should be aware of the fact that this oscillator might take
as long as one second to stabilize. The user is advised to wait for at least one
second before using Timer/Counter0 after power-up or wake-up from Power-down.
The content of all Timer/Counter0 registers must be considered lost after a wake-up
from Power-down due to the unstable clock signal upon start-up, no matter whether
the oscillator is in use or a clock signal is applied to the TOSC pin.
•Description of wake-up from Power-save mode when the timer is clocked
asynchronously: When the interrupt condition is met, the wake-up process is started
on the following cycle of the timer clock, that is, the timer is always advanced by at
least one before the processor can read the counter value. To execute the
corresponding Timer/Counter0 interrupt routine, the global interrupt bit in SREG
must have been set. Otherwise, the part will still wake up from Power-down, but
continues to execute the Sleep command. The interrupt flags are updated three
processor cycles after the processor clock has started. During these cycles, the
processor executes instructions, but the interrupt condition is not readable and the
interrupt routine has not started yet.
•During asynchronous operation, the synchronization of the interrupt flags for the
asynchronous timer takes three processor cycles plus one timer cycle. The timer is
therefore advanced by at least one before the processor can read the timer value
causing the setting of the interrupt flag. The output compare pin is changed on the
timer clock, and is not synchronized to the processor clock.
•After waking up from Power-save mode with the asynchronous timer enabled, there
will be a short interval of which TCNT0 will read as the same value as before Powersave mode was entered. After an edge on the asynchronous clock, TCNT0 will read
correctly. (The compare and overflow functions of the timer are not affected by this
behavior.) Safe procedure to ensure correct value is read:
0945G–09/01
43
1. Write any value to either of the registers OCR0 or TCCR0
2. Wait for the corresponding Update Busy Flag to be cleared
3. Read TCNT0
Note that OCR0 and TCCR0 are never modified by hardware, and will always read
correctly.
16-bit Timer/Counter1Figure 33 shows the block diagram for Timer/Counter1.
The 16-bit Timer/Counter1 can select clock source from CK, prescaled CK or an external pin. In addition, it can be stopped as described in the specification for the
Timer/Counter1 Control Register (TCCR1B). The different status flags (overflow, compare match and capture event) are found in the Timer/Counter Interrupt Flag Register
(TIFR). Control signals are found in the Timer/Counter1 Control Registers – TCCR1A
and TCCR1B. The interrupt enable/disable settings for Timer/Counter1 are found in the
Timer/Counter Interrupt Mask Register (TIMSK).
When Timer/Counter1 is externally clocked, the external signal is synchronized with the
oscillator frequency of the CPU. To assure proper sampling of the external clock, the
minimum time between two external clock transitions must be at least one internal CPU
clock period. The external clock signal is sampled on the rising edge of the internal CPU
clock.
The 16-bit Timer/Counter1 features both a high-resolution and a high-accuracy usage
with the lower prescaling opportunities. Similarly, the high prescaling opportunities
makes the Timer/Counter1 useful for lower speed functions or exact timing functions
with infrequent actions.
The Timer/Counter1 supports two Output Compare functions using the Output Compare
registers 1A and 1B (OCR1A and OCR1B) as the data sources to be compared to the
Timer/Counter1 contents. The Output Compare functions include optional clearing of
the counter on compareA match, and actions on the Output Compare pins on both compare matches.
44
ATmega103(L)
0945G–09/01
Figure 33. Timer/Counter1 Block Diagram
TA BUS
8-BIT DA
OCIE0
TOIE0
T/C1 COMPARE
MATCHA IRQ
OCF2
7
8
T/C1 OVER-
FLOW IRQ
TOIE1
OCIE1A
OCIE1B
TICIE1
TOIE2
OCIE2
TIMER INT. MASK
REGISTER (TIMSK)
15
T/C1 INPUT CAPTURE REGISTER (ICR1)
TOV2
TIMER INT. FLAG
REGISTER (TIFR)
ICF1
ICF1
OCF1B
OCF1B
T/C1 COMPARE
MATCHB IRQ
TOV0
OCF0
OCF1A
TOV1
TOV1
OCF1A
CAPTURE
TRIGGER
T/C1 INPUT
CAPTURE IRQ
T/C1 CONTROL
REGISTER A (TCCR1A)
COM1B1
COM1A1
COM1A0
0
COM1B0
ATmega103(L)
T/C1 CONTROL
REGISTER B (TCCR1B)
CS11
CS12
CS10
PWM11
PWM10
CONTROL
LOGIC
ICNC1
ICES1
CTC1
CK
T1
15
TIMER/COUNTER1 (TCNT1)
15
15
TIMER/COUNTER1 OUTPUT COMPARE REGISTER A
7
8
7
8
16-BIT COMPARATOR
7
8
0
T/C CLEAR
T/C CLOCK SOURCE
UP/DOWN
7
15
0
15
0
TIMER/COUNTER1 OUTPUT COMPARE REGISTER B
8
16-BIT COMPARATOR
7
8
0
0
Timer/Counter1 can also be used as an 8-, 9- or 10-bit Pulse Width Modulator. In this
mode the counter and the OCR1A/OCR1B registers serve as a dual glitch-free standalone PWM with centered pulses. Refer to page 50 for a detailed description of this
function.
The Input Capture function of Timer/Counter1 provides a capture of the Timer/Counter1
contents to the Input Capture Register (ICR1), triggered by an external event on the
Input Capture pin – PD4/(IC1). The actual capture event settings are defined by the
Timer/Counter1 Control Register (TCCR1B). In addition, the Analog Comparator can be
set to trigger the input capture. Refer to “Analog Comparator” on page 70 for details on
this. The ICP pin logic is shown in Figure 34.
0945G–09/01
Figure 34. ICP Pin Schematic Diagram
45
Timer/Counter1 Control
Register A – TCCR1A
If the Noise Canceler function is enabled, the actual trigger condition for the capture
event is monitored over four samples, and all four must be equal to activate the capture
flag.
The COM1A1 and COM1A0 control bits determine any output pin action following a
compare match in Timer/Counter1. Any output pin actions affect pin OC1A – Output
CompareA pin 1. This is an alternative function to an I/O port, and the corresponding
direction control bit must be set (one) to control an output pin. The control configuration
is shown in Table 15
The COM1B1 and COM1B0 control bits determine any output pin action following a
compare match in Timer/Counter1. Any output pin actions affect pin OC1B – Output
CompareB. Since this is an alternative function to an I/O port, the corresponding direction control bit must be set (one) to control an output pin. The control configuration is
given in Table 15.
Table 15. Compare1 Mode Select
COM1X1COM1X0Description
00Timer/Counter1 disconnected from output pin OC1X
01Toggle the OC1X output line.
10Clear the OC1X output line (to zero).
11Set the OC1X output line (to one).
Note:X = A or B.
In PWM mode, these bits have a different function. Refer to Table 16 for a detailed
description.
• Bits 3..2 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and always read as zero.
When the ICNC1 bit is cleared (zero), the Input Capture Trigger Noise Canceler function
is disabled. The input capture is triggered at the first rising/falling edge sampled on the
input capture pin PD4(IC1) as specified. When the ICNC1 bit is set (one), four successive samples are measured on PD4(IC1), and all samples must be high/low according to
the input capture trigger specification in the ICES1 bit. The actual sampling frequency is
XTAL clock frequency.
• Bit 6 – ICES1: Input Capture1 Edge Select
While the ICES1 bit is cleared (zero), the Timer/Counter1 contents are transferred to the
Input Capture Register (ICR1) on the falling edge of the input capture pin – PD4(IC1).
While the ICES1 bit is set (one), the Timer/Counter1 contents are transferred to the
Input Capture Register on the rising edge of the input capture pin – PD4(IC1).
• Bits 5, 4 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and always read as zero.
• Bit 3 – CTC1: Clear Timer/Counter1 on Compare Match
When the CTC1 control bit is set (one), the Timer/Counter1 is reset to $0000 in the clock
cycle after a compareA match. If the CTC1 control bit is cleared, Timer/Counter1 continues counting and is unaffected by a compare match. Since the compare match is
detected in the CPU clock cycle following the match, this function will behave differently
when a prescaling higher than 1 is used for the timer. When a prescaling of 1 is used
and the compareA register is set to C, the timer will count as follows if CTC1 is set:
... | C-2 | C-1 | C | 0 | 1 | ...
When the prescaler is set to divide by 8, the timer will count like this:
The lock Select1 bits 2, 1 and 0 define the prescaling source of Timer/Counter1.
Table 17. Clock1 Prescale Select
CS12CS11CS10Description
000Stop, the Timer/Counter1 is stopped.
001CK
010CK/8
0945G–09/01
011CK/64
100CK/256
101CK/1024
110External Pin T1, falling edge
111External Pin T1, rising edge
47
Timer/Counter1 – TCNT1H and
TCNT1L
The Stop condition provides a Timer Enable/Disable function. The CK down divided
modes are scaled directly from the CK CPU clock. If the external pin modes are used for
Timer/Counter1, transitions on PD6/(T1) will clock the counter even if the pin is configured as an output. This feature can give the user software control of the counting.
Bit151413121110 9 8
$2D ($4D)MSBTCNT1H
$2C ($4C)LSBTCNT1L
76543210
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
R/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
00000000
This 16-bit register contains the prescaled value of the 16-bit Timer/Counter1. To
ensure that both the high and low bytes are read and written simultaneously when the
CPU accesses these registers, the access is performed using an 8-bit temporary register (TEMP). This temporary register is also used when accessing OCR1A, OCR1B and
ICR1. If the main program and interrupt routines perform access to registers using
TEMP, interrupts must be disabled during access from the main program (and from
interrupt routines if interrupts are allowed from within interrupt routines).
•TCNT1 Timer/Counter1 Write:
When the CPU writes to the high byte TCNT1H, the written data is placed in the
TEMP register. Next, when the CPU writes the low byte TCNT1L, this byte of data is
combined with the byte data in the TEMP register, and all 16 bits are written to the
TCNT1 Timer/Counter1 register simultaneously. Consequently, the high byte
TCNT1H must be accessed first for a full 16-bit register write operation. When using
Timer/Counter1 as an 8-bit timer, it is sufficient to write the low byte only.
•TCNT1 Timer/Counter1 Read:
When the CPU reads the low byte TCNT1L, the data of TCNT1L is sent to the CPU
and the data of the high byte TCNT1H is placed in the TEMP register. When the
CPU reads the data in the high byte TCNT1H, the CPU receives the data in the
TEMP register. Consequently, the low byte TCNT1L must be accessed first for a full
16-bit register read operation. When using Timer/Counter1 as an 8-bit timer, it is
sufficient to read the low byte only.
Timer/Counter1 Output
Compare Register – OCR1AH
and OCR1AL
48
ATmega103(L)
The Timer/Counter1 is realized as an up or up/down (in PWM mode) counter with read
and write access. If Timer/Counter1 is written to and a clock source is selected, the
Timer/Counter1 continues counting in the clock cycle after it is preset with the written
value.
Bit151413121110 9 8
$2BMSBOCR1AH
$2ALSBOCR1AL
76543210
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
R/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
00000000
0945G–09/01
Timer/Counter1 Output
Compare Register – OCR1BH
and OCR1BL
ATmega103(L)
Bit151413121110 9 8
$29MSBOCR1BH
$28LSBOCR1BL
76543210
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
R/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
00000000
The Output Compare registers are 16-bit read/write registers.
The Timer/Counter1 Output Compare registers contain the data to be continuously compared with Timer/Counter1. Actions on compare matches are specified in the
Timer/Counter1 Control and Status registers. A compare match occurs only if
Timer/Counter1 counts to the OCR value. A software write that sets TCNT1 and OCR1A
or OCR1B to the same value does not generate a compare match.
A compare match will set the compare interrupt flag in the CPU clock cycle following the
compare event.
Since the Output Compare registers (OCR1A and OCR1B) are 16-bit registers, a temporary register TEMP is used when OCR1A/B are written to ensure that both bytes are
updated simultaneously. When the CPU writes the high byte, OCR1AH or OCR1BH, the
data is temporarily stored in the TEMP register. When the CPU writes the low byte,
OCR1AL or OCR1BL, the TEMP register is simultaneously written to OCR1AH or
OCR1BH. Consequently, the high byte OCR1AH or OCR1BH must be written first for a
full 16-bit register write operation.
Timer/Counter1 Input Capture
Register – ICR1H and ICR1L
The TEMP register is also used when accessing TCNT1 and ICR1. If the main program
and interrupt routines perform access to registers using TEMP, interrupts must be disabled during access from the main program.
Bit151413121110 9 8
$27 ($37)MSBICR1H
$26 ($36)LSBICR1L
76543210
Read/WriteRRRRRRRR
RRRRRRRR
Initial Value00000000
00000000
The Input Capture Register is a 16-bit read-only register.
When the rising or falling edge (according to the input capture edge setting (ICES1)) of
the signal at the input capture pin – PD4(IC1) – is detected, the current value of the
Timer/Counter1 is transferred to the Input Capture Register (ICR1). At the same time,
the input capture flag (ICF1) is set (one).
Since the Input Capture Register (ICR1) is a 16-bit register, a temporary register TEMP
is used when ICR1 is read to ensure that both bytes are read simultaneously. When the
CPU reads the low byte ICR1L, the data is sent to the CPU and the data of the high byte
ICR1H is placed in the TEMP register. When the CPU reads the data in the high byte
ICR1H, the CPU receives the data in the TEMP register. Consequently, the low byte
ICR1L must be accessed first for a full 16-bit register read operation.
0945G–09/01
49
The TEMP register is also used when accessing TCNT1, OCR1A and OCR1B. If the
main program and interrupt routines perform access to registers using TEMP, interrupts
must be disabled during access from the main program.
Timer/Counter1 in PWM ModeWhen the PWM mode is selected, Timer/Counter1, the Output Compare Register1A
(OCR1A) and the Output Compare Register1B (OCR1B) form a dual 8-, 9- or 10-bit,
free-running, glitch-free and phase-correct PWM with outputs on the PB5(OC1A) and
PB6(OC1B) pins. Timer/Counter1 acts as an up/down counter, counting up from $0000
to TOP (see Table 16), where it turns and counts down again to zero before the cycle is
repeated. When the counter value matches the contents of the 10 least significant bits of
OCR1A or OCR1B, the PB5(OC1A)/PB6(OC1B) pins are set or cleared according to the
settings of the COM1A1/COM1A0 or COM1B1/COM1B0 bits in the Timer/Counter1
Control Register, TCCR1A. Refer to Table 19 for details.
Table 18. Timer TOP Values and PWM Frequency
PWM ResolutionTimer TOP valueFrequency
8-bit$00FF (255)f
9-bit$01FF (511)f
10-bit$03FF (1023)f
TCK1
TCK1
TCK1
/510
/1022
/2046
Table 19. Compare1 Mode Select in PWM Mode
COM1X1COM1X0Effect on OCX1
00Not connected
01Not connected
10
11
Note:X = A or B
Cleared on compare match, up-counting. Set on compare match,
down-counting (non-inverted PWM).
Cleared on compare match, down-counting. Set on compare match,
up-counting (inverted PWM).
Note that in the PWM mode, the 10 least significant OCR1A/OCR1B bits, when written,
are transferred to a temporary location. They are latched when Timer/Counter1 reaches
the value TOP. This prevents the occurrence of odd-length PWM pulses (glitches) in the
event of an unsynchronized OCR1A/OCR1B write. See Figure 35 for an example.
50
ATmega103(L)
0945G–09/01
Figure 35. Effects on Unsynchronized OCR1 Latching
Compare V
SynchronizedOCR1X Latch
alue changes
ATmega103(L)
Counter
Compare V
PWM
alue
V
alue
Output OC1X
Compare Value changes
Note: X = A or B
UnsynchronizedOCR1X Latch
Glitch
Counter
Compare Value
PWM Output OC1X
Value
During the time between the write and the latch operation, a read from OCR1A or
OCR1B will read the contents of the temporary location. This means that the most
recently written value always will read out of OCR1A/B.
When the OCR1A/OCR1B contains $0000 or TOP, the output OC1A/OC1B is updated
to low or high on the next compare match according to the settings of
COM1A1/COM1A0 or COM1B1/COM1B0. This is shown in Table 20.
Note:If the compare register contains the TOP value and the prescaler is not in use
(CS12..CS10 = 001), the PWM output will not produce any pulse at all, because the upcounting and down-counting value is reached simultaneously. When the prescaler is in
use (CS12..CS10 ≠ 001 or 000), the PWM output goes active when the counter reaches
the TOP value, but the down-counting compare match is not interpreted to be reached
before the next time the counter reaches the TOP value, making a one-period PWM
pulse.
Table 20. PWM Outputs OCR1X = $0000 or TOP
0945G–09/01
COM1X1COM1X0OCR1XOutput OC1X
10$0000L
10TOP H
11$0000H
11TOP L
Note:X = A or B
In PWM mode, the Timer Overflow Flag1, TOV1, is set when the counter advances from
$0000. Timer Overflow Interrupt1 operates exactly as in normal Timer/Counter mode,
i.e., it is executed when TOV1 is set provided that Timer Overflow Interrupt1 and global
interrupts are enabled. This does also apply to the Timer Output Compare1 flags and
interrupts.
51
Watchdog TimerThe Watchdog Timer is clocked from a separate On-chip oscillator. By controlling the
Watchdog Timer prescaler, the Watchdog reset interval can be adjusted as shown in
Table 21. See characterization data for typical values at other V
(Watchdog Reset) instruction resets the Watchdog Timer. From the Watchdog reset,
eight different clock cycle periods can be selected to determine the reset period. If the
reset period expires without another Watchdog reset, the ATmega103(L) resets and
executes from the reset vector. For timing details on the Watchdog reset, refer to page
28.
To prevent unintentional disabling of the Watchdog, a special turn-off procedure must
be followed when the Watchdog is disabled. Refer to the description of the Watchdog
Timer Control Register for details.
Figure 36. Watchdog Timer
Oscillator
levels. The WDR
CC
Watchdog Timer Control
Register – WDTCR
1 MHz at V
350 kHz at V
Bit765 43210
$21 ($41)–––WDTOEWDEWDP2WDP1WDP0WDTCR
Read/WriteRRRR/WR/WR/WR/WR/W
Initial Value00000000
CC
CC
= 5V
= 3V
• Bits 7..5 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and will always read as zero.
• Bit 4 – WDTOE: Watchdog Turn-off Enable
52
This bit must be set (one) when the WDE bit is cleared, Otherwise, the Watchdog will
not be disabled. Once set, hardware will clear this bit to zero after four clock cycles.
Refer to the description of the WDE bit for a Watchdog disable procedure.
• Bit 3 – WDE: Watchdog Enable
When the WDE is set (one), the Watchdog Timer is enabled and if the WDE is cleared
(zero), the Watchdog Timer function is disabled. WDE can only be cleared if the
WDTOE bit is set (one). To disable an enabled Watchdog Timer, the following procedure must be followed:
ATmega103(L)
0945G–09/01
ATmega103(L)
1. In the same operation, write a logical “1” to WDTOE and WDE. A logical “1” must
be written to WDE even though it is set to one before the disable operation
starts.
2. Within the next four clock cycles, write a logical “0” to WDE. This disables the
Watchdog.
The WDP2, WDP1 and WDP0 bits determine the Watchdog Timer prescaling when the
Watchdog Timer is enabled. The different prescaling values and their corresponding
time-out periods are shown in Table 21.
Table 21. Watchdog Timer Prescale Select
Number of WDT
WDP2WDP1WDP0
00016K cycles47 ms15 ms
00132K cycles94 ms30 ms
01064K cycles0.19 s60 ms
011128K cycles0.38 s0.12 s
100256K cycles0.75 s0,24 s
101512K cycles1.5 s0.49 s
1101,024K cycles3.0 s0.97 s
1112,048K cycles6.0 s1.9 s
Note:The frequency of the Watchdog oscillator is voltage-dependent as shown in the Electrical
Characteristics section.
The WDR (Watchdog Reset) instruction should always be executed before the Watchdog
Timer is enabled. This ensures that the reset period will be in accordance with the
Watchdog Timer prescale settings. If the Watchdog Timer is enabled without reset, the
Watchdog Timer may not start counting from zero.
To avoid unintentional MCU reset, the Watchdog Timer should be disabled or reset
before changing the Watchdog Timer Prescale Select.
Oscillator Cycles
Typical Time-out
at VCC = 3.0V
Typical Time-out
at VCC = 5.0V
0945G–09/01
53
EEPROM Read/Write
Access
EEPROM Address Register –
EEARH, EEARL
The EEPROM access registers are accessible in the I/O space.
The write access time is in the range of 2.5 - 4 ms, depending on the V
voltages. A
CC
self-timing function lets the user software detect when the next byte can be written. A
special EEPROM Ready interrupt can be set to trigger when the EEPROM is ready to
accept new data.
In order to prevent unintentional EEPROM writes, a specific write procedure must be followed. Refer to the description of the EEPROM control register for details on this.
When the EEPROM is written, the CPU is halted for two clock cycles before the next
instruction is executed. When it is read, the CPU is halted for four clock cycles.
The EEPROM Address Registers (EEARH and EEARL) specify the EEPROM address
in the 4 KB EEPROM space. The EEPROM data bytes are addressed linearly between
0 and 4095.
EEPROM Data Register –
EEDR
EEPROM Control Register –
EECR
Bit76543210
$1D ($3D)MSBLSBEEDR
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
• Bits 7..0 – EEDR7..0: EEPROM Data:
For the EEPROM write operation, the EEDR register contains the data to be written to
the EEPROM in the address given by the EEAR register. For the EEPROM read operation, the EEDR contains the data read out from the EEPROM at the address given by
EEAR.
Bit76543 2 10
$1C ($3C)––––EERIEEEMWEEEWEEEREEECR
Read/WriteRRRRR/WR/WR/WR/W
Initial Value00000000
• Bits 7..4 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and will always be read as zero.
• Bit 3 – EERIE: EEPROM Ready Interrupt Enable
When the I-bit in SREG and EERIE are set (one), the EEPROM Ready interrupt is
enabled. When cleared (zero), the interrupt is disabled. The EEPROM Ready interrupt
constantly generates an interrupt request when EEWE is cleared (zero).
54
ATmega103(L)
0945G–09/01
ATmega103(L)
• Bit 2 – EEMWE: EEPROM Master Write Enable
The EEMWE bit determines whether setting EEWE to one causes the EEPROM to be
written. When EEMWE is set (one), setting EEWE will write data to the EEPROM at the
selected address. If EEMWE is zero, setting EEWE will have no effect. When EEMWE
has been set (one) by software, hardware clears the bit to zero after four clock cycles.
See the description of the EEWE bit for a EEPROM write procedure.
• Bit 1 – EEWE: EEPROM Write Enable
The EEPROM Write Enable signal (EEWE) is the write strobe to the EEPROM. When
address and data are correctly set up, the EEWE bit must be set to write the value into
the EEPROM. The EEMWE bit must be set when the logical “1” is written to EEWE, otherwise no EEPROM write takes place. The following procedure should be followed
when writing the EEPROM (the order of steps 2 and 3 is unessential):
1. Wait until EEWE becomes zero.
2. Write new EEPROM address to EEAR (optional).
3. Write new EEPROM data to EEDR (optional).
4. Write a logical “1” to the EEMWE bit in EECR (to be able to write an logical “1” to
the EEMWE bit, the EEWE bit must be written to zero in the same cycle).
5. Within four clock cycles after setting EEMWE, write a logical “1” to EEWE.
Caution: An interrupt between step 4 and step 5 will make the write cycle fail, since the
EEPROM Master Write Enable will time-out. If an interrupt routine accessing the
EEPROM is interrupting another EEPROM access, the EEAR and EEDR registers will
be modified, causing the interrupted EEPROM access to fail. It is recommended to have
the global interrupt flag cleared during the four last steps to avoid these problems.
When the write access time (typically 2.5 ms at V
elapsed, the EEWE bit is cleared (zero) by hardware. The user software can poll this bit
and wait for a zero before writing the next byte. When EEWE has been set, the CPU is
halted for two cycles before the next instruction is executed.
• Bit 0 – EERE: EEPROM Read Enable
The EEPROM Read Enable signal (EERE) is the read strobe to the EEPROM. When
the correct address is set up in the EEAR register, the EERE bit must be set. When the
EERE bit is cleared (zero) by hardware, requested data is found in the EEDR register.
The EEPROM read access takes one instruction and there is no need to poll the EERE
bit. When EERE has been set, the CPU is halted for four cycles before the next instruction is executed.
The user should poll the EEWE bit before starting the read operation. If a write operation
is in progress when new data or address is written to the EEPROM I/O registers, the
write operation will be interrupted and the result is undefined.
= 5V or 4 ms at VCC = 2.7V) has
CC
0945G–09/01
55
Prevent EEPROM
Corruption
During periods of low VCC, the EEPROM data can be corrupted because the supply voltage is too low for the CPU and the EEPROM to operate properly. These issues are the
same as for board-level systems using the EEPROM and the same design solutions
should be applied.
An EEPROM data corruption can be caused by two situations when the voltage is too
low. First, a regular write sequence to the EEPROM requires a minimum voltage to
operate correctly. Second, the CPU itself can execute instructions incorrectly if the supply voltage for executing instructions is too low.
EEPROM data corruption can easily be avoided by following these design recommendations (one is sufficient):
1. Keep the AVR RESET active (low) during periods of insufficient power supply
voltage. This is best done by an external low V
referred to as a Brown-out Detector (BOD). Please refer to application note “AVR
180” for design considerations regarding power-on reset and low-voltage
detection.
2. Keep the AVR core in Power-down Sleep Mode during periods of low V
will prevent the CPU from attempting to decode and execute instructions, effectively protecting the EEPROM registers from unintentional writes.
3. Store constants in Flash memory if the ability to change memory contents from
software is not required. Flash memory cannot be updated by the CPU and will
not be subject to corruption.
Reset Protection circuit, often
CC
. This
CC
56
ATmega103(L)
0945G–09/01
ATmega103(L)
Serial Peripheral
Interface – SPI
The Serial Peripheral Interface (SPI) allows high-speed synchronous data transfer
between the ATmega103(L) and peripheral devices or between several AVR devices.
The ATmega103(L) SPI features include the following:
Full-duplex, 3-wire Synchronous Data Transfer
•
• Master or Slave Operation
• LSB First or MSB First Data Transfer
• Four Programmable Bit Rates
• End-of-Transmission Interrupt Flag
• Write Collision Flag Protection
• Wake-up from Idle Mode (Slave Mode only)
Figure 37. SPI Block Diagram
0945G–09/01
The interconnection between master and slave CPUs with SPI is shown in Figure 38.
The PB1 (SCK) pin is the clock output in the master mode and is the clock input in the
slave mode. Writing to the SPI Data Register of the master CPU starts the SPI clock
generator, and the data written shifts out of the PB2 (MOSI) pin and into the PB2 (MOSI)
pin of the slave CPU. After shifting one byte, the SPI clock generator stops, setting the
end-of-transmission flag (SPIF). If the SPI interrupt enable bit (SPIE) in the SPCR register is set, an interrupt is requested. The Slave Select input, PB0(SS
an individual slave SPI device. The two shift registers in the master and the slave can be
considered as one distributed 16-bit circular shift register. This is shown in Figure 38.
When data is shifted from the master to the slave, data is also shifted in the opposite
direction, simultaneously. This means that during one shift cycle, data in the master and
the slave are interchanged.
), is set low to select
57
Figure 38. SPI Master-slave Interconnection
MSBLSB
8-BIT SHIFT REGISTER
SPI
CLOCK GENERATOR
MASTER
MISO
MISO
MOSI MOSI
SCK
SCK
SSSS
V
CC
MSBLSB
8-BIT SHIFT REGISTER
SLAVE
The system is single-buffered in the transmit direction and double-buffered in the
receive direction. This means that characters to be transmitted cannot be written to the
SPI Data Register before the entire shift cycle is completed. When receiving data, however, a received byte must be read from the SPI Data Register before the next byte has
been completely shifted in. Otherwise, the first byte is lost.
When the SPI is enabled, the data direction of the MOSI, MISO, SCK and SS
pins is
overridden according to the following table:
Table 22. SPI Pin Overrides
PINDirection, Master SPIDirection, Slave SPI
MOSIUser DefinedInput
MISOInputUser Defined
SCKUser DefinedInput
SSUser DefinedInput
Note:See “Alternate Functions of Port B” on page 84 for a detailed description and how to
define the direction of the user-defined SPI pins.
SS Pin FunctionalityWhen the SPI is configured as a master (MSTR in SPCR is set), the user can determine
the direction of the SS
pin that does not affect the SPI system. If SS
high to ensure Master SPI operation. If the SS
when the SPI is configured as master with the SS
pin. If SS is configured as an output, the pin is a general output
is configured as an input, it must be held
pin is driven low by peripheral circuitry
pin defined as an input, the SPI system interprets this as another master selecting the SPI as a slave and starts to send
data to it. To avoid bus contention, the SPI system takes the following actions:
1. The MSTR bit in SPCR is cleared and the SPI system becomes a slave. As a
result of the SPI becoming a slave, the MOSI and SCK pins become inputs.
2. The SPIF flag in SPSR is set, and if the SPI interrupt is enabled and the I-bit in
SREG is set, the interrupt routine will be executed.
58
Thus, when interrupt-driven SPI transmittal is used in master mode and there exists a
possibility that SS
still set. Once the MSTR bit has been cleared by a slave select, it must be set by the
user to re-enable SPI master mode.
When the SPI is configured as a slave, the SS
the SPI is activated and MISO becomes an output if configured so by the user. All other
pins are inputs. When SS
means that it will not receive incoming data. Note that the SPI logic will be reset once
ATmega103(L)
is driven low, the interrupt should always check that the MSTR bit is
pin is always input. When SS is held low,
is driven high, all pins are inputs and the SPI is passive, which
0945G–09/01
ATmega103(L)
the SS pin is brought high. If the SS pin is brought high during a transmission, the SPI
will stop sending and receiving immediately and both data received and data sent must
be considered as lost.
Data ModesThere are four combinations of SCK phase and polarity with respect to serial data that
are determined by control bits CPHA and CPOL. The SPI data transfer formats are
shown in Figure 39 and Figure 40.
Figure 39. SPI Transfer Format with CPHA = 0 and DORD = 0
SCK CYCLE #
(FOR REFERENCE)
SCK (CPOL=0)
SCK (CPOL=1)
12345678
SPI Control Register – SPCR
(FROM MASTER)
MOSI
MISO
(FROM SLAVE)
SS (TO SLAVE)
SAMPLE
* Not defined but normally MSB of character just received.
MSB654321LSB
MSB
654321LSB
Figure 40. SPI Transfer Format with CPHA = 1 and DORD = 0
SCK CYCLE #
(FOR REFERENCE)
SCK (CPOL=0)
SCK (CPOL=1)
(FROM MASTER)
(FROM SLAVE)
SS (TO SLAVE)
Bit76543210
$0D ($2D)SPIESPEDORDMSTRCPOLCPHASPR1SPR0SPCR
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
MOSI
MISO
SAMPLE
* Not defined but normally LSB of previously transmitted character.
1
MSBLSB
*
3
2
6
5
6
5
5
4
4
3
4
3
6
2
2
78
1
1
LSBMSB
*
0945G–09/01
• Bit 7 – SPIE: SPI Interrupt Enable
This bit causes the SPI interrupt to be executed if SPIF bit in the SPSR register is set
and the global interrupts are enabled.
• Bit 6 – SPE: SPI Enable
When the SPE bit is set (one), the SPI is enabled and SS
, MOSI, MISO and SCK are
connected to pins PB0, PB1, PB2 and PB3.
• Bit 5 – DORD: Data Order
When the DORD bit is set (one), the LSB of the data word is transmitted first.
When the DORD bit is cleared (zero), the MSB of the data word is transmitted first.
59
• Bit 4 – MSTR: Master/Slave Select
This bit selects Master SPI mode when set (one), and Slave SPI mode when cleared
(zero). If SS is configured as an input and is driven low while MSTR is set, MSTR will be
cleared and SPIF in SPSR will become set. The user will then have to set MSTR to reenable SPI master mode.
• Bit 3 – CPOL: Clock Polarity
When this bit is set (one), SCK is high when idle. When CPOL is cleared (zero), SCK is
low when idle. Refer to Figure 39 and Figure 40 for additional information.
• Bit 2 – CPHA: Clock Phase
Refer to Figure 39 or Figure 40 for the functionality of this bit.
These two bits control the SCK rate of the device configured as a master. SPR1 and
SPR0 have no effect on the slave. The relationship between SCK and the CPU Clock
frequency (f
) is shown in Table 23.
cl
Table 23. Relationship between SCK and the Oscillator Frequency
SPR1SPR0SCK Frequency
SPI Status Register – SPSR
00
01
10
11
Note:Observe that CPU clock frequency can be lower than the XTAL frequency if the XTAL
divider is enabled.
Bit76543210
$0ESPIFWCOL––––––SPSR
Read/WriteRRRRRRRR
Initial Value00000000
fcl/4
fcl/16
fcl/64
fcl/128
• Bit 7 – SPIF: SPI Interrupt Flag
When a serial transfer is complete, the SPIF bit is set (one) and an interrupt is generated if SPIE in SPCR is set (one) and global interrupts are enabled. SPIF is cleared by
hardware when executing the corresponding interrupt handling vector. Alternatively, the
SPIF bit is cleared by first reading the SPI Status Register with SPIF set (one), then
accessing the SPI Data Register (SPDR).
• Bit 6 – WCOL: Write Collision Flag
The WCOL bit is set if the SPI Data Register (SPDR) is written during a data transfer.
The WCOL bit (and the SPIF bit) are cleared (zero) by first reading the SPI Status Register with WCOL set (one), and then accessing the SPI Data Register.
• Bits 5..0 – Res: Reserved Bits
60
These bits are reserved bits in the ATmega103(L) and will always read as zero.
ATmega103(L)
0945G–09/01
SPI Data Register – SPDR
ATmega103(L)
Bit76543210
$0F ($2F)MSBLSBSPDR
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial ValueXXXXXXXXUndefined
The SPI Data Register is a read/write register used for data transfer between the register file and the SPI Shift Register. Writing to the register initiates data transmission.
Reading the register causes the Shift Register Receive buffer to be read.
0945G–09/01
61
UARTThe ATmega103(L) features a full duplex (separate receive and transmit registers) Uni-
versal Asynchronous Receiver and Transmitter (UART). The main features are:
Baud Rate Generator that can Generate a large Number of Baud Rates (bps)
•
• High Baud Rates at Low XTAL Frequencies
• 8 or 9 Bits Data
• Noise Filtering
• Overrun Detection
• Framing Error Detection
• False Start Bit Detection
• Three separate Interrupts on TX Complete, TX Data Register Empty and RX Complete
Data TransmissionA block schematic of the UART transmitter is shown in Figure 41.
Data transmission is initiated by writing the data to be transmitted to the UART I/O Data
Register, UDR. Data is transferred from UDR to the Transmit Shift register when:
•A new character has been written to UDR after the stop bit from the previous
character has been shifted out. The shift register is loaded immediately.
•A new character has been written to UDR before the stop bit from the previous
character has been shifted out. The shift register is loaded when the stop bit of the
character currently being transmitted has been shifted out.
If the 10(11)-bit Transmit Shift register is empty, data is transferred from UDR to the shift
register. At this time the UDRE (UART Data Register Empty) bit in the UART Status
Register, USR, is set. When this bit is set (one), the UART is ready to receive the next
character. Writing to UDR clears UDRE. At the same time as the data is transferred from
UDR to the 10(11)-bit shift register, bit 0 of the shift register is cleared (start bit) and bit 9
or 10 is set (stop bit). If 9-bit data word is selected (the CHR9 bit in the UART Control
Register, UCR is set), the TXB8 bit in UCR is transferred to bit 9 in the Transmit Shift
register.
62
ATmega103(L)
0945G–09/01
Figure 41. UART Transmitter
ATmega103(L)
DATA BUS
XTAL
CONTROL LOGIC
BAUD RATE
GENERATOR
BAUD x 16
STORE UDR
SHIFT ENABLE
IDLE
RXCIE
/16
BAUD
RXB8
TXEN
CHR9
RXEN
UART CONTROL
REGISTER (UCR)
UDRIE
DATA BUS
TXCIE
UART I/O DATA
REGISTER (UDR)
10(11)-BIT TX
SHIFT REGISTER
TXC
TXB8
RXC
UART STATUS
REGISTER (USR)
TXC
UDRE
FE
UDRE
PIN CONTROL
LOGIC
1
TXD
OR
TXC
IRQ
UDRE
IRQ
On the baud rate clock following the transfer operation to the shift register, the start bit is
shifted out on the TXD pin, followed by the data, LSB first. When the stop bit has been
shifted out, the shift register is loaded if any new data has been written to the UDR during the transmission. During loading, UDRE is set. If there is no new data in the UDR
register to send when the stop bit is shifted out, the UDRE flag will remain set. In this
case, after the stop bit has been present on TXD for one bit length, the TX Complete
Flag (TXC) in USR is set.
The TXEN bit in UCR enables the UART transmitter when set (one). When this bit is
cleared (zero), the PE1 pin can be used for general I/O. When TXEN is set, the UART
transmitter will be connected to PE1, which is forced to be an output pin regardless of
the setting of the DDE1 bit in DDRE.
0945G–09/01
63
Data ReceptionFigure 42. UART Receiver
XTAL
BAUD RATE
GENERATOR
PIN CONTROL
LOGIC
DATA BUS
BAUD X 16BAUD
/16
STORE UDR
UART I/O DATA
REGISTER (UDR)
RXD
DATA RECOVERY
LOGIC
RXEN
UART CONTROL
REGISTER (UCR)
UDRIE
RXCIE
TXCIE
RXB8
TXEN
CHR9
DATA BUS
TXB8
10(11)-BIT RX
SHIFT REGISTER
UDRE
RXC
TXC
FE
UART STATUS
REGISTER (USR)
RXC
RXC
IRQ
DOR
64
The receiver front-end logic samples the signal on the RXD pin at a frequency 16 times
the baud rate. While the line is idle, one single sample of logical “0” will be interpreted as
the falling edge of a start bit, and the start bit detection sequence is initiated. Let sample
1 denote the first zero-sample. Following the 1-to-0 transition, the receiver samples the
RXD pin at samples 8, 9, and 10. If two or more of these three samples are found to be
logical “1”s, the start bit is rejected as a noise spike and the receiver starts looking for
the next 1-to-0 transition.
If, however, a valid start bit is detected, sampling of the data bits following the start bit is
performed. These bits are also sampled at samples 8, 9 and 10. The logical value found
in at least two of the three samples is taken as the bit value. All bits are shifted into the
transmitter shift register as they are sampled. Sampling of an incoming character is
shown in Figure 43.
ATmega103(L)
0945G–09/01
ATmega103(L)
Figure 43. Sampling Received Data
RXD
START BITD0D1D2D3D4D5D6D7STOP BIT
RECEIVER
SAMPLING
When the stop bit enters the receiver, the majority of the three samples must be one to
accept the stop bit. If two or more samples are logical “0”s, the Framing Error (FE) flag in
the UART Status Register (USR) is set when the received byte is transferred to UDR.
Before reading the UDR register, the user should always check the FE bit to detect
Framing Errors. FE is cleared when UDR is read.
Whether or not a valid stop bit is detected at the end of a character reception cycle, the
data is transferred to UDR and the RXC flag in USR is set. UDR is in fact two physically
separate registers, one for transmitted data and one for received data. When UDR is
read, the Receive Data register is accessed, and when UDR is written, the Transmit
Data register is accessed. If 9-bit data word is selected (the CHR9 bit in the UART Control Register, UCR is set), the RXB8 bit in UCR is loaded with bit 9 in the Transmit Shift
register when data is transferred to UDR.
If, after having received a character, the UDR register has not been accessed since the
last receive, the OverRun (OR) flag in USR is set. This means that the new data transferred to the shift register could not be transferred to UDR and is lost. The OR bit is
buffered, and is available when the valid data byte in UDR has been read. The user
should always check the OR after reading from the UDR register in order to detect any
overruns if the baud rate is high or CPU load is high.
When the RXEN bit in the UCR register is cleared (zero), the receiver is disabled. This
means that the PE0 pin can be used as a general I/O pin. When RXEN is set, the UART
Receiver will be connected to PE0, which is forced to be an input pin regardless of the
setting of the DDE0 bit in DDRE. When PE0 is forced to input by the UART, the
PORTE0 bit can still be used to control the pull-up resistor on the pin.
When the CHR9 bit in the UCR register is set, transmitted and received characters are 9
bits long plus start and stop bits. The 9th data bit to be transmitted is the TXB8 bit in
UCR register. This bit must be set to the wanted value before a transmission is initated
by writing to the UDR register.
0945G–09/01
65
UART Control
UART I/O Data Register – UDR
UART Status Register – USR
Bit76543210
$0C ($2C)MSBLSBUDR
Read/WriteR/WR/WR/WR/WR/WR/WR/WR/W
Initial Value00000000
The UDR register is actually two physically separate registers sharing the same I/O
address. When writing to the register, the UART Transmit Data register is written. When
reading from UDR, the UART Receive Data register is read.
Bit76543210
$0B ($2B)RXCTXCUDREFEOR–––USR
Read/WriteRR/WRRRRRR
Initial Value00100000
The USR register is a read-only register providing information on the UART Status.
• Bit 7 – RXC: UART Receive Complete
This bit is set (one) when a received character is transferred from the Receiver Shift register to UDR. The bit is set regardless of any detected framing errors. When the RXCIE
bit in UCR is set, the UART Receive Complete interrupt will be executed when RXC is
set (one). RXC is cleared by reading UDR. When interrupt-driven data reception is used,
the UART Receive Complete Interrupt routine must read UDR in order to clear RXC,
otherwise a new interrupt will occur once the interrupt routine terminates.
• Bit 6 – TXC: UART Transmit Complete
This bit is set (one) when the entire character (including the stop bit) in the Transmit
Shift register has been shifted out and no new data has been written to the UDR. This
flag is especially useful in half-duplex communications interfaces, where a transmitting
application must enter receive mode and free the communications bus immediately after
completing the transmission.
When the TXCIE bit in UCR is set, setting of TXC causes the UART Transmit Complete
interrupt to be executed. TXC is cleared by hardware when executing the corresponding
interrupt handling vector. Alternatively, the TXC bit is cleared (zero) by writing a logical
“1” to the bit.
• Bit 5 – UDRE: UART Data Register Empty
This bit is set (one) when a character written to UDR is transferred to the Transmit Shift
register. Setting of this bit indicates that the transmitter is ready to receive a new character for transmission.
When the UDRIE bit in UCR is set, the UART Transmit Complete interrupt to be executed as long as UDRE is set. UDRE is cleared by writing UDR. When interrupt-driven
data transmittal is used, the UART Data Register Empty Interrupt routine must write
UDR in order to clear UDRE, otherwise a new interrupt will occur once the interrupt routine terminates.
UDRE is set (one) during reset to indicate that the transmitter is ready.
• Bit 4 – FE: Framing Error
This bit is set if a Framing Error condition is detected, i.e., when the stop bit of an incoming character is zero.
66
ATmega103(L)
0945G–09/01
UART Control Register – UCR
ATmega103(L)
The FE bit is cleared when the stop bit of received data is one.
• Bit 3 – OR: OverRun
This bit is set if an overrun condition is detected, i.e., when a character already present
in the UDR register is not read before the next character is transferred from the Receiver
Shift register. The OR bit is buffered, which means that it will be set once the valid data
still in UDRE is read.
The OR bit is cleared (zero) when data is received and transferred to UDR.
• Bits 2..0 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and will always read as zero.
Bit76543210
$0A ($2A)RXCIETXCIEUDRIERXENTXENCHR9RXB8TXB8UCR
Read/WriteR/WR/WR/WR/WR/WR/WRW
Initial Value00000010
• Bit 7 – RXCIE: RX Complete Interrupt Enable
When this bit is set (one), a setting of the RXC bit in USR will cause the Receive Complete Interrupt routine to be executed, provided that global interrupts are enabled.
• Bit 6 – TXCIE: TX Complete Interrupt Enable
When this bit is set (one), a setting of the TXC bit in USR will cause the Transmit Complete Interrupt routine to be executed, provided that global interrupts are enabled.
• Bit 5 – UDRIE: UART Data Register Empty Interrupt Enable
When this bit is set (one), a setting of the UDRE bit in USR will cause the UART Data
Register Empty Interrupt routine to be executed, provided that global interrupts are
enabled.
• Bit 4 – RXEN: Receiver Enable
This bit enables the UART receiver when set (one). When the receiver is disabled, the
RXC, OR and FE status flags cannot become set. If these flags are set, turning off
RXEN does not cause them to be cleared.
• Bit 3 – TXEN: Transmitter Enable
This bit enables the UART transmitter when set (one). When disabling the transmitter
while transmitting a character, the transmitter is not disabled before the character in the
shift register plus any following character in UDR has been completely transmitted.
• Bit 2 – CHR9: 9-bit Characters
When this bit is set (one), transmitted and received characters are nine bits long, plus
start and stop bits. The ninth bit is read and written by using the RXB8 and TXB8 bits in
UCR, respectively. The ninth data bit can be used as an extra stop bit or a parity bit.
• Bit 1 – RXB8: Receive Data Bit 8
When CHR9 is set (one), RXB8 is the ninth data bit of the received character.
• Bit 0 – TXB8: Transmit Data Bit 8
0945G–09/01
When CHR9 is set (one), TXB8 is the ninth data bit in the character to be transmitted.
67
Baud Rate GeneratorThe baud rate generator is a frequency divider that generates baud rates according to
the following equation:
f
CK
BAUD
•BAUD = baud rate
•f
•UBRR = contents of the UART baud rate register, UBRR (0 - 255)
For standard crystal frequencies, the most commonly used baud rates can be generated
by using the UBRR settings in Table 24. Observe that CPU clock frequency can be
lower than the XTAL frequency if the XTAL divider is enabled. UBRR values that yield an
actual baud rate differing less than 2% from the target baud rate are in boldface in the
table. However, using baud rates that have more than 1% error is not recommended.
High error ratings give less noise resistance.
The UBRR is an 8-bit read/write register that specifies the UART baud rate according to
the description on the previous page.
0945G–09/01
69
Analog ComparatorThe analog comparator compares the input values on the positive input PE2 (AC+) and
negative input PE3 (AC-). When the voltage on the positive input PE2 (AC+) is higher
than the voltage on the negative input PE3 (AC-), the Analog Comparator Output (ACO)
is set (one). The output of the comparator can be set to trigger the Timer/Counter1 Input
Capture function. In addition, the comparator can trigger a separate interrupt, exclusive
to the analog comparator. The user can select interrupt triggering on comparator output
rise, fall or toggle. A block diagram of the comparator and its surrounding logic is shown
in Figure 44.
Figure 44. Analog Comparator Block Diagram
VCC
ACD
ACIE
PE2
(AC+)
PE3
(AC-)
+
-
INTERRUPT
SELECT
ACIS1ACIS0
ACIC
ANALOG
COMPARATOR
IRQ
ACI
Analog Comparator Control
and Status Register – ACSR
TO T/C1 CAPTURE
ACO
Bit76543210
$08 ($28)ACD–ACOACIACIEACICACIS1ACIS0ACSR
Read/WriteR/WRRR/WR/WR/WR/WR/W
Initial Value00X00000
TRIGGER MUX
• Bit 7 – ACD: Analog Comparator Disable
When this bit is set (one), the power to the analog comparator is switched off. This bit
can be set at any time to turn off the analog comparator. This will reduce power consumption in active and idle mode. When changing the ACD bit, the Analog Comparator
interrupt must be disabled by clearing the ACIE bit in ACSR. Otherwise, an interrupt can
occur when the bit is changed.
• Bit 6 – Res: Reserved Bit
This bit is a reserved bit in the ATmega103(L) and will always read as zero.
• Bit 5 – ACO: Analog Comparator Output
ACO is directly connected to the comparator output.
• Bit 4 – ACI: Analog Comparator Interrupt Flag
This bit is set (one) when a comparator output event triggers the interrupt mode defined
by ACI1 and ACI0. The Analog Comparator Interrupt routine is executed if the ACIE bit
is set (one) and the I-bit in SREG is set (one). ACI is cleared by hardware when executing the corresponding interrupt handling vector. Alternatively, ACI is cleared by writing a
logical “1” to the flag. Observe, however, that if another bit in this register is modified
70
ATmega103(L)
0945G–09/01
ATmega103(L)
using the SBI or CBI instruction, ACI will be cleared if it has become set before the
operation.
• Bit 3 – ACIE: Analog Comparator Interrupt Enable
When the ACIE bit is set (one) and the I-bit in the Status Register is set (one), the Analog Comparator interrupt is activated. When cleared (zero), the interrupt is disabled.
• Bit 2 – ACIC: Analog Comparator Input Capture Enable
When set (one), this bit enables the Input Capture function in Timer/Counter1 to be triggered by the analog comparator. The comparator output is, in this case, directly
connected to the Input Capture front-end logic, making the comparator utilize the noise
canceler and edge select features of the Timer/Counter1 Input Capture interrupt. When
cleared (zero), no connection between the analog comparator and the Input Capture
function is given. To make the comparator trigger the Timer/Counter1 Input Capture
interrupt, the TICIE1 bit in the Timer Interrupt Mask Register (TIMSK) must be set (one).
These bits determine which comparator events that trigger the Analog Comparator interrupt. The different settings are shown in Table 25.
Table 25. ACIS1/ACIS0 Settings
ACIS1ACIS0Interrupt Mode
00Comparator Interrupt on Output Toggle
01Reserved
10Comparator Interrupt on Falling Output Edge
11Comparator Interrupt on Rising Output Edge
When changing the ACIS1/ACIS0 bits, the Analog Comparator interrupt must be disabled by clearing its Interrupt Enable bit in the ACSR register. Otherwise, an interrupt
can occur when the bits are changed.
Caution: Using the SBI or CBI instruction on other bits than ACI in this register will write
a one back into ACI if it is read as set, thus clearing the flag.
0945G–09/01
71
Analog-to-Digital
Converter
Feature list:• 10-bit Resolution
• ±2 LSB Absolute Accuracy
• 0.5 LSB Integral Non-linearity
• 70 - 280 µs Conversion Time
• Up to 14 kSPS
• 8 Multiplexed Input Channels
• Interrupt on ADC Conversion Complete
• Sleep Mode Noise Canceler
The ATmega103(L) features a 10-bit successive approximation ADC. The ADC is connected to an 8-channel Analog Multiplexer, which allows each pin of Port F to be used
as an input for the ADC. The ADC contains a Sample and Hold Amplifier, which ensures
that the input voltage to the ADC is held at a constant level during conversion. A block
diagram of the ADC is shown in Figure 45.
The ADC has two separate analog supply voltage pins, AV
be connected to GND, and the voltage on AV
. See the section “ADC Noise Canceling Techniques” on page 77 on how to connect
V
CC
must not differ more than ± 0.3V from
CC
and AGND. AGND must
CC
these pins.
An external reference voltage must be applied to the AREF pin. This voltage must be in
the range AGND - AV
OperationThe ADC operates in Single Conversion mode, and each conversion will have to be ini-
tiated by the user.
The ADC is enabled by writing a logical “1” to the ADC Enable bit, ADEN in ADCSR.
The first conversion that is started after enabling the ADC will be preceded by a dummy
conversion to initialize the ADC. To the user, the only difference will be that this conversion takes 13 more ADC clock pulses than a normal conversion (see Figure 48).
A conversion is started by writing a logical “1” to the ADC Start Conversion bit, ADSC.
This bit will stay high as long as the conversion is in progress and be set to zero by hardware when the conversion is completed. If a different data channel is selected while a
conversion is in progress, the ADC will finish the current conversion before performing
the channel change.
As the ADC generates a 10-bit result, two data registers, ADCH and ADCL, must be
read to get the result when the conversion is complete. Special data protection logic is
used to ensure that the contents of the data registers belong to the same result when
they are read. This mechanism works as follows:
When reading data, ADCL must be read first. Once ADCL is read, ADC access to data
registers is blocked. This means that if ADCL has been read, and a conversion completes before ADCH is read, none of the registers are updated and the result from the
conversion is lost. When ADCH is read, ADC access to the ADCH and ADCL registers
is re-enabled.
The ADC has its own interrupt, ADIF, which can be triggered when a conversion completes. When ADC access to the data registers is prohibited between reading of ADCL
and ADCH, the interrupt will trigger even if the result is lost.
PrescalingFigure 46. ADC Prescaler
ADEN
CK
ADPS0
ADPS1
ADPS2
The ADC contains a prescaler, which divides the system clock to an acceptable ADC
clock frequency. The ADC accepts input clock frequencies in the range 50 - 200 kHz.
Applying a higher input frequency will result in poorer accuracy (see “ADC DC Characteristics” on page 78).
Reset
7-BIT ADC PRESCALER
CK/2
CK/4
CK/8
ADC CLOCK SOURCE
CK/16
CK/32
CK/64
CK/128
0945G–09/01
The ADPS0 - ADPS2 bits in ADCSR are used to generate a proper ADC clock input frequency from any XTAL frequency above 100 kHz. The prescaler starts counting from
the moment the ADC is switched on by setting the ADEN bit in ADCSR. The prescaler
73
keeps running for as long as the ADEN bit is set and is continuously reset when ADEN
is low.
When initiating a conversion by setting the ADSC bit in ADCSR, the conversion starts at
the following falling edge of the ADC clock cycle. The actual sample-and-hold takes
place one ADC clock cycle after the start of the conversion. The result is ready and written to the ADC Result Register after 13 cycles. The ADC needs two more clock cycles
before a new conversion can be started. If ADSC is set high in this period, the ADC will
start the new conversion immediately. For a summary of conversion times, see Table
26.
Figure 47. ADC Timing Diagram, First Conversion
Cycle number
ADC clock
ADEN
ADSC
Hold strobe
ADIF
ADCH
ADCL
1212
Dummy ConversionActual Conversion
13
14 15
16 17
18
19 20 2122 23
24 25 2627 28 1
MSB of result
LSB of result
Table 26. ADC Conversion Time
Sample
Cycle
Condition
Number
1st Conversion142628140 - 560
Single Conversion1131575 - 300
Result Ready
(Cycle
Number)
Tot al
Conversion
Time (Cycles)
Total
Conversion
Time (µs)
Figure 48. ADC Timing Diagram
Cycle number
12345678
10111213
9
141512
Second
Conversion
2
74
ADC clock
ADSC
Hold strobe
ADIF
ADCH
ADCL
ATmega103(L)
MSB of result
LSB of result
One ConversionNext Conversion
0945G–09/01
ATmega103(L)
ADC Noise Canceler
Function
ADC Multiplexer Select
Register – ADMUX
The ADC features a noise canceler that enables conversion during idle mode to reduce
noise induced from the CPU core. To make use of this feature, the following procedure
should be used:
1. Turn off the ADC by clearing ADEN.
2. Turn on the ADC and simultaneously start a conversion by setting ADEN and
ADSC. This starts a dummy conversion that will be followed by a valid
conversion.
3. Within 14 ADC clock cycles, enter idle mode.
4. If no other interrupts occur before the ADC conversion completes, the ADC interrupt will wake up the MCU and execute the ADC conversion complete interrupt
routine.
Bit76543210
$07 ($27)–––––MUX2MUX1MUX0ADMUX
Read/WriteRRRRRR/WR/WR/W
Initial Value00000000
• Bits 7..3 – Res: Reserved Bits
These bits are reserved bits in the ATmega103(L) and always read as zero.
The value of these three bits selects which analog input 7 - 0 is connected to the ADC.
ADC Control and Status
Register – ADCSR
Bit76543210
$06 ($26)ADENADSC–ADIFADIEADPS2ADPS1ADPS0ADCSR
Read/WriteR/WR/WRR/WR/WR/WR/WR/W
Initial Value00000000
• Bit 7 – ADEN: ADC Enable
Writing a logical “1” to this bit enables the ADC. By clearing this bit to zero, the ADC is
turned off. Turning the ADC off while a conversion is in progress will terminate this
conversion.
• Bit 6 – ADSC: ADC Start Conversion
A logical “1” must be written to this bit to start each conversion. The first time ADSC has
been written after the ADC has been enabled, or if ADSC is written at the same time as
the ADC is enabled, a dummy conversion will precede the initiated conversion. This
dummy conversion performs initialization of the ADC.
ADSC remains high during the conversion. ADSC goes low after the conversion is complete, but before the result is written to the ADC Data registers. This allows a new
conversion to be initiated before the current conversion is complete. The new conversion will then start immediately after the current conversion completes. When a dummy
conversion precedes a real conversion, ADSC will stay high until the real conversion
completes.
Writing a zero to this bit has no effect.
• Bit 5 – Res: Reserved Bit
0945G–09/01
This bit is reserved in the ATmega103(L). Warning: When writing ADCSR, a logical “0”
must be written to this bit.
75
• Bit 4 – ADIF: ADC Interrupt Flag
This bit is set (one) when an ADC conversion is complete and the result is written to the
ADC Data registers are updated. The ADC Conversion Complete interrupt is executed if
the ADIE bit and the I-bit in SREG are set (one). ADIF is cleared by hardware when executing the corresponding interrupt handling vector. Alternatively, ADIF is cleared by
writing a logical “1” to the flag. Beware that if doing a read-modify-write on ADCSR, a
pending interrupt can be disabled. This also applies if the SBI and CBI instructions are
used.
• Bit 3 – ADIE: ADC Interrupt Enable
When this bit is set (one) and the I-bit in SREG is set (one), the ADC Conversion Complete interrupt is activated.
These bits determine the division factor between the XTAL frequency and the input
clock to the ADC.
Table 27. ADC Prescaler Selections
ADPS2ADPS1ADPS0Division Factor
000Invalid
0012
0104
0118
ADC Data Register – ADCL
and ADCH
10016
10132
11064
111128
Bit151413121110 9 8
$05 ($25)––––––ADC9ADC8ADCH
$04 ($24)ADC7ADC6ADC5ADC4ADC3ADC2ADC1ADC0ADCL
76543210
Read/WriteRRRRRRRR
RRRRRRRR
Initial Value00000000
00000000
When an ADC conversion is complete, the result is found in these two registers. It is
essential that both registers are read and that ADCL is read before ADCH.
76
ATmega103(L)
0945G–09/01
ATmega103(L)
ADC Noise Canceling
Techniques
Digital circuitry inside and outside the ATmega103(L) generates EMI, which might affect
the accuracy of analog measurements. If conversion accuracy is critical, the noise level
can be reduced by applying the following techniques:
1. The analog part of the ATmega103(L) and all analog components in the application should have a separate analog ground plane on the PCB. This ground plane
is connected to the digital ground plane via a single point on the PCB.
2. Keep analog signal paths as short as possible. Make sure analog tracks run over
the analog ground plane, and keep them well away from high-speed switching
digital tracks.
3. The AV
pin on the ATmega103(L) should have its own decoupling capacitor as
CC
shown in Figure 49.
4. Use the ADC Noise Canceler function to reduce induced noise from the CPU.
5. If some Port F pins are used as digital inputs, it is essential that these do not
switch while a conversion is in progress.
Figure 49. ADC Power Connections
(AD0) PA0
(ADC7) PF7
VCC
GND
51
52
53
54
10nF
Analog Ground Plane
(ADC6) PF6
(ADC5) PF5
(ADC4) PF4
(ADC3) PF3
(ADC2) PF2
(ADC1) PF1
(ADC0) PF0
AREF
AGND
AVCC
55
56
57
58
59
60
6161
6262
6363
6464
ATmega103(L)
1
PEN
0945G–09/01
77
ADC DC Characteristics
TA = -40°C to 85°C
SymbolParameterConditionMinTypMaxUnits
Resolution10Bits
AV
V
REF
CC
Absolute
accuracy
Absolute
accuracy
Absolute
accuracy
Integral
Non-linearity
Differential
Non-linearity
Zero Error
(Offset)
Conversion
Time
Clock
Frequency
Analog
Supply
Volt age
Reference
Volt age
VREF = 4V, V
ADC clock = 200 kHz
VREF = 4V, V
ADC clock = 1 MHz
VREF = 4V, V
ADC clock = 2 MHz
CC
CC
CC
= 4V
= 4V
= 4V
12LSB
4LSB
16LSB
VREF > 2V0.5LSB
VREF > 2V0.5LSB
1LSB
70280µs
50200kHz
(1)
- 0.3
V
CC
VCC + 0.3
2AV
CC
(2)
V
V
Reference
R
REF
Input
Resistance
R
AIN
Analog Input
Resistance
Notes:1. Minimum for AVCC is 2.7V.
2. Maximum for AV
is 6.0V.
CC
61013 kΩ
100MΩ
78
ATmega103(L)
0945G–09/01
ATmega103(L)
Interface to External
SRAM
The interface to the SRAM consists of:
Port A: multiplexed low-order address bus and data bus
Port C: high-order address bus
The ALE pin: address latch enable
The RD
and WR pin: read and write strobes
The external data SRAM is enabled by setting the external SRAM enable bit (SRE) of
the MCU Control Register (MCUCR) and will override the setting of the data direction
register (DDRA). When the SRE bit is cleared (zero), the external data SRAM is disabled and the normal pin and data direction settings are used. When SRE is cleared
(zero), the address space above the internal SRAM boundary is not mapped into the
internal SRAM as AVR parts do not have an interface to the external SRAM.
When ALE goes from high to low, there is a valid address on Port A. ALE is low during a
data transfer. RD
and WR are active when accessing the external SRAM only.
When the external SRAM is enabled, the ALE signal may have short pulses when
accessing the internal RAM, but the ALE signal is stable when accessing the external
SRAM.
Figure 50 shows how to connect an external SRAM to the AVR using eight latches that
are transparent when G is high.
By default, the external SRAM access is a three-cycle scheme as depicted in Figure 51.
When one extra wait state is needed in the access cycle, set the SRW bit (one) in the
MCUCR register. The resulting access scheme is shown in Figure 52. In both cases,
note that Port A is data bus in one cycle only. As soon as the data access finishes, Port
A becomes a low-order address bus again.
Note:If a read is followed by a write, or vice versa, there is no extra insertion of wait states in
between. The user may insert a NOP between consecutive read and write operations to
the external RAM, because such short time for releasing the bus is difficult to obtain without making bus contention.
0945G–09/01
For details on the timing for the SRAM interface, please refer to Figure 79, Table 45,
Table 46, Table 47, and Table 48 in the section “DC Characteristics” on page 113 and
refer to “Architectural Overview” on page 8 for a description of the memory map, including address space for SRAM.
Figure 50. External SRAM Connected to the AVR
D[7:0]
A[7:0]
SRAM
A[15:8]
RD
WR
AVR
Por t A
ALE
Por t C
RD
WR
DQ
G
79
Figure 51. External SRAM Access Cycle without Wait States
T1T2T3
System Clock Ø
ALE
Address [15..8]
Data / Address [7..0]
Prev. Address
Prev. Address
Address
WR
Data / Address [7..0]
Prev. Address
Address
Data
RD
Figure 52. External SRAM Access Cycle with Wait State
T1T2T3T4
System Clock Ø
ALE
Address [15..8]
Data / Address [7..0]
Prev. Address
Prev. Address
Address
WR
Data / Address [7..0]
Prev. Address
Address
RD
Address
Data
Address
Data
Data
Address
Address
Addr
Addr
Write
Read
.
Write
.
Read
80
ATmega103(L)
0945G–09/01
ATmega103(L)
I/O PortsAll AVR ports have true read-modify-write functionality when used as general digital I/O
ports. This means that the direction of one port pin can be changed without unintentionally changing the direction of any other pin with the SBI and CBI instructions. The same
applies for changing drive value (if configured as output) or enabling/disabling of pull-up
resistors (if configured as input).
Port APort A is an 8-bit bi-directional I/O port with internal pull-ups.
Three I/O memory address locations are allocated for Port A, one each for the Data
Register – PORTA, $1B($3B), Data Direction Register – DDRA, $1A($3A) and the Port
A Input Pins – PINA, $19($39). The Port A Input Pins address is read-only, while the
Data Register and the Data Direction Register are read/write.
All port pins have individually selectable pull-up resistors. The Port A output buffers can
sink 20 mA and thus drive LED displays directly. When pins PA0 to PA7 are used as
inputs and are externally pulled low, they will source current if the internal pull-up resistors are activated.
The Port A pins have alternate functions related to the optional external data SRAM.
Port A can be configured to be the multiplexed low-order address/data bus during
accesses to the byte.
When Port A is set to the alternate function by the SRE (External SRAM Enable) bit in
the MCUCR (MCU Control Register), the alternate settings override the data direction
register.
The Port A Input Pins address (PINA) is not a register; this address enables access to
the physical value on each Port A pin. When reading PORTA the Port A Data Latch is
read and when reading PINA, the logical values present on the pins are read.
Port A as General Digital I/OAll eight pins in Port A have equal functionality when used as digital I/O pins.
PAn, general I/O pin: The DDAn bit in the DDRA register selects the direction of this pin.
If DDAn is set (one), PAn is configured as an output pin. If DDAn is cleared (zero), PAn
is configured as an input pin. If PORTAn is set (one) when the pin configured as an input
pin, the MOS pull-up resistor is activated. To switch the pull-up resistor off, PORTAn has
0945G–09/01
81
to be cleared (zero) or the pin has to be configured as an output pin. The port pins are
tri-stated when a reset condition becomes active, even if the clock is not running.
Table 28. DDAn Effects on Port A Pins
DDAnPORTAnI/OPull-upComment
00InputNoTri-state (high-Z)
01InputYesPAn will source current if ext. pulled low.
10OutputNoPush-pull Zero Output
11OutputNoPush-pull One Output
Note:n: 7,6...0, pin number
Port A SchematicsNote that all port pins are synchronized. The synchronization latch is, however, not
shown in the figure.
Figure 53. Port A Schematic Diagrams (Pins PA0 - PA7)
RD
MOS
PULLUP
PAn
RL
RESET
Q
DDAn
RESET
Q
PORTAn
R
D
C
WD
R
D
C
WP
A BUS
DAT
R
SRE
W
Dn
An
WRITE PORTA
WP:
WRITE DDRA
WD:
READ PORTA LATCH
RL:
READ PORTA PIN
RP:
READ DDRA
RD:
EXT. SRAM ENABLE
SRE:
ADDRESS
A:
DATA
D:
WRITE
W:
READ
R:
0-7
n:
Port BPort B is an 8-bit bi-directional I/O port with internal pull-ups.
Three I/O memory address locations are allocated for Port B, one each for the Data
Register – PORTB, $18($38), Data Direction Register – DDRB, $17($37) and the Port B
Input Pins – PINB, $16($36). The Port B Input Pins address is read-only, while the Data
Register and the Data Direction Register are read/write.
All port pins have individually selectable pull-up resistors. The Port B output buffers can
sink 20 mA and thus drive LED displays directly. When pins PB0 to PB7 are used as
RP
SRE
R
W
Dn
An
82
ATmega103(L)
0945G–09/01
ATmega103(L)
inputs and are externally pulled low, they will source current if the internal pull-up resistors are activated.
The Port B pins with alternate functions are shown in Table 29.
Table 29. Port B Pin Alternate Functions
Port PinAlternate Functions
Port B Data Register – PORTB
Port B Data Direction Register
– DDRB
PB0SS
PB1SCK (SPI Bus Serial Clock)
PB2MOSI (SPI Bus Master Output/Slave Input)
PB3MISO (SPI Bus Master Input/Slave Output)
PB4OC0/PWM0 (Output Compare and PWM Output for Timer/Counter0)
PB5OC1A/PWM1A (Output Compare and PWM Output A for Timer/Counter1)
PB6OC1B/PWM1B (Output Compare and PWM Output B for Timer/Counter1)
PB7OC2/PWM2 (Output Compare and PWM Output for Timer/Counter2)
(SPI Slave Select input)
When the pins are used for the alternate function, the DDRB and PORTB registers have
to be set according to the alternate function description.
The Port B Input Pins address (PINB) is not a register; this address enables access to
the physical value on each Port B pin. When reading PORTB, the Port B Data Latch is
read and when reading PINB, the logical values present on the pins are read.
Port B as General Digital I/OAll eight pins in Port B have equal functionality when used as digital I/O pins.
PBn, general I/O pin: The DDBn bit in the DDRB register selects the direction of this pin.
If DDBn is set (one), PBn is configured as an output pin. If DDBn is cleared (zero), PBn
is configured as an input pin. If PORTBn is set (one) when the pin configured as an input
pin, the MOS pull-up resistor is activated. To switch the pull-up resistor off, the PORTBn
has to be cleared (zero) or the pin has to be configured as an output pin. The port pins
are tri-stated when a reset condition becomes active, even if the clock is not running.
0945G–09/01
83
Table 30. DDBn Effects on Port B Pins
DDBnPORTBnI/OPull-upComment
00InputNoTri-state (high-Z)
01InputYesPBn will source current if ext. pulled low
10OutputNoPush-pull Zero Output
11OutputNoPush-pull One Output
Note:n: 7,6...0, pin number
Alternate Functions of Port BThe alternate pin configuration is as follows:
• OC2/PWM2, Bit 7
OC2/PWM2, Output Compare output for Timer/Counter2 or PWM output when
Timer/Counter2 is in PWM Mode. The pin has to be configured as an output to serve
this function.
• OC1B/PWM1B, Bit 6
OC1B/PWM1B, Output Compare output B for Timer/Counter1 or PWM output B when
Timer/Counter1 is in PWM Mode. The pin has to be configured as an output to serve
this function.
• OC1A/PWM1A, Bit 5
OC1A/PWM1A, Output Compare output A for Timer/Counter1 or PWM output A when
Timer/Counter1 is in PWM Mode. The pin has to be configured as an output to serve
this function.
• OC0/PWM0, Bit 4
OC0/PWM0, Output Compare output for Timer/Counter0 or PWM output when
Timer/Counter0 is in PWM Mode. The pin has to be configured as an output to serve
this function.
• MISO – Port B, Bit 3
MISO: Master data input, slave data output pin for SPI channel. When the SPI is
enabled as a master, this pin is configured as an input regardless of the setting of
DDB3. When the SPI is enabled as a slave, the data direction of this pin is controlled by
DDB3. When the pin is forced to be an input, the pull-up can still be controlled by the
PORTB3 bit. See the description of the SPI port for further details.
• MOSI – Port B, Bit 2
MOSI: SPI Master data output, slave data input for SPI channel. When the SPI is
enabled as a slave, this pin is configured as an input regardless of the setting of DDB2.
When the SPI is enabled as a master, the data direction of this pin is controlled by
DDB2. When the pin is forced to be an input, the pull-up can still be controlled by the
PORTB2 bit. See the description of the SPI port for further details.
• SCK – Port B, Bit 1
SCK: Master clock output, slave clock input pin for SPI channel. When the SPI is
enabled as a slave, this pin is configured as an input regardless of the setting of DDB1.
When the SPI is enabled as a master, the data direction of this pin is controlled by
DDB1. When the pin is forced to be an input, the pull-up can still be controlled by the
PORTB1 bit. See the description of the SPI port for further details.
84
ATmega103(L)
0945G–09/01
ATmega103(L)
• SS – Port B, Bit 0
SS
: Slave port select input. When the SPI is enabled as a slave, this pin is configured as
an input regardless of the setting of DDB0. As a slave, the SPI is activated when this pin
is driven low. When the SPI is enabled as a master, the data direction of this pin is controlled by DDB0. When the pin is forced to be an input, the pull-up can still be controlled
by the PORTB0 bit. See the description of the SPI port for further details.
Port B SchematicsNote that all port pins are synchronized. The synchronization latches are, however, not
shown in the figures.
Figure 54. Port B Schematic Diagram (Pin PB0)
0945G–09/01
85
Figure 55. Port B Schematic Diagram (Pin PB1)
Figure 56. Port B Schematic Diagram (Pin PB2)
86
ATmega103(L)
0945G–09/01
Figure 57. Port B Schematic Diagram (Pin PB3)
ATmega103(L)
Figure 58. Port B Schematic Diagram (Pin PB4)
MOS
PULLUP
PB4
WRITE PORTB
WP:
WRITE DDRB
WD:
READ PORTB LATCH
RL:
READ PORTB PIN
RP:
READ DDRB
RD:
RL
RP
MODE SELECT
OUTPUT
RESET
Q
DDB4
RESET
Q
PORTB4
RD
R
D
C
WD
R
D
C
WP
COM01
COM01
COMP. MATCH 0
TA B US
DA
0945G–09/01
87
Figure 59. Port B Schematic Diagram (Pins PB5 and PB6)
MOS
PULLUP
PBn
RL
RP
RESET
Q
DDBn
WD
RESET
Q
PORTBn
WP
RD
R
D
C
R
D
C
TA B US
DA
WRITE PORTB
WP:
WRITE DDRB
WD:
READ PORTB LATCH
RL:
READ PORTB PIN
RP:
READ DDRB
RD:
5, 6
n:
A, B
X:
Figure 60. Port B Schematic Diagram (Pin PB7)
MOS
PULLUP
PB7
MODE SELECT
RL
RP
OUTPUT
RESET
Q
DDB7
RESET
Q
PORTB7
RD
WD
WP
COM1X0
COM1X1
COMP. MATCH 1X
R
D
C
R
D
C
TA B US
DA
88
ATmega103(L)
WRITE PORTB
WP:
WRITE DDRB
WD:
READ PORTB LATCH
RL:
READ PORTB PIN
RP:
READ DDRB
RD:
OUTPUT
MODE SELECT
COM20
COM21
COMP. MATCH 2
0945G–09/01
Port CPort C is an 8-bit output port.
U
The Port C pins have alternate functions related to the optional external data SRAM.
When using the device with external SRAM, Port C outputs the high-order address byte
during accesses to external data memory. When a reset condition becomes active, the
port pins are not tri-stated, but the pins will assume their initial value after two stable
clock cycles.
The Port C Data Register –
PORTC
Port C SchematicsFigure 61. Port C Schematic Diagram (Pins PC0 - PC7)
Port DPort D is an 8-bit bi-directional I/O port with internal pull-up resistors.
Three I/O memory address locations are allocated for the Port D, one each for the Data
Register – PORTD, $12($32), Data Direction Register – DDRD, $11($31) and the Port D
Input Pins – PIND, $10($30). The Port D Input Pins address is read-only, while the Data
Register and the Data Direction Register are read/write.
The Port D output buffers can sink 20 mA. As inputs, Port D pins that are externally
pulled low will source current if the pull-up resistors are activated.
Some Port D pins have alternate functions as shown in Table 31.
Table 31. Port D Pin Alternate Functions
S
DATA B
SRE
An
0945G–09/01
Port PinAlternate Function
PD0INT0
(External Interrupt0 Input)
PD1INT1 (External Interrupt1 Input)
PD2INT2
PD3INT3
(External Interrupt2 Input)
(External Interrupt3 Input)
PD4IC1 (Timer/Counter1 Input Capture Trigger)
PD6T1 (Timer/Counter1 Clock Input)
PD7T2 (Timer/Counter2 Clock Input)
89
Port D Data Register – PORTD
Port D Data Direction Register
– DDRD
Port D Input Pins Address –
PIND
When the pins are used for the alternate function, the DDRD and PORTD registers have
to be set according to the alternate function description.
The Port D Input Pins address (PIND) is not a register, and this address enables access
to the physical value on each Port D pin. When reading PORTD, the Port D Data Latch
is read, and when reading PIND, the logical values present on the pins are read.
Port D as General Digital I/OPDn, General I/O pin: The DDDn bit in the DDRD register selects the direction of this
pin. If DDDn is set (one), PDn is configured as an output pin. If DDDn is cleared (zero),
PDn is configured as an input pin. If PDn is set (one) when configured as an input pin
the MOS pull-up resistor is activated. To switch the pull-up resistor off the PDn has to be
cleared (zero) or the pin has to be configured as an output pin. The port pins are tristated when a reset condition becomes active, even if the clock is not running.
Table 32. DDDn Bits on Port D Pins
DDDnPORTDnI/OPull-upComment
00InputNoTri-state (high-Z)
01InputYesPDn will source current if ext. pulled low.
10OutputNoPush-pull Zero Output
11OutputNoPush-pull One Output
Note:n: 7,6...0, pin number
Alternate Functions of Port DThe alternate pin functions of Port D are:
• INT0..INT3 – Port D, Bits 0..3
External Interrupt sources 0 - 3. The PD0 - PD3 pins can serve as external active low
interrupt sources to the MCU. The internal pull-up MOS resistors can be activated as
described above. See the interrupt description for further details, and how to enable the
sources.
• IC1 – Port D, Bit 4
90
IC1, Input Capture pin for Timer/Counter1. When a positive or negative (selectable)
edge is applied to this pin, the contents of Timer/Counter1 is transferred to the
Timer/Counter1 Input Capture Register. The pin has to be configured as an input to
ATmega103(L)
0945G–09/01
ATmega103(L)
U
serve this function. See the Timer/Counter1 description on how to operate this function.
The internal pull-up MOS resistor can be activated as described above.
• T1 – Port D, Bit 6
T1, Timer/Counter1 counter source. See the timer description for further details.
• T2 – Port D, Bit 7
T2, Timer/Counter2 counter source. See the timer description for further details.
Port D SchematicsNote that all port pins are synchronized. The synchronization latches are, however, not
shown in the figures.
Figure 62. Port D Schematic Diagram (Pins PD0, PD1, PD2 and PD3)
Figure 65. Port D Schematic Diagram (Pins PD6 and PD7)
MOS
PULLUP
PDn
RL
RP
WRITE PORTD
WP:
WRITE DDRD
WD:
READ PORTD LATCH
RL:
READ PORTD PIN
RP:
READ DDRD
RD:
6, 7
n:
1, 2
m:
SENSE CONTROL
CSm2
CSm1
CSm0
ATmega103(L)
RD
RESET
R
D
Q
DDDn
C
WD
RESET
Q
PORTDn
WP
R
C
D
TIMERm CLOCK
SOURCE MUX
DATA BUS
Port EPort E is an 8-bit bi-directional I/O port with internal pull-up resistors.
Three I/O memory address locations are allocated for the Port E, one each for the Data
Register – PORTE, $03($23), Data Direction Register – DDRE, $02($22) and the Port E
Input Pins – PINE, $01($21). The Port E Input Pins address is read-only, while the Data
Register and the Data Direction Register are read/write.
The Port E output buffers can sink 20 mA. As inputs, Port E pins that are externally
pulled low will source current if the pull-up resistors are activated.
All Port E pins have alternate functions as shown in Table 33.
Table 33. Port E Pin Alternate Functions
Port PinAlternate Function
PE0PDI/RXD (Programming Data Input or UART Receive Pin)
PE1PDO/TXD (Programming Data Output or UART Transmit Pin)
PE2AC+ (Analog Comparator Positive Input)
PE3AC- (Analog Comparator Negative Input)
PE4INT4 (External Interrupt4 Input)
PE5INT5 (External Interrupt5 Input)
PE6INT6 (External Interrupt6 Input)
PE7INT7 (External Interrupt7 Input)
0945G–09/01
When the pins are used for the alternate function, the DDRE and PORTE registers have
to be set according to the alternate function description.
The Port E Input Pins address (PINE) is not a register; this address enables access to
the physical value on each Port E pin. When reading PORTE, the Port E Data Latch is
read and when reading PINE, the logical values present on the pins are read.
Port E as General Digital I/OPEn, general I/O pin: The DDEn bit in the DDRE register selects the direction of this pin.
If DDEn is set (one), PEn is configured as an output pin. If DDEn is cleared (zero), PEn
is configured as an input pin. If PEn is set (one) when configured as an input pin, the
MOS pull-up resistor is activated. To switch the pull-up resistor off, the PEn has to be
cleared (zero) or the pin has to be configured as an output pin.The port pins are tristated when a reset condition becomes active, even if the clock is not running.
Table 34. DDEn Bits on Port E Pins
DDEnPORTEnI/OPull-upComment
00InputNoTri-state (high-Z)
01InputYesPDn will source current if ext. pulled low.
10OutputNoPush-pull Zero Output
11OutputNoPush-pull One Output
Note:n: 7,6...0, pin number
Alternate Functions of Port EThe alternate pin functions of Port E are:
• PDI/RXD – Port E, Bit 0
PDI, Serial Programming Data Input. During Serial Program downloading, this pin is
used as data input line for the ATmega103(L).
RXD, UART Receive Pin. Receive Data (Data input pin for the UART). When the UART
receiver is enabled, this pin is configured as an input regardless of the value of DDRD0.
When the UART forces this pin to be an input, a logical “1” in PORTD0 will turn on the
internal pull-up.
• PDO/TXD – Port E, Bit 1
PDO, Serial Programming Data Output. During Serial Program downloading, this pin is
used as data output line for the ATmega103(L).
TXD, UART Transmit Pin.
94
ATmega103(L)
0945G–09/01
ATmega103(L)
• AC+ – Port E, Bit 2
AC+, Analog Comparator Positive Input. This pin is directly connected to the positive
input of the analog comparator.
• AC- – Port E, Bit 3
AC-, Analog Comparator Negative Input. This pin is directly connected to the negative
input of the analog comparator.
• INT4..INT7 – Port E, Bits 4 - 7
INT4..INT7, External Interrupt sources 4 - 7: The PE4 - PE7 pins can serve as external
interrupt sources to the MCU. Interrupts can be triggered by low level or positive or negative edge on these pins. The internal pull-up MOS resistors can be activated as
described above. See the interrupt description for further details, and how to enable the
sources.
Port E SchematicsNote that all port pins are synchronized. The synchronization latches are, however, not
shown in the figures.
Figure 66. Port E Schematic Diagram (Pin PE0)
0945G–09/01
95
Figure 67. Port E Schematic Diagram (Pin PE1)
Figure 68. Port E Schematic Diagram (Pin PE2)
MOS
PULLUP
PE2
RL
RP
WP:
WRITE PORTE
WD:
WRITE DDRE
RL:
READ PORTE LATCH
RP:
READ PORTE PIN
RD:
READ DDRE
RD
RESET
Q
DDE2
C
WD
RESET
Q
PORTE2
C
WP
TO COMPARATOR
D
TA B US
D
DA
AC+
96
ATmega103(L)
0945G–09/01
Figure 69. Port E Schematic Diagram (Pin PE3)
U
MOS
PULLUP
PE3
RL
RP
ATmega103(L)
RD
RESET
D
Q
DDE3
C
WD
RESET
Q
PORTE3
WP
D
C
DATA BUS
WP:
WRITE PORTE
WD:
WRITE DDRE
RL:
READ PORTE LATCH
RP:
READ PORTE PIN
RD:
READ DDRE
TO COMPARATOR
Figure 70. Port E Schematic Diagram (Pins PE4, PE5, PE6 and PE7)
RD
MOS
PULLUP
PEn
RL
RP
RESET
Q
DDEn
RESET
Q
PORTEn
R
D
C
WD
R
D
C
WP
AC-
S
TA B
DA
0945G–09/01
WP:
WD:
RL:
RP:
RD:
n:
WRITE PORTE
WRITE DDRE
READ PORTE LATCH
READ PORTE PIN
READ DDRE
4, 5, 6, 7
SENSE CONTROLINTn
ISCn1ISCn0
97
Port FPort F is an 8-bit input port.
U
One I/O memory location is allocated for Port F, the Port F Input Pins – PINF, $00 ($20).
All Port F pins are connected to the analog multiplexer, which further is connected to the
A/D converter. The digital input function of Port F can be used together with the A/D
converter, allowing the user to use some pins of Port F and digital inputs and other as
analog inputs, at the same time.
The Port F Input Pins address (PINF) is not a register; this address enables access to
the physical value on each Port F pin.
Figure 71. Port F Schematic Diagram (Pins PF7 - PF0)
RP
PFn
RP:n:READ PORTF PIN
0 - 7
TO ADC MUX
S
DATA B
AINn
98
ATmega103(L)
0945G–09/01
Memory
Programming
ATmega103(L)
Program and Data
Memory Lock Bits
The ATmega103(L) MCU provides two Lock bits that can be left unprogrammed (“1”) or
can be programmed (“0”) to obtain the additional features listed in Table 35. The Lock
bits can only be erased to “1” with the Chip Erase command.
Table 35. Lock Bit Protection Modes
Memory Lock BitsProtection Type
ModeLB1LB2
111No memory lock features enabled.
201Further programming of the Flash and EEPROM is disabled.
300Same as mode 2, and verify is also disabled.
Note:1. In Parallel mode, programming of the Fuse bits are also disabled. Program the Fuse
bits before programming the Lock bits.
Fuse BitsThe ATmega103(L) has four Fuse bits, SPIEN, SUT1..0 and EESAVE.
•When the SPIEN Fuse is programmed (“0”), Serial Program and Data Downloading
is enabled. Default value is programmed (“0”). The SPIEN Fuse is not accessible in
serial programming mode.
•When EESAVE is programmed, the EEPROM memory is preserved through the
Chip Erase cycle. Default value is unprogrammed (“1”). The EESAVE Fuse bit
cannot be programmed if any of the Lock bits are programmed.
•SUT1..0 Fuses: Determine the MCU start-up time. See Table 5 on page 26 for
further details. Default value is unprogrammed (“11”), which gives a nominal start-up
time of 16 ms.
(1)
The status of the Fuse bits is not affected by Chip Erase.
Signature BytesAll Atmel microcontrollers have a 3-byte signature code that identifies the device. This
code can be read in both serial and parallel mode. The three bytes reside in a separate
address space.
For the ATmega103 they are:
1. $000: $1E (indicates manufactured by Atmel)
2. $001: $97 (indicates 128K bytes Flash memory)
3. $002: $01 (indicates ATmega103 when signature byte $001 is $97)
Programming the Flash
and EEPROM
Atmel’s ATmega103(L) offers 128K bytes of In-System reprogrammable Flash memory
and 4K bytes of EEPROM data memory.
The ATmega103(L) is shipped with the On-chip Flash program and EEPROM data
memory arrays in the erased state (i.e., contents = $FF) and ready to be programmed.
This device supports a parallel programming mode and a serial programming mode.
The +12V supplied to the RESET
ming enable only, and no current of significance is drawn by this pin. The serial
programming mode provides a convenient way to download program and data into the
ATmega103(L) inside the user’s system.
pin in parallel programming mode is used for program-
0945G–09/01
99
The Flash program memory array on the ATmega103(L) is organized as 512 pages of
256 bytes each. When programming the Flash, the program data is latched into a page
buffer. This allows one page of program data to be programmed simultaneously in either
programming mode.
The EEPROM data memory array on the ATmega103(L) is programmed byte-by-byte in
either programming mode. An auto-erase cycle is provided within the self-timed
EEPROM write instruction in the serial programming mode.
During programming, the supply voltage must be in accordance with Table 36.
Table 36. Supply Voltage during Programming
PartSerial ProgrammingParallel Programming
ATmega1034.0 - 5.0V4.0 - 5.0V
ATmega103L3.2 - 3.6V3.2 - 5.0V
Parallel ProgrammingThis section describes how to parallel program and verify Flash program memory,
EEPROM data memory, Lock bits and Fuse bits in the ATmega103(L). Pulses are
assumed to be at least 500 ns unless otherwise noted.
Signal NamesIn this section, some pins of the ATmega103(L) are referenced by signal names describ-
ing their function during parallel programming (see Figure 72 and Table 37). Pins not
described in Table 37 are referenced by pin names.
The XA1/XA0 pins determine the action executed when the XTAL1 pin is given a positive pulse. The bit coding is shown in Table 38.
When pulsing WR
or OE, the command loaded determines the action executed. The
command is a byte where the different bits are assigned functions, as shown in Table
39.
Figure 72. Parallel Programming
RDY/BSY
OE
WR
BS1
XA0
XA1
PAGEL
PD1
PD2
PD3
PD4
PD5
PD6
PA0
RESET+12V
PD7
XTAL1
GND
ATmega103(L)
VCC
PB7 - PB0DATA
V
CC
100
ATmega103(L)
0945G–09/01
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.