ATMEGA64L [ATMEL]
8-bit AVR Microcontroller with 64K Bytes In-System Programmable Flash; 8位AVR微控制器具有64K字节的系统内可编程闪存型号: | ATMEGA64L |
厂家: | ATMEL |
描述: | 8-bit AVR Microcontroller with 64K Bytes In-System Programmable Flash |
文件: | 总363页 (文件大小:3295K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
Features
• High-performance, Low-power AVR® 8-bit Microcontroller
• Advanced RISC Architecture
– 130 Powerful Instructions – Most Single Clock Cycle Execution
– 32 x 8 General Purpose Working Registers + Peripheral Control Registers
– Fully Static Operation
– Up to 16 MIPS Throughput at 16 MHz
– On-chip 2-cycle Multiplier
• Non-volatile Program and Data Memories
– 64K Bytes of In-System Reprogrammable Flash
Endurance: 10,000 Write/Erase Cycles
– Optional Boot Code Section with Independent Lock Bits
In-System Programming by On-chip Boot Program
True Read-While-Write Operation
8-bit
Microcontroller
with 64K Bytes
In-System
Programmable
Flash
– 2K Bytes EEPROM
Endurance: 100,000 Write/Erase Cycles
– 4K Bytes Internal SRAM
– Up to 64K Bytes Optional External Memory Space
– Programming Lock for Software Security
– SPI Interface for In-System Programming
• JTAG (IEEE std. 1149.1 Compliant) Interface
– Boundary-scan Capabilities According to the JTAG Standard
– Extensive On-chip Debug Support
– Programming of Flash, EEPROM, Fuses, and Lock Bits through the JTAG Interface
• Peripheral Features
– Two 8-bit Timer/Counters with Separate Prescalers and Compare Modes
– Two Expanded 16-bit Timer/Counters with Separate Prescaler, Compare Mode, and
Capture Mode
– Real Time Counter with Separate Oscillator
– Two 8-bit PWM Channels
ATmega64
ATmega64L
– 6 PWM Channels with Programmable Resolution from 1 to 16 Bits
– 8-channel, 10-bit ADC
Preliminary
8 Single-ended Channels
7 Differential Channels
2 Differential Channels with Programmable Gain (1x, 10x, 200x)
– Byte-oriented Two-wire Serial Interface
– Dual Programmable Serial USARTs
– Master/Slave SPI Serial Interface
– Programmable Watchdog Timer with On-chip Oscillator
– On-chip Analog Comparator
• Special Microcontroller Features
– Power-on Reset and Programmable Brown-out Detection
– Internal Calibrated RC Oscillator
– External and Internal Interrupt Sources
– Six Sleep Modes: Idle, ADC Noise Reduction, Power-save, Power-down, Standby
and Extended Standby
– Software Selectable Clock Frequency
– ATmega103 Compatibility Mode Selected by a Fuse
– Global Pull-up Disable
• I/O and Packages
– 53 Programmable I/O Lines
– 64-lead TQFP and 64-pad MLF
• Operating Voltages
– 2.7 - 5.5V for ATmega64L
– 4.5 - 5.5V for ATmega64
• Speed Grades
– 0 - 8 MHz for ATmega64L
– 0 - 16 MHz for ATmega64
2490G–AVR–03/04
Pin Configuration
Figure 1. Pinout ATmega64
TQFP/MLF
PEN
RXD0/(PDI) PE0
(TXD0/PDO) PE1
(XCK0/AIN0) PE2
(OC3A/AIN1) PE3
(OC3B/INT4) PE4
(OC3C/INT5) PE5
(T3/INT6) PE6
1
2
3
4
5
6
7
8
9
48 PA3 (AD3)
47 PA4 (AD4)
46 PA5 (AD5)
45 PA6 (AD6)
44 PA7 (AD7)
43 PG2(ALE)
42 PC7 (A15)
41 PC6 (A14)
40 PC5 (A13)
39 PC4 (A12)
38 PC3 (A11)
37 PC2 (A10
36 PC1 (A9)
35 PC0 (A8)
34 PG1(RD)
33 PG0(WR)
(IC3/INT7) PE7
(SS) PB0 10
(SCK) PB1 11
(MOSI) PB2 12
(MISO) PB3 13
(OC0) PB4 14
(OC1A) PB5 15
(OC1B) PB6
16
Disclaimer
Typical values contained in this data sheet are based on simulations and characteriza-
tion of other AVR microcontrollers manufactured on the same process technology. Min
and Max values will be available after the device is characterized.
2
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Overview
The ATmega64 is a low-power CMOS 8-bit microcontroller based on the AVR enhanced RISC architecture. By executing
powerful instructions in a single clock cycle, the ATmega64 achieves throughputs approaching 1 MIPS per MHz, allowing
the system designer to optimize power consumption versus processing speed.
Block Diagram
Figure 2. Block Diagram
PF0 - PF7
PA0 - PA7
PC0 - PC7
VCC
GND
PORTA DRIVERS
PORTF DRIVERS
PORTC DRIVERS
AVCC
DATA REGISTER
PORTF
DATA DIR.
REG. PORTF
DATA REGISTER
PORTA
DATA DIR.
REG. PORTA
DATA REGISTER
PORTC
DATA DIR.
REG. PORTC
8-BIT DATA BUS
AGND
AREF
XTAL1
XTAL2
CALIB. OSC
INTERNAL
OSCILLATOR
ADC
OSCILLATOR
OSCILLATOR
PROGRAM
COUNTER
STACK
POINTER
WATCHDOG
TIMER
JTAG TAP
TIMING AND
CONTROL
PROGRAM
FLASH
MCU CONTROL
REGISTER
SRAM
ON-CHIP DEBUG
RESET
BOUNDARY-
SCAN
INSTRUCTION
REGISTER
TIMER/
COUNTERS
GENERAL
PURPOSE
REGISTERS
X
Y
Z
PROGRAMMING
LOGIC
INSTRUCTION
DECODER
INTERRUPT
UNIT
PEN
CONTROL
LINES
ALU
EEPROM
STATUS
REGISTER
2-WIRE SERIAL
INTERFACE
SPI
USART0
USART1
DATA REGISTER
PORTE
DATA DIR.
REG. PORTE
DATA REGISTER
PORTB
DATA DIR.
REG. PORTB
DATA REGISTER
PORTD
DATA DIR.
REG. PORTD
DATA REG. DATA DIR.
PORTG
REG. PORTG
PORTB DRIVERS
PORTD DRIVERS
PORTG DRIVERS
PORTE DRIVERS
PE0 - PE7
PB0 - PB7
PD0 - PD7
PG0 - PG4
The AVR core 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.
3
2490G–AVR–03/04
The ATmega64 provides the following features: 64K bytes of In-System Programmable
Flash with Read-While-Write capabilities, 2K bytes EEPROM, 4K bytes SRAM, 53 gen-
eral purpose I/O lines, 32 general purpose working registers, Real Time Counter (RTC),
four flexible Timer/Counters with compare modes and PWM, two USARTs, a byte ori-
ented Two-wire Serial Interface, an 8-channel, 10-bit ADC with optional differential input
stage with programmable gain, programmable Watchdog Timer with internal Oscillator,
an SPI serial port, IEEE std. 1149.1 compliant JTAG test interface, also used for
accessing the On-chip Debug system and programming, and six 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 asyn-
chronous timer continues to run, allowing the user to maintain a timer base while the
rest of the device is sleeping. The ADC Noise Reduction mode stops the CPU and all
I/O modules except asynchronous timer and ADC, to minimize switching noise during
ADC conversions. In Standby mode, the crystal/resonator Oscillator is running while the
rest of the device is sleeping. This allows very fast start-up combined with low power
consumption. In Extended Standby mode, both the main Oscillator and the asynchro-
nous timer continue to run.
The device is manufactured using Atmel’s high-density non-volatile memory technology.
The On-chip ISP Flash allows the program memory to be reprogrammed In-System
through an SPI serial interface, by a conventional non-volatile memory programmer, or
by an On-chip Boot program running on the AVR core. The Boot Program can use any
interface to download the Application Program in the Application Flash memory. Soft-
ware in the Boot Flash section will continue to run while the Application Flash section is
updated, providing true Read-While-Write operation. By combining an 8-bit RISC CPU
with In-System Self-Programmable Flash on a monolithic chip, the Atmel ATmega64 is
a powerful microcontroller that provides a highly-flexible and cost-effective solution to
many embedded control applications.
The ATmega64 AVR is supported with a full suite of program and system development
tools including: C compilers, macro assemblers, program debugger/simulators, In-Cir-
cuit Emulators, and evaluation kits.
ATmega103 and
ATmega64 Compatibility
The ATmega64 is a highly complex microcontroller where the number of I/O locations
supersedes the 64 I/O location reserved in the AVR instruction set. To ensure backward
compatibility with the ATmega103, all I/O locations present in ATmega103 have the
same location in ATmega64. Most additional I/O locations are added in an Extended I/O
space starting from 0x60 to 0xFF (i.e., in the ATmega103 internal RAM space). These
location can be reached by using LD/LDS/LDD and ST/STS/STD instructions only, not
by using IN and OUT instructions. The relocation of the internal RAM space may still be
a problem for ATmega103 users. Also, the increased number of Interrupt Vectors might
be a problem if the code uses absolute addresses. To solve these problems, an
ATmega103 compatibility mode can be selected by programming the fuse M103C. In
this mode, none of the functions in the Extended I/O space are in use, so the internal
RAM is located as in ATmega103. Also, the extended Interrupt Vectors are removed.
The ATmega64 is 100% pin compatible with ATmega103, and can replace the
ATmega103 on current printed circuit boards. The application note “Replacing
ATmega103 by ATmega64” describes what the user should be aware of replacing the
ATmega103 by an ATmega64.
4
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
ATmega103 Compatibility
Mode
By programming the M103C Fuse, the ATmega64 will be compatible with the
ATmega103 regards to RAM, I/O pins and Interrupt Vectors as described above. How-
ever, some new features in ATmega64 are not available in this compatibility mode,
these features are listed below:
•
One USART instead of two, asynchronous mode only. Only the eight least
significant bits of the Baud Rate Register is available.
•
One 16 bits Timer/Counter with two compare registers instead of two 16 bits
Timer/Counters with three compare registers.
•
•
•
•
•
•
Two-wire serial interface is not supported.
Port G serves alternate functions only (not a general I/O port).
Port F serves as digital input only in addition to analog input to the ADC.
Boot Loader capabilities is not supported.
It is not possible to adjust the frequency of the internal calibrated RC Oscillator.
The External Memory Interface can not release any Address pins for general I/O,
neither configure different wait states to different External Memory Address
sections.
•
•
•
Only EXTRF and PORF exist in the MCUCSR Register.
No timed sequence is required for Watchdog Timeout change.
Only low-level external interrupts can be used on four of the eight External Interrupt
sources.
•
•
•
Port C is output only.
USART has no FIFO buffer, so Data OverRun comes earlier.
The user must have set unused I/O bits to 0 in ATmega103 programs.
Pin Descriptions
VCC
Digital supply voltage.
Ground.
GND
Port A (PA7..PA0)
Port A is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for each
bit). The Port A output buffers have symmetrical drive characteristics with both high sink
and source capability. As inputs, Port A pins that are externally pulled low will source
current if the pull-up resistors are activated. The Port A pins are tri-stated when a reset
condition becomes active, even if the clock is not running.
Port A also serves the functions of various special features of the ATmega64 as listed
on page 71.
Port B (PB7..PB0)
Port B is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for each
bit). The Port B output buffers have symmetrical drive characteristics with both high sink
and source capability. As inputs, Port B pins that are externally pulled low will source
current if the pull-up resistors are activated. The Port B pins are tri-stated when a reset
condition becomes active, even if the clock is not running.
Port B also serves the functions of various special features of the ATmega64 as listed
on page 72.
5
2490G–AVR–03/04
Port C (PC7..PC0)
Port C is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for each
bit). The Port C output buffers have symmetrical drive characteristics with both high sink
and source capability. As inputs, Port C pins that are externally pulled low will source
current if the pull-up resistors are activated. The Port C pins are tri-stated when a reset
condition becomes active, even if the clock is not running.
Port C also serves the functions of special features of the ATmega64 as listed on page
75. In ATmega103 compatibility mode, Port C is output only, and the port C pins are not
tri-stated when a reset condition becomes active.
Port D (PD7..PD0)
Port E (PE7..PE0)
Port F (PF7..PF0)
Port D is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for each
bit). The Port D output buffers have symmetrical drive characteristics with both high sink
and source capability. As inputs, Port D pins that are externally pulled low will source
current if the pull-up resistors are activated. The Port D pins are tri-stated when a reset
condition becomes active, even if the clock is not running.
Port D also serves the functions of various special features of the ATmega64 as listed
on page 76.
Port E is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for each
bit). The Port E output buffers have symmetrical drive characteristics with both high sink
and source capability. As inputs, Port E pins that are externally pulled low will source
current if the pull-up resistors are activated. The Port E pins are tri-stated when a reset
condition becomes active, even if the clock is not running.
Port E also serves the functions of various special features of the ATmega64 as listed
on page 79.
Port F serves as the analog inputs to the A/D Converter.
Port F also serves as an 8-bit bi-directional I/O port, if the A/D Converter is not used.
Port pins can provide internal pull-up resistors (selected for each bit). The Port F output
buffers have symmetrical drive characteristics with both high sink and source capability.
As inputs, Port F pins that are externally pulled low will source current if the pull-up
resistors are activated. The Port F pins are tri-stated when a reset condition becomes
active, even if the clock is not running. If the JTAG interface is enabled, the pull-up resis-
tors on pins PF7(TDI), PF5(TMS) and PF4(TCK) will be activated even if a reset occurs.
The TDO pin is tri-stated unless TAP states that shift out data are entered.
Port F also serves the functions of the JTAG interface.
In ATmega103 compatibility mode, Port F is an input port only.
Port G (PG4..PG0)
Port G is a 5-bit bi-directional I/O port with internal pull-up resistors (selected for each
bit). The Port G output buffers have symmetrical drive characteristics with both high sink
and source capability. As inputs, Port G pins that are externally pulled low will source
current if the pull-up resistors are activated. The Port G pins are tri-stated when a reset
condition becomes active, even if the clock is not running.
Port G also serves the functions of various special features.
In ATmega103 compatibility mode, these pins only serves as strobes signals to the
external memory as well as input to the 32 kHz Oscillator, and the pins are initialized to
PG0 = 1, PG1 = 1, and PG2 = 0 asynchronously when a reset condition becomes
active, even if the clock is not running. PG3 and PG4 are Oscillator pins.
6
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
RESET
Reset input. A low level on this pin for longer than the minimum pulse length will gener-
ate a reset, even if the clock is not running. The minimum pulse length is given in Table
19 on page 50. Shorter pulses are not guaranteed to generate a reset.
XTAL1
XTAL2
AVCC
Input to the inverting Oscillator amplifier and input to the internal clock operating circuit.
Output from the inverting Oscillator amplifier.
AVCC is the supply voltage pin for Port F and the A/D Converter. It should be externally
connected to VCC, even if the ADC is not used. If the ADC is used, it should be con-
nected to VCC through a low-pass filter.
AREF
PEN
AREF is the analog reference pin for the A/D Converter.
This is a programming enable pin for the SPI Serial Programming mode. By holding this
pin low during a Power-on Reset, the device will enter the SPI Serial Programming
mode. PEN has no function during normal operation.
About Code
Examples
This datasheet contains simple code examples that briefly show how to use various
parts of the device. These code examples assume that the part specific header file is
included before compilation. Be aware that not all C compiler vendors include bit defini-
tions in the header files and interrupt handling in C is compiler dependent. Please
confirm with the C compiler documentation for more details.
7
2490G–AVR–03/04
AVR CPU Core
Introduction
This section discusses the AVR core architecture in general. The main function of the
CPU core is to ensure correct program execution. The CPU must therefore be able to
access memories, perform calculations, control peripherals, and handle interrupts.
Architectural Overview
Figure 3. Block Diagram of the AVR MCU Architecture
Data Bus 8-bit
Program
Counter
Status
and Control
Flash
Program
Memory
Interrupt
Unit
32 x 8
General
Purpose
Registrers
Instruction
Register
SPI
Unit
Instruction
Decoder
Watchdog
Timer
ALU
Analog
Comparator
Control Lines
I/O Module1
I/O Module 2
I/O Module n
Data
SRAM
EEPROM
I/O Lines
In order to maximize performance and parallelism, the AVR uses a Harvard architecture
– with separate memories and buses for program and data. Instructions in the program
memory are executed with a single level pipelining. While one instruction is being exe-
cuted, 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 Reprogrammable Flash memory.
The fast-access Register File contains 32 x 8-bit general purpose working registers with
a single clock cycle access time. This allows single-cycle Arithmetic Logic Unit (ALU)
operation. In a typical ALU operation, two operands are output from the Register File,
the operation is executed, and the result is stored back in the Register File – in one
clock cycle.
8
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Six of the 32 registers can be used as three 16-bit indirect address register pointers for
Data Space addressing – enabling efficient address calculations. One of the these
address pointers can also be used as an address pointer for look up tables in Flash pro-
gram memory. These added function registers are the 16-bit X-, Y-, and Z-register,
described later in this section.
The ALU supports arithmetic and logic operations between registers or between a con-
stant and a register. Single register operations can also be executed in the ALU. After
an arithmetic operation, the Status Register is updated to reflect information about the
result of the operation.
Program flow is provided by conditional and unconditional jump and call instructions,
able to directly address the whole address space. Most AVR instructions have a single
16-bit word format. Every program memory address contains a 16- or 32-bit instruction.
Program Flash memory space is divided in two sections, the Boot program section and
the Application program section. Both sections have dedicated Lock bits for write and
read/write protection. The SPM instruction that writes into the Application Flash memory
section must reside in the Boot program section.
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 Stack Pointer SP is read/write accessible in the I/O
space. The data SRAM can easily be accessed through the five different addressing
modes supported in the AVR architecture.
The memory spaces in the AVR architecture are all linear and regular memory maps.
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 interrupts have a separate Interrupt
Vector in the Interrupt Vector table. The interrupts have priority in accordance with their
Interrupt Vector position. The lower the Interrupt Vector address, the higher the priority.
The I/O memory space contains 64 addresses which can be accessed directly, or as the
Data Space locations following those of the Register File, 0x20 - 0x5F. In addition, the
ATmega64 has Extended I/O space from 0x60 - 0xFF in SRAM where only the
ST/STS/STD and LD/LDS/LDD instructions can be used.
ALU – Arithmetic Logic
Unit
The high-performance AVR ALU operates in direct connection with all the 32 general
purpose working registers. Within a single clock cycle, arithmetic operations between
general purpose registers or between a register and an immediate are executed. The
ALU operations are divided into three main categories – arithmetic, logical, and bit-func-
tions. Some implementations of the architecture also provide a powerful multiplier
supporting both signed/unsigned multiplication and fractional format. See the “Instruc-
tion Set” section for a detailed description.
9
2490G–AVR–03/04
Status Register
The Status Register contains information about the result of the most recently executed
arithmetic instruction. This information can be used for altering program flow in order to
perform conditional operations. Note that the Status Register is updated after all ALU
operations, as specified in the Instruction Set Reference. This will in many cases
remove the need for using the dedicated compare instructions, resulting in faster and
more compact code.
The Status Register is not automatically stored when entering an interrupt routine and
restored when returning from an interrupt. This must be handled by software.
The AVR Status Register – SREG – is defined as:
Bit
7
I
6
T
5
H
4
S
3
V
2
N
1
Z
0
C
SREG
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
• Bit 7 – I: Global Interrupt Enable
The Global Interrupt Enable bit must be set for the interrupts to be enabled. The individ-
ual interrupt enable control is then performed in separate control registers. If the Global
Interrupt Enable Register is cleared, 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. The I-
bit can also be set and cleared in software with the SEI and CLI instructions, as
described in the instruction set reference.
• Bit 6 – T: Bit Copy Storage
The Bit Copy instructions BLD (Bit LoaD) and BST (Bit STore) use the T-bit as source or
destination for the operated bit. A bit from a register in the Register File can be copied
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. Half Carry is
useful in BCD arithmetic. See the “Instruction Set Description” for detailed information.
• Bit 4 – S: Sign Bit, S = N ⊕ V
The S-bit is always an exclusive or between the Negative Flag N and the Two’s Comple-
ment Overflow Flag V. See the “Instruction Set Description” 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” for detailed information.
• Bit 2 – N: Negative Flag
The Negative Flag N indicates a negative result in an arithmetic or logic operation. See
the “Instruction Set Description” for detailed information.
• Bit 1 – Z: Zero Flag
The Zero Flag Z indicates a zero result in an arithmetic or logic operation. See the
“Instruction Set Description” for detailed information.
10
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• Bit 0 – C: Carry Flag
The Carry Flag C indicates a carry in an arithmetic or logic operation. See the “Instruc-
tion Set Description” for detailed information.
General Purpose
Register File
The Register File is optimized for the AVR Enhanced RISC instruction set. In order to
achieve the required performance and flexibility, the following input/output schemes are
supported by the Register File:
•
•
•
•
One 8-bit output operand and one 8-bit result input.
Two 8-bit output operands and one 8-bit result input.
Two 8-bit output operands and one 16-bit result input.
One 16-bit output operand and one 16-bit result input.
Figure 4 shows the structure of the 32 general purpose working registers in the CPU.
Figure 4. AVR CPU General Purpose Working Registers
7
0
Addr.
0x00
0x01
0x02
R0
R1
R2
…
R13
R14
R15
R16
R17
…
0x0D
0x0E
0x0F
0x10
0x11
General
Purpose
Working
Registers
R26
R27
R28
R29
R30
R31
0x1A
0x1B
0x1C
0x1D
0x1E
0x1F
X-register Low Byte
X-register High Byte
Y-register Low Byte
Y-register High Byte
Z-register Low Byte
Z-register High Byte
Most of the instructions operating on the Register File have direct access to all registers,
and most of them are single cycle instructions.
As shown in Figure 4, 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 physi-
cally implemented as SRAM locations, this memory organization provides great
flexibility in access of the registers, as the X-, Y-, and Z-pointer registers can be set to
index any register in the file.
11
2490G–AVR–03/04
X-, Y-, and Z-register
The registers R26..R31 have some added functions to their general purpose usage.
These registers are 16-bit address pointers for indirect addressing of the data space.
The three indirect address registers X, Y, and Z are defined as described in Figure 5.
Figure 5. The X-, Y-, and Z-Registers
15
XH
XL
0
0
X - register
7
0
7
R27 (0x1B)
R26 (0x1A)
15
YH
ZH
YL
ZL
0
0
Y - register
Z - register
7
0
0
7
R29 (0x1D)
R28 (0x1C)
15
0
0
7
7
R31 (0x1F)
R30 (0x1E)
In the different addressing modes these address registers have functions as fixed dis-
placement, automatic increment, and automatic decrement (see the Instruction Set
Reference for details).
Stack Pointer
The Stack is mainly used for storing temporary data, for storing local variables and for
storing return addresses after interrupts and subroutine calls. The Stack Pointer Regis-
ter always points to the top of the Stack. Note that the Stack is implemented as growing
from higher memory locations to lower memory locations. This implies that a Stack
PUSH command decreases the Stack Pointer. If software reads the Program Counter
from the Stack after a call or an interrupt, unused bits (bit 15) should be masked out.
The Stack Pointer points to the data SRAM Stack area where the subroutine and inter-
rupt 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 0x60. The Stack Pointer is decremented by one
when data is pushed onto the Stack with the PUSH instruction, and it is decremented by
two when the return address is pushed onto the Stack with subroutine call or interrupt.
The Stack Pointer is incremented by one when data is popped from the Stack with the
POP instruction, and it is incremented by two when data is popped from the Stack with
return from subroutine RET or return from interrupt RETI.
The AVR Stack Pointer is implemented as two 8-bit registers in the I/O space. The num-
ber of bits actually used is implementation dependent. Note that the data space in some
implementations of the AVR architecture is so small that only SPL is needed. In this
case, the SPH Register will not be present.
Bit
15
SP15
SP7
7
14
SP14
SP6
6
13
SP13
SP5
5
12
SP12
SP4
4
11
SP11
SP3
3
10
SP10
SP2
2
9
SP9
SP1
1
8
SP8
SP0
0
SPH
SPL
Read/Write
Initial Value
R/W
R/W
0
R/W
R/W
0
R/W
R/W
0
R/W
R/W
0
R/W
R/W
0
R/W
R/W
0
R/W
R/W
0
R/W
R/W
0
0
0
0
0
0
0
0
0
12
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Instruction Execution
Timing
This section describes the general access timing concepts for instruction execution. The
AVR CPU is driven by the CPU clock clkCPU, directly generated from the selected clock
source for the chip. No internal clock division is used.
Figure 6 shows the parallel instruction fetches and instruction executions enabled by the
Harvard architecture and the fast-access Register File concept. This is the basic pipelin-
ing 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 6. The Parallel Instruction Fetches and Instruction Executions
T1
T2
T3
T4
clkCPU
1st Instruction Fetch
1st Instruction Execute
2nd Instruction Fetch
2nd Instruction Execute
3rd Instruction Fetch
3rd Instruction Execute
4th Instruction Fetch
Figure 7 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 7. Single Cycle ALU Operation
T1
T2
T3
T4
clkCPU
Total Execution Time
Register Operands Fetch
ALU Operation Execute
Result Write Back
Reset and Interrupt
Handling
The AVR provides several 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 which must be written logic one together
with the Global Interrupt Enable bit in the Status Register in order to enable the interrupt.
Depending on the Program Counter value, interrupts may be automatically disabled
when Boot Lock bits BLB02 or BLB12 are programmed. This feature improves software
security. See the section “Memory Programming” on page 290 for details.
The lowest addresses in the program memory space are by default defined as the Reset
and Interrupt Vectors. The complete list of vectors is shown in “Interrupts” on page 59.
The list also determines the priority levels of the different interrupts. The lower the
address the higher is the priority level. RESET has the highest priority, and next is INT0
– the External Interrupt Request 0. The Interrupt Vectors can be moved to the start of
the Boot Flash section by setting the IVSEL bit in the MCU Control Register (MCUCR).
Refer to “Interrupts” on page 59 for more information. The Reset Vector can also be
13
2490G–AVR–03/04
moved to the start of the Boot Flash section by programming the BOOTRST Fuse, see
“Boot Loader Support – Read-While-Write Self-programming” on page 277.
When an interrupt occurs, the Global Interrupt Enable I-bit is cleared and all interrupts
are disabled. The user software can write logic one to the I-bit to enable nested inter-
rupts. All enabled interrupts can then interrupt the current interrupt routine. The I-bit is
automatically set when a Return from Interrupt instruction – RETI – is executed.
There are basically two types of interrupts. The first type is triggered by an event that
sets the interrupt flag. For these interrupts, the Program Counter is vectored to the
actual Interrupt Vector in order to execute the interrupt handling routine, and hardware
clears the corresponding interrupt flag. Interrupt flags can also be cleared by writing a
logic one to the flag bit position(s) to be cleared. If an interrupt condition occurs while the
corresponding interrupt enable bit is cleared, the interrupt flag will be set and remem-
bered until the interrupt is enabled, or the flag is cleared by software. Similarly, if one or
more interrupt conditions occur while the Global Interrupt Enable bit is cleared, the cor-
responding interrupt flag(s) will be set and remembered until the Global Interrupt Enable
bit is set, and will then be executed by order of priority.
The second type of interrupts will trigger as long as the interrupt condition is present.
These interrupts do not necessarily have interrupt flags. If the interrupt condition disap-
pears before the interrupt is enabled, the interrupt will not be triggered.
When the AVR exits from an interrupt, it will always return to the main program and exe-
cute one more instruction before any pending interrupt is served.
Note that the Status Register is not automatically stored when entering an interrupt rou-
tine, nor restored when returning from an interrupt routine. This must be handled by
software.
When using the CLI instruction to disable interrupts, the interrupts will be immediately
disabled. No interrupt will be executed after the CLI instruction, even if it occurs simulta-
neously with the CLI instruction. The following example shows how this can be used to
avoid interrupts during the timed EEPROM write sequence.
Assembly Code Example
in r16, SREG
; store SREG value
cli ; disable interrupts during timed sequence
sbi EECR, EEMWE
sbi EECR, EEWE
out SREG, r16
; start EEPROM write
; restore SREG value (I-bit)
C Code Example
char cSREG;
cSREG = SREG; /* store SREG value */
/* disable interrupts during timed sequence */
_CLI();
EECR |= (1<<EEMWE); /* start EEPROM write */
EECR |= (1<<EEWE);
SREG = cSREG; /* restore SREG value (I-bit) */
14
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
When using the SEI instruction to enable interrupts, the instruction following SEI will be
executed before any pending interrupts, as shown in this example.
Assembly Code Example
sei
; set global interrupt enable
sleep ; enter sleep, waiting for interrupt
; note: will enter sleep before any pending
; interrupt(s)
C Code Example
_SEI(); /* set global interrupt enable */
_SLEEP(); /* enter sleep, waiting for interrupt */
/* note: will enter sleep before any pending interrupt(s) */
Interrupt Response Time
The interrupt execution response for all the enabled AVR interrupts is four clock cycles
minimum. After four clock cycles the program vector address for the actual interrupt
handling routine is executed. During this four clock cycle period, the Program Counter is
pushed onto the Stack. 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. If an interrupt
occurs when the MCU is in sleep mode, the interrupt execution response time is
increased by four clock cycles. This increase comes in addition to the start-up time from
the selected sleep mode.
A return from an interrupt handling routine takes four clock cycles. During these four
clock cycles, the Program Counter (two bytes) is popped back from the Stack, the Stack
Pointer is incremented by two, and the I-bit in SREG is set.
15
2490G–AVR–03/04
AVR ATmega64
Memories
This section describes the different memories in the ATmega64. The AVR architecture
has two main memory spaces, the Data Memory and the Program Memory space. In
addition, the ATmega64 features an EEPROM Memory for data storage. All three mem-
ory spaces are linear and regular.
In-System
Reprogrammable Flash
Program Memory
The ATmega64 contains 64K bytes On-chip In-System Reprogrammable Flash memory
for program storage. Since all AVR instructions are 16 or 32 bits wide, the Flash is orga-
nized as 32K x 16. For software security, the Flash Program memory space is divided
into two sections, Boot Program section and Application Program section.
The Flash memory has an endurance of at least 10,000 write/erase cycles. The
ATmega64 Program Counter (PC) is 15 bits wide, thus addressing the 32K program
memory locations. The operation of Boot Program section and associated Boot Lock
bits for software protection are described in detail in “Boot Loader Support – Read-
While-Write Self-programming” on page 277. “Memory Programming” on page 290 con-
tains a detailed description on Flash programming in SPI, JTAG, or Parallel
Programming mode.
Constant tables can be allocated within the entire program memory address space (see
the LPM – Load Program Memory instruction description).
Timing diagrams for instruction fetch and execution are presented in “Instruction Execu-
tion Timing” on page 13.
Figure 8. Program Memory Map
$0000
Application Flash Section
Boot Flash Section
$7FFF
16
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
SRAM Data Memory
The ATmega64 supports two different configurations for the SRAM data memory as
listed in Table 1.
Table 1. Memory Configurations
Internal SRAM
Data Memory
External SRAM
Data Memory
Configuration
Normal mode
4096
4000
up to 64K
up to 64K
ATmega103 compatibility mode
Figure 9 on page 18 shows how the ATmega64 SRAM Memory is organized.
The ATmega64 is a complex microcontroller with more peripheral units than can be sup-
ported within the 64 locations reserved in the Opcode for the IN and OUT instructions.
For the Extended I/O space from 0x60 - 0xFF in SRAM, only the ST/STS/STD and
LD/LDS/LDD instructions can be used. The Extended I/O space does not exist when the
ATmega64 is in the ATmega103 compatibility mode.
The first 4,352 data memory locations address both the Register File, the I/O memory,
Extended I/O memory, and the internal data SRAM. The first 32 locations address the
Register File, the next 64 location the standard I/O memory, then 160 locations of
Extended I/O memory, and the next 4,096 locations address the internal data SRAM.
In ATmega103 compatibility mode, the first 4,096 data memory locations address both
the Register File, the I/O memory and the internal data SRAM. The first 32 locations
address the Register File, the next 64 location the standard I/O memory, and the next
4,000 locations address the internal data SRAM.
An optional external data SRAM can be used with the ATmega64. 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. The Register File, I/O, Extended I/O
and internal SRAM occupy the lowest 4,352 bytes in Normal mode, and the lowest
4,096 bytes in the ATmega103 compatibility mode (Extended I/O not present), so when
using 64KB (65,536 bytes) of External memory, 61,184 Bytes of External memory are
available in Normal mode, and 61,440 Bytes in ATmega103 compatibility mode. See
“External Memory Interface” on page 25 for details on how to take advantage of the
external memory map.
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 (PG0 and PG1) are inactive during the whole access
cycle. 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, LDD, STD,
PUSH, and POP take one additional clock cycle. If the Stack is placed in external
SRAM, interrupts, subroutine calls and returns take three clock cycles extra because the
2-byte Program Counter is pushed and popped, and external memory access does not
take advantage of the internal pipeline memory access. When external SRAM interface
is used with wait state, one-byte external access takes two, three, or four additional
clock cycles for one, two, and three wait states respectively. Interrupt, subroutine calls
and returns will need five, seven, or nine clock cycles more than specified in the AVR
Instruction Set manual for one, two, and three waitstates.
17
2490G–AVR–03/04
The five different addressing modes for the data memory cover: Direct, Indirect with Dis-
placement, Indirect, Indirect with Pre-decrement, and Indirect with Post-increment. In
the Register File, registers R26 to R31 feature the indirect addressing pointer registers.
The direct addressing reaches the entire data space.
The Indirect with Displacement mode reaches 63 address locations from the base
address given by the Y- or Z-register.
When using register indirect addressing modes with automatic pre-decrement and post-
increment, the address registers X, Y, and Z are decremented or incremented.
The 32 general purpose working registers, 64 I/O Registers, 160 extended I/O Regis-
ters, and the 4,096 bytes of internal data SRAM in the ATmega64 are all accessible
through all these addressing modes. The Register File is described in “General Purpose
Register File” on page 11.
Figure 9. Data Memory Map
Memory Configuration A
Memory Configuration B
Data Memory
Data Memory
$0000 - $001F
$0020 - $005F
160 Ext I/O Reg. $0060 - $00FF
32 Registers
64 I/O Registers $0020 - $005F
$0060
$0000 - $001F
32 Registers
64 I/O Registers
Internal SRAM
(4000 x 8)
$0100
Internal SRAM
$0FFF
$1000
(4096 x 8)
$10FF
$1100
External SRAM
(0 - 64K x 8)
External SRAM
(0 - 64K x 8)
$FFFF
$FFFF
18
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Data Memory Access Times
This section describes the general access timing concepts for internal memory access.
The internal data SRAM access is performed in two clkCPU cycles as described in Figure
10.
Figure 10. On-chip Data SRAM Access Cycles
T1
T2
T3
clkCPU
Address Valid
Compute Address
Address
Data
WR
Data
RD
Memory Access Instruction
Next Instruction
EEPROM Data Memory
The ATmega64 contains 2K bytes of data EEPROM memory. It is organized as a sepa-
rate 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 in the following, specifying the EEPROM Address Registers,
the EEPROM Data Register, and the EEPROM Control Register.
“Memory Programming” on page 290 contains a detailed description on EEPROM pro-
gramming in SPI, JTAG, or Parallel Programming mode.
EEPROM Read/Write Access
The EEPROM Access Registers are accessible in the I/O space.
The write access time for the EEPROM is given in Table 2 on page 22. A self-timing
function, however, lets the user software detect when the next byte can be written. If the
user code contains instructions that write the EEPROM, some precautions must be
taken. In heavily filtered power supplies, VCC is likely to rise or fall slowly on Power-
up/down. This causes the device for some period of time to run at a voltage lower than
specified as minimum for the clock frequency used. See “Preventing EEPROM Corrup-
tion” on page 24. for details on how to avoid problems in these situations.
In order to prevent unintentional EEPROM writes, a specific write procedure must be fol-
lowed. Refer to the description of the EEPROM Control Register for details on this.
When the EEPROM is read, the CPU is halted for four clock cycles before the next
instruction is executed. When the EEPROM is written, the CPU is halted for two clock
cycles before the next instruction is executed.
19
2490G–AVR–03/04
EEPROM Address Register –
EEARH and EEARL
Bit
15
14
13
12
11
10
EEAR10
EEAR2
2
9
EEAR9
EEAR1
1
8
EEAR8
EEAR0
0
–
–
–
–
–
EEARH
EEARL
EEAR7
7
R
EEAR6
6
R
EEAR5
5
R
EEAR4
4
R
EEAR3
3
R
Read/Write
Initial Value
R/W
R/W
X
R/W
R/W
X
R/W
R/W
X
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
X
X
X
X
X
X
X
X
• Bits 15..11 – Res: Reserved Bits
These are reserved bits and will always read as zero. When writing to this address loca-
tion, write these bits to zero for compatibility with future devices.
• Bits 10..0 – EEAR10..0: EEPROM Address
The EEPROM Address Registers – EEARH and EEARL specify the EEPROM address
in the 2K bytes EEPROM space. The EEPROM data bytes are addressed linearly
between 0 and 2,048. The Initial Value of EEAR is undefined. A proper value must be
written before the EEPROM may be accessed.
EEPROM Data Register –
EEDR
Bit
7
6
5
4
3
2
1
0
MSB
R/W
0
LSB
R/W
0
EEDR
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
• 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 oper-
ation, the EEDR contains the data read out from the EEPROM at the address given by
EEAR.
EEPROM Control Register –
EECR
Bit
7
–
R
0
6
–
R
0
5
–
R
0
4
–
R
0
3
EERIE
R/W
0
2
EEMWE
R/W
0
1
EEWE
R/W
X
0
EERE
R/W
0
EECR
Read/Write
Initial Value
• Bits 7..4 – Res: Reserved Bits
These bits are reserved bits in the ATmega64 and will always read as zero.
• Bit 3 – EERIE: EEPROM Ready Interrupt Enable
Writing EERIE to one enables the EEPROM Ready Interrupt if the I-bit in SREG is set.
Writing EERIE to zero disables the interrupt. The EEPROM Ready Interrupt generates a
constant interrupt when EEWE is cleared.
20
ATmega64(L)
2490G–AVR–03/04
ATmega64(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 written to one, writing EEWE to one within four clock cycles
will write data to the EEPROM at the selected address. If EEMWE is zero, writing EEWE
to one will have no effect. When EEMWE has been written to one by software, hardware
clears the bit to zero after four clock cycles. See the description of the EEWE bit for an
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 one is written to EEWE,
otherwise no EEPROM write takes place. The following procedure should be followed
when writing the EEPROM (the order of steps 3 and 4 is not essential):
1. Wait until EEWE becomes zero.
2. Wait until SPMEN in SPMCSR becomes zero.
3. Write new EEPROM address to EEAR (optional).
4. Write new EEPROM data to EEDR (optional).
5. Write a logical one to the EEMWE bit while writing a zero to EEWE in EECR.
6. Within four clock cycles after setting EEMWE, write a logical one to EEWE.
The EEPROM can not be programmed during a CPU write to the Flash memory. The
software must check that the Flash programming is completed before initiating a new
EEPROM write. Step 2 is only relevant if the software contains a Boot Loader allowing
the CPU to program the Flash. If the Flash is never being updated by the CPU, step 2
can be omitted. See “Boot Loader Support – Read-While-Write Self-programming” on
page 277 for details about Boot programming.
Caution: An interrupt between step 5 and step 6 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 or EEDR Register 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 has elapsed, the EEWE bit is cleared 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 written to a logic
one to trigger the EEPROM read. The EEPROM read access takes one instruction, and
the requested data is available immediately. When the EEPROM is read, 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, it is neither possible to read the EEPROM, nor to change the EEAR
Register.
The calibrated Oscillator is used to time the EEPROM accesses. Table 2 lists the typical
programming time for EEPROM access from the CPU.
21
2490G–AVR–03/04
Table 2. EEPROM Programming Time(1)
Number of Calibrated RC
Symbol
Oscillator Cycles
Typ Programming Time
EEPROM write (from CPU)
8448
8.4 ms
Note:
1. Uses 1 MHz clock, independent of CKSEL Fuse settings.
The following code examples show one assembly and one C function for writing to the
EEPROM. The examples assume that interrupts are controlled (e.g., by disabling inter-
rupts globally) so that no interrupts will occur during execution of these functions. The
examples also assume that no Flash boot loader is present in the software. If such code
is present, the EEPROM write function must also wait for any ongoing SPM command to
finish.
Assembly Code Example
EEPROM_write:
; Wait for completion of previous write
sbic EECR,EEWE
rjmp EEPROM_write
; Set up address (r18:r17) in address register
out EEARH, r18
out EEARL, r17
; Write data (r16) to data register
out EEDR,r16
; Write logical one to EEMWE
sbi EECR,EEMWE
; Start eeprom write by setting EEWE
sbi EECR,EEWE
ret
C Code Example
void EEPROM_write(unsigned int uiAddress, unsigned char ucData)
{
/* Wait for completion of previous write */
while(EECR & (1<<EEWE))
;
/* Set up address and data registers */
EEAR = uiAddress;
EEDR = ucData;
/* Write logical one to EEMWE */
EECR |= (1<<EEMWE);
/* Start eeprom write by setting EEWE */
EECR |= (1<<EEWE);
}
22
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
The next code examples show assembly and C functions for reading the EEPROM. The
examples assume that interrupts are controlled so that no interrupts will occur during
execution of these functions.
Assembly Code Example
EEPROM_read:
; Wait for completion of previous write
sbic EECR,EEWE
rjmp EEPROM_read
; Set up address (r18:r17) in address register
out EEARH, r18
out EEARL, r17
; Start eeprom read by writing EERE
sbi EECR,EERE
; Read data from data register
in
r16,EEDR
ret
C Code Example
unsigned char EEPROM_read(unsigned int uiAddress)
{
/* Wait for completion of previous write */
while(EECR & (1<<EEWE))
;
/* Set up address register */
EEAR = uiAddress;
/* Start eeprom read by writing EERE */
EECR |= (1<<EERE);
/* Return data from data register */
return EEDR;
}
EEPROM Write During Power- When entering Power-down Sleep mode while an EEPROM write operation is active,
down Sleep Mode
the EEPROM write operation will continue, and will complete before the Write Access
time has passed. However, when the write operation is completed, the crystal oscillator
continues running, and as a consequence, the device does not enter Power-down
entirely. It is therefore recommended to verify that the EEPROM write operation is com-
pleted before entering Power-down.
23
2490G–AVR–03/04
Preventing EEPROM
Corruption
During periods of low VCC, the EEPROM data can be corrupted because the supply volt-
age is too low for the CPU and the EEPROM to operate properly. These issues are the
same as for board level systems using 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. Secondly, the CPU itself can execute instructions incorrectly, if the
supply voltage is too low.
EEPROM data corruption can easily be avoided by following this design
recommendation:
Keep the AVR RESET active (low) during periods of insufficient power supply volt-
age. This can be done by enabling the internal Brown-out Detector (BOD). If the
detection level of the internal BOD does not match the needed detection level, an
external low VCC Reset Protection circuit can be used. If a reset occurs while a write
operation is in progress, the write operation will be completed provided that the
power supply voltage is sufficient.
I/O Memory
The I/O space definition of the ATmega64 is shown in “Register Summary” on page 342.
All ATmega64 I/Os and peripherals are placed in the I/O space. All I/O locations may be
accessed by the LD/LDS/LDD and ST/STS/STD instructions, transferring data between
the 32 general purpose working registers and the I/O space. I/O Registers within the
address range 0x00 - 0x1F are directly bit-accessible using the SBI and CBI instruc-
tions. In these registers, the value of single bits can be checked by using the SBIS and
SBIC instructions. Refer to the instruction set section for more details. When using the
I/O specific commands IN and OUT, the I/O addresses 0x00 - 0x3F must be used.
When addressing I/O Registers as data space using LD and ST instructions, 0x20 must
be added to these addresses. The ATmega64 is a complex microcontroller with more
peripheral units than can be supported within the 64 location reserved in Opcode for the
IN and OUT instructions. For the Extended I/O space from 0x60 - 0xFF in SRAM, only
the ST/STS/STD and LD/LDS/LDD instructions can be used. The Extended I/O space is
replaced with SRAM locations when the ATmega64 is in the ATmega103 compatibility
mode.
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 one 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 reg-
isters 0x00 to 0x1F only.
The I/O and peripherals control registers are explained in later sections.
24
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
External Memory
Interface
With all the features that the External Memory Interface provides, it is well suited to
operate as an interface to memory devices such as external SRAM and Flash, and
peripherals such as LCD-display, A/D, and D/A. The main features are:
•
•
Four different wait-state settings (Including no wait-state).
Independent wait-state setting for different external memory sectors (configurable
sector size).
•
•
The number of bits dedicated to address high byte is selectable.
Bus Keepers on data lines to minimize current consumption (optional).
Overview
When the eXternal MEMory (XMEM) is enabled, address space outside the internal
SRAM becomes available using the dedicated external memory pins (see Figure 1 on
page 2, Table 27 on page 71, Table 33 on page 75, and Table 45 on page 83). The
memory configuration is shown in Figure 11.
Figure 11. External Memory with Sector Select(1)
Memory Configuration A
Memory Configuration B
0x0000
0x0000
Internal Memory
Internal Memory
0x0FFF
0x1000
0x10FF
0x1100
Lower Sector
SRW01
SRW00
SRW10
SRL[2..0]
External Memory
(0-60K x 8)
External Memory
(0-60K x 8)
Upper Sector
SRW11
SRW10
0xFFFF
0xFFFF
Note:
1. ATmega64 in non ATmega103 compatibility mode: Memory Configuration A is avail-
able (Memory Configuration B N/A).
ATmega64 in mega103 compatibility mode: Memory Configuration B is available
(Memory Configuration A N/A).
25
2490G–AVR–03/04
ATmega103 Compatibility
Both External Memory Control Registers, XMCRA and XMCRB, are placed in Extended
I/O space. In ATmega103 compatibility mode, these registers are not available, and the
features selected by these registers are not available. The device is still ATmega103
compatible, as these features did not exist in ATmega103. The limitations in
ATmega103 compatibility mode are:
•
•
•
Only two wait-state settings are available (SRW1n = 0b00 and SRW1n = 0b01).
The number of bits that are assigned to address high byte are fixed.
The external memory section cannot be divided into sectors with different wait-state
settings.
•
•
Bus Keeper is not available.
RD, WR, and ALE pins are output only (Port G in ATmega64).
Using the External Memory
Interface
The interface consists of:
•
•
•
•
•
AD7:0: Multiplexed low-order address bus and data bus.
A15:8: High-order address bus (configurable number of bits).
ALE: Address latch enable.
RD: Read strobe.
WR: Write strobe.
The control bits for the External Memory Interface are located in three registers, the
MCU Control Register – MCUCR, the External Memory Control Register A – XMCRA,
and the External Memory Control Register B – XMCRB.
When the XMEM interface is enabled, the XMEM interface will override the setting in the
Data Direction Registers that corresponds to the ports dedicated to the XMEM interface.
For details about the port override, see the alternate functions in section “I/O Ports” on
page 64. The XMEM interface will auto-detect whether an access is internal or external.
If the access is external, the XMEM interface will output address, data, and the control
signals on the ports according to Figure 13 (this figure shows the wave forms without
wait states). When ALE goes from high-to-low, there is a valid address on AD7:0. ALE is
low during a data transfer. When the XMEM interface is enabled, also an internal access
will cause activity on address-, data- and ALE ports, but the RD and WR strobes will not
toggle during internal access. When the external memory interface is disabled, the nor-
mal pin and data direction settings are used. Note that when the XMEM interface is
disabled, the address space above the internal SRAM boundary is not mapped into the
internal SRAM. Figure 12 illustrates how to connect an external SRAM to the AVR using
an octal latch (typically 74 x 573 or equivalent) which is transparent when G is high.
Address Latch Requirements Due to the high-speed operation of the XRAM interface, the address latch must be
selected with care for system frequencies above 8 MHz @ 4V and 4 MHz @ 2.7V.
When operating at conditions above these frequencies, the typical old style 74HC series
latch becomes inadequate. The external memory interface is designed in compliance to
the 74AHC series latch. However, most latches can be used as long they comply with
the main timing parameters. The main parameters for the address latch are:
•
•
•
D to Q propagation delay (tpd).
Data setup time before G low (tsu).
Data (address) hold time after G low (th).
The external memory interface is designed to guaranty minimum address hold time after
G is asserted low of th = 5 ns (refer to tLAXX_LD/tLLAXX_ST in Table 138 to Table 145 on
page 338). The D to Q propagation delay (tpd) must be taken into consideration when
calculating the access time requirement of the external component. The data setup time
26
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
before G low (tsu) must not exceed address valid to ALE low (tAVLLC) minus PCB wiring
delay (dependent on the capacitive load).
Figure 12. External SRAM Connected to the AVR
D[7:0]
AD7:0
ALE
D
G
Q
A[7:0]
SRAM
A[15:8]
AVR
A15:8
RD
RD
WR
WR
Pull-up and Bus Keeper
The pull-ups on the AD7:0 ports may be activated if the corresponding Port Register is
written to one. To reduce power consumption in sleep mode, it is recommended to dis-
able the pull-ups by writing the Port Register to zero before entering sleep.
The XMEM interface also provides a Bus Keeper on the AD7:0 lines. The Bus Keeper
can be disabled and enabled in software as described in “External Memory Control Reg-
ister B – XMCRB” on page 32. When enabled, the Bus Keeper will keep the previous
value on the AD7:0 bus while these lines are tri-stated by the XMEM interface.
Timing
External memory devices have different timing requirements. To meet these require-
ments, the ATmega64 XMEM interface provides four different wait states as shown in
Table 4. It is important to consider the timing specification of the external memory
device before selecting the wait-state. The most important parameters are the access
time for the external memory compared to the set-up requirement of the ATmega64.
The access time for the external memory is defined to be the time from receiving the
chip select/address until the data of this address actually is driven on the bus. The
access time cannot exceed the time from the ALE pulse is asserted low until data must
be stable during a read sequence (tLLRL+ tRLRH - tDVRH in Table 138 to Table 145 on page
338). The different wait states are set up in software. As an additional feature, it is possi-
ble to divide the external memory space in two sectors with individual wait-state
settings. This makes it possible to connect two different memory devices with different
timing requirements to the same XMEM interface. For XMEM interface timing details,
please refer to Figure 159 to Figure 162, and Table 138 to Table 145.
Note that the XMEM interface is asynchronous and that the waveforms in the following
figures are related to the internal system clock. The skew between the internal and
external clock (XTAL1) is not guaranteed (varies between devices, temperature, and
supply voltage). Consequently the XMEM interface is not suited for synchronous
operation.
27
2490G–AVR–03/04
Figure 13. External Data Memory Cycles without Wait State(1)
(SRWn1 = 0 and SRWn0 =0)
T1
T2
T3
T4
System Clock (CLKCPU
)
ALE
A15:8 Prev. Addr.
DA7:0 Prev. Data
WR
Address
Data
Address
Address
XX
DA7:0 (XMBK = 0) Prev. Data
DA7:0 (XMBK = 1) Prev. Data
RD
Data
Data
Address
Note:
1. SRWn1 = SRW11 (upper sector) or SRW01 (lower sector), SRWn0 = SRW10 (upper
sector) or SRW00 (lower sector).
The ALE pulse in period T4 is only present if the next instruction accesses the RAM
(internal or external).
Figure 14. External Data Memory Cycles with SRWn1 = 0 and SRWn0 = 1(1)
T1
T2
T3
T4
T5
System Clock (CLKCPU
)
ALE
A15:8 Prev. Addr.
DA7:0 Prev. Data
WR
Address
Data
Address
Address
XX
DA7:0 (XMBK = 0) Prev. Data
DA7:0 (XMBK = 1) Prev. Data
RD
Data
Data
Address
Note:
1. SRWn1 = SRW11 (upper sector) or SRW01 (lower sector), SRWn0 = SRW10 (upper
sector) or SRW00 (lower sector).
The ALE pulse in period T5 is only present if the next instruction accesses the RAM
(internal or external).
28
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 15. External Data Memory Cycles with SRWn1 = 1 and SRWn0 = 0(1)
T1
T2
T3
T4
T5
T6
System Clock (CLKCPU
)
ALE
A15:8 Prev. Addr.
DA7:0 Prev. Data
WR
Address
Data
Address
Address
XX
DA7:0 (XMBK = 0) Prev. Data
DA7:0 (XMBK = 1) Prev. Data
RD
Data
Data
Address
Note:
1. SRWn1 = SRW11 (upper sector) or SRW01 (lower sector), SRWn0 = SRW10 (upper
sector) or SRW00 (lower sector).
The ALE pulse in period T6 is only present if the next instruction accesses the RAM
(internal or external).
Figure 16. External Data Memory Cycles with SRWn1 = 1 and SRWn0 = 1(1)
T1
T2
T3
T4
T5
T6
T7
System Clock (CLKCPU
)
ALE
A15:8 Prev. Addr.
DA7:0 Prev. Data
WR
Address
Address XX
Data
DA7:0 (XMBK = 0) Prev. Data
DA7:0 (XMBK = 1) Prev. Data
RD
Address
Address
Data
Data
Note:
1. SRWn1 = SRW11 (upper sector) or SRW01 (lower sector), SRWn0 = SRW10 (upper
sector) or SRW00 (lower sector).
The ALE pulse in period T7 is only present if the next instruction accesses the RAM
(internal or external).
29
2490G–AVR–03/04
XMEM Register
Description
MCU Control Register –
MCUCR
Bit
7
6
SRW10
R/W
0
5
SE
R/W
0
4
3
2
1
IVSEL
R/W
0
0
IVCE
R/W
0
SRE
R/W
0
SM1
R/W
0
SM0
R/W
0
SM2
R/W
0
MCUCR
Read/Write
Initial Value
• Bit 7 – SRE: External SRAM/XMEM Enable
Writing SRE to one enables the External Memory Interface.The pin functions AD7:0,
A15:8, ALE, WR, and RD are activated as the alternate pin functions. The SRE bit over-
rides any pin direction settings in the respective data direction registers. Writing SRE to
zero, disables the External Memory Interface and the normal pin and data direction set-
tings are used.
• Bit 6 – SRW10: Wait State Select Bit
For a detailed description in non ATmega103 compatibility mode, see common descrip-
tion for the SRWn bits below (XMRA description). In ATmega103 compatibility mode,
writing SRW10 to one enables the wait state and one extra cycle is added during
read/write strobe as shown in Figure 14.
External Memory Control
Register A – XMCRA
Bit
7
–
R
0
6
SRL2
R/W
0
5
SRL1
R/W
0
4
SRL0
R/W
0
3
SRW01
R/W
0
2
SRW00
R/W
0
1
SRW11
R/W
0
0
–
R
0
XMCRA
Read/Write
Initial Value
• Bit 7 – Res: Reserved Bit
This is a reserved bit and will always read as zero. When writing to this address location,
write this bit to zero for compatibility with future devices.
• Bit 6..4 – SRL2, SRL1, SRL0: Wait State Sector Limit
It is possible to configure different wait states for different external memory addresses.
The external memory address space can be divided in two sectors that have separate
wait-state bits. The SRL2, SRL1, and SRL0 bits select the split of the sectors, see Table
3 and Figure 11. By default, the SRL2, SRL1, and SRL0 bits are set to zero and the
entire external memory address space is treated as one sector. When the entire SRAM
address space is configured as one sector, the wait states are configured by the SRW11
and SRW10 bits.
30
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 3. Sector Limits with Different Settings of SRL2..0
SRL2
SRL1
SRL0
Sector Limits
0
0
0
Lower sector = N/A
Upper sector = 0x1100 - 0xFFFF
0
0
0
1
1
1
1
0
1
1
0
0
1
1
1
0
1
0
1
0
1
Lower sector = 0x1100 - 0x1FFF
Upper sector = 0x2000 - 0xFFFF
Lower sector = 0x1100 - 0x3FFF
Upper sector = 0x4000 - 0xFFFF
Lower sector = 0x1100 - 0x5FFF
Upper sector = 0x6000 - 0xFFFF
Lower sector = 0x1100 - 0x7FFF
Upper sector = 0x8000 - 0xFFFF
Lower sector = 0x1100 - 0x9FFF
Upper sector = 0xA000 - 0xFFFF
Lower sector = 0x1100 - 0xBFFF
Upper sector = 0xC000 - 0xFFFF
Lower sector = 0x1100 - 0xDFFF
Upper sector = 0xE000 - 0xFFFF
• Bit 1 and Bit 6 MCUCR – SRW11, SRW10: Wait State Select Bits for Upper
Sector
The SRW11 and SRW10 bits control the number of wait states for the upper sector of
the external memory address space, see Table 4.
• Bit 3..2 – SRW01, SRW00: Wait State Select Bits for Lower Sector
The SRW01 and SRW00 bits control the number of wait states for the lower sector of
the external memory address space, see Table 4.
Table 4. Wait States(1)
SRWn1 SRWn0 Wait States
0
0
1
0
1
0
No wait states
Wait one cycle during read/write strobe
Wait two cycles during read/write strobe
Wait two cycles during read/write and wait one cycle before driving out
new address
1
1
Note:
1. n = 0 or 1 (lower/upper sector).
For further details of the timing and wait states of the External Memory Interface, see
Figure 13 to Figure 16 how the setting of the SRW bits affects the timing.
• Bit 0 – Res: Reserved Bit
This is a reserved bit and will always read as zero. When writing to this address location,
write this bit to zero for compatibility with future devices.
31
2490G–AVR–03/04
External Memory Control
Register B – XMCRB
Bit
7
XMBK
R/W
0
6
–
R
0
5
–
R
0
4
–
R
0
3
–
R
0
2
XMM2
R/W
0
1
XMM1
R/W
0
0
XMM0
R/W
0
XMCRB
Read/Write
Initial Value
• Bit 7 – XMBK: External Memory Bus Keeper Enable
Writing XMBK to one enables the Bus Keeper on the AD7:0 lines. When the Bus Keeper
is enabled, AD7:0 will keep the last driven value on the lines even if the XMEM interface
has tri-stated the lines. Writing XMBK to zero disables the Bus Keeper. XMBK is not
qualified with SRE, so even if the XMEM interface is disabled, the Bus Keepers are still
activated as long as XMBK is one.
• Bit 6..3 – Res: Reserved Bits
These are reserved bits and will always read as zero. When writing to this address loca-
tion, write these bits to zero for compatibility with future devices.
• Bit 2..0 – XMM2, XMM1, XMM0: External Memory High Mask
When the External Memory is enabled, all Port C pins are default used for the high
address byte. If the full 60KB address space is not required to access the external mem-
ory, some, or all, Port C pins can be released for normal port pin function as described in
Table 5. As described in “Using all 64KB Locations of External Memory” on page 34, it is
possible to use the XMMn bits to access all 64KB locations of the external memory.
Table 5. Port C Pins Released as Normal Port Pins when the External Memory is
Enabled
XMM2 XMM1 XMM0 # Bits for External Memory Address
Released Port Pins
None
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
8 (Full 60 KB space)
7
PC7
6
PC7 - PC6
PC7 - PC5
PC7 - PC4
PC7 - PC3
PC7 - PC2
Full Port C
5
4
3
2
No Address high bits
32
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Using all Locations of
Since the external memory is mapped after the internal memory as shown in Figure 11,
External Memory Smaller than the external memory is not addressed when addressing the first 4,352 bytes of data
64 KB
space. It may appear that the first 4,352 bytes of the external memory are inaccessible
(external memory addresses 0x0000 to 0x10FF). However, when connecting an exter-
nal memory smaller than 64 KB, for example 32 KB, these locations are easily accessed
simply by addressing from address 0x8000 to 0x90FF. Since the External Memory
Address bit A15 is not connected to the external memory, addresses 0x8000 to 0x90FF
will appear as addresses 0x0000 to 0x10FF for the external memory. Addressing above
address 0x90FF is not recommended, since this will address an external memory loca-
tion that is already accessed by another (lower) address. To the Application software,
the external 32 KB memory will appear as one linear 32 KB address space from 0x1100
to 0x90FF. This is illustrated in Figure 17. Memory configuration B refers to the
ATmega103 compatibility mode, configuration A to the non-compatible mode.
When the device is set in ATmega103 compatibility mode, the internal address space is
4,096 bytes. This implies that the first 4,096 bytes of the external memory can be
accessed at addresses 0x8000 to 0x8FFF. To the Application software, the external 32
KB memory will appear as one linear 32 KB address space from 0x1000 to 0x8FFF.
Figure 17. Address Map with 32 KB External Memory
Memory Configuration B
Memory Configuration A
AVR Memory Map
External 32K SRAM
AVR Memory Map
External 32K SRAM
0x0000
0x0000
0x0000
0x0000
Internal Memory
Internal Memory
0x0FFF
0x1000
0x0FFF
0x1000
0x10FF
0x1100
0x10FF
0x1100
External
Memory
External
Memory
0x7FFF
0x8000
0x7FFF
0x8000
0x7FFF
0x7FFF
0x90FF
0x9100
0x8FFF
0x9000
(Unused)
(Unused)
0xFFFF
0xFFFF
33
2490G–AVR–03/04
Using all 64KB Locations of
External Memory
Since the external memory is mapped after the internal memory as shown in Figure 11,
only 60KB of external memory is available by default (address space 0x0000 to 0x10FF
is reserved for internal memory). However, it is possible to take advantage of the entire
external memory by masking the higher address bits to zero. This can be done by using
the XMMn bits and controlled by software the most significant bits of the address. By
setting Port C to output 0x00, and releasing the most significant bits for normal Port Pin
operation, the Memory Interface will address 0x0000 - 0x1FFF. See code examples
below.
Assembly Code Example(1)
; OFFSET is defined to 0x2000 to ensure
; external memory access
; Configure Port C (address high byte) to
; output 0x00 when the pins are released
; for normal Port Pin operation
ldi
out
ldi
out
r16, 0xFF
DDRC, r16
r16, 0x00
PORTC, r16
; release PC7:5
ldi
sts
r16, (1<<XMM1)|(1<<XMM0)
XMCRB, r16
; write 0xAA to address 0x0001 of external
; memory
ldi
sts
r16, 0xaa
0x0001+OFFSET, r16
; re-enable PC7:5 for external memory
ldi
sts
r16, (0<<XMM1)|(0<<XMM0)
XMCRB, r16
; store 0x55 to address (OFFSET + 1) of
; external memory
ldi
sts
r16, 0x55
0x0001+OFFSET, r16
C Code Example(1)
#define OFFSET 0x2000
void XRAM_example(void)
{
unsigned char *p = (unsigned char *) (OFFSET + 1);
DDRC = 0xFF;
PORTC = 0x00;
XMCRB = (1<<XMM1) | (1<<XMM0);
*p = 0xaa;
XMCRB = 0x00;
*p = 0x55;
}
Note:
1. The example code assumes that the part specific header file is included.
Care must be exercised using this option as most of the memory is masked away.
34
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
System Clock and
Clock Options
Clock Systems and their Figure 18 presents the principal clock systems in the AVR and their distribution. All of
the clocks need not be active at a given time. In order to reduce power consumption, the
Distribution
clocks to modules not being used can be halted by using different sleep modes, as
described in “Power Management and Sleep Modes” on page 44. The clock systems
are detailed below.
Figure 18. Clock Distribution
Asynchronous
Timer/Counter
General I/O
Modules
Flash and
EEPROM
ADC
CPU Core
RAM
clkADC
clkI/O
clkCPU
AVR Clock
Control Unit
clkASY
clkFLASH
Reset Logic
Watchdog Timer
Source Clock
Watchdog Clock
Clock
Multiplexer
Watchdog
Oscillator
Timer/Counter
Oscillator
External RC
Oscillator
Crystal
Oscillator
Low-frequency
Crystal Oscillator
Calibrated RC
Oscillator
External Clock
CPU Clock – clkCPU
I/O Clock – clkI/O
The CPU clock is routed to parts of the system concerned with operation of the AVR
core. Examples of such modules are the General Purpose Register File, the Status Reg-
ister and the data memory holding the Stack Pointer. Halting the CPU clock inhibits the
core from performing general operations and calculations.
The I/O clock is used by the majority of the I/O modules, like Timer/Counters, SPI, and
USART. The I/O clock is also used by the External Interrupt module, but note that some
external interrupts are detected by asynchronous logic, allowing such interrupts to be
detected even if the I/O clock is halted. Also note that address recognition in the TWI
module is carried out asynchronously when clkI/O is halted, enabling TWI address recep-
tion in all sleep modes.
Flash Clock – clkFLASH
The Flash clock controls operation of the Flash interface. The Flash clock is usually
active simultaneously with the CPU clock.
35
2490G–AVR–03/04
Asynchronous Timer Clock –
clkASY
The Asynchronous Timer clock allows the Asynchronous Timer/Counter to be clocked
directly from an external 32 kHz clock crystal. The dedicated clock domain allows using
this Timer/Counter as a real-time counter even when the device is in sleep mode.
ADC Clock – clkADC
The ADC is provided with a dedicated clock domain. This allows halting the CPU and
I/O clocks in order to reduce noise generated by digital circuitry. This gives more accu-
rate ADC conversion results.
Clock Sources
The device has the following clock source options, selectable by Flash Fuse bits as
shown below. The clock from the selected source is input to the AVR clock generator,
and routed to the appropriate modules.
Table 6. Device Clocking Options Select(1)
Device Clocking Option
External Crystal/Ceramic Resonator
External Low-frequency Crystal
External RC Oscillator
CKSEL3..0
1111 - 1010
1001
1000 - 0101
0100 - 0001
0000
Calibrated Internal RC Oscillator
External Clock
Note:
1. For all fuses “1” means unprogrammed while “0” means programmed.
The various choices for each clocking option is given in the following sections. When the
CPU wakes up from Power-down or Power-save, the selected clock source is used to
time the start-up, ensuring stable Oscillator operation before instruction execution starts.
When the CPU starts from reset, there is as an additional delay allowing the power to
reach a stable level before commencing normal operation. The Watchdog Oscillator is
used for timing this real-time part of the start-up time. The number of WDT Oscillator
cycles used for each time-out is shown in Table 7. The frequency of the Watchdog Oscil-
lator is voltage dependent as shown in the “ATmega64 Typical Characteristics –
Preliminary Data” on page 341.
Table 7. Number of Watchdog Oscillator Cycles
Typ Time-out (VCC = 5.0V)
Typ Time-out (VCC = 3.0V)
Number of Cycles
4K (4,096)
4.1 ms
65 ms
4.3 ms
69 ms
64K (65,536)
Default Clock Source
The device is shipped with CKSEL = “0001” and SUT = “10”. The default clock source
setting is therefore the Internal RC Oscillator with longest startup time. This default set-
ting ensures that all users can make their desired clock source setting using an In-
System or Parallel Programmer.
36
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Crystal Oscillator
XTAL1 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 19. Either a quartz
crystal or a ceramic resonator may be used. The CKOPT Fuse selects between two dif-
ferent Oscillator amplifier modes. When CKOPT is programmed, the Oscillator output
will oscillate a full rail-to-rail swing on the output. This mode is suitable when operating
in a very noisy environment or when the output from XTAL2 drives a second clock
buffer. This mode has a wide frequency range. When CKOPT is unprogrammed, the
Oscillator has a smaller output swing. This reduces power consumption considerably.
This mode has a limited frequency range and it cannot be used to drive other clock
buffers.
For resonators, the maximum frequency is 8 MHz with CKOPT unprogrammed and
16 MHz with CKOPT programmed. C1 and C2 should always be equal for both crystals
and resonators. The optimal value of the capacitors depends on the crystal or resonator
in use, the amount of stray capacitance, and the electromagnetic noise of the environ-
ment. Some initial guidelines for choosing capacitors for use with crystals are given in
Table 8. For ceramic resonators, the capacitor values given by the manufacturer should
be used.
Figure 19. Crystal Oscillator Connections
C2
XTAL2
C1
XTAL1
GND
The Oscillator can operate in three different modes, each optimized for a specific fre-
quency range. The operating mode is selected by the fuses CKSEL3..1 as shown in
Table 8.
Table 8. Crystal Oscillator Operating Modes
Frequency Range(1)
(MHz)
Recommended Range for Capacitors
C1 and C2 for Use with Crystals (pF)
CKOPT
CKSEL3..1
101(2)
1
1
1
0
0.4 - 0.9
0.9 - 3.0
3.0 - 8.0
1.0 -
–
110
12 - 22
12 - 22
12 - 22
111
101, 110, 111
Notes: 1. The frequency ranges are preliminary values. Actual values are TBD.
2. This option should not be used with crystals, only with ceramic resonators.
The CKSEL0 Fuse together with the SUT1..0 Fuses select the start-up times as shown
in Table 9.
37
2490G–AVR–03/04
Table 9. Start-up Times for the Crystal Oscillator Clock Selection
Start-up Time from
Power-down and
Power-save
Additional Delay
from Reset
CKSEL0 SUT1..0
(VCC = 5.0V)
Recommended Usage
Ceramic resonator, fast
rising power
0
0
0
0
1
1
1
1
00
01
10
11
00
01
10
11
258 CK(1)
258 CK(1)
1K CK(2)
1K CK(2)
1K CK(2)
16K CK
16K CK
16K CK
4.1 ms
65 ms
–
Ceramic resonator,
slowly rising power
Ceramic resonator, BOD
enabled
Ceramic resonator, fast
rising power
4.1 ms
65 ms
–
Ceramic resonator,
slowly rising power
Crystal Oscillator, BOD
enabled
Crystal Oscillator, fast
rising power
4.1 ms
65 ms
Crystal Oscillator, slowly
rising power
Notes: 1. These options should only be used when not operating close to the maximum fre-
quency of the device, and only if frequency stability at start-up is not important for the
application. These options are not suitable for crystals.
2. These options are intended for use with ceramic resonators and will ensure fre-
quency stability at start-up. They can also be used with crystals when not operating
close to the maximum frequency of the device, and if frequency stability at start-up is
not important for the application.
Low-frequency Crystal
Oscillator
To use a 32.768 kHz watch crystal as the clock source for the device, the Low-fre-
quency crystal Oscillator must be selected by setting the CKSEL Fuses to “1001”. The
crystal should be connected as shown in Figure 19. By programming the CKOPT Fuse,
the user can enable internal capacitors on XTAL1 and XTAL2, thereby removing the
need for external capacitors. The internal capacitors have a nominal value of 36 pF.
When this Oscillator is selected, start-up times are determined by the SUT Fuses as
shown in Table 10.
Table 10. Start-up Times for the Low-frequency Crystal Oscillator Clock Selection
Start-up Time
from Power-down
and Power-save
Additional Delay
from Reset
SUT1..0
00
(VCC = 5.0V)
Recommended Usage
1K CK(1)
1K CK(1)
32K CK
4.1 ms
65 ms
65 ms
Fast rising power or BOD enabled
Slowly rising power
01
10
Stable frequency at start-up
11
Reserved
Note:
1. These options should only be used if frequency stability at start-up is not important
for the application.
38
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
External RC Oscillator
For timing insensitive applications, the external RC configuration shown in Figure 20
can be used. The frequency is roughly estimated by the equation f = 1/(3RC). C should
be at least 22 pF. By programming the CKOPT Fuse, the user can enable an internal 36
pF capacitor between XTAL1 and GND, thereby removing the need for an external
capacitor. For more information on Oscillator operation and details on how to choose R
and C, refer to the External RC Oscillator application note.
Figure 20. External RC Configuration
V
CC
NC
XTAL2
XTAL1
GND
R
C
The Oscillator can operate in four different modes, each optimized for a specific fre-
quency range. The operating mode is selected by the fuses CKSEL3..0 as shown in
Table 11.
Table 11. External RC Oscillator Operating Modes
CKSEL3..0
0101
Frequency Range (MHz)
- 0.9
0110
0.9 - 3.0
3.0 - 8.0
8.0 - 12.0
0111
1000
When this Oscillator is selected, start-up times are determined by the SUT Fuses as
shown in Table 12.
Table 12. Start-up Times for the External RC Oscillator Clock Selection
Start-up Time from
Power-down and
Power-save
Additional Delay
from Reset
SUT1..0
00
(VCC = 5.0V)
Recommended Usage
BOD enabled
18 CK
18 CK
18 CK
6 CK(1)
–
01
4.1 ms
65 ms
4.1 ms
Fast rising power
10
Slowly rising power
11
Fast rising power or BOD enabled
Note:
1. This option should not be used when operating close to the maximum frequency of
the device.
39
2490G–AVR–03/04
Calibrated Internal RC
Oscillator
The calibrated internal RC Oscillator provides a fixed 1.0, 2.0, 4.0, or 8.0 MHz clock. All
frequencies are nominal values at 5V and 25°C. This clock may be selected as the sys-
tem clock by programming the CKSEL Fuses as shown in Table 13. If selected, it will
operate with no external components. The CKOPT Fuse should always be unpro-
grammed when using this clock option. During reset, hardware loads the calibration byte
into the OSCCAL Register and thereby automatically calibrates the RC Oscillator. At 5V,
25°C and 1.0 MHz Oscillator frequency selected, this calibration gives a frequency
within 3% of the nominal frequency. Using run-time calibration methods as described
in application notes available at www.atmel.com/avr it is possible to achieve 1% accu-
racy at any given VCC and Temperature. When this Oscillator is used as the chip clock,
the Watchdog Oscillator will still be used for the Watchdog Timer and for the Reset
Time-out. For more information on the preprogrammed calibration value, see the section
“Calibration Byte” on page 293.
Table 13. Internal Calibrated RC Oscillator Operating Modes
CKSEL3..0
0001(1)
0010
Nominal Frequency (MHz)
1.0
2.0
4.0
8.0
0011
0100
Note:
1. The device is shipped with this option selected.
When this Oscillator is selected, start-up times are determined by the SUT Fuses as
shown in Table 14. XTAL1 and XTAL2 should be left unconnected (NC).
Table 14. Start-up Times for the Internal Calibrated RC Oscillator Clock Selection
Start-up Time from Power-
down and Power-save
Additional Delay from
Reset (VCC = 5.0V)
SUT1..0
00
Recommended Usage
BOD enabled
6 CK
6 CK
6 CK
–
4.1 ms
01
Fast rising power
Slowly rising power
10(1)
65 ms
11
Reserved
Note:
1. The device is shipped with this option selected.
Oscillator Calibration Register
– OSCCAL(1)
Bit
7
6
5
4
3
2
1
0
CAL7
R/W
CAL6
R/W
CAL5
R/W
CAL4
R/W
CAL3
R/W
CAL2
R/W
CAL1
R/W
CAL0
R/W
OSCCAL
Read/Write
Initial Value
Device Specific Calibration Value
Note:
1. The OSCCAL Register is not available in ATmega103 compatibility mode.
• Bits 7..0 – CAL7..0: Oscillator Calibration Value
Writing the calibration byte to this address will trim the internal Oscillator to remove pro-
cess variations from the Oscillator frequency. During Reset, the 1 MHz calibration value
which is located in the signature row high byte (address 0x00) is automatically loaded
into the OSCCAL Register. If the internal RC is used at other frequencies, the calibration
values must be loaded manually. This can be done by first reading the signature row by
a programmer, and then store the calibration values in the Flash or EEPROM. Then the
40
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
value can be read by software and loaded into the OSCCAL Register. When OSCCAL is
zero, the lowest available frequency is chosen. Writing non-zero values to this register
will increase the frequency of the internal Oscillator. Writing 0xFF to the register gives
the highest available frequency. The calibrated Oscillator is used to time EEPROM and
Flash access. If EEPROM or Flash is written, do not calibrate to more than 10% above
the nominal frequency. Otherwise, the EEPROM or Flash write may fail. Note that the
Oscillator is intended for calibration to 1.0, 2.0, 4.0, or 8.0 MHz. Tuning to other values is
not guaranteed, as indicated in Table 15.
Table 15. Internal RC Oscillator Frequency Range
Min Frequency in Percentage of
Nominal Frequency (%)
Max Frequency in Percentage of
Nominal Frequency (%)
OSCCAL Value
0x00
50
75
100
150
200
0x7F
0xFF
100
External Clock
To drive the device from an external clock source, XTAL1 should be driven as shown in
Figure 21. To run the device on an external clock, the CKSEL Fuses must be pro-
grammed to “0000”. By programming the CKOPT Fuse, the user can enable an internal
36 pF capacitor between XTAL1 and GND.
Figure 21. External Clock Drive Configuration
EXTERNAL
CLOCK
SIGNAL
When this clock source is selected, start-up times are determined by the SUT Fuses as
shown in Table 16.
Table 16. Start-up Times for the External Clock Selection
Start-up Time from Power-
down and Power-save
Additional Delay from
Reset (VCC = 5.0V)
SUT1..0
00
Recommended Usage
BOD enabled
6 CK
6 CK
6 CK
–
4.1 ms
01
Fast rising power
Slowly rising power
10
65 ms
11
Reserved
When applying an external clock, it is required to avoid sudden changes in the applied
clock frequency to ensure stable operation of the MCU. A variation in frequency of more
than 2% from one clock cycle to the next can lead to unpredictable behavior. It is
41
2490G–AVR–03/04
required to ensure that the MCU is kept in Reset during such changes in the clock
frequency.
42
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Timer/Counter Oscillator For AVR microcontrollers with Timer/Counter 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 kHz watch crystal. Applying an external
clock source to TOSC1 is not recommended.
XTAL Divide Control Register The XTAL Divide Control Register is used to divide the source clock frequency by a
– XDIV
number in the range 2 - 129. This feature can be used to decrease power consumption
when the requirement for processing power is low.
Bit
7
XDIVEN
R/W
0
6
XDIV6
R/W
0
5
XDIV5
R/W
0
4
XDIV4
R/W
0
3
XDIV3
R/W
0
2
XDIV2
R/W
0
1
XDIV1
R/W
0
0
XDIV0
R/W
0
XDIV
Read/Write
Initial Value
• Bit 7 – XDIVEN: XTAL Divide Enable
When the XDIVEN bit is written one, the clock frequency of the CPU and all peripherals
(clkI/O, clkADC, clkCPU, clkFLASH) is divided by the factor defined by the setting of XDIV6 -
XDIV0. This bit can be written run-time to vary the clock frequency as suitable to the
application.
• Bits 6..0 – XDIV6..XDIV0: XTAL Divide Select Bits 6 - 0
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 and
peripherals clock frequency fclk:
Source clock
f
= ---------------------------------
CLK
129 – d
The value of these bits can only be changed when XDIVEN is zero. When XDIVEN is
written to one, the value written simultaneously into XDIV6..XDIV0 is taken as the divi-
sion factor. When XDIVEN is written 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.
Note:
When the system clock is divided, Timer/Counter0 can be used with Asynchronous clock
only. The frequency of the asynchronous clock must be lower than 1/4th of the frequency
of the scaled down Source clock. Otherwise, interrupts may be lost, and accessing the
Timer/Counter0 registers may fail.
43
2490G–AVR–03/04
Power Management
and Sleep Modes
Sleep modes enable the application to shut down unused modules in the MCU, thereby
saving power. The AVR provides various sleep modes allowing the user to tailor the
power consumption to the application’s requirements.
To enter any of the six sleep modes, the SE-bit in MCUCR must be written to logic one
and a SLEEP instruction must be executed. The SM2, SM1, and SM0 bits in the
MCUCR Register select which sleep mode (Idle, ADC Noise Reduction, Power-down,
Power-save, Standby, or Extended Standby) will be activated by the SLEEP instruction.
See Table 17 for a summary. If an enabled interrupt occurs while the MCU is in a sleep
mode, the MCU wakes up. The MCU is then halted for four cycles in addition to the
start-up time, it executes the interrupt routine, and resumes execution from the instruc-
tion following SLEEP. The contents of the Register File and SRAM are unaltered when
the device wakes up from sleep. If a reset occurs during sleep mode, the MCU wakes
up and executes from the Reset Vector.
Figure 18 on page 35 presents the different clock systems in the ATmega64, and their
distribution. This figure is helpful in selecting an appropriate sleep mode.
MCU Control Register –
MCUCR
The MCU Control Register contains control bits for power management.
Bit
7
6
SRW10
R/W
0
5
SE
R/W
0
4
3
2
1
IVSEL
R/W
0
0
IVCE
R/W
0
SRE
R/W
0
SM1
R/W
0
SM0
R/W
0
SM2
R/W
0
MCUCR
Read/Write
Initial Value
• Bit 5 – SE: Sleep Enable
The SE bit must be written to logic 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 programmers purpose, it is recommended to write the Sleep Enable (SE) bit to one
just before the execution of the SLEEP instruction and to clear it immediately after wak-
ing up.
• Bits 4..2 – SM2..0: Sleep Mode Select Bits 2, 1, and 0
These bits select between the six available sleep modes as shown in Table 17.
Table 17. Sleep Mode Select
SM2
SM1
SM0
Sleep Mode
Idle
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
ADC Noise Reduction
Power-down
Power-save
Reserved
Reserved
Standby(1)
Extended Standby(1)
Note:
1. Standby mode and Extended Standby mode are only available with external crystals
or resonators.
44
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Idle Mode
When the SM2..0 bits are written to 000, the SLEEP instruction makes the MCU enter
Idle mode, stopping the CPU but allowing SPI, USART, Analog Comparator, ADC, Two-
wire Serial Interface, Timer/Counters, Watchdog, and the interrupt system to continue
operating. This sleep mode basically halts clkCPU and clkFLASH, while allowing the other
clocks to run.
Idle mode enables the MCU to wake up from external triggered interrupts as well as
internal ones like the Timer Overflow and USART Transmit 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 Sta-
tus Register – ACSR. This will reduce power consumption in Idle mode. If the ADC is
enabled, a conversion starts automatically when this mode is entered.
ADC Noise Reduction
Mode
When the SM2..0 bits are written to 001, the SLEEP instruction makes the MCU enter
ADC Noise Reduction mode, stopping the CPU but allowing the ADC, the external inter-
rupts, the Two-wire Serial Interface address watch, Timer/Counter0 and the Watchdog
to continue operating (if enabled). This sleep mode basically halts clkI/O, clkCPU, and
clk-FLASH, while allowing the other clocks to run.
This improves the noise environment for the ADC, enabling higher resolution measure-
ments. If the ADC is enabled, a conversion starts automatically when this mode is
entered. Apart form the ADC Conversion Complete interrupt, only an External Reset, a
Watchdog Reset, a Brown-out Reset, a Two-wire Serial Interface address match inter-
rupt, a Timer/Counter0 interrupt, an SPM/EEPROM ready interrupt, an external level
interrupt on INT7:4, or an External Interrupt on INT3:0 can wake up the MCU from ADC
Noise Reduction mode.
Power-down Mode
When the SM2..0 bits are written to 010, the SLEEP instruction makes the MCU enter
Power-down mode. In this mode, the external Oscillator is stopped, while the external
interrupts, the Two-wire Serial Interface address watch, and the Watchdog continue
operating (if enabled). Only an External Reset, a Watchdog Reset, a Brown-out Reset, a
Two-wire Serial Interface address match interrupt, an external level interrupt on INT7:4,
or an External Interrupt on INT3:0 can wake up the MCU. This sleep mode basically
halts all generated clocks, allowing operation of asynchronous modules only.
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. Refer to “External Inter-
rupts” on page 88 for details.
When waking up from Power-down mode, there is 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 CKSEL
Fuses that define the Reset Time-out period, as described in “Clock Sources” on page
36.
Power-save Mode
When the SM2..0 bits are written to 011, the SLEEP instruction makes the MCU enter
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. The device can wake up from either Timer Over-
flow or Output Compare event from Timer/Counter0 if the corresponding
Timer/Counter0 interrupt enable bits are set in TIMSK, and the Global Interrupt Enable
bit in SREG is set.
If the asynchronous timer is NOT clocked asynchronously, Power-down mode is recom-
mended instead of Power-save mode because the contents of the registers in the
45
2490G–AVR–03/04
asynchronous timer should be considered undefined after wake-up in Power-save mode
if AS0 is 0.
This sleep mode basically halts all clocks except clkASY, allowing operation only of asyn-
chronous modules, including Timer/Counter0 if clocked asynchronously.
Standby Mode
When the SM2..0 bits are 110 and an external crystal/resonator clock option is selected,
the SLEEP instruction makes the MCU enter Standby mode. This mode is identical to
Power-down with the exception that the Oscillator is kept running. From Standby mode,
the device wakes up in six clock cycles.
Extended Standby Mode When the SM2..0 bits are 111 and an external crystal/resonator clock option is selected,
the SLEEP instruction makes the MCU enter Extended Standby mode. This mode is
identical to Power-save mode with the exception that the Oscillator is kept running.
From Extended Standby mode, the device wakes up in six clock cycles.
Table 18. Active Clock Domains and Wake Up Sources in the Different Sleep Modes
Active Clock Domains
Oscillators
Wake Up Sources
Main
I
Clock
Source
Timer
Osc
N
T
TWI
Address
Match
SPM/
EEPROM
Ready
A
D
C
Sleep
Mode
Other
I/O
clkCPU clkFLASH clkIO clkADC clkASY Enabled Enabled 7:0
Timer0
X
Idle
X
X
X
X
X
X(2)
X(2)
X
X
X
X
X
ADC
Noise
Reduction
X
X
X(3)
X
X
X
X
Power-
down
X(3)
X
Power-
save
X(2)
X(2)
X(3)
X(3)
X(3)
X
X
X
X(2)
Standby(1)
X
X
Extended
Standby(1)
X(2)
X(2)
X(2)
Notes: 1. External Crystal or resonator selected as clock source.
2. If AS0 bit in ASSR is set.
3. Only INT3:0 or level interrupt INT7:4.
46
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Minimizing Power
Consumption
There are several issues to consider when trying to minimize the power consumption in
an AVR controlled system. In general, sleep modes should be used as much as possi-
ble, and the sleep mode should be selected so that as few as possible of the device’s
functions are operating. All functions not needed should be disabled. In particular, the
following modules may need special consideration when trying to achieve the lowest
possible power consumption.
Analog to Digital Converter
Analog Comparator
If enabled, the ADC will be enabled in all sleep modes. To save power, the ADC should
be disabled before entering any sleep mode. When the ADC is turned off and on again,
the next conversion will be an extended conversion. Refer to “Analog to Digital Con-
verter” on page 230 for details on ADC operation.
When entering Idle mode, the Analog Comparator should be disabled if not used. When
entering ADC Noise Reduction mode, the Analog Comparator should be disabled. In the
other sleep modes, the Analog Comparator is automatically disabled. However, if the
Analog Comparator is set up to use the internal voltage reference as input, the Analog
Comparator should be disabled in all sleep modes. Otherwise, the internal voltage refer-
ence will be enabled, independent of sleep mode. Refer to “Analog Comparator” on
page 227 for details on how to configure the Analog Comparator.
Brown-out Detector
If the Brown-out Detector is not needed in the application, this module should be turned
off. If the Brown-out Detector is enabled by the BODEN Fuse, it will be enabled in all
sleep modes, and hence, always consume power. In the deeper sleep modes, this will
contribute significantly to the total current consumption. Refer to “Brown-out Detector”
on page 47 for details on how to configure the Brown-out Detector.
Internal Voltage Reference
The internal voltage reference will be enabled when needed by the Brown-out Detector,
the Analog Comparator or the ADC. If these modules are disabled as described in the
sections above, the internal voltage reference will be disabled and it will not be consum-
ing power. When turned on again, the user must allow the reference to start up before
the output is used. If the reference is kept on in sleep mode, the output can be used
immediately. Refer to “Internal Voltage Reference” on page 54 for details on the start-up
time.
Watchdog Timer
Port Pins
If the Watchdog Timer is not needed in the application, this module should be turned off.
If the Watchdog Timer is enabled, it will be enabled in all sleep modes, and hence,
always consume power. In the deeper sleep modes, this will contribute significantly to
the total current consumption. Refer to “Watchdog Timer” on page 54 for details on how
to configure the Watchdog Timer.
When entering a sleep mode, all port pins should be configured to use minimum power.
The most important thing is then to ensure that no pins drive resistive loads. In sleep
modes where the both the I/O clock (clkI/O) and the ADC clock (clkADC) are stopped, the
input buffers of the device will be disabled. This ensures that no power is consumed by
the input logic when not needed. In some cases, the input logic is needed for detecting
wake-up conditions, and it will then be enabled. Refer to the section “Digital Input
Enable and Sleep Modes” on page 68 for details on which pins are enabled. If the input
buffer is enabled and the input signal is left floating or have an analog signal level close
to VCC/2, the input buffer will use excessive power.
47
2490G–AVR–03/04
JTAG Interface and
On-chip Debug System
If the On-chip debug system is enabled by the OCDEN Fuse and the chip enter Power
down or Power save sleep mode, the main clock source remains enabled. In these
sleep modes, this will contribute significantly to the total current consumption. There are
three alternative ways to avoid this:
•
•
•
Disable OCDEN Fuse.
Disable JTAGEN Fuse.
Write one to the JTD bit in MCUCSR.
The TDO pin is left floating when the JTAG interface is enabled while the JTAG TAP
controller is not shifting data. If the hardware connected to the TDO pin does not pull up
the logic level, power consumption will increase. Note that the TDI pin for the next
device in the scan chain contains a pull-up that avoids this problem. Writing the JTD bit
in the MCUCSR register to one or leaving the JTAG fuse unprogrammed disables the
JTAG interface.
48
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
System Control and
Reset
Resetting the AVR
During Reset, all I/O Registers are set to their initial values, and the program starts exe-
cution from the Reset Vector. The instruction placed at the Reset Vector 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. This is also the case if the Reset Vector is in the
Application section while the Interrupt Vectors are in the Boot section or vice versa. The
circuit diagram in Figure 22 shows the Reset logic. Table 19 defines the electrical
parameters of the Reset circuitry.
The I/O ports of the AVR are immediately Reset to their initial state when a reset source
goes active. This does not require any clock source to be running.
After all reset sources have gone inactive, a delay counter is invoked, stretching the
Internal Reset. This allows the power to reach a stable level before normal operation
starts. The Time-out period of the delay counter is defined by the user through the
CKSEL Fuses. The different selections for the delay period are presented in “Clock
Sources” on page 36.
Reset Sources
The ATmega64 has five sources of reset:
•
•
•
•
•
Power-on Reset. The MCU is reset when the supply voltage is below the Power-on
Reset threshold (VPOT).
External Reset. The MCU is reset when a low level is present on the RESET pin for
longer than the minimum pulse length.
Watchdog Reset. The MCU is reset when the Watchdog Timer period expires and
the Watchdog is enabled.
Brown-out Reset. The MCU is reset when the supply voltage VCC is below the
Brown-out Reset threshold (VBOT) and the Brown-out Detector is enabled.
JTAG AVR Reset. The MCU is reset as long as there is a logic one in the Reset
Register, one of the scan chains of the JTAG system. Refer to the section “IEEE
1149.1 (JTAG) Boundary-scan” on page 254 for details.
49
2490G–AVR–03/04
Figure 22. Reset Logic
DATA BUS
D
Q
Q
MCU Control and Status
Register (MCUCSR)
PEN
L
Pull-up Resistor
Power-On Reset
Circuit
Brown-Out
Reset Circuit
BODEN
BODLEVEL
Pull-up Resistor
SPIKE
FILTER
Reset Circuit
RESET
JTAG Reset
Register
Watchdog
Timer
Watchdog
Oscillator
Delay Counters
Clock
CK
TIMEOUT
Generator
CKSEL[3:0]
SUT[1:0]
Table 19. Reset Characteristics
Symbol Parameter
Condition
Min
Typ
Max
Units
Power-on Reset
Threshold Voltage
1.4
2.3
V
(rising)
VPOT
Power-on Reset
Threshold Voltage
1.3
2.3
V
(falling)(1)
RESET Pin Threshold
Voltage
VRST
0.2 VCC
0.85 VCC
V
Minimum pulse width on
RESET Pin
tRST
50
ns
Brown-out Reset
BODLEVEL = 1
BODLEVEL = 0
BODLEVEL = 1
2.5
3.7
2.7
4.0
2
3.2
4.5
Threshold Voltage(2)
VBOT
V
Minimum low voltage
period for Brown-out
Detection
µs
µs
tBOD
BODLEVEL = 0
2
Brown-out Detector
hysteresis
VHYST
120
mV
50
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Notes: 1. The Power-on Reset will not work unless the supply voltage has been below VPOT
(falling).
2. VBOT may be below nominal minimum operating voltage for some devices. For
devices where this is the case, the device is tested down to VCC = VBOT during the
production test. This guarantees that a Brown-out Reset will occur before VCC drops
to a voltage where correct operation of the microcontroller is no longer guaranteed.
The test is performed using BODLEVEL=1 for ATmega64L and BODLEVEL=0 for
ATmega64. BODLEVEL=1 is not applicable for ATmega64.
Power-on Reset
A Power-on Reset (POR) pulse is generated by an On-chip Detection circuit. The detec-
tion level is defined in Table 19. The POR is activated whenever VCC is below the
detection level. The POR circuit can be used to trigger the Start-up Reset, as well as to
detect a failure in supply voltage.
A Power-on Reset (POR) circuit ensures that the device is reset from Power-on. Reach-
ing the Power-on Reset threshold voltage invokes the delay counter, which determines
how long the device is kept in RESET after VCC rise. The RESET signal is activated
again, without any delay, when VCC decreases below the detection level.
Figure 23. MCU Start-up, RESET Tied to VCC
VPOT
VCC
VRST
RESET
tTOUT
TIME-OUT
INTERNAL
RESET
Figure 24. MCU Start-up, RESET Extended Externally
VPOT
VCC
VRST
RESET
tTOUT
TIME-OUT
INTERNAL
RESET
51
2490G–AVR–03/04
External Reset
An External Reset is generated by a low level on the RESET pin. Reset pulses longer
than the minimum pulse width (see Table 19) 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 – VRST on its positive edge, the delay
counter starts the MCU after the Time-out period tTOUT has expired.
Figure 25. External Reset during Operation
CC
Brown-out Detection
ATmega64 has an On-chip Brown-out Detection (BOD) circuit for monitoring the VCC
level during operation by comparing it to a fixed trigger level. The trigger level for the
BOD can be selected by the fuse BODLEVEL to be 2.7V (BODLEVEL unprogrammed),
or 4.0V (BODLEVEL programmed). The trigger level has a hysteresis to ensure spike
free Brown-out Detection. The hysteresis on the detection level should be interpreted as
V
BOT+ = VBOT + VHYST/2 and VBOT- = VBOT - VHYST/2.
The BOD circuit can be enabled/disabled by the fuse BODEN. When the BOD is
enabled (BODEN programmed), and VCC decreases to a value below the trigger level
(VBOT- in Figure 26), the Brown-out Reset is immediately activated. When VCC increases
above the trigger level (VBOT+ in Figure 26), the delay counter starts the MCU after the
Time-out period tTOUT has expired.
The BOD circuit will only detect a drop in VCC if the voltage stays below the trigger level
for longer than tBOD given in Table 19.
Figure 26. Borwn-out Reset During Operation
VBOT+
VCC
VBOT-
RESET
t
TOUT
TIME-OUT
INTERNAL
RESET
52
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Watchdog Reset
When the Watchdog times out, it will generate a short reset pulse of one CK cycle dura-
tion. On the falling edge of this pulse, the delay timer starts counting the Time-out period
tTOUT. Refer to page 54 for details on operation of the Watchdog Timer.
Figure 27. Watchdog Reset During Operation
CC
CK
MCU Control and Status
Register – MCUCSR(1)
The MCU Control and Status Register provides information on which reset source
caused an MCU Reset.
Bit
7
6
–
R
0
5
–
R
0
4
3
2
1
0
JTD
R/W
0
JTRF
R/W
WDRF
R/W
BORF
R/W
EXTRF
R/W
PORF
R/W
MCUCSR
Read/Write
Initial Value
See Bit Description
Note:
1. Only EXTRF and PORF are available in mega103 compatibility mode.
• Bit 4 – JTRF: JTAG Reset Flag
This bit is set if a reset is being caused by a logic one in the JTAG Reset Register
selected by the JTAG instruction AVR_RESET. This bit is reset by a Brown-out Reset,
or by writing a logic zero to the flag.
• Bit 3 – WDRF: Watchdog Reset Flag
This bit is set if a Watchdog Reset occurs. The bit is reset by a Power-on Reset, or by
writing a logic zero to the flag.
• Bit 2 – BORF: Brown-out Reset Flag
This bit is set if a Brown-out Reset occurs. The bit is reset by a Power-on Reset, or by
writing a logic zero to the flag.
• Bit 1 – EXTRF: External Reset Flag
This bit is set if an External Reset occurs. The bit is reset by a Power-on Reset, or by
writing a logic zero to the flag.
53
2490G–AVR–03/04
• Bit 0 – PORF: Power-on Reset Flag
This bit is set if a Power-on Reset occurs. The bit is reset only by writing a logic zero to
the flag.
To make use of the reset flags to identify a reset condition, the user should read and
then reset the MCUCSR as early as possible in the program. If the register is cleared
before another reset occurs, the source of the reset can be found by examining the
Reset Flags.
Internal Voltage
Reference
ATmega64 features an internal bandgap reference. This reference is used for Brown-
out Detection, and it can be used as an input to the Analog Comparator or the ADC. The
2.56V reference to the ADC is generated from the internal bandgap reference.
Voltage Reference Enable
Signals and Start-up Time
The voltage reference has a start-up time that may influence the way it should be used.
The start-up time is given in Table 20. To save power, the reference is not always turned
on. The reference is on during the following situations:
1. When the BOD is enabled (by programming the BODEN Fuse).
2. When the bandgap reference is connected to the Analog Comparator (by setting
the ACBG bit in ACSR).
3. When the ADC is enabled.
Thus, when the BOD is not enabled, after setting the ACBG bit or enabling the ADC, the
user must always allow the reference to start up before the output from the Analog Com-
parator or ADC is used. To reduce power consumption in Power-down mode, the user
can avoid the three conditions above to ensure that the reference is turned off before
entering Power-down mode.
Table 20. Internal Voltage Reference Characteristics
Symbol
VBG
Parameter
Min
Typ
1.23
40
Max
1.40
70
Units
V
Bandgap reference voltage
Bandgap reference start-up time
Bandgap reference current consumption
1.15
tBG
µs
IBG
10
µA
Watchdog Timer
The Watchdog Timer is clocked from a separate On-chip Oscillator which runs at 1 Mhz.
This is the typical value at VCC = 5V. See characterization data for typical values at other
V
CC levels. By controlling the Watchdog Timer prescaler, the Watchdog Reset interval
can be adjusted as shown in Table 22 on page 56. The WDR – Watchdog Reset –
instruction resets the Watchdog Timer. The Watchdog Timer is also reset when it is dis-
abled and when a Chip Reset occurs. Eight different clock cycle periods can be selected
to determine the reset period. If the reset period expires without another Watchdog
Reset, the ATmega64 resets and executes from the Reset Vector. For timing details on
the Watchdog Reset, refer to page 53.
To prevent unintentional disabling of the Watchdog or unintentional change of Time-out
period, three different safety levels are selected by the fuses M103C and WDTON as
shown in Table 21. Safety level 0 corresponds to the setting in ATmega103. There is no
restriction on enabling the WDT in any of the safety levels. Refer to “Timed Sequences
for Changing the Configuration of the Watchdog Timer” on page 58 for details.
54
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 21. WDT Configuration as a Function of the Fuse Settings of M103C and
WDTON
WDT
Safety Initial
How to
Disable the
WDT
How to
Change
Time-out
M103C
WDTON
Level
State
Unprogrammed Unprogrammed
1
Disabled
Timed
Timed
sequence
sequence
Unprogrammed Programmed
2
0
2
Enabled
Disabled
Enabled
Always
enabled
Timed
sequence
Programmed
Programmed
Unprogrammed
Programmed
Timed
sequence
No restriction
Always
Timed
enabled
sequence
Figure 28. Watchdog Timer
WATCHDOG
OSCILLATOR
Watchdog Timer Control
Register – WDTCR
Bit
7
–
R
0
6
5
4
3
WDE
R/W
0
2
WDP2
R/W
0
1
WDP1
R/W
0
0
–
R
0
–
R
0
WDCE
R/W
0
WDP0
R/W
0
WDTCR
Read/Write
Initial Value
• Bits 7..5 – Res: Reserved Bits
These bits are reserved bits in the ATmega64 and will always read as zero.
• Bit 4 – WDCE: Watchdog Change Enable
This bit must be set when the WDE bit is written to logic zero. Otherwise, the Watchdog
will not be disabled. Once written to one, hardware will clear this bit after four clock
cycles. Refer to the description of the WDE bit for a Watchdog disable procedure. In
Safety Level 1 and 2, this bit must also be set when changing the prescaler bits. See
“Timed Sequences for Changing the Configuration of the Watchdog Timer” on page 58.
55
2490G–AVR–03/04
• Bit 3 – WDE: Watchdog Enable
When the WDE is written to logic one, the Watchdog Timer is enabled, and if the WDE is
written to logic zero, the Watchdog Timer function is disabled. WDE can only be cleared
if the WDCE bit has logic level one. To disable an enabled Watchdog Timer, the follow-
ing procedure must be followed:
1. In the same operation, write a logic one to WDCE and WDE. A logic one 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 logic 0 to WDE. This disables the
Watchdog.
In safety level 2, it is not possible to disable the Watchdog Timer, even with the algo-
rithm described above. See “Timed Sequences for Changing the Configuration of the
Watchdog Timer” on page 58.
• Bits 2..0 – WDP2, WDP1, WDP0: Watchdog Timer Prescaler 2, 1, and 0
The WDP2, WDP1, and WDP0 bits determine the Watchdog Timer prescaling when the
Watchdog Timer is enabled. The different prescaling values and their corresponding
Timeout Periods are shown in Table 22.
Table 22. Watchdog Timer Prescale Select
Number of WDT
Oscillator Cycles
TypicalTime-out
at VCC = 3.0V
Typical Time-out
at VCC = 5.0V
WDP2 WDP1 WDP0
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
16K (16,384)
32K (32,768)
17.1 ms
34.3 ms
68.5 ms
0.14 s
0.27 s
0.55 s
1.1 s
16.3 ms
32.5 ms
65 ms
0.13 s
0.26 s
0.52 s
1.0 s
64K (65,536)
128K (131,072)
256K (262,144)
512K (524,288)
1,024K (1,048,576)
2,048K (2,097,152)
2.2 s
2.1 s
56
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
The following code examples show one assembly and one C function for turning off the
WDT. The examples assume that interrupts are controlled (e.g., by disabling interrupts
globally) so that no interrupts will occur during execution of these functions.
Assembly Code Example
WDT_off:
; Write logical one to WDCE and WDE
ldi r16, (1<<WDCE)|(1<<WDE)
out WDTCR, r16
; Turn off WDT
ldi r16, (0<<WDE)
out WDTCR, r16
ret
C Code Example
void WDT_off(void)
{
/* Write logical one to WDCE and WDE */
WDTCR = (1<<WDCE) | (1<<WDE);
/* Turn off WDT */
WDTCR = 0x00;
}
57
2490G–AVR–03/04
Timed Sequences for
Changing the
The sequence for changing configuration differs slightly between the three safety levels.
Separate procedures are described for each level.
Configuration of the
Watchdog Timer
Safety Level 0
This mode is compatible with the Watchdog operation found in ATmega103. The Watch-
dog Timer is initially disabled, but can be enabled by writing the WDE bit to 1 without
any restriction. The Time-out period can be changed at any time without restriction. To
disable an enabled Watchdog Timer, the procedure described on page 56 (WDE bit
description) must be followed.
Safety Level 1
In this mode, the Watchdog Timer is initially disabled, but can be enabled by writing the
WDE bit to 1 without any restriction. A timed sequence is needed when changing the
Watchdog Time-out period or disabling an enabled Watchdog Timer. To disable an
enabled Watchdog Timer, and/or changing the Watchdog Time-out, the following proce-
dure must be followed:
1. In the same operation, write a logic one to WDCE and WDE. A logic one must be
written to WDE regardless of the previous value of the WDE bit.
2. Within the next four clock cycles, in the same operation, write the WDE and WDP
bits as desired, but with the WDCE bit cleared.
Safety Level 2
In this mode, the Watchdog Timer is always enabled, and the WDE bit will always read
as one. A timed sequence is needed when changing the Watchdog Time-out period. To
change the Watchdog Time-out, the following procedure must be followed:
1. In the same operation, write a logical one to WDCE and WDE. Even though the
WDE always is set, the WDE must be written to one to start the timed sequence.
2. Within the next four clock cycles, in the same operation, write the WDP bits as
desired, but with the WDCE bit cleared. The value written to the WDE bit is
irrelevant.
58
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Interrupts
This section describes the specifics of the interrupt handling as performed in
ATmega64. For a general explanation of the AVR interrupt handling, refer to “Reset and
Interrupt Handling” on page 13.
Interrupt Vectors in
ATmega64
Table 23. Reset and Interrupt Vectors
Vector
No.
Program
Address(2) Source
Interrupt Definition
1
0x0000(1)
RESET
External Pin, Power-on Reset, Brown-out Reset,
Watchdog Reset, and JTAG AVR Reset
2
0x0002
0x0004
0x0006
0x0008
0x000A
0x000C
0x000E
0x0010
0x0012
0x0014
0x0016
0x0018
0x001A
0x001C
0x001E
0x0020
0x0022
0x0024
0x0026
0x0028
0x002A
0x002C
0x002E
0x0030(3)
0x0032(3)
0x0034(3)
0x0036(3)
0x0038(3)
0x003A(3)
0x003C(3)
INT0
External Interrupt Request 0
External Interrupt Request 1
External Interrupt Request 2
External Interrupt Request 3
External Interrupt Request 4
External Interrupt Request 5
External Interrupt Request 6
External Interrupt Request 7
Timer/Counter2 Compare Match
Timer/Counter2 Overflow
3
INT1
4
INT2
5
INT3
6
INT4
7
INT5
8
INT6
9
INT7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
TIMER2 COMP
TIMER2 OVF
TIMER1 CAPT
TIMER1 COMPA
Timer/Counter1 Capture Event
Timer/Counter1 Compare Match A
TIMER1 COMPB Timer/Counter1 Compare Match B
TIMER1 OVF
TIMER0 COMP
TIMER0 OVF
SPI, STC
Timer/Counter1 Overflow
Timer/Counter0 Compare Match
Timer/Counter0 Overflow
SPI Serial Transfer Complete
USART0, Rx Complete
USART0 Data Register Empty
USART0, Tx Complete
ADC Conversion Complete
EEPROM Ready
USART0, RX
USART0, UDRE
USART0, TX
ADC
EE READY
ANALOG COMP
Analog Comparator
TIMER1 COMPC Timer/Countre1 Compare Match C
TIMER3 CAPT
Timer/Counter3 Capture Event
TIMER3 COMPA
Timer/Counter3 Compare Match A
TIMER3 COMPB Timer/Counter3 Compare Match B
TIMER3 COMPC Timer/Counter3 Compare Match C
TIMER3 OVF
USART1, RX
Timer/Counter3 Overflow
USART1, Rx Complete
59
2490G–AVR–03/04
Table 23. Reset and Interrupt Vectors (Continued)
Vector
No.
Program
Address(2) Source
Interrupt Definition
32
33
34
35
0x003E(3)
0x0040(3)
0x0042(3)
0x0044(3)
USART1, UDRE
USART1 Data Register Empty
USART1, Tx Complete
USART1, TX
TWI
Two-wire Serial Interface
Store Program Memory Ready
SPM READY
Notes: 1. When the BOOTRST Fuse is programmed, the device will jump to the Boot Loader
address at reset, see “Boot Loader Support – Read-While-Write Self-programming”
on page 277.
2. When the IVSEL bit in MCUCR is set, Interrupt Vectors will be moved to the start of
the Boot Flash section. The address of each Interrupt Vector will then be address in
this table added to the start address of the Boot Flash section.
3. The Interrupts on address 0x0030 - 0x0044 do not exist in ATmega103 compatibility
mode.
Table 24 shows Reset and Interrupt Vectors placement for the various combinations of
BOOTRST and IVSEL settings. If the program never enables an interrupt source, the
Interrupt Vectors are not used, and regular program code can be placed at these loca-
tions. This is also the case if the Reset Vector is in the Application section while the
Interrupt Vectors are in the Boot section or vice versa.
Table 24. Reset and Interrupt Vectors Placement(1)
BOOTRST
IVSEL
Reset Address
0x0000
Interrupt Vectors Start Address
0x0002
1
1
0
0
0
1
0
1
0x0000
Boot Reset Address + 0x0002
0x0002
Boot Reset Address
Boot Reset Address
Boot Reset Address + 0x0002
Note:
1. The Boot Reset Address is shown in Table 113 on page 289. For the BOOTRST
Fuse “1” means unprogrammed while “0” means programmed.
The most typical and general program setup for the Reset and Interrupt Vector
Addresses in ATmega64 is:
AddressLabels Code
Comments
0x0000
0x0002
0x0004
0x0006
0x0008
0x000A
0x000C
0x000E
0x0010
0x0012
0x0014
0x0016
0x0018
0x001A
0x001C
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
RESET
; Reset Handler
EXT_INT0
EXT_INT1
EXT_INT2
EXT_INT3
EXT_INT4
EXT_INT5
EXT_INT6
EXT_INT7
TIM2_COMP
TIM2_OVF
TIM1_CAPT
TIM1_COMPA
TIM1_COMPB
TIM1_OVF
; IRQ0 Handler
; IRQ1 Handler
; IRQ2 Handler
; IRQ3 Handler
; IRQ4 Handler
; IRQ5 Handler
; IRQ6 Handler
; IRQ7 Handler
; Timer2 Compare Handler
; Timer2 Overflow Handler
; Timer1 Capture Handler
; Timer1 CompareA Handler
; Timer1 CompareB Handler
; Timer1 Overflow Handler
60
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
0x001E
0x0020
0x0022
0x0024
0x0026
0x0028
0x002A
0x002C
0x002E
0x0030
0x0032
0x0034
0x0036
0x0038
0x003A
0x003C
0x003E
0x0040
0x0042
0x0044
;
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
jmp
TIM0_COMP
TIM0_OVF
SPI_STC
; Timer0 Compare Handler
; Timer0 Overflow Handler
; SPI Transfer Complete Handler
; USART0 RX Complete Handler
; USART0,UDR Empty Handler
; USART0 TX Complete Handler
; ADC Conversion Complete Handler
; EEPROM Ready Handler
USART0_RXC
USART0_DRE
USART0_TXC
ADC
EE_RDY
ANA_COMP
TIM1_COMPC
TIM3_CAPT
TIM3_COMPA
TIM3_COMPB
TIM3_COMPC
TIM3_OVF
USART1_RXC
USART1_DRE
USART1_TXC
TWI
; Analog Comparator Handler
; Timer1 CompareC Handler
; Timer3 Capture Handler
; Timer3 CompareA Handler
; Timer3 CompareB Handler
; Timer3 CompareC Handler
; Timer3 Overflow Handler
; USART1 RX Complete Handler
; USART1,UDR Empty Handler
; USART1 TX Complete Handler
; Two-wire Serial Interface Handler
; SPM Ready Handler
SPM_RDY
0x0046 RESET: ldi
r16, high(RAMEND); Main program start
0x0047
0x0048
out
ldi
SPH,r16
; Set Stack Pointer to top of RAM
r16, low(RAMEND)
0x0049
0x004A
out
sei
SPL,r16
; Enable interrupts
0x004B
...
<instr> xxx
... ...
...
When the BOOTRST Fuse is unprogrammed, the Boot section size set to 8K bytes and
the IVSEL bit in the MCUCR Register is set before any interrupts are enabled, the most
typical and general program setup for the Reset and Interrupt Vector Addresses is:
AddressLabels Code
Comments
0x0000 RESET: ldi
r16,high(RAMEND); Main program start
0x0001
0x0002
out
ldi
SPH,r16
; Set Stack Pointer to top of RAM
r16,low(RAMEND)
SPL,r16
0x0003
0x0004
out
sei
; Enable interrupts
0x0005
;
<instr> xxx
.org 0x7002
0x7002
0x7004
...
jmp
jmp
...
jmp
EXT_INT0
; IRQ0 Handler
; IRQ1 Handler
;
EXT_INT1
...
0x7044
SPM_RDY ; Store Program Memory Ready Handler
61
2490G–AVR–03/04
When the BOOTRST Fuse is programmed and the Boot section size set to 8K bytes, the
most typical and general program setup for the Reset and Interrupt Vector Addresses is:
AddressLabels Code
Comments
.org 0x0002
0x0002
0x0004
...
jmp
jmp
...
jmp
EXT_INT0 ; IRQ0 Handler
EXT_INT1 ; IRQ1 Handler
...
;
0x0044
;
SPM_RDY ; Store Program Memory Ready Handler
.org 0x7000
0x7000 RESET: ldi
r16,high(RAMEND); Main program start
SPH,r16 ; Set Stack Pointer to top of RAM
r16,low(RAMEND)
0x7001
0x7002
out
ldi
0x7003
0x7004
out
sei
SPL,r16
; Enable interrupts
0x7005
<instr> xxx
When the BOOTRST Fuse is programmed, the Boot section size set to 8K bytes and the
IVSEL bit in the MCUCR Register is set before any interrupts are enabled, the most typ-
ical and general program setup for the Reset and Interrupt Vector Addresses is:
AddressLabels Code
Comments
;
.org 0x7000
0x7000
0x7002
jmp
jmp
RESET
; Reset handler
EXT_INT0 ; IRQ0 Handler
0x7004
...
jmp
...
jmp
EXT_INT1 ; IRQ1 Handler
...
;
0x7044
SPM_RDY; Store Program Memory Ready Handler
r16,high(RAMEND); Main program start
SPH,r16 ; Set Stack Pointer to top of RAM
r16,low(RAMEND)
0x7046 RESET: ldi
0x7047
0x7048
out
ldi
0x7049
0x704A
out
sei
SPL,r16
; Enable interrupts
0x704B
<instr> xxx
Moving Interrupts Between
Application and Boot Space
The General Interrupt Control Register controls the placement of the Interrupt Vector
table.
MCU Control Register –
MCUCR
Bit
7
6
SRW10
R/W
0
5
SE
R/W
0
4
3
2
1
IVSEL
R/W
0
0
IVCE
R/W
0
SRE
R/W
0
SM1
R/W
0
SM0
R/W
0
SM2
R/W
0
MCUCR
Read/Write
Initial Value
• Bit 1 – IVSEL: Interrupt Vector Select
When the IVSEL bit is cleared (zero), the Interrupt Vectors are placed at the start of the
Flash memory. When this bit is set (one), the Interrupt Vectors are moved to the begin-
ning of the Boot Loader section of the Flash. The actual address of the start of the Boot
Flash section is determined by the BOOTSZ Fuses. Refer to the section “Boot Loader
Support – Read-While-Write Self-programming” on page 277 for details. To avoid unin-
tentional changes of Interrupt Vector tables, a special write procedure must be followed
to change the IVSEL bit:
62
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
1. Write the Interrupt Vector Change Enable (IVCE) bit to one.
2. Within four cycles, write the desired value to IVSEL while writing a zero to IVCE.
Interrupts will automatically be disabled while this sequence is executed. Interrupts are
disabled in the cycle IVCE is set, and they remain disabled until after the instruction fol-
lowing the write to IVSEL. If IVSEL is not written, interrupts remain disabled for four
cycles. The I-bit in the Status Register is unaffected by the automatic disabling.
Note:
If Interrupt Vectors are placed in the Boot Loader section and Boot Lock bit BLB02 is pro-
grammed, interrupts are disabled while executing from the Application section. If
Interrupt Vectors are placed in the Application section and Boot Lock bit BLB12 is pro-
gramed, interrupts are disabled while executing from the Boot Loader section. Refer to
the section “Boot Loader Support – Read-While-Write Self-programming” on page 277
for details on Boot Lock bits.
• Bit 0 – IVCE: Interrupt Vector Change Enable
The IVCE bit must be written to logic one to enable change of the IVSEL bit. IVCE is
cleared by hardware four cycles after it is written or when IVSEL is written. Setting the
IVCE bit will disable interrupts, as explained in the IVSEL description above. See code
examples below.
Assembly Code Example
Move_interrupts:
; Enable change of Interrupt Vectors
ldi r16, (1<<IVCE)
out MCUCR, r16
; Move interrupts to boot Flash section
ldi r16, (1<<IVSEL)
out MCUCR, r16
ret
C Code Example
void Move_interrupts(void)
{
/* Enable change of Interrupt Vectors */
MCUCR = (1<<IVCE);
/* Move interrupts to boot Flash section */
MCUCR = (1<<IVSEL);
}
63
2490G–AVR–03/04
I/O Ports
Introduction
All 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 uninten-
tionally changing the direction of any other pin with the SBI and CBI instructions. The
same applies when changing drive value (if configured as output) or enabling/disabling
of pull-up resistors (if configured as input). Each output buffer has symmetrical drive
characteristics with both high sink and source capability. The pin driver is strong enough
to drive LED displays directly. All port pins have individually selectable pull-up resistors
with a supply voltage invariant resistance. All I/O pins have protection diodes to both
V
CC and Ground as indicated in Figure 29. Refer to “Electrical Characteristics” on page
326 for a complete list of parameters.
Figure 29. I/O Pin Equivalent Schematic
Rpu
Pxn
Logic
Cpin
See Figure
"General Digital I/O" for
Details
All registers and bit references in this section are written in general form. A lower case
“x” represents the numbering letter for the port, and a lower case “n” represents the bit
number. However, when using the register or bit defines in a program, the precise form
must be used (i.e., PORTB3 for bit no. 3 in Port B, here documented generally as
PORTxn). The physical I/O Registers and bit locations are listed in “Register Description
for I/O Ports” on page 85.
Three I/O memory address locations are allocated for each port, one each for the Data
Register – PORTx, Data Direction Register – DDRx, and the Port Input Pins – PINx. The
Port Input Pins I/O location is read only, while the Data Register and the Data Direction
Register are read/write. In addition, the Pull-up Disable – PUD bit in SFIOR disables the
pull-up function for all pins in all ports when set.
Using the I/O port as general digital I/O is described in “Ports as General Digital I/O” on
page 65. Most port pins are multiplexed with alternate functions for the peripheral fea-
tures on the device. How each alternate function interferes with the port pin is described
in “Alternate Port Functions” on page 69. Refer to the individual module sections for a
full description of the alternate functions.
Note that enabling the alternate function of some of the port pins does not affect the use
of the other pins in the port as general digital I/O.
64
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Ports as General Digital
I/O
The ports are bi-directional I/O ports with optional internal pull-ups. Figure 30 shows a
functional description of one I/O-port pin, here generically called Pxn.
Figure 30. General Digital I/O(1)
PUD
Q
D
DDxn
Q CLR
WDx
RDx
RESET
Q
D
Pxn
PORTxn
Q CLR
WPx
RRx
RESET
SLEEP
SYNCHRONIZER
RPx
D
Q
D
L
Q
Q
PINxn
Q
clk I/O
WDx:
RDx:
WPx:
RRx:
RPx:
WRITE DDRx
PUD:
PULLUP DISABLE
SLEEP CONTROL
I/O CLOCK
READ DDRx
SLEEP:
WRITE PORTx
clkI/O
:
READ PORTx REGISTER
READ PORTx PIN
Note:
1. WPx, WDx, RRx, RPx, and RDx are common to all pins within the same port. clkI/O
,
SLEEP, and PUD are common to all ports.
Configuring the Pin
Each port pin consists of three register bits: DDxn, PORTxn, and PINxn. As shown in
“Register Description for I/O Ports” on page 85, the DDxn bits are accessed at the DDRx
I/O address, the PORTxn bits at the PORTx I/O address, and the PINxn bits at the PINx
I/O address.
The DDxn bit in the DDRx Register selects the direction of this pin. If DDxn is written
logic one, Pxn is configured as an output pin. If DDxn is written logic zero, Pxn is config-
ured as an input pin.
If PORTxn is written logic one when the pin is configured as an input pin, the pull-up
resistor is activated. To switch the pull-up resistor off, PORTxn has to be written logic
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 no clocks are running.
If PORTxn is written logic one when the pin is configured as an output pin, the port pin is
driven high (one). If PORTxn is written logic zero when the pin is configured as an out-
put pin, the port pin is driven low (zero).
65
2490G–AVR–03/04
When switching between tri-state ({DDxn, PORTxn} = 0b00) and output high ({DDxn,
PORTxn} = 0b11), an intermediate state with either pull-up enabled ({DDxn, PORTxn} =
0b01) or output low ({DDxn, PORTxn} = 0b10) must occur. Normally, the pull-up
enabled state is fully acceptable, as a high-impedant environment will not notice the dif-
ference between a strong high driver and a pull-up. If this is not the case, the PUD bit in
the SFIOR Register can be written to one to disable all pull-ups in all ports.
Switching between input with pull-up and output low generates the same problem. The
user must use either the tri-state ({DDxn, PORTxn} = 0b00) or the output high state
({DDxn, PORTxn} = 0b11) as an intermediate step.
Table 25 summarizes the control signals for the pin value.
Table 25. Port Pin Configurations
PUD
DDxn PORTxn (in SFIOR)
I/O
Pull-up Comment
0
0
0
1
X
0
Input
Input
No
Tri-state (Hi-Z)
Yes
Pxn will source current if ext. pulled
low.
0
1
1
1
0
1
1
X
X
Input
Output
Output
No
No
No
Tri-state (Hi-Z)
Output Low (Sink)
Output High (Source)
Reading the Pin Value
Independent of the setting of Data Direction bit DDxn, the port pin can be read through
the PINxn Register bit. As shown in Figure 30, the PINxn Register bit and the preceding
latch constitute a synchronizer. This is needed to avoid metastability if the physical pin
changes value near the edge of the internal clock, but it also introduces a delay. Figure
31 shows a timing diagram of the synchronization when reading an externally applied
pin value. The maximum and minimum propagation delays are denoted tpd,max and tpd,min
respectively.
Figure 31. Synchronization when Reading an Externally Applied Pin Value
SYSTEM CLK
in r17, PINx
XXX
XXX
INSTRUCTIONS
SYNC LATCH
PINxn
0xFF
r17
0x00
tpd, max
tpd, min
66
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Consider the clock period starting shortly after the first falling edge of the system clock.
The latch is closed when the clock is low, and goes transparent when the clock is high,
as indicated by the shaded region of the “SYNC LATCH” signal. The signal value is
latched when the system clock goes low. It is clocked into the PINxn Register at the suc-
ceeding positive clock edge. As indicated by the two arrows tpd,max and tpd,min, a single
signal transition on the pin will be delayed between ½ and 1-½ system clock period
depending upon the time of assertion.
When reading back a software assigned pin value, a nop instruction must be inserted as
indicated in Figure 32. The out instruction sets the “SYNC LATCH” signal at the positive
edge of the clock. In this case, the delay tpd through the synchronizer is one system
clock period.
Figure 32. Synchronization when Reading a Software Assigned Pin Value
SYSTEM CLK
r16
0xFF
nop
in r17, PINx
out PORTx, r16
INSTRUCTIONS
SYNC LATCH
PINxn
0xFF
0x00
tpd
r17
67
2490G–AVR–03/04
The following code example show how to set Port B pins 0 and 1 high, 2 and 3 low, and
define the port pins from 4 to 7 as input with pull-ups assigned to port pins 6 and 7. The
resulting pin values are read back again, but as previously discussed, a nop instruction
is included to be able to read back the value recently assigned to some of the pins.
Assembly Code Example(1)
...
; Define pull-ups and set outputs high
; Define directions for port pins
ldi
ldi
out
out
r16,(1<<PB7)|(1<<PB6)|(1<<PB1)|(1<<PB0)
r17,(1<<DDB3)|(1<<DDB2)|(1<<DDB1)|(1<<DDB0)
PORTB,r16
DDRB,r17
; Insert nop for synchronization
nop
; Read port pins
in
r16,PINB
...
C Code Example
unsigned char i;
...
/* Define pull-ups and set outputs high */
/* Define directions for port pins */
PORTB = (1<<PB7)|(1<<PB6)|(1<<PB1)|(1<<PB0);
DDRB = (1<<DDB3)|(1<<DDB2)|(1<<DDB1)|(1<<DDB0);
/* Insert nop for synchronization*/
_NOP();
/* Read port pins */
i = PINB;
...
Note:
1. For the assembly program, two temporary registers are used to minimize the time
from pull-ups are set on pins 0, 1, 6, and 7, until the direction bits are correctly set,
defining bit 2 and 3 as low and redefining bits 0 and 1 as strong high drivers.
Digital Input Enable and Sleep As shown in Figure 30, the digital input signal can be clamped to ground at the input of
Modes
the Schmitt Trigger. The signal denoted SLEEP in the figure, is set by the MCU Sleep
Controller in Power-down mode, Power-save mode, Standby mode, and Extended
Standby mode to avoid high power consumption if some input signals are left floating, or
have an analog signal level close to VCC/2.
SLEEP is overridden for port pins enabled as External Interrupt pins. If the External
Interrupt request is not enabled, SLEEP is active also for these pins. SLEEP is also
overridden by various other alternate functions as described in “Alternate Port Func-
tions” on page 69.
If a logic high level (“one”) is present on an asynchronous External Interrupt pin config-
ured as “Interrupt on Any Logic Change on Pin” while the External Interrupt is not
enabled, the corresponding External Interrupt Flag will be set when resuming from the
above mentioned sleep modes, as the clamping in these sleep modes produces the
requested logic change.
68
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Alternate Port Functions Most port pins have alternate functions in addition to being general digital I/Os. Figure
33 shows how the port pin control signals from the simplified Figure 30 can be overrid-
den by alternate functions. The overriding signals may not be present in all port pins, but
the figure serves as a generic description applicable to all port pins in the AVR micro-
controller family.
Figure 33. Alternate Port Functions(1)
PUOExn
PUOVxn
1
PUD
0
DDOExn
DDOVxn
1
Q
D
0
DDxn
Q CLR
WDx
RDx
PVOExn
PVOVxn
RESET
1
0
Pxn
Q
D
PORTxn
Q CLR
DIEOExn
DIEOVxn
SLEEP
WPx
RRx
RESET
1
0
SYNCHRONIZER
RPx
SET
D
L
Q
Q
D
Q
PINxn
CLR Q
CLR
clk I/O
DIxn
AIOxn
PUOExn: Pxn PULL-UP OVERRIDE ENABLE
PUOVxn: Pxn PULL-UP OVERRIDE VALUE
DDOExn: Pxn DATA DIRECTION OVERRIDE ENABLE
DDOVxn: Pxn DATA DIRECTION OVERRIDE VALUE
PVOExn: Pxn PORT VALUE OVERRIDE ENABLE
PVOVxn: Pxn PORT VALUE OVERRIDE VALUE
PUD:
WDx:
RDx:
RRx:
WPx:
RPx:
PULLUP DISABLE
WRITE DDRx
READ DDRx
READ PORTx REGISTER
WRITE PORTx
READ PORTx PIN
DIEOExn: Pxn DIGITAL INPUT-ENABLE OVERRIDE ENABLE
DIEOVxn: Pxn DIGITAL INPUT-ENABLE OVERRIDE VALUE
clkI/O
DIxn:
AIOxn:
:
I/O CLOCK
DIGITAL INPUT PIN n ON PORTx
ANALOG INPUT/OUTPUT PIN n ON PORTx
SLEEP:
SLEEP CONTROL
Note:
1. WPx, WDx, RLx, RPx, and RDx are common to all pins within the same port. clkI/O
,
SLEEP, and PUD are common to all ports. All other signals are unique for each pin.
Table 26 summarizes the function of the overriding signals. The pin and port indexes
from Figure 33 are not shown in the succeeding tables. The overriding signals are gen-
erated internally in the modules having the alternate function.
69
2490G–AVR–03/04
Table 26. Generic Description of Overriding Signals for Alternate Functions
Signal
Name
Full Name
Description
PUOE
Pull-up Override
Enable
If this signal is set, the pull-up enable is controlled by the
PUOV signal. If this signal is cleared, the pull-up is enabled
when {DDxn, PORTxn, PUD} = 0b010.
PUOV
DDOE
DDOV
PVOE
Pull-up Override
Value
If PUOE is set, the pull-up is enabled/disabled when PUOV
is set/cleared, regardless of the setting of the DDxn,
PORTxn, and PUD Register bits.
Data Direction
Override Enable
If this signal is set, the Output Driver Enable is controlled by
the DDOV signal. If this signal is cleared, the Output driver is
enabled by the DDxn Register bit.
Data Direction
Override Value
If DDOE is set, the Output Driver is enabled/disabled when
DDOV is set/cleared, regardless of the setting of the DDxn
Register bit.
Port Value Override
Enable
If this signal is set and the Output Driver is enabled, the port
value is controlled by the PVOV signal. If PVOE is cleared,
and the Output Driver is enabled, the port Value is controlled
by the PORTxn Register bit.
PVOV
Port Value Override
Value
If PVOE is set, the port value is set to PVOV, regardless of
the setting of the PORTxn Register bit.
DIEOE
Digital Input Enable
Override Enable
If this bit is set, the Digital Input Enable is controlled by the
DIEOV signal. If this signal is cleared, the Digital Input
Enable is determined by MCU state (Normal mode, sleep
modes).
DIEOV
DI
Digital Input Enable
Override Value
If DIEOE is set, the Digital Input is enabled/disabled when
DIEOV is set/cleared, regardless of the MCU state (Normal
mode, sleep modes).
Digital Input
This is the Digital Input to alternate functions. In the figure,
the signal is connected to the output of the Schmitt Trigger
but before the synchronizer. Unless the Digital Input is used
as a clock source, the module with the alternate function will
use its own synchronizer.
AIO
Analog Input/output
This is the Analog Input/output to/from alternate functions.
The signal is connected directly to the pad, and can be used
bi-directionally.
The following subsections shortly describe the alternate functions for each port, and
relate the overriding signals to the alternate function. Refer to the alternate function
description for further details.
Special Function IO Register –
SFIOR
Bit
7
6
–
R
0
5
–
R
0
4
–
R
0
3
ACME
R/W
0
2
1
PSR0
R/W
0
0
PSR321
R/W
0
TSM
R/W
0
PUD
R/W
0
SFIOR
Read/Write
Initial Value
• Bit 2 – PUD: Pull-up disable
When this bit is written to one, the pull-ups in the I/O ports are disabled even if the DDxn
and PORTxn Registers are configured to enable the pull-ups ({DDxn, PORTxn} = 0b01).
See “Configuring the Pin” on page 65 for more details about this feature.
70
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Alternate Functions of Port A The Port A has an alternate function as the address low byte and data lines for the
External Memory Interface.
Table 27. Port A Pins Alternate Functions
Port Pin
PA7
Alternate Function
AD7 (External memory interface address and data bit 7)
AD6 (External memory interface address and data bit 6)
AD5 (External memory interface address and data bit 5)
AD4 (External memory interface address and data bit 4)
AD3 (External memory interface address and data bit 3)
AD2 (External memory interface address and data bit 2)
AD1 (External memory interface address and data bit 1)
AD0 (External memory interface address and data bit 0)
PA6
PA5
PA4
PA3
PA2
PA1
PA0
Table 28 and Table 29 relates the alternate functions of Port A to the overriding signals
shown in Figure 33 on page 69.
Table 28. Overriding Signals for Alternate Functions in PA7..PA4
Signal
Name
PUOE
PUOV
PA7/AD7
PA6/AD6
PA5/AD5
PA4/AD4
SRE
SRE
SRE
SRE
~(WR | ADA(1)) •
PORTA7 • PUD
~(WR | ADA) •
PORTA6 • PUD
~(WR | ADA) •
PORTA5 • PUD
~(WR | ADA) •
PORTA4 • PUD
SRE
WR | ADA
SRE
SRE
WR | ADA
SRE
SRE
WR | ADA
SRE
SRE
WR | ADA
SRE
DDOE
DDOV
PVOE
PVOV
A7 • ADA | D7
OUTPUT • WR
A6 • ADA | D6
OUTPUT • WR
A5 • ADA | D5
OUTPUT • WR
A4 • ADA | D4
OUTPUT • WR
DIEOE
DIEOV
DI
0
0
0
0
0
D7 INPUT
–
0
D6 INPUT
–
0
D5 INPUT
–
0
D4 INPUT
–
AIO
71
2490G–AVR–03/04
Table 29. Overriding Signals for Alternate Functions in PA3..PA0(1)
Signal
Name
PUOE
PUOV
PA3/AD3
PA2/AD2
PA1/AD1
PA0/AD0
SRE
SRE
SRE
SRE
~(WR | ADA) •
PORTA3 • PUD
~(WR | ADA) •
PORTA2 • PUD
~(WR | ADA) •
PORTA1 • PUD
~(WR | ADA) •
PORTA0 • PUD
SRE
WR | ADA
SRE
SRE
WR | ADA
SRE
SRE
WR | ADA
SRE
SRE
WR | ADA
SRE
DDOE
DDOV
PVOE
PVOV
A3 • ADA | D3
OUTPUT • WR
A2• ADA | D2
OUTPUT • WR
A1 • ADA | D1
OUTPUT • WR
A0 • ADA | D0
OUTPUT • WR
0
0
0
0
DIEOE
DIEOV
DI
0
D3 INPUT
–
0
D2 INPUT
–
0
D1 INPUT
–
0
D0 INPUT
–
AIO
Note:
1. ADA is short for ADdress Active and represents the time when address is output. See
“External Memory Interface” on page 25 for details.
Alternate Functions of Port B The Port B pins with alternate functions are shown in Table 30.
Table 30. Port B Pins Alternate Functions
Port Pin Alternate Functions
PB7
OC2/OC1C(1) (Output Compare and PWM Output for Timer/Counter2 or Output
Compare and PWM Output C for Timer/Counter1)
OC1B (Output Compare and PWM Output B for Timer/Counter1)
OC1A (Output Compare and PWM Output A for Timer/Counter1)
OC0 (Output Compare and PWM Output for Timer/Counter0)
MISO (SPI Bus Master Input/Slave Output)
PB6
PB5
PB4
PB3
PB2
PB1
PB0
MOSI (SPI Bus Master Output/Slave Input)
SCK (SPI Bus Serial Clock)
SS (SPI Slave Select input)
Note:
1. OC1C not applicable in ATmega103 compatibility mode.
The alternate pin configuration is as follows:
• OC2/OC1C, Bit 7
OC2, Output Compare Match output: The PB7 pin can serve as an external output for
the Timer/Counter2 Output Compare. The pin has to be configured as an output (DDB7
set (one)) to serve this function. The OC2 pin is also the output pin for the PWM mode
timer function.
OC1C, Output Compare Match C output: The PB7 pin can serve as an external output
for the Timer/Counter1 Output Compare C. The pin has to be configured as an output
(DDB7 set (one)) to serve this function. The OC1C pin is also the output pin for the
PWM mode timer function.
72
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• OC1B, Bit 6
OC1B, Output Compare Match B output: The PB6 pin can serve as an external output
for the Timer/Counter1 Output Compare B. The pin has to be configured as an output
(DDB6 set (one)) to serve this function. The OC1B pin is also the output pin for the PWM
mode timer function.
• OC1A, Bit 5
OC1A, Output Compare Match A output: The PB5 pin can serve as an external output
for the Timer/Counter1 Output Compare A. The pin has to be configured as an output
(DDB5 set (one)) to serve this function. The OC1A pin is also the output pin for the PWM
mode timer function.
• OC0, Bit 4
OC0, Output Compare Match output: The PB4 pin can serve as an external output for
the Timer/Counter0 Output Compare. The pin has to be configured as an output (DDB4
set (one)) to serve this function. The OC0 pin is also the output pin for the PWM mode
timer 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.
• 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.
• 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.
• 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 con-
trolled by the PORTB0 bit.
Table 31 and Table 32 relate the alternate functions of Port B to the overriding signals
shown in Figure 33 on page 69. SPI MSTR INPUT and SPI SLAVE OUTPUT constitute
the MISO signal, while MOSI is divided into SPI MSTR OUTPUT and SPI SLAVE
INPUT.
73
2490G–AVR–03/04
Table 31. Overriding Signals for Alternate Functions in PB7..PB4
Signal
Name
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
PB7/OC2/OC1C
PB6/OC1B
PB5/OC1A
PB4/OC0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
OC2/OC1C ENABLE(1)
OC1B ENABLE
OC1A ENABLE
OC0 ENABLE
OC2/OC1C(1)
OC1B
OC1A
OC0B
0
0
–
–
0
0
–
–
0
0
–
–
0
0
–
–
AIO
Note:
1. See “Output Compare Modulator (OCM1C2)” on page 159 for details. OC1C does
not exist in ATmega103 compatibility mode.
Table 32. Overriding Signals for Alternate Functions in PB3..PB0
Signal
Name
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
PB3/MISO
SPE • MSTR
PORTB3 • PUD
SPE • MSTR
0
PB2/MOSI
SPE • MSTR
PORTB2 • PUD
SPE • MSTR
0
PB1/SCK
PB0/SS
SPE • MSTR
SPE • MSTR
PORTB1 • PUD PORTB0 • PUD
SPE • MSTR
0
SPE • MSTR
0
SPE • MSTR
SPE • MSTR
SPE • MSTR
0
SPI SLAVE OUTPUT SPI MSTR OUTPUT SCK OUTPUT
0
0
0
0
0
0
0
0
0
SPI MSTR INPUT
–
SPI SLAVE INPUT
–
SCK INPUT
–
SPI SS
–
AIO
74
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Alternate Functions of Port C In ATmega103 compatibility mode, Port C is output only. The Port C has an alternate
function as the address high byte for the External Memory Interface
Table 33. Port C Pins Alternate Functions
Port Pin
PC7
Alternate Function
A15
A14
A13
A12
A11
A10
A9
PC6
PC5
PC4
PC3
PC2
PC1
PC0
A8
Table 34 and Table 35 relate the alternate functions of Port C to the overriding signals
shown in Figure 33 on page 69.
Table 34. Overriding Signals for Alternate Functions in PC7..PC4
Signal Name
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
PC7/A15
PC6/A14
PC5/A13
PC4/A12
SRE • (XMM(1)<1)
SRE • (XMM<2)
SRE • (XMM<3)
SRE • (XMM<4)
0
0
0
0
SRE • (XMM<1)
SRE • (XMM<2)
SRE • (XMM<3)
SRE • (XMM<4)
1
1
1
1
SRE • (XMM<1)
SRE • (XMM<2)
SRE • (XMM<3)
SRE • (XMM<4)
A11
0
A10
0
A9
0
A8
0
0
0
0
0
–
–
–
–
AIO
–
–
–
–
75
2490G–AVR–03/04
Table 35. Overriding Signals for Alternate Functions in PC3..PC0(1)
Signal Name
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
PC3/A11
PC2/A10
PC1/A9
PC0/A8
SRE • (XMM<5)
SRE • (XMM<6)
SRE • (XMM<7)
SRE • (XMM<7)
0
0
0
0
SRE • (XMM<5)
SRE • (XMM<6)
SRE • (XMM<7)
SRE • (XMM<7)
1
1
1
1
SRE • (XMM<5)
SRE • (XMM<6)
SRE • (XMM<7)
SRE • (XMM<7)
A11
0
A10
0
A9
0
A8
0
0
0
0
0
–
–
–
–
AIO
–
–
–
–
Note:
1. XMM = 0 in ATmega103 compatibility mode.
Alternate Functions of Port D The Port D pins with alternate functions are shown in Table 36.
Table 36. Port D Pins Alternate Functions
Port Pin
PD7
Alternate Function
T2 (Timer/Counter2 Clock Input)
PD6
T1 (Timer/Counter1 Clock Input)
PD5
XCK1(1) (USART1 External Clock Input/Output)
IC1 (Timer/Counter1 Input Capture Trigger)
INT3/TXD1(1) (External Interrupt3 Input or UART1 Transmit Pin)
INT2/RXD1(1) (External Interrupt2 Input or UART1 Receive Pin)
INT1/SDA(1) (External Interrupt1 Input or TWI Serial DAta)
INT0/SCL(1) (External Interrupt0 Input or TWI Serial CLock)
PD4
PD3
PD2
PD1
PD0
Note:
1. XCK1, TXD1, RXD1, SDA, and SCL not applicable in ATmega103 compatibility
mode.
The alternate pin configuration is as follows:
• T2 – Port D, Bit 7
T2, Timer/Counter2 Counter Source.
• T1 – Port D, Bit 6
T1, Timer/Counter1 Counter Source.
• XCK1 – Port D, Bit 5
XCK1, USART1 External Clock. The Data Direction Register (DDD5) controls whether
the clock is output (DDD5 set) or input (DDD5 cleared). The XCK1 pin is active only
when the USART1 operates in synchronous mode.
76
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• IC1 – Port D, Bit 4
IC1 – Input Capture Pin1: The PD4 pin can act as an Input Capture pin for
Timer/Counter1.
• INT3/TXD1 – Port D, Bit 3
INT3, External Interrupt Source 3: The PD3 pin can serve as an External Interrupt
source to the MCU.
TXD1, Transmit Data (Data output pin for the USART1). When the USART1 transmitter
is enabled, this pin is configured as an output regardless of the value of DDD3.
• INT2/RXD1 – Port D, Bit 2
INT2, External Interrupt source 2. The PD2 pin can serve as an External Interrupt
source to the MCU.
RXD1, Receive Data (Data input pin for the USART1). When the USART1 receiver is
enabled this pin is configured as an input regardless of the value of DDD2. When the
USART forces this pin to be an input, the pull-up can still be controlled by the PORTD2
bit.
• INT1/SDA – Port D, Bit 1
INT1, External Interrupt Source 1. The PD1 pin can serve as an External Interrupt
source to the MCU.
SDA, Two-wire Serial Interface Data: When the TWEN bit in TWCR is set (one) to
enable the Two-wire Serial Interface, pin PD1 is disconnected from the port and
becomes the serial data I/O pin for the Two-wire Serial Interface. In this mode, there is a
spike filter on the pin to suppress spikes shorter than 50 ns on the input signal, and the
pin is driven by an open drain driver with slew-rate limitation.
• INT0/SCL – Port D, Bit 0
INT0, External Interrupt Source 0. The PD0 pin can serve as an External Interrupt
source to the MCU.
SCL, Two-wire Serial Interface Clock: When the TWEN bit in TWCR is set (one) to
enable the Two-wire Serial Interface, pin PD0 is disconnected from the port and
becomes the serial clock I/O pin for the Two-wire Serial Interface. In this mode, there is
a spike filter on the pin to suppress spikes shorter than 50 ns on the input signal, and the
pin is driven by an open drain driver with slew-rate limitation.
Table 37 and Table 38 relates the alternate functions of Port D to the overriding signals
shown in Figure 33 on page 69.
77
2490G–AVR–03/04
Table 37. Overriding Signals for Alternate Functions PD7..PD4
Signal Name
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
PD7/T2
PD6/T1
PD5/XCK1
PD4/IC1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
UMSEL1
0
0
0
XCK1 OUTPUT
0
0
0
0
0
0
0
0
0
T2 INPUT
–
T1 INPUT
–
XCK1 INPUT
–
IC1 INPUT
–
AIO
Table 38. Overriding Signals for Alternate Functions in PD3..PD0(1)
Signal Name PD3/INT3/TXD1 PD2/INT2/RXD1
PD1/INT1/SDA
PD0/INT0/SCL
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
TXEN1
RXEN1
TWEN
TWEN
0
PORTD2 • PUD
PORTD1 • PUD PORTD0 • PUD
TXEN1
RXEN1
TWEN
TWEN
1
0
SDA_OUT
SCL_OUT
TWEN
TXEN1
0
TWEN
TXD1
0
0
0
INT3 ENABLE
INT2 ENABLE
1
INT1 ENABLE
1
INT0 ENABLE
1
1
INT3 INPUT
–
INT2 INPUT/RXD1 INT1 INPUT
– SDA INPUT
INT0 INPUT
SCL INPUT
AIO
Note:
1. When enabled, the Two-wire Serial Interface enables Slew-rate controls on the output
pins PD0 and PD1. This is not shown on the figure. In addition, spike filters are con-
nected between the AIO outputs shown in the port figure and the digital logic of the
TWI module.
78
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Alternate Functions of Port E
The Port E pins with alternate functions are shown in Table 39.
Table 39. Port E Pins Alternate Functions
Port Pin Alternate Function
PE7
PE6
PE5
INT7/IC3(1) (External Interrupt 7 Input or Timer/Counter3 Input Capture Trigger)
INT6/ T3(1) (External Interrupt 6 Input or Timer/Counter3 Clock Input)
INT5/OC3C(1) (External Interrupt 5 Input or Output Compare and PWM Output C
for Timer/Counter3)
PE4
PE3
PE2
INT4/OC3B(1) (External Interrupt 4 Input or Output Compare and PWM Output B for
Timer/Counter3)
AIN1/OC3A (1) (Analog Comparator Negative Input or Output Compare and PWM
Output A for Timer/Counter3)
AIN0/XCK0(1) (Analog Comparator Positive Input or USART0 external clock
input/output)
PE1
PE0
PDO/TXD0 (Programming Data Output or UART0 Transmit Pin)
PDI/RXD0 (Programming Data Input or UART0 Receive Pin)
Note:
1. IC3, T3, OC3C, OC3B, OC3B, OC3A, and XCK0 not applicable in ATmega103 com-
patibility mode.
• INT7/IC3 – Port E, Bit 7
INT7, External Interrupt Source 7: The PE7 pin can serve as an External Interrupt
source.
IC3 – Input Capture Pin3: The PE7 pin can act as an Input Capture pin for
Timer/Counter3.
• INT6/T3 – Port E, Bit 6
INT6, External Interrupt Source 6: The PE6 pin can serve as an External Interrupt
source.
T3, Timer/Counter3 Counter Source.
• INT5/OC3C – Port E, Bit 5
INT5, External Interrupt Source 5: The PE5 pin can serve as an External Interrupt
source.
OC3C, Output Compare Match C output: The PE5 pin can serve as an external output
for the Timer/Counter3 Output Compare C. The pin has to be configured as an output
(DDE5 set – one) to serve this function. The OC3C pin is also the output pin for the
PWM mode timer function.
• INT4/OC3B – Port E, Bit 4
INT4, External Interrupt Source 4: The PE4 pin can serve as an External Interrupt
source.
OC3B, Output Compare Match B output: The PE4 pin can serve as an external output
for the Timer/Counter3 Output Compare B. The pin has to be configured as an output
(DDE4 set – one) to serve this function. The OC3B pin is also the output pin for the
PWM mode timer function.
79
2490G–AVR–03/04
• AIN1/OC3A – Port E, Bit 3
AIN1 – Analog Comparator Negative input. This pin is directly connected to the negative
input of the Analog Comparator.
OC3A, Output Compare Match A output: The PE3 pin can serve as an external output
for the Timer/Counter3 Output Compare A. The pin has to be configured as an output
(DDE3 set – one) to serve this function. The OC3A pin is also the output pin for the
PWM mode timer function.
• AIN0/XCK0 – Port E, Bit 2
AIN0 – Analog Comparator Positive input. This pin is directly connected to the positive
input of the Analog Comparator.
XCK0, USART0 External Clock. The Data Direction Register (DDE2) controls whether
the clock is output (DDE2 set) or input (DDE2 cleared). The XCK0 pin is active only
when the USART0 operates in synchronous mode.
• PDO/TXD0 – Port E, Bit 1
PDO, SPI Serial Programming Data output. During Serial Program Downloading, this
pin is used as data output line for the ATmega64.
TXD0, UART0 Transmit Pin.
• PDI/RXD0 – Port E, Bit 0
PDI, SPI Serial Programming Data input. During serial program downloading, this pin is
used as data input line for the ATmega64.
RXD0, USART0 Receive pin. Receive Data (Data Input pin for the USART0). When the
USART0 Receiver is enabled this pin is configured as an input regardless of the value of
DDRE0. When the USART0 forces this pin to be an input, a logical one in PORTE0 will
turn on the internal pull-up.
Table 40 and Table 41 relates the alternate functions of Port E to the overriding signals
shown in Figure 33 on page 69.
Table 40. Overriding Signals for Alternate Functions PE7..PE4
Signal
Name
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
PE7/INT7/IC3
PE6/INT6/T3
PE5/INT5/OC3C
PE4/INT4/OC3B
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
OC3C ENABLE
OC3C
OC3B ENABLE
OC3B
0
0
INT7 ENABLE
1
INT6 ENABLE
1
INT5 ENABLE
1
INT4 ENABLE
1
INT7 INPUT/IC3
INPUT
INT7 INPUT/T3
INPUT
INT5 INPUT
INT4 INPUT
AIO
–
–
–
–
80
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 41. Overriding Signals for Alternate Functions in PE3..PE0
Signal
Name
PE3/AIN1/OC3A
PE2/AIN0/XCK0
PE1/PDO/TXD0
PE0/PDI/RXD0
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
0
0
TXEN0
RXEN0
0
0
0
PORTE0 • PUD
0
0
TXEN0
RXEN0
0
0
1
0
OC3B ENABLE
UMSEL0
XCK0 OUTPUT
0
TXEN0
0
OC3B
TXD0
0
0
0
0
–
–
0
0
0
0
0
XCK0 INPUT
AIN0 INPUT
RXD0
–
AIO
AIN1 INPUT
Alternate Functions of Port F
The Port F has an alternate function as analog input for the ADC as shown in Table 42.
If some Port F pins are configured as outputs, it is essential that these do not switch
when a conversion is in progress. This might corrupt the result of the conversion. In
ATmega103 compatibility mode Port F is input only. If the JTAG interface is enabled, the
pull-up resistors on pins PF7(TDI), PF5(TMS) and PF4(TCK) will be activated even if a
reset occurs.
Table 42. Port F Pins Alternate Functions
Port Pin
PF7
Alternate Function
ADC7/TDI (ADC input channel 7 or JTAG Test Data Input)
ADC6/TDO (ADC input channel 6 or JTAG Test Data Output)
ADC5/TMS (ADC input channel 5 or JTAG Test mode Select)
ADC4/TCK (ADC input channel 4 or JTAG Test Clock)
ADC3 (ADC input channel 3)
PF6
PF5
PF4
PF3
PF2
ADC2 (ADC input channel 2)
PF1
ADC1 (ADC input channel 1)
PF0
ADC0 (ADC input channel 0)
• TDI, ADC7 – Port F, Bit 7
ADC7, Analog to Digital Converter, Channel 7.
TDI, JTAG Test Data In: Serial input data to be shifted in to the Instruction Register or
Data Register (scan chains). When the JTAG interface is enabled, this pin can not be
used as an I/O pin.
• TDO, ADC6 – Port F, Bit 6
ADC6, Analog to Digital Converter, Channel 6.
TDO, JTAG Test Data Out: Serial output data from Instruction Register or Data Regis-
ter. When the JTAG interface is enabled, this pin can not be used as an I/O pin.
The TDO pin is tri-stated unless TAP states that shift out data are entered.
81
2490G–AVR–03/04
• TMS, ADC5 – Port F, Bit 5
ADC5, Analog to Digital Converter, Channel 5.
TMS, JTAG Test mode Select: This pin is used for navigating through the TAP-controller
state machine. When the JTAG interface is enabled, this pin can not be used as an I/O
pin.
• TCK, ADC4 – Port F, Bit 4
ADC4, Analog to Digital Converter, Channel 4.
TCK, JTAG Test Clock: JTAG operation is synchronous to TCK. When the JTAG inter-
face is enabled, this pin can not be used as an I/O pin.
• ADC3 - ADC0 – Port F, Bit 3..0
Analog to Digital Converter, Channel 3..0.
Table 43. Overriding Signals for Alternate Functions in PF7..PF4
Signal
Name
PUOE
PUOV
DDOE
DDOV
PF7/ADC7/TDI
PF6/ADC6/TDO
JTAGEN
0
PF5/ADC5/TMS
PF4/ADC4/TCK
JTAGEN
JTAGEN
JTAGEN
1
1
1
JTAGEN
0
JTAGEN
JTAGEN
0
JTAGEN
0
SHIFT_IR +
SHIFT_DR
PVOE
PVOV
DIEOE
DIEOV
DI
0
JTAGEN
0
0
0
TDO
0
0
JTAGEN
JTAGEN
JTAGEN
JTAGEN
0
0
0
–
0
–
–
–
AIO
TDI/ADC7 INPUT
ADC6 INPUT
TMS/ADC5
INPUT
TCKADC4 INPUT
82
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 44. Overriding Signals for Alternate Functions in PF3..PF0
Signal
Name
PF3/ADC3
PF2/ADC2
PF1/ADC1
PF0/ADC0
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
–
–
–
–
AIO
ADC3 INPUT
ADC2 INPUT
ADC1 INPUT
ADC0 INPUT
Alternate Functions of Port G In ATmega103 compatibility mode, only the alternate functions are the defaults for Port
G, and Port G cannot be used as General Digital Port Pins. The alternate pin configura-
tion is as follows:
Table 45. Port G Pins Alternate Functions
Port Pin
PG4
Alternate Function
TOSC1 (RTC Oscillator Timer/Counter0)
TOSC2 (RTC Oscillator Timer/Counter0)
ALE (Address Latch Enable to external memory)
RD (Read strobe to external memory)
WR (Write strobe to external memory)
PG3
PG2
PG1
PG0
• TOSC1 – Port G, Bit 4
TOSC2, Timer Oscillator pin 1: When the AS0 bit in ASSR is set (one) to enable asyn-
chronous clocking of Timer/Counter0, pin PG4 is disconnected from the port, and
becomes the inverting output of the Oscillator amplifier. In this mode, a crystal Oscillator
is connected to this pin, and the pin can not be used as an I/O pin.
• TOSC2 – Port G, Bit 3
TOSC2, Timer Oscillator pin 2: When the AS0 bit in ASSR is set (one) to enable asyn-
chronous clocking of Timer/Counter0, pin PG3 is disconnected from the port, and
becomes the input of the inverting Oscillator amplifier. In this mode, a crystal Oscillator
is connected to this pin, and the pin cannot be used as an I/O pin.
• ALE – Port G, Bit 2
ALE is the external data memory Address Latch Enable signal.
• RD – Port G, Bit 1
RD is the external data memory read control strobe.
83
2490G–AVR–03/04
• WR – Port G, Bit 0
WR is the external data memory write control strobe.
Table 46 and Table 47 relates the alternate functions of Port G to the overriding signals
shown in Figure 33 on page 69.
Table 46. Overriding Signals for Alternate Functions in PG4..PG1
Signal Name
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
PG4/TOSC1
PG3/TOSC2
PG2/ALE
PG1/RD
AS0
AS0
SRE
0
SRE
0
0
0
AS0
AS0
SRE
1
SRE
1
0
0
0
0
SRE
ALE
0
SRE
RD
0
0
0
AS0
AS0
0
0
0
0
–
–
–
–
AIO
T/C0 OSC INPUT
T/C0 OSC OUTPUT
–
–
Table 47. Overriding Signals for Alternate Functions in PG0
Signal Name
PUOE
PUOV
DDOE
DDOV
PVOE
PVOV
DIEOE
DIEOV
DI
PG0/WR
SRE
0
SRE
1
SRE
WR
0
0
–
AIO
–
84
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Register Description for
I/O Ports
Port A Data Register – PORTA
Bit
7
PORTA7
R/W
0
6
PORTA6
R/W
0
5
PORTA5
R/W
0
4
PORTA4
R/W
0
3
PORTA3
R/W
0
2
PORTA2
R/W
0
1
PORTA1
R/W
0
0
PORTA0
R/W
0
PORTA
DDRA
PINA
Read/Write
Initial Value
Port A Data Direction Register
– DDRA
Bit
7
DDA7
R/W
0
6
DDA6
R/W
0
5
DDA5
R/W
0
4
DDA4
R/W
0
3
DDA3
R/W
0
2
DDA2
R/W
0
1
DDA1
R/W
0
0
DDA0
R/W
0
Read/Write
Initial Value
Port A Input Pins Address –
PINA
Bit
7
PINA7
R
6
PINA6
R
5
PINA5
R
4
PINA4
R
3
PINA3
R
2
PINA2
R
1
PINA1
R
0
PINA0
R
Read/Write
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Port B Data Register – PORTB
Bit
7
PORTB7
R/W
0
6
PORTB6
R/W
0
5
PORTB5
R/W
0
4
PORTB4
R/W
0
3
PORTB3
R/W
0
2
PORTB2
R/W
0
1
PORTB1
R/W
0
0
PORTB0
R/W
0
PORTB
DDRB
PINB
Read/Write
Initial Value
Port B Data Direction Register
– DDRB
Bit
7
DDB7
R/W
0
6
DDB6
R/W
0
5
DDB5
R/W
0
4
DDB4
R/W
0
3
DDB3
R/W
0
2
DDB2
R/W
0
1
DDB1
R/W
0
0
DDB0
R/W
0
Read/Write
Initial Value
Port B Input Pins Address –
PINB
Bit
7
PINB7
R
6
PINB6
R
5
PINB5
R
4
PINB4
R
3
PINB3
R
2
PINB2
R
1
PINB1
R
0
PINB0
R
Read/Write
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Port C Data Register – PORTC
Bit
7
PORTC7
R/W
0
6
PORTC6
R/W
0
5
PORTC5
R/W
0
4
PORTC4
R/W
0
3
PORTC3
R/W
0
2
PORTC2
R/W
0
1
PORTC1
R/W
0
0
PORTC0
R/W
0
PORTC
DDRC
Read/Write
Initial Value
Port C Data Direction Register
– DDRC
Bit
7
DDC7
R/W
0
6
DDC6
R/W
0
5
DDC5
R/W
0
4
DDC4
R/W
0
3
DDC3
R/W
0
2
DDC2
R/W
0
1
DDC1
R/W
0
0
DDC0
R/W
0
Read/Write
Initial Value
85
2490G–AVR–03/04
Port C Input Pins Address –
PINC
Bit
7
PINC7
R
6
PINC6
R
5
PINC5
R
4
PINC4
R
3
PINC3
R
2
PINC2
R
1
PINC1
R
0
PINC0
R
PINC
Read/Write
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
In ATmega103 compatibility mode, DDRC and PINC Registers are initialized to being
Push-pull Zero Output. The port pins assumes their Initial Value, even if the clock is not
running. Note that the DDRC and PINC registers are available in ATmega103 compati-
bility mode, and should not be used for 100% backward compatibility.
Port D Data Register – PORTD
Bit
7
PORTD7
R/W
0
6
PORTD6
R/W
0
5
PORTD5
R/W
0
4
PORTD4
R/W
0
3
PORTD3
R/W
0
2
PORTD2
R/W
0
1
PORTD1
R/W
0
0
PORTD0
R/W
0
PORTD
DDRD
PIND
Read/Write
Initial Value
Port D Data Direction Register
– DDRD
Bit
7
DDD7
R/W
0
6
DDD6
R/W
0
5
DDD5
R/W
0
4
DDD4
R/W
0
3
DDD3
R/W
0
2
DDD2
R/W
0
1
DDD1
R/W
0
0
DDD0
R/W
0
Read/Write
Initial Value
Port D Input Pins Address –
PIND
Bit
7
PIND7
R
6
PIND6
R
5
PIND5
R
4
PIND4
R
3
PIND3
R
2
PIND2
R
1
PIND1
R
0
PIND0
R
Read/Write
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Port E Data Register – PORTE
Bit
7
PORTE7
R/W
0
6
PORTE6
R/W
0
5
PORTE5
R/W
0
4
PORTE4
R/W
0
3
PORTE3
R/W
0
2
PORTE2
R/W
0
1
PORTE1
R/W
0
0
PORTE0
R/W
0
PORTE
DDRE
PINF
Read/Write
Initial Value
Port E Data Direction Register
– DDRE
Bit
7
DDE7
R/W
0
6
DDE6
R/W
0
5
DDE5
R/W
0
4
DDE4
R/W
0
3
DDE3
R/W
0
2
DDE2
R/W
0
1
DDE1
R/W
0
0
DDE0
R/W
0
Read/Write
Initial Value
Port E Input Pins Address –
PINE
Bit
7
PINE7
R
6
PINE6
R
5
PINE5
R
4
PINE4
R
3
PINE3
R
2
PINE2
R
1
PINE1
R
0
PINE0
R
Read/Write
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Port F Data Register – PORTF
Bit
7
PORTF7
R/W
0
6
PORTF6
R/W
0
5
PORTF5
R/W
0
4
PORTF4
R/W
0
3
PORTF3
R/W
0
2
PORTF2
R/W
0
1
PORTF1
R/W
0
0
PORTF0
R/W
0
PORTF
Read/Write
Initial Value
86
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Port F Data Direction Register
– DDRF
Bit
7
DDF7
R/W
0
6
DDF6
R/W
0
5
DDF5
R/W
0
4
DDF4
R/W
0
3
DDF3
R/W
0
2
DDF2
R/W
0
1
DDF1
R/W
0
0
DDF0
R/W
0
DDRF
Read/Write
Initial Value
Port F Input Pins Address –
PINF
Bit
7
PINF7
R
6
PINF6
R
5
PINF5
R
4
PINF4
R
3
PINF3
R
2
PINF2
R
1
PINF1
R
0
PINF0
R
PINF
Read/Write
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Note that PORTF and DDRF Registers are not available in ATmega103 compatibility
mode where Port F serves as digital input only.
Port G Data Register – PORTG
Bit
7
–
6
–
5
–
4
PORTG4
R/W
0
3
PORTG3
R/W
0
2
PORTG2
R/W
0
1
PORTG1
R/W
0
0
PORTG0
R/W
0
PORTG
DDRG
PING
Read/Write
Initial Value
R
0
R
0
R
0
Port G Data Direction Register
– DDRG
Bit
7
–
R
0
6
–
R
0
5
–
R
0
4
DDG4
R/W
0
3
DDG3
R/W
0
2
DDG2
R/W
0
1
DDG1
R/W
0
0
DDG0
R/W
0
Read/Write
Initial Value
Port G Input Pins Address –
PING
Bit
7
–
R
0
6
–
R
0
5
–
R
0
4
PING4
R
3
PING3
R
2
PING2
R
1
PING1
R
0
PING0
R
Read/Write
Initial Value
N/A
N/A
N/A
N/A
N/A
Note that PORTG, DDRG, and PING are not available in ATmega103 compatibility
mode. In the ATmega103 compatibility mode Port G serves its alternate functions only
(TOSC1, TOSC2, WR, RD and ALE).
87
2490G–AVR–03/04
External Interrupts
The External Interrupts are triggered by the INT7:0 pins. Observe that, if enabled, the
interrupts will trigger even if the INT7:0 pins are configured as outputs. This feature pro-
vides a way of generating a software interrupt. The external interrupts can be triggered
by a falling or rising edge or a low level. This is set up as indicated in the specification for
the External Interrupt Control Registers – EICRA (INT3:0) and EICRB (INT7:4). When
the External Interrupt is enabled and is configured as level triggered, the interrupt will
trigger as long as the pin is held low. Note that recognition of falling or rising edge inter-
rupts on INT7:4 requires the presence of an I/O clock, described in “Clock Systems and
their Distribution” on page 35. Low level interrupts and the edge interrupt on INT3:0 are
detected asynchronously. This implies that these interrupts can be used for waking the
part also from sleep modes other than Idle mode. The I/O clock is halted in all sleep
modes except Idle mode.
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. The period of the Watchdog Oscillator is 1 µs (nominal) at 5.0V and 25°C. The
frequency of the Watchdog Oscillator is voltage dependent as shown in the “Electrical
Characteristics” on page 326. The MCU will wake up if the input has the required level
during this sampling or if it is held until the end of the start-up time. The start-up time is
defined by the SUT Fuses as described in “Clock Systems and their Distribution” on
page 35. If the level is sampled twice by the Watchdog Oscillator clock but disappears
before the end of the start-up time, the MCU will still wake up, but no interrupt will be
generated. The required level must be held long enough for the MCU to complete the
wake up to trigger the level interrupt.
External Interrupt Control
Register A – EICRA
Bit
7
ISC31
R/W
0
6
ISC30
R/W
0
5
ISC21
R/W
0
4
ISC20
R/W
0
3
ISC11
R/W
0
2
ISC10
R/W
0
1
ISC01
R/W
0
0
ISC00
R/W
0
EICRA
Read/Write
Initial Value
This Register can not be reached in ATmega103 compatibility mode, but the Initial
Value defines INT3:0 as low level interrupts, as in ATmega103.
• Bits 7..0 – ISC31, ISC30 - ISC00, ISC00: External Interrupt 3 - 0 Sense Control
Bits
The External Interrupts 3 - 0 are activated by the external pins INT3:0 if the SREG I-flag
and the corresponding interrupt mask in the EIMSK is set. The level and edges on the
external pins that activate the interrupts are defined in Table 48. Edges on INT3..INT0
are registered asynchronously. Pulses on INT3:0 pins wider than the minimum pulse
width given in Table 49 will generate an interrupt. Shorter pulses are not guaranteed to
generate an interrupt. 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. When changing the ISCn bit, an interrupt can occur. Therefore, it is recommended
to first disable INTn by clearing its Interrupt Enable bit in the EIMSK Register. Then, the
ISCn bit can be changed. Finally, the INTn interrupt flag should be cleared by writing a
logical one to its Interrupt Flag bit (INTFn) in the EIFR Register before the interrupt is re-
enabled.
88
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 48. Interrupt Sense Control(1)
ISCn1
ISCn0 Description
0
0
1
1
0
1
0
1
The low level of INTn generates an interrupt request.
Reserved
The falling edge of INTn generates asynchronously an interrupt request.
The rising edge of INTn generates asynchronously an interrupt request.
Note:
1. n = 3, 2, 1or 0.
When changing the ISCn1/ISCn0 bits, the interrupt must be disabled by clearing its
Interrupt Enable bit in the EIMSK Register. Otherwise an interrupt can occur when
the bits are changed.
Table 49. Asynchronous External Interrupt Characteristics
Symbol Parameter Condition Min
tINT Minimum pulse width for
Typ
Max
Units
50
ns
asynchronous External Interrupt
External Interrupt Control
Register B – EICRB
Bit
7
ISC71
R/W
0
6
ISC70
R/W
0
5
ISC61
R/W
0
4
ISC60
R/W
0
3
ISC51
R/W
0
2
ISC50
R/W
0
1
0
ISC41
R/W
0
ISC40
R/W
0
EICRB
Read/Write
Initial Value
• Bits 7..0 – ISC71, ISC70 - ISC41, ISC40: External Interrupt 7 - 4 Sense Control
Bits
The External Interrupts 7 - 4 are activated by the external pins INT7:4 if the SREG I-flag
and the corresponding interrupt mask in the EIMSK is set. The level and edges on the
external pins that activate the interrupts are defined in Table 50. The value on the
INT7:4 pins are sampled before detecting edges. If edge or toggle interrupt is selected,
pulses that last longer than one 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.
Table 50. Interrupt Sense Control(1)
ISCn1 ISCn0 Description
0
0
1
0
1
0
The low level of INTn generates an interrupt request.
Any logical change on INTn generates an interrupt request
The falling edge between two samples of INTn generates an interrupt
request.
1
1
The rising edge between two samples of INTn generates an interrupt
request.
Note:
1. n = 7, 6, 5 or 4.
When changing the ISCn1/ISCn0 bits, the interrupt must be disabled by clearing its
Interrupt Enable bit in the EIMSK Register. Otherwise an interrupt can occur when
the bits are changed.
89
2490G–AVR–03/04
External Interrupt Mask
Register – EIMSK
Bit
7
6
5
4
3
2
1
0
INT7
R/W
0
INT6
R/W
0
INT5
R/W
0
INT4
R/W
0
INT3
R/W
0
INT2
R/W
0
INT1
R/W
0
INT0
R/W
0
EIMSK
Read/Write
Initial Value
• Bits 7..4 – INT7 - INT0: External Interrupt Request 7 - 0 Enable
When an INT7 - INT4 bit is written to one and the I-bit in the Status Register (SREG) is
set (one), the corresponding external pin interrupt is enabled. The Interrupt Sense Con-
trol bits in the External Interrupt Control Registers – EICRA and EICRB defines whether
the External Interrupt is activated on rising or falling edge or 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.
External Interrupt Flag
Register – EIFR
Bit
7
INTF7
R/W
0
6
INTF6
R/W
0
5
INTF5
R/W
0
4
INTF4
R/W
0
3
INTF3
R/W
0
2
INTF2
R/W
0
1
INTF1
R/W
0
0
INTF0
R/W
0
EIFR
Read/Write
Initial Value
• Bits 7..0 – INTF7 - INTF0: External Interrupt Flags 7 - 0
When an edge or logic change on the INT7:0 pin triggers an interrupt request, INTF7:0
becomes set (one). If the I-bit in SREG and the corresponding Interrupt Enable bit,
INT7:0 in EIMSK, are set (one), the MCU will jump to the Interrupt Vector. The flag is
cleared when the interrupt routine is executed. Alternatively, the flag can be cleared by
writing a logical one to it. These flags are always cleared when INT7:0 are configured as
level interrupt. Note that when entering sleep mode with the INT3:0 interrupts disabled,
the input buffers on these pins will be disabled. This may cause a logic change in inter-
nal signals which will set the INTF3:0 flags. See “Digital Input Enable and Sleep Modes”
on page 68 for more information.
90
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
8-bit Timer/Counter0 Timer/Counter0 is a general purpose, single-channel, 8-bit Timer/Counter module. The
main features are:
• Single Channel Counter
• Clear Timer on Compare Match (Auto Reload)
• Glitch-free, Phase Correct Pulse Width Modulator (PWM)
with PWM and
Asynchronous
Operation
• Frequency Generator
• 10-bit Clock Prescaler
• Overflow and Compare Match Interrupt Sources (TOV0 and OCF0)
• Allows Clocking from External 32 kHz Watch Crystal Independent of the I/O Clock
Overview
A simplified block diagram of the 8-bit Timer/Counter is shown in Figure 34. For the
actual placement of I/O pins, refer to “Pin Configuration” on page 2. CPU accessible I/O
Registers, including I/O bits and I/O pins, are shown in bold. The device-specific I/O
Register and bit locations are listed in the “8-bit Timer/Counter Register Description” on
page 102.
Figure 34. 8-bit Timer/Counter Block Diagram
TCCRn
count
TOVn
(Int. Req.)
clear
Control Logic
TOP
direction
clkTn
TOSC1
TOSC2
BOTTOM
T/C
Oscillator
Prescaler
Timer/Counter
TCNTn
= 0
= 0xFF
clkI/O
OCn
OCn
(Int. Req.)
Waveform
Generation
=
OCRn
clkI/O
Synchronized Status Flags
Synchronization Unit
clkASY
Status Flags
ASSRn
Asynchronous Mode
Select (ASn)
Registers
The Timer/Counter (TCNT0) and Output Compare Register (OCR0) are 8-bit registers.
Interrupt request (shorten as Int.Req.) signals are all visible in the Timer Interrupt Flag
Register (TIFR). All interrupts are individually masked with the Timer Interrupt Mask
Register (TIMSK). TIFR and TIMSK are not shown in the figure since these registers are
shared by other timer units.
The Timer/Counter can be clocked internally, via the prescaler, or asynchronously
clocked from the TOSC1/2 pins, as detailed later in this section. The asynchronous
91
2490G–AVR–03/04
operation is controlled by the Asynchronous Status Register (ASSR). The Clock Select
logic block controls which clock source the Timer/Counter uses to increment (or decre-
ment) its value. The Timer/Counter is inactive when no clock source is selected. The
output from the Clock Select logic is referred to as the timer clock (clkT0).
The double buffered Output Compare Register (OCR0) is compared with the
Timer/Counter value at all times. The result of the compare can be used by the Wave-
form Generator to generate a PWM or variable frequency output on the Output Compare
pin (OC0). See “Output Compare Unit” on page 93. for details. The Compare Match
event will also set the Compare Flag (OCF0) which can be used to generate an Output
Compare interrupt request.
Definitions
Many register and bit references in this datasheet are written in general form. A lower
case “n” replaces the Timer/Counter number, in this case 0. However, when using the
register or bit defines in a program, the precise form must be used i.e. TCNT0 for
accessing Timer/Counter0 counter value and so on.
The definitions in Table 51 are also used extensively throughout this section.
Table 51. Definitions
BOTTOM The counter reaches the BOTTOM when it becomes zero (0x00).
MAX
TOP
The counter reaches its MAXimum when it becomes 0xFF (decimal 255).
The counter reaches the TOP when it becomes equal to the highest
value in the count sequence. The TOP value can be assigned to be the
fixed value 0xFF (MAX) or the value stored in the OCR0 Register. The
assignment is dependent on the mode of operation.
Timer/Counter Clock
Sources
The Timer/Counter can be clocked by an internal synchronous or an external asynchro-
nous clock source. The clock source clkT0 is by default equal to the MCU clock, clkI/O.
When the AS0 bit in the ASSR Register is written to logic one, the clock source is taken
from the Timer/Counter Oscillator connected to TOSC1 and TOSC2. For details on
asynchronous operation, see “Asynchronous Status Register – ASSR” on page 105. For
details on clock sources and prescaler, see “Timer/Counter Prescaler” on page 108.
92
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Counter Unit
The main part of the 8-bit Timer/Counter is the programmable bi-directional counter unit.
Figure 35 shows a block diagram of the counter and its surrounding environment.
Figure 35. Counter Unit Block Diagram
TOVn
(Int.Req.)
DATA BUS
TOSC1
count
T/C
Oscillator
clk Tn
clear
TCNTn
Control Logic
Prescaler
direction
TOSC2
clk
bottom
top
I/O
Signal description (internal signals):
count Increment or decrement TCNT0 by 1.
direction Selects between increment and decrement.
clear
clkT0
top
Clear TCNT0 (set all bits to zero).
Timer/Counter clock.
Signalizes that TCNT0 has reached maximum value.
bottom
Signalizes that TCNT0 has reached minimum value (zero).
Depending on the mode of operation used, the counter is cleared, incremented, or dec-
remented at each timer clock (clkT0). clkT0 can be generated from an external or internal
clock source, selected by the Clock Select bits (CS02:0). When no clock source is
selected (CS02:0 = 0) the timer is stopped. However, the TCNT0 value can be accessed
by the CPU, regardless of whether clkT0 is present or not. A CPU write overrides (has
priority over) all counter clear or count operations.
The counting sequence is determined by the setting of the WGM01 and WGM00 bits
located in the Timer/Counter Control Register (TCCR0). There are close connections
between how the counter behaves (counts) and how waveforms are generated on the
Output Compare output OC0. For more details about advanced counting sequences
and waveform generation, see “Modes of Operation” on page 96.
The Timer/Counter Overflow Flag (TOV0) is set according to the mode of operation
selected by the WGM01:0 bits. TOV0 can be used for generating a CPU interrupt.
Output Compare Unit
The 8-bit comparator continuously compares TCNT0 with the Output Compare Register
(OCR0). Whenever TCNT0 equals OCR0, the comparator signals a match. A match will
set the Output Compare Flag (OCF0) at the next timer clock cycle. If enabled (OCIE0 =
1), the Output Compare Flag generates an Output Compare interrupt. The OCF0 flag is
automatically cleared when the interrupt is executed. Alternatively, the OCF0 flag can
be cleared by software by writing a logical one to its I/O bit location. The Waveform Gen-
erator uses the match signal to generate an output according to operating mode set by
the WGM01:0 bits and Compare Output mode (COM01:0) bits. The max and bottom sig-
nals are used by the Waveform Generator for handling the special cases of the extreme
values in some modes of operation (“Modes of Operation” on page 96). Figure 36 shows
a block diagram of the Output Compare unit.
93
2490G–AVR–03/04
Figure 36. Output Compare Unit, Block Diagram
DATA BUS
OCRn
TCNTn
= (8-bit Comparator )
OCFn (Int.Req.)
top
bottom
Waveform Generator
OCxy
FOCn
WGMn1:0
COMn1:0
The OCR0 Register is double buffered when using any of the Pulse Width Modulation
(PWM) modes. For the normal and Clear Timer on Compare (CTC) modes of operation,
the double buffering is disabled. The double buffering synchronizes the update of the
OCR0 Compare Register to either top or bottom of the counting sequence. The synchro-
nization prevents the occurrence of odd-length, non-symmetrical PWM pulses, thereby
making the output glitch-free.
The OCR0 Register access may seem complex, but this is not case. When the double
buffering is enabled, the CPU has access to the OCR0 Buffer Register, and if double
buffering is disabled the CPU will access the OCR0 directly.
Force Output Compare
In non-PWM waveform generation modes, the match output of the comparator can be
forced by writing a one to the Force Output Compare (FOC0) bit. Forcing Compare
Match will not set the OCF0 flag or reload/clear the timer, but the OC0 pin will be
updated as if a real Compare Match had occurred (the COM01:0 bits settings define
whether the OC0 pin is set, cleared or toggled).
Compare Match Blocking by
TCNT0 Write
All CPU write operations to the TCNT0 Register will block any Compare Match that
occurs in the next timer clock cycle, even when the timer is stopped. This feature allows
OCR0 to be initialized to the same value as TCNT0 without triggering an interrupt when
the Timer/Counter clock is enabled.
Using the Output Compare
Unit
Since writing TCNT0 in any mode of operation will block all Compare Matches for one
timer clock cycle, there are risks involved when changing TCNT0 when using the Output
Compare channel, independently of whether the Timer/Counter is running or not. If the
value written to TCNT0 equals the OCR0 value, the Compare Match will be missed,
resulting in incorrect waveform generation. Similarly, do not write the TCNT0 value
equal to BOTTOM when the counter is downcounting.
The setup of the OC0 should be performed before setting the Data Direction Register for
the port pin to output. The easiest way of setting the OC0 value is to use the Force Out-
put Compare (FOC0) strobe bit in Normal mode. The OC0 Register keeps its value even
when changing between waveform generation modes.
Be aware that the COM01:0 bits are not double buffered together with the compare
value. Changing the COM01:0 bits will take effect immediately.
94
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Compare Match Output
Unit
The Compare Output mode (COM01:0) bits have two functions. The Waveform Genera-
tor uses the COM01:0 bits for defining the Output Compare (OC0) state at the next
Compare Match. Also, the COM01:0 bits control the OC0 pin output source. Figure 37
shows a simplified schematic of the logic affected by the COM01:0 bit setting. The I/O
Registers, I/O bits, and I/O pins in the figure are shown in bold. Only the parts of the
general I/O port control registers (DDR and PORT) that are affected by the COM01:0
bits are shown. When referring to the OC0 state, the reference is for the internal OC0
Register, not the OC0 pin.
Figure 37. Compare Match Output Unit, Schematic
COMn1
Waveform
Generator
COMn0
FOCn
D
Q
Q
1
0
OCn
Pin
OCn
D
PORT
D
Q
DDR
clkI/O
The general I/O port function is overridden by the Output Compare (OC0) from the
Waveform Generator if either of the COM01:0 bits are set. However, the OC0 pin direc-
tion (input or output) is still controlled by the Data Direction Register (DDR) for the port
pin. The Data Direction Register bit for the OC0 pin (DDR_OC0) must be set as output
before the OC0 value is visible on the pin. The port override function is independent of
the Waveform Generation mode.
The design of the Output Compare pin logic allows initialization of the OC0 state before
the output is enabled. Note that some COM01:0 bit settings are reserved for certain
modes of operation. See “8-bit Timer/Counter Register Description” on page 102.
Compare Output Mode and
Waveform Generation
The Waveform Generator uses the COM01:0 bits differently in Normal, CTC, and PWM
modes. For all modes, setting the COM01:0 = 0 tells the Waveform Generator that no
action on the OC0 Register is to be performed on the next Compare Match. For com-
pare output actions in the non-PWM modes refer to Table 53 on page 103. For fast
PWM mode, refer to Table 54 on page 103, and for phase correct PWM refer to Table
55 on page 103.
95
2490G–AVR–03/04
A change of the COM01:0 bits state will have effect at the first Compare Match after the
bits are written. For non-PWM modes, the action can be forced to have immediate effect
by using the FOC0 strobe bits.
Modes of Operation
The mode of operation, i.e., the behavior of the Timer/Counter and the Output Compare
pins, is defined by the combination of the Waveform Generation mode (WGM01:0) and
Compare Output mode (COM01:0) bits. The Compare Output mode bits do not affect
the counting sequence, while the Waveform Generation mode bits do. The COM01:0
bits control whether the PWM output generated should be inverted or not (inverted or
non-inverted PWM). For non-PWM modes the COM01:0 bits control whether the output
should be set, cleared, or toggled at a Compare Match (See “Compare Match Output
Unit” on page 95.).
For detailed timing information refer to “Timer/Counter Timing Diagrams” on page 100.
Normal Mode
The simplest mode of operation is the Normal mode (WGM01:0 = 0). In this mode the
counting direction is always up (incrementing), and no counter clear is performed. The
counter simply overruns when it passes its maximum 8-bit value (TOP = 0xFF) and then
restarts from the bottom (0x00). In normal operation the Timer/Counter Overflow Flag
(TOV0) will be set in the same timer clock cycle as the TCNT0 becomes zero. The
TOV0 flag in this case behaves like a ninth bit, except that it is only set, not cleared.
However, combined with the timer overflow interrupt that automatically clears the TOV0
flag, the timer resolution can be increased by software. There are no special cases to
consider in the Normal mode, a new counter value can be written anytime.
The Output Compare unit can be used to generate interrupts at some given time. Using
the Output Compare to generate waveforms in Normal mode is not recommended,
since this will occupy too much of the CPU time.
Clear Timer on Compare
Match (CTC) Mode
In Clear Timer on Compare or CTC mode (WGM01:0 = 2), the OCR0 Register is used to
manipulate the counter resolution. In CTC mode the counter is cleared to zero when the
counter value (TCNT0) matches the OCR0. The OCR0 defines the top value for the
counter, hence also its resolution. This mode allows greater control of the Compare
Match output frequency. It also simplifies the operation of counting external events.
The timing diagram for the CTC mode is shown in Figure 38. The counter value
(TCNT0) increases until a Compare Match occurs between TCNT0 and OCR0, and then
counter (TCNT0) is cleared.
Figure 38. CTC Mode, Timing Diagram
OCn Interrupt Flag Set
TCNTn
OCn
(Toggle)
(COMn1:0 = 1)
1
2
3
4
Period
96
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
An interrupt can be generated each time the counter value reaches the TOP value by
using the OCF0 flag. If the interrupt is enabled, the interrupt handler routine can be used
for updating the TOP value. However, changing the TOP to a value close to BOTTOM
when the counter is running with none or a low prescaler value must be done with care
since the CTC mode does not have the double buffering feature. If the new value written
to OCR0 is lower than the current value of TCNT0, the counter will miss the Compare
Match. The counter will then have to count to its maximum value (0xFF) and wrap
around starting at 0x00 before the Compare Match can occur.
For generating a waveform output in CTC mode, the OC0 output can be set to toggle its
logical level on each Compare Match by setting the Compare Output mode bits to toggle
mode (COM01:0 = 1). The OC0 value will not be visible on the port pin unless the data
direction for the pin is set to output. The waveform generated will have a maximum fre-
quency of fOC0 = fclk_I/O/2 when OCR0 is set to zero (0x00). The waveform frequency is
defined by the following equation:
f
clk_I/O
f
= ----------------------------------------------
OCn
2 ⋅ N ⋅ (1 + OCRn)
The N variable represents the prescale factor (1, 8, 32, 64, 128, 256, or 1024).
As for the Normal mode of operation, the TOV0 flag is set in the same timer clock cycle
that the counter counts from MAX to 0x00.
Fast PWM Mode
The fast Pulse Width Modulation or fast PWM mode (WGM01:0 = 3) provides a high fre-
quency PWM waveform generation option. The fast PWM differs from the other PWM
option by its single-slope operation. The counter counts from BOTTOM to MAX then
restarts from BOTTOM. In non-inverting Compare Output mode, the Output Compare
(OC0) is cleared on the Compare Match between TCNT0 and OCR0, and set at BOT-
TOM. In inverting Compare Output mode, the output is set on Compare Match and
cleared at BOTTOM. Due to the single-slope operation, the operating frequency of the
fast PWM mode can be twice as high as the phase correct PWM mode that uses dual-
slope operation. This high frequency makes the fast PWM mode well suited for power
regulation, rectification, and DAC applications. High frequency allows physically small
sized external components (coils, capacitors), and therefore reduces total system cost.
In fast PWM mode, the counter is incremented until the counter value matches the MAX
value. The counter is then cleared at the following timer clock cycle. The timing diagram
for the fast PWM mode is shown in Figure 39. The TCNT0 value is in the timing diagram
shown as a histogram for illustrating the single-slope operation. The diagram includes
non-inverted and inverted PWM outputs. The small horizontal line marks on the TCNT0
slopes represent Compare Matches between OCR0 and TCNT0.
97
2490G–AVR–03/04
Figure 39. Fast PWM Mode, Timing Diagram
OCRn Interrupt Flag Set
OCRn Update and
TOVn Interrupt Flag Set
TCNTn
(COMn1:0 = 2)
(COMn1:0 = 3)
OCn
OCn
1
2
3
4
5
6
7
Period
The Timer/Counter Overflow Flag (TOV0) is set each time the counter reaches MAX. If
the interrupt is enabled, the interrupt handler routine can be used for updating the com-
pare value.
In fast PWM mode, the compare unit allows generation of PWM waveforms on the OC0
pin. Setting the COM01:0 bits to two will produce a non-inverted PWM and an inverted
PWM output can be generated by setting the COM01:0 to three (See Table 54 on page
103). The actual OC0 value will only be visible on the port pin if the data direction for the
port pin is set as output. The PWM waveform is generated by setting (or clearing) the
OC0 Register at the Compare Match between OCR0 and TCNT0, and clearing (or set-
ting) the OC0 Register at the timer clock cycle the counter is cleared (changes from
MAX to BOTTOM).
The PWM frequency for the output can be calculated by the following equation:
f
clk_I/O
f
= -----------------
OCnPWM
N ⋅ 256
The N variable represents the prescale factor (1, 8, 32, 64, 128, 256, or 1024).
The extreme values for the OCR0 Register represent special cases when generating a
PWM waveform output in the fast PWM mode. If the OCR0 is set equal to BOTTOM, the
output will be a narrow spike for each MAX+1 timer clock cycle. Setting the OCR0 equal
to MAX will result in a constantly high or low output (depending on the polarity of the out-
put set by the COM01:0 bits.)
A frequency (with 50% duty cycle) waveform output in fast PWM mode can be achieved
by setting OC0 to toggle its logical level on each Compare Match (COM01:0 = 1). The
waveform generated will have a maximum frequency of foc0 = fclk_I/O/2 when OCR0 is set
to zero. This feature is similar to the OC0 toggle in CTC mode, except the double buffer
feature of the Output Compare unit is enabled in the fast PWM mode.
98
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Phase Correct PWM Mode
The phase correct PWM mode (WGM01:0 = 1) provides a high resolution phase correct
PWM waveform generation option. The phase correct PWM mode is based on a dual-
slope operation. The counter counts repeatedly from BOTTOM to MAX and then from
MAX to BOTTOM. In non-inverting Compare Output mode, the Output Compare (OC0)
is cleared on the Compare Match between TCNT0 and OCR0 while upcounting, and set
on the Compare Match while downcounting. In inverting Output Compare mode, the
operation is inverted. The dual-slope operation has lower maximum operation frequency
than single slope operation. However, due to the symmetric feature of the dual-slope
PWM modes, these modes are preferred for motor control applications.
The PWM resolution for the phase correct PWM mode is fixed to eight bits. In phase
correct PWM mode the counter is incremented until the counter value matches MAX.
When the counter reaches MAX, it changes the count direction. The TCNT0 value will
be equal to MAX for one timer clock cycle. The timing diagram for the phase correct
PWM mode is shown on Figure 40. The TCNT0 value is in the timing diagram shown as
a histogram for illustrating the dual-slope operation. The diagram includes non-inverted
and inverted PWM outputs. The small horizontal line marks on the TCNT0 slopes repre-
sent Compare Matches between OCR0 and TCNT0.
Figure 40. Phase Correct PWM Mode, Timing Diagram
OCn Interrupt
Flag Set
OCRn Update
TOVn Interrupt
Flag Set
TCNTn
(COMn1:0 = 2)
OCn
(COMn1:0 = 3)
OCn
1
2
3
Period
The Timer/Counter Overflow Flag (TOV0) is set each time the counter reaches BOT-
TOM. The interrupt flag can be used to generate an interrupt each time the counter
reaches the BOTTOM value.
In phase correct PWM mode, the compare unit allows generation of PWM waveforms on
the OC0 pin. Setting the COM01:0 bits to two will produce a non-inverted PWM. An
inverted PWM output can be generated by setting the COM01:0 to three (See Table 55
on page 103). The actual OC0 value will only be visible on the port pin if the data direc-
tion for the port pin is set as output. The PWM waveform is generated by clearing (or
setting) the OC0 Register at the Compare Match between OCR0 and TCNT0 when the
counter increments, and setting (or clearing) the OC0 Register at Compare Match
99
2490G–AVR–03/04
between OCR0 and TCNT0 when the counter decrements. The PWM frequency for the
output when using phase correct PWM can be calculated by the following equation:
f
clk_I/O
f
= -----------------
OCnPCPWM
N ⋅ 510
The N variable represents the prescale factor (1, 8, 32, 64, 128, 256, or 1024).
The extreme values for the OCR0 Register represent special cases when generating a
PWM waveform output in the phase correct PWM mode. If the OCR0 is set equal to
BOTTOM, the output will be continuously low and if set equal to MAX the output will be
continuously high for non-inverted PWM mode. For inverted PWM the output will have
the opposite logic values.
At the very start of period 2 in Figure 40 OCn has a transition from high to low even
though there is no Compare Match. The point of this transition is to guarantee symmetry
around BOTTOM. There are two cases that give a transition without Compare Match.
•
OCR0 changes its value from MAX, like in Figure 40. When the OCR0 value is MAX
the OCn pin value is the same as the result of a down-counting Compare Match. To
ensure symmetry around BOTTOM the OCn value at MAX must correspond to the
result of an up-counting Compare Match.
•
The timer starts counting from a higher value than the one in OCR0, and for that
reason misses the Compare Match and hence the OCn change that would have
happened on the way up.
Timer/Counter Timing
Diagrams
Figure 41 and Figure 42 contain timing data for the Timer/Counter operation. The
Timer/Counter is a synchronous design and the timer clock (clkT0) is therefore shown as
a clock enable signal. The figure shows the count sequence close to the MAX value.
Figure 43 and Figure 44 show the same timing data, but with the prescaler enabled. The
figures illustrate when interrupt flags are set.
The following figures show the Timer/Counter in synchronous mode, and the timer clock
(clkT0) is therefore shown as a clock enable signal. In asynchronous mode, clkI/O should
be replaced by the Timer/Counter Oscillator clock. The figures include information on
when interrupt flags are set. Figure 41 contains timing data for basic Timer/Counter
operation. The figure shows the count sequence close to the MAX value in all modes
other than phase correct PWM mode.
Figure 41. Timer/Counter Timing Diagram, no Prescaling
clkI/O
clkTn
(clkI/O/1)
TCNTn
TOVn
MAX - 1
MAX
BOTTOM
BOTTOM + 1
Figure 42 shows the same timing data, but with the prescaler enabled.
100
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 42. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O/8)
TCNTn
TOVn
MAX - 1
MAX
BOTTOM
BOTTOM + 1
Figure 43 shows the setting of OCF0 in all modes except CTC mode.
Figure 43. Timer/Counter Timing Diagram, Setting of OCF0, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O/8)
TCNTn
OCRn
OCFn
OCRn - 1
OCRn
OCRn + 1
OCRn + 2
OCRn Value
Figure 44 shows the setting of OCF0 and the clearing of TCNT0 in CTC mode.
Figure 44. Timer/Counter Timing Diagram, Clear Timer on Compare Match Mode, with
Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O/8)
TCNTn
(CTC)
TOP - 1
TOP
BOTTOM
BOTTOM + 1
OCRn
TOP
OCFn
101
2490G–AVR–03/04
8-bit Timer/Counter
Register Description
Timer/Counter Control
Register – TCCR0
Bit
7
FOC0
W
6
WGM00
R/W
0
5
COM01
R/W
0
4
COM00
R/W
0
3
WGM01
R/W
0
2
CS02
R/W
0
1
CS01
R/W
0
0
CS00
R/W
0
TCCR0
Read/Write
Initial Value
0
• Bit 7 – FOC0: Force Output Compare
The FOC0 bit is only active when the WGM bits specify a non-PWM mode. However, for
ensuring compatibility with future devices, this bit must be set to zero when TCCR0 is
written when operating in PWM mode. When writing a logical one to the FOC0 bit, an
immediate Compare Match is forced on the waveform generation unit. The OC0 output
is changed according to its COM01:0 bits setting. Note that the FOC0 bit is implemented
as a strobe. Therefore it is the value present in the COM01:0 bits that determines the
effect of the forced compare.
A FOC0 strobe will not generate any interrupt, nor will it clear the timer in CTC mode
using OCR0 as TOP.
The FOC0 bit is always read as zero.
• Bit 6, 3 – WGM01:0: Waveform Generation Mode
These bits control the counting sequence of the counter, the source for the maximum
(TOP) counter value, and what type of waveform generation to be used. Modes of oper-
ation supported by the Timer/Counter unit are: Normal mode, Clear Timer on Compare
match (CTC) mode, and two types of Pulse Width Modulation (PWM) modes. See Table
52 and “Modes of Operation” on page 96.
Table 52. Waveform Generation Mode Bit Description(1)
WGM01 WGM00 Timer/Counter Mode of
Update of TOV0 Flag
OCR0 at Set on
Mode
(CTC0)
(PWM0) Operation
TOP
0xFF
0xFF
0
1
2
3
0
0
1
1
0
1
0
1
Normal
Immediate MAX
TOP BOTTOM
PWM, Phase Correct
CTC
OCR0 Immediate MAX
0xFF TOP MAX
Fast PWM
Note:
1. The CTC0 and PWM0 bit definition names are now obsolete. Use the WGM01:0 def-
initions. However, the functionality and location of these bits are compatible with
previous versions of the timer.
• Bit 5:4 – COM01:0: Compare Match Output Mode
These bits control the Output Compare pin (OC0) behavior. If one or both of the
COM01:0 bits are set, the OC0 output overrides the normal port functionality of the I/O
pin it is connected to. However, note that the Data Direction Register (DDR) bit corre-
sponding to OC0 pin must be set in order to enable the output driver.
When OC0 is connected to the pin, the function of the COM01:0 bits depends on the
WGM01:0 bit setting. Table 53 shows the COM01:0 bit functionality when the WGM01:0
bits are set to a Normal or CTC mode (non-PWM).
102
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 53. Compare Output Mode, non-PWM Mode
COM01
COM00
Description
0
0
1
1
0
1
0
1
Normal port operation, OC0 disconnected.
Toggle OC0 on Compare Match.
Clear OC0 on Compare Match.
Set OC0 on Compare Match.
Table 54 shows the COM01:0 bit functionality when the WGM01:0 bits are set to fast
PWM mode.
Table 54. Compare Output Mode, Fast PWM Mode(1)
COM01
COM00
Description
0
0
1
1
0
1
0
1
Normal port operation, OC0 disconnected.
Reserved
Clear OC0 on Compare Match, set OC0 at TOP.
Set OC0 on Compare Match, clear OC0 at TOP.
Note:
1. A special case occurs when OCR0 equals TOP and COM01 is set. In this case, the
Compare Match is ignored, but the set or clear is done at TOP. See “Fast PWM
Mode” on page 97 for more details.
Table 55 shows the COM01:0 bit functionality when the WGM01:0 bits are set to phase
correct PWM mode.
Table 55. Compare Output Mode, Phase Correct PWM Mode(1)
COM01 COM00 Description
0
0
1
0
1
0
Normal port operation, OC0 disconnected.
Reserved.
Clear OC0 on Compare Match when up-counting. Set OC0 on Compare
Match when downcounting.
1
1
Set OC0 on Compare Match when up-counting. Clear OC0 on Compare
Match when downcounting.
Note:
1. A special case occurs when OCR0 equals TOP and COM01 is set. In this case, the
Compare Match is ignored, but the set or clear is done at TOP. See “Phase Correct
PWM Mode” on page 99 for more details.
103
2490G–AVR–03/04
• Bit 2:0 – CS02:0: Clock Select
The three Clock Select bits select the clock source to be used by the Timer/Counter, see
Table 56.
Table 56. Clock Select Bit Description
CS02
CS01
CS00
Description
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
No clock source (Timer/counter stopped)
clkT0S/(No prescaling)
clkT0S/8 (From prescaler)
clkT0S/32 (From prescaler)
clkT0S/64 (From prescaler)
clkT0S/128 (From prescaler)
clkT S/256 (From prescaler)
0
clkT S/1024 (From prescaler)
0
Timer/Counter Register –
TCNT0
Bit
7
6
5
4
3
2
1
0
TCNT0[7:0]
TCNT0
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
The Timer/Counter Register gives direct access, both for read and write operations, to
the Timer/Counter unit 8-bit counter. Writing to the TCNT0 Register blocks (removes)
the Compare Match on the following timer clock. Modifying the counter (TCNT0) while
the counter is running, introduces a risk of missing a Compare Match between TCNT0
and the OCR0 Register.
Output Compare Register –
OCR0
Bit
7
6
5
4
3
2
1
0
OCR0[7:0]
OCR0
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
The Output Compare Register contains an 8-bit value that is continuously compared
with the counter value (TCNT0). A match can be used to generate an Output Compare
interrupt, or to generate a waveform output on the OC0 pin.
104
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
AsynchronousOperation
of the Timer/Counter
Asynchronous Status
Register – ASSR
Bit
7
–
R
0
6
–
R
0
5
–
R
0
4
–
R
0
3
2
1
0
AS0
R/W
0
TCN0UB
OCR0UB
TCR0UB
ASSR
Read/Write
Initial Value
R
0
R
0
R
0
• Bit 3 – AS0: Asynchronous Timer/Counter0
When AS0 is written to zero, Timer/Counter0 is clocked from the I/O clock, clkI/O. When
AS0 is written to one, Timer/Counter 0 is clocked from a crystal Oscillator connected to
the Timer Oscillator 1 (TOSC1) pin. When the value of AS0 is changed, the contents of
TCNT0, OCR0, and TCCR0 might be corrupted.
• Bit 2 – TCN0UB: Timer/Counter0 Update Busy
When Timer/Counter0 operates asynchronously and TCNT0 is written, this bit becomes
set. When TCNT0 has been updated from the temporary storage register, this bit is
cleared by hardware. A logical zero 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. When OCR0 has been updated from the temporary storage register, this bit is
cleared by hardware. A logical zero 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. When TCCR0 has been updated from the temporary storage register, this bit is
cleared by hardware. A logical zero 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, the updated value might get corrupted and cause an unintentional interrupt to
occur.
The mechanisms for reading TCNT0, OCR0, and TCCR0 are different. When reading
TCNT0, the actual timer value is read. When reading OCR0 or TCCR0, the value in the
temporary storage register is read.
Asynchronous Operation of
Timer/Counter0
When Timer/Counter0 operates asynchronously, some considerations must be taken.
•
Warning: When switching between asynchronous and synchronous clocking of
Timer/Counter0, the timer registers TCNT0, OCR0, and TCCR0 might be corrupted.
A safe procedure for switching clock source is:
1. Disable the Timer/Counter0 interrupts by clearing OCIE0 and TOIE0.
2. Select clock source by setting AS0 as appropriate.
3. Write new values to TCNT0, OCR0, and TCCR0.
4. To switch to asynchronous operation: Wait for TCN0UB, OCR0UB, and
TCR0UB.
5. Clear the Timer/Counter0 interrupt flags.
105
2490G–AVR–03/04
6. Enable interrupts, if needed.
•
•
The Oscillator is optimized for use with a 32.768 kHz watch crystal. Applying an
external clock to the TOSC1 pin may result in incorrect Timer/Counter0 operation.
The CPU main clock frequency must be more than four times the Oscillator
frequency.
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
have been transferred to its destination. Each of the three mentioned registers have
their individual temporary register, for example, writing to TCNT0 does not disturb
an OCR0 write in progress. To detect that a transfer to the destination register has
taken place, the Asynchronous Status Register – ASSR has been implemented.
•
When entering Power-save or Extended Standby 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
enter sleep mode before the changes are effective. This is particularly important if
the Output Compare0 interrupt is used to wake up the device, since the Output
Compare function is disabled during writing to OCR0 or TCNT0. If the write cycle is
not finished, and the MCU enters sleep mode before the OCR0UB bit returns to
zero, the device will never receive a Compare Match interrupt, and the MCU will not
wake up.
•
If Timer/Counter0 is used to wake the device up from Power-save or Extended
Standby mode, precautions must be taken if the user wants to reenter one of these
modes: The interrupt logic needs one TOSC1 cycle to be reset. If the time between
wake-up and re-entering sleep 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 or Extended Standby mode 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 or Extended Standby mode.
•
When the asynchronous operation is selected, the 32.768 kHz Oscillator for
Timer/Counter0 is always running, except in Power-down and Standby modes. After
a Power-up Reset or wake-up from Power-down or Standby mode, 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 or Standby mode. The contents of all
Timer/Counter0 registers must be considered lost after a wake-up from Power-down
or Standby mode due to unstable clock signal upon start-up, no matter whether the
Oscillator is in use or a clock signal is applied to the TOSC1 pin.
•
•
Description of wake up from Power-save or Extended Standby 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.
After wake-up, the MCU is halted for four cycles, it executes the interrupt routine,
and resumes execution from the instruction following SLEEP.
Reading of the TCNT0 Register shortly after wake-up from Power-save may give an
incorrect result. Since TCNT0 is clocked on the asynchronous TOSC clock, reading
TCNT0 must be done through a register synchronized to the internal I/O clock
domain. Synchronization takes place for every rising TOSC1 edge. When waking up
from Power-save mode, and the I/O clock (clkI/O) again becomes active, TCNT0 will
106
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
read as the previous value (before entering sleep) until the next rising TOSC1 edge.
The phase of the TOSC clock after waking up from Power-save mode is essentially
unpredictable, as it depends on the wake-up time. The recommended procedure for
reading TCNT0 is thus as follows:
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.
•
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.
Timer/Counter Interrupt Mask
Register – TIMSK
Bit
7
OCIE2
R/W
0
6
TOIE2
R/W
0
5
TICIE1
R/W
0
4
OCIE1A
R/W
0
3
OCIE1B
R/W
0
2
TOIE1
R/W
0
1
OCIE0
R/W
0
0
TOIE0
R/W
0
TIMSK
Read/Write
Initial Value
• Bit 1 – OCIE0: Timer/Counter0 Output Compare Match Interrupt Enable
When the OCIE0 bit is written to one, and the I-bit in the Status Register is set (one), the
Timer/Counter0 Compare Match interrupt is enabled. The corresponding interrupt 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 – TIFR.
• Bit 0 – TOIE0: Timer/Counter0 Overflow Interrupt Enable
When the TOIE0 bit is written to one, and the I-bit in the Status Register is set (one), the
Timer/Counter0 Overflow interrupt is enabled. The corresponding interrupt is executed if
an overflow in Timer/Counter0 occurs, i.e., when the TOV0 bit is set in the
Timer/Counter Interrupt Flag Register – TIFR.
Timer/Counter Interrupt Flag
Register – TIFR
Bit
7
OCF2
R/W
0
6
TOV2
R/W
0
5
4
OCF1A
R/W
0
3
OCF1B
R/W
0
2
TOV1
R/W
0
1
OCF0
R/W
0
0
TOV0
R/W
0
ICF1
R/W
0
TIFR
Read/Write
Initial Value
• Bit 1 – OCF0: Output Compare Flag 0
The OCF0 bit is set (one) when a Compare Match occurs between the Timer/Counter0
and the data in OCR0 – Output Compare Register0. OCF0 is cleared by hardware when
executing the corresponding interrupt handling vector. Alternatively, OCF0 is cleared by
writing a logic one to the flag. When the I-bit in SREG, OCIE0 (Timer/Counter0 Com-
pare Match Interrupt Enable), and OCF0 are set (one), the Timer/Counter0 Compare
Match Interrupt is executed.
107
2490G–AVR–03/04
• 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 logic one to the flag. When the SREG I-bit, 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 changes counting direction at 0x00.
Timer/Counter Prescaler Figure 45. Prescaler for Timer/Counter0
clkOSC
clkT0S
10-BIT T/C PRESCALER
Clear
TOSC1
AS0
PSR0
0
CS00
CS01
CS02
TIMER/COUNTER0 CLOCK SOURCE
clkT0
The clock source for Timer/Counter0 is named clkT0S. clkT0S is by default connected to
the main system clock clkOSC. By setting the AS0 bit in ASSR, Timer/Counter0 is asyn-
chronously clocked from the TOSC1 pin. This enables use of Timer/Counter0 as a Real
Time Counter (RTC). When AS0 is set, pins TOSC1 and TOSC2 are disconnected from
Port C. A crystal can then be connected between the TOSC1 and TOSC2 pins to serve
as an independent clock source for Timer/Counter0. The Oscillator is optimized for use
with a 32.768 kHz crystal. Applying an external clock source to TOSC1 is not
recommended.
For Timer/Counter0, the possible prescaled selections are: clkT0S/8, clkT0S/32, clkT0S/64,
clkT0S/128, clkT0S/256, and clkT0S/1024. Additionally, clkT0S as well as 0 (stop) may be
selected. Setting the PSR0 bit in SFIOR resets the prescaler. This allows the user to
operate with a predictable prescaler.
108
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Special Function IO Register –
SFIOR
Bit
7
6
–
R
0
5
–
R
0
4
–
R
0
3
ACME
R/W
0
2
1
PSR0
R/W
0
0
PSR321
R/W
0
TSM
R/W
0
PUD
R/W
0
SFIOR
Read/Write
Initial Value
• Bit 7 – TSM: Timer/Counter Synchronization Mode
Writing TSM bit to one activates the Timer/Counter Synchronization mode. In this mode,
the value that is written to PSR0 and PSR321 bits is kept, hence keeping the corre-
sponding prescaler reset signals asserted. This ensures that the corresponding
Timer/Counters are halted and can be configured to the same value without the risk of
one of them advancing during configuration. When the TSM bit written zero, the PSR0
and PSR321 bits are cleared by hardware, and the Timer/Counters start counting
simultaneously.
• Bit 1 – PSR0: Prescaler Reset Timer/Counter0
When this bit is one, the Timer/Counter0 prescaler will be reset. The bit is normally
cleared immediately by hardware. If this bit is written when Timer/Counter0 is operating
in Asynchronous mode, the bit will remain one until the prescaler has been reset. The bit
will not be cleared by hardware if the TSM bit is set.
109
2490G–AVR–03/04
16-bit Timer/Counter The 16-bit Timer/Counter unit allows accurate program execution timing (event man-
agement), wave generation, and signal timing measurement. The main features are:
• True 16-bit Design (i.e., Allows 16-bit PWM)
• Three Independent Output Compare Units
• Double Buffered Output Compare Registers
• One Input Capture Unit
(Timer/Counter1 and
Timer/Counter3)
• Input Capture Noise Canceler
• Clear Timer on Compare Match (Auto Reload)
• Glitch-free, Phase Correct Pulse Width Modulator (PWM)
• Variable PWM Period
• Frequency Generator
• External Event Counter
• Ten Independent Interrupt Sources (TOV1, OCF1A, OCF1B, OCF1C, ICF1, TOV3, OCF3A,
OCF3B, OCF3C, and ICF3)
Restrictions in ATmega103
Compatibility Mode
Note that in ATmega103 compatibility mode, only one 16-bit Timer/Counter is available
(Timer/Counter1). Also note that in ATmega103 compatibility mode, the Timer/Counter1
has two compare registers (Compare A and Compare B) only.
Overview
Most register and bit references in this datasheet are written in general form. A lower
case “n” replaces the Timer/Counter number, and a lower case “x” replaces the Output
Compare unit channel. However, when using the register or bit defines in a program, the
precise form must be used (i.e,. TCNT1 for accessing Timer/Counter1 counter value
and so on). The physical I/O Register and bit locations for ATmega64 are listed in the
“16-bit Timer/Counter Register Description” on page 131.
A simplified block diagram of the 16-bit Timer/Counter is shown in Figure 46. CPU
accessible I/O Registers, including I/O bits and I/O pins, are shown in bold.
110
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 46. 16-bit Timer/Counter Block Diagram(1)
Count
TOVx
(Int.Req.)
Clear
Control Logic
Clock Select
Direction
TCLK
Edge
Detector
Tx
TOP BOTTOM
( From Prescaler )
Timer/Counter
TCNTx
=
= 0
OCFxA
(Int.Req.)
Waveform
Generation
OCxA
OCxB
OCxC
=
OCRxA
OCFxB
(Int.Req.)
Fixed
TOP
Values
Waveform
Generation
=
OCRxB
OCFxC
(Int.Req.)
Waveform
Generation
=
OCRxC
( From Analog
Comparator Ouput )
ICFx (Int.Req.)
Edge
Detector
Noise
Canceler
ICRx
ICPx
TCCRxA
TCCRxB
TCCRxC
Note:
1. Refer to Figure 1 on page 2, Table 30 on page 72, and Table 39 on page 79 for
Timer/Counter1 and 3 pin placement and description.
Registers
The Timer/Counter (TCNTn), Output Compare Registers (OCRnA/B/C), and Input Cap-
ture Register (ICRn) are all 16-bit registers. Special procedures must be followed when
accessing the 16-bit registers. These procedures are described in the section “Access-
ing 16-bit Registers” on page 113. The Timer/Counter Control Registers (TCCRnA/B/C)
are 8-bit registers and have no CPU access restrictions. Interrupt requests (shorten as
Int.Req.) signals are all visible in the Timer Interrupt Flag Register (TIFR) and Extended
Timer Interrupt Flag Register (ETIFR). All interrupts are individually masked with the
Timer Interrupt Mask Register (TIMSK) and Extended Timer Interrupt Mask Register
(ETIMSK). (E)TIFR and (E)TIMSK are not shown in the figure since these registers are
shared by other timer units.
The Timer/Counter can be clocked internally, via the prescaler, or by an external clock
source on the Tn pin. The Clock Select logic block controls which clock source and edge
the Timer/Counter uses to increment (or decrement) its value. The Timer/Counter is
inactive when no clock source is selected. The output from the Clock Select logic is
referred to as the timer clock (clk ).
n
T
111
2490G–AVR–03/04
The double buffered Output Compare Registers (OCRnA/B/C) are compared with the
Timer/Counter value at all time. The result of the compare can be used by the Waveform
Generator to generate a PWM or variable frequency output on the Output Compare Pin
(OCnA/B/C). See “Output Compare Units” on page 119. The Compare Match event will
also set the Compare Match Flag (OCFnA/B/C) which can be used to generate an Out-
put Compare interrupt request.
The Input Capture Register can capture the Timer/Counter value at a given external
(edge triggered) event on either the Input Capture pin (ICPn) or on the Analog Compar-
ator pins (See “Analog Comparator” on page 227.) The Input Capture unit includes a
digital filtering unit (Noise Canceler) for reducing the chance of capturing noise spikes.
The TOP value, or maximum Timer/Counter value, can in some modes of operation be
defined by either the OCRnA Register, the ICRn Register, or by a set of fixed values.
When using OCRnA as TOP value in a PWM mode, the OCRnA Register can not be
used for generating a PWM output. However, the TOP value will in this case be double
buffered allowing the TOP value to be changed in run time. If a fixed TOP value is
required, the ICRn Register can be used as an alternative, freeing the OCRnA to be
used as PWM output.
Definitions
The following definitions are used extensively throughout this section:
Table 57. Definitions
BOTTOM The counter reaches the BOTTOM when it becomes 0x0000.
MAX
The counter reaches its MAXimum when it becomes 0xFFFF (decimal
65535).
TOP
The counter reaches the TOP when it becomes equal to the highest
value in the count sequence. The TOP value can be assigned to be one
of the fixed values: 0x00FF, 0x01FF, or 0x03FF, or to the value stored in
the OCRnA or ICRn Register. The assignment is dependent of the mode
of operation.
Compatibility
The 16-bit Timer/Counter has been updated and improved from previous versions of the
16-bit AVR Timer/Counter. This 16-bit Timer/Counter is fully compatible with the earlier
version regarding:
•
•
•
All 16-bit Timer/Counter related I/O Register address locations, including Timer
Interrupt Registers.
Bit locations inside all 16-bit Timer/Counter registers, including Timer Interrupt
Registers.
Interrupt Vectors.
The following control bits have changed name, but have same functionality and register
location:
•
•
•
PWMn0 is changed to WGMn0.
PWMn1 is changed to WGMn1.
CTCn is changed to WGMn2.
The following registers are added to the 16-bit Timer/Counter:
•
•
Timer/Counter Control Register C (TCCRnC).
Output Compare Register C, OCRnCH and OCRnCL, combined OCRnC.
The following bits are added to the 16-bit Timer/Counter control registers:
COM1C1:0 are added to TCCR1A.
•
112
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
•
•
FOCnA, FOCnB, and FOCnC are added in the new TCCRnC Register.
WGMn3 is added to TCCRnB.
Interrupt flag and mask bits for Output Compare unit C are added.
The 16-bit Timer/Counter has improvements that will affect the compatibility in some
special cases.
Accessing 16-bit
Registers
The TCNTn, OCRnA/B/C, and ICRn are 16-bit registers that can be accessed by the
AVR CPU via the 8-bit data bus. The 16-bit register must be byte accessed using two
read or write operations. Each 16-bit timer has a single 8-bit register for temporary stor-
ing of the high byte of the 16-bit access. The same temporary register is shared
between all 16-bit registers within each 16-bit timer. Accessing the low byte triggers the
16-bit read or write operation. When the low byte of a 16-bit register is written by the
CPU, the high byte stored in the temporary register, and the low byte written are both
copied into the 16-bit register in the same clock cycle. When the low byte of a 16-bit reg-
ister is read by the CPU, the high byte of the 16-bit register is copied into the temporary
register in the same clock cycle as the low byte is read.
Not all 16-bit accesses uses the temporary register for the high byte. Reading the
OCRnA/B/C 16-bit registers does not involve using the temporary register.
To do a 16-bit write, the high byte must be written before the low byte. For a 16-bit read,
the low byte must be read before the high byte.
The following code examples show how to access the 16-bit timer registers assuming
that no interrupts updates the temporary register. The same principle can be used
directly for accessing the OCRnA/B/C and ICRn Registers. Note that when using “C”,
the compiler handles the 16-bit access.
Assembly Code Examples(1)
...
; Set TCNTn to 0x01FF
ldi r17,0x01
ldi r16,0xFF
out TCNTnH,r17
out TCNTnL,r16
; Read TCNTn into r17:r16
in r16,TCNTnL
in r17,TCNTnH
...
C Code Examples(1)
unsigned int i;
...
/* Set TCNTn to 0x01FF */
TCNTn = 0x1FF;
/* Read TCNTn into i */
i = TCNTn;
...
Note:
1. The example code assumes that the part specific header file is included.
For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”,
and “SBI” instructions must be replaced with instructions that allow access to
113
2490G–AVR–03/04
extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
The assembly code example returns the TCNTn value in the r17:r16 register pair.
It is important to notice that accessing 16-bit registers are atomic operations. If an inter-
rupt occurs between the two instructions accessing the 16-bit register, and the interrupt
code updates the temporary register by accessing the same or any other of the 16-bit
timer registers, then the result of the access outside the interrupt will be corrupted.
Therefore, when both the main code and the interrupt code update the temporary regis-
ter, the main code must disable the interrupts during the 16-bit access.
The following code examples show how to do an atomic read of the TCNTn Register
contents. Reading any of the OCRnA/B/C or ICRn Registers can be done by using the
same principle.
Assembly Code Example(1)
TIM16_ReadTCNTn:
; Save global interrupt flag
in r18,SREG
; Disable interrupts
cli
; Read TCNTn into r17:r16
in r16,TCNTnL
in r17,TCNTnH
; Restore global interrupt flag
out SREG,r18
ret
C Code Example(1)
unsigned int TIM16_ReadTCNTn( void )
{
unsigned char sreg;
unsigned int i;
/* Save global interrupt flag */
sreg = SREG;
/* Disable interrupts */
_CLI();
/* Read TCNTn into i */
i = TCNTn;
/* Restore global interrupt flag */
SREG = sreg;
return i;
}
Note:
1. The example code assumes that the part specific header file is included.
For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”,
and “SBI” instructions must be replaced with instructions that allow access to
extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
The assembly code example returns the TCNTn value in the r17:r16 register pair.
114
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
The following code examples show how to do an atomic write of the TCNTn Register
contents. Writing any of the OCRnA/B/C or ICRn Registers can be done by using the
same principle.
Assembly Code Example(1)
TIM16_WriteTCNTn:
; Save global interrupt flag
in r18,SREG
; Disable interrupts
cli
; Set TCNTn to r17:r16
out TCNTnH,r17
out TCNTnL,r16
; Restore global interrupt flag
out SREG,r18
ret
C Code Example(1)
void TIM16_WriteTCNTn( unsigned int i )
{
unsigned char sreg;
unsigned int i;
/* Save global interrupt flag */
sreg = SREG;
/* Disable interrupts */
_CLI();
/* Set TCNTn to i */
TCNTn = i;
/* Restore global interrupt flag */
SREG = sreg;
}
Note:
1. The example code assumes that the part specific header file is included.
For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”,
and “SBI” instructions must be replaced with instructions that allow access to
extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
The assembly code example requires that the r17:r16 register pair contains the value to
be written to TCNTn.
Reusing the Temporary High
Byte Register
If writing to more than one 16-bit register where the high byte is the same for all registers
written, then the high byte only needs to be written once. However, note that the same
rule of atomic operation described previously also applies in this case.
Timer/Counter Clock
Sources
The Timer/Counter can be clocked by an internal or an external clock source. The clock
source is selected by the Clock Select logic which is controlled by the Clock Select
(CSn2:0) bits located in the Timer/Counter Control Register B (TCCRnB). For details on
clock sources and prescaler, see “Timer/Counter3, Timer/Counter2 and Timer/Counter1
Prescalers” on page 142.
115
2490G–AVR–03/04
Counter Unit
The main part of the 16-bit Timer/Counter is the programmable 16-bit bi-directional
counter unit. Figure 47 shows a block diagram of the counter and its surroundings.
Figure 47. Counter Unit Block Diagram
DATA BUS (8-bit)
TOVn
(Int.Req.)
TEMP (8-bit)
Clock Select
Count
Clear
Edge
Detector
Tn
TCNTnH (8-bit)
TCNTnL (8-bit)
clkTn
Control Logic
Direction
TCNTn (16-bit Counter)
( From Prescaler )
TOP
BOTTOM
Signal description (internal signals):
Count Increment or decrement TCNTn by 1.
Direction Select between increment and decrement.
Clear
Clear TCNTn (set all bits to zero).
Timer/counter clock.
clkT
n
TOP
Signalize that TCNTn has reached maximum value.
BOTTOM Signalize that TCNTn has reached minimum value (zero).
The 16-bit counter is mapped into two 8-bit I/O memory locations: Counter High
(TCNTnH) containing the upper eight bits of the counter, and Counter Low (TCNTnL)
containing the lower eight bits. The TCNTnH Register can only be indirectly accessed
by the CPU. When the CPU does an access to the TCNTnH I/O location, the CPU
accesses the high byte temporary register (TEMP). The temporary register is updated
with the TCNTnH value when the TCNTnL is read, and TCNTnH is updated with the
temporary register value when TCNTnL is written. This allows the CPU to read or write
the entire 16-bit counter value within one clock cycle via the 8-bit data bus. It is impor-
tant to notice that there are special cases of writing to the TCNTn Register when the
counter is counting that will give unpredictable results. The special cases are described
in the sections where they are of importance.
Depending on the mode of operation used, the counter is cleared, incremented, or dec-
remented at each Timer Clock (clk ). The clk n can be generated from an external or
n
T
T
internal clock source, selected by the Clock Select bits (CSn2:0). When no clock source
is selected (CSn2:0 = 0) the timer is stopped. However, the TCNTn value can be
accessed by the CPU, independent of whether clkTn is present or not. A CPU write over-
rides (has priority over) all counter clear or count operations.
The counting sequence is determined by the setting of the Waveform Generation mode
bits (WGMn3:0) located in the Timer/Counter Control Registers A and B (TCCRnA and
TCCRnB). There are close connections between how the counter behaves (counts) and
how waveforms are generated on the Output Compare outputs OCnx. For more details
about advanced counting sequences and waveform generation, see “Modes of Opera-
tion” on page 122.
The Timer/Counter Overflow Flag (TOVn) is set according to the mode of operation
selected by the WGMn3:0 bits. TOVn can be used for generating a CPU interrupt.
116
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Input Capture Unit
The Timer/Counter incorporates an Input Capture unit that can capture external events
and give them a time-stamp indicating time of occurrence. The external signal indicating
an event, or multiple events, can be applied via the ICPn pin or alternatively, for the
Timer/Counter1 only, via the Analog Comparator unit. The time-stamps can then be
used to calculate frequency, duty-cycle, and other features of the signal applied. Alter-
natively the time-stamps can be used for creating a log of the events.
The Input Capture unit is illustrated by the block diagram shown in Figure 48. The ele-
ments of the block diagram that are not directly a part of the Input Capture unit are gray
shaded. The small “n” in register and bit names indicates the Timer/Counter number.
Figure 48. Input Capture Unit Block Diagram(1)
DATA BUS (8-bit)
TEMP (8-bit)
ICRnH (8-bit)
ICRnL (8-bit)
TCNTnH (8-bit)
TCNTnL (8-bit)
ICRn (16-bit Register)
TCNTn (16-bit Counter)
WRITE
ACO*
ACIC*
ICNC
ICES
Analog
Comparator
Noise
Canceler
Edge
Detector
ICFn (Int.Req.)
ICPn
Note:
1. The Analog Comparator Output (ACO) can only trigger the Timer/Counter1 ICP – not
Timer/Counter3.
When a change of the logic level (an event) occurs on the Input Capture pin (ICPn),
alternatively on the Analog Comparator output (ACO), and this change confirms to the
setting of the edge detector, a capture will be triggered. When a capture is triggered, the
16-bit value of the counter (TCNTn) is written to the Input Capture Register (ICRn). The
Input Capture Flag (ICFn) is set at the same system clock as the TCNTn value is copied
into ICRn Register. If enabled (TICIEn = 1), the Input Capture Flag generates an Input
Capture interrupt. The ICFn flag is automatically cleared when the interrupt is executed.
Alternatively the ICFn flag can be cleared by software by writing a logical one to its I/O
bit location.
Reading the 16-bit value in the Input Capture Register (ICRn) is done by first reading the
low byte (ICRnL) and then the high byte (ICRnH). When the low byte is read the high
byte is copied into the high byte temporary register (TEMP). When the CPU reads the
ICRnH I/O location it will access the TEMP Register.
The ICRn Register can only be written when using a Waveform Generation mode that
utilizes the ICRn Register for defining the counter’s TOP value. In these cases the
Waveform Generation mode (WGMn3:0) bits must be set before the TOP value can be
117
2490G–AVR–03/04
written to the ICRn Register. When writing the ICRn Register the high byte must be writ-
ten to the ICRnH I/O location before the low byte is written to ICRnL.
For more information on how to access the 16-bit registers refer to “Accessing 16-bit
Registers” on page 113.
Input Capture Trigger Source
The main trigger source for the Input Capture unit is the Input Capture pin (ICPn).
Timer/Counter1 can alternatively use the Analog Comparator output as trigger source
for the Input Capture unit. The Analog Comparator is selected as trigger source by set-
ting the Analog Comparator Input Capture (ACIC) bit in the Analog Comparator Control
and Status Register (ACSR). Be aware that changing trigger source can trigger a cap-
ture. The Input Capture Flag must therefore be cleared after the change.
Both the Input Capture pin (ICPn) and the Analog Comparator output (ACO) inputs are
sampled using the same technique as for the Tn pin (Figure 59 on page 142). The edge
detector is also identical. However, when the noise canceler is enabled, additional logic
is inserted before the edge detector, which increases the delay by four system clock
cycles. Note that the input of the noise canceler and edge detector is always enabled
unless the Timer/Counter is set in a Waveform Generation mode that uses ICRn to
define TOP.
An Input Capture can be triggered by software by controlling the port of the ICPn pin.
Noise Canceler
The Noise Canceler improves noise immunity by using a simple digital filtering scheme.
The Noise Canceler input is monitored over four samples, and all four must be equal for
changing the output that in turn is used by the edge detector.
The Noise Canceler is enabled by setting the Input Capture Noise Canceler (ICNCn) bit
in Timer/Counter Control Register B (TCCRnB). When enabled the Noise Canceler
introduces additional four system clock cycles of delay from a change applied to the
input, to the update of the ICRn Register. The Noise Canceler uses the system clock
and is therefore not affected by the prescaler.
Using the Input Capture Unit
The main challenge when using the Input Capture unit is to assign enough processor
capacity for handling the incoming events. The time between two events is critical. If the
processor has not read the captured value in the ICRn Register before the next event
occurs, the ICRn will be overwritten with a new value. In this case the result of the cap-
ture will be incorrect.
When using the Input Capture interrupt, the ICRn Register should be read as early in the
interrupt handler routine as possible. Even though the Input Capture interrupt has rela-
tively high priority, the maximum interrupt response time is dependent on the maximum
number of clock cycles it takes to handle any of the other interrupt requests.
Using the Input Capture unit in any mode of operation when the TOP value (resolution)
is actively changed during operation, is not recommended.
Measurement of an external signal’s duty cycle requires that the trigger edge is changed
after each capture. Changing the edge sensing must be done as early as possible after
the ICRn Register has been read. After a change of the edge, the Input Capture Flag
(ICFn) must be cleared by software (writing a logical one to the I/O bit location). For
measuring frequency only, the clearing of the ICFn flag is not required (if an interrupt
handler is used).
118
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Output Compare Units
The 16-bit comparator continuously compares TCNTn with the Output Compare Regis-
ter (OCRnx). If TCNT equals OCRnx the comparator signals a match. A match will set
the Output Compare Flag (OCFnx) at the next timer clock cycle. If enabled (OCIEnx =
1), the Output Compare Flag generates an Output Compare interrupt. The OCFnx flag is
automatically cleared when the interrupt is executed. Alternatively the OCFnx flag can
be cleared by software by writing a logical one to its I/O bit location. The Waveform Gen-
erator uses the match signal to generate an output according to operating mode set by
the Waveform Generation mode (WGMn3:0) bits and Compare Output mode
(COMnx1:0) bits. The TOP and BOTTOM signals are used by the Waveform Generator
for handling the special cases of the extreme values in some modes of operation (See
“Modes of Operation” on page 122.)
A special feature of Output Compare unit A allows it to define the Timer/Counter TOP
value (i.e., counter resolution). In addition to the counter resolution, the TOP value
defines the period time for waveforms generated by the Waveform Generator.
Figure 49 shows a block diagram of the Output Compare unit. The small “n” in the regis-
ter and bit names indicates the device number (n = n for Timer/Counter n), and the “x”
indicates Output Compare unit (A/B/C). The elements of the block diagram that are not
directly a part of the Output Compare unit are gray shaded.
Figure 49. Output Compare Unit, Block Diagram
DATA BUS (8-bit)
TEMP (8-bit)
OCRnxH Buf. (8-bit)
OCRnxL Buf. (8-bit)
TCNTnH (8-bit)
TCNTnL (8-bit)
OCRnx Buffer (16-bit Register)
TCNTn (16-bit Counter)
OCRnxH (8-bit)
OCRnxL (8-bit)
OCRnx (16-bit Register)
=
(16-bit Comparator )
OCFnx (Int.Req.)
TOP
OCnx
Waveform Generator
BOTTOM
WGMn3:0
COMnx1:0
The OCRnx Register is double buffered when using any of the twelve Pulse Width Mod-
ulation (PWM) modes. For the normal and Clear Timer on Compare (CTC) modes of
operation, the double buffering is disabled. The double buffering synchronizes the
update of the OCRnx Compare Register to either TOP or BOTTOM of the counting
sequence. The synchronization prevents the occurrence of odd-length, non-symmetrical
PWM pulses, thereby making the output glitch-free.
The OCRnx Register access may seem complex, but this is not case. When the double
buffering is enabled, the CPU has access to the OCRnx Buffer Register, and if double
119
2490G–AVR–03/04
buffering is disabled the CPU will access the OCRnx directly. The content of the OCR1x
(Buffer or Compare) Register is only changed by a write operation (the Timer/Counter
does not update this register automatically as the TCNTn – and ICRn Register). There-
fore OCRnx is not read via the high byte temporary register (TEMP). However, it is a
good practice to read the low byte first as when accessing other 16-bit registers. Writing
the OCRnx registers must be done via the TEMP Register since the compare of all 16
bits is done continuously. The high byte (OCRnxH) has to be written first. When the high
byte I/O location is written by the CPU, the TEMP Register will be updated by the value
written. Then when the low byte (OCRnxL) is written to the lower eight bits, the high byte
will be copied into the upper eight bits of either the OCRnx Buffer or OCRnx Compare
Register in the same system clock cycle.
For more information of how to access the 16-bit registers refer to “Accessing 16-bit
Registers” on page 113.
Force Output Compare
In non-PWM waveform generation modes, the match output of the comparator can be
forced by writing a one to the Force Output Compare (FOCnx) bit. Forcing Compare
Match will not set the OCFnx flag or reload/clear the timer, but the OCnx pin will be
updated as if a real Compare Match had occurred (the COMn1:0 bits settings define
whether the OCnx pin is set, cleared or toggled).
Compare Match Blocking by
TCNTn Write
All CPU writes to the TCNTn Register will block any Compare Match that occurs in the
next timer clock cycle, even when the timer is stopped. This feature allows OCRnx to be
initialized to the same value as TCNTn without triggering an interrupt when the
Timer/Counter clock is enabled.
Using the Output Compare
Unit
Since writing TCNTn in any mode of operation will block all Compare Matches for one
timer clock cycle, there are risks involved when changing TCNTn when using any of the
Output Compare channels, independent of whether the Timer/Counter is running or not.
If the value written to TCNTn equals the OCRnx value, the Compare Match will be
missed, resulting in incorrect waveform generation. Do not write the TCNTn equal to
TOP in PWM modes with variable TOP values. The Compare Match for the TOP will be
ignored and the counter will continue to 0xFFFF. Similarly, do not write the TCNTn value
equal to BOTTOM when the counter is downcounting.
The setup of the OCnx should be performed before setting the Data Direction Register
for the port pin to output. The easiest way of setting the OCnx value is to use the Force
Output Compare (FOCnx) strobe bits in Normal mode. The OCnx Register keeps its
value even when changing between waveform generation modes.
Be aware that the COMnx1:0 bits are not double buffered together with the compare
value. Changing the COMnx1:0 bits will take effect immediately.
Compare Match Output
Unit
The Compare Output mode (COMnx1:0) bits have two functions. The Waveform Gener-
ator uses the COMnx1:0 bits for defining the Output Compare (OCnx) state at the next
Compare Match. Secondly the COMnx1:0 bits control the OCnx pin output source. Fig-
ure 50 shows a simplified schematic of the logic affected by the COMnx1:0 bit setting.
The I/O Registers, I/O bits, and I/O pins in the figure are shown in bold. Only the parts of
the general I/O port control registers (DDR and PORT) that are affected by the
COMnx1:0 bits are shown. When referring to the OCnx state, the reference is for the
internal OCnx Register, not the OCnx pin. If a System Reset occur, the OCnx Register is
reset to “0”.
120
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 50. Compare Match Output Unit, Schematic
COMnx1
Waveform
Generator
COMnx0
FOCnx
D
Q
1
OCnx
Pin
OCnx
0
D
Q
PORT
D
Q
DDR
clkI/O
The general I/O port function is overridden by the Output Compare (OCnx) from the
Waveform Generator if either of the COMnx1:0 bits are set. However, the OCnx pin
direction (input or output) is still controlled by the Data Direction Register (DDR) for the
port pin. The Data Direction Register bit for the OCnx pin (DDR_OCnx) must be set as
output before the OCnx value is visible on the pin. The port override function is generally
independent of the Waveform Generation mode, but there are some exceptions. Refer
to Table 58, Table 59 and Table 60 for details.
The design of the Output Compare pin logic allows initialization of the OCnx state before
the output is enabled. Note that some COMnx1:0 bit settings are reserved for certain
modes of operation. See “16-bit Timer/Counter Register Description” on page 131.
The COMnx1:0 bits have no effect on the Input Capture unit.
Compare Output Mode and
Waveform Generation
The Waveform Generator uses the COMnx1:0 bits differently in Normal, CTC, and PWM
modes. For all modes, setting the COMnx1:0 = 0 tells the Waveform Generator that no
action on the OCnx Register is to be performed on the next Compare Match. For com-
pare output actions in the non-PWM modes refer to Table 58 on page 132. For fast
PWM mode refer to Table 59 on page 132, and for phase correct and phase and fre-
quency correct PWM refer to Table 60 on page 133.
A change of the COMnx1:0 bits state will have effect at the first Compare Match after
the bits are written. For non-PWM modes, the action can be forced to have immediate
effect by using the FOCnx strobe bits.
121
2490G–AVR–03/04
Modes of Operation
The mode of operation, i.e., the behavior of the Timer/Counter and the Output Compare
pins, is defined by the combination of the Waveform Generation mode (WGMn3:0) and
Compare Output mode (COMnx1:0) bits. The Compare Output mode bits do not affect
the counting sequence, while the Waveform Generation mode bits do. The COMnx1:0
bits control whether the PWM output generated should be inverted or not (inverted or
non-inverted PWM). For non-PWM modes the COMnx1:0 bits control whether the out-
put should be set, cleared or toggle at a Compare Match (See “Compare Match Output
Unit” on page 120.)
For detailed timing information refer to “Timer/Counter Timing Diagrams” on page 129.
Normal Mode
The simplest mode of operation is the Normal mode (WGMn3:0 = 0). In this mode the
counting direction is always up (incrementing), and no counter clear is performed. The
counter simply overruns when it passes its maximum 16-bit value (MAX = 0xFFFF) and
then restarts from the BOTTOM (0x0000). In normal operation the Timer/Counter Over-
flow Flag (TOVn) will be set in the same timer clock cycle as the TCNTn becomes zero.
The TOVn flag in this case behaves like a 17th bit, except that it is only set, not cleared.
However, combined with the timer overflow interrupt that automatically clears the TOVn
flag, the timer resolution can be increased by software. There are no special cases to
consider in the Normal mode, a new counter value can be written anytime.
The Input Capture unit is easy to use in Normal mode. However, observe that the maxi-
mum interval between the external events must not exceed the resolution of the counter.
If the interval between events are too long, the timer overflow interrupt or the prescaler
must be used to extend the resolution for the capture unit.
The Output Compare units can be used to generate interrupts at some given time. Using
the Output Compare to generate waveforms in Normal mode is not recommended,
since this will occupy too much of the CPU time.
Clear Timer on Compare
Match (CTC) Mode
In Clear Timer on Compare or CTC mode (WGMn3:0 = 4 or 12), the OCRnA or ICRn
Register are used to manipulate the counter resolution. In CTC mode the counter is
cleared to zero when the counter value (TCNTn) matches either the OCRnA (WGMn3:0
= 4) or the ICRn (WGMn3:0 = 12). The OCRnA or ICRn define the top value for the
counter, hence also its resolution. This mode allows greater control of the Compare
Match output frequency. It also simplifies the operation of counting external events.
The timing diagram for the CTC mode is shown in Figure 51. The counter value
(TCNTn) increases until a Compare Match occurs with either OCRnA or ICRn, and then
counter (TCNTn) is cleared.
122
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 51. CTC Mode, Timing Diagram
OCnA Interrupt Flag Set
or ICFn Interrupt Flag Set
(Interrupt on TOP)
TCNTn
OCnA
(Toggle)
(COMnA1:0 = 1)
1
2
3
4
Period
An interrupt can be generated at each time the counter value reaches the TOP value by
either using the OCFnA or ICFn flag according to the register used to define the TOP
value. If the interrupt is enabled, the interrupt handler routine can be used for updating
the TOP value. However, changing the TOP to a value close to BOTTOM when the
counter is running with none or a low prescaler value must be done with care since the
CTC mode does not have the double buffering feature. If the new value written to
OCRnA or ICRn is lower than the current value of TCNTn, the counter will miss the
Compare Match. The counter will then have to count to its maximum value (0xFFFF)
and wrap around starting at 0x0000 before the Compare Match can occur. In many
cases this feature is not desirable. An alternative will then be to use the fast PWM mode
using OCRnA for defining TOP (WGMn3:0 = 15) since the OCRnA then will be double
buffered.
For generating a waveform output in CTC mode, the OCnA output can be set to toggle
its logical level on each Compare Match by setting the Compare Output mode bits to
toggle mode (COMnA1:0 = 1). The OCnA value will not be visible on the port pin unless
the data direction for the pin is set to output (DDR_OCnA = 1). The waveform generated
will have a maximum frequency of fOC A = fclk_I/O/2 when OCRnA is set to zero (0x0000).
n
The waveform frequency is defined by the following equation:
f
clk_I/O
f
= --------------------------------------------------
OCnA
2 ⋅ N ⋅ (1 + OCRnA)
The N variable represents the prescaler factor (1, 8, 64, 256, or 1024).
As for the Normal mode of operation, the TOVn flag is set in the same timer clock cycle
that the counter counts from MAX to 0x0000.
Fast PWM Mode
The fast Pulse Width Modulation or fast PWM mode (WGMn3:0 = 5, 6, 7, 14, or 15) pro-
vides a high frequency PWM waveform generation option. The fast PWM differs from
the other PWM options by its single-slope operation. The counter counts from BOTTOM
to TOP then restarts from BOTTOM. In non-inverting Compare Output mode, the Output
Compare (OCnx) is set on the Compare Match between TCNTn and OCRnx, and
cleared at TOP. In inverting Compare Output mode output is cleared on Compare Match
and set at TOP. Due to the single-slope operation, the operating frequency of the fast
PWM mode can be twice as high as the phase correct and phase and frequency correct
PWM modes that use dual-slope operation. This high frequency makes the fast PWM
mode well suited for power regulation, rectification, and DAC applications. High fre-
123
2490G–AVR–03/04
quency allows physically small sized external components (coils, capacitors), hence
reduces total system cost.
The PWM resolution for fast PWM can be fixed to 8-, 9-, or 10-bit, or defined by either
ICRn or OCRnA. The minimum resolution allowed is 2-bit (ICRn or OCRnA set to
0x0003), and the maximum resolution is 16-bit (ICRn or OCRnA set to MAX). The PWM
resolution in bits can be calculated by using the following equation:
log(TOP + 1)
= ----------------------------------
R
FPWM
log(2)
In fast PWM mode the counter is incremented until the counter value matches either
one of the fixed values 0x00FF, 0x01FF, or 0x03FF (WGMn3:0 = 5, 6, or 7), the value in
ICRn (WGMn3:0 = 14), or the value in OCRnA (WGMn3:0 = 15). The counter is then
cleared at the following timer clock cycle. The timing diagram for the fast PWM mode is
shown in Figure 52. The figure shows fast PWM mode when OCRnA or ICRn is used to
define TOP. The TCNTn value is in the timing diagram shown as a histogram for illus-
trating the single-slope operation. The diagram includes non-inverted and inverted PWM
outputs. The small horizontal line marks on the TCNTn slopes represent Compare
Matches between OCRnx and TCNTn. The OCnx interrupt flag will be set when a Com-
pare Match occurs.
Figure 52. Fast PWM Mode, Timing Diagram
OCRnx / TOP Update
and TOVn Interrupt Flag
Set and OCnA Interrupt
Flag Set or ICFn
Interrupt Flag Set
(Interrupt on TOP)
TCNTn
(COMnx1:0 = 2)
OCnx
(COMnx1:0 = 3)
OCnx
1
2
3
4
5
6
7
8
Period
The Timer/Counter Overflow Flag (TOVn) is set each time the counter reaches TOP. In
addition the OCnA or ICFn flag is set at the same timer clock cycle as TOVn is set when
either OCRnA or ICRn is used for defining the TOP value. If one of the interrupts are
enabled, the interrupt handler routine can be used for updating the TOP and compare
values.
When changing the TOP value the program must ensure that the new TOP value is
higher or equal to the value of all of the compare registers. If the TOP value is lower
than any of the compare registers, a Compare Match will never occur between the
TCNTn and the OCRnx. Note that when using fixed TOP values the unused bits are
masked to zero when any of the OCRnx Registers are written.
The procedure for updating ICRn differs from updating OCRnA when used for defining
the TOP value. The ICRn Register is not double buffered. This means that if ICRn is
124
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
changed to a low value when the counter is running with none or a low prescaler value,
there is a risk that the new ICRn value written is lower than the current value of TCNTn.
The result will then be that the counter will miss the Compare Match at the TOP value.
The counter will then have to count to the MAX value (0xFFFF) and wrap around start-
ing at 0x0000 before the Compare Match can occur. The OCRnA Register however, is
double buffered. This feature allows the OCRnA I/O location to be written anytime.
When the OCRnA I/O location is written the value written will be put into the OCRnA
Buffer Register. The OCRnA Compare Register will then be updated with the value in
the buffer register at the next timer clock cycle the TCNTn matches TOP. The update is
done at the same timer clock cycle as the TCNTn is cleared and the TOVn flag is set.
Using the ICRn Register for defining TOP works well when using fixed TOP values. By
using ICRn, the OCRnA Register is free to be used for generating a PWM output on
OCnA. However, if the base PWM frequency is actively changed (by changing the TOP
value), using the OCRnA as TOP is clearly a better choice due to its double buffer
feature.
In fast PWM mode, the compare units allow generation of PWM waveforms on the
OCnx pins. Setting the COMnx1:0 bits to two will produce a non-inverted PWM and an
inverted PWM output can be generated by setting the COMnx1:0 to three (See Table 59
on page 132). The actual OCnx value will only be visible on the port pin if the data direc-
tion for the port pin is set as output (DDR_OCnx). The PWM waveform is generated by
setting (or clearing) the OCnx Register at the Compare Match between OCRnx and
TCNTn, and clearing (or setting) the OCnx Register at the timer clock cycle the counter
is cleared (changes from TOP to BOTTOM).
The PWM frequency for the output can be calculated by the following equation:
f
clk_I/O
f
= ----------------------------------
OCnxPWM
N ⋅ (1 + TOP)
The N variable represents the prescaler divider (1, 8, 64, 256, or 1024).
The extreme values for the OCRnx Register represents special cases when generating
a PWM waveform output in the fast PWM mode. If the OCRnx is set equal to BOTTOM
(0x0000) the output will be a narrow spike for each TOP+1 timer clock cycle. Setting the
OCRnx equal to TOP will result in a constant high or low output (depending on the polar-
ity of the output set by the COMnx1:0 bits.)
A frequency (with 50% duty cycle) waveform output in fast PWM mode can be achieved
by setting OCnA to toggle its logical level on each Compare Match (COMnA1:0 = 1).
This applies only if OCRnA is used to define the TOP value (WGMn3:0 = 15). The wave-
form generated will have a maximum frequency of fOC A = fclk_I/O/2 when OCRnA is set to
n
zero (0x0000). This feature is similar to the OCnA toggle in CTC mode, except the dou-
ble buffer feature of the Output Compare unit is enabled in the fast PWM mode.
Phase Correct PWM Mode
The phase correct Pulse Width Modulation or phase correct PWM mode (WGMn3:0 = 1,
2, 3, 10, or 11) provides a high resolution phase correct PWM waveform generation
option. The phase correct PWM mode is, like the phase and frequency correct PWM
mode, based on a dual-slope operation. The counter counts repeatedly from BOTTOM
(0x0000) to TOP and then from TOP to BOTTOM. In non-inverting Compare Output
mode, the Output Compare (OCnx) is cleared on the Compare Match between TCNTn
and OCRnx while upcounting, and set on the Compare Match while downcounting. In
inverting Output Compare mode, the operation is inverted. The dual-slope operation has
lower maximum operation frequency than single slope operation. However, due to the
symmetric feature of the dual-slope PWM modes, these modes are preferred for motor
control applications.
125
2490G–AVR–03/04
The PWM resolution for the phase correct PWM mode can be fixed to 8-, 9-, or 10-bit, or
defined by either ICRn or OCRnA. The minimum resolution allowed is 2-bit (ICRn or
OCRnA set to 0x0003), and the maximum resolution is 16-bit (ICRn or OCRnA set to
MAX). The PWM resolution in bits can be calculated by using the following equation:
log(TOP + 1)
= ----------------------------------
R
PCPWM
log(2)
In phase correct PWM mode the counter is incremented until the counter value matches
either one of the fixed values 0x00FF, 0x01FF, or 0x03FF (WGMn3:0 = 1, 2, or 3), the
value in ICRn (WGMn3:0 = 10), or the value in OCRnA (WGMn3:0 = 11). The counter
has then reached the TOP and changes the count direction. The TCNTn value will be
equal to TOP for one timer clock cycle. The timing diagram for the phase correct PWM
mode is shown on Figure 53. The figure shows phase correct PWM mode when OCRnA
or ICRn is used to define TOP. The TCNTn value is in the timing diagram shown as a
histogram for illustrating the dual-slope operation. The diagram includes non-inverted
and inverted PWM outputs. The small horizontal line marks on the TCNTn slopes repre-
sent Compare Matches between OCRnx and TCNTn. The OCnx interrupt flag will be set
when a Compare Match occurs.
Figure 53. Phase Correct PWM Mode, Timing Diagram
OCRnx / TOP Update and
OCnA Interrupt Flag Set
or ICFn Interrupt Flag Set
(Interrupt on TOP)
TOVn Interrupt Flag Set
(Interrupt on Bottom)
TCNTn
(COMnx1:0 = 2)
OCnx
(COMnx1:0 = 3)
OCnx
1
2
3
4
Period
The Timer/Counter Overflow Flag (TOVn) is set each time the counter reaches BOT-
TOM. When either OCRnA or ICRn is used for defining the TOP value, the OCnA or
ICFn flag is set accordingly at the same timer clock cycle as the OCRnx Registers are
updated with the double buffer value (at TOP). The interrupt flags can be used to gener-
ate an interrupt each time the counter reaches the TOP or BOTTOM value.
When changing the TOP value the program must ensure that the new TOP value is
higher or equal to the value of all of the compare registers. If the TOP value is lower
than any of the compare registers, a Compare Match will never occur between the
TCNTn and the OCRnx. Note that when using fixed TOP values, the unused bits are
masked to zero when any of the OCRnx Registers are written. As the third period shown
in Figure 53 illustrates, changing the TOP actively while the Timer/Counter is running in
the phase correct mode can result in an unsymmetrical output. The reason for this can
be found in the time of update of the OCRnx Register. Since the OCRnx update occurs
126
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
at TOP, the PWM period starts and ends at TOP. This implies that the length of the fall-
ing slope is determined by the previous TOP value, while the length of the rising slope is
determined by the new TOP value. When these two values differ the two slopes of the
period will differ in length. The difference in length gives the unsymmetrical result on the
output.
It is recommended to use the phase and frequency correct mode instead of the phase
correct mode when changing the TOP value while the Timer/Counter is running. When
using a static TOP value there are practically no differences between the two modes of
operation.
In phase correct PWM mode, the compare units allow generation of PWM waveforms on
the OCnx pins. Setting the COMnx1:0 bits to two will produce a non-inverted PWM and
an inverted PWM output can be generated by setting the COMnx1:0 to three (See Table
60 on page 133). The actual OCnx value will only be visible on the port pin if the data
direction for the port pin is set as output (DDR_OCnx). The PWM waveform is gener-
ated by setting (or clearing) the OCnx Register at the Compare Match between OCRnx
and TCNTn when the counter increments, and clearing (or setting) the OCnx Register at
Compare Match between OCRnx and TCNTn when the counter decrements. The PWM
frequency for the output when using phase correct PWM can be calculated by the fol-
lowing equation:
f
clk_I/O
f
= ---------------------------
OCnxPCPWM
2 ⋅ N ⋅ TOP
The N variable represents the prescaler divider (1, 8, 64, 256, or 1024).
The extreme values for the OCRnx Register represents special cases when generating
a PWM waveform output in the phase correct PWM mode. If the OCRnx is set equal to
BOTTOM the output will be continuously low and if set equal to TOP the output will be
continuously high for non-inverted PWM mode. For inverted PWM the output will have
the opposite logic values. If OCRnA is used to define the TOP value (WGMn3:0 = 11)
and COMnA1:0 = 1, the OCnA output will toggle with a 50% duty cycle.
Phase and Frequency Correct The phase and frequency correct Pulse Width Modulation, or phase and frequency cor-
PWM Mode
rect PWM mode (WGMn3:0 = 8 or 9) provides a high resolution phase and frequency
correct PWM waveform generation option. The phase and frequency correct PWM
mode is, like the phase correct PWM mode, based on a dual-slope operation. The
counter counts repeatedly from BOTTOM (0x0000) to TOP and then from TOP to BOT-
TOM. In non-inverting Compare Output mode, the Output Compare (OCnx) is cleared
on the Compare Match between TCNTn and OCRnx while upcounting, and set on the
Compare Match while downcounting. In inverting Compare Output mode, the operation
is inverted. The dual-slope operation gives a lower maximum operation frequency com-
pared to the single-slope operation. However, due to the symmetric feature of the dual-
slope PWM modes, these modes are preferred for motor control applications.
The main difference between the phase correct, and the phase and frequency correct
PWM mode is the time the OCRnx Register is updated by the OCRnx Buffer Register,
(see Figure 53 and Figure 54).
The PWM resolution for the phase and frequency correct PWM mode can be defined by
either ICRn or OCRnA. The minimum resolution allowed is 2-bit (ICRn or OCRnA set to
0x0003), and the maximum resolution is 16-bit (ICRn or OCRnA set to MAX). The PWM
resolution in bits can be calculated using the following equation:
log(TOP + 1)
= ----------------------------------
R
PFCPWM
log(2)
127
2490G–AVR–03/04
In phase and frequency correct PWM mode the counter is incremented until the counter
value matches either the value in ICRn (WGMn3:0 = 8), or the value in OCRnA
(WGMn3:0 = 9). The counter has then reached the TOP and changes the count direc-
tion. The TCNTn value will be equal to TOP for one timer clock cycle. The timing
diagram for the phase correct and frequency correct PWM mode is shown on Figure 54.
The figure shows phase and frequency correct PWM mode when OCRnA or ICRn is
used to define TOP. The TCNTn value is in the timing diagram shown as a histogram for
illustrating the dual-slope operation. The diagram includes non-inverted and inverted
PWM outputs. The small horizontal line marks on the TCNTn slopes represent Compare
Matches between OCRnx and TCNTn. The OCnx interrupt flag will be set when a Com-
pare Match occurs.
Figure 54. Phase and Frequency Correct PWM Mode, Timing Diagram
OCnA Interrupt Flag Set
or ICFn Interrupt Flag Set
(Interrupt on TOP)
OCRnx / TOP Update and
TOVn Interrupt Flag Set
(Interrupt on Bottom)
TCNTn
(COMnx1:0 = 2)
OCnx
(COMnx1:0 = 3)
OCnx
1
2
3
4
Period
The Timer/Counter Overflow Flag (TOVn) is set at the same timer clock cycle as the
OCRnx Registers are updated with the double buffer value (at BOTTOM). When either
OCRnA or ICRn is used for defining the TOP value, the OCnA or ICFn flag set when
TCNTn has reached TOP. The interrupt flags can then be used to generate an interrupt
each time the counter reaches the TOP or BOTTOM value.
When changing the TOP value the program must ensure that the new TOP value is
higher or equal to the value of all of the compare registers. If the TOP value is lower
than any of the compare registers, a Compare Match will never occur between the
TCNTn and the OCRnx.
As Figure 54 shows the output generated is, in contrast to the phase correct mode, sym-
metrical in all periods. Since the OCRnx registers are updated at BOTTOM, the length
of the rising and the falling slopes will always be equal. This gives symmetrical output
pulses and is therefore frequency correct.
Using the ICRn Register for defining TOP works well when using fixed TOP values. By
using ICRn, the OCRnA Register is free to be used for generating a PWM output on
OCnA. However, if the base PWM frequency is actively changed by changing the TOP
value, using the OCRnA as TOP is clearly a better choice due to its double buffer
feature.
In phase and frequency correct PWM mode, the compare units allow generation of
PWM waveforms on the OCnx pins. Setting the COMnx1:0 bits to two will produce a
128
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
non-inverted PWM and an inverted PWM output can be generated by setting the
COMnx1:0 to three (See Table 60 on page 133). The actual OCnx value will only be vis-
ible on the port pin if the data direction for the port pin is set as output (DDR_OCnx). The
PWM waveform is generated by setting (or clearing) the OCnx Register at the Compare
Match between OCRnx and TCNTn when the counter increments, and clearing (or set-
ting) the OCnx Register at Compare Match between OCRnx and TCNTn when the
counter decrements. The PWM frequency for the output when using phase and fre-
quency correct PWM can be calculated by the following equation:
f
clk_I/O
f
= ---------------------------
OCnxPFCPWM
2 ⋅ N ⋅ TOP
The N variable represents the prescaler divider (1, 8, 64, 256, or 1024).
The extreme values for the OCRnx Register represent special cases when generating a
PWM waveform output in the phase correct PWM mode. If the OCRnx is set equal to
BOTTOM the output will be continuously low and if set equal to TOP the output will be
set to high for non-inverted PWM mode. For inverted PWM the output will have the
opposite logic values. If OCnA is used to define the TOP value (WGMn3:0 = 9) and
COMnA1:0 = 1, the OCnA output will toggle with a 50% duty cycle.
Timer/Counter Timing
Diagrams
The Timer/Counter is a synchronous design and the timer clock (clkTn) is therefore
shown as a clock enable signal in the following figures. The figures include information
on when interrupt flags are set, and when the OCRnx Register is updated with the
OCRnx buffer value (only for modes utilizing double buffering). Figure 55 shows a timing
diagram for the setting of OCFnx.
Figure 55. Timer/Counter Timing Diagram, Setting of OCFnx, no Prescaling
clkI/O
clkTn
(clkI/O/1)
TCNTn
OCRnx
OCFnx
OCRnx - 1
OCRnx
OCRnx + 1
OCRnx + 2
OCRnx Value
Figure 56 shows the same timing data, but with the prescaler enabled.
129
2490G–AVR–03/04
Figure 56. Timer/Counter Timing Diagram, Setting of OCFnx, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O/8)
TCNTn
OCRnx
OCFnx
OCRnx - 1
OCRnx
OCRnx + 1
OCRnx + 2
OCRnx Value
Figure 57 shows the count sequence close to TOP in various modes. When using phase
and frequency correct PWM mode the OCRnx Register is updated at BOTTOM. The
timing diagrams will be the same, but TOP should be replaced by BOTTOM, TOP-1 by
BOTTOM+1 and so on. The same renaming applies for modes that set the TOVn flag at
BOTTOM.
Figure 57. Timer/Counter Timing Diagram, no Prescaling
clkI/O
clkTn
(clkI/O/1)
TCNTn
TOP - 1
TOP - 1
TOP
TOP
BOTTOM
TOP - 1
BOTTOM + 1
TOP - 2
(CTC and FPWM)
TCNTn
(PC and PFC PWM)
TOVn (FPWM)
and ICFn (if used
as TOP)
OCRnx
(Update at TOP)
Old OCRnx Value
New OCRnx Value
Figure 58 shows the same timing data, but with the prescaler enabled.
130
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 58. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O/8)
TCNTn
TOP - 1
TOP - 1
TOP
TOP
BOTTOM
TOP - 1
BOTTOM + 1
TOP - 2
(CTC and FPWM)
TCNTn
(PC and PFC PWM)
TOVn (FPWM)
and ICFn (if used
as TOP)
OCRnx
(Update at TOP)
Old OCRnx Value
New OCRnx Value
16-bit Timer/Counter
Register Description
Timer/Counter1 Control
Register A – TCCR1A
Bit
7
COM1A1
R/W
6
COM1A0
R/W
5
COM1B1
R/W
4
3
COM1C1
R/W
2
1
WGM11
R/W
0
0
COM1B0
R/W
0
COM1C0
R/W
0
WGM10
R/W
0
TCCR1A
TCCR3A
Read/Write
Initial Value
0
0
0
0
Timer/Counter3 Control
Register A – TCCR3A
Bit
7
COM3A1
R/W
6
COM3A0
R/W
5
COM3B1
R/W
4
COM3B0
R/W
3
COM3C1
R/W
2
COM3C0
R/W
1
WGM31
R/W
0
0
WGM30
R/W
0
Read/Write
Initial Value
0
0
0
0
0
0
• Bit 7:6 – COMnA1:0: Compare Output Mode for Channel A
• Bit 5:4 – COMnB1:0: Compare Output Mode for Channel B
• Bit 3:2 – COMnC1:0: Compare Output Mode for Channel C
The COMnA1:0, COMnB1:0, and COMnC1:0 control the Output Compare pins (OCnA,
OCnB, and OCnC respectively) behavior. If one or both of the COMnA1:0 bits are writ-
ten to one, the OCnA output overrides the normal port functionality of the I/O pin it is
connected to. If one or both of the COMnB1:0 bits are written to one, the OCnB output
overrides the normal port functionality of the I/O pin it is connected to. If one or both of
the COMnC1:0 bits are written to one, the OCnC output overrides the normal port func-
tionality of the I/O pin it is connected to. However, note that the Data Direction Register
(DDR) bit corresponding to the OCnA, OCnB or OCnC pin must be set in order to
enable the output driver.
When the OCnA, OCnB or OCnC is connected to the pin, the function of the COMnx1:0
bits is dependent of the WGMn3:0 bits setting. Table 58 shows the COMnx1:0 bit func-
tionality when the WGMn3:0 bits are set to a Normal or a CTC mode (non-PWM).
131
2490G–AVR–03/04
Table 58. Compare Output Mode, non-PWM
COMnA1/
COMnB1/
COMnC1
COMnA0/
COMnB0/
COMnC0
Description
0
0
1
0
1
0
Normal port operation, OCnA/OCnB/OCnC disconnected.
Toggle OCnA/OCnB/OCnC on Compare Match.
Clear OCnA/OCnB/OCnC on Compare Match (Set output to
low level).
1
1
Set OCnA/OCnB/OCnC on Compare Match (Set output to high
level).
Table 59 shows the COMnx1:0 bit functionality when the WGMn3:0 bits are set to the
fast PWM mode
Table 59. Compare Output Mode, Fast PWM(1)
COMnA1/
COMnB1/
COMnC0
COMnA0/
COMnB0/
COMnC0
Description
0
0
0
1
Normal port operation, OCnA/OCnB/OCnC disconnected.
WGMn3:0 = 15: Toggle OCnA on Compare Match,
OCnB/OCnC disconnected (normal port operation).
For all other WGMn settings, normal port operation,
OCnA/OCnB/OCnC disconnected.
1
1
0
1
Clear OCnA/OCnB/OCnC on Compare Match, set
OCnA/OCnB/OCnC at TOP.
Set OCnA/OCnB/OCnC on Compare Match, clear
OCnA/OCnB/OCnC at TOP.
Note:
1. A special case occurs when OCRnA/OCRnB/OCRnC equals TOP and
COMnA1/COMnB1/COMnC1 is set. In this case the Compare Match is ignored, but
the set or clear is done at TOP. See “Fast PWM Mode” on page 123. for more details.
Table 59 shows the COMnx1:0 bit functionality when the WGMn3:0 bits are set to the
phase correct and frequency correct PWM mode.
132
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 60. Compare Output Mode, Phase Correct and Phase and Frequency Correct
PWM(1)
COMnA1/
COMnB1/
COMnC1
COMnA0/
COMnB0/
COMnC0
Description
0
0
0
1
Normal port operation, OCnA/OCnB/OCnC disconnected.
WGMn3:0 = 9 or 14: Toggle OCnA on Compare Match,
OCnB/OCnC disconnected (normal port operation).
Forr all other WGMn settings, normal port operation,
OCnA/OCnB/OCnC disconnected.
1
1
0
1
Clear OCnA/OCnB/OCnC on Compare Match when up-
counting. Set OCnA/OCnB/OCnC on Compare Match when
downcounting.
Set OCnA/OCnB/OCnC on Compare Match when up-counting.
Clear OCnA/OCnB/OCnC on Compare Match when
downcounting.
Note:
1. A special case occurs when OCRnA/OCRnB/OCRnC equals TOP and
COMnA1/COMnB1/COMnC1 is set. See “Phase Correct PWM Mode” on page 125.
for more details.
• Bit 1:0 – WGMn1:0: Waveform Generation Mode
Combined with the WGMn3:2 bits found in the TCCRnB Register, these bits control the
counting sequence of the counter, the source for maximum (TOP) counter value, and
what type of waveform generation to be used, see Table 61. Modes of operation sup-
ported by the Timer/Counter unit are: Normal mode (counter), Clear Timer on Compare
match (CTC) mode, and three types of Pulse Width Modulation (PWM) modes. (See
“Modes of Operation” on page 122.)
133
2490G–AVR–03/04
Table 61. Waveform Generation Mode Bit Description
WGMn2
(CTCn)
WGMn1
WGMn0
Timer/Counter Mode of
Update of TOVn Flag
OCRnx at Set on
Mode WGMn3
(PWMn1) (PWMn0) Operation
TOP
0
1
2
3
4
5
6
7
8
0
0
0
0
0
0
0
0
1
0
0
0
0
1
1
1
1
0
0
0
1
1
0
0
1
1
0
0
1
0
1
0
1
0
1
0
Normal
0xFFFF Immediate MAX
PWM, Phase Correct, 8-bit
PWM, Phase Correct, 9-bit
PWM, Phase Correct, 10-bit
CTC
0x00FF
0x01FF
0x03FF
OCRnA
0x00FF
0x01FF
0x03FF
TOP
TOP
TOP
BOTTOM
BOTTOM
BOTTOM
Immediate MAX
Fast PWM, 8-bit
TOP
TOP
TOP
TOP
TOP
TOP
Fast PWM, 9-bit
Fast PWM, 10-bit
PWM, Phase and Frequency
Correct
ICRn
BOTTOM
BOTTOM
BOTTOM
BOTTOM
9
1
0
0
1
PWM, Phase and Frequency
Correct
OCRnA
10
11
1
1
1
1
1
1
0
0
1
1
1
1
1
1
0
0
1
1
0
1
0
1
0
1
PWM, Phase Correct
PWM, Phase Correct
CTC
ICRn
OCRnA
ICRn
–
TOP
TOP
BOTTOM
BOTTOM
12
Immediate MAX
13
(Reserved)
–
–
14
Fast PWM
ICRn
OCRnA
TOP
TOP
TOP
TOP
15
Fast PWM
Note:
The CTCn and PWMn1:0 bit definition names are obsolete. Use the WGMn2:0 definitions. However, the functionality and loca-
tion of these bits are compatible with previous versions of the timer.
Timer/Counter1 Control
Register B – TCCR1B
Bit
7
ICNC1
R/W
0
6
ICES1
R/W
0
5
–
R
0
4
WGM13
R/W
0
3
WGM12
R/W
0
2
CS12
R/W
0
1
CS11
R/W
0
0
CS10
R/W
0
TCCR1B
Read/Write
Initial Value
Timer/Counter3 Control
Register B – TCCR3B
Bit
7
ICNC3
R/W
0
6
ICES3
R/W
0
5
–
R
0
4
WGM33
R/W
0
3
WGM32
R/W
0
2
CS32
R/W
0
1
CS31
R/W
0
0
CS30
R/W
0
TCCR3B
Read/Write
Initial Value
• Bit 7 – ICNCn: Input Capture Noise Canceler
Setting this bit (to one) activates the Input Capture Noise Canceler. When the noise can-
celer is activated, the input from the Input Capture pin (ICPn) is filtered. The filter
function requires four successive equal valued samples of the ICPn pin for changing its
output. The Input Capture is therefore delayed by four Oscillator cycles when the noise
canceler is enabled.
134
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• Bit 6 – ICESn: Input Capture Edge Select
This bit selects which edge on the Input Capture pin (ICPn) that is used to trigger a cap-
ture event. When the ICESn bit is written to zero, a falling (negative) edge is used as
trigger, and when the ICESn bit is written to one, a rising (positive) edge will trigger the
capture.
When a capture is triggered according to the ICESn setting, the counter value is copied
into the Input Capture Register (ICRn). The event will also set the Input Capture Flag
(ICFn), and this can be used to cause an Input Capture Interrupt, if this interrupt is
enabled.
When the ICRn is used as TOP value (see description of the WGMn3:0 bits located in
the TCCRnA and the TCCRnB Register), the ICPn is disconnected and consequently
the Input Capture function is disabled.
• Bit 5 – Reserved Bit
This bit is reserved for future use. For ensuring compatibility with future devices, this bit
must be written to zero when TCCRnB is written.
• Bit 4:3 – WGMn3:2: Waveform Generation Mode
See TCCRnA Register description.
• Bit 2:0 – CSn2:0: Clock Select
The three Clock Select bits select the clock source to be used by the Timer/Counter, see
Figure 55 and Figure 56.
Table 62. Clock Select Bit Description
CSn2
CSn1
CSn0
Description
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
No clock source (Timer/counter stopped).
clkI/O/1 (No prescaling)
clkI/O/8 (From prescaler)
clkI/O/64 (From prescaler)
clkI/O/256 (From prescaler)
clkI/O/1024 (From prescaler)
External clock source on Tn pin. Clock on falling edge.
External clock source on Tn pin. Clock on rising edge.
If external pin modes are used for the Timer/Countern, transitions on the Tn pin will
clock the counter even if the pin is configured as an output. This feature allows software
control of the counting.
Timer/Counter1 Control
Register C – TCCR1C
Bit
7
FOC1A
W
6
FOC1B
W
5
FOC1C
W
4
–
R
0
3
–
R
0
2
–
R
0
1
–
R
0
0
–
R
0
TCCR1C
Read/Write
Initial Value
0
0
0
135
2490G–AVR–03/04
Timer/Counter3 Control
Register C – TCCR3C
Bit
7
FOC3A
W
6
FOC3B
W
5
FOC3C
W
4
–
R
0
3
–
R
0
2
–
R
0
1
–
R
0
0
–
R
0
TCCR3C
Read/Write
Initial Value
0
0
0
• Bit 7 – FOCnA: Force Output Compare for Channel A
• Bit 6 – FOCnB: Force Output Compare for Channel B
• Bit 5 – FOCnC: Force Output Compare for Channel C
The FOCnA/FOCnB/FOCnC bits are only active when the WGMn3:0 bits specifies a
non-PWM mode. When writing a logical one to the FOCnA/FOCnB/FOCnC bit, an
immediate Compare Match is forced on the waveform generation unit. The
OCnA/OCnB/OCnC output is changed according to its COMnx1:0 bits setting. Note that
the FOCnA/FOCnB/FOCnC bits are implemented as strobes. Therefore it is the value
present in the COMnx1:0 bits that determine the effect of the forced compare.
A FOCnA/FOCnB/FOCnC strobe will not generate any interrupt nor will it clear the timer
in Clear Timer on Compare match (CTC) mode using OCRnA as TOP.
The FOCnA/FOCnB/FOCnB bits are always read as zero.
• Bit 4:0 – Reserved Bits
These bits are reserved for future use. For ensuring compatibility with future devices,
these bits must be written to zero when TCCRnC is written.
Timer/Counter1 – TCNT1H
and TCNT1L
Bit
7
6
5
4
3
2
1
0
TCNT1[15:8]
TCNT1[7:0]
TCNT1H
TCNT1L
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
R/W
0
R/W
0
R/W
0
0
0
Timer/Counter3 – TCNT3H
and TCNT3L
Bit
7
6
5
4
3
2
1
0
TCNT3[15:8]
TCNT3[7:0]
TCNT3H
TCNT3L
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
R/W
0
R/W
0
R/W
0
0
0
The two Timer/Counter I/O locations (TCNTnH and TCNTnL, combined TCNTn) give
direct access, both for read and for write operations, to the Timer/Counter unit 16-bit
counter. 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 tempo-
rary High Byte Register (TEMP). This temporary register is shared by all the other 16-bit
registers. See “Accessing 16-bit Registers” on page 113.
Modifying the counter (TCNTn) while the counter is running introduces a risk of missing
a Compare Match between TCNTn and one of the OCRnx Registers.
Writing to the TCNTn Register blocks (removes) the Compare Match on the following
timer clock for all compare units.
136
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Output Compare Register 1 A
– OCR1AH and OCR1AL
Bit
7
6
5
4
3
2
1
0
OCR1A[15:8]
OCR1A[7:0]
OCR1AH
OCR1AL
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
R/W
0
R/W
0
R/W
0
0
0
Output Compare Register 1 B
– OCR1BH and OCR1BL
Bit
7
6
5
4
3
2
1
0
OCR1B[15:8]
OCR1B[7:0]
OCR1BH
OCR1BL
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
R/W
0
R/W
0
R/W
0
0
0
Output Compare Register 1 C
– OCR1CH and OCR1CL
Bit
7
6
5
4
3
2
1
0
OCR1C[15:8]
OCR1C[7:0]
OCR1CH
OCR1CL
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
R/W
0
R/W
0
R/W
0
0
0
Output Compare Register 3 A
– OCR3AH and OCR3AL
Bit
7
6
5
4
3
2
1
0
OCR3A[15:8]
OCR3A[7:0]
OCR3AH
OCR3AL
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
R/W
0
R/W
0
R/W
0
0
0
Output Compare Register 3 B
– OCR3BH and OCR3BL
Bit
7
6
5
4
3
2
1
0
OCR3B[15:8]
OCR3B[7:0]
OCR3BH
OCR3BL
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
R/W
0
R/W
0
R/W
0
0
0
Output Compare Register 3 C
– OCR3CH and OCR3CL
Bit
7
6
5
4
3
2
1
0
OCR3C[15:8]
OCR3C[7:0]
OCR3CH
OCR3CL
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
R/W
0
R/W
0
R/W
0
0
0
The Output Compare Registers contain a 16-bit value that is continuously compared
with the counter value (TCNTn). A match can be used to generate an Output Compare
interrupt, or to generate a waveform output on the OCnx pin.
The Output Compare Registers are 16-bit in size. To ensure that both the high and low
bytes are written simultaneously when the CPU writes to these registers, the access is
performed using an 8-bit temporary high byte register (TEMP). This temporary register
is shared by all the other 16-bit registers. See “Accessing 16-bit Registers” on page 113.
137
2490G–AVR–03/04
Input Capture Register 1 –
ICR1H and ICR1L
Bit
7
6
5
4
3
2
1
0
ICR1[15:8]
ICR1[7:0]
ICR1H
ICR1L
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
0
R/W
0
R/W
0
R/W
0
0
Input Capture Register 3 –
ICR3H and ICR3L
Bit
7
6
5
4
3
2
1
0
ICR3[15:8]
ICR3[7:0]
ICR3H
ICR3L
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
R/W
0
R/W
0
R/W
0
R/W
0
0
The Input Capture is updated with the counter (TCNTn) value each time an event occurs
on the ICPn pin (or optionally on the Analog Comparator output for Timer/Counter1).
The Input Capture can be used for defining the counter TOP value.
The Input Capture Register is 16-bit in size. To ensure that both the high and low bytes
are read simultaneously when the CPU accesses these registers, the access is per-
formed using an 8-bit temporary high byte register (TEMP). This temporary register is
shared by all the other 16-bit registers. See “Accessing 16-bit Registers” on page 113.
Timer/Counter Interrupt Mask
Register – TIMSK(1)
Bit
7
OCIE2
R/W
0
6
TOIE2
R/W
0
5
TICIE1
R/W
0
4
OCIE1A
R/W
0
3
OCIE1B
R/W
0
2
TOIE1
R/W
0
1
OCIE0
R/W
0
0
TOIE0
R/W
0
TIMSK
Read/Write
Initial Value
Note:
1. This register contains interrupt control bits for several Timer/Counters, but only
Timer1 bits are described in this section. The remaining bits are described in their
respective timer sections.
• Bit 5 – TICIE1: Timer/Counter1, Input Capture Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter1 Input Capture interrupt is enabled. The
corresponding Interrupt Vector (see “Interrupts” on page 59) is executed when the ICF1
flag, located in TIFR, is set.
• Bit 4 – OCIE1A: Timer/Counter1, Output Compare A Match Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter1 Output Compare A Match interrupt is enabled. The
corresponding Interrupt Vector (see “Interrupts” on page 59) is executed when the
OCF1A flag, located in TIFR, is set.
• Bit 3 – OCIE1B: Timer/Counter1, Output Compare B Match Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter1 Output Compare B Match interrupt is enabled. The
corresponding Interrupt Vector (see “Interrupts” on page 59) is executed when the
OCF1B flag, located in TIFR, is set.
138
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• Bit 2 – TOIE1: Timer/Counter1, Overflow Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter1 Overflow Interrupt is enabled. The corresponding
Interrupt Vector (see “Interrupts” on page 59) is executed when the TOV1 flag, located
in TIFR, is set.
Extended Timer/Counter
Interrupt Mask Register –
ETIMSK(1)
Bit
7
–
6
–
5
TICIE3
R/W
0
4
OCIE3A
R/W
0
3
OCIE3B
R/W
0
2
TOIE3
R/W
0
1
OCIE3C
R/W
0
0
OCIE1C
R/W
0
ETIMSK
Read/Write
Initial Value
R
0
R
0
Note:
1. This register is not available in ATmega103 compatibility mode.
• Bit 7:6 – Reserved Bits
These bits are reserved for future use. For ensuring compatibility with future devices,
these bits must be set to zero when ETIMSK is written.
• Bit 5 – TICIE3: Timer/Counter3, Input Capture Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter3 Input Capture interrupt is enabled. The
corresponding Interrupt Vector (see “Interrupts” on page 59) is executed when the ICF3
flag, located in ETIFR, is set.
• Bit 4 – OCIE3A: Timer/Counter3, Output Compare A Match Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter3 Output Compare A Match interrupt is enabled. The
corresponding Interrupt Vector (see “Interrupts” on page 59) is executed when the
OCF3A flag, located in ETIFR, is set.
• Bit 3 – OCIE3B: Timer/Counter3, Output Compare B Match Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter3 Output Compare B Match interrupt is enabled. The
corresponding Interrupt Vector (see “Interrupts” on page 59) is executed when the
OCF3B flag, located in ETIFR, is set.
• Bit 2 – TOIE3: Timer/Counter3, Overflow Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter3 Overflow Interrupt is enabled. The corresponding
Interrupt Vector (see “Interrupts” on page 59) is executed when the TOV3 flag, located
in ETIFR, is set.
• Bit 1 – OCIE3C: Timer/Counter3, Output Compare C Match Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter3 Output Compare C Match interrupt is enabled. The
corresponding Interrupt Vector (see “Interrupts” on page 59) is executed when the
OCF3C flag, located in ETIFR, is set.
139
2490G–AVR–03/04
• Bit 0 – OCIE1C: Timer/Counter1, Output Compare C Match Interrupt Enable
When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-
bally enabled), the Timer/Counter1 Output Compare C Match interrupt is enabled. The
corresponding Interrupt Vector (see “Interrupts” on page 59) is executed when the
OCF1C flag, located in ETIFR, is set.
Timer/Counter Interrupt Flag
Register – TIFR(1)
Bit
7
OCF2
R/W
0
6
TOV2
R/W
0
5
4
OCF1A
R/W
0
3
OCF1B
R/W
0
2
TOV1
R/W
0
1
OCF0
R/W
0
0
TOV0
R/W
0
ICF1
R/W
0
TIFR
Read/Write
Initial Value
Note:
1. This register contains flag bits for several Timer/Counters, but only Timer1 bits are
described in this section. The remaining bits are described in their respective timer
sections.
• Bit 5 – ICF1: Timer/Counter1, Input Capture Flag
This flag is set when a capture event occurs on the ICP1 pin. When the Input Capture
Register (ICR1) is set by the WGMn3:0 to be used as the TOP value, the ICF1 flag is set
when the counter reaches the TOP value.
ICF1 is automatically cleared when the Input Capture Interrupt Vector is executed. Alter-
natively, ICF1 can be cleared by writing a logic one to its bit location.
• Bit 4 – OCF1A: Timer/Counter1, Output Compare A Match Flag
This flag is set in the timer clock cycle after the counter (TCNT1) value matches the Out-
put Compare Register A (OCR1A).
Note that a Forced Output Compare (FOC1A) strobe will not set the OCF1A flag.
OCF1A is automatically cleared when the Output Compare Match A Interrupt Vector is
executed. Alternatively, OCF1A can be cleared by writing a logic one to its bit location.
• Bit 3 – OCF1B: Timer/Counter1, Output Compare B Match Flag
This flag is set in the timer clock cycle after the counter (TCNT1) value matches the Out-
put Compare Register B (OCR1B).
Note that a Forced Output Compare (FOC1B) strobe will not set the OCF1B flag.
OCF1B is automatically cleared when the Output Compare Match B Interrupt Vector is
executed. Alternatively, OCF1B can be cleared by writing a logic one to its bit location.
• Bit 2 – TOV1: Timer/Counter1, Overflow Flag
The setting of this flag is dependent of the WGMn3:0 bits setting. In Normal and CTC
modes, the TOV1 flag is set when the timer overflows. Refer to Table 61 on page 134
for the TOV1 flag behavior when using another WGMn3:0 bit setting.
TOV1 is automatically cleared when the Timer/Counter1 Overflow Interrupt Vector is
executed. Alternatively, TOV1 can be cleared by writing a logic one to its bit location.
Extended Timer/Counter
Interrupt Flag Register –
ETIFR
Bit
7
–
R/W
0
6
–
R/W
0
5
4
OCF3A
R/W
0
3
OCF3B
R/W
0
2
TOV3
R/W
0
1
OCF3C
R/W
0
0
OCF1C
R/W
0
ICF3
R/W
0
ETIFR
Read/Write
Initial Value
140
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• Bit 7:6 – Reserved Bits
These bits are reserved for future use. For ensuring compatibility with future devices,
these bits must be set to zero when ETIFR is written.
• Bit 5 – ICF3: Timer/Counter3, Input Capture Flag
This flag is set when a capture event occurs on the ICP3 pin. When the Input Capture
Register (ICR3) is set by the WGM3:0 to be used as the TOP value, the ICF3 flag is set
when the counter reaches the TOP value.
ICF3 is automatically cleared when the Input Capture 3 Interrupt Vector is executed.
Alternatively, ICF3 can be cleared by writing a logic one to its bit location.
• Bit 4 – OCF3A: Timer/Counter3, Output Compare A Match Flag
This flag is set in the timer clock cycle after the counter (TCNT3) value matches the Out-
put Compare Register A (OCR3A).
Note that a Forced Output Compare (FOC3A) strobe will not set the OCF3A flag.
OCF3A is automatically cleared when the Output Compare Match 3 A Interrupt Vector is
executed. Alternatively, OCF3A can be cleared by writing a logic one to its bit location.
• Bit 3 – OCF3B: Timer/Counter3, Output Compare B Match Flag
This flag is set in the timer clock cycle after the counter (TCNT3) value matches the Out-
put Compare Register B (OCR3B).
Note that a Forced Output Compare (FOC3B) strobe will not set the OCF3B flag.
OCF3B is automatically cleared when the Output Compare Match 3 B Interrupt Vector is
executed. Alternatively, OCF3B can be cleared by writing a logic one to its bit location.
• Bit 2 – TOV3: Timer/Counter3, Overflow Flag
The setting of this flag is dependent of the WGM3:0 bits setting. In Normal and CTC
modes, the TOV3 flag is set when the timer overflows. Refer to Table 52 on page 102
for the TOV3 flag behavior when using another WGM3:0 bit setting.
TOV3 is automatically cleared when the Timer/Counter3 Overflow Interrupt Vector is
executed. Alternatively, OCF3B can be cleared by writing a logic one to its bit location.
• Bit 1 – OCF3C: Timer/Counter3, Output Compare C Match Flag
This flag is set in the timer clock cycle after the counter (TCNT3) value matches the Out-
put Compare Register C (OCR3C).
Note that a Forced Output Compare (FOC3C) strobe will not set the OCF3C flag.
OCF3C is automatically cleared when the Output Compare Match 3 C Interrupt Vector is
executed. Alternatively, OCF3C can be cleared by writing a logic one to its bit location.
• Bit 0 – OCF1C: Timer/Counter1, Output Compare C Match Flag
This flag is set in the timer clock cycle after the counter (TCNT1) value matches the Out-
put Compare Register C (OCR1C).
Note that a Forced Output Compare (FOC1C) strobe will not set the OCF1C flag.
OCF1C is automatically cleared when the Output Compare Match 1 C Interrupt Vector is
executed. Alternatively, OCF1C can be cleared by writing a logic one to its bit location.
141
2490G–AVR–03/04
Timer/Counter3,
Timer/Counter2 and
Timer/Counter1
Prescalers
Timer/Counter3, Timer/Counter2 and Timer/Counter1 share the same prescaler mod-
ule, but the Timer/Counters can have different prescaler settings. The description below
applies to all of the mentioned Timer/Counters.
Internal Clock Source
The Timer/Counter can be clocked directly by the system clock (by setting the
CSn2:0 = 1). This provides the fastest operation, with a maximum Timer/Counter clock
frequency equal to system clock frequency (fCLK_I/O). Alternatively, one of four taps from
the prescaler can be used as a clock source. The prescaled clock has a frequency of
either fCLK_I/O/8, fCLK_I/O/64, fCLK_I/O/256, or fCLK_I/O/1024.
Prescaler Reset
The prescaler is free running, for example, it operates independently of the Clock Select
logic of the Timer/Counter, and it is shared by Timer/Counter1, Timer/Counter2, and
Timer/Counter3. Since the prescaler is not affected by the Timer/Counter’s clock select,
the state of the prescaler will have implications for situations where a prescaled clock is
used. One example of prescaling artifacts occurs when the timer is enabled and clocked
by the prescaler (6 > CSn2:0 > 1). The number of system clock cycles from when the
timer is enabled to the first count occurs can be from 1 to N+1 system clock cycles,
where N equals the prescaler divisor (8, 64, 256, or 1024).
It is possible to use the Prescaler Reset for synchronizing the Timer/Counter to program
execution. However, care must be taken if the other Timer/Counter that shares the
same prescaler also use prescaling. A Prescaler Reset will affect the prescaler period
for all Timer/Counters it is connected to.
External Clock Source
An external clock source applied to the Tn pin can be used as Timer/Counter clock
(clkT1/clkT2/clkT3). The Tn pin is sampled once every system clock cycle by the pin syn-
chronization logic. The synchronized (sampled) signal is then passed through the edge
detector. Figure 59 shows a functional equivalent block diagram of the Tn synchroniza-
tion and edge detector logic. The registers are clocked at the positive edge of the
internal system clock (clkI/O). The latch is transparent in the high period of the internal
system clock.
The edge detector generates one clk /clk /clk 3 pulse for each positive (CSn2:0 = 7) or
2
T1
T
T
negative (CSn2:0 = 6) edge it detects.
Figure 59. Tn Pin Sampling
Tn_sync
(To Clock
Tn
D
Q
D
Q
D
Q
Select Logic)
LE
clkI/O
Edge Detector
The synchronization and edge detector logic introduces a delay of 2.5 to 3.5 system
clock cycles from an edge has been applied to the Tn pin to the counter is updated.
Enabling and disabling of the clock input must be done when Tn has been stable for at
least one system clock cycle, otherwise it is a risk that a false Timer/Counter clock pulse
is generated.
142
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Each half period of the external clock applied must be longer than one system clock
cycle to ensure correct sampling. The external clock must be guaranteed to have less
than half the system clock frequency (fExtClk < fclk_I/O/2) given a 50/50% duty cycle. Since
the edge detector uses sampling, the maximum frequency of an external clock it can
detect is half the sampling frequency (Nyquist sampling theorem). However, due to vari-
ation of the system clock frequency and duty cycle caused by Oscillator source (crystal,
resonator, and capacitors) tolerances, it is recommended that maximum frequency of an
external clock source is less than fclk_I/O/2.5.
An external clock source can not be prescaled.
Figure 60. Prescaler for Timer/Counter1, Timer/Counter2, and Timer/Counter3(1)
CK
10-BIT T/C PRESCALER
Clear
PSR321
T3
T2
T1
0
0
0
CS30
CS31
CS32
CS20
CS21
CS22
CS10
CS11
CS12
TIMER/COUNTER3 CLOCK SOURCE
TIMER/COUNTER2 CLOCK SOURCE
TIMER/COUNTER1 CLOCK SOURCE
clkT3
clkT2
clkT1
Note:
1. The synchronization logic on the input pins (T3/T2/T1) is shown in Figure 59.
Special Function IO Register –
SFIOR
Bit
7
6
–
R
0
5
–
R
0
4
–
R
0
3
ACME
R/W
0
2
1
PSR0
R/W
0
0
PSR321
R/W
0
TSM
R/W
0
PUD
R/W
0
SFIOR
Read/Write
Initial Value
• Bit 7 – TSM: Timer/Counter Synchronization Mode
Writing TSM bit to one activates the Timer/Counter Synchronization mode. In this mode,
the value that is written to PSR0 and PSR321 bits is kept, hence keeping the corre-
sponding prescaler reset signals asserted. This ensures that the corresponding
Timer/Counters are halted and can be configured to the same value without the risk of
one of them advancing during configuration. When the TSM bit written zero, the PSR0
and PSR321 bits are cleared by hardware, and the Timer/Counters start counting
simultaneously.
• Bit 0 – PSR321: Prescaler Reset Timer/Counter3, Timer/Counter2, and
Timer/Counter1
When this bit is one, the Timer/Counter3, Timer/Counter2, and Timer/Counter1 pres-
caler will be reset. The bit is normally cleared immediately by hardware, except if the
TSM bit is set. Note that Timer/Counter3 Timer/Counter2, and Timer/Counter1 share the
same prescaler and a reset of this prescaler will affect all three timers.
143
2490G–AVR–03/04
8-bit Timer/Counter2 Timer/Counter2 is a general purpose, single-channel, 8-bit Timer/Counter module. The
main features are:
• Single Channel Counter
with PWM
• Clear Timer on Compare Match (Auto Reload)
• Glitch-free, Phase Correct Pulse width Modulator (PWM)
• Frequency Generator
• External Event Counter
• 10-bit Clock Prescaler
• Overflow and Compare Match Interrupt Sources (TOV2 and OCF2)
Overview
A simplified block diagram of the 8-bit Timer/Counter is shown in Figure 61. For the
actual placement of I/O pins, refer to “Pin Configuration” on page 2. CPU accessible I/O
Registers, including I/O bits and I/O pins, are shown in bold. The device-specific I/O
Register and bit locations are listed in the “8-bit Timer/Counter Register Description” on
page 155.
Figure 61. 8-bit Timer/Counter Block Diagram
TCCRn
count
TOVn
(Int.Req.)
clear
Control Logic
TOP
Clock Select
direction
clk
Tn
Edge
Detector
Tn
BOTTOM
( From Prescaler )
Timer/Counter
TCNTn
= 0
= 0xFF
OCn
(Int.Req.)
Waveform
Generation
OCn
=
OCRn
Registers
The Timer/Counter (TCNT2) and Output Compare Register (OCR2) are 8-bit registers.
Interrupt request (abbreviated to Int.Req. in the figure) signals are all visible in the Timer
Interrupt Flag Register (TIFR). All interrupts are individually masked with the Timer
Interrupt Mask Register (TIMSK). TIFR and TIMSK are not shown in the figure since
these registers are shared by other timer units.
The Timer/Counter can be clocked internally, via the prescaler, or by an external clock
source on the T2 pin. The Clock Select logic block controls which clock source and edge
the Timer/Counter uses to increment (or decrement) its value. The Timer/Counter is
inactive when no clock source is selected. The output from the Clock Select logic is
referred to as the timer clock (clkT2).
144
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
The double buffered Output Compare Register (OCR2) is compared with the
Timer/Counter value at all times. The result of the compare can be used by the Wave-
form Generator to generate a PWM or variable frequency output on the Output Compare
pin (OC2). For details, see “Output Compare Unit” on page 146. The Compare Match
event will also set the Compare Flag (OCF2) which can be used to generate an Output
Compare interrupt request.
Definitions
Many register and bit references in this document are written in general form. A lower
case “n” replaces the Timer/Counter number, in this case 2. However, when using the
register or bit defines in a program, the precise form must be used (i.e., TCNT2 for
accessing Timer/Counter2 counter value and so on).
The definitions in Table 63 are also used extensively throughout this section.
Table 63. Definitions
BOTTOM The counter reaches the BOTTOM when it becomes 0x00.
MAX
TOP
The counter reaches its MAXimum when it becomes 0xFF (decimal 255).
The counter reaches the TOP when it becomes equal to the highest
value in the count sequence. The TOP value can be assigned to be the
fixed value 0xFF (MAX) or the value stored in the OCR2 Register. The
assignment is dependent on the mode of operation.
Timer/Counter Clock
Sources
The Timer/Counter can be clocked by an internal or an external clock source. The clock
source is selected by the Clock Select logic which is controlled by the Clock Select
(CS22:0) bits located in the Timer/Counter Control Register (TCCR2). For details on
clock sources and prescaler, see “Timer/Counter3, Timer/Counter2 and Timer/Counter1
Prescalers” on page 142.
145
2490G–AVR–03/04
Counter Unit
The main part of the 8-bit Timer/Counter is the programmable bi-directional counter unit.
Figure 62 shows a block diagram of the counter and its surroundings.
Figure 62. Counter Unit Block Diagram
TOVn
(Int.Req.)
DATA BUS
Clock Select
count
clear
Edge
Detector
Tn
clkTn
TCNTn
Control Logic
direction
( From Prescaler )
bottom
top
Signal description (internal signals):
count Increment or decrement TCNT2 by 1.
direction Select between increment and decrement.
clear
clkTn
top
Clear TCNT2 (set all bits to zero).
Timer/counter clock, referred to as clkT0 in the following.
Signalize that TCNT2 has reached maximum value.
Signalize that TCNT2 has reached minimum value (zero).
bottom
Depending of the mode of operation used, the counter is cleared, incremented, or dec-
remented at each timer clock (clkT2). clkT2 can be generated from an external or internal
clock source, selected by the Clock Select bits (CS22:0). When no clock source is
selected (CS22:0 = 0) the timer is stopped. However, the TCNT2 value can be accessed
by the CPU, regardless of whether clkT2 is present or not. A CPU write overrides (has
priority over) all counter clear or count operations.
The counting sequence is determined by the setting of the WGM01 and WGM00 bits
located in the Timer/Counter Control Register (TCCR2). There are close connections
between how the counter behaves (counts) and how waveforms are generated on the
Output Compare output OC2. For more details about advanced counting sequences
and waveform generation, see “Modes of Operation” on page 149.
The Timer/Counter Overflow Flag (TOV2) is set according to the mode of operation
selected by the WGM21:0 bits. TOV2 can be used for generating a CPU interrupt.
Output Compare Unit
The 8-bit comparator continuously compares TCNT2 with the Output Compare Register
(OCR2). Whenever TCNT2 equals OCR2, the comparator signals a match. A match will
set the Output Compare Flag (OCF2) at the next timer clock cycle. If enabled (OCIE2 =
1 and Global Interrupt Flag in SREG is set), the Output Compare Flag generates an
Output Compare interrupt. The OCF2 flag is automatically cleared when the interrupt is
executed. Alternatively, the OCF2 flag can be cleared by software by writing a logical
one to its I/O bit location. The Waveform Generator uses the match signal to generate
an output according to operating mode set by the WGM21:0 bits and Compare Output
mode (COM21:0) bits. The max and bottom signals are used by the Waveform Genera-
tor for handling the special cases of the extreme values in some modes of operation
146
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
(see “Modes of Operation” on page 149). Figure 63 shows a block diagram of the Output
Compare unit.
Figure 63. Output Compare Unit, Block Diagram
DATA BUS
OCRn
TCNTn
= (8-bit Comparator )
OCFn (Int.Req.)
top
bottom
FOCn
Waveform Generator
OCn
WGMn1:0
COMn1:0
The OCR2 Register is double buffered when using any of the Pulse Width Modulation
(PWM) modes. For the normal and Clear Timer on Compare (CTC) modes of operation,
the double buffering is disabled. The double buffering synchronizes the update of the
OCR2 Compare Register to either top or bottom of the counting sequence. The synchro-
nization prevents the occurrence of odd-length, non-symmetrical PWM pulses, thereby
making the output glitch-free.
The OCR2 Register access may seem complex, but this is not case. When the double
buffering is enabled, the CPU has access to the OCR2 Buffer Register, and if double
buffering is disabled the CPU will access the OCR2 directly.
Force Output Compare
In non-PWM waveform generation modes, the match output of the comparator can be
forced by writing a one to the Force Output Compare (FOC2) bit. Forcing Compare
Match will not set the OCF2 flag or reload/clear the timer, but the OC2 pin will be
updated as if a real Compare Match had occurred (the COM21:0 bits settings define
whether the OC2 pin is set, cleared or toggled).
Compare Match Blocking by
TCNT2 Write
All CPU write operations to the TCNT2 Register will block any Compare Match that
occur in the next timer clock cycle, even when the timer is stopped. This feature allows
OCR2 to be initialized to the same value as TCNT2 without triggering an interrupt when
the Timer/Counter clock is enabled.
147
2490G–AVR–03/04
Using the Output Compare
Unit
Since writing TCNT2 in any mode of operation will block all Compare Matches for one
timer clock cycle, there are risks involved when changing TCNT2 when using the Output
Compare channel, independently of whether the Timer/Counter is running or not. If the
value written to TCNT2 equals the OCR2 value, the Compare Match will be missed,
resulting in incorrect waveform generation. Similarly, do not write the TCNT2 value
equal to BOTTOM when the counter is downcounting.
The setup of the OC2 should be performed before setting the Data Direction Register for
the port pin to output. The easiest way of setting the OC2 value is to use the Force Out-
put Compare (FOC2) strobe bits in Normal mode. The OC2 Register keeps its value
even when changing between Waveform Generation modes.
Be aware that the COM21:0 bits are not double buffered together with the compare
value. Changing the COM21:0 bits will take effect immediately.
Compare Match Output
Unit
The Compare Output mode (COM21:0) bits have two functions. The Waveform Genera-
tor uses the COM21:0 bits for defining the Output Compare (OC2) state at the next
Compare Match. Also, the COM21:0 bits control the OC2 pin output source. Figure 64
shows a simplified schematic of the logic affected by the COM21:0 bit setting. The I/O
Registers, I/O bits, and I/O pins in the figure are shown in bold. Only the parts of the
general I/O port control registers (DDR and PORT) that are affected by the COM21:0
bits are shown. When referring to the OC2 state, the reference is for the internal OC2
Register, not the OC2 pin. If a System Reset occur, the OC2 Register is reset to “0”.
Figure 64. Compare Match Output Unit, Schematic
COMn1
Waveform
Generator
COMn0
FOCn
D
Q
Q
1
0
OCn
Pin
OCn
D
PORT
D
Q
DDR
clkI/O
The general I/O port function is overridden by the Output Compare (OC2) from the
Waveform Generator if either of the COM21:0 bits are set. However, the OC2 pin direc-
tion (input or output) is still controlled by the Data Direction Register (DDR) for the port
148
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
pin. The Data Direction Register bit for the OC2 pin (DDR_OC2) must be set as output
before the OC2 value is visible on the pin. The port override function is independent of
the Waveform Generation mode.
The design of the Output Compare pin logic allows initialization of the OC2 state before
the output is enabled. Note that some COM21:0 bit settings are reserved for certain
modes of operation. See “8-bit Timer/Counter Register Description” on page 155.
Compare Output Mode and
Waveform Generation
The Waveform Generator uses the COM21:0 bits differently in Normal, CTC, and PWM
modes. For all modes, setting the COM21:0 = 0 tells the Waveform Generator that no
action on the OC2 Register is to be performed on the next Compare Match. For com-
pare output actions in the non-PWM modes refer to Table 65 on page 156. For fast
PWM mode, refer to Table 66 on page 156, and for phase correct PWM refer to Table
67 on page 157.
A change of the COM21:0 bits state will have effect at the first Compare Match after the
bits are written. For non-PWM modes, the action can be forced to have immediate effect
by using the FOC2 strobe bits.
Modes of Operation
The mode of operation, i.e., the behavior of the Timer/Counter and the Output Compare
pins, is defined by the combination of the Waveform Generation mode (WGM21:0) and
Compare Output mode (COM21:0) bits. The Compare Output mode bits do not affect
the counting sequence, while the Waveform Generation mode bits do. The COM21:0
bits control whether the PWM output generated should be inverted or not (inverted or
non-inverted PWM). For non-PWM modes the COM21:0 bits control whether the output
should be set, cleared, or toggled at a Compare Match (see “Compare Match Output
Unit” on page 148).
For detailed timing information refer to Figure 68, Figure 69, Figure 70, and Figure 71 in
“Timer/Counter Timing Diagrams” on page 153.
Normal Mode
The simplest mode of operation is the Normal mode (WGM21:0 = 0). In this mode the
counting direction is always up (incrementing), and no counter clear is performed. The
counter simply overruns when it passes its maximum 8-bit value (TOP = 0xFF) and then
restarts from the bottom (0x00). In normal operation the Timer/Counter Overflow Flag
(TOV2) will be set in the same timer clock cycle as the TCNT2 becomes zero. The
TOV2 flag in this case behaves like a ninth bit, except that it is only set, not cleared.
However, combined with the timer overflow interrupt that automatically clears the TOV2
flag, the timer resolution can be increased by software. There are no special cases to
consider in the Normal mode, a new counter value can be written anytime.
The Output Compare unit can be used to generate interrupts at some given time. Using
the Output Compare to generate waveforms in Normal mode is not recommended,
since this will occupy too much of the CPU time.
Clear Timer on Compare
Match (CTC) Mode
In Clear Timer on Compare or CTC mode (WGM21:0 = 2), the OCR2 Register is used to
manipulate the counter resolution. In CTC mode the counter is cleared to zero when the
counter value (TCNT2) matches the OCR2. The OCR2 defines the top value for the
counter, hence also its resolution. This mode allows greater control of the Compare
Match output frequency. It also simplifies the operation of counting external events.
The timing diagram for the CTC mode is shown in Figure 65. The counter value
(TCNT2) increases until a Compare Match occurs between TCNT2 and OCR2, and then
counter (TCNT2) is cleared.
149
2490G–AVR–03/04
Figure 65. CTC Mode, Timing Diagram
OCn Interrupt Flag Set
TCNTn
OCn
(Toggle)
(COMn1:0 = 1)
1
2
3
4
Period
An interrupt can be generated each time the counter value reaches the TOP value by
using the OCF2 flag. If the interrupt is enabled, the interrupt handler routine can be used
for updating the TOP value. However, changing the TOP to a value close to BOTTOM
when the counter is running with none or a low prescaler value must be done with care
since the CTC mode does not have the double buffering feature. If the new value written
to OCR2 is lower than the current value of TCNT2, the counter will miss the Compare
Match. The counter will then have to count to its maximum value (0xFF) and wrap
around starting at 0x00 before the Compare Match can occur.
For generating a waveform output in CTC mode, the OC2 output can be set to toggle its
logical level on each Compare Match by setting the Compare Output mode bits to toggle
mode (COM21:0 = 1). The OC2 value will not be visible on the port pin unless the data
direction for the pin is set to output. The waveform generated will have a maximum fre-
quency of fOC2 = fclk_I/O/2 when OCR2 is set to zero (0x00). The waveform frequency is
defined by the following equation:
f
clk_I/O
f
= ----------------------------------------------
OCn
2 ⋅ N ⋅ (1 + OCRn)
The N variable represents the prescale factor (1, 8, 64, 256, or 1024).
As for the Normal mode of operation, the TOV2 flag is set in the same timer clock cycle
that the counter counts from MAX to 0x00.
Fast PWM Mode
The fast Pulse Width Modulation or fast PWM mode (WGM21:0 = 3) provides a high fre-
quency PWM waveform generation option. The fast PWM differs from the other PWM
option by its single-slope operation. The counter counts from BOTTOM to MAX then
restarts from BOTTOM. In non-inverting Compare Output mode, the Output Compare
(OC2) is cleared on the Compare Match between TCNT2 and OCR2, and set at BOT-
TOM. In inverting Compare Output mode, the output is set on Compare Match and
cleared at BOTTOM. Due to the single-slope operation, the operating frequency of the
fast PWM mode can be twice as high as the phase correct PWM mode that use dual-
slope operation. This high frequency makes the fast PWM mode well suited for power
regulation, rectification, and DAC applications. High frequency allows physically small
sized external components (coils, capacitors), and therefore reduces total system cost.
In fast PWM mode, the counter is incremented until the counter value matches the MAX
value. The counter is then cleared at the following timer clock cycle. The timing diagram
for the fast PWM mode is shown in Figure 66. The TCNT2 value is in the timing diagram
shown as a histogram for illustrating the single-slope operation. The diagram includes
150
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
non-inverted and inverted PWM outputs. The small horizontal line marks on the TCNT2
slopes represent Compare Matches between OCR2 and TCNT2.
Figure 66. Fast PWM Mode, Timing Diagram
OCRn Interrupt Flag Set
OCRn Update and
TOVn Interrupt Flag Set
TCNTn
(COMn1:0 = 2)
(COMn1:0 = 3)
OCn
OCn
1
2
3
4
5
6
7
Period
The Timer/Counter Overflow Flag (TOV2) is set each time the counter reaches MAX. If
the interrupt is enabled, the interrupt handler routine can be used for updating the com-
pare value.
In fast PWM mode, the compare unit allows generation of PWM waveforms on the OC2
pin. Setting the COM21:0 bits to two will produce a non-inverted PWM and an inverted
PWM output can be generated by setting the COM21:0 to three (see Table 66 on page
156). The actual OC2 value will only be visible on the port pin if the data direction for the
port pin is set as output. The PWM waveform is generated by setting (or clearing) the
OC2 Register at the Compare Match between OCR2 and TCNT2, and clearing (or set-
ting) the OC2 Register at the timer clock cycle the counter is cleared (changes from
MAX to BOTTOM).
The PWM frequency for the output can be calculated by the following equation:
f
clk_I/O
f
= -----------------
OCnPWM
N ⋅ 256
The N variable represents the prescale factor (1, 8, 64, 256, or 1024).
The extreme values for the OCR2 Register represents special cases when generating a
PWM waveform output in the fast PWM mode. If the OCR2 is set equal to BOTTOM, the
output will be a narrow spike for each MAX+1 timer clock cycle. Setting the OCR2 equal
to MAX will result in a constantly high or low output (depending on the polarity of the out-
put set by the COM21:0 bits.)
A frequency (with 50% duty cycle) waveform output in fast PWM mode can be achieved
by setting OC2 to toggle its logical level on each Compare Match (COM21:0 = 1). The
waveform generated will have a maximum frequency of fOC2 = fclk_I/O/2 when OCR2 is
set to zero. This feature is similar to the OC2 toggle in CTC mode, except the double
buffer feature of the Output Compare unit is enabled in the fast PWM mode.
151
2490G–AVR–03/04
Phase Correct PWM Mode
The phase correct PWM mode (WGM21:0 = 1) provides a high resolution phase correct
PWM waveform generation option. The phase correct PWM mode is based on a dual-
slope operation. The counter counts repeatedly from BOTTOM to MAX and then from
MAX to BOTTOM. In non-inverting Compare Output mode, the Output Compare (OC2)
is cleared on the Compare Match between TCNT2 and OCR2 while upcounting, and set
on the Compare Match while downcounting. In inverting Output Compare mode, the
operation is inverted. The dual-slope operation has lower maximum operation frequency
than single slope operation. However, due to the symmetric feature of the dual-slope
PWM modes, these modes are preferred for motor control applications.
The PWM resolution for the phase correct PWM mode is fixed to eight bits. In phase
correct PWM mode the counter is incremented until the counter value matches MAX.
When the counter reaches MAX, it changes the count direction. The TCNT2 value will
be equal to MAX for one timer clock cycle. The timing diagram for the phase correct
PWM mode is shown on Figure 67. The TCNT2 value is in the timing diagram shown as
a histogram for illustrating the dual-slope operation. The diagram includes non-inverted
and inverted PWM outputs. The small horizontal line marks on the TCNT2 slopes repre-
sent Compare Matches between OCR2 and TCNT2.
Figure 67. Phase Correct PWM Mode, Timing Diagram
OCn Interrupt
Flag Set
OCRn Update
TOVn Interrupt
Flag Set
TCNTn
(COMn1:0 = 2)
OCn
(COMn1:0 = 3)
OCn
1
2
3
Period
The Timer/Counter Overflow Flag (TOV2) is set each time the counter reaches BOT-
TOM. The interrupt flag can be used to generate an interrupt each time the counter
reaches the BOTTOM value.
In phase correct PWM mode, the compare unit allows generation of PWM waveforms on
the OC2 pin. Setting the COM21:0 bits to two will produce a non-inverted PWM. An
inverted PWM output can be generated by setting the COM21:0 to three (see Table 67
on page 157). The actual OC2 value will only be visible on the port pin if the data direc-
tion for the port pin is set as output. The PWM waveform is generated by clearing (or
setting) the OC2 Register at the Compare Match between OCR2 and TCNT2 when the
counter increments, and setting (or clearing) the OC2 Register at Compare Match
152
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
between OCR2 and TCNT2 when the counter decrements. The PWM frequency for the
output when using phase correct PWM can be calculated by the following equation:
f
clk_I/O
f
= -----------------
OCnPCPWM
N ⋅ 510
The N variable represents the prescale factor (1, 8, 64, 256, or 1024).
The extreme values for the OCR2 Register represent special cases when generating a
PWM waveform output in the phase correct PWM mode. If the OCR2 is set equal to
BOTTOM, the output will be continuously low and if set equal to MAX the output will be
continuously high for non-inverted PWM mode. For inverted PWM the output will have
the opposite logic values.
At the very start of period 2 in Figure 67 OCn has a transition from high to low even
though there is no Compare Match. The point of this transition is to guarantee symmetry
around BOTTOM. There are two cases that give a transition without a Compare Match.
•
OCR2 changes its value from MAX, like in Figure 67. When the OCR2 value is MAX
the OCn pin value is the same as the result of a down-counting Compare Match. To
ensure symmetry around BOTTOM the OCn value at MAX must correspond to the
result of an up-counting Compare Match.
•
The timer starts counting from a higher value than the one in OCR2, and for that
reason misses the Compare Match and hence the OCn change that would have
happened on the way up.
Timer/Counter Timing
Diagrams
The Timer/Counter is a synchronous design and the timer clock (clkT2) is therefore
shown as a clock enable signal in the following figures. The figures include information
on when interrupt flags are set. Figure 68 contains timing data for basic Timer/Counter
operation. The figure shows the count sequence close to the MAX value in all modes
other than phase correct PWM mode.
Figure 68. Timer/Counter Timing Diagram, no Prescaling
clkI/O
clkTn
(clkI/O/1)
TCNTn
TOVn
MAX - 1
MAX
BOTTOM
BOTTOM + 1
Figure 69 shows the same timing data, but with the prescaler enabled.
153
2490G–AVR–03/04
Figure 69. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O/8)
TCNTn
TOVn
MAX - 1
MAX
BOTTOM
BOTTOM + 1
Figure 70 shows the setting of OCF2 in all modes except CTC mode.
Figure 70. Timer/Counter Timing Diagram, Setting of OCF2, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O/8)
TCNTn
OCRn
OCFn
OCRn - 1
OCRn
OCRn + 1
OCRn + 2
OCRn Value
Figure 71 shows the setting of OCF2 and the clearing of TCNT2 in CTC mode.
154
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 71. Timer/Counter Timing Diagram, Clear Timer on Compare Match Mode, with
Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O/8)
TCNTn
(CTC)
TOP - 1
TOP
BOTTOM
BOTTOM + 1
OCRn
TOP
OCFn
8-bit Timer/Counter
Register Description
Timer/Counter Control
Register – TCCR2
Bit
7
FOC2
W
6
5
COM21
R/W
0
4
COM20
R/W
0
3
WGM21
R/W
0
2
1
CS21
R/W
0
0
WGM20
R/W
0
CS22
R/W
0
CS20
R/W
0
TCCR2
Read/Write
Initial Value
0
• Bit 7 – FOC2: Force Output Compare
The FOC2 bit is only active when the WGM20 bit specifies a non-PWM mode. However,
for ensuring compatibility with future devices, this bit must be set to zero when TCCR2 is
written when operating in PWM mode. When writing a logical one to the FOC2 bit, an
immediate Compare Match is forced on the waveform generation unit. The OC2 output
is changed according to its COM21:0 bits setting. Note that the FOC2 bit is implemented
as a strobe. Therefore it is the value present in the COM21:0 bits that determines the
effect of the forced compare.
A FOC2 strobe will not generate any interrupt, nor will it clear the timer in CTC mode
using OCR2 as TOP.
The FOC2 bit is always read as zero.
• Bit 6, 3 – WGM21:0: Waveform Generation Mode
These bits control the counting sequence of the counter, the source for the maximum
(TOP) counter value, and what type of waveform generation to be used. Modes of oper-
ation supported by the Timer/Counter unit are: Normal mode, Clear Timer on Compare
match (CTC) mode, and two types of Pulse Width Modulation (PWM) modes. See Table
64 and “Modes of Operation” on page 149.
155
2490G–AVR–03/04
Table 64. Waveform Generation Mode Bit Description(1)
WGM21
(CTC2)
WGM20 Timer/Counter Mode
(PWM2) of Operation
Update of
OCR2
TOV2 Flag
Set on
Mode
TOP
0xFF
0xFF
0
1
2
3
0
0
1
1
0
1
0
1
Normal
Immediate
TOP
MAX
PWM, Phase Correct
CTC
BOTTOM
MAX
OCR2 Immediate
0xFF TOP
Fast PWM
MAX
Note:
1. The CTC2 and PWM2 bit definition names are now obsolete. Use the WGM21:0 def-
initions. However, the functionality and location of these bits are compatible with
previous versions of the timer.
• Bit 5:4 – COM21:0: Compare Match Output Mode
These bits control the Output Compare pin (OC2) behavior. If one or both of the
COM21:0 bits are set, the OC2 output overrides the normal port functionality of the I/O
pin it is connected to. However, note that the Data Direction Register (DDR) bit corre-
sponding to the OC2 pin must be set in order to enable the output driver.
When OC2 is connected to the pin, the function of the COM21:0 bits depends on the
WGM21:0 bit setting. Table 65 shows the COM21:0 bit functionality when the WGM21:0
bits are set to a Normal or CTC mode (non-PWM).
Table 65. Compare Output Mode, non-PWM Mode
COM21
COM20
Description
0
0
1
1
0
1
0
1
Normal port operation, OC2 disconnected.
Toggle OC2 on Compare Match.
Clear OC2 on Compare Match.
Set OC2 on Compare Match.
Table 66 shows the COM21:0 bit functionality when the WGM21:0 bits are set to fast
PWM mode.
Table 66. Compare Output Mode, Fast PWM Mode(1)
COM21
COM20
Description
0
0
1
1
0
1
0
1
Normal port operation, OC2 disconnected.
Reserved
Clear OC2 on Compare Match, set OC2 at TOP.
Set OC2 on Compare Match, clear OC2 at TOP.
Note:
1. A special case occurs when OCR2 equals TOP and COM21 is set. In this case, the
Compare Match is ignored, but the set or clear is done at TOP. See “Fast PWM
Mode” on page 150 for more details.
Table 67 shows the COM21:0 bit functionality when the WGM21:0 bits are set to phase
correct PWM mode.
156
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 67. Compare Output Mode, Phase Correct PWM Mode(1)
COM21 COM20 Description
0
0
1
0
1
0
Normal port operation, OC2 disconnected.
Reserved
Clear OC2 on Compare Match when up-counting. Set OC2 on Compare
Match when downcounting.
1
1
Set OC2 on Compare Match when up-counting. Clear OC2 on Compare
Match when downcounting.
Note:
1. A special case occurs when OCR2 equals TOP and COM21 is set. In this case, the
Compare Match is ignored, but the set or clear is done at TOP. See “Phase Correct
PWM Mode” on page 152 for more details.
• Bit 2:0 – CS22:0: Clock Select
The three Clock Select bits select the clock source to be used by the Timer/Counter.
Table 68. Clock Select Bit Description
CS22
CS21
CS20
Description
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
No clock source (Timer/counter stopped).
clkI/O/(No prescaling)
clkI/O/8 (From prescaler)
clkI/O/64 (From prescaler)
clkI/O/256 (From prescaler)
clkI/O/1024 (From prescaler)
External clock source on T2 pin. Clock on falling edge.
External clock source on T2 pin. Clock on rising edge.
If external pin modes are used for the Timer/Counter2, transitions on the T2 pin will
clock the counter even if the pin is configured as an output. This feature allows software
control of the counting.
Timer/Counter Register –
TCNT2
Bit
7
6
5
4
3
2
1
0
TCNT2[7:0]
TCNT2
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
The Timer/Counter Register gives direct access, both for read and write operations, to
the Timer/Counter unit 8-bit counter. Writing to the TCNT2 Register blocks (removes)
the Compare Match on the following timer clock. Modifying the counter (TCNT2) while
the counter is running, introduces a risk of missing a Compare Match between TCNT2
and the OCR2 Register.
157
2490G–AVR–03/04
Output Compare Register –
OCR2
Bit
7
6
5
4
3
2
1
0
OCR2[7:0]
OCR2
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
The Output Compare Register contains an 8-bit value that is continuously compared
with the counter value (TCNT2). A match can be used to generate an Output Compare
interrupt, or to generate a waveform output on the OC2 pin.
Timer/Counter Interrupt Mask
Register – TIMSK
Bit
7
OCIE2
R/W
0
6
TOIE2
R/W
0
5
TICIE1
R/W
0
4
OCIE1A
R/W
0
3
OCIE1B
R/W
0
2
TOIE1
R/W
0
1
OCIE0
R/W
0
0
TOIE0
R/W
0
TIMSK
Read/Write
Initial Value
• Bit 7 – OCIE2: Timer/Counter2 Output Compare Match Interrupt Enable
When the OCIE2 bit is written to one, and the I-bit in the Status Register is set (one), the
Timer/Counter2 Compare Match Interrupt is enabled. The corresponding interrupt is
executed if a Compare Match in Timer/Counter2 occurs, for example, 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 written to one, and the I-bit in the Status Register is set (one), the
Timer/Counter2 Overflow Interrupt is enabled. The corresponding interrupt is executed if
an overflow in Timer/Counter2 occurs, for example, when the TOV2 bit is set in the
Timer/Counter Interrupt Flag Register – TIFR.
Timer/Counter Interrupt Flag
Register – TIFR
Bit
7
OCF2
R/W
0
6
TOV2
R/W
0
5
4
OCF1A
R/W
0
3
OCF1B
R/W
0
2
TOV1
R/W
0
1
OCF0
R/W
0
0
TOV0
R/W
0
ICF1
R/W
0
TIFR
Read/Write
Initial Value
• Bit 7 – OCF2: Output Compare Flag 2
The OCF2 bit is set (one) when a Compare Match occurs between the Timer/Counter2
and the data in OCR2 – Output Compare Register2. OCF2 is cleared by hardware when
executing the corresponding interrupt handling vector. Alternatively, OCF2 is cleared by
writing a logic one to the flag. When the I-bit in SREG, OCIE2 (Timer/Counter2 Com-
pare Match Interrupt Enable), and OCF2 are set (one), the Timer/Counter2 Compare
match Interrupt is executed.
• Bit 6 – TOV2: Timer/Counter2 Overflow Flag
The bit TOV2 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 logic one to the flag. When the SREG I-bit, TOIE2
(Timer/Counter2 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 changes counting direction at 0x00.
158
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Output Compare
Modulator (OCM1C2)
Overview
The Output Compare Modulator (OCM) allows generation of waveforms modulated with
a carrier frequency. The modulator uses the outputs from the Output Compare Unit C of
the 16-bit Timer/Counter1 and the Output Compare Unit of the 8-bit Timer/Counter2. For
more details about these Timer/Counters see “16-bit Timer/Counter (Timer/Counter1
and Timer/Counter3)” on page 110 and “8-bit Timer/Counter2 with PWM” on page 144.
Note that this feature is not available in ATmega103 compatibility mode.
Figure 72. Output Compare Modulator, Block Diagram
OC1C
OC2
Timer/Counter1
Timer/Counter2
Pin
OC1C/
OC2/PB7
When the modulator is enabled, the two Output Compare channels are modulated
together as shown in the block diagram (Figure 72).
Description
The Output Compare unit 1C and Output Compare unit 2 shares the PB7 port pin for
output. The outputs of the Output Compare units (OC1C and OC2) overrides the normal
PORTB7 Register when one of them is enabled (i.e., when COMnx1:0 is not equal to
zero). When both OC1C and OC2 are enabled at the same time, the modulator is auto-
matically enabled.
The functional equivalent schematic of the modulator is shown on Figure 73. The sche-
matic includes part of the Timer/Counter units and the Port B pin 7 output driver circuit.
Figure 73. Output Compare Modulator, Schematic
COM21
Vcc
COM20
COM1C1
COM1C0
Modulator
0
1
( From Waveform Generator )
D
Q
1
0
OC1C
Pin
OC1C /
OC2 / PB7
( From Waveform Generator )
D
Q
Q
OC2
D
D
Q
PORTB7
DDRB7
DATA BUS
159
2490G–AVR–03/04
When the modulator is enabled the type of modulation (logical AND or OR) can be
selected by the PORTB7 Register. Note that the DDRB7 controls the direction of the
port independent of the COMnx1:0 bit setting.
Timing Example
Figure 74 illustrates the modulator in action. In this example the Timer/Counter1 is set to
operate in fast PWM mode (non-inverted) and Timer/Counter2 uses CTC waveform
mode with toggle Compare Output mode (COMnx1:0 = 1).
Figure 74. Output Compare Modulator, Timing Diagram
clkI/O
OC1C
(FPWM Mode)
OC2
(CTC Mode)
PB7
(PORTB7 = 0)
PB7
(PORTB7 = 1)
1
2
3
(Period)
In this example, Timer/Counter2 provides the carrier, while the modulating signal is gen-
erated by the Output Compare unit C of the Timer/Counter1.
The resolution of the PWM signal (OC1C) is reduced by the modulation. The reduction
factor is equal to the number of system clock cycles of one period of the carrier (OC2).
In this example the resolution is reduced by a factor of two. The reason for the reduction
is illustrated in Figure 74 at the second and third period of the PB7 output when
PORTB7 equals zero. The period 2 high time is one cycle longer than the period three
high time, but the result on the PB7 output is equal in both periods.
160
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Serial Peripheral
Interface – SPI
The Serial Peripheral Interface (SPI) allows high-speed synchronous data transfer
between the ATmega64 and peripheral devices or between several AVR devices. The
ATmega64 SPI includes the following features:
• Full-duplex, Three-wire Synchronous Data Transfer
• Master or Slave Operation
• LSB First or MSB First Data Transfer
• Seven Programmable Bit Rates
• End of Transmission Interrupt Flag
• Write Collision Flag Protection
• Wake-up from Idle Mode
• Double Speed (CK/2) Master SPI Mode
Figure 75. SPI Block Diagram(1)
DIVIDER
/2/4/8/16/32/64/128
Note:
1. Refer to Figure 1 on page 2, and Table 30 on page 72 for SPI pin placement.
The interconnection between Master and Slave CPUs with SPI is shown in Figure 76.
The system consists of two Shift Registers, and a Master clock generator. The SPI Mas-
ter initiates the communication cycle when pulling low the Slave Select SS pin of the
desired Slave. Master and Slave prepare the data to be sent in their respective Shift
Registers, and the Master generates the required clock pulses on the SCK line to inter-
change data. Data is always shifted from Master to Slave on the Master Out – Slave In,
MOSI, line, and from Slave to Master on the Master In – Slave Out, MISO, line. After
each data packet, the Master will synchronize the Slave by pulling high the Slave Select,
SS, line.
When configured as a Master, the SPI interface has no automatic control of the SS line.
This must be handled by user software before communication can start. When this is
161
2490G–AVR–03/04
done, writing a byte to the SPI Data Register starts the SPI clock generator, and the
hardware shifts the eight bits into the Slave. After shifting one byte, the SPI clock gener-
ator 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 Master may continue
to shift the next byte by writing it into SPDR, or signal the end of packet by pulling high
the Slave Select, SS line. The last incoming byte will be kept in the buffer register for
later use.
When configured as a Slave, the SPI interface will remain sleeping with MISO tri-stated
as long as the SS pin is driven high. In this state, software may update the contents of
the SPI Data Register, SPDR, but the data will not be shifted out by incoming clock
pulses on the SCK pin until the SS pin is driven low. As one byte has been completely
shifted, the end of transmission flag, SPIF is set. If the SPI Interrupt Enable bit, SPIE, in
the SPCR Register is set, an interrupt is requested. The Slave may continue to place
new data to be sent into SPDR before reading the incoming data. The last incoming byte
will be kept in the buffer register for later use.
Figure 76. SPI Master-Slave Interconnection
MSB
MASTER
LSB
MSB
SLAVE
LSB
MISO
MOSI
MISO
MOSI
8 BIT SHIFT REGISTER
8 BIT SHIFT REGISTER
SHIFT
ENABLE
SPI
SCK
SS
SCK
SS
CLOCK GENERATOR
V
CC
The system is single buffered in the transmit direction and double buffered in the receive
direction. This means that bytes to be transmitted cannot be written to the SPI Data
Register before the entire shift cycle is completed. When receiving data, however, a
received character must be read from the SPI Data Register before the next character
has been completely shifted in. Otherwise, the first byte is lost.
In SPI Slave mode, the control logic will sample the incoming signal of the SCK pin. To
ensure correct sampling of the clock signal, the frequency of the SPI clock should never
exceed fosc/4.
When the SPI is enabled, the data direction of the MOSI, MISO, SCK, and SS pins is
overridden according to Table 69. For more details on automatic port overrides, refer to
“Alternate Port Functions” on page 69.
Table 69. SPI Pin Overrides(1)
Pin
MOSI
MISO
SCK
SS
Direction, Master SPI
User Defined
Input
Direction, Slave SPI
Input
User Defined
Input
User Defined
User Defined
Input
Note:
1. See “Alternate Functions of Port B” on page 72 for a detailed description of how to
define the direction of the user defined SPI pins.
162
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
The following code examples show how to initialize the SPI as a Master and how to per-
form a simple transmission. DDR_SPI in the examples must be replaced by the actual
Data Direction Register controlling the SPI pins. DD_MOSI, DD_MISO and DD_SCK
must be replaced by the actual data direction bits for these pins. For example, if MOSI is
placed on pin PB5, replace DD_MOSI with DDB5 and DDR_SPI with DDRB.
Assembly Code Example(1)
SPI_MasterInit:
; Set MOSI and SCK output, all others input
ldi
out
r17,(1<<DD_MOSI)|(1<<DD_SCK)
DDR_SPI,r17
; Enable SPI, Master, set clock rate fck/16
ldi
out
ret
r17,(1<<SPE)|(1<<MSTR)|(1<<SPR0)
SPCR,r17
SPI_MasterTransmit:
; Start transmission of data (r16)
out
SPDR,r16
Wait_Transmit:
; Wait for transmission complete
sbis SPSR,SPIF
rjmp Wait_Transmit
ret
C Code Example(1)
void SPI_MasterInit(void)
{
/* Set MOSI and SCK output, all others input */
DDR_SPI = (1<<DD_MOSI)|(1<<DD_SCK);
/* Enable SPI, Master, set clock rate fck/16 */
SPCR = (1<<SPE)|(1<<MSTR)|(1<<SPR0);
}
void SPI_MasterTransmit(char cData)
{
/* Start transmission */
SPDR = cData;
/* Wait for transmission complete */
while(!(SPSR & (1<<SPIF)))
;
}
Note:
1. The example code assumes that the part specific header file is included.
163
2490G–AVR–03/04
The following code examples show how to initialize the SPI as a Slave and how to per-
form a simple reception.
Assembly Code Example(1)
SPI_SlaveInit:
; Set MISO output, all others input
ldi
out
r17,(1<<DD_MISO)
DDR_SPI,r17
; Enable SPI
ldi
out
ret
r17,(1<<SPE)
SPCR,r17
SPI_SlaveReceive:
; Wait for reception complete
sbis SPSR,SPIF
rjmp SPI_SlaveReceive
; Read received data and return
in
r16,SPDR
ret
C Code Example(1)
void SPI_SlaveInit(void)
{
/* Set MISO output, all others input */
DDR_SPI = (1<<DD_MISO);
/* Enable SPI */
SPCR = (1<<SPE);
}
char SPI_SlaveReceive(void)
{
/* Wait for reception complete */
while(!(SPSR & (1<<SPIF)))
;
/* Return data register */
return SPDR;
}
Note:
1. The example code assumes that the part specific header file is included.
164
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
SS Pin Functionality
Slave Mode
When the SPI is configured as a Slave, the Slave Select (SS) pin is always input. When
SS is held low, the SPI is activated, and MISO becomes an output if configured so by
the user. All other pins are inputs. When SS is driven high, all pins are inputs, and the
SPI is passive, which means that it will not receive incoming data. Note that the SPI
logic will be reset once the SS pin is driven high.
The SS pin is useful for packet/byte synchronization to keep the slave bit counter syn-
chronous with the Master clock generator. When the SS pin is driven high, the SPI Slave
will immediately reset the send and receive logic, and drop any partially received data in
the Shift Register.
Master Mode
When the SPI is configured as a Master (MSTR in SPCR is set), the user can determine
the direction of the SS pin.
If SS is configured as an output, the pin is a general output pin which does not affect the
SPI system. Typically, the pin will be driving the SS pin of the SPI Slave.
If SS is configured as an input, it must be held high to ensure Master SPI operation. If
the SS pin is driven low by peripheral circuitry when the SPI is configured as a Master
with the SS pin defined as an input, the SPI system interprets this as another Master
selecting the SPI as a Slave and starting 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.
Thus, when interrupt-driven SPI transmission is used in Master mode, and there exists a
possibility that SS is driven low, the interrupt should always check that the MSTR bit is
still set. If the MSTR bit has been cleared by a slave select, it must be set by the user to
re-enable SPI Master mode.
SPI Control Register – SPCR
Bit
7
SPIE
R/W
0
6
5
DORD
R/W
0
4
MSTR
R/W
0
3
CPOL
R/W
0
2
CPHA
R/W
0
1
SPR1
R/W
0
0
SPR0
R/W
0
SPE
R/W
0
SPCR
Read/Write
Initial Value
• 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 if the Global Interrupt Enable bit in SREG is set.
• Bit 6 – SPE: SPI Enable
When the SPE bit is written to one, the SPI is enabled. This bit must be set to enable
any SPI operations.
• Bit 5 – DORD: Data Order
When the DORD bit is written to one, the LSB of the data word is transmitted first.
When the DORD bit is written to zero, the MSB of the data word is transmitted first.
165
2490G–AVR–03/04
• Bit 4 – MSTR: Master/Slave Select
This bit selects Master SPI mode when written to one, and Slave SPI mode when written
logic 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
re-enable SPI Master mode.
• Bit 3 – CPOL: Clock Polarity
When this bit is written to one, SCK is high when idle. When CPOL is written to zero,
SCK is low when idle. Refer to Figure 77 and Figure 78 for an example. The CPOL func-
tionality is summarized below:
Table 70. CPOL Functionality
CPOL
Leading Edge
Rising
Trailing Edge
Falling
0
1
Falling
Rising
• Bit 2 – CPHA: Clock Phase
The settings of the Clock Phase bit (CPHA) determine if data is sampled on the leading
(first) or trailing (last) edge of SCK. Refer to Figure 77 and Figure 78 for an example.
The CPHA functionality is summarized below:
Table 71. CPHA Functionality
CPHA
Leading Edge
Sample
Trailing Edge
Setup
0
1
Setup
Sample
• Bits 1, 0 – SPR1, SPR0: SPI Clock Rate Select 1 and 0
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 Oscillator
Clock frequency fosc is shown in Table 72.
Table 72. Relationship Between SCK and the Oscillator Frequency
SPI2X
SPR1
SPR0
SCK Frequency
fosc/4
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
fosc/16
fosc/64
fosc/128
fosc/2
fosc/8
fosc/32
fosc/64
166
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
SPI Status Register – SPSR
Bit
7
SPIF
R
6
5
–
R
0
4
–
R
0
3
–
R
0
2
–
R
0
1
–
R
0
0
SPI2X
R/W
0
WCOL
R
0
SPSR
Read/Write
Initial Value
0
• Bit 7 – SPIF: SPI Interrupt Flag
When a serial transfer is complete, the SPIF flag is set. An interrupt is generated if SPIE
in SPCR is set and global interrupts are enabled. If SS is an input and is driven low
when the SPI is in Master mode, this will also set the SPIF flag. SPIF is cleared by hard-
ware when executing the corresponding interrupt handling vector. Alternatively, the
SPIF bit is cleared by first reading the SPI Status Register with SPIF set, 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 by first reading the SPI Status Register
with WCOL set, and then accessing the SPI Data Register.
• Bit 5..1 – Res: Reserved Bits
These bits are reserved bits in the ATmega64 and will always read as zero.
• Bit 0 – SPI2X: Double SPI Speed Bit
When this bit is written logic one the SPI speed (SCK Frequency) will be doubled when
the SPI is in Master mode (see Table 72). This means that the minimum SCK period will
be two CPU clock periods. When the SPI is configured as Slave, the SPI is only guaran-
teed to work at fosc/4 or lower.
The SPI interface on the ATmega64 is also used for program memory and EEPROM
downloading or uploading. See page 306 for SPI Serial Programming and verification.
SPI Data Register – SPDR
Bit
7
6
5
4
3
2
1
0
MSB
R/W
X
LSB
R/W
X
SPDR
Read/Write
Initial Value
R/W
X
R/W
X
R/W
X
R/W
X
R/W
X
R/W
X
Undefined
The SPI Data Register is a read/write register used for data transfer between the Regis-
ter 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.
167
2490G–AVR–03/04
Data Modes
There are four combinations of SCK phase and polarity with respect to serial data,
which are determined by control bits CPHA and CPOL. The SPI data transfer formats
are shown in Figure 77 and Figure 78. Data bits are shifted out and latched in on oppo-
site edges of the SCK signal, ensuring sufficient time for data signals to stabilize. This is
clearly seen by summarizing Table 70 and Table 71, as done below:
Table 73. CPOL and CPHA Functionality
Leading Edge
Sample (Rising)
Setup (Rising)
Sample (Falling)
Setup (Falling)
Trailing Edge
Setup (Falling)
Sample (Falling)
Setup (Rising)
Sample (Rising)
SPI Mode
CPOL = 0, CPHA = 0
CPOL = 0, CPHA = 1
CPOL = 1, CPHA = 0
CPOL = 1, CPHA = 1
0
1
2
3
Figure 77. SPI Transfer Format with CPHA = 0
SCK (CPOL = 0)
mode 0
SCK (CPOL = 1)
mode 2
SAMPLE I
MOSI/MISO
CHANGE 0
MOSI PIN
CHANGE 0
MISO PIN
SS
MSB first (DORD = 0) MSB
LSB first (DORD = 1) LSB
Bit 6
Bit 1
Bit 5
Bit 2
Bit 4
Bit 3
Bit 3
Bit 4
Bit 2
Bit 5
Bit 1
Bit 6
LSB
MSB
Figure 78. SPI Transfer Format with CPHA = 1
SCK (CPOL = 0)
mode 1
SCK (CPOL = 1)
mode 3
SAMPLE I
MOSI/MISO
CHANGE 0
MOSI PIN
CHANGE 0
MISO PIN
SS
MSB first (DORD = 0)
LSB first (DORD = 1)
MSB
LSB
Bit 6
Bit 1
Bit 5
Bit 2
Bit 4
Bit 3
Bit 3
Bit 4
Bit 2
Bit 5
Bit 1
Bit 6
LSB
MSB
168
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
USART
The Universal Synchronous and Asynchronous serial Receiver and Transmitter
(USART) is a highly flexible serial communication device. The main features are:
• Full Duplex Operation (Independent Serial Receive and Transmit Registers)
• Asynchronous or Synchronous Operation
• Master or Slave Clocked Synchronous Operation
• High Resolution Baud Rate Generator
• Supports Serial Frames with 5, 6, 7, 8, or 9 Data Bits and 1 or 2 Stop Bits
• Odd or Even Parity Generation and Parity Check Supported by Hardware
• Data OverRun Detection
• Framing Error Detection
• Noise Filtering Includes False Start Bit Detection and Digital Low Pass Filter
• Three Separate Interrupts on TX Complete, TX Data Register Empty and RX Complete
• Multi-processor Communication Mode
• Double Speed Asynchronous Communication Mode
Dual USART
The ATmega64 has two USART’s, USART0 and USART1. The functionality for both
USART’s is described below. USART0 and USART1 have different I/O Registers as
shown in “Register Summary” on page 342. Note that in ATmega103 compatibility
mode, USART1 is not available, neither is the UBRR0H or UCRS0C registers. This
means that in ATmega103 compatibility mode, the ATmega64 supports asynchronous
operation of USART0 only.
Overview
A simplified block diagram of the USART Transmitter is shown in Figure 79. CPU acces-
sible I/O Registers and I/O pins are shown in bold.
Figure 79. USART Block Diagram(1)
Clock Generator
UBRR[H:L]
OSC
BAUD RATE GENERATOR
SYNC LOGIC
PIN
XCK
CONTROL
Transmitter
TX
UDR (Transmit)
CONTROL
PARITY
GENERATOR
PIN
CONTROL
TRANSMIT SHIFT REGISTER
TxD
Receiver
CLOCK
RX
RECOVERY
CONTROL
DATA
RECOVERY
PIN
CONTROL
RECEIVE SHIFT REGISTER
UDR (Receive)
RxD
PARITY
CHECKER
UCSRA
UCSRB
UCSRC
Note:
1. Refer to Figure 1 on page 2, Table 36 on page 76, and Table 39 on page 79 for
USART pin placement.
169
2490G–AVR–03/04
The dashed boxes in the block diagram separate the three main parts of the USART
(listed from the top): Clock generator, Transmitter and Receiver. Control registers are
shared by all units. The Clock Generation logic consists of synchronization logic for
external clock input used by synchronous slave operation, and the baud rate generator.
The XCK (Transfer Clock) pin is only used by synchronous transfer mode. The Trans-
mitter consists of a single write buffer, a serial Shift Register, Parity Generator and
Control Logic for handling different serial frame formats. The write buffer allows a contin-
uous transfer of data without any delay between frames. The Receiver is the most
complex part of the USART module due to its clock and data recovery units. The recov-
ery units are used for asynchronous data reception. In addition to the recovery units, the
Receiver includes a Parity Checker, Control Logic, a Shift Register and a two level
receive buffer (UDR). The Receiver supports the same frame formats as the Transmit-
ter, and can detect Frame Error, Data OverRun and Parity Errors.
AVR USART vs. AVR UART –
Compatibility
The USART is fully compatible with the AVR UART regarding:
•
•
•
•
•
Bit locations inside all USART Registers
Baud Rate Generation.
Transmitter Operation.
Transmit Buffer Functionality.
Receiver Operation.
However, the receive buffering has two improvements that will affect the compatibility in
some special cases:
•
A second buffer register has been added. The two buffer registers operate as a
circular FIFO buffer. Therefore the UDR must only be read once for each incoming
data! More important is the fact that the error flags (FE and DOR) and the ninth data
bit (RXB8) are buffered with the data in the receive buffer. Therefore the status bits
must always be read before the UDR Register is read. Otherwise the error status
will be lost since the buffer state is lost.
•
The Receiver Shift Register can now act as a third buffer level. This is done by
allowing the received data to remain in the serial Shift Register (see Figure 79) if the
buffer registers are full, until a new start bit is detected. The USART is therefore
more resistant to Data OverRun (DOR) error conditions.
The following control bits have changed name, but have same functionality and register
location:
•
•
CHR9 is changed to UCSZ2.
OR is changed to DOR.
Clock Generation
The Clock Generation logic generates the base clock for the Transmitter and Receiver.
The USART supports four modes of clock operation: Normal asynchronous, Double
Speed asynchronous, Master synchronous and Slave synchronous mode. The UMSEL
bit in USART Control and Status Register C (UCSRC) selects between asynchronous
and synchronous operation. Double Speed (asynchronous mode only) is controlled by
the U2X found in the UCSRB Register. When using synchronous mode (UMSEL = 1),
the Data Direction Register for the XCK pin (DDR_XCK) controls whether the clock
source is internal (Master mode) or external (Slave mode). The XCK pin is only active
when using synchronous mode.
Figure 80 shows a block diagram of the Clock Generation logic.
170
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 80. Clock Generation Logic, Block Diagram
UBRR
U2X
fosc
UBRR+1
Prescaling
Down-Counter
/ 2
/ 4
/ 2
0
1
0
OSC
txclk
1
DDR_XCK
Sync
Edge
Register
Detector
xcki
0
UMSEL
1
XCK
Pin
xcko
DDR_XCK
UCPOL
1
rxclk
0
Signal description:
txclk Transmitter clock (Internal Signal).
rxclk Receiver base clock (Internal Signal).
xcki
Input from XCK pin (internal Signal). Used for synchronous slave operation.
xcko Clock output to XCK pin (Internal Signal). Used for synchronous master
operation.
fosc
XTAL pin frequency (System Clock).
Internal Clock Generation –
The Baud Rate Generator
Internal clock generation is used for the asynchronous and the synchronous master
modes of operation. The description in this section refers to Figure 80.
The USART Baud Rate Register (UBRR) and the down-counter connected to it function
as a programmable prescaler or baud rate generator. The down-counter, running at sys-
tem clock (fOSC), is loaded with the UBRR value each time the counter has counted
down to zero or when the UBRRL Register is written. A clock is generated each time the
counter reaches zero. This clock is the baud rate generator clock output
(= fOSC/(UBRR+1)). The transmitter divides the baud rate generator clock output by
2, 8, or 16 depending on mode. The baud rate generator output is used directly by the
receiver’s clock and data recovery units. However, the recovery units use a state
machine that uses 2, 8 or 16 states depending on mode set by the state of the UMSEL,
U2X and DDR_XCK bits.
Table 74 contains equations for calculating the baud rate (in bits per second) and for
calculating the UBRR value for each mode of operation using an internally generated
clock source.
171
2490G–AVR–03/04
Table 74. Equations for Calculating Baud Rate Register Setting
Equation for Calculating
Baud Rate(1)
Equation for Calculating
UBRR Value
Operating Mode
Asynchronous Normal
mode (U2X = 0)
f
f
OSC
OSC
BAUD = --------------------------------------
16(UBRR + 1)
UBRR = ----------------------- – 1
16BAUD
Asynchronous Double
Speed mode (U2X = 1)
f
f
OSC
OSC
BAUD = -----------------------------------
8(UBRR + 1)
UBRR = -------------------- – 1
8BAUD
Synchronous Master
mode
f
f
OSC
OSC
BAUD = -----------------------------------
UBRR = -------------------- – 1
2BAUD
2(UBRR + 1)
Note:
BAUD Baud rate (in bits per second, bps)
fOSC System Oscillator clock frequency
1. The baud rate is defined to be the transfer rate in bit per second (bps).
UBRR Contents of the UBRRH and UBRRL Registers, (0 - 4095)
Some examples of UBRR values for some system clock frequencies are found in Table
82 on page 192 to Table 85 on page 195.
Double Speed Operation
(U2X)
The transfer rate can be doubled by setting the U2X bit in UCSRB. Setting this bit only
has effect for the asynchronous operation. Set this bit to zero when using synchronous
operation.
Setting this bit will reduce the divisor of the baud rate divider from 16 to 8, effectively
doubling the transfer rate for asynchronous communication. Note however that the
Receiver will in this case only use half the number of samples (reduced from 16 to 8) for
data sampling and clock recovery, and therefore a more accurate baud rate setting and
system clock are required when this mode is used. For the Transmitter, there are no
downsides.
External Clock
External clocking is used by the synchronous slave modes of operation. The description
in this section refers to Figure 80 for details.
External clock input from the XCK pin is sampled by a synchronization register to mini-
mize the chance of meta-stability. The output from the synchronization register must
then pass through an edge detector before it can be used by the Transmitter and
Receiver. This process introduces a two CPU clock period delay and therefore the max-
imum external XCK clock frequency is limited by the following equation:
f
OSC
4
-----------
<
f
XCK
Note that fosc depends on the stability of the system clock source. It is therefore recom-
mended to add some margin to avoid possible loss of data due to frequency variations.
Synchronous Clock Operation When synchronous mode is used (UMSEL = 1), the XCK pin will be used as either clock
input (Slave) or clock output (Master). The dependency between the clock edges and
data sampling or data change is the same. The basic principle is that data input (on
RxD) is sampled at the opposite XCK clock edge of the edge the data output (TxD) is
changed.
172
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 81. Synchronous Mode XCK Timing
UCPOL = 1
XCK
RxD / TxD
Sample
UCPOL = 0
XCK
RxD / TxD
Sample
The UCPOL bit UCRSC selects which XCK clock edge is used for data sampling and
which is used for data change. As Figure 81 shows, when UCPOL is zero the data will
be changed at rising XCK edge and sampled at falling XCK edge. If UCPOL is set, the
data will be changed at falling XCK edge and sampled at rising XCK edge.
Frame Formats
A serial frame is defined to be one character of data bits with synchronization bits (start
and stop bits), and optionally a parity bit for error checking. The USART accepts all 30
combinations of the following as valid frame formats:
•
•
•
•
1 start bit
5, 6, 7, 8, or 9 data bits
no, even or odd parity bit
1 or 2 stop bits
A frame starts with the start bit followed by the least significant data bit. Then the next
data bits, up to a total of nine, are succeeding, ending with the most significant bit. If
enabled, the parity bit is inserted after the data bits, before the stop bits. When a com-
plete frame is transmitted, it can be directly followed by a new frame, or the
communication line can be set to an idle (high) state. Figure 82 illustrates the possible
combinations of the frame formats. Bits inside brackets are optional.
Figure 82. Frame Formats
FRAME
(IDLE)
St
0
1
2
3
4
[5]
[6]
[7]
[8]
[P] Sp1 [Sp2] (St / IDLE)
St
(n)
P
Start bit, always low.
Data bits (0 to 8).
Parity bit. Can be odd or even.
Stop bit, always high.
Sp
IDLE No transfers on the communication line (RxD or TxD). An IDLE line must be
high.
The frame format used by the USART is set by the UCSZ2:0, UPM1:0 and USBS bits in
UCSRB and UCSRC. The Receiver and Transmitter use the same setting. Note that
173
2490G–AVR–03/04
changing the setting of any of these bits will corrupt all ongoing communication for both
the Receiver and Transmitter.
The USART Character SiZe (UCSZ2:0) bits select the number of data bits in the frame.
The USART Parity mode (UPM1:0) bits enable and set the type of parity bit. The selec-
tion between one or two stop bits is done by the USART Stop Bit Select (USBS) bit. The
receiver ignores the second stop bit. An FE (Frame Error) will therefore only be detected
in the cases where the first stop bit is zero.
Parity Bit Calculation
The parity bit is calculated by doing an exclusive-or of all the data bits. If odd parity is
used, the result of the exclusive or is inverted. The relation between the parity bit and
data bits is as follows::
P
= d
= d
⊕ … ⊕ d ⊕ d ⊕ d ⊕ d ⊕ 0
3 2 1 0
even
odd
n – 1
n – 1
P
⊕ … ⊕ d ⊕ d ⊕ d ⊕ d ⊕ 1
3 2 1 0
Peven Parity bit using even parity
Podd
dn
Parity bit using odd parity
Data bit n of the character
If used, the parity bit is located between the last data bit and first stop bit of a serial
frame.
USART Initialization
The USART has to be initialized before any communication can take place. The initial-
ization process normally consists of setting the baud rate, setting frame format and
enabling the Transmitter or the Receiver depending on the usage. For interrupt driven
USART operation, the Global Interrupt Flag should be cleared (and interrupts globally
disabled) when doing the initialization.
Before doing a re-initialization with changed baud rate or frame format, be sure that
there are no ongoing transmissions during the period the registers are changed. The
TXC flag can be used to check that the Transmitter has completed all transfers, and the
RXC flag can be used to check that there are no unread data in the receive buffer. Note
that the TXC flag must be cleared before each transmission (before UDR is written) if it
is used for this purpose.
The following simple USART initialization code examples show one assembly and one
C function that are equal in functionality. The examples assume asynchronous opera-
tion using polling (no interrupts enabled) and a fixed frame format. The baud rate is
given as a function parameter. For the assembly code, the baud rate parameter is
assumed to be stored in the r17:r16 registers.
174
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Assembly Code Example(1)
USART_Init:
; Set baud rate
out
out
UBRRH, r17
UBRRL, r16
; Enable receiver and transmitter
ldi
out
r16, (1<<RXEN)|(1<<TXEN)
UCSRB,r16
; Set frame format: 8data, 2stop bit
ldi
out
ret
r16, (1<<USBS)|(3<<UCSZ0)
UCSRC,r16
C Code Example(1)
void USART_Init( unsigned int baud )
{
/* Set baud rate */
UBRRH = (unsigned char)(baud>>8);
UBRRL = (unsigned char)baud;
/* Enable receiver and transmitter */
UCSRB = (1<<RXEN)|(1<<TXEN);
/* Set frame format: 8data, 2stop bit */
UCSRC = (1<<USBS)|(3<<UCSZ0);
}
Note:
1. The example code assumes that the part specific header file is included.
For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”,
and “SBI” instructions must be replaced with instructions that allow access to
extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
More advanced initialization routines can be made that include frame format as parame-
ters, disable interrupts and so on. However, many applications use a fixed setting of the
baud and control registers, and for these types of applications the initialization code can
be placed directly in the main routine, or be combined with initialization code for other
I/O modules.
Data Transmission – The The USART Transmitter is enabled by setting the Transmit Enable (TXEN) bit in the
UCSRB Register. When the Transmitter is enabled, the normal port operation of the
USART Transmitter
TxD pin is overridden by the USART and given the function as the transmitter’s serial
output. The baud rate, mode of operation and frame format must be set up once before
doing any transmissions. If synchronous operation is used, the clock on the XCK pin will
be overridden and used as transmission clock.
Sending Frames with 5 to 8
Data Bits
A data transmission is initiated by loading the transmit buffer with the data to be trans-
mitted. The CPU can load the transmit buffer by writing to the UDR I/O location. The
buffered data in the transmit buffer will be moved to the Shift Register when the Shift
Register is ready to send a new frame. The Shift Register is loaded with new data if it is
in idle state (no ongoing transmission) or immediately after the last stop bit of the previ-
ous frame is transmitted. When the Shift Register is loaded with new data, it will transfer
one complete frame at the rate given by the baud register, U2X bit or by XCK depending
on mode of operation.
175
2490G–AVR–03/04
The following code examples show a simple USART transmit function based on polling
of the Data Register Empty (UDRE) flag. When using frames with less than eight bits,
the most significant bits written to the UDR are ignored. The USART has to be initialized
before the function can be used. For the assembly code, the data to be sent is assumed
to be stored in register R16
Assembly Code Example(1)
USART_Transmit:
; Wait for empty transmit buffer
sbis UCSRA,UDRE
rjmp USART_Transmit
; Put data (r16) into buffer, sends the data
out
ret
UDR,r16
C Code Example(1)
void USART_Transmit( unsigned char data )
{
/* Wait for empty transmit buffer */
while ( !( UCSRA & (1<<UDRE)) )
;
/* Put data into buffer, sends the data */
UDR = data;
}
Note:
1. The example code assumes that the part specific header file is included.
For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”,
and “SBI” instructions must be replaced with instructions that allow access to
extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
The function simply waits for the transmit buffer to be empty by checking the UDRE flag,
before loading it with new data to be transmitted. If the Data Register Empty Interrupt is
utilized, the interrupt routine writes the data into the buffer.
176
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Sending Frames with 9 Data
Bits
If 9-bit characters are used (UCSZ = 7), the ninth bit must be written to the TXB8 bit in
UCSRB before the low byte of the character is written to UDR. The following code
examples show a transmit function that handles 9-bit characters. For the assembly
code, the data to be sent is assumed to be stored in registers r17:r16.
Assembly Code Example(1)
USART_Transmit:
; Wait for empty transmit buffer
sbis UCSRA,UDRE
rjmp USART_Transmit
; Copy ninth bit from r17 to TXB8
cbi
UCSRB,TXB8
sbrc r17,0
sbi
UCSRB,TXB8
; Put LSB data (r16) into buffer, sends the data
out
ret
UDR,r16
C Code Example(1)
void USART_Transmit( unsigned int data )
{
/* Wait for empty transmit buffer */
while ( !( UCSRA & (1<<UDRE)) )
;
/* Copy ninth bit to TXB8 */
UCSRB &= ~(1<<TXB8);
if ( data & 0x0100 )
UCSRB |= (1<<TXB8);
/* Put data into buffer, sends the data */
UDR = data;
}
Note:
1. These transmit functions are written to be general functions. They can be optimized if
the contents of the UCSRB is static. For example, only the TXB8 bit of the UCSRB
Register is used after initialization.
For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”,
and “SBI” instructions must be replaced with instructions that allow access to
extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
The ninth bit can be used for indicating an address frame when using Multi-processor
Communication mode or for other protocol handling as for example synchronization.
177
2490G–AVR–03/04
Transmitter Flags and
Interrupts
The USART Transmitter has two flags that indicate its state: USART Data Register
Empty (UDRE) and Transmit Complete (TXC). Both flags can be used for generating
interrupts.
The Data Register Empty (UDRE) flag indicates whether the transmit buffer is ready to
receive new data. This bit is set when the transmit buffer is empty, and cleared when the
transmit buffer contains data to be transmitted that has not yet been moved into the Shift
Register. For compatibility with future devices, always write this bit to zero when writing
the UCSRA Register.
When the Data Register empty Interrupt Enable (UDRIE) bit in UCSRB is written to one,
the USART Data Register Empty Interrupt will be executed as long as UDRE is set (pro-
vided that global interrupts are enabled). UDRE is cleared by writing UDR. When
interrupt-driven data transmission is used, the Data Register Empty Interrupt routine
must either write new data to UDR in order to clear UDRE or disable the Data Register
Empty Interrupt, otherwise a new interrupt will occur once the interrupt routine
terminates.
The Transmit Complete (TXC) flag bit is set one when the entire frame in the Transmit
Shift Register has been shifted out and there are no new data currently present in the
transmit buffer. The TXC flag bit is automatically cleared when a transmit complete inter-
rupt is executed, or it can be cleared by writing a one to its bit location. The TXC flag is
useful in half-duplex communication interfaces (like the RS-485 standard), where a
transmitting application must enter Receive mode and free the communication bus
immediately after completing the transmission.
When the Transmit Compete Interrupt Enable (TXCIE) bit in UCSRB is set, the USART
Transmit Complete Interrupt will be executed when the TXC flag becomes set (provided
that global interrupts are enabled). When the transmit complete interrupt is used, the
interrupt handling routine does not have to clear the TXC flag, this is done automatically
when the interrupt is executed.
Parity Generator
The Parity Generator calculates the parity bit for the serial frame data. When parity bit is
enabled (UPM1 = 1), the Transmitter Control logic inserts the parity bit between the last
data bit and the first stop bit of the frame that is sent.
Disabling the Transmitter
The disabling of the Transmitter (setting the TXEN to zero) will not become effective
until ongoing and pending transmissions are completed, i.e., when the Transmit Shift
Register and Transmit Buffer Register do not contain data to be transmitted. When dis-
abled, the transmitter will no longer override the TxD pin.
178
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Data Reception – The
USART Receiver
The USART Receiver is enabled by writing the Receive Enable (RXEN) bit in the
UCSRB Register to one. When the Receiver is enabled, the normal pin operation of the
RxD pin is overridden by the USART and given the function as the receiver’s serial
input. The baud rate, mode of operation and frame format must be set up once before
any serial reception can be done. If synchronous operation is used, the clock on the
XCK pin will be used as transfer clock.
Receiving Frames with 5 to 8
Data Bits
The Receiver starts data reception when it detects a valid start bit. Each bit that follows
the start bit will be sampled at the baud rate or XCK clock, and shifted into the Receive
Shift Register until the first stop bit of a frame is received. A second stop bit will be
ignored by the Receiver. When the first stop bit is received, i.e., a complete serial frame
is present in the Receive Shift Register, the contents of the Shift Register will be moved
into the receive buffer. The receive buffer can then be read by reading the UDR I/O
location.
The following code example shows a simple USART receive function based on polling
of the Receive Complete (RXC) flag. When using frames with less than eight bits the
most significant bits of the data read from the UDR will be masked to zero. The USART
has to be initialized before the function can be used.
Assembly Code Example(1)
USART_Receive:
; Wait for data to be received
sbis UCSRA, RXC
rjmp USART_Receive
; Get and return received data from buffer
in
r16, UDR
ret
C Code Example(1)
unsigned char USART_Receive( void )
{
/* Wait for data to be received */
while ( !(UCSRA & (1<<RXC)) )
;
/* Get and return received data from buffer */
return UDR;
}
Note:
1. The example code assumes that the part specific header file is included.
For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”,
and “SBI” instructions must be replaced with instructions that allow access to
extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
The function simply waits for data to be present in the receive buffer by checking the
RXC flag, before reading the buffer and returning the value.
179
2490G–AVR–03/04
Receiving Frames with 9 Data If 9-bit characters are used (UCSZ=7) the ninth bit must be read from the RXB8 bit in
Bits
UCSRB before reading the low bits from the UDR. This rule applies to the FE, DOR,
and UPE status flags as well. Read status from UCSRA, then data from UDR. Reading
the UDR I/O location will change the state of the receive buffer FIFO and consequently
the TXB8, FE, DOR, and UPE bits, which all are stored in the FIFO, will change. The fol-
lowing code example shows a simple USART receive function that handles both nine bit
characters and the status bits.
Assembly Code Example(1)
USART_Receive:
; Wait for data to be received
sbis UCSRA, RXC
rjmp USART_Receive
; Get status and ninth bit, then data from buffer
in
in
in
r18, UCSRA
r17, UCSRB
r16, UDR
; If error, return -1
andi r18,(1<<FE)|(1<<DOR)|(1<<UPE)
breq USART_ReceiveNoError
ldi
ldi
r17, HIGH(-1)
r16, LOW(-1)
USART_ReceiveNoError:
; Filter the ninth bit, then return
lsr
r17
andi r17, 0x01
ret
C Code Example(1)
unsigned int USART_Receive( void )
{
unsigned char status, resh, resl;
/* Wait for data to be received */
while ( !(UCSRA & (1<<RXC)) )
;
/* Get status and ninth bit, then data */
/* from buffer */
status = UCSRA;
resh = UCSRB;
resl = UDR;
/* If error, return -1 */
if ( status & (1<<FE)|(1<<DOR)|(1<<UPE) )
return -1;
/* Filter the ninth bit, then return */
resh = (resh >> 1) & 0x01;
return ((resh << 8) | resl);
}
Note:
1. The code assumes that the part specific header file is included. For I/O Registers
located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”, and “SBI” instruc-
180
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
tions must be replaced with instructions that allow access to extended I/O. Typically
“LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and “CBR”.
The receive function example reads all the I/O Registers into the Register File before
any computation is done. This gives an optimal receive buffer utilization since the buffer
location read will be free to accept new data as early as possible.
Receive Compete Flag and
Interrupt
The USART Receiver has one flag that indicates the receiver state.
The Receive Complete (RXC) flag indicates if there are unread data present in the
receive buffer. This flag is one when unread data exist in the receive buffer, and zero
when the receive buffer is empty (i.e. does not contain any unread data). If the receiver
is disabled (RXEN = 0), the receive buffer will be flushed and consequently the RXC bit
will become zero.
When the Receive Complete Interrupt Enable (RXCIE) in UCSRB is set, the USART
Receive Complete Interrupt will be executed as long as the RXC flag is set (provided
that global interrupts are enabled). When interrupt-driven data reception is used, the
receive complete routine must read the received data from UDR in order to clear the
RXC flag, otherwise a new interrupt will occur once the interrupt routine terminates.
Receiver Error Flags
The USART Receiver has three error flags: Frame Error (FE), Data OverRun (DOR) and
USART Parity Error (UPE). All can be accessed by reading UCSRA. Common for the
error flags is that they are located in the receive buffer together with the frame for which
they indicate the error status. Due to the buffering of the error flags, the UCSRA must be
read before the receive buffer (UDR), since reading the UDR I/O location changes the
buffer read location. Another equality for the error flags is that they can not be altered by
software doing a write to the flag location. However, all flags must be set to zero when
the UCSRA is written for upward compatibility of future USART implementations. None
of the error flags can generate interrupts.
The Frame Error (FE) flag indicates the state of the first stop bit of the next readable
frame stored in the receive buffer. The FE flag is zero when the stop bit was correctly
read (as one), and the FE flag will be one when the stop bit was incorrect (zero). This
flag can be used for detecting out-of-sync conditions, detecting break conditions and
protocol handling. The FE flag is not affected by the setting of the USBS bit in UCSRC
since the Receiver ignores all, except for the first, stop bits. For compatibility with future
devices, always set this bit to zero when writing to UCSRA.
The Data OverRun (DOR) flag indicates data loss due to a receiver buffer full condition.
A Data OverRun occurs when the receive buffer is full (two characters), it is a new char-
acter waiting in the Receive Shift Register, and a new start bit is detected. If the DOR
flag is set there was one or more serial frame lost between the frame last read from
UDR, and the next frame read from UDR. For compatibility with future devices, always
write this bit to zero when writing to UCSRA. The DOR flag is cleared when the frame
received was successfully moved from the Shift Register to the receive buffer.
The USART Parity Error (UPE) flag indicates that the next frame in the receive buffer
had a Parity Error when received. If parity check is not enabled the UPE bit will always
be read zero. For compatibility with future devices, always set this bit to zero when writ-
ing to UCSRA. For more details see “Parity Bit Calculation” on page 174 and “Parity
Checker” on page 181.
Parity Checker
The Parity Checker is active when the high USART Parity mode (UPM1) bit is set. Type
of parity check to be performed (odd or even) is selected by the UPM0 bit. When
enabled, the Parity Checker calculates the parity of the data bits in incoming frames and
compares the result with the parity bit from the serial frame. The result of the check is
181
2490G–AVR–03/04
stored in the receive buffer together with the received data and stop bits. The Parity
Error (UPE) flag can then be read by software to check if the frame had a Parity Error.
The UPE bit is set if the next character that can be read from the receive buffer had a
Parity Error when received and the parity checking was enabled at that point
(UPM1 = 1). This bit is valid until the receive buffer (UDR) is read.
Disabling the Receiver
In contrast to the Transmitter, disabling of the Receiver will be immediate. Data from
ongoing receptions will therefore be lost. When disabled (i.e. the RXEN is set to zero)
the receiver will no longer override the normal function of the RxD port pin. The receiver
buffer FIFO will be flushed when the Receiver is disabled. Remaining data in the buffer
will be lost
Flushing the Receive Buffer
The receiver buffer FIFO will be flushed when the Receiver is disabled, i.e., the buffer
will be emptied of its contents. Unread data will be lost. If the buffer has to be flushed
during normal operation, due to for instance an error condition, read the UDR I/O loca-
tion until the RXC flag is cleared. The following code examples show how to flush the
receive buffer.
Assembly Code Example(1)
USART_Flush:
sbis UCSRA, RXC
ret
in
r16, UDR
rjmp USART_Flush
C Code Example(1)
void USART_Flush( void )
{
unsigned char dummy;
while ( UCSRA & (1<<RXC) ) dummy = UDR;
}
Note:
1. The example code assumes that the part specific header file is included.
For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”,
and “SBI” instructions must be replaced with instructions that allow access to
extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
Asynchronous Data
Reception
The USART includes a clock recovery and a data recovery unit for handling asynchro-
nous data reception. The clock recovery logic is used for synchronizing the internally
generated baud rate clock to the incoming asynchronous serial frames at the RxD pin.
The data recovery logic samples and low pass filters each incoming bit, thereby improv-
ing the noise immunity of the Receiver. The asynchronous reception operational range
depends on the accuracy of the internal baud rate clock, the rate of the incoming
frames, and the frame size in number of bits.
Asynchronous Clock
Recovery
The Clock Recovery logic synchronizes internal clock to the incoming serial frames. Fig-
ure 83 illustrates the sampling process of the start bit of an incoming frame. The sample
rate is 16 times the baud rate for Normal mode, and eight times the baud rate for Double
Speed mode. The horizontal arrows illustrate the synchronization variation due to the
sampling process. Note the larger time variation when using the Double Speed mode
182
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
(U2X = 1) of operation. Samples denoted zero are samples done when the RxD line is
idle (i.e., no communication activity).
Figure 83. Start Bit Sampling
RxD
IDLE
START
BIT 0
Sample
(U2X = 0)
0
0
1
1
2
3
2
4
5
3
6
7
4
8
9
5
10
11
6
12
13
7
14
15
8
16
1
1
2
3
Sample
(U2X = 1)
0
2
When the Clock Recovery logic detects a high (idle) to low (start) transition on the RxD
line, the start bit detection sequence is initiated. Let sample 1 denote the first zero-sam-
ple as shown in the figure. The Clock Recovery logic then uses samples 8, 9 and 10 for
Normal mode, and samples 4, 5 and 6 for Double Speed mode (indicated with sample
numbers inside boxes on the figure), to decide if a valid start bit is received. If two or
more of these three samples have logical high levels (the majority wins), the start bit is
rejected as a noise spike and the Receiver starts looking for the next high to low-transi-
tion. If however, a valid start bit is detected, the clock recovery logic is synchronized and
the data recovery can begin. The synchronization process is repeated for each start bit.
Asynchronous Data Recovery When the receiver clock is synchronized to the start bit, the data recovery can begin.
The data recovery unit uses a state machine that has 16 states for each bit in Normal
mode and eight states for each bit in Double Speed mode. Figure 84 shows the sam-
pling of the data bits and the parity bit. Each of the samples is given a number that is
equal to the state of the recovery unit.
Figure 84. Sampling of Data and Parity Bit
RxD
BIT n
Sample
(U2X = 0)
1
1
2
3
2
4
5
3
6
7
4
8
9
5
10
11
6
12
13
7
14
15
8
16
1
1
Sample
(U2X = 1)
The decision of the logic level of the received bit is taken by doing a majority voting of
the logic value to the three samples in the center of the received bit. The center samples
are emphasized on the figure by having the sample number inside boxes. The majority
voting process is done as follows: If two or all three samples have high levels, the
received bit is registered to be a logic 1. If two or all three samples have low levels, the
received bit is registered to be a logic 0. This majority voting process acts as a low pass
filter for the incoming signal on the RxD pin. The recovery process is then repeated until
a complete frame is received. Including the first stop bit. Note that the receiver only uses
the first stop bit of a frame. Figure 85 shows the sampling of the stop bit and the earliest
possible beginning of the start bit of the next frame.
183
2490G–AVR–03/04
Figure 85. Stop Bit Sampling and Next Start Bit Sampling
(A)
(B)
(C)
RxD
STOP 1
Sample
(U2X = 0)
1
1
2
3
2
4
5
3
6
7
4
8
9
5
10
0/1 0/1 0/1
Sample
(U2X = 1)
6
0/1
The same majority voting is done to the stop bit as done for the other bits in the frame. If
the stop bit is registered to have a logic 0 value, the Frame Error (FE) flag will be set.
A new high to low transition indicating the start bit of a new frame can come right after
the last of the bits used for majority voting. For Normal Speed mode, the first low level
sample can be at point marked (A) in Figure 85. For Double Speed mode the first low
level must be delayed to (B). (C) marks a stop bit of full length. The early start bit detec-
tion influences the operational range of the Receiver.
Asynchronous Operational
Range
The operational range of the Receiver is dependent on the mismatch between the
received bit rate and the internally generated baud rate. If the Transmitter is sending
frames at too fast or too slow bit rates, or the internally generated baud rate of the
receiver does not have a similar (see Table 75) base frequency, the Receiver will not be
able to synchronize the frames to the start bit.
The following equations can be used to calculate the ratio of the incoming data rate and
internal receiver baud rate.
(D + 1)S
= ------------------------------------------
(D + 2)S
= -----------------------------------
R
R
fast
slow
S – 1 + D ⋅ S + S
(D + 1)S + S
M
F
D
S
Sum of character size and parity size (D = 5 to 10 bit)
Samples per bit. S = 16 for Normal Speed mode and S = 8 for Double Speed
mode.
SF
First sample number used for majority voting. SF = 8 for Normal Speed and SF = 4
for Double Speed mode.
SM Middle sample number used for majority voting. SM = 9 for Normal Speed and
SM = 5 for Double Speed mode.
Rslow is the ratio of the slowest incoming data rate that can be accepted in relation to the
receiver baud rate. Rfast is the ratio of the fastest incoming data rate that can be
accepted in relation to the receiver baud rate.
Table 75 and Table 76 list the maximum Receiver baud rate error that can be tolerated.
Note that Normal Speed mode has higher toleration of baud rate variations.
184
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 75. Recommended Maximum Receiver Baud Rate Error for Normal Speed Mode
(U2X = 0)
D
Max Total
Error (%)
Recommended Max
Receiver Error (%)
# (Data+Parity Bit)
R
slow (%)
93.20
94.12
94.81
95.36
95.81
96.17
Rfast (%)
106.67
105.79
105.11
104.58
104.14
103.78
5
6
+6.67/-6.8
+5.79/-5.88
+5.11/-5.19
+4.58/-4.54
+4.14/-4.19
+3.78/-3.83
3.0
2.5
2.0
2.0
1.5
1.5
7
8
9
10
Table 76. Recommended Maximum Receiver Baud Rate Error for Double Speed Mode
(U2X = 1)
D
Max Total
Error (%)
Recommended Max
Receiver Error (%)
# (Data+Parity Bit) Rslow (%) Rfast (%)
5
6
94.12
94.92
95.52
96.00
96.39
96.70
105.66
104.92
104.35
103.90
103.53
103.23
+5.66/-5.88
+4.92/-5.08
+4.35/-4.48
+3.90/-4.00
+3.53/-3.61
+3.23/-3.30
2.5
2.0
1.5
1.5
1.5
1.0
7
8
9
10
The recommendations of the maximum receiver baud rate error was made under the
assumption that the Receiver and Transmitter equally divides the maximum total error.
There are two possible sources for the receivers baud rate error. The receiver’s system
clock (XTAL) will always have some minor instability over the supply voltage range and
the temperature range. When using a crystal to generate the system clock, this is rarely
a problem, but for a resonator the system clock may differ more than 2% depending of
the resonators tolerance. The second source for the error is more controllable. The baud
rate generator can not always do an exact division of the system frequency to get the
baud rate wanted. In this case an UBRR value that gives an acceptable low error can be
used if possible.
Multi-processor
Communication Mode
Setting the Multi-processor Communication mode (MPCM) bit in UCSRA enables a fil-
tering function of incoming frames received by the USART Receiver. Frames that do not
contain address information will be ignored and not put into the receive buffer. This
effectively reduces the number of incoming frames that has to be handled by the CPU,
in a system with multiple MCUs that communicate via the same serial bus. The Trans-
mitter is unaffected by the MPCM setting, but has to be used differently when it is a part
of a system utilizing the Multi-processor Communication mode.
If the Receiver is set up to receive frames that contain five to eight data bits, then the
first stop bit indicates if the frame contains data or address information. If the receiver is
set up for frames with nine data bits, then the ninth bit (RXB8) is used for identifying
address and data frames. When the frame type bit (the first stop or the ninth bit) is one,
the frame contains an address. When the frame type bit is zero the frame is a data
frame.
185
2490G–AVR–03/04
The Multi-processor Communication mode enables several Slave MCUs to receive data
from a Master MCU. This is done by first decoding an address frame to find out which
MCU has been addressed. If a particular Slave MCU has been addressed, it will receive
the following data frames as normal, while the other slave MCUs will ignore the received
frames until another address frame is received.
Using MPCM
For an MCU to act as a Master MCU, it can use a 9-bit character frame format
(UCSZ = 7). The ninth bit (TXB8) must be set when an address frame (TXB8 = 1) or
cleared when a data frame (TXB = 0) is being transmitted. The Slave MCUs must in this
case be set to use a 9-bit character frame format.
The following procedure should be used to exchange data in Multi-processor Communi-
cation mode:
1. All Slave MCUs are in Multi-processor Communication mode (MPCM in UCSRA
is set).
2. The Master MCU sends an address frame, and all slaves receive and read this
frame. In the Slave MCUs, the RXC flag in UCSRA will be set as normal.
3. Each Slave MCU reads the UDR Register and determines if it has been
selected. If so, it clears the MPCM bit in UCSRA, otherwise it waits for the next
address byte and keeps the MPCM setting.
4. The addressed MCU will receive all data frames until a new address frame is
received. The other Slave MCUs, which still have the MPCM bit set, will ignore
the data frames.
5. When the last data frame is received by the addressed MCU, the addressed
MCU sets the MPCM bit and waits for a new address frame from Master. The
process then repeats from 2.
Using any of the 5- to 8-bit character frame formats is possible, but impractical since the
Receiver must change between using n and n+1 character frame formats. This makes
full duplex operation difficult since the Transmitter and Receiver uses the same charac-
ter size setting. If 5- to 8-bit character frames are used, the Transmitter must be set to
use two stop bit (USBS = 1) since the first stop bit is used for indicating the frame type.
Do not use Read-Modify-Write instructions (SBI and CBI) to set or clear the MPCM bit.
The MPCM bit shares the same I/O location as the TXC flag and this might accidentally
be cleared when using SBI or CBI instructions.
USART Register
Description
USART I/O Data Register –
UDR
Bit
7
6
5
4
3
2
1
0
RXB[7:0]
TXB[7:0]
UDR (Read)
UDR (Write)
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
The USART Transmit Data Buffer Register and USART Receive Data Buffer Registers
share the same I/O address referred to as USART Data Register or UDR. The Transmit
Data Buffer Register (TXB) will be the destination for data written to the UDR Register
location. Reading the UDR Register location will return the contents of the Receive Data
Buffer Register (RXB).
For 5-, 6-, or 7-bit characters the upper unused bits will be ignored by the Transmitter
and set to zero by the Receiver.
186
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
The transmit buffer can only be written when the UDRE flag in the UCSRA Register is
set. Data written to UDR when the UDRE flag is not set, will be ignored by the USART
transmitter. When data is written to the transmit buffer, and the Transmitter is enabled,
the Transmitter will load the data into the Transmit Shift Register when the Shift Register
is empty. Then the data will be serially transmitted on the TxD pin.
The receive buffer consists of a two level FIFO. The FIFO will change its state whenever
the receive buffer is accessed. Due to this behavior of the receive buffer, do not use
read modify write instructions (SBI and CBI) on this location. Be careful when using bit
test instructions (SBIC and SBIS), since these also will change the state of the FIFO.
USART Control and Status
Register A – UCSRA
Bit
7
RXC
R
6
5
UDRE
R
4
FE
R
3
DOR
R
2
UPE
R
1
0
MPCM
R/W
0
TXC
R/W
0
U2X
R/W
0
UCSRA
Read/Write
Initial Value
0
1
0
0
0
• Bit 7 – RXC: USART Receive Complete
This flag bit is set when there are unread data in the receive buffer and cleared when the
receive buffer is empty (i.e., does not contain any unread data). If the receiver is dis-
abled, the receive buffer will be flushed and consequently the RXC bit will become zero.
The RXC flag can be used to generate a Receive Complete interrupt (see description of
the RXCIE bit).
• Bit 6 – TXC: USART Transmit Complete
This flag bit is set when the entire frame in the Transmit Shift Register has been shifted
out and there are no new data currently present in the transmit buffer (UDR). The TXC
flag bit is automatically cleared when a transmit complete interrupt is executed, or it can
be cleared by writing a one to its bit location. The TXC flag can generate a Transmit
Complete interrupt (see description of the TXCIE bit).
• Bit 5 – UDRE: USART Data Register Empty
The UDRE flag indicates if the transmit buffer (UDR) is ready to receive new data. If
UDRE is one, the buffer is empty, and therefore ready to be written. The UDRE flag can
generate a Data Register Empty interrupt (see description of the UDRIE bit).
UDRE is set after a reset to indicate that the Transmitter is ready.
• Bit 4 – FE: Frame Error
This bit is set if the next character in the receive buffer had a Frame Error when
received. For example, when the first stop bit of the next character in the receive buffer
is zero. This bit is valid until the receive buffer (UDR) is read. The FE bit is zero when
the stop bit of received data is one. Always set this bit to zero when writing to UCSRA.
• Bit 3 – DOR: Data OverRun
This bit is set if a Data OverRun condition is detected. A Data OverRun occurs when the
receive buffer is full (two characters), it is a new character waiting in the Receive Shift
Register, and a new start bit is detected. This bit is valid until the receive buffer (UDR) is
read. Always set this bit to zero when writing to UCSRA.
187
2490G–AVR–03/04
• Bit 2 – UPE: USART Parity Error
This bit is set if the next character in the receive buffer had a Parity Error when received
and the Parity Checking was enabled at that point (UPM1 = 1). This bit is valid until the
receive buffer (UDR) is read. Always set this bit to zero when writing to UCSRA.
• Bit 1 – U2X: Double the USART Transmission Speed
This bit only has effect for the asynchronous operation. Write this bit to zero when using
synchronous operation.
Writing this bit to one will reduce the divisor of the baud rate divider from 16 to 8 effec-
tively doubling the transfer rate for asynchronous communication.
• Bit 0 – MPCM: Multi-processor Communication Mode
This bit enables the Multi-processor Communication Mode. When the MPCM bit is writ-
ten to one, all the incoming frames received by the USART Receiver that do not contain
address information will be ignored. The Transmitter is unaffected by the MPCM setting.
For more detailed information see “Multi-processor Communication Mode” on page 185.
USART Control and Status
Register B – UCSRB
Bit
7
RXCIE
R/W
0
6
TXCIE
R/W
0
5
UDRIE
R/W
0
4
RXEN
R/W
0
3
TXEN
R/W
0
2
UCSZ2
R/W
0
1
RXB8
R
0
TXB8
R/W
0
UCSRB
Read/Write
Initial Value
0
• Bit 7 – RXCIE: RX Complete Interrupt Enable
Writing this bit to one enables interrupt on the RXC flag. A USART Receive Complete
interrupt will be generated only if the RXCIE bit is written to one, the Global Interrupt
Flag in SREG is written to one and the RXC bit in UCSRA is set.
• Bit 6 – TXCIE: TX Complete Interrupt Enable
Writing this bit to one enables interrupt on the TXC flag. A USART Transmit Complete
interrupt will be generated only if the TXCIE bit is written to one, the Global Interrupt
Flag in SREG is written to one and the TXC bit in UCSRA is set.
• Bit 5 – UDRIE: USART Data Register Empty Interrupt Enable
Writing this bit to one enables interrupt on the UDRE flag. A Data Register Empty inter-
rupt will be generated only if the UDRIE bit is written to one, the Global Interrupt Flag in
SREG is written to one and the UDRE bit in UCSRA is set.
• Bit 4 – RXEN: Receiver Enable
Writing this bit to one enables the USART receiver. The Receiver will override normal
port operation for the RxD pin when enabled. Disabling the Receiver will flush the
receive buffer invalidating the FE, DOR, and UPE flags.
• Bit 3 – TXEN: Transmitter Enable
Writing this bit to one enables the USART Transmitter. The Transmitter will override nor-
mal port operation for the TxD pin when enabled. The disabling of the Transmitter
(writing TXEN to zero) will not become effective until ongoing and pending transmis-
sions are completed, i.e., when the Transmit Shift Register and Transmit Buffer Register
188
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
do not contain data to be transmitted. When disabled, the Transmitter will no longer
override the TxD port.
• Bit 2 – UCSZ2: Character Size
The UCSZ2 bits combined with the UCSZ1:0 bit in UCSRC sets the number of data bits
(Character Size) in a frame the Receiver and Transmitter use.
• Bit 1 – RXB8: Receive Data Bit 8
RXB8 is the ninth data bit of the received character when operating with serial frames
with nine data bits. Must be read before reading the low bits from UDR.
• Bit 0 – TXB8: Transmit Data Bit 8
TXB8 is the ninth data bit in the character to be transmitted when operating with serial
frames with nine data bits. Must be written before writing the low bits to UDR.
USART Control and Status
Register C – UCSRC(1)
Bit
7
–
R/W
0
6
UMSEL
R/W
0
5
UPM1
R/W
0
4
UPM0
R/W
0
3
USBS
R/W
0
2
UCSZ1
R/W
1
1
UCSZ0
R/W
1
0
UCPOL
R/W
0
UCSRC
Read/Write
Initial Value
Note:
1. This register is not available in ATmega103 compatibility mode.
• Bit 7 – Reserved Bit
This bit is reserved for future use. For compatibility with future devices, this bit must be
written to zero when UCSRC is written.
• Bit 6 – UMSEL: USART Mode Select
This bit selects between asynchronous and synchronous mode of operation.
Table 77. UMSEL Bit Settings
UMSEL
Mode
0
1
Asynchronous Operation
Synchronous Operation
• Bit 5:4 – UPM1:0: Parity Mode
These bits enable and set type of parity generation and check. If enabled, the Transmit-
ter will automatically generate and send the parity of the transmitted data bits within
each frame. The Receiver will generate a parity value for the incoming data and com-
pare it to the UPM0 setting. If a mismatch is detected, the UPE flag in UCSRB will be
set.
Table 78. UPM Bits Settings
UPM1
UPM0
Parity Mode
0
0
1
1
0
1
0
1
Disabled
Reserved
Enabled, Even Parity
Enabled, Odd Parity
189
2490G–AVR–03/04
• Bit 3 – USBS: Stop Bit Select
This bit selects the number of stop bits to be inserted by the Transmitter. The Receiver
ignores this setting.
Table 79. USBS Bit Settings
USBS
Stop Bit(s)
1-bit
0
1
2-bit
• Bit 2:1 – UCSZ1:0: Character Size
The UCSZ1:0 bits combined with the UCSZ2 bit in UCSRB sets the number of data bits
(Character Size) in a frame the Receiver and Transmitter use.
Table 80. UCSZ Bits Settings
UCSZ2
UCSZ1
UCSZ0
Character Size
5-bit
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
6-bit
7-bit
8-bit
Reserved
Reserved
Reserved
9-bit
• Bit 0 – UCPOL: Clock Polarity
This bit is used for synchronous mode only. Write this bit to zero when asynchronous
mode is used. The UCPOL bit sets the relationship between data output change and
data input sample, and the synchronous clock (XCK).
Table 81. UCPOL Bit Settings
Transmitted Data Changed
(Output of TxD Pin)
Received Data Sampled
(Input on RxD Pin)
UCPOL
0
1
Rising XCK Edge
Falling XCK Edge
Falling XCK Edge
Rising XCK Edge
190
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
USART Baud Rate Registers –
UBRRL and UBRRH(1)
Bit
15
14
13
12
11
10
9
8
–
–
–
–
UBRR[11:8]
UBRRH
UBRRL
UBRR[7:0]
7
R
6
R
5
R
4
R
3
R/W
R/W
0
2
R/W
R/W
0
1
R/W
R/W
0
0
R/W
R/W
0
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
0
0
0
0
0
0
0
0
Note:
1. UBRRH is not available in mega103 compatibility mode
• Bit 15:12 – Reserved Bits
These bits are reserved for future use. For compatibility with future devices, these bit
must be written to zero when UBRRH is written.
• Bit 11:0 – UBRR11:0: USART Baud Rate Register
This is a 12-bit register which contains the USART baud rate. The UBRRH contains the
four most significant bits, and the UBRRL contains the eight least significant bits of the
USART baud rate. Ongoing transmissions by the Transmitter and Receiver will be cor-
rupted if the baud rate is changed. Writing UBRRL will trigger an immediate update of
the baud rate prescaler.
Examples of Baud Rate
Setting
For standard crystal and resonator frequencies, the most commonly used baud rates for
asynchronous operation can be generated by using the UBRR settings in Table 82 to
Table 85. UBRR values which yield an actual baud rate differing less than 0.5% from the
target baud rate, are bold in the table. Higher error ratings are acceptable, but the
receiver will have less noise resistance when the error ratings are high, especially for
large serial frames (see “Asynchronous Operational Range” on page 184). The error
values are calculated using the following equation:
BaudRateClosest Match
Error[%] = ------------------------------------------------------- – 1 • 100%
BaudRate
191
2490G–AVR–03/04
Table 82. Examples of UBRR Settings for Commonly Used Oscillator Frequencies
fosc = 1.0000 MHz fosc = 1.8432 MHz
U2X = 0 U2X = 1 U2X = 0 U2X = 1
UBRR UBRR UBRR UBRR
fosc = 2.0000 MHz
U2X = 0 U2X = 1
UBRR UBRR
Baud
Rate
(bps)
Error
0.2%
0.2%
-7.0%
8.5%
8.5%
8.5%
-18.6%
8.5%
–
Error
0.2%
0.2%
0.2%
-3.5%
-7.0%
8.5%
8.5%
8.5%
-18.6%
8.5%
–
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-25.0%
0.0%
–
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
–
Error
0.2%
0.2%
0.2%
-3.5%
-7.0%
8.5%
8.5%
8.5%
-18.6%
8.5%
–
Error
0.2%
0.2%
0.2%
2.1%
0.2%
-3.5%
-7.0%
8.5%
8.5%
8.5%
–
2400
25
12
6
51
25
12
8
47
23
11
7
95
47
23
15
11
7
51
25
12
8
103
51
25
16
12
8
4800
9600
14.4k
19.2k
28.8k
38.4k
57.6k
76.8k
115.2k
230.4k
250k
3
2
6
5
6
1
3
3
3
1
2
2
5
2
6
0
1
1
3
1
3
–
1
1
2
1
2
–
–
0
0
1
0
1
–
–
–
–
0
–
–
–
–
–
–
–
–
–
–
–
0
0.0%
Max (1)
62.5 kbps
UBRR = 0, Error = 0.0%
125 kbps
115.2 kbps
230.4 kbps
125 kbps
250 kbps
1.
192
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 83. Examples of UBRR Settings for Commonly Used Oscillator Frequencies (Continued)
fosc = 3.6864 MHz fosc = 4.0000 MHz
U2X = 0 U2X = 1 U2X = 0 U2X = 1
UBRR UBRR UBRR UBRR
fosc = 7.3728 MHz
Baud
Rate
(bps)
U2X = 0
U2X = 1
UBRR
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-7.8%
–
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-7.8%
-7.8%
–
Error
0.2%
0.2%
0.2%
2.1%
0.2%
-3.5%
-7.0%
8.5%
8.5%
8.5%
8.5%
0.0%
–
Error
0.2%
0.2%
0.2%
-0.8%
0.2%
2.1%
0.2%
-3.5%
-7.0%
8.5%
8.5%
0.0%
0.0%
–
UBRR
191
95
47
31
23
15
11
7
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-7.8%
-7.8%
–
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-7.8%
-7.8%
-7.8%
2400
4800
9600
14.4k
19.2k
28.8k
38.4k
57.6k
76.8k
115.2k
230.4k
250k
95
47
23
15
11
7
191
95
47
31
23
15
11
7
103
51
25
16
12
8
207
103
51
34
25
16
12
8
383
191
95
63
47
31
23
15
11
7
5
6
3
3
2
5
2
6
5
1
3
1
3
3
0
1
0
1
1
3
0
1
0
1
1
3
0.5M
1M
–
0
–
0
0
1
–
–
–
–
–
–
–
0
Max (1)
230.4 kbps
UBRR = 0, Error = 0.0%
460.8 kbps
250 kbps
0.5 Mbps
460.8 kbps
921.6 kbps
1.
193
2490G–AVR–03/04
Table 84. Examples of UBRR Settings for Commonly Used Oscillator Frequencies (Continued)
fosc = 8.0000 MHz fosc = 11.0592 MHz fosc = 14.7456 MHz
U2X = 0 U2X = 1 U2X = 0 U2X = 1 U2X = 0 U2X = 1
UBRR UBRR UBRR UBRR UBRR UBRR
Baud
Rate
(bps)
Error
0.2%
0.2%
0.2%
-0.8%
0.2%
2.1%
0.2%
-3.5%
-7.0%
8.5%
8.5%
0.0%
0.0%
–
Error
-0.1%
0.2%
0.2%
0.6%
0.2%
-0.8%
0.2%
2.1%
0.2%
-3.5%
8.5%
0.0%
0.0%
0.0%
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-7.8%
–
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-7.8%
-7.8%
–
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-7.8%
-7.8%
-7.8%
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
5.3%
-7.8%
-7.8%
2400
4800
9600
14.4k
19.2k
28.8k
38.4k
57.6k
76.8k
115.2k
230.4k
250k
207
103
51
34
25
16
12
8
416
207
103
68
51
34
25
16
12
8
287
143
71
47
35
23
17
11
8
575
287
143
95
71
47
35
23
17
11
5
383
191
95
63
47
31
23
15
11
7
767
383
191
127
95
63
47
31
23
15
7
6
3
5
1
3
2
3
1
3
2
5
3
6
0.5M
1M
0
1
–
2
1
3
–
0
–
–
–
0
1
Max (1)
0.5 Mbps
UBRR = 0, Error = 0.0%
1 Mbps
691.2 kbps
1.3824 Mbps
921.6 kbps
1.8432 Mbps
1.
194
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 85. Examples of UBRR Settings for Commonly Used Oscillator Frequencies (Continued)
fosc = 16.0000 MHz fosc = 18.4320 MHz
U2X = 0 U2X = 1 U2X = 0 U2X = 1
UBRR UBRR UBRR UBRR
fosc = 20.0000 MHz
Baud
Rate
(bps)
U2X = 0
U2X = 1
UBRR
Error
-0.1%
0.2%
0.2%
0.6%
0.2%
-0.8%
0.2%
2.1%
0.2%
-3.5%
8.5%
0.0%
0.0%
0.0%
Error
0.0%
-0.1%
0.2%
-0.1%
0.2%
0.6%
0.2%
-0.8%
0.2%
2.1%
-3.5%
0.0%
0.0%
0.0%
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-7.8%
–
Error
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
2.4%
-7.8%
–
UBRR
520
259
129
86
64
42
32
21
15
10
4
Error
0.0%
0.2%
0.2%
-0.2%
0.2%
0.9%
-1.4%
-1.4%
1.7%
-1.4%
8.5%
0.0%
–
Error
0.0%
0.0%
0.2%
-0.2%
0.2%
-0.2%
0.2%
0.9%
-1.4%
-1.4%
-1.4%
0.0%
0.0%
–
2400
4800
9600
14.4k
19.2k
28.8k
38.4k
57.6k
76.8k
115.2k
230.4k
250k
416
207
103
68
51
34
25
16
12
8
832
416
207
138
103
68
51
34
25
16
8
479
239
119
79
59
39
29
19
14
9
959
479
239
159
119
79
59
39
29
19
9
1041
520
259
173
129
86
64
42
32
21
3
4
10
3
7
4
8
4
9
0.5M
1M
1
3
–
4
–
4
0
1
–
–
–
–
–
–
Max (1)
1 Mbps
UBRR = 0, Error = 0.0%
2 Mbps
1.152 Mbps
2.304 Mbps
1.25 Mbps
2.5 Mbps
1.
195
2490G–AVR–03/04
Two-wire Serial
Interface
Features
• Simple yet Powerful and Flexible Communication Interface, Only Two Bus Lines Needed
• Both Master and Slave Operation Supported
• Device can Operate as Transmitter or Receiver
• 7-bit Address Space allows up to 128 Different Slave Addresses
• Multi-master Arbitration Support
• Up to 400 kHz Data Transfer Speed
• Slew-rate Limited Output Drivers
• Noise Suppression Circuitry Rejects Spikes on Bus Lines
• Fully Programmable Slave Address with General Call Support
• Address Recognition Causes Wake-up when AVR is in Sleep Mode
Two-wire Serial Interface The Two-wire Serial Interface (TWI) is ideally suited for typical microcontroller applica-
tions. The TWI protocol allows the systems designer to interconnect up to 128 different
Bus Definition
devices using only two bi-directional bus lines, one for clock (SCL) and one for data
(SDA). The only external hardware needed to implement the bus is a single pull-up
resistor for each of the TWI bus lines. All devices connected to the bus have individual
addresses, and mechanisms for resolving bus contention are inherent in the TWI
protocol.
Figure 86. TWI Bus Interconnection
VCC
Device 1
Device 3
Device 2
Device n
R1
R2
........
SDA
SCL
TWI Terminology
The following definitions are frequently encountered in this section.
Table 86. TWI Terminology
Term
Description
Master
The device that initiates and terminates a transmission. The Master also
generates the SCL clock.
Slave
The device addressed by a Master.
The device placing data on the bus.
The device reading data from the bus.
Transmitter
Receiver
196
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Electrical Interconnection
As depicted in Figure 86, both bus lines are connected to the positive supply voltage
through pull-up resistors. The bus drivers of all TWI-compliant devices are open-drain or
open-collector. This implements a wired-AND function which is essential to the opera-
tion of the interface. A low level on a TWI bus line is generated when one or more TWI
devices output a zero. A high level is output when all TWI devices tri-state their outputs,
allowing the pull-up resistors to pull the line high. Note that all AVR devices connected to
the TWI bus must be powered in order to allow any bus operation.
The number of devices that can be connected to the bus is only limited by the bus
capacitance limit of 400 pF and the 7-bit slave address space. A detailed specification of
the electrical characteristics of the TWI is given in “Two-wire Serial Interface Character-
istics” on page 329. Two different sets of specifications are presented there, one
relevant for bus speeds below 100 kHz, and one valid for bus speeds up to 400 kHz.
Data Transfer and Frame
Format
Transferring Bits
Each data bit transferred on the TWI bus is accompanied by a pulse on the clock line.
The level of the data line must be stable when the clock line is high. The only exception
to this rule is for generating start and stop conditions.
Figure 87. Data Validity
SDA
SCL
Data Stable
Data Stable
Data Change
START and STOP Conditions
The Master initiates and terminates a data transmission. The transmission is initiated
when the Master issues a START condition on the bus, and it is terminated when the
Master issues a STOP condition. Between a START and a STOP condition, the bus is
considered busy, and no other Master should try to seize control of the bus. A special
case occurs when a new START condition is issued between a START and STOP con-
dition. This is referred to as a REPEATED START condition, and is used when the
Master wishes to initiate a new transfer without relinquishing control of the bus. After a
REPEATED START, the bus is considered busy until the next STOP. This is identical to
the START behavior, and therefore START is used to describe both START and
REPEATED START for the remainder of this data sheet, unless otherwise noted. As
depicted below, START and STOP conditions are signalled by changing the level of the
SDA line when the SCL line is high.
197
2490G–AVR–03/04
Figure 88. START, REPEATED START, and STOP Conditions
SDA
SCL
START
STOP START
REPEATED START
STOP
Address Packet Format
All address packets transmitted on the TWI bus are nine bits long, consisting of seven
address bits, one READ/WRITE control bit and an acknowledge bit. If the READ/WRITE
bit is set, a read operation is to be performed, otherwise a write operation should be per-
formed. When a slave recognizes that it is being addressed, it should acknowledge by
pulling SDA low in the ninth SCL (ACK) cycle. If the addressed Slave is busy, or for
some other reason can not service the Master’s request, the SDA line should be left
high in the ACK clock cycle. The Master can then transmit a STOP condition, or a
REPEATED START condition to initiate a new transmission. An address packet consist-
ing of a slave address and a READ or a WRITE bit is called SLA+R or SLA+W,
respectively.
The MSB of the address byte is transmitted first. Slave addresses can freely be allo-
cated by the designer, but the address 0000 000 is reserved for a general call.
When a general call is issued, all slaves should respond by pulling the SDA line low in
the ACK cycle. A general call is used when a Master wishes to transmit the same mes-
sage to several slaves in the system. When the general call address followed by a write
bit is transmitted on the bus, all slaves set up to acknowledge the general call will pull
the SDA line low in the ack cycle. The following data packets will then be received by all
the slaves that acknowledged the general call. Note that transmitting the general call
address followed by a Read bit is meaningless, as this would cause contention if several
slaves started transmitting different data.
All addresses of the format 1111 xxx should be reserved for future purposes.
Figure 89. Address Packet Format
Addr MSB
Addr LSB
R/W
ACK
SDA
SCL
1
2
7
8
9
START
198
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Data Packet Format
All data packets transmitted on the TWI bus are nine bits long, consisting of one data
byte and an acknowledge bit. During a data transfer, the Master generates the clock and
the START and STOP conditions, while the Receiver is responsible for acknowledging
the reception. An Acknowledge (ACK) is signalled by the Receiver pulling the SDA line
low during the ninth SCL cycle. If the Receiver leaves the SDA line high, a NACK is sig-
nalled. When the Receiver has received the last byte, or for some reason cannot receive
any more bytes, it should inform the Transmitter by sending a NACK after the final byte.
The MSB of the data byte is transmitted first.
Figure 90. Data Packet Format
Data MSB
Data LSB
ACK
Aggregate
SDA
SDA from
Transmitter
SDA from
Receiver
SCL from
Master
1
2
7
8
9
STOP, REPEATED
START, or Next
Data Byte
SLA+R/W
Data Byte
Combining Address and Data A transmission basically consists of a START condition, a SLA+R/W, one or more data
Packets Into a Transmission
packets and a STOP condition. An empty message, consisting of a START followed by
a STOP condition, is illegal. Note that the wired-ANDing of the SCL line can be used to
implement handshaking between the Master and the Slave. The Slave can extend the
SCL low period by pulling the SCL line low. This is useful if the clock speed set up by the
Master is too fast for the Slave, or the Slave needs extra time for processing between
the data transmissions. The Slave extending the SCL low period will not affect the SCL
high period, which is determined by the Master. As a consequence, the Slave can
reduce the TWI data transfer speed by prolonging the SCL duty cycle.
Figure 91 shows a typical data transmission. Note that several data bytes can be trans-
mitted between the SLA+R/W and the STOP condition, depending on the software
protocol implemented by the application software.
Figure 91. Typical Data Transmission
Addr MSB
Addr LSB
R/W
ACK
Data MSB
Data LSB
ACK
SDA
SCL
1
2
7
SLA+R/W
8
9
1
2
7
8
9
START
Data Byte
STOP
199
2490G–AVR–03/04
Multi-master Bus
Systems, Arbitration and
Synchronization
The TWI protocol allows bus systems with several masters. Special concerns have
been taken in order to ensure that transmissions will proceed as normal, even if two or
more masters initiate a transmission at the same time. Two problems arise in multi-mas-
ter systems:
•
An algorithm must be implemented allowing only one of the masters to complete the
transmission. All other masters should cease transmission when they discover that
they have lost the selection process. This selection process is called arbitration.
When a contending master discovers that it has lost the arbitration process, it
should immediately switch to Slave mode to check whether it is being addressed by
the winning master. The fact that multiple masters have started transmission at the
same time should not be detectable to the slaves (i.e., the data being transferred on
the bus must not be corrupted).
•
Different masters may use different SCL frequencies. A scheme must be devised to
synchronize the serial clocks from all masters, in order to let the transmission
proceed in a lockstep fashion. This will facilitate the arbitration process.
The wired-ANDing of the bus lines is used to solve both these problems. The serial
clocks from all masters will be wired-ANDed, yielding a combined clock with a high
period equal to the one from the master with the shortest high period. The low period of
the combined clock is equal to the low period of the master with the longest low period.
Note that all masters listen to the SCL line, effectively starting to count their SCL high
and low Time-out periods when the combined SCL line goes high or low, respectively.
Figure 92. SCL Synchronization between Multiple Masters
TA low
TA high
SCL from
Master A
SCL from
Master B
SCL bus
Line
TBlow
TBhigh
Masters Start
Masters Start
Counting Low Period
Counting High Period
Arbitration is carried out by all masters continuously monitoring the SDA line after out-
putting data. If the value read from the SDA line does not match the value the master
had output, it has lost the arbitration. Note that a master can only lose arbitration when it
outputs a high SDA value while another master outputs a low value. The losing master
should immediately go to Slave mode, checking if it is being addressed by the winning
master. The SDA line should be left high, but losing masters are allowed to generate a
clock signal until the end of the current data or address packet. Arbitration will continue
until only one master remains, and this may take many bits. If several masters are trying
to address the same slave, arbitration will continue into the data packet.
200
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 93. Arbitration between Two Masters
START
Master A Loses
Arbitration, SDAA SDA
SDA from
Master A
SDA from
M
SDA Line
Synchronized
SCL Line
Note that arbitration is not allowed between:
•
•
•
A REPEATED START condition and a data bit.
A STOP condition and a data bit.
A REPEATED START and a STOP condition.
It is the user software’s responsibility to ensure that these illegal arbitration conditions
never occur. This implies that in multi-master systems, all data transfers must use the
same composition of SLA+R/W and data packets. In other words: All transmissions
must contain the same number of data packets, otherwise the result of the arbitration is
undefined.
201
2490G–AVR–03/04
Overview of the TWI
Module
The TWI module is comprised of several submodules, as shown in Figure 94. All regis-
ters drawn in a thick line are accessible through the AVR data bus.
Figure 94. Overview of the TWI Module
SCL
SDA
Spike
Filter
Spike
Filter
Slew-rate
Control
Slew-rate
Control
Bus Interface Unit
Bit Rate Generator
START / STOP
Spike Suppression
Prescaler
Control
Address/Data Shift
Register (TWDR)
Bit Rate Register
(TWBR)
Arbitration Detection
Ack
Address Match Unit
Control Unit
Address Register
(TWAR)
Status Register
(TWSR)
Control Register
(TWCR)
State Machine and
Status Control
Address Comparator
SCL and SDA Pins
These pins interface the AVR TWI with the rest of the MCU system. The output drivers
contain a slew-rate limiter in order to conform to the TWI specification. The input stages
contain a spike suppression unit removing spikes shorter than 50 ns. Note that the inter-
nal pull-ups in the AVR pads can be enabled by setting the PORT bits corresponding to
the SCL and SDA pins, as explained in the I/O Port section. The internal pull-ups can in
some systems eliminate the need for external ones.
Bit Rate Generator Unit
This unit controls the period of SCL when operating in a Master mode. The SCL period
is controlled by settings in the TWI Bit Rate Register (TWBR) and the Prescaler bits in
the TWI Status Register (TWSR). Slave operation does not depend on Bit Rate or Pres-
caler settings, but the CPU clock frequency in the slave must be at least 16 times higher
than the SCL frequency. Note that slaves may prolong the SCL low period, thereby
reducing the average TWI bus clock period. The SCL frequency is generated according
to the following equation:
CPU Clock frequency
SCL frequency = -----------------------------------------------------------
TWPS
16 + 2(TWBR) ⋅ 4
•
TWBR = Value of the TWI Bit Rate Register.
202
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
•
TWPS = Value of the prescaler bits in the TWI Status Register.
Note:
TWBR should be 10 or higher if the TWI operates in Master mode. If TWBR is lower than
10, the master may produce an incorrect output on SDA and SCL for the reminder of the
byte. The problem occurs when operating the TWI in Master mode, sending Start + SLA
+ R/W to a slave (a slave does not need to be connected to the bus for the condition to
happen).
Bus Interface Unit
This unit contains the Data and Address Shift Register (TWDR), a START/STOP Con-
troller and Arbitration detection hardware. The TWDR contains the address or data
bytes to be transmitted, or the address or data bytes received. In addition to the 8-bit
TWDR, the Bus Interface Unit also contains a register containing the (N)ACK bit to be
transmitted or received. This (N)ACK Register is not directly accessible by the applica-
tion software. However, when receiving, it can be set or cleared by manipulating the
TWI Control Register (TWCR). When in Transmitter mode, the value of the received
(N)ACK bit can be determined by the value in the TWSR.
The START/STOP Controller is responsible for generation and detection of START,
REPEATED START, and STOP conditions. The START/STOP controller is able to
detect START and STOP conditions even when the AVR MCU is in one of the sleep
modes, enabling the MCU to wake up if addressed by a Master.
If the TWI has initiated a transmission as Master, the Arbitration Detection hardware
continuously monitors the transmission trying to determine if arbitration is in process. If
the TWI has lost an arbitration, the Control Unit is informed. Correct action can then be
taken and appropriate status codes generated.
Address Match Unit
The Address Match unit checks if received address bytes match the 7-bit address in the
TWI Address Register (TWAR). If the TWI General Call Recognition Enable (TWGCE)
bit in the TWAR is written to one, all incoming address bits will also be compared
against the General Call address. Upon an address match, the Control unit is informed,
allowing correct action to be taken. The TWI may or may not acknowledge its address,
depending on settings in the TWCR. The Address Match unit is able to compare
addresses even when the AVR MCU is in sleep mode, enabling the MCU to wake-up if
addressed by a Master.
Control Unit
The Control unit monitors the TWI bus and generates responses corresponding to set-
tings in the TWI Control Register (TWCR). When an event requiring the attention of the
application occurs on the TWI bus, the TWI Interrupt Flag (TWINT) is asserted. In the
next clock cycle, the TWI Status Register (TWSR) is updated with a status code identify-
ing the event. The TWSR only contains relevant status information when the TWI
interrupt flag is asserted. At all other times, the TWSR contains a special status code
indicating that no relevant status information is available. As long as the TWINT flag is
set, the SCL line is held low. This allows the application software to complete its tasks
before allowing the TWI transmission to continue.
The TWINT flag is set in the following situations:
•
•
•
•
•
•
•
After the TWI has transmitted a START/REPEATED START condition.
After the TWI has transmitted SLA+R/W.
After the TWI has transmitted an address byte.
After the TWI has lost arbitration.
After the TWI has been addressed by own slave address or general call.
After the TWI has received a data byte.
After a STOP or REPEATED START has been received while still addressed as a
Slave.
203
2490G–AVR–03/04
•
When a bus error has occurred due to an illegal START or STOP condition.
204
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
TWI Register Description
TWI Bit Rate Register – TWBR
Bit
7
TWBR7
R/W
0
6
TWBR6
R/W
0
5
TWBR5
R/W
0
4
TWBR4
R/W
0
3
TWBR3
R/W
0
2
TWBR2
R/W
0
1
TWBR1
R/W
0
0
TWBR0
R/W
0
TWBR
Read/Write
Initial Value
• Bits 7..0 – TWI Bit Rate Register
TWBR selects the division factor for the bit rate generator. The bit rate generator is a
frequency divider which generates the SCL clock frequency in the Master modes. See
“Bit Rate Generator Unit” on page 202 for calculating bit rates.
TWI Control Register – TWCR
Bit
7
TWINT
R/W
0
6
TWEA
R/W
0
5
TWSTA
R/W
0
4
TWSTO
R/W
0
3
2
TWEN
R/W
0
1
–
R
0
0
TWIE
R/W
0
TWWC
R
0
TWCR
Read/Write
Initial Value
The TWCR is used to control the operation of the TWI. It is used to enable the TWI, to
initiate a Master access by applying a START condition to the bus, to generate a
Receiver acknowledge, to generate a stop condition, and to control halting of the bus
while the data to be written to the bus are written to the TWDR. It also indicates a write
collision if data is attempted written to TWDR while the register is inaccessible.
• Bit 7 – TWINT: TWI Interrupt Flag
This bit is set by hardware when the TWI has finished its current job and expects appli-
cation software response. If the I-bit in SREG and TWIE in TWCR are set, the MCU will
jump to the TWI Interrupt Vector. While the TWINT flag is set, the SCL low period is
stretched. The TWINT flag must be cleared by software by writing a logic one to it. Note
that this flag is not automatically cleared by hardware when executing the interrupt rou-
tine. Also note that clearing this flag starts the operation of the TWI, so all accesses to
the TWI Address Register (TWAR), TWI Status Register (TWSR), and TWI Data Regis-
ter (TWDR) must be complete before clearing this flag.
• Bit 6 – TWEA: TWI Enable Acknowledge Bit
The TWEA bit controls the generation of the acknowledge pulse. If the TWEA bit is writ-
ten to one, the ACK pulse is generated on the TWI bus if the following conditions are
met:
1. The device’s own slave address has been received.
2. A general call has been received, while the TWGCE bit in the TWAR is set.
3. A data byte has been received in Master Receiver or Slave Receiver mode.
By writing the TWEA bit to zero, the device can be virtually disconnected from the Two-
wire Serial Bus temporarily. Address recognition can then be resumed by writing the
TWEA bit to one again.
• Bit 5 – TWSTA: TWI START Condition Bit
The application writes the TWSTA bit to one when it desires to become a Master on the
Two-wire Serial Bus. The TWI hardware checks if the bus is available, and generates a
START condition on the bus if it is free. However, if the bus is not free, the TWI waits
until a STOP condition is detected, and then generates a new START condition to claim
205
2490G–AVR–03/04
the Bus Master status. TWSTA must be cleared by software when the START condition
has been transmitted.
• Bit 4 – TWSTO: TWI STOP Condition Bit
Writing the TWSTO bit to one in Master mode will generate a STOP condition on the
Two-wire Serial Bus. When the STOP condition is executed on the bus, the TWSTO bit
is cleared automatically. In Slave mode, setting the TWSTO bit can be used to recover
from an error condition. This will not generate a STOP condition, but the TWI returns to
a well-defined unaddressed Slave mode and releases the SCL and SDA lines to a high
impedance state.
• Bit 3 – TWWC: TWI Write Collision Flag
The TWWC bit is set when attempting to write to the TWI Data Register – TWDR when
TWINT is low. This flag is cleared by writing the TWDR Register when TWINT is high.
• Bit 2 – TWEN: TWI Enable Bit
The TWEN bit enables TWI operation and activates the TWI interface. When TWEN is
written to one, the TWI takes control over the I/O pins connected to the SCL and SDA
pins, enabling the slew-rate limiters and spike filters. If this bit is written to zero, the TWI
is switched off and all TWI transmissions are terminated, regardless of any ongoing
operation.
• Bit 1 – Res: Reserved Bit
This bit is a reserved bit and will always read as zero.
• Bit 0 – TWIE: TWI Interrupt Enable
When this bit is written to one, and the I-bit in SREG is set, the TWI interrupt request will
be activated for as long as the TWINT flag is high.
TWI Status Register – TWSR
Bit
7
TWS7
R
6
TWS6
R
5
TWS5
R
4
TWS4
R
3
TWS3
R
2
–
R
0
1
TWPS1
R/W
0
0
TWPS0
R/W
0
TWSR
Read/Write
Initial Value
1
1
1
1
1
• Bits 7..3 – TWS: TWI Status
These five bits reflect the status of the TWI logic and the Two-wire Serial Bus. The dif-
ferent status codes are described later in this section. Note that the value read from
TWSR contains both the 5-bit status value and the 2-bit prescaler value. The application
designer should mask the prescaler bits to zero when checking the status bits. This
makes status checking independent of prescaler setting. This approach is used in this
datasheet, unless otherwise noted.
• Bit 2 – Res: Reserved Bit
This bit is reserved and will always read as zero.
206
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• Bits 1..0 – TWPS: TWI Prescaler Bits
These bits can be read and written, and control the bit rate prescaler.
Table 87. TWI Bit Rate Prescaler
TWPS1
TWPS0
Prescaler Value
0
0
1
1
0
1
0
1
1
4
16
64
To calculate bit rates, see “Bit Rate Generator Unit” on page 202. The value of
TWPS1..0 is used in the equation.
TWI Data Register – TWDR
Bit
7
TWD7
R/W
1
6
TWD6
R/W
1
5
TWD5
R/W
1
4
TWD4
R/W
1
3
TWD3
R/W
1
2
TWD2
R/W
1
1
TWD1
R/W
1
0
TWD0
R/W
1
TWDR
Read/Write
Initial Value
In Transmit mode, TWDR contains the next byte to be transmitted. In Receive mode, the
TWDR contains the last byte received. It is writable while the TWI is not in the process of
shifting a byte. This occurs when the TWI Interrupt Flag (TWINT) is set by hardware.
Note that the data register cannot be initialized by the user before the first interrupt
occurs. The data in TWDR remains stable as long as TWINT is set. While data is shifted
out, data on the bus is simultaneously shifted in. TWDR always contains the last byte
present on the bus, except after a wake-up from a sleep mode by the TWI interrupt. In
this case, the contents of TWDR is undefined. In the case of a lost bus arbitration, no
data is lost in the transition from Master to Slave. Handling of the ACK bit is controlled
automatically by the TWI logic, the CPU cannot access the ACK bit directly.
• Bits 7..0 – TWD: TWI Data Register
These eight bits constitute the next data byte to be transmitted, or the latest data byte
received on the Two-wire Serial Bus.
TWI (Slave) Address Register
– TWAR
Bit
7
TWA6
R/W
1
6
TWA5
R/W
1
5
TWA4
R/W
1
4
TWA3
R/W
1
3
TWA2
R/W
1
2
TWA1
R/W
1
1
TWA0
R/W
1
0
TWGCE
R/W
0
TWAR
Read/Write
Initial Value
The TWAR should be loaded with the 7-bit slave address (in the seven most significant
bits of TWAR) to which the TWI will respond when programmed as a slave transmitter or
Receiver, and not needed in the Master modes. In multimaster systems, TWAR must be
set in masters which can be addressed as slaves by other masters.
The LSB of TWAR is used to enable recognition of the general call address (0x00).
There is an associated address comparator that looks for the slave address (or general
call address if enabled) in the received serial address. If a match is found, an interrupt
request is generated.
• Bits 7..1 – TWA: TWI (Slave) Address Register
These seven bits constitute the slave address of the TWI unit.
207
2490G–AVR–03/04
• Bit 0 – TWGCE: TWI General Call Recognition Enable Bit
If set, this bit enables the recognition of a General Call given over the Two-wire Serial
Bus.
Using the TWI
The AVR TWI is byte-oriented and interrupt based. Interrupts are issued after all bus
events, like reception of a byte or transmission of a START condition. Because the TWI
is interrupt-based, the application software is free to carry on other operations during a
TWI byte transfer. Note that the TWI Interrupt Enable (TWIE) bit in TWCR together with
the Global Interrupt Enable bit in SREG allow the application to decide whether or not
assertion of the TWINT flag should generate an interrupt request. If the TWIE bit is
cleared, the application must poll the TWINT flag in order to detect actions on the TWI
bus.
When the TWINT flag is asserted, the TWI has finished an operation and awaits applica-
tion response. In this case, the TWI Status Register (TWSR) contains a value indicating
the current state of the TWI bus. The application software can then decide how the TWI
should behave in the next TWI bus cycle by manipulating the TWCR and TWDR
registers.
Figure 95 is a simple example of how the application can interface to the TWI hardware.
In this example, a Master wishes to transmit a single data byte to a Slave. This descrip-
tion is quite abstract, a more detailed explanation follows later in this section. A simple
code example implementing the desired behavior is also presented.
Figure 95. Interfacing the Application to the TWI in a Typical Transmission
1. Application writes
to TWCR to initiate
transmission of
START
3. Check TWSR to see if START was
sent. Application loads SLA+W into
TWDR, and loads appropriate control
signals into TWCR, making sure that
TWINT is written to one, and
5. Check TWSR to see if SLA+W was
sent and ACK received.
Application loads data into TWDR,
and loads appropriate control signals
into TWCR, making sure that
TWINT is written to one
7. Check TWSR to see if data was sent
and ACK received. Application loads
appropriate control signals to send
STOP into TWCR, making sure that
TWINT is written to one
TWSTA is written to zero.
TWI bus
START
SLA+W
A
Data
A
STOP
Indicates
TWINT set
4. TWINT set.
2. TWINT set.
Status code indicates
START condition sent
6. TWINT set.
Status code indicates
data sent, ACK received
Status code indicates
SLA+W sent, ACK
received
1. The first step in a TWI transmission is to transmit a START condition. This is
done by writing a specific value into TWCR, instructing the TWI hardware to
transmit a START condition. Which value to write is described later on. However,
it is important that the TWINT bit is set in the value written. Writing a one to
TWINT clears the flag. The TWI will not start any operation as long as the
TWINT bit in TWCR is set. Immediately after the application has cleared TWINT,
the TWI will initiate transmission of the START condition.
2. When the START condition has been transmitted, the TWINT flag in TWCR is
set, and TWSR is updated with a status code indicating that the START condition
has successfully been sent.
208
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
3. The application software should now examine the value of TWSR, to make sure
that the START condition was successfully transmitted. If TWSR indicates other-
wise, the application software might take some special action, like calling an
error routine. Assuming that the status code is as expected, the application must
load SLA+W into TWDR. Remember that TWDR is used both for address and
data. After TWDR has been loaded with the desired SLA+W, a specific value
must be written to TWCR, instructing the TWI hardware to transmit the SLA+W
present in TWDR. Which value to write is described later on. However, it is
important that the TWINT bit is set in the value written. Writing a one to TWINT
clears the flag. The TWI will not start any operation as long as the TWINT bit in
TWCR is set. Immediately after the application has cleared TWINT, the TWI will
initiate transmission of the address packet.
4. When the address packet has been transmitted, the TWINT flag in TWCR is set,
and TWSR is updated with a status code indicating that the address packet has
successfully been sent. The status code will also reflect whether a slave
acknowledged the packet or not.
5. The application software should now examine the value of TWSR, to make sure
that the address packet was successfully transmitted, and that the value of the
ACK bit was as expected. If TWSR indicates otherwise, the application software
might take some special action, like calling an error routine. Assuming that the
status code is as expected, the application must load a data packet into TWDR.
Subsequently, a specific value must be written to TWCR, instructing the TWI
hardware to transmit the data packet present in TWDR. Which value to write is
described later on. However, it is important that the TWINT bit is set in the value
written. Writing a one to TWINT clears the flag. The TWI will not start any opera-
tion as long as the TWINT bit in TWCR is set. Immediately after the application
has cleared TWINT, the TWI will initiate transmission of the data packet.
6. When the data packet has been transmitted, the TWINT flag in TWCR is set, and
TWSR is updated with a status code indicating that the data packet has success-
fully been sent. The status code will also reflect whether a slave acknowledged
the packet or not.
7. The application software should now examine the value of TWSR, to make sure
that the data packet was successfully transmitted, and that the value of the ACK
bit was as expected. If TWSR indicates otherwise, the application software might
take some special action, like calling an error routine. Assuming that the status
code is as expected, the application must write a specific value to TWCR,
instructing the TWI hardware to transmit a STOP condition. Which value to write
is described later on. However, it is important that the TWINT bit is set in the
value written. Writing a one to TWINT clears the flag. The TWI will not start any
operation as long as the TWINT bit in TWCR is set. Immediately after the appli-
cation has cleared TWINT, the TWI will initiate transmission of the STOP
condition. Note that TWINT is NOT set after a STOP condition has been sent.
Even though this example is simple, it shows the principles involved in all TWI transmis-
sions. These can be summarized as follows:
•
When the TWI has finished an operation and expects application response, the
TWINT flag is set. The SCL line is pulled low until TWINT is cleared.
•
When the TWINT flag is set, the user must update all TWI registers with the value
relevant for the next TWI bus cycle. As an example, TWDR must be loaded with the
value to be transmitted in the next bus cycle.
•
After all TWI Register updates and other pending application software tasks have
been completed, TWCR is written. When writing TWCR, the TWINT bit should be
209
2490G–AVR–03/04
set. Writing a one to TWINT clears the flag. The TWI will then commence executing
whatever operation was specified by the TWCR setting.
In the following an assembly and C implementation of the example is given. Note that
the code below assumes that several definitions have been made for example by using
include-files.
210
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Assembly code example(1)
C example(1)
Comments
ldi
r16, (1<<TWINT)|(1<<TWSTA)|
TWCR = (1<<TWINT)|(1<<TWSTA)|
1
2
Send START condition
(1<<TWEN)
TWCR, r16
(1<<TWEN)
out
wait1:
while (!(TWCR & (1<<TWINT)))
Wait for TWINT flag set. This
indicates that the START
in
r16,TWCR
;
condition has been transmitted
sbrs r16,TWINT
rjmp wait1
in
andi r16, 0xF8
cpi r16, START
brne ERROR
r16,TWSR
if ((TWSR & 0xF8) != START)
3
4
Check value of TWI Status
Register. Mask prescaler bits. If
status different from START go to
ERROR
ERROR();
ldi
out
ldi
out
r16, SLA_W
TWDR = SLA_W;
Load SLA_W into TWDR
Register. Clear TWINT bit in
TWCR to start transmission of
address
TWDR, r16
TWCR = (1<<TWINT) | (1<<TWEN);
r16, (1<<TWINT) | (1<<TWEN)
TWCR, r16
wait2:
while (!(TWCR & (1<<TWINT)))
Wait for TWINT flag set. This
indicates that the SLA+W has
been transmitted, and
in
r16,TWCR
;
sbrs r16,TWINT
rjmp wait2
ACK/NACK has been received.
in
r16,TWSR
if ((TWSR & 0xF8) != MT_SLA_ACK)
5
Check value of TWI Status
Register. Mask prescaler bits. If
status different from
andi r16, 0xF8
ERROR();
cpi
r16, MT_SLA_ACK
MT_SLA_ACK go to ERROR
brne ERROR
ldi
out
ldi
out
r16, DATA
TWDR = DATA;
Load DATA into TWDR Register.
Clear TWINT bit in TWCR to
start transmission of data
TWDR, r16
TWCR = (1<<TWINT) | (1<<TWEN);
r16, (1<<TWINT) | (1<<TWEN)
TWCR, r16
wait3:
while (!(TWCR & (1<<TWINT)))
6
7
Wait for TWINT flag set. This
indicates that the DATA has been
transmitted, and ACK/NACK has
been received.
in
r16,TWCR
;
sbrs r16,TWINT
rjmp wait3
in
r16,TWSR
if ((TWSR & 0xF8) != MT_DATA_ACK)
Check value of TWI Status
Register. Mask prescaler bits. If
status different from
andi r16, 0xF8
ERROR();
cpi
r16, MT_DATA_ACK
MT_DATA_ACK go to ERROR
brne ERROR
ldi
out
r16, (1<<TWINT)|(1<<TWEN)|
TWCR = (1<<TWINT)|(1<<TWEN)|
(1<<TWSTO);
Transmit STOP condition
(1<<TWSTO)
TWCR, r16
Note:
1. For I/O Registers located in extended I/O map, “IN”, “OUT”, “SBIS”, “SBIC”, “CBI”, and “SBI” instructions must be replaced
with instructions that allow access to extended I/O. Typically “LDS” and “STS” combined with “SBRS”, “SBRC”, “SBR”, and
“CBR”.
211
2490G–AVR–03/04
Transmission Modes
The TWI can operate in one of four major modes. These are named Master Transmitter
(MT), Master Receiver (MR), Slave Transmitter (ST) and Slave Receiver (SR). Several
of these modes can be used in the same application. As an example, the TWI can use
MT mode to write data into a TWI EEPROM, MR mode to read the data back from the
EEPROM. If other masters are present in the system, some of these might transmit data
to the TWI, and then SR mode would be used. It is the application software that decides
which modes are legal.
The following sections describe each of these modes. Possible status codes are
described along with figures detailing data transmission in each of the modes. These fig-
ures contain the following abbreviations:
S:
START condition
Rs:
R:
REPEATED START condition
Read bit (high level at SDA)
Write bit (low level at SDA)
Acknowledge bit (low level at SDA)
Not acknowledge bit (high level at SDA)
8-bit data byte
W:
A:
A:
Data:
P:
STOP condition
SLA:
Slave Address
In Figure 97 to Figure 103, circles are used to indicate that the TWINT flag is set. The
numbers in the circles show the status code held in TWSR, with the prescaler bits
masked to zero. At these points, actions must be taken by the application to continue or
complete the TWI transfer. The TWI transfer is suspended until the TWINT flag is
cleared by software.
When the TWINT flag is set, the status code in TWSR is used to determine the appropri-
ate software action. For each status code, the required software action and details of the
following serial transfer are given in Table 88 to Table 91. Note that the prescaler bits
are masked to zero in these tables.
212
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Master Transmitter Mode
In the Master Transmitter mode, a number of data bytes are transmitted to a slave
receiver (see Figure 96). In order to enter a Master mode, a START condition must be
transmitted. The format of the following address packet determines whether Master
Transmitter or Master Receiver mode is to be entered. If SLA+W is transmitted, MT
mode is entered, if SLA+R is transmitted, MR mode is entered. All the status codes
mentioned in this section assume that the prescaler bits are zero or are masked to zero.
Figure 96. Data Transfer in Master Transmitter Mode
VCC
Device 1
MASTER
TRANSMITTER
Device 2
SLAVE
RECEIVER
Device 3
R1
R2
Device n
........
SDA
SCL
A START condition is sent by writing the following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
1
0
X
1
0
X
TWEN must be set to enable the Two-wire Serial Interface, TWSTA must be written to
one to transmit a START condition and TWINT must be written to one to clear the
TWINT flag. The TWI will then test the Two-wire Serial Bus and generate a START con-
dition as soon as the bus becomes free. After a START condition has been transmitted,
the TWINT flag is set by hardware, and the status code in TWSR will be 0x08 (see Table
88). In order to enter MT mode, SLA+W must be transmitted. This is done by writing
SLA+W to TWDR. Thereafter the TWINT bit should be cleared (by writing it to one) to
continue the transfer. This is accomplished by writing the following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
0
0
X
1
0
X
When SLA+W have been transmitted and an acknowledgment bit has been received,
TWINT is set again and a number of status codes in TWSR are possible. Possible sta-
tus codes in Master mode are 0x18, 0x20, or 0x38. The appropriate action to be taken
for each of these status codes is detailed in Table 88.
When SLA+W has been successfully transmitted, a data packet should be transmitted.
This is done by writing the data byte to TWDR. TWDR must only be written when
TWINT is high. If not, the access will be discarded, and the Write Collision bit (TWWC)
will be set in the TWCR Register. After updating TWDR, the TWINT bit should be
cleared (by writing it to one) to continue the transfer. This is accomplished by writing the
following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
0
0
X
1
0
X
213
2490G–AVR–03/04
This scheme is repeated until the last byte has been sent and the transfer is ended by
generating a STOP condition or a repeated START condition. A STOP condition is gen-
erated by writing the following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
0
1
X
1
0
X
A REPEATED START condition is generated by writing the following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
1
0
X
1
0
X
After a repeated START condition (state 0x10) the Two-wire Serial Interface can access
the same slave again, or a new slave without transmitting a STOP condition. Repeated
START enables the master to switch between slaves, Master Transmitter mode and
Master Receiver mode without losing control of the bus.
Table 88. Status Codes for Master Transmitter Mode
Status Code
(TWSR)
Application Software Response
To TWCR
Status of the Two-wire Serial
Bus and Two-wire Serial Inter-
face Hardware
Prescaler Bits
are 0
STA
STO
TWINT
TWEA
X
Next Action Taken by TWI Hardware
To/from TWDR
Load SLA+W
0x08
A START condition has been
transmitted
0
0
1
SLA+W will be transmitted;
ACK or NOT ACK will be received
0x10
A repeated START condition
Load SLA+W or
Load SLA+R
0
0
0
0
1
1
X
X
SLA+W will be transmitted;
has been transmitted
ACK or NOT ACK will be received
SLA+R will be transmitted;
Logic will switch to Master Receiver mode
0x18
0x20
0x28
0x30
0x38
SLA+W has been transmitted;
ACK has been received
Load data byte or
0
0
1
X
Data byte will be transmitted and ACK or NOT ACK will
be received
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
Repeated START will be transmitted
STOP condition will be transmitted and
TWSTO flag will be reset
No TWDR action
Load data byte or
1
0
1
0
1
1
X
X
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
SLA+W has been transmitted;
NOT ACK has been received
Data byte will be transmitted and ACK or NOT ACK will
be received
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
Repeated START will be transmitted
STOP condition will be transmitted and
TWSTO flag will be reset
No TWDR action
Load data byte or
1
0
1
0
1
1
X
X
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
Data byte has been transmit-
ted;
Data byte will be transmitted and ACK or NOT ACK will
be received
ACK has been received
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
Repeated START will be transmitted
STOP condition will be transmitted and
TWSTO flag will be reset
No TWDR action
Load data byte or
1
0
1
0
1
1
X
X
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
Data byte has been transmit-
ted;
Data byte will be transmitted and ACK or NOT ACK will
be received
NOT ACK has been received
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
Repeated START will be transmitted
STOP condition will be transmitted and
TWSTO flag will be reset
No TWDR action
1
1
1
X
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
Arbitration lost in SLA+W or
data bytes
No TWDR action or
No TWDR action
0
1
0
0
1
1
X
X
Two-wire Serial Bus will be released and not ad-
dressed slave mode entered
A START condition will be transmitted when the bus
becomes free
214
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 97. Formats and States in the Master Transmitter Mode
MT
Successfull
transmission
to a slave
receiver
S
SLA
W
A
DATA
A
P
$08
$18
$28
Next transfer
started with a
repeated start
condition
RS
SLA
W
R
$10
Not acknowledge
received after the
slave address
A
P
$20
MR
Not acknowledge
received after a data
byte
A
P
$30
Arbitration lost in slave
address or data byte
Other master
continues
Other master
continues
A or A
A or A
$38
A
$38
Arbitration lost and
addressed as slave
Other master
continues
To corresponding
states in slave mode
$68 $78 $B0
Any number of data bytes
and their associated acknowledge bits
From master to slave
From slave to master
DATA
A
This number (contained in TWSR) corresponds
to a defined state of the Two-wire Serial Bus. The
prescaler bits are zero or masked to zero
n
215
2490G–AVR–03/04
Master Receiver Mode
In the Master Receiver mode, a number of data bytes are received from a slave trans-
mitter (see Figure 98). In order to enter a Master mode, a START condition must be
transmitted. The format of the following address packet determines whether Master
Transmitter or Master Receiver mode is to be entered. If SLA+W is transmitted, MT
mode is entered, if SLA+R is transmitted, MR mode is entered. All the status codes
mentioned in this section assume that the prescaler bits are zero or are masked to zero.
Figure 98. Data Transfer in Master Receiver Mode
VCC
Device 1
MASTER
RECEIVER
Device 2
SLAVE
TRANSMITTER
Device 3
R1
R2
Device n
........
SDA
SCL
A START condition is sent by writing the following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
1
0
X
1
0
X
TWEN must be written to one to enable the Two-wire Serial Interface, TWSTA must be
written to one to transmit a START condition and TWINT must be set to clear the TWINT
flag. The TWI will then test the Two-wire Serial Bus and generate a START condition as
soon as the bus becomes free. After a START condition has been transmitted, the
TWINT flag is set by hardware, and the status code in TWSR will be 0x08 (see Table
88). In order to enter MR mode, SLA+R must be transmitted. This is done by writing
SLA+R to TWDR. Thereafter the TWINT bit should be cleared (by writing it to one) to
continue the transfer. This is accomplished by writing the following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
0
0
X
1
0
X
When SLA+R have been transmitted and an acknowledgment bit has been received,
TWINT is set again and a number of status codes in TWSR are possible. Possible sta-
tus codes in Master mode are 0x38, 0x40, or 0x48. The appropriate action to be taken
for each of these status codes is detailed in Table 89. Received data can be read from
the TWDR Register when the TWINT flag is set high by hardware. This scheme is
repeated until the last byte has been received. After the last byte has been received, the
MR should inform the ST by sending a NACK after the last received data byte. The
transfer is ended by generating a STOP condition or a repeated START condition. A
STOP condition is generated by writing the following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
0
1
X
1
0
X
A REPEATED START condition is generated by writing the following value to TWCR:
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
1
0
X
1
0
X
216
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
After a repeated START condition (state 0x10) the Two-wire Serial Interface can access
the same slave again, or a new slave without transmitting a STOP condition. Repeated
START enables the master to switch between slaves, Master Transmitter mode and
Master Receiver mode without losing control over the bus.
Table 89. Status Codes for Master Receiver Mode
Status Code
(TWSR)
Application Software Response
To TWCR
Status of the Two-wire Serial
Bus and Two-wire Serial Inter-
face Hardware
Prescaler Bits
are 0
STA
STO
TWINT
TWEA
X
Next Action Taken by TWI Hardware
To/from TWDR
Load SLA+R
0x08
A START condition has been
transmitted
0
0
1
SLA+R will be transmitted
ACK or NOT ACK will be received
0x10
A repeated START condition
Load SLA+R or
Load SLA+W
0
0
0
0
1
1
X
X
SLA+R will be transmitted
has been transmitted
ACK or NOT ACK will be received
SLA+W will be transmitted
Logic will switch to Master Transmitter mode
0x38
Arbitration lost in SLA+R or
NOT ACK bit
No TWDR action or
No TWDR action
0
1
0
0
1
1
X
X
Two-wire Serial Bus will be released and not ad-
dressed Slave mode will be entered
A START condition will be transmitted when the bus
becomes free
0x40
0x48
SLA+R has been transmitted;
ACK has been received
No TWDR action or
No TWDR action
0
0
0
0
1
1
0
1
Data byte will be received and NOT ACK will be
returned
Data byte will be received and ACK will be returned
SLA+R has been transmitted;
NOT ACK has been received
No TWDR action or
1
0
1
X
Repeated START will be transmitted
STOP condition will be transmitted and TWSTO flag
will be reset
No TWDR action or
0
1
1
X
No TWDR action
1
1
1
X
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
0x50
0x58
Data byte has been received;
ACK has been returned
Read data byte or
Read data byte
0
0
0
0
1
1
0
1
Data byte will be received and NOT ACK will be
returned
Data byte will be received and ACK will be returned
Data byte has been received;
NOT ACK has been returned
Read data byte or
1
0
1
X
Repeated START will be transmitted
STOP condition will be transmitted and TWSTO flag
will be reset
Read data byte or
0
1
1
X
Read data byte
1
1
1
X
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
217
2490G–AVR–03/04
Figure 99. Formats and States in the Master Receiver Mode
MR
Successfull
S
SLA
R
A
DATA
A
DATA
A
P
reception
from a slave
receiver
$08
$40
$50
$58
Next transfer
started with a
repeated start
condition
RS
SLA
R
$10
Not acknowledge
received after the
slave address
W
A
P
$48
MT
Arbitration lost in slave
address or data byte
Other master
continues
Other master
continues
A or A
A
$38
A
$38
Arbitration lost and
addressed as slave
Other master
continues
To corresponding
states in slave mode
$68 $78 $B0
Any number of data bytes
and their associated acknowledge bits
From master to slave
From slave to master
DATA
A
This number (contained in TWSR) corresponds
to a defined state of the Two-wire Serial Bus. The
prescaler bits are zero or masked to zero
n
218
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Slave Receiver Mode
In the Slave Receiver mode, a number of data bytes are received from a master trans-
mitter (see Figure 100). All the status codes mentioned in this section assume that the
prescaler bits are zero or are masked to zero.
Figure 100. Data Transfer in Slave Receiver Mode
VCC
Device 1
SLAVE
RECEIVER
Device 2
MASTER
TRANSMITTER
Device 3
R1
R2
Device n
........
SDA
SCL
To initiate the Slave Receiver mode, TWAR and TWCR must be initialized as follows:
TWAR
TWA6
TWA5
TWA4
TWA3
TWA2
TWA1
TWA0
TWGCE
Value
Device’s Own Slave Address
The upper seven bits are the address to which the Two-wire Serial Interface will respond
when addressed by a master. If the LSB is set, the TWI will respond to the general call
address (0x00), otherwise it will ignore the general call address.
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
0
1
0
0
0
1
0
X
TWEN must be written to one to enable the TWI. The TWEA bit must be written to one
to enable the acknowledgment of the device’s own slave address or the general call
address. TWSTA and TWSTO must be written to zero.
When TWAR and TWCR have been initialized, the TWI waits until it is addressed by its
own slave address (or the general call address if enabled) followed by the data direction
bit. If the direction bit is “0” (write), the TWI will operate in SR mode, otherwise ST mode
is entered. After its own slave address and the write bit have been received, the TWINT
flag is set and a valid status code can be read from TWSR. The status code is used to
determine the appropriate software action. The appropriate action to be taken for each
status code is detailed in Table 90. The Slave Receiver mode may also be entered if
arbitration is lost while the TWI is in the Master mode (see states 0x68 and 0x78).
If the TWEA bit is reset during a transfer, the TWI will return a “Not Acknowledge” (“1”)
to SDA after the next received data byte. This can be used to indicate that the slave is
not able to receive any more bytes. While TWEA is zero, the TWI does not acknowledge
its own slave address. However, the Two-wire Serial Bus is still monitored and address
recognition may resume at any time by setting TWEA. This implies that the TWEA bit
may be used to temporarily isolate the TWI from the Two-wire Serial Bus.
In all sleep modes other than Idle mode, the clock system to the TWI is turned off. If the
TWEA bit is set, the interface can still acknowledge its own slave address or the general
call address by using the Two-wire Serial Bus clock as a clock source. The part will then
wake-up from sleep and the TWI will hold the SCL clock low during the wake up and
until the TWINT flag is cleared (by writing it to one). Further data reception will be car-
219
2490G–AVR–03/04
ried out as normal, with the AVR clocks running as normal. Observe that if the AVR is
set up with a long start-up time, the SCL line may be held low for a long time, blocking
other data transmissions.
Note that the Two-wire Serial Interface Data Register – TWDR does not reflect the last
byte present on the bus when waking up from these Sleep modes.
Table 90. Status Codes for Slave Receiver Mode
Status Code
(TWSR)
Application Software Response
To TWCR
Status of the Two-wire Serial Bus
and Two-wire Serial Interface
Hardware
Prescaler Bits
Are 0
STA
STO
TWINT
TWEA
0
Next Action Taken by TWI Hardware
To/from TWDR
0x60
0x68
0x70
0x78
Own SLA+W has been received;
ACK has been returned
No TWDR action or
X
0
1
Data byte will be received and NOT ACK will be
returned
No TWDR action
No TWDR action or
X
X
0
0
1
1
1
0
Data byte will be received and ACK will be returned
Arbitration lost in SLA+R/W as
master; own SLA+W has been
received; ACK has been returned
Data byte will be received and NOT ACK will be
returned
No TWDR action
No TWDR action or
X
X
0
0
1
1
1
0
Data byte will be received and ACK will be returned
General call address has been
Data byte will be received and NOT ACK will be
received; ACK has been returned
returned
No TWDR action
No TWDR action or
X
X
0
0
1
1
1
0
Data byte will be received and ACK will be returned
Arbitration lost in SLA+R/W as
master; General call address has
been received; ACK has been
returned
Previously addressed with own
SLA+W; data has been received;
ACK has been returned
Previously addressed with own
SLA+W; data has been received;
NOT ACK has been returned
Data byte will be received and NOT ACK will be
returned
No TWDR action
Read data byte or
X
X
0
0
1
1
1
0
Data byte will be received and ACK will be returned
0x80
0x88
Data byte will be received and NOT ACK will be
returned
Read data byte
Read data byte or
X
0
0
0
1
1
1
0
Data byte will be received and ACK will be returned
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA
Switched to the not addressed Slave mode;
own SLA will be recognized;
Read data byte or
0
0
1
1
GCA will be recognized if TWGCE = “1”
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA;
a START condition will be transmitted when the bus
becomes free
Read data byte or
1
0
1
0
Read data byte
1
0
0
1
1
1
0
Switched to the not addressed Slave mode;
own SLA will be recognized;
GCA will be recognized if TWGCE = “1”;
a START condition will be transmitted when the bus
becomes free
0x90
0x98
Previously addressed with
Read data byte or
X
Data byte will be received and NOT ACK will be
general call; data has been re-
returned
ceived; ACK has been returned
Read data byte
Read data byte or
X
0
0
0
1
1
1
0
Data byte will be received and ACK will be returned
Previously addressed with
general call; data has been
received; NOT ACK has been
returned
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA
Switched to the not addressed Slave mode;
own SLA will be recognized;
Read data byte or
0
0
1
1
GCA will be recognized if TWGCE = “1”
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA;
a START condition will be transmitted when the bus
becomes free
Read data byte or
1
0
1
0
Read data byte
No Action
1
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized;
GCA will be recognized if TWGCE = “1”;
a START condition will be transmitted when the bus
becomes free
0xA0
A STOP condition or repeated
START condition has been
received while still addressed as
slave
0
0
0
0
1
1
0
1
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA
Switched to the not addressed Slave mode;
own SLA will be recognized;
GCA will be recognized if TWGCE = “1”
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA;
a START condition will be transmitted when the bus
becomes free
1
1
0
0
1
1
0
1
Switched to the not addressed Slave mode;
own SLA will be recognized;
GCA will be recognized if TWGCE = “1”;
a START condition will be transmitted when the bus
becomes free
220
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 101. Formats and States in the Slave Receiver Mode
Reception of the own
S
SLA
W
A
DATA
A
DATA
A
P or S
slave address and one or
more data bytes. All are
acknowledged
$60
$80
$80
A
$A0
Last data byte received
is not acknowledged
P or S
$88
Arbitration lost as master
and addressed as slave
A
$68
A
Reception of the general call
address and one or more data
bytes
General Call
DATA
A
DATA
A
P or S
$70
$90
$90
A
$A0
Last data byte received is
not acknowledged
P or S
$98
Arbitration lost as master and
addressed as slave by general call
A
$78
Any number of data bytes
and their associated acknowledge bits
From master to slave
From slave to master
DATA
A
This number (contained in TWSR) corresponds
to a defined state of the Two-wire Serial Bus. The
prescaler bits are zero or masked to zero
n
221
2490G–AVR–03/04
Slave Transmitter Mode
In the Slave Transmitter mode, a number of data bytes are transmitted to a master
receiver (see Figure 102). All the status codes mentioned in this section assume that the
prescaler bits are zero or are masked to zero.
Figure 102. Data Transfer in Slave Transmitter Mode
VCC
Device 1
SLAVE
TRANSMITTER
Device 2
MASTER
RECEIVER
Device 3
R1
R2
Device n
........
SDA
SCL
To initiate the Slave Transmitter mode, TWAR and TWCR must be initialized as follows:
TWAR
TWA6
TWA5
TWA4
TWA3
TWA2
TWA1
TWA0
TWGCE
Value
Device’s Own Slave Address
The upper seven bits are the address to which the Two-wire Serial Interface will respond
when addressed by a master. If the LSB is set, the TWI will respond to the general call
address (0x00), otherwise it will ignore the general call address.
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
0
1
0
0
0
1
0
X
TWEN must be written to one to enable the TWI. The TWEA bit must be written to one
to enable the acknowledgment of the device’s own slave address or the general call
address. TWSTA and TWSTO must be written to zero.
When TWAR and TWCR have been initialized, the TWI waits until it is addressed by its
own slave address (or the general call address if enabled) followed by the data direction
bit. If the direction bit is “1” (read), the TWI will operate in ST mode, otherwise SR mode
is entered. After its own slave address and the write bit have been received, the TWINT
flag is set and a valid status code can be read from TWSR. The status code is used to
determine the appropriate software action. The appropriate action to be taken for each
status code is detailed in Table 91. The Slave Transmitter mode may also be entered if
arbitration is lost while the TWI is in the Master mode (see state 0xB0).
If the TWEA bit is written to zero during a transfer, the TWI will transmit the last byte of
the transfer. State 0xC0 or state 0xC8 will be entered, depending on whether the master
receiver transmits a NACK or ACK after the final byte. The TWI is switched to the not
addressed Slave mode, and will ignore the master if it continues the transfer. Thus the
master receiver receives all “1” as serial data. State 0xC8 is entered if the master
demands additional data bytes (by transmitting ACK), even though the slave has trans-
mitted the last byte (TWEA zero and expecting NACK from the master).
While TWEA is zero, the TWI does not respond to its own slave address. However, the
Two-wire Serial Bus is still monitored and address recognition may resume at any time
by setting TWEA. This implies that the TWEA bit may be used to temporarily isolate the
TWI from the Two-wire Serial Bus.
222
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
In all sleep modes other than Idle mode, the clock system to the TWI is turned off. If the
TWEA bit is set, the interface can still acknowledge its own slave address or the general
call address by using the Two-wire Serial Bus clock as a clock source. The part will then
wake up from sleep and the TWI will hold the SCL clock will low during the wake up and
until the TWINT flag is cleared (by writing it to one). Further data transmission will be
carried out as normal, with the AVR clocks running as normal. Observe that if the AVR is
set up with a long start-up time, the SCL line may be held low for a long time, blocking
other data transmissions.
Note that the Two-wire Serial Interface Data Register – TWDR – does not reflect the last
byte present on the bus when waking up from these sleep modes.
Table 91. Status Codes for Slave Transmitter Mode
Status Code
(TWSR)
Application Software Response
To TWCR
Status of the Two-wire Serial Bus
and Two-wire Serial Interface
Hardware
Prescaler
To/from TWDR
Load data byte or
Load data byte
STA
STO
TWINT
TWEA
0
Bits are 0
Next Action Taken by TWI Hardware
0xA8
0xB0
0xB8
0xC0
Own SLA+R has been received;
X
0
1
Last data byte will be transmitted and NOT ACK should
ACK has been returned
be received
X
0
1
1
Data byte will be transmitted and ACK should be re-
ceived
Arbitration lost in SLA+R/W as
master; own SLA+R has been
received; ACK has been returned
Load data byte or
Load data byte
X
X
0
0
1
1
0
1
Last data byte will be transmitted and NOT ACK should
be received
Data byte will be transmitted and ACK should be re-
ceived
Data byte in TWDR has been
transmitted; ACK has been
received
Load data byte or
Load data byte
X
X
0
0
1
1
0
1
Last data byte will be transmitted and NOT ACK should
be received
Data byte will be transmitted and ACK should be re-
ceived
Data byte in TWDR has been
transmitted; NOT ACK has been
received
No TWDR action or
No TWDR action or
0
0
0
0
1
1
0
1
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA
Switched to the not addressed Slave mode;
own SLA will be recognized;
GCA will be recognized if TWGCE = “1”
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA;
a START condition will be transmitted when the bus
becomes free
No TWDR action or
No TWDR action
1
1
0
0
1
1
0
1
Switched to the not addressed Slave mode;
own SLA will be recognized;
GCA will be recognized if TWGCE = “1”;
a START condition will be transmitted when the bus
becomes free
0xC8
Last data byte in TWDR has been
transmitted (TWEA = “0”); ACK
has been received
No TWDR action or
No TWDR action or
0
0
0
0
1
1
0
1
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA
Switched to the not addressed Slave mode;
own SLA will be recognized;
GCA will be recognized if TWGCE = “1”
Switched to the not addressed Slave mode;
no recognition of own SLA or GCA;
a START condition will be transmitted when the bus
becomes free
No TWDR action or
No TWDR action
1
1
0
0
1
1
0
1
Switched to the not addressed Slave mode;
own SLA will be recognized;
GCA will be recognized if TWGCE = “1”;
a START condition will be transmitted when the bus
becomes free
223
2490G–AVR–03/04
Figure 103. Formats and States in the Slave Transmitter Mode
Reception of the own
slave address and one or
more data bytes
S
SLA
R
A
DATA
A
DATA
A
P or S
$A8
A
$B8
$C0
Arbitration lost as master
and addressed as slave
$B0
Last data byte transmitted.
Switched to not addressed
slave (TWEA = '0')
A
All 1's
P or S
$C8
Any number of data bytes
and their associated acknowledge bits
From master to slave
From slave to master
DATA
A
This number (contained in TWSR) corresponds
to a defined state of the Two-wire Serial Bus. The
prescaler bits are zero or masked to zero
n
Miscellaneous States
There are two status codes that do not correspond to a defined TWI state, see Table 92.
Status 0xF8 indicates that no relevant information is available because the TWINT flag
is not set. This occurs between other states, and when the TWI is not involved in a serial
transfer.
Status 0x00 indicates that a bus error has occurred during a Two-wire Serial Bus trans-
fer. A bus error occurs when a START or STOP condition occurs at an illegal position in
the format frame. Examples of such illegal positions are during the serial transfer of an
address byte, a data byte, or an acknowledge bit. When a bus error occurs, TWINT is
set. To recover from a bus error, the TWSTO flag must set and TWINT must be cleared
by writing a logic one to it. This causes the TWI to enter the not addressed Slave mode
and to clear the TWSTO flag (no other bits in TWCR are affected). The SDA and SCL
lines are released, and no STOP condition is transmitted.
Table 92. Miscellaneous States
Status Code
Application Software Response
(TWSR)
Status of the Two-wire Serial
To TWCR
STO TWINT
Prescaler Bits
are 0
Bus and Two-wire Serial Inter-
face hardware
STA
TWEA
X
Next Action Taken by TWI Hardware
Wait or proceed current transfer
To/from TWDR
0xF8
No relevant state information
available; TWINT = “0”
No TWDR action
No TWCR action
0x00
Bus error due to an illegal
No TWDR action
0
1
1
Only the internal hardware is affected, no STOP condi-
tion is sent on the bus. In all cases, the bus is released
and TWSTO is cleared.
START or STOP condition
224
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Combining Several TWI
Modes
In some cases, several TWI modes must be combined in order to complete the desired
action. Consider for example reading data from a serial EEPROM. Typically, such a
transfer involves the following steps:
1. The transfer must be initiated.
2. The EEPROM must be instructed what location should be read.
3. The reading must be performed.
4. The transfer must be finished.
Note that data is transmitted both from Master to Slave and vice versa. The Master must
instruct the slave what location it wants to read, requiring the use of the MT mode. Sub-
sequently, data must be read from the slave, implying the use of the MR mode. Thus,
the transfer direction must be changed. The Master must keep control of the bus during
all these steps, and the steps should be carried out as an atomic operation. If this princi-
ple is violated in a multimaster system, another master can alter the data pointer in the
EEPROM between steps 2 and 3, and the master will read the wrong data location.
Such a change in transfer direction is accomplished by transmitting a REPEATED
START between the transmission of the address byte and reception of the data. After a
REPEATED START, the master keeps ownership of the bus. The following figure shows
the flow in this transfer.
Figure 104. Combining Several TWI Modes to Access a Serial EEPROM
Master Transmitter
Master Receiver
S
SLA+W
A
ADDRESS
A
Rs
SLA+R
A
DATA
A
P
S = START
Transmitted from master to slave
Rs = REPEATED START
Transmitted from slave to master
P = STOP
Multi-master Systems
and Arbitration
If multiple masters are connected to the same bus, transmissions may be initiated simul-
taneously by one or more of them. The TWI standard ensures that such situations are
handled in such a way that one of the masters will be allowed to proceed with the trans-
fer, and that no data will be lost in the process. An example of an arbitration situation is
depicted below, where two masters are trying to transmit data to a slave receiver.
Figure 105. An Arbitration Example
VCC
Device 1
MASTER
TRANSMITTER
Device 3
SLAVE
RECEIVER
Device 2
MASTER
TRANSMITTER
R1
R2
Device n
........
SDA
SCL
225
2490G–AVR–03/04
Several different scenarios may arise during arbitration, as described below:
•
Two or more masters are performing identical communication with the same slave.
In this case, neither the slave nor any of the masters will know about the bus
contention.
•
Two or more masters are accessing the same slave with different data or direction
bit. In this case, arbitration will occur, either in the READ/WRITE bit or in the data
bits. The masters trying to output a one on SDA while another master outputs a zero
will lose the arbitration. Losing masters will switch to not addressed Slave mode or
wait until the bus is free and transmit a new START condition, depending on
application software action.
•
Two or more masters are accessing different slaves. In this case, arbitration will
occur in the SLA bits. Masters trying to output a one on SDA while another master
outputs a zero will lose the arbitration. Masters losing arbitration in SLA will switch to
Slave mode to check if they are being addressed by the winning master. If
addressed, they will switch to SR or ST mode, depending on the value of the
READ/WRITE bit. If they are not being addressed, they will switch to not addressed
Slave mode or wait until the bus is free and transmit a new START condition,
depending on application software action.
This is summarized in Figure 106. Possible status values are given in circles.
Figure 106. Possible Status Codes Caused by Arbitration
START
SLA
Data
STOP
Arbitration lost in SLA
Arbitration lost in Data
Own
No
38
TWI bus will be released and not addressed slave mode will be entered
A START condition will be transmitted when the bus becomes free
Address / General Call
received
Yes
Write
68/78
B0
Data byte will be received and NOT ACK will be returned
Data byte will be received and ACK will be returned
Direction
Read
Last data byte will be transmitted and NOT ACK should be received
Data byte will be transmitted and ACK should be received
226
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Analog Comparator
The Analog Comparator compares the input values on the positive pin AIN0 and nega-
tive pin AIN1. When the voltage on the positive pin AIN0 is higher than the voltage on
the negative pin AIN1, the Analog Comparator output, ACO, is set. The comparator’s
output 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 dia-
gram of the comparator and its surrounding logic is shown in Figure 107.
Figure 107. Analog Comparator Block Diagram(1)(2)
BANDGAP
REFERENCE
ACBG
ACME
ADEN
ADC MULTIPLEXER
OUTPUT 1)
Notes: 1. See Table 94 on page 229.
2. Refer to Figure 1 on page 2 and Table 30 on page 72 for Analog Comparator pin
placement.
Special Function IO Register –
SFIOR
Bit
7
6
–
R
0
5
–
R
0
4
–
R
0
3
ACME
R/W
0
2
1
PSR2
R/W
0
0
PSR10
R/W
0
TSM
R/W
0
PUD
R/W
0
SFIOR
Read/Write
Initial Value
• Bit 3 – ACME: Analog Comparator Multiplexer Enable
When this bit is written logic one and the ADC is switched off (ADEN in ADCSRA is
zero), the ADC multiplexer selects the negative input to the Analog Comparator. When
this bit is written logic zero, AIN1 is applied to the negative input of the Analog Compar-
ator. For a detailed description of this bit, see “Analog Comparator Multiplexed Input” on
page 229.
227
2490G–AVR–03/04
Analog Comparator Control
and Status Register – ACSR
Bit
7
6
ACBG
R/W
0
5
ACO
R
4
ACI
R/W
0
3
ACIE
R/W
0
2
ACIC
R/W
0
1
ACIS1
R/W
0
0
ACIS0
R/W
0
ACD
R/W
0
ACSR
Read/Write
Initial Value
N/A
• Bit 7 – ACD: Analog Comparator Disable
When this bit is written logic 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 Compar-
ator Interrupt must be disabled by clearing the ACIE bit in ACSR. Otherwise an interrupt
can occur when the bit is changed.
• Bit 6 – ACBG: Analog Comparator Bandgap Select
When this bit is set, a fixed bandgap reference voltage replaces the positive input to the
Analog Comparator. When this bit is cleared, AIN0 is applied to the positive input of the
Analog Comparator. See “Internal Voltage Reference” on page 54.
• Bit 5 – ACO: Analog Comparator Output
The output of the Analog Comparator is synchronized and then directly connected to
ACO. The synchronization introduces a delay of 1 - 2 clock cycles.
• Bit 4 – ACI: Analog Comparator Interrupt Flag
This bit is set by hardware when a comparator output event triggers the interrupt mode
defined by ACIS1 and ACIS0. The Analog Comparator Interrupt routine is executed if
the ACIE bit is set and the I-bit in SREG is set. ACI is cleared by hardware when execut-
ing the corresponding interrupt handling vector. Alternatively, ACI is cleared by writing a
logic one to the flag.
• Bit 3 – ACIE: Analog Comparator Interrupt Enable
When the ACIE bit is written logic one and the I-bit in the Status Register is set, the Ana-
log Comparator interrupt is activated. When written logic zero, the interrupt is disabled.
• Bit 2 – ACIC: Analog Comparator Input Capture Enable
When written logic 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
written logic zero, no connection between the Analog Comparator and the Input Capture
function exists. To make the comparator trigger the Timer/Counter1 Input Capture inter-
rupt, the TICIE1 bit in the Timer Interrupt Mask Register (TIMSK) must be set.
228
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• Bits 1, 0 – ACIS1, ACIS0: Analog Comparator Interrupt Mode Select
These bits determine which comparator events that trigger the Analog Comparator inter-
rupt. The different settings are shown in Table 93.
Table 93. ACIS1/ACIS0 Settings
ACIS1
ACIS0
Interrupt Mode
0
0
1
1
0
1
0
1
Comparator Interrupt on Output Toggle.
Reserved
Comparator Interrupt on Falling Output Edge.
Comparator Interrupt on Rising Output Edge.
When changing the ACIS1/ACIS0 bits, the Analog Comparator Interrupt must be dis-
abled by clearing its Interrupt Enable bit in the ACSR Register. Otherwise an interrupt
can occur when the bits are changed.
Analog Comparator
Multiplexed Input
It is possible to select any of the ADC7..0 pins to replace the negative input to the Ana-
log Comparator. The ADC multiplexer is used to select this input, and consequently, the
ADC must be switched off to utilize this feature. If the Analog Comparator Multiplexer
Enable bit (ACME in SFIOR) is set and the ADC is switched off (ADEN in ADCSRA is
zero), MUX2..0 in ADMUX select the input pin to replace the negative input to the Ana-
log Comparator, as shown in Table 94. If ACME is cleared or ADEN is set, AIN1 is
applied to the negative input to the Analog Comparator.
Table 94. Analog Comparator Multiplexed Input
ACME
ADEN
MUX2..0
xxx
Analog Comparator Negative Input
0
1
1
1
1
1
1
1
1
1
x
1
0
0
0
0
0
0
0
0
AIN1
xxx
AIN1
000
ADC0
ADC1
ADC2
ADC3
ADC4
ADC5
ADC6
ADC7
001
010
011
100
101
110
111
229
2490G–AVR–03/04
Analog to Digital
Converter
Features
• 10-bit Resolution
• 0.5 LSB Integral Non-linearity
•
2 LSB Absolute Accuracy
• 65 - 260 µs Conversion Time
• Up to 15 kSPS at Maximum Resolution
• Eight Multiplexed Single Ended Input Channels
• Seven Differential Input Channels
• Two Differential Input Channels with Optional Gain of 10x and 200x
• Optional Left Adjustment for ADC Result Readout
• 0 - VCC ADC Input Voltage Range
• Selectable 2.56V ADC Reference Voltage
• Free Running or Single Conversion Mode
• ADC Start Conversion by Auto Triggering on Interrupt Sources
• Interrupt on ADC Conversion Complete
• Sleep Mode Noise Canceler
The ATmega64 features a 10-bit successive approximation ADC. The ADC is con-
nected to an 8-channel Analog Multiplexer which allows eight single-ended voltage
inputs constructed from the pins of Port F. The single-ended voltage inputs refer to 0V
(GND).
The device also supports 16 differential voltage input combinations. Two of the differen-
tial inputs (ADC1, ADC0 and ADC3, ADC2) are equipped with a programmable gain
stage, providing amplification steps of 0 dB (1x), 20 dB (10x), or 46 dB (200x) on the dif-
ferential input voltage before the A/D conversion. Seven differential analog input
channels share a common negative terminal (ADC1), while any other ADC input can be
selected as the positive input terminal. If 1x or 10x gain is used, 8-bit resolution can be
expected. If 200x gain is used, 7-bit resolution can be expected.
The ADC contains a Sample and Hold circuit 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 108.
The ADC has a separate analog supply voltage pin, AVCC. AVCC must not differ more
than 0.3V from VCC. See the paragraph “ADC Noise Canceler” on page 238 on how to
connect this pin.
Internal reference voltages of nominally 2.56V or AVCC are provided On-chip. The volt-
age reference may be externally decoupled at the AREF pin by a capacitor for better
noise performance.
230
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 108. Analog to Digital Converter Block Schematic
ADC CONVERSION
COMPLETE IRQ
INTERRUPT
FLAGS
ADTS[2:0]
8-BIT DATA BUS
15
0
ADC MULTIPLEXER
SELECT (ADMUX)
ADC CTRL. & STATUS
REGISTER (ADCSRA)
ADC DATA REGISTER
(ADCH/ADCL)
TRIGGER
SELECT
MUX DECODER
PRESCALER
START
CONVERSION LOGIC
AVCC
INTERNAL 2.56V
REFERENCE
SAMPLE & HOLD
COMPARATOR
AREF
GND
10-BIT DAC
-
+
BANDGAP
REFERENCE
ADC7
ADC6
ADC5
ADC4
ADC3
ADC2
ADC1
ADC0
SINGLE ENDED / DIFFERENTIAL SELECTION
POS.
INPUT
MUX
ADC MULTIPLEXER
OUTPUT
GAIN
AMPLIFIER
+
-
NEG.
INPUT
MUX
Operation
The ADC converts an analog input voltage to a 10-bit digital value through successive
approximation. The minimum value represents GND and the maximum value represents
the voltage on the AREF pin minus 1 LSB. Optionally, AVCC or an internal 2.56V refer-
ence voltage may be connected to the AREF pin by writing to the REFSn bits in the
ADMUX Register. The internal voltage reference may thus be decoupled by an external
capacitor at the AREF pin to improve noise immunity.
The analog input channel and differential gain are selected by writing to the MUX bits in
ADMUX. Any of the ADC input pins, as well as GND and a fixed bandgap voltage refer-
ence, can be selected as single ended inputs to the ADC. A selection of ADC input pins
can be selected as positive and negative inputs to the differential gain amplifier.
If differential channels are selected, the differential gain stage amplifies the voltage dif-
ference between the selected input channel pair by the selected gain factor. This
amplified value then becomes the analog input to the ADC. If single ended channels are
used, the gain amplifier is bypassed altogether.
231
2490G–AVR–03/04
The ADC is enabled by setting the ADC Enable bit, ADEN in ADCSRA. Voltage refer-
ence and input channel selections will not go into effect until ADEN is set. The ADC
does not consume power when ADEN is cleared, so it is recommended to switch off the
ADC before entering power saving sleep modes.
The ADC generates a 10-bit result which is presented in the ADC Data Registers,
ADCH and ADCL. By default, the result is presented right adjusted, but can optionally
be presented left adjusted by setting the ADLAR bit in ADMUX.
If the result is left adjusted and no more than 8-bit precision is required, it is sufficient to
read ADCH. Otherwise, ADCL must be read first, then ADCH, to ensure that the content
of the data registers belongs to the same conversion. 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, neither register is updated and the result from the con-
version is lost. When ADCH is read, ADC access to the ADCH and ADCL Registers is
re-enabled.
The ADC has its own interrupt which can be triggered when a conversion completes.
When ADC access to the data registers is prohibited between reading of ADCH and
ADCL, the interrupt will trigger even if the result is lost.
Starting a Conversion
A single conversion is started by writing a logical one to the ADC Start Conversion bit,
ADSC. This bit stays high as long as the conversion is in progress and will be cleared 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.
Alternatively, a conversion can be triggered automatically by various sources. Auto Trig-
gering is enabled by setting the ADC Auto Trigger Enable bit, ADATE in ADCSRA. The
trigger source is selected by setting the ADC Trigger Select bits, ADTS in ADCSRB (see
description of the ADTS bits for a list of the trigger sources). When a positive edge
occurs on the selected trigger signal, the ADC prescaler is reset and a conversion is
started. This provides a method of starting conversions at fixed intervals. If the trigger
signal still is set when the conversion completes, a new conversion will not be started. If
another positive edge occurs on the trigger signal during conversion, the edge will be
ignored. Note that an interrupt flag will be set even if the specific interrupt is disabled or
the Global Interrupt Enable bit in SREG is cleared. A conversion can thus be triggered
without causing an interrupt. However, the interrupt flag must be cleared in order to trig-
ger a new conversion at the next interrupt event.
232
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 109. ADC Auto Trigger Logic
ADTS[2:0]
PRESCALER
CLKADC
START
ADIF
ADATE
SOURCE 1
.
.
.
CONVERSION
LOGIC
.
EDGE
DETECTOR
SOURCE n
ADSC
Using the ADC Interrupt Flag as a trigger source makes the ADC start a new conversion
as soon as the ongoing conversion has finished. The ADC then operates in Free Run-
ning mode, constantly sampling and updating the ADC Data Register. The first
conversion must be started by writing a logical one to the ADSC bit in ADCSRA. In this
mode the ADC will perform successive conversions independently of whether the ADC
Interrupt Flag, ADIF is cleared or not.
If Auto Triggering is enabled, single conversions can be started by writing ADSC in
ADCSRA to one. ADSC can also be used to determine if a conversion is in progress.
The ADSC bit will be read as one during a conversion, independently of how the conver-
sion was started.
Prescaling and
Figure 110. ADC Prescaler
Conversion Timing
ADEN
START
Reset
7-BIT ADC PRESCALER
CK
ADPS0
ADPS1
ADPS2
ADC CLOCK SOURCE
By default, the successive approximation circuitry requires an input clock frequency
between 50 kHz and 200 kHz to get maximum resolution. If a lower resolution than 10
bits is needed, the input clock frequency to the ADC can be higher than 200 kHz to get a
higher sample rate.
233
2490G–AVR–03/04
The ADC module contains a prescaler, which generates an acceptable ADC clock fre-
quency from any CPU frequency above 100 kHz. The prescaling is set by the ADPS bits
in ADCSRA. The prescaler starts counting from the moment the ADC is switched on by
setting the ADEN bit in ADCSRA. The prescaler keeps running for as long as the ADEN
bit is set, and is continuously reset when ADEN is low.
When initiating a single ended conversion by setting the ADSC bit in ADCSRA, the con-
version starts at the following rising edge of the ADC clock cycle. See “Differential Gain
Channels” on page 236 for details on differential conversion timing.
A normal conversion takes 13 ADC clock cycles. The first conversion after the ADC is
switched on (ADEN in ADCSRA is set) takes 25 ADC clock cycles in order to initialize
the analog circuitry.
The actual sample-and-hold takes place 1.5 ADC clock cycles after the start of a normal
conversion and 13.5 ADC clock cycles after the start of a first conversion. When a con-
version is complete, the result is written to the ADC Data Registers, and ADIF is set. In
Single Conversion mode, ADSC is cleared simultaneously. The software may then set
ADSC again, and a new conversion will be initiated on the first rising ADC clock edge.
When Auto Triggering is used, the prescaler is reset when the trigger event occurs. This
assures a fixed delay from the trigger event to the start of conversion. In this mode, the
sample-and-hold takes place two ADC clock cycles after the rising edge on the trigger
source signal. Three additional CPU clock cycles are used for synchronization logic.
When using Differential mode, along with auto trigging from a source other that the ADC
Conversion Complete, each conversion will require 25 ADC clocks. This is because the
ADC must be disabled and re-enabled after every conversion.
In Free Running mode, a new conversion will be started immediately after the conver-
sion completes, while ADSC remains high. For a summary of conversion times, see
Table 95.
Figure 111. ADC Timing Diagram, First Conversion (Single Conversion Mode)
Next
First Conversion
Conversion
Cycle Number
1
2
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1
2
3
ADC Clock
ADEN
ADSC
ADIF
MSB of Result
LSB of Result
ADCH
ADCL
MUX and REFS
Update
Conversion
Complete
MUX and REFS
Update
Sample & Hold
234
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 112. ADC Timing Diagram, Single Conversion
One Conversion
Next Conversion
1
2
3
4
5
6
7
8
9
10
11
12
13
1
2
3
Cycle Number
ADC Clock
ADSC
ADIF
ADCH
MSB of Result
LSB of Result
ADCL
Sample & Hold
Conversion
Complete
MUX and REFS
Update
MUX and REFS
Update
Figure 113. ADC Timing Diagram, Auto Triggered Conversion
One Conversion
Next Conversion
1
2
3
4
5
6
7
8
9
10
11
12
13
1
2
Cycle Number
ADC Clock
Trigger
Source
ADATE
ADIF
ADCH
ADCL
MSB of Result
LSB of Result
Sample &
Hold
Prescaler
Reset
Conversion
Complete
Prescaler
Reset
MUX and REFS
Update
Figure 114. ADC Timing Diagram, Free Running Conversion
One Conversion
Next Conversion
11
12
13
1
2
3
4
Cycle Number
ADC Clock
ADSC
ADIF
ADCH
ADCL
MSB of Result
LSB of Result
Sample & Hold
MUX and REFS
Update
Conversion
Complete
235
2490G–AVR–03/04
Table 95. ADC Conversion Time
Sample & Hold (Cycles from
Start of Conversion)
ConversionTime
(Cycles)
Condition
First conversion
14.5
1.5
25
13
Normal conversions, single ended
Auto Triggered conversions
Normal conversions, differential
2
13.5
13/14
1.5/2.5
Differential Gain Channels
When using differential gain channels, certain aspects of the conversion need to be
taken into consideration.
Differential conversions are synchronized to the internal clock CKADC2 equal to half the
ADC clock. This synchronization is done automatically by the ADC interface in such a
way that the sample-and-hold occurs at a specific phase of CKADC2. A conversion initi-
ated by the user (i.e., all single conversions, and the first free running conversion) when
CKADC2 is low will take the same amount of time as a single ended conversion (13 ADC
clock cycles from the next prescaled clock cycle). A conversion initiated by the user
when CKADC2 is high will take 14 ADC clock cycles due to the synchronization mecha-
nism. In Free Running mode, a new conversion is initiated immediately after the
previous conversion completes, and since CKADC2 is high at this time, all automatically
started (i.e., all but the first) free running conversions will take 14 ADC clock cycles.
The gain stage is optimized for a bandwidth of 4 kHz at all gain settings. Higher frequen-
cies may be subjected to non-linear amplification. An external low-pass filter should be
used if the input signal contains higher frequency components than the gain stage band-
width. Note that the ADC clock frequency is independent of the gain stage bandwidth
limitation. For example, the ADC clock period may be 6 µs, allowing a channel to be
sampled at 12 kSPS, regardless of the bandwidth of this channel.
If differential gain channels are used and conversions are started by Auto Triggering, the
ADC must be switched off between conversions. When Auto Triggering is used, the
ADC prescaler is reset before the conversion is started. Since the gain stage is depen-
dent of a stable ADC clock prior to the conversion, this conversion will not be valid. By
disabling and then re-enabling the ADC between each conversion (writing ADEN in
ADCSRA to “0” then to “1”), only extended conversions are performed. The result from
the extended conversions will be valid. See “Prescaling and Conversion Timing” on
page 233 for timing details.
Changing Channel or
Reference Selection
The MUXn and REFS1:0 bits in the ADMUX Register are single buffered through a tem-
porary register to which the CPU has random access. This ensures that the channels
and reference selection only takes place at a safe point during the conversion. The
channel and reference selection is continuously updated until a conversion is started.
Once the conversion starts, the channel and reference selection is locked to ensure a
sufficient sampling time for the ADC. Continuous updating resumes in the last ADC
clock cycle before the conversion completes (ADIF in ADCSRA is set). Note that the
conversion starts on the following rising ADC clock edge after ADSC is written. The user
is thus advised not to write new channel or reference selection values to ADMUX until
one ADC clock cycle after ADSC is written.
If Auto Triggering is used, the exact time of the triggering event can be indeterministic.
Special care must be taken when updating the ADMUX Register, in order to control
which conversion will be affected by the new settings.
236
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
If both ADATE and ADEN is written to one, an interrupt event can occur at any time. If
the ADMUX Register is changed in this period, the user cannot tell if the next conversion
is based on the old or the new settings. ADMUX can be safely updated in the following
ways:
1. When ADATE or ADEN is cleared.
2. During conversion, minimum one ADC clock cycle after the trigger event.
3. After a conversion, before the interrupt flag used as trigger source is cleared.
When updating ADMUX in one of these conditions, the new settings will affect the next
ADC conversion.
Special care should be taken when changing differential channels. Once a differential
channel has been selected, the gain stage may take as much as 125 µs to stabilize to
the new value. Thus conversions should not be started within the first 125 µs after
selecting a new differential channel. Alternatively, conversion results obtained within this
period should be discarded.
The same settling time should be observed for the first differential conversion after
changing ADC reference (by changing the REFS1:0 bits in ADMUX).
If the JTAG interface is enabled, the function of ADC channels on PORTF7:4 is overrid-
den. Refer to Table 42, “Port F Pins Alternate Functions,” on page 81.
ADC Input Channels
ADC Voltage Reference
2490G–AVR–03/04
When changing channel selections, the user should observe the following guidelines to
ensure that the correct channel is selected:
In Single Conversion mode, always select the channel before starting the conversion.
The channel selection may be changed one ADC clock cycle after writing one to ADSC.
However, the simplest method is to wait for the conversion to complete before changing
the channel selection.
In Free Running mode, always select the channel before starting the first conversion.
The channel selection may be changed one ADC clock cycle after writing one to ADSC.
However, the simplest method is to wait for the first conversion to complete, and then
change the channel selection. Since the next conversion has already started automati-
cally, the next result will reflect the previous channel selection. Subsequent conversions
will reflect the new channel selection.
When switching to a differential gain channel, the first conversion result may have a
poor accuracy due to the required settling time for the automatic offset cancellation cir-
cuitry. The user should preferably disregard the first conversion result.
The reference voltage for the ADC (VREF) indicates the conversion range for the ADC.
Single ended channels that exceed VREF will result in codes close to 0x3FF. VREF can be
selected as either AVCC, internal 2.56V reference, or external AREF pin.
AVCC is connected to the ADC through a passive switch. The internal 2.56V reference
is generated from the internal bandgap reference (VBG) through an internal amplifier. In
either case, the external AREF pin is directly connected to the ADC, and the reference
voltage can be made more immune to noise by connecting a capacitor between the
AREF pin and ground. VREF can also be measured at the AREF pin with a high impedant
voltmeter. Note that VREF is a high impedant source, and only a capacitive load should
be connected in a system.
If the user has a fixed voltage source connected to the AREF pin, the user may not use
the other reference voltage options in the application, as they will be shorted to the
external voltage. If no external voltage is applied to the AREF pin, the user may switch
237
between AVCC and 2.56V as reference selection. The first ADC conversion result after
switching reference voltage source may be inaccurate, and the user is advised to dis-
card this result.
If differential channels are used, the selected reference should not be closer to AVCC
than indicated in Table 137 on page 334.
ADC Noise Canceler
The ADC features a noise canceler that enables conversion during sleep mode to
reduce noise induced from the CPU core and other I/O peripherals. The noise canceler
can be used with ADC Noise Reduction and Idle mode. To make use of this feature, the
following procedure should be used:
1. Make sure that the ADC is enabled and is not busy converting. Single Con-
version mode must be selected and the ADC conversion complete interrupt
must be enabled.
2. Enter ADC Noise Reduction mode (or Idle mode). The ADC will start a con-
version once the CPU has been halted.
3. If no other interrupts occur before the ADC conversion completes, the ADC
interrupt will wake up the CPU and execute the ADC Conversion Complete
interrupt routine. If another interrupt wakes up the CPU before the ADC con-
version is complete, that interrupt will be executed, and an ADC Conversion
Complete interrupt request will be generated when the ADC conversion
completes. The CPU will remain in Active mode until a new sleep command
is executed.
Note that the ADC will not be automatically turned off when entering other sleep modes
than Idle mode and ADC Noise Reduction mode. The user is advised to write zero to
ADEN before entering such sleep modes to avoid excessive power consumption. If the
ADC is enabled in such sleep modes and the user wants to perform differential conver-
sions, the user is advised to switch the ADC off and on after waking up from sleep to
prompt an extended conversion to get a valid result.
Analog Input Circuitry
The analog input circuitry for single ended channels is illustrated in Figure 115. An ana-
log source applied to ADCn is subjected to the pin capacitance and input leakage of that
pin, regardless of whether that channel is selected as input for the ADC. When the chan-
nel is selected, the source must drive the S/H capacitor through the series resistance
(combined resistance in the input path).
The ADC is optimized for analog signals with an output impedance of approximately
10 kΩ or less. If such a source is used, the sampling time will be negligible. If a source
with higher impedance is used, the sampling time will depend on how long time the
source needs to charge the S/H capacitor, with can vary widely. The user is recom-
mended to only use low impedant sources with slowly varying signals, since this
minimizes the required charge transfer to the S/H capacitor.
If differential gain channels are used, the input circuitry looks somewhat different,
although source impedances of a few hundred kΩ or less is recommended.
Signal components higher than the Nyquist frequency (fADC/2) should not be present for
either kind of channels, to avoid distortion from unpredictable signal convolution. The
user is advised to remove high frequency components with a low-pass filter before
applying the signals as inputs to the ADC.
238
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 115. Analog Input Circuitry
IIH
ADCn
1..100 kΩ
CS/H= 14 pF
VCC/2
IIL
Analog Noise Canceling
Techniques
Digital circuitry inside and outside the device 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. 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.
2. The AVCC pin on the device should be connected to the digital VCC supply
voltage via an LC network as shown in Figure 116.
3. Use the ADC noise canceler function to reduce induced noise from the CPU.
4. If any ADC port pins are used as digital outputs, it is essential that these do
not switch while a conversion is in progress.
Figure 116. ADC Power Connections
(AD0) PA0 51
VCC
52
GND 53
(ADC7) PF7 54
(ADC6) PF6 55
(ADC5) PF5 56
(ADC4) PF4 57
(ADC3) PF3 58
(ADC2) PF2 59
(ADC1) PF1
(ADC0) PF0
60
61
10 µΗ
62
63
64
AREF
GND
AVCC
100 nF
1
Analog Ground Plane
239
2490G–AVR–03/04
Offset Compensation
Schemes
The gain stage has a built-in offset cancellation circuitry that nulls the offset of differen-
tial measurements as much as possible. The remaining offset in the analog path can be
measured directly by selecting the same channel for both differential inputs. This offset
residue can be then subtracted in software from the measurement results. Using this
kind of software based offset correction, offset on any channel can be reduced below
one LSB.
ADC Accuracy Definitions
An n-bit single-ended ADC converts a voltage linearly between GND and VREF in 2n
steps (LSBs). The lowest code is read as 0, and the highest code is read as 2n - 1.
Several parameters describe the deviation from the ideal behavior:
•
Offset: The deviation of the first transition (0x000 to 0x001) compared to the ideal
transition (at 0.5 LSB). Ideal value: 0 LSB.
Figure 117. Offset Error
Output Code
Ideal ADC
Actual ADC
Offset
Error
VREF
Input Voltage
•
Gain Error: After adjusting for offset, the Gain Error is found as the deviation of the
last transition (0x3FE to 0x3FF) compared to the ideal transition (at 1.5 LSB below
maximum). Ideal value: 0 LSB
Figure 118. Gain Error
Gain
Error
Output Code
Ideal ADC
Actual ADC
VREF
Input Voltage
240
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
•
Integral Non-linearity (INL): After adjusting for Offset and Gain Error, the INL is the
maximum deviation of an actual transition compared to an ideal transition for any
code. Ideal value: 0 LSB.
Figure 119. Integral Non-linearity (INL)
Output Code
Ideal ADC
Actual ADC
VREF Input Voltage
•
Differential Non-linearity (DNL): The maximum deviation of the actual code width
(the interval between two adjacent transitions) from the ideal code width (1 LSB).
Ideal value: 0 LSB.
Figure 120. Differential Non-linearity (DNL)
Output Code
0x3FF
1 LSB
DNL
0x000
0
VREF Input Voltage
•
•
Quantization Error: Due to the quantization of the input voltage into a finite number
of codes, a range of input voltages (1 LSB wide) will code to the same value. Always
0.5 LSB.
Absolute Accuracy: The maximum deviation of an actual (unadjusted) transition
compared to an ideal transition for any code. This is the compound effect of Offset,
Gain Error, Differential Error, Non-linearity, and Quantization Error. Ideal value:
0.5 LSB.
241
2490G–AVR–03/04
ADC Conversion Result
After the conversion is complete (ADIF is high), the conversion result can be found in
the ADC Result registers (ADCL, ADCH).
For single ended conversion, the result is
V
⋅ 1024
IN
V
ADC = --------------------------
REF
where VIN is the voltage on the selected input pin and VREF the selected voltage refer-
ence (see Table 97 on page 243 and Table 98 on page 244). 0x000 represents analog
ground, and 0x3FF represents the selected reference voltage minus one LSB.
If differential channels are used, the result is
(V
– V
) ⋅ GAIN ⋅ 512
NEG
POS
ADC = ------------------------------------------------------------------------
V
REF
where VPOS is the voltage on the positive input pin, VNEG the voltage on the negative
input pin, GAIN the selected gain factor, and VREF the selected voltage reference. The
result is presented in two’s complement form, from 0x200 (-512d) through 0x1FF
(+511d). Note that if the user wants to perform a quick polarity check of the results, it is
sufficient to read the MSB of the result (ADC9 in ADCH). If this bit is one, the result is
negative, and if this bit is zero, the result is positive. Figure 121 shows the decoding of
the differential input range.
Table 96 shows the resulting output codes if the differential input channel pair (ADCn -
ADCm) is selected with a gain of GAIN and a reference voltage of VREF
.
Figure 121. Differential Measurement Range
Output Code
0x1FF
0x000
0
Differential Input
Voltage (Volts)
- VREF/GAIN
V
REF/GAIN
0x3FF
0x200
242
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 96. Correlation Between Input Voltage and Output Codes
VADCn
Read Code
0x1FF
0x1FF
0x1FE
...
Corresponding Decimal Value
VADCm + VREF/GAIN
511
511
510
...
VADCm + 0.999 VREF/GAIN
VADCm + 0.998 VREF/GAIN
...
V
ADCm + 0.001 VREF/GAIN
VADCm
ADCm - 0.001 VREF/GAIN
0x001
0x000
0x3FF
...
1
0
V
-1
...
...
VADCm - 0.999 VREF/GAIN
0x201
0x200
-511
-512
VADCm - VREF/GAIN
Example:
ADMUX = 0xED (ADC3 - ADC2, 10x gain, 2.56V reference, left adjusted result).
Voltage on ADC3 is 300 mV, voltage on ADC2 is 500 mV.
ADCR = 512 * 10 * (300 - 500) / 2560 = -400 = 0x270.
ADCL will thus read 0x00, and ADCH will read 0x9C. Writing zero to ADLAR right
adjusts the result: ADCL = 0x70, ADCH = 0x02.
ADC Multiplexer Selection
Register – ADMUX
Bit
7
REFS1
R/W
0
6
REFS0
R/W
0
5
ADLAR
R/W
0
4
MUX4
R/W
0
3
MUX3
R/W
0
2
MUX2
R/W
0
1
MUX1
R/W
0
0
MUX0
R/W
0
ADMUX
Read/Write
Initial Value
• Bit 7:6 – REFS1:0: Reference Selection Bits
These bits select the voltage reference for the ADC, as shown in Table 97. If these bits
are changed during a conversion, the change will not go in effect until this conversion is
complete (ADIF in ADCSRA is set). The internal voltage reference options may not be
used if an external reference voltage is being applied to the AREF pin.
Table 97. Voltage Reference Selections for ADC
REFS1
REFS0 Voltage Reference Selection
0
0
1
1
0
1
0
1
AREF, Internal Vref turned off.
AVCC with external capacitor at AREF pin.
Reserved
Internal 2.56V Voltage Reference with external capacitor at AREF pin.
243
2490G–AVR–03/04
•
Bit 5 – ADLAR: ADC Left Adjust Result
The ADLAR bit affects the presentation of the ADC conversion result in the ADC Data
Register. Write one to ADLAR to left adjust the result. Otherwise, the result is right
adjusted. Changing the ADLAR bit will affect the ADC Data Register immediately,
regardless of any ongoing conversions. For a complete description of this bit, see “The
ADC Data Register – ADCL and ADCH” on page 246.
• Bits 4:0 – MUX4:0: Analog Channel and Gain Selection Bits
The value of these bits selects which combination of analog inputs are connected to the
ADC. These bits also select the gain for the differential channels. See Table 98 for
details. If these bits are changed during a conversion, the change will not go in effect
until this conversion is complete (ADIF in ADCSRA is set).
Table 98. Input Channel and Gain Selections
Positive Differential
Input
Negative Differential
Input
MUX4..0
00000
00001
00010
00011
00100
00101
00110
00111
01000
01001
01010
01011
01100
01101
01110
01111
10000
10001
10010
10011
10100
10101
10110
10111
11000
11001
Single Ended Input
ADC0
Gain
ADC1
ADC2
ADC3
N/A
ADC4
ADC5
ADC6
ADC7
ADC0
ADC1
ADC0
ADC1
ADC2
ADC3
ADC2
ADC3
ADC0
ADC1
ADC2
ADC3
ADC4
ADC5
ADC6
ADC7
ADC0
ADC1
ADC0
ADC0
ADC0
ADC0
ADC2
ADC2
ADC2
ADC2
ADC1
ADC1
ADC1
ADC1
ADC1
ADC1
ADC1
ADC1
ADC2
ADC2
10x
10x
200x
200x
10x
10x
200x
200x
1x
1x
N/A
1x
1x
1x
1x
1x
1x
1x
1x
244
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 98. Input Channel and Gain Selections (Continued)
Positive Differential
Negative Differential
MUX4..0
11010
11011
11100
11101
11110
11111
Single Ended Input
Input
ADC2
ADC3
ADC4
ADC5
N/A
Input
ADC2
ADC2
ADC2
ADC2
Gain
1x
1x
1x
1x
1.22V (VBG
)
0V (GND)
ADC Control and Status
Register A – ADCSRA
Bit
7
ADEN
R/W
0
6
5
4
ADIF
R/W
0
3
ADIE
R/W
0
2
1
ADPS1
R/W
0
0
ADPS0
R/W
0
ADSC
R/W
0
ADATE
R/W
0
ADPS2
R/W
0
ADCSRA
Read/Write
Initial Value
• Bit 7 – ADEN: ADC Enable
Writing this bit to one enables the ADC. By writing it to zero, the ADC is turned off. Turn-
ing the ADC off while a conversion is in progress, will terminate this conversion.
• Bit 6 – ADSC: ADC Start Conversion
In Single Conversion mode, write this bit to one to start each conversion. In Free Run-
ning mode, write this bit to one to start the first conversion. The first conversion after
ADSC has been written after the ADC has been enabled, or if ADSC is written at the
same time as the ADC is enabled, will take 25 ADC clock cycles instead of the normal
13. This first conversion performs initialization of the ADC.
ADSC will read as one as long as a conversion is in progress. When the conversion is
complete, it returns to zero. Writing zero to this bit has no effect.
• Bit 5 – ADATE: ADC Auto Trigger Enable
When this bit is written to one, Auto Triggering of the ADC is enabled. The ADC will start
a conversion on a positive edge of the selected trigger signal. The trigger source is
selected by setting the ADC Trigger Select bits, ADTS in ADCSRB.
• Bit 4 – ADIF: ADC Interrupt Flag
This bit is set when an ADC conversion completes and the data registers are updated.
The ADC Conversion Complete Interrupt is executed if the ADIE bit and the I-bit in
SREG are set. ADIF is cleared by hardware when executing the corresponding interrupt
handling vector. Alternatively, ADIF is cleared by writing a logical one to the flag.
Beware that if doing a Read-Modify-Write on ADCSRA, a pending interrupt can be dis-
abled. This also applies if the SBI and CBI instructions are used.
• Bit 3 – ADIE: ADC Interrupt Enable
When this bit is written to one and the I-bit in SREG is set, the ADC Conversion Com-
plete Interrupt is activated.
245
2490G–AVR–03/04
• Bits 2:0 – ADPS2:0: ADC Prescaler Select Bits
These bits determine the division factor between the XTAL frequency and the input
clock to the ADC.
Table 99. ADC Prescaler Selections
ADPS2
ADPS1
ADPS0
Division Factor
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
2
2
4
8
16
32
64
128
The ADC Data Register –
ADCL and ADCH
ADLAR = 0
Bit
15
14
13
12
11
10
9
8
–
–
–
–
–
–
ADC9
ADC1
ADC8
ADC0
ADCH
ADCL
ADC7
ADC6
ADC5
ADC4
ADC3
ADC2
7
R
R
0
6
R
R
0
5
R
R
0
4
R
R
0
3
R
R
0
2
R
R
0
1
R
R
0
0
R
R
0
Read/Write
Initial Value
0
0
0
0
0
0
0
0
ADLAR = 1
Bit
15
14
13
12
11
10
9
8
ADC9
ADC1
7
R
R
0
ADC8
ADC0
6
R
R
0
ADC7
–
5
ADC6
–
4
ADC5
–
3
ADC4
–
2
ADC3
–
1
ADC2
–
0
ADCH
ADCL
Read/Write
Initial Value
R
R
0
R
R
0
R
R
0
R
R
0
R
R
0
R
R
0
0
0
0
0
0
0
0
0
When an ADC conversion is complete, the result is found in these two registers. If differ-
ential channels are used, the result is presented in two’s complement form.
When ADCL is read, the ADC Data Register is not updated until ADCH is read. Conse-
quently, if the result is left adjusted and no more than 8-bit precision is required, it is
sufficient to read ADCH. Otherwise, ADCL must be read first, then ADCH.
The ADLAR bit in ADMUX, and the MUXn bits in ADMUX affect the way the result is
read from the registers. If ADLAR is set, the result is left adjusted. If ADLAR is cleared
(default), the result is right adjusted.
246
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
• ADC9:0: ADC Conversion Result
These bits represent the result from the conversion, as detailed in “ADC Conversion
Result” on page 242.
ADC Control and Status
Register B – ADCSRB
Bit
7
–
R
0
6
–
R
0
5
–
R
0
4
–
R
0
3
–
R
0
2
ADTS2
R/W
0
1
ADTS1
R/W
0
0
ADTS0
R/W
0
ADCSRB
Read/Write
Initial Value
• Bits 7:3 – Res: Reserved Bits
These bits are reserved bits in the ATmega64 and will always read as zero.
• Bit 2:0 – ADTS2:0: ADC Auto Trigger Source
If ADATE in ADCSRA is written to one, the value of these bits selects which source will
trigger an ADC conversion. If ADATE is cleared, the ADTS2:0 settings will have no
effect. A conversion will be triggered by the rising edge of the selected interrupt flag.
Note that switching from a trigger source that is cleared to a trigger source that is set,
will generate a positive edge on the trigger signal. If ADEN in ADCSRA is set, this will
start a conversion. Switching to Free Running mode (ADTS[2:0]=0) will not cause a trig-
ger event, even if the ADC Interrupt Flag is set.
Figure 122. ADC Auto Trigger Source Selections
ADTS2
ADTS1
ADTS0
Trigger Source
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
Free Running mode
Analog Comparator
External Interrupt Request 0
Timer/Counter0 Compare Match
Timer/Counter0 Overflow
Timer/Counter1 Compare Match B
Timer/Counter1 Overflow
Timer/Counter1 Capture Event
247
2490G–AVR–03/04
JTAG Interface and
On-chip Debug
System
Features
• JTAG (IEEE std. 1149.1 Compliant) Interface
• Boundary-scan Capabilities According to the IEEE std. 1149.1 (JTAG) Standard
• Debugger Access to:
– All Internal Peripheral Units
– Internal and External RAM
– The Internal Register File
– Program Counter
– EEPROM and Flash Memories
• Extensive On-chip Debug Support for Break Conditions, Including
– AVR Break Instruction
– Break on Change of Program Memory Flow
– Single Step Break
– Program Memory Break Points on Single Address or Address Range
– Data Memory Break Points on Single Address or Address Range
• Programming of Flash, EEPROM, Fuses, and Lock Bits through the JTAG Interface
• On-chip Debugging Supported by AVR Studio®
Overview
The AVR IEEE std. 1149.1 compliant JTAG interface can be used for:
•
•
•
Testing PCBs by using the JTAG Boundary-scan capability.
Programming the non-volatile memories, Fuses and Lock bits.
On-chip debugging.
A brief description is given in the following sections. Detailed descriptions for Program-
ming via the JTAG interface, and using the Boundary-scan chain can be found in the
sections “Programming Via the JTAG Interface” on page 312 and “IEEE 1149.1 (JTAG)
Boundary-scan” on page 254, respectively. The On-chip Debug support is considered
being private JTAG instructions, and distributed within ATMEL and to selected third
party vendors only.
Figure 123 shows a block diagram of the JTAG interface and the On-chip Debug sys-
tem. The TAP Controller is a state machine controlled by the TCK and TMS signals. The
TAP Controller selects either the JTAG Instruction Register or one of several data regis-
ters as the scan chain (Shift Register) between the TDI – input and TDO – output. The
Instruction Register holds JTAG instructions controlling the behavior of a data register.
The ID-Register, Bypass Register, and the Boundary-scan Chain are the data registers
used for board-level testing. The JTAG Programming Interface (actually consisting of
several physical and virtual data registers) is used for serial programming via the JTAG
interface. The Internal Scan Chain and Break Point Scan Chain are used for On-chip
debugging only.
Test Access Port – TAP
The JTAG interface is accessed through four of the AVR’s pins. In JTAG terminology,
these pins constitute the Test Access Port – TAP. These pins are:
•
TMS: Test mode select. This pin is used for navigating through the TAP-controller
state machine.
•
•
TCK: Test clock. JTAG operation is synchronous to TCK.
TDI: Test Data In. Serial input data to be shifted in to the Instruction Register or Data
Register (Scan Chains).
•
TDO: Test Data Out. Serial output data from Instruction Register or Data Register.
248
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
The IEEE std. 1149.1 also specifies an optional TAP signal; TRST – Test ReSeT –
which is not provided.
When the JTAGEN fuse is unprogrammed, these four TAP pins are normal port pins
and the TAP controller is in reset. When programmed and the JTD bit in MCUCSR is
cleared, the TAP input signals are internally pulled high and the JTAG is enabled for
Boundary-scan and programming. In this case, the TAP output pin (TDO) is left floating
in states where the JTAG TAP controller is not shifting data, and must therefore be con-
nected to a pull-up resistor or other hardware having pull-ups (for instance the TDI-input
of the next device in the scan chain). The device is shipped with this fuse programmed.
For the On-chip Debug system, in addition to the JTAG interface pins, the RESET pin is
monitored by the debugger to be able to detect External Reset sources. The debugger
can also pull the RESET pin low to reset the whole system, assuming only open collec-
tors on the reset line are used in the application.
Figure 123. Block Diagram
I/O PORT 0
DEVICE BOUNDARY
BOUNDARY SCAN CHAIN
TDI
TDO
TCK
TMS
JTAG PROGRAMMING
INTERFACE
TAP
CONTROLLER
AVR CPU
INTERNAL
FLASH
MEMORY
Address
Data
SCAN
CHAIN
PC
Instruction
INSTRUCTION
REGISTER
ID
REGISTER
BREAKPOINT
UNIT
M
U
X
FLOW CONTROL
UNIT
BYPASS
REGISTER
DIGITAL
PERIPHERAL
UNITS
ANALOG
PERIPHERIAL
UNITS
BREAKPOINT
SCAN CHAIN
JTAG / AVR CORE
COMMUNICATION
INTERFACE
ADDRESS
DECODER
OCD STATUS
AND CONTROL
I/O PORT n
249
2490G–AVR–03/04
Figure 124. TAP Controller State Diagram
1
Test-Logic-Reset
0
1
1
1
0
Run-Test/Idle
Select-DR Scan
Select-IR Scan
0
0
1
1
Capture-DR
Capture-IR
0
0
Shift-IR
1
Shift-DR
0
0
1
Exit1-DR
0
1
1
Exit1-IR
0
Pause-DR
1
0
Pause-IR
1
0
0
0
Exit2-DR
1
Exit2-IR
1
Update-DR
Update-IR
1
1
0
0
TAP Controller
The TAP controller is a 16-state finite state machine that controls the operation of the
Boundary-scan circuitry, JTAG programming circuitry, or On-chip Debug system. The
state transitions depicted in Figure 124 depends on the signal present on TMS (shown
adjacent to each state transition) at the time of the rising edge at TCK. The initial state
after a Power-on Reset is Test-Logic-Reset.
As a definition in this datasheet, the LSB is shifted in and out first for all Shift Registers.
Assuming Run-Test/Idle is the present state, a typical scenario for using the JTAG inter-
face is:
•
At the TMS input, apply the sequence 1, 1, 0, 0 at the rising edges of TCK to enter
the Shift Instruction Register – Shift-IR state. While in this state, shift the four bits of
the JTAG instructions into the JTAG instruction register from the TDI input at the
rising edge of TCK. The TMS input must be held low during input of the 3 LSBs in
order to remain in the Shift-IR state. The MSB of the instruction is shifted in when
this state is left by setting TMS high. While the instruction is shifted in from the TDI
250
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
pin, the captured IR-state 0x01 is shifted out on the TDO pin. The JTAG Instruction
selects a particular Data Register as path between TDI and TDO and controls the
circuitry surrounding the selected data register.
•
•
Apply the TMS sequence 1, 1, 0 to reenter the Run-Test/Idle state. The instruction is
latched onto the parallel output from the Shift Register path in the Update-IR state.
The Exit-IR, Pause-IR, and Exit2-IR states are only used for navigating the state
machine.
At the TMS input, apply the sequence 1, 0, 0 at the rising edges of TCK to enter the
Shift Data Register – Shift-DR state. While in this state, upload the selected data
register (selected by the present JTAG instruction in the JTAG Instruction Register)
from the TDI input at the rising edge of TCK. In order to remain in the Shift-DR state,
the TMS input must be held low during input of all bits except the MSB. The MSB of
the data is shifted in when this state is left by setting TMS high. While the data
register is shifted in from the TDI pin, the parallel inputs to the data register captured
in the Capture-DR state is shifted out on the TDO pin.
•
Apply the TMS sequence 1, 1, 0 to reenter the Run-Test/Idle state. If the selected
data register has a latched parallel-output, the latching takes place in the Update-
DR state. The Exit-DR, Pause-DR, and Exit2-DR states are only used for navigating
the state machine.
As shown in the state diagram, the Run-Test/Idle state need not be entered between
selecting JTAG instruction and using data registers, and some JTAG instructions may
select certain functions to be performed in the Run-Test/Idle, making it unsuitable as an
Idle state.
Note:
Independent of the initial state of the TAP Controller, the Test-Logic-Reset state can
always be entered by holding TMS high for five TCK clock periods.
For detailed information on the JTAG specification, refer to the literature listed in “Bibli-
ography” on page 253.
Using the Boundary -
scan Chain
A complete description of the Boundary-scan capabilities are given in the section “IEEE
1149.1 (JTAG) Boundary-scan” on page 254.
Using the On-chip Debug As shown in Figure 123, the hardware support for On-chip Debugging consists mainly
of:
system
•
A scan chain on the interface between the internal AVR CPU and the internal
peripheral units.
•
•
Break Point unit.
Communication interface between the CPU and JTAG system.
All read or modify/write operations needed for implementing the Debugger are done by
applying AVR instructions via the internal AVR CPU Scan Chain. The CPU sends the
result to an I/O memory mapped location which is part of the communication interface
between the CPU and the JTAG system.
The Break Point Unit implements Break on Change of Program Flow, Single Step
Break, two Program Memory Break Points, and two combined Break Points. Together,
the four Break Points can be configured as either:
•
•
•
4 Single Program Memory Break Points.
3 Single Program Memory Break Points + 1 Single Data Memory Break Point.
2 Single Program Memory Break Points + 2 Single Data Memory Break Points.
251
2490G–AVR–03/04
•
•
2 Single Program Memory Break Points + 1 Program Memory Break Point with
mask (“range Break Point”).
2 Single Program Memory Break Points + 1 Data Memory Break Point with mask
(“range Break Point”).
A debugger, like the AVR Studio®, may however use one or more of these resources for
its internal purpose, leaving less flexibility to the end-user.
A list of the On-chip Debug specific JTAG instructions is given in “On-chip Debug Spe-
cific JTAG Instructions” on page 252.
The JTAGEN Fuse must be programmed to enable the JTAG Test Access Port. In addi-
tion, the OCDEN Fuse must be programmed and no Lock bits must be set for the On-
chip Debug system to work. As a security feature, the On-chip Debug system is disabled
when any Lock bits are set. Otherwise, the On-chip Debug system would have provided
a back-door into a secured device.
The AVR Studio enables the user to fully control execution of programs on an AVR
device with On-chip Debug capability, AVR In-Circuit Emulator, or the built-in AVR
Instruction Set Simulator. AVR Studio supports source level execution of Assembly pro-
grams assembled with Atmel AVR Assembler and C programs compiled with third party
vendors’ compilers.
AVR Studio runs under Microsoft® Windows® 95/98/2000 and Microsoft Windows NT®.
For a full description of the AVR Studio, please refer to the AVR Studio User Guide.
Only highlights are presented in this document.
All necessary execution commands are available in AVR Studio, both on source level
and on disassembly level. The user can execute the program, single step through the
code either by tracing into or stepping over functions, step out of functions, place the
cursor on a statement and execute until the statement is reached, stop the execution,
and reset the execution target. In addition, the user can have an unlimited number of
code Break Points (using the BREAK instruction) and up to two data memory Break
Points, alternatively combined as a mask (range) Break Point.
On-chip Debug Specific
JTAG Instructions
The On-chip debug support is considered being private JTAG instructions, and distrib-
uted within ATMEL and to selected third party vendors only. Instruction opcodes are
listed for reference.
PRIVATE0; 0x8
PRIVATE1; 0x9
PRIVATE2; 0xA
PRIVATE3; 0xB
Private JTAG instruction for accessing On-chip Debug system.
Private JTAG instruction for accessing On-chip Debug system.
Private JTAG instruction for accessing On-chip Debug system.
Private JTAG instruction for accessing On-chip Debug system.
252
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
On-chip Debug Related
Register in I/O Memory
On-chip Debug Register –
OCDR
Bit
7
MSB/IDRD
R/W
6
5
4
3
2
1
0
LSB
R/W
0
OCDR
Read/Write
Initial Value
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
0
The OCDR Register provides a communication channel from the running program in the
microcontroller to the debugger. The CPU can transfer a byte to the debugger by writing
to this location. At the same time, an internal flag; I/O Debug Register Dirty – IDRD – is
set to indicate to the debugger that the register has been written. When the CPU reads
the OCDR Register the 7 LSB will be from the OCDR Register, while the MSB is the
IDRD bit. The debugger clears the IDRD bit when it has read the information.
In some AVR devices, this register is shared with a standard I/O location. In this case,
the OCDR Register can only be accessed if the OCDEN Fuse is programmed, and the
debugger enables access to the OCDR Register. In all other cases, the standard I/O
location is accessed.
Refer to the debugger documentation for further information on how to use this register.
Using the JTAG
Programming
Capabilities
Programming of AVR parts via JTAG is performed via the 4-pin JTAG port, TCK, TMS,
TDI, and TDO. These are the only pins that need to be controlled/observed to perform
JTAG programming (in addition to power pins). It is not required to apply 12V externally.
The JTAGEN Fuse must be programmed and the JTD bit in the MCUSR Register must
be cleared to enable the JTAG Test Access Port.
The JTAG programming capability supports:
•
•
•
•
Flash Programming and verifying
EEPROM Programming and verifying
Fuse Programming and verifying
Lock bit Programming and verifying
The Lock bit security is exactly as in Parallel Programming mode. If the Lock bits LB1 or
LB2 are programmed, the OCDEN Fuse cannot be programmed unless first doing a
Chip Erase. This is a security feature that ensures no back-door exists for reading out
the content of a secured device.
The details on programming through the JTAG interface and programming specific
JTAG instructions are given in the section “Programming Via the JTAG Interface” on
page 312.
Bibliography
For more information about general Boundary-scan, the following literature can be
consulted:
•
IEEE: IEEE Std 1149.1 - 1990. IEEE Standard Test Access Port and Boundary-scan
Architecture, IEEE, 1993.
•
Colin Maunder: The Board Designers Guide to Testable Logic Circuits, Addison –
Wesley, 1992.
253
2490G–AVR–03/04
IEEE 1149.1 (JTAG)
Boundary-scan
Features
• JTAG (IEEE std. 1149.1 Compliant) Interface
• Boundary-scan Capabilities According to the JTAG Standard
• Full Scan of all Port Functions as well as Analog Circuitry Having Off-chip Connections
• Supports the Optional IDCODE Instruction
• Additional Public AVR_RESET Instruction to Reset the AVR
System Overview
The Boundary-scan chain has the capability of driving and observing the logic levels on
the digital I/O pins, as well as the boundary between digital and analog logic for analog
circuitry having Off-chip connections. At system level, all ICs having JTAG capabilities
are connected serially by the TDI/TDO signals to form a long Shift Register. An external
controller sets up the devices to drive values at their output pins, and observe the input
values received from other devices. The controller compares the received data with the
expected result. In this way, Boundary-scan provides a mechanism for testing intercon-
nections and integrity of components on Printed Circuits Boards by using the four TAP
signals only.
The four IEEE 1149.1 defined mandatory JTAG instructions IDCODE, BYPASS, SAM-
PLE/PRELOAD, and EXTEST, as well as the AVR specific public JTAG instruction
AVR_RESET can be used for testing the printed circuit board. Initial scanning of the
data register path will show the ID-Code of the device, since IDCODE is the default
JTAG instruction. It may be desirable to have the AVR device in reset during test mode.
If not reset, inputs to the device may be determined by the scan operations, and the
internal software may be in an undetermined state when exiting the test mode. Entering
reset, the outputs of any Port Pin will instantly enter the high impedance state, making
the HIGHZ instruction redundant. If needed, the BYPASS instruction can be issued to
make the shortest possible scan chain through the device. The device can be set in the
reset state either by pulling the external RESET pin low, or issuing the AVR_RESET
instruction with appropriate setting of the Reset Data Register.
The EXTEST instruction is used for sampling external pins and loading output pins with
data. The data from the output latch will be driven out on the pins as soon as the
EXTEST instruction is loaded into the JTAG IR-Register. Therefore, the SAMPLE/PRE-
LOAD should also be used for setting initial values to the scan ring, to avoid damaging
the board when issuing the EXTEST instruction for the first time. SAMPLE/PRELOAD
can also be used for taking a snapshot of the external pins during normal operation of
the part.
The JTAGEN Fuse must be programmed and the JTD bit in the I/O Register MCUCSR
must be cleared to enable the JTAG Test Access Port.
When using the JTAG interface for Boundary-scan, using a JTAG TCK clock frequency
higher than the internal chip frequency is possible. The chip clock is not required to run.
Data Registers
The data registers relevant for Boundary-scan operations are:
•
•
•
•
Bypass Register
Device Identification Register
Reset Register
Boundary-scan Chain
254
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Bypass Register
The Bypass Register consists of a single Shift Register stage. When the Bypass Regis-
ter is selected as path between TDI and TDO, the register is reset to 0 when leaving the
Capture-DR controller state. The Bypass Register can be used to shorten the scan
chain on a system when the other devices are to be tested.
Device Identification Register Figure 125 shows the structure of the Device Identification Register.
Figure 125. The Format of the Device Identification Register
LSB
0
MSB
31
Bit
Device ID
28
27
12
11
1
Version
4 bits
Part Number
16 bits
Manufacturer ID
1
1-bit
11 bits
Version
Version is a 4-bit number identifying the revision of the component. The relevant version
number is shown in Table 100.
Table 100. JTAG Version Numbers
Version
JTAG Version Number (Hex)
ATmega64 Revision A
0x0
Part Number
The part number is a 16-bit code identifying the component. The JTAG Part Number for
ATmega64 is listed in Table 101.
Table 101. AVR JTAG Part Number
Part Number
JTAG Part Number (Hex)
ATmega64
0x9602
Manufacturer ID
The Manufacturer ID is a 11-bit code identifying the manufacturer. The JTAG manufac-
turer ID for Atmel is listed in Table 102.
Table 102. Manufacturer ID
Manufacturer
JTAG Man. ID (Hex)
Atmel
0x01F
255
2490G–AVR–03/04
Reset Register
The Reset Register is a Test Data Register used to reset the part. Since the AVR tri-
states port pins when reset, the Reset Register can also replace the function of the
unimplemented optional JTAG instruction HIGHZ.
A high value in the Reset Register corresponds to pulling the External Reset low. The
part is reset as long as there is a high value present in the Reset Register. Depending
on the Fuse settings for the clock options, the part will remain reset for a Reset Time-out
Period (refer to “Clock Sources” on page 36) after releasing the Reset Register. The
output from this data register is not latched, so the reset will take place immediately, as
shown in Figure 126.
Figure 126. Reset Register
To
TDO
From Other Internal and
External Reset Sources
From
TDI
Internal Reset
D
Q
ClockDR · AVR_RESET
Boundary-scan Chain
The Boundary-scan Chain has the capability of driving and observing the logic levels on
the digital I/O pins, as well as the boundary between digital and analog logic for analog
circuitry having Off-chip connections.
See “Boundary-scan Chain” on page 258 for a complete description.
Boundary-scan Specific The instruction register is 4-bit wide, supporting up to 16 instructions. Listed below are
the JTAG instructions useful for Boundary-scan operation. Note that the optional HIGHZ
JTAG Instructions
instruction is not implemented, but all outputs with tri-state capability can be set in high-
impedant state by using the AVR_RESET instruction, since the initial state for all port
pins is tri-state.
As a definition in this datasheet, the LSB is shifted in and out first for all Shift Registers.
The OPCODE for each instruction is shown behind the instruction name in hex format.
The text describes which data register is selected as path between TDI and TDO for
each instruction.
EXTEST; 0x0
Mandatory JTAG instruction for selecting the Boundary-scan Chain as data register for
testing circuitry external to the AVR package. For port-pins, Pull-up Disable, Output
Control, Output Data, and Input Data are all accessible in the scan chain. For analog cir-
cuits having Off-chip connections, the interface between the analog and the digital logic
is in the scan chain. The contents of the latched outputs of the Boundary-scan Chain is
driven out as soon as the JTAG IR-Register is loaded with the EXTEST instruction.
256
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
The active states are:
•
•
•
Capture-DR: Data on the external pins are sampled into the Boundary-scan Chain.
Shift-DR: The internal scan chain is shifted by the TCK input.
Update-DR: Data from the scan chain is applied to output pins.
IDCODE; 0x1
Optional JTAG instruction selecting the 32-bit ID-Register as data register. The ID-Reg-
ister consists of a version number, a device number and the manufacturer code chosen
by JEDEC. This is the default instruction after Power-up.
The active states are:
•
Capture-DR: Data in the IDCODE Register is sampled into the Boundary-scan
Chain.
•
Shift-DR: The IDCODE scan chain is shifted by the TCK input.
SAMPLE_PRELOAD; 0x2
Mandatory JTAG instruction for taking a snap-shot of the input/output pins without
affecting the system operation, and pre-loading the output latches. However, the output
latches are not connected to the pins. The Boundary-scan Chain is selected as data
register.
The active states are:
•
•
•
Capture-DR: Data on the external pins are sampled into the Boundary-scan Chain.
Shift-DR: The Boundary-scan Chain is shifted by the TCK input.
Update-DR: Data from the Boundary-scan Chain is applied to the output latches.
However, the output latches are not connected to the pins.
AVR_RESET; 0xC
The AVR specific public JTAG instruction for forcing the AVR device into the Reset
mode or releasing the JTAG Reset source. The TAP controller is not reset by this
instruction. The one bit Reset Register is selected as data register. Note that the reset
will be active as long as there is a logic “one” in the Reset Chain. The output from this
chain is not latched.
The active states are:
•
Shift-DR: The Reset Register is shifted by the TCK input.
BYPASS; 0xF
Mandatory JTAG instruction selecting the Bypass Register for Data Register.
The active states are:
•
•
Capture-DR: Loads a logic “0” into the Bypass Register.
Shift-DR: The Bypass Register cell between TDI and TDO is shifted.
257
2490G–AVR–03/04
Boundary-scan Related
Register in I/O Memory
MCU Control and Status
Register – MCUCSR
The MCU Control and Status Register contains control bits for general MCU functions,
and provides information on which reset source caused an MCU Reset.
Bit
7
6
–
R
0
5
–
R
0
4
3
2
1
0
JTD
R/W
0
JTRF
R/W
WDRF
R/W
BORF
R/W
EXTRF
R/W
PORF
R/W
MCUCSR
Read/Write
Initial Value
See Bit Description
• Bit 7 – JTD: JTAG Interface Disable
When this bit is zero, the JTAG interface is enabled if the JTAGEN Fuse is programmed.
If this bit is one, the JTAG interface is disabled. In order to avoid unintentional disabling
or enabling of the JTAG interface, a timed sequence must be followed when changing
this bit: The application software must write this bit to the desired value twice within four
cycles to change its value.
If the JTAG interface is left unconnected to other JTAG circuitry, the JTD bit should be
set to one. The reason for this is to avoid static current at the TDO pin in the JTAG
interface.
• Bit 4 – JTRF: JTAG Reset Flag
This bit is set if a reset is being caused by a logic one in the JTAG Reset Register
selected by the JTAG instruction AVR_RESET. This bit is reset by a Brown-out Reset,
or by writing a logic zero to the flag.
Boundary-scan Chain
The Boundary-scan Chain has the capability of driving and observing the logic levels on
the digital I/O pins, as well as the boundary between digital and analog logic for analog
circuitry having Off-chip connection.
Scanning the Digital Port Pins Figure 127 shows the Boundary-scan Cell for a bi-directional port pin with pull-up func-
tion. The cell consists of a standard Boundary-scan cell for the Pull-up Enable – PUExn
– function, and a bi-directional pin cell that combines the three signals, Output Control –
OCxn, Output Data – ODxn, and Input Data – IDxn, into only a two-stage Shift Register.
The port and pin indexes are not used in the following description.
The Boundary-scan logic is not included in the figures in this Datasheet. Figure 128
shows a simple digital Port Pin as described in the section “I/O Ports” on page 64. The
Boundary-scan details from Figure 127 replaces the dashed box in Figure 128.
When no alternate port function is present, the Input Data – ID corresponds to the PINxn
Register value (but ID has no synchronizer), Output Data corresponds to the PORT
Register, Output Control corresponds to the Data Direction – DD Register, and the Pull-
up Enable – PUExn – corresponds to logic expression PUD · DDxn · PORTxn.
Digital alternate port functions are connected outside the dotted box in Figure 128 to
make the scan chain read the actual pin value. For analog function, there is a direct con-
nection from the external pin to the analog circuit, and a scan chain is inserted on the
interface between the digital logic and the analog circuitry.
258
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 127. Boundary-scan Cell for Bi-directional Port Pin with Pull-up Function
ShiftDR
To Next Cell
EXTEST
Vcc
0
1
FF2
Q
LD2
0
1
D
D
Q
G
FF1
D Q
LD1
0
1
0
1
D
G
Q
0
1
FF0
D
LD0
0
1
0
1
Q
D
G
Q
From Last Cell
ClockDR
UpdateDR
259
2490G–AVR–03/04
Figure 128. General Port Pin Schematic Diagram
See Boundary-scan Description
for Details!
PUExn
PUD
Q
D
DDxn
Q CLR
WDx
RDx
RESET
OCxn
Q
D
Pxn
PORTxn
ODxn
Q CLR
WPx
RRx
IDxn
RESET
SLEEP
SYNCHRONIZER
RPx
D
Q
D
L
Q
Q
PINxn
Q
CLK I/O
PUD:
PULLUP DISABLE
WDx:
RDx:
WPx:
RRx:
RPx:
WRITE DDRx
READ DDRx
WRITE PORTx
PUExn:
OCxn:
ODxn:
IDxn:
PULLUP ENABLE for pin Pxn
OUTPUT CONTROL for pin Pxn
OUTPUT DATA to pin Pxn
INPUT DATA from pin Pxn
SLEEP CONTROL
READ PORTx REGISTER
READ PORTx PIN
I/O CLOCK
SLEEP:
CLK I/O
:
Boundary-scan and the Two-
wire Interface
The two Two-wire Interface pins SCL and SDA have one additional control signal in the
scan-chain; Two-wire Interface Enable – TWIEN. As shown in Figure 129, the TWIEN
signal enables a tri-state buffer with slew-rate control in parallel with the ordinary digital
port pins. A general scan cell as shown in Figure 133 is attached to the TWIEN signal.
Notes: 1. A separate scan chain for the 50 ns spike filter on the input is not provided. The ordi-
nary scan support for digital port pins suffice for connectivity tests. The only reason
for having TWIEN in the scan path, is to be able to disconnect the slew-rate control
buffer when doing boundary-scan.
2. Make sure the OC and TWIEN signals are not asserted simultaneously, as this will
lead to drive contention.
260
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 129. Additional Scan Signal for the Two-wire Interface
PUExn
OCxn
ODxn
TWIEN
Pxn
SRC
Slew-rate limited
IDxn
Scanning the RESET Pin
The RESET pin accepts 5V active low logic for standard reset operation, and 12V active
high logic for High Voltage Parallel programming. An observe-only cell as shown in Fig-
ure 130 is inserted both for the 5V reset signal; RSTT, and the 12V reset signal;
RSTHV.
Figure 130. Observe-only Cell
To
Next
ShiftDR
Cell
FF1
0
1
D
Q
From
ClockDR
Previous
Cell
261
2490G–AVR–03/04
Scanning the Clock Pins
The AVR devices have many clock options selectable by fuses. These are: Internal RC
Oscillator, External RC, External Clock, (High Frequency) Crystal Oscillator, Low-fre-
quency Crystal Oscillator, and Ceramic Resonator.
Figure 131 shows how each Oscillator with external connection is supported in the scan
chain. The Enable signal is supported with a general boundary-scan cell, while the
Oscillator/clock output is attached to an observe-only cell. In addition to the main clock,
the timer Oscillator is scanned in the same way. The output from the internal RC Oscilla-
tor is not scanned, as this Oscillator does not have external connections.
Figure 131. Boundary-scan Cells for Oscillators and Clock Options
XTAL1/TOSC1
XTAL2/TOSC2
To
Next
Cell
To
ShiftDR
EXTEST
Next
Cell
Oscillator
ShiftDR
0
1
ENABLE
OUTPUT
0
1
FF1
D
Q
D
G
Q
0
1
D
Q
From
Previous
Cell
ClockDR UpdateDR
From
ClockDR
Previous
Cell
Table 103 summaries the scan registers for the external clock pin XTAL1, oscillators
with XTAL1/XTAL2 connections as well as 32 kHz Timer Oscillator.
Table 103. Scan Signals for the Oscillators(1)(2)(3)
Enable
Signal
Scanned Clock
Line
Scanned Clock Line
when Not Used
Clock Option
EXTCLKEN
OSCON
EXTCLK (XTAL1)
OSCCK
External Clock
0
0
External Crystal
External Ceramic Resonator
RCOSCEN
OSC32EN
TOSKON
RCCK
External RC
0
1
0
OSC32CK
TOSCK
Low Freq. External Crystal
32 kHz Timer Oscillator
Notes: 1. Do not enable more than one clock source as main clock at a time.
2. Scanning an Oscillator output gives unpredictable results as there is a frequency drift
between the internal Oscillator and the JTAG TCK clock. If possible, scanning an
external clock is preferred.
3. The clock configuration is programmed by fuses. As a fuse does not change run-time,
the clock configuration is considered fixed for a given application. The user is advised
to scan the same clock option as to be used in the final system. The enable signals
are supported in the scan chain because the system logic can disable clock options
262
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
in sleep modes, thereby disconnecting the Oscillator pins from the scan path if not
provided. The INTCAP Fuses are not supported in the scan-chain, so the boundary
scan chain cannot make a XTAL Oscillator requiring internal capacitors to run unless
the fuse is correctly programmed.
Scanning the Analog
Comparator
The relevant Comparator signals regarding Boundary-scan are shown in Figure 132.
The Boundary-scan cell from Figure 133 is attached to each of these signals. The sig-
nals are described in Table 104.
The Comparator needs not be used for pure connectivity testing, since all analog inputs
are shared with a digital port pin as well.
Figure 132. Analog Comparator
BANDGAP
REFERENCE
ACBG
ACO
AC_IDLE
ACME
ADCEN
ADC MULTIPLEXER
OUTPUT
263
2490G–AVR–03/04
Figure 133. General Boundary-scan Cell used for Signals for Comparator and ADC
To
Next
ShiftDR
Cell
EXTEST
0
1
0
1
D
Q
D
G
Q
From
ClockDR
UpdateDR
Previous
Cell
Table 104. Boundary-scan Signals for the Analog Comparator
Direction as
Seen from the
Comparator
Recommended
Input when Not
in Use
Output Values when
Recommended
Inputs are Used
Signal
Name
Description
AC_IDLE
Input
Output
Input
Input
Turns off Analog
Comparator
when true
1
Depends upon µC
code being executed
ACO
Analog
Comparator
Output
Will become
input to µC code
being executed
0
ACME
ACBG
Uses output
signal from ADC
mux when true
0
Depends upon µC
code being executed
Bandgap
Reference
enable
0
Depends upon µC
code being executed
264
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Scanning the ADC
Figure 134 shows a block diagram of the ADC with all relevant control and observe sig-
nals. The Boundary-scan cell from Figure 130 is attached to each of these signals. The
ADC need not be used for pure connectivity testing, since all analog inputs are shared
with a digital port pin as well.
Figure 134. Analog to Digital Converter
VCCREN
AREF
IREFEN
2.56V
Ref
TO COMPARATOR
PASSEN
MUXEN_7
ADC_7
MUXEN_6
ADC_6
MUXEN_5
ADC_5
MUXEN_4
ADC_4
ADCBGEN
SCTEST
1.22V
Ref
EXTCH
MUXEN_3
ADC_3
PRECH
AREF
AREF
DACOUT
COMP
MUXEN_2
ADC_2
MUXEN_1
ADC_1
MUXEN_0
ADC_0
DAC_9..0
+
-
10-bit DAC
G10
G20
ADCEN
HOLD
ACTEN
+
-
+
-
20x
10x
NEGSEL_2
NEGSEL_1
NEGSEL_0
ADC_2
ADC_1
GNDEN
ST
ADC_0
ACLK
AMPEN
The signals are described briefly in Table 105.
265
2490G–AVR–03/04
Table 105. Boundary-scan Signals for the ADC(1)
Direction
Output Values when
Recommended RecommendedInputs
as Seen
from the
ADC
Signal
Name
Input when not
in Use
are Used, and CPU is
not Using the ADC
Description
COMP
ACLK
Output
Input
Comparator Output
0
0
0
0
Clock signal to gain
stages implemented
as Switch-cap filters
ACTEN
Input
Input
Enable path from
gain stages to the
Comparator
0
0
0
0
ADCBGEN
Enable Band-gap
reference as
negative input to
Comparator
ADCEN
AMPEN
DAC_9
DAC_8
DAC_7
DAC_6
DAC_5
DAC_4
DAC_3
DAC_2
DAC_1
DAC_0
EXTCH
Input
Input
Input
Input
Input
Input
Input
Input
Input
Input
Input
Input
Input
Power-on signal to
the ADC
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
1
Power-on signal to
the gain stages
Bit nine of digital
value to DAC
Bit eight of digital
value to DAC
Bit seven of digital
value to DAC
Bit six of digital
value to DAC
Bit five of digital
value to DAC
Bit four of digital
value to DAC
Bit three of digital
value to DAC
Bit two of digital
value to DAC
Bit 1 of digital value
to DAC
Bit 0 of digital value
to DAC
Connect ADC
channels 0 - 3 to
bypass path around
gain stages
G10
G20
Input
Input
Enable 10x gain
Enable 20x gain
0
0
0
0
266
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 105. Boundary-scan Signals for the ADC(1) (Continued)
Direction
Output Values when
as Seen
from the
ADC
Recommended RecommendedInputs
Signal
Name
Input when not
in Use
are Used, and CPU is
not Using the ADC
Description
GNDEN
Input
Ground the negative
input to comparator
when true
0
1
0
1
HOLD
Input
Sample&Hold
signal. Sample
analog signal when
low. Hold signal
when high. If gain
stages are used,
this signal must go
active when ACLK
is high.
IREFEN
Input
Enables Band-gap
reference as AREF
signal to DAC
0
0
MUXEN_7
MUXEN_6
MUXEN_5
MUXEN_4
MUXEN_3
MUXEN_2
MUXEN_1
MUXEN_0
Input
Input
Input
Input
Input
Input
Input
Input
Input Mux bit 7
Input Mux bit 6
Input Mux bit 5
Input Mux bit 4
Input Mux bit 3
Input Mux bit 2
Input Mux bit 1
Input Mux bit 0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
NEGSEL_2 Input
NEGSEL_1 Input
NEGSEL_0 Input
Input Mux for
negative input for
differential signal,
bit 2
Input Mux for
negative input for
differential signal,
bit 1
0
0
0
0
Input Mux for
negative input for
differential signal,
bit 0
PASSEN
PRECH
Input
Input
Enable pass-gate of
gain stages.
1
1
1
1
Precharge output
latch of comparator
(Active low)
267
2490G–AVR–03/04
Table 105. Boundary-scan Signals for the ADC(1) (Continued)
Direction
Output Values when
Recommended RecommendedInputs
as Seen
from the
ADC
Signal
Name
Input when not
in Use
are Used, and CPU is
not Using the ADC
Description
SCTEST
Input
Switch-cap TEST
enable. Output from
x10 gain stage send
out to Port Pin
0
0
having ADC_4
ST
Input
Output of gain
0
0
stages will settle
faster if this signal is
high first two ACLK
periods after
AMPEN goes high.
VCCREN
Input
Selects Vcc as the
ACC reference
voltage.
0
0
Note:
1. Incorrect setting of the switches in Figure 134 will make signal contention and may
damage the part. There are several input choices to the S&H circuitry on the negative
input of the output comparator in Figure 134. Make sure only one path is selected
from either one ADC pin, Bandgap reference source, or Ground.
If the ADC is not to be used during scan, the recommended input values from Table 105
should be used. The user is recommended not to use the Differential Gain stages dur-
ing scan. Switch-cap based gain stages require fast operation and accurate timing
which is difficult to obtain when used in a scan chain. Details concerning operations of
the differential gain stage is therefore not provided.
The AVR ADC is based on the analog circuitry shown in Figure 134 with a successive
approximation algorithm implemented in the digital logic. When used in Boundary-scan,
the problem is usually to ensure that an applied analog voltage is measured within some
limits. This can easily be done without running a successive approximation algorithm:
apply the lower limit on the digital DAC[9:0] lines, make sure the output from the com-
parator is low, then apply the upper limit on the digital DAC[9:0] lines, and verify the
output from the comparator to be high.
The ADC needs not be used for pure connectivity testing, since all analog inputs are
shared with a digital port pin as well.
When using the ADC, remember the following:
•
The Port Pin for the ADC channel in use must be configured to be an input with pull-
up disabled to avoid signal contention.
•
In Normal mode, a dummy conversion (consisting of 10 comparisons) is performed
when enabling the ADC. The user is advised to wait at least 200 ns after enabling
the ADC before controlling/observing any ADC signal, or perform a dummy
conversion before using the first result.
•
The DAC values must be stable at the midpoint value 0x200 when having the HOLD
signal low (Sample mode).
268
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
As an example, consider the task of verifying a 1.5V 5% input signal at ADC channel 3
when the power supply is 5.0V and AREF is externally connected to VCC
.
The lower limit is:
The upper limit is:
1024 ⋅ 1.5V ⋅ 0,95 ⁄ 5V = 291 = 0x123
1024 ⋅ 1.5V ⋅ 1.05 ⁄ 5V = 323 = 0x143
The recommended values from Table 105 are used unless other values are given in the
algorithm in Table 106. Only the DAC and Port Pin values of the Scan-chain are shown.
The column “Actions” describes what JTAG instruction to be used before filling the
Boundary-scan Register with the succeeding columns. The verification should be done
on the data scanned out when scanning in the data on the same row in the table.
Table 106. Algorithm for Using the ADC(1)
PA3.
Pull-
Ste
p
PA3.
PA3.
up_
Enable
Actions
ADCEN
DAC
MUXEN HOLD PRECH Data Control
1
2
SAMPLE_PRELOAD
EXTEST
1
1
1
1
1
1
1
1
1
1
1
0x200
0x200
0x200
0x123
0x123
0x200
0x200
0x200
0x143
0x143
0x200
0x08
0x08
0x08
0x08
0x08
0x08
0x08
0x08
0x08
0x08
0x08
1
0
1
1
1
1
0
1
1
1
1
1
1
1
1
0
1
1
1
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
4
5
6
Verify the COMP bit scanned out to be 0
7
8
9
10
11
Note:
Verify the COMP bit scanned out to be 1
1. Using this algorithm, the timing constraint on the HOLD signal constrains the TCK clock frequency. As the algorithm keeps
HOLD high for five steps, the TCK clock frequency has to be at least five times the number of scan bits divided by the maxi-
mum hold time, thold,max
.
269
2490G–AVR–03/04
ATmega64 Boundary-
scan Order
Table 107 shows the Scan order between TDI and TDO when the Boundary-scan Chain
is selected as data path. Bit 0 is the LSB; the first bit scanned in, and the first bit
scanned out. The scan order follows the pinout order as far as possible. Therefore, the
bits of Port A are scanned in the opposite bit order of the other ports. Exceptions from
the rules are the scan chains for the analog circuits, which constitute the most significant
bits of the scan chain regardless of which physical pin they are connected to. In Figure
127, PXn, Data corresponds to FF0, PXn. Control corresponds to FF1, and PXn.
Pullup_enable corresponds to FF2. Bit 2, 3, 4, and 5 of Port C is not in the scan chain,
since these pins constitute the TAP pins when the JTAG is enabled.
Table 107. ATmega64 Boundary-scan Order
Bit Number
204
203
202
201
200
199
198
197
196
195
194
193
192
191
190
189
188
187
186
185
184
183
182
181
180
179
178
177
176
Signal Name
AC_IDLE
ACO
Module
Comparator
ACME
AINBG
COMP
ADC
PRIVATE_SIGNAL1(1)
ACLK
ACTEN
PRIVATE_SIGNAL2(2)
ADCBGEN
ADCEN
AMPEN
DAC_9
DAC_8
DAC_7
DAC_6
DAC_5
DAC_4
DAC_3
DAC_2
DAC_1
DAC_0
EXTCH
G10
G20
GNDEN
HOLD
IREFEN
MUXEN_7
270
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 107. ATmega64 Boundary-scan Order (Continued)
Bit Number
175
174
173
172
171
170
169
168
167
166
165
164
163
162
161
160
159
158
157
156
155
154
153
152
151
150
149
148
147
146
145
144
143
142
141
140
Signal Name
MUXEN_6
Module
ADC
MUXEN_5
MUXEN_4
MUXEN_3
MUXEN_2
MUXEN_1
MUXEN_0
NEGSEL_2
NEGSEL_1
NEGSEL_0
PASSEN
PRECH
SCTEST
ST
VCCREN
PEN
Programming Enable (Observe-only)
Port E
PE0.Data
PE0.Control
PE0.Pullup_Enable
PE1.Data
PE1.Control
PE1.Pullup_Enable
PE2.Data
PE2.Control
PE2.Pullup_Enable
PE3.Data
PE3.Control
PE3.Pullup_Enable
PE4.Data
PE4.Control
PE4.Pullup_Enable
PE5.Data
PE5.Control
PE5.Pullup_Enable
PE6.Data
PE6.Control
271
2490G–AVR–03/04
Table 107. ATmega64 Boundary-scan Order (Continued)
Bit Number
139
138
137
136
135
134
133
132
131
130
129
128
127
126
125
124
123
122
121
120
119
118
117
116
115
114
113
112
111
110
109
108
107
106
105
104
Signal Name
PE6.Pullup_Enable
PE7.Data
Module
Port E
PE7.Control
PE7.Pullup_Enable
PB0.Data
Port B
PB0.Control
PB0.Pullup_Enable
PB1.Data
PB1.Control
PB1.Pullup_Enable
PB2.Data
PB2.Control
PB2.Pullup_Enable
PB3.Data
PB3.Control
PB3.Pullup_Enable
PB4.Data
PB4.Control
PB4.Pullup_Enable
PB5.Data
PB5.Control
PB5.Pullup_Enable
PB6.Data
PB6.Control
PB6.Pullup_Enable
PB7.Data
PB7.Control
PB7.Pullup_Enable
PG3.Data
Port G
PG3.Control
PG3.Pullup_Enable
PG4.Data
PG4.Control
PG4.Pullup_Enable
TOSC
32 kHz Timer Oscillator
TOSCON
272
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 107. ATmega64 Boundary-scan Order (Continued)
Bit Number
103
102
101
100
99
Signal Name
RSTT
Module
Reset Logic
(Observe-only)
RSTHV
EXTCLKEN
OSCON
Enable Signals for Main Clock/Oscillators
RCOSCEN
98
OSC32EN
97
EXTCLK (XTAL1)
OSCCK
Clock Input and Oscillators for the Main Clock
(Observe-only)
96
95
RCCK
94
OSC32CK
93
TWIEN
TWI
92
PD0.Data
Port D
91
PD0.Control
PD0.Pullup_Enable
PD1.Data
90
89
88
PD1.Control
PD1.Pullup_Enable
PD2.Data
87
86
85
PD2.Control
PD2.Pullup_Enable
PD3.Data
84
83
82
PD3.Control
PD3.Pullup_Enable
PD4.Data
81
80
79
PD4.Control
PD4.Pullup_Enable
PD5.Data
78
77
76
PD5.Control
PD5.Pullup_Enable
PD6.Data
75
74
73
PD6.Control
PD6.Pullup_Enable
PD7.Data
72
71
70
PD7.Control
PD7.Pullup_Enable
PG0.Data
69
68
Port G
273
2490G–AVR–03/04
Table 107. ATmega64 Boundary-scan Order (Continued)
Bit Number
67
Signal Name
PG0.Control
Module
Port G
66
PG0.Pullup_Enable
PG1.Data
65
64
PG1.Control
63
PG1.Pullup_Enable
PC0.Data
62
Port C
61
PC0.Control
60
PC0.Pullup_Enable
PC1.Data
59
58
PC1.Control
57
PC1.Pullup_Enable
PC2.Data
56
55
PC2.Control
54
PC2.Pullup_Enable
PC3.Data
53
52
PC3.Control
51
PC3.Pullup_Enable
PC4.Data
50
49
PC4.Control
48
PC4.Pullup_Enable
PC5.Data
47
46
PC5.Control
45
PC5.Pullup_Enable
PC6.Data
44
43
PC6.Control
42
PC6.Pullup_Enable
PC7.Data
41
40
PC7.Control
39
PC7.Pullup_Enable
PG2.Data
38
Port G
Port A
37
PG2.Control
36
PG2.Pullup_Enable
PA7.Data
35
34
PA7.Control
33
PA7.Pullup_Enable
PA6.Data
32
274
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 107. ATmega64 Boundary-scan Order (Continued)
Bit Number
Signal Name
PA6.Control
Module
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
Port A
PA6.Pullup_Enable
PA5.Data
PA5.Control
PA5.Pullup_Enable
PA4.Data
PA4.Control
PA4.Pullup_Enable
PA3.Data
PA3.Control
PA3.Pullup_Enable
PA2.Data
PA2.Control
PA2.Pullup_Enable
PA1.Data
PA1.Control
PA1.Pullup_Enable
PA0.Data
PA0.Control
PA0.Pullup_Enable
PF3.Data
Port F
PF3.Control
PF3.Pullup_Enable
PF2.Data
8
7
PF2.Control
6
PF2.Pullup_Enable
PF1.Data
5
4
PF1.Control
3
PF1.Pullup_Enable
PF0.Data
2
1
PF0.Control
0
PF0.Pullup_Enable
Notes: 1. PRIVATE_SIGNAL1 should always scanned in as zero.
2. PRIVATE_SIGNAL2 should always scanned in as zero.
275
2490G–AVR–03/04
Boundary-scan
Description Language
Files
Boundary-scan Description Language (BSDL) files describe Boundary-scan capable
devices in a standard format used by automated test-generation software. The order
and function of bits in the Boundary-scan Data Register are included in this description.
276
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Boot Loader Support The Boot Loader Support provides a real Read-While-Write Self-programming mecha-
nism for downloading and uploading program code by the MCU itself. This feature
allows flexible application software updates controlled by the MCU using a Flash-resi-
dent Boot Loader program. The Boot Loader program can use any available data
– Read-While-Write
Self-programming
interface and associated protocol to read code and write (program) that code into the
Flash memory, or read the code from the program memory. The program code within
the Boot Loader section has the capability to write into the entire Flash, including the
Boot Loader Memory. The Boot Loader can thus even modify itself, and it can also
erase itself from the code if the feature is not needed anymore. The size of the Boot
Loader Memory is configurable with Fuses and the Boot Loader has two separate sets
of Boot Lock bits which can be set independently. This gives the user a unique flexibility
to select different levels of protection.
Features
• Read-While-Write Self-programming
• Flexible Boot Memory Size
• High Security (Separate Boot Lock Bits for a Flexible Protection)
• Separate Fuse to Select Reset Vector
• Optimized Page(1) Size
• Code Efficient Algorithm
• Efficient Read-Modify-Write Support
Note:
1. A page is a section in the Flash consisting of several bytes (see Table 124 on page
296) used during programming. The page organization does not affect normal
operation.
Application and Boot
Loader Flash Sections
The Flash memory is organized in two main sections, the Application section and the
Boot Loader section (see Figure 136). The size of the different sections is configured by
the BOOTSZ Fuses as shown in Table 113 on page 289 and Figure 136. These two
sections can have different levels of protection since they have different sets of Lock
bits.
Application Section
The Application section is the section of the Flash that is used for storing the application
code. The protection level for the Application section can be selected by the application
Boot Lock bits (Boot Lock bits 0), see Table 109 on page 280. The Application section
can never store any Boot Loader code since the SPM instruction is disabled when exe-
cuted from the Application section.
BLS – Boot Loader Section
While the Application section is used for storing the application code, the Boot Loader
software must be located in the BLS since the SPM instruction can initiate a program-
ming when executing from the BLS only. The SPM instruction can access the entire
Flash, including the BLS itself. The protection level for the Boot Loader section can be
selected by the Boot Loader Lock bits (Boot Lock bits 1), see Table 110 on page 280.
Read-While-Write and No Whether the CPU supports Read-While-Write or if the CPU is halted during a Boot
Loader software update is dependent on which address that is being programmed. In
addition to the two sections that are configurable by the BOOTSZ Fuses as described
above, the Flash is also divided into two fixed sections, the Read-While-Write (RWW)
section and the No Read-While-Write (NRWW) section. The limit between the RWW-
and NRWW sections is given in “ATmega64 Boot Loader Parameters” on page 289 and
Figure 136 on page 279. The main difference between the two sections is:
Read-While-Write Flash
Sections
•
When erasing or writing a page located inside the RWW section, the NRWW section
can be read during the operation.
•
When erasing or writing a page located inside the NRWW section, the CPU is halted
during the entire operation.
277
2490G–AVR–03/04
Note that the user software can never read any code that is located inside the RWW
section during a Boot Loader software operation. The syntax “Read-While-Write sec-
tion” refers to which section that is being programmed (erased or written), not which
section that actually is being read during a Boot Loader software update.
RWW – Read-While-Write
Section
If a Boot Loader software update is programming a page inside the RWW section, it is
possible to read code from the Flash, but only code that is located in the NRWW sec-
tion. During an ongoing programming, the software must ensure that the RWW section
never is being read. If the user software is trying to read code that is located inside the
RWW section (i.e., by a call/jmp/lpm or an interrupt) during programming, the software
might end up in an unknown state. To avoid this, the interrupts should either be disabled
or moved to the Boot Loader section. The Boot Loader section is always located in the
NRWW section. The RWW section Busy Bit (RWWSB) in the Store Program Memory
Control Register (SPMCSR) will be read as logical one as long as the RWW section is
blocked for reading. After a programming is completed, the RWWSB must be cleared by
software before reading code located in the RWW section. See “Store Program Memory
Control Register – SPMCSR” on page 281. for details on how to clear RWWSB.
NRWW – No Read-While-Write The code located in the NRWW section can be read when the Boot Loader software is
Section
updating a page in the RWW section. When the Boot Loader code updates the NRWW
section, the CPU is halted during the entire Page Erase or Page Write operation.
Table 108. Read-While-Write Features
Which Section does the Z-
pointer Address During the
Programming?
Which Section Can
be Read During
Programming?
Is the
CPU
Halted?
Read-While-
Write
Supported?
RWW section
NRWW section
None
No
Yes
No
NRWW section
Yes
Figure 135. Read-While-Write vs. No Read-While-Write
Read-While-Write
(RWW) Section
Z-pointer
Addresses NRWW
Section
Z-pointer
No Read-While-Write
(NRWW) Section
Addresses RWW
Section
CPU is Halted
During the Operation
Code Located in
NRWW Section
Can be Read During
the Operation
278
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 136. Memory Sections(1)
Program Memory
BOOTSZ = '10'
Program Memory
BOOTSZ = '11'
$0000
$0000
Application Flash Section
Application Flash Section
End RWW
End RWW
Start NRWW
Start NRWW
Application Flash Section
Boot Loader Flash Section
Application Flash Section
Boot Loader Flash Section
End Application
End Application
Start Boot Loader
Flashend
Start Boot Loader
Flashend
$0000
Program Memory
BOOTSZ = '01'
Program Memory
BOOTSZ = '00'
$0000
Application Flash Section
Application flash Section
End RWW, End Application
End RWW
Start NRWW, Start Boot Loader
Start NRWW
Application Flash Section
Boot Loader Flash Section
End Application
Boot Loader Flash Section
Start Boot Loader
Flashend
Flashend
Note:
1. The parameters are given in Table 113 on page 289.
Boot Loader Lock Bits
If no Boot Loader capability is needed, the entire Flash is available for application code.
The Boot Loader has two separate sets of Boot Lock bits which can be set indepen-
dently. This gives the user a unique flexibility to select different levels of protection.
The user can select:
•
•
•
•
To protect the entire Flash from a software update by the MCU.
To protect only the Boot Loader Flash section from a software update by the MCU.
To protect only the Application Flash section from a software update by the MCU.
Allow software update in the entire Flash.
See Table 109 and Table 110 for further details. The Boot Lock bits can be set in soft-
ware and in Serial or Parallel Programming mode, but they can be cleared by a chip
erase command only. The general Write Lock (Lock bit mode 2) does not control the
programming of the Flash memory by SPM instruction. Similarly, the general
279
2490G–AVR–03/04
Read/Write Lock (Lock bit mode 3) does not control reading nor writing by LPM/SPM, if
it is attempted.
Table 109. Boot Lock Bit0 Protection Modes (Application Section)(1)
BLB0 Mode BLB02 BLB01 Protection
1
1
1
No restrictions for SPM or LPM accessing the Application
section.
2
3
1
0
0
0
SPM is not allowed to write to the Application section.
SPM is not allowed to write to the Application section, and
LPM executing from the Boot Loader section is not
allowed to read from the Application section. If Interrupt
Vectors are placed in the Boot Loader section, interrupts
are disabled while executing from the Application section.
4
0
1
LPM executing from the Boot Loader section is not
allowed to read from the Application section. If Interrupt
Vectors are placed in the Boot Loader section, interrupts
are disabled while executing from the Application section.
Note:
1. “1” means unprogrammed, “0” means programmed
Table 110. Boot Lock Bit1 Protection Modes (Boot Loader Section)(1)
BLB1 Mode BLB12 BLB11 Protection
1
1
1
No restrictions for SPM or LPM accessing the Boot Loader
section.
2
3
1
0
0
0
SPM is not allowed to write to the Boot Loader section.
SPM is not allowed to write to the Boot Loader section,
and LPM executing from the Application section is not
allowed to read from the Boot Loader section. If Interrupt
Vectors are placed in the Application section, interrupts
are disabled while executing from the Boot Loader section.
4
0
1
LPM executing from the Application section is not allowed
to read from the Boot Loader section. If Interrupt Vectors
are placed in the Application section, interrupts are
disabled while executing from the Boot Loader section.
Note:
1. “1” means unprogrammed, “0” means programmed
280
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Entering the Boot Loader Entering the Boot Loader takes place by a jump or call from the application program.
This may be initiated by a trigger such as a command received via USART, or SPI inter-
Program
face. Alternatively, the Boot Reset Fuse can be programmed so that the Reset Vector is
pointing to the Boot Flash start address after a reset. In this case, the Boot Loader is
started after a reset. After the application code is loaded, the program can start execut-
ing the application code. Note that the fuses cannot be changed by the MCU itself. This
means that once the Boot Reset Fuse is programmed, the Reset Vector will always
point to the Boot Loader Reset and the fuse can only be changed through the serial or
parallel programming interface.
Table 111. Boot Reset Fuse(1)
BOOTRST
Reset Address
1
0
Reset Vector = Application Reset (address 0x0000)
Reset Vector = Boot Loader Reset (see Table 113 on page 289)
Note:
1. “1” means unprogrammed, “0” means programmed
Store Program Memory
Control Register – SPMCSR
The Store Program Memory Control Register contains the control bits needed to control
the Boot Loader operations.
Bit
7
SPMIE
R/W
0
6
5
–
4
RWWSRE
R/W
3
BLBSET
R/W
0
2
PGWRT
R/W
0
1
PGERS
R/W
0
0
SPMEN
R/W
0
RWWSB
SPMCSR
Read/Write
Initial Value
R
0
R
0
0
• Bit 7 – SPMIE: SPM Interrupt Enable
When the SPMIE bit is written to one, and the I-bit in the Status Register is set (one), the
SPM ready interrupt will be enabled. The SPM ready interrupt will be executed as long
as the SPMEN bit in the SPMCSR Register is cleared.
• Bit 6 – RWWSB: Read-While-Write Section Busy
When a Self-programming (Page Erase or Page Write) operation to the RWW section is
initiated, the RWWSB will be set (one) by hardware. When the RWWSB bit is set, the
RWW section cannot be accessed. The RWWSB bit will be cleared if the RWWSRE bit
is written to one after a Self-programming operation is completed. Alternatively the
RWWSB bit will automatically be cleared if a page load operation is initiated.
• Bit 5 – Res: Reserved Bit
This bit is a reserved bit in the ATmega64 and always read as zero.
• Bit 4 – RWWSRE: Read-While-Write Section Read Enable
When programming (Page Erase or Page Write) to the RWW section, the RWW section
is blocked for reading (the RWWSB will be set by hardware). To re-enable the RWW
section, the user software must wait until the programming is completed (SPMEN will be
cleared). Then, if the RWWSRE bit is written to one at the same time as SPMEN, the
next SPM instruction within four clock cycles re-enables the RWW section. The RWW
section cannot be re-enabled while the Flash is busy with a Page Erase or a Page Write
(SPMEN is set). If the RWWSRE bit is written while the Flash is being loaded, the Flash
load operation will abort and the data loaded will be lost.
281
2490G–AVR–03/04
• Bit 3 – BLBSET: Boot Lock Bit Set
If this bit is written to one at the same time as SPMEN, the next SPM instruction within
four clock cycles sets Boot Lock bits, according to the data in R0. The data in R1 and
the address in the Z-pointer are ignored. The BLBSET bit will automatically be cleared
upon completion of the Lock bit set, or if no SPM instruction is executed within four clock
cycles.
An LPM instruction within three cycles after BLBSET and SPMEN are set in the
SPMCSR Register, will read either the Lock bits or the Fuse bits (depending on Z0 in
the Z-pointer) into the destination register. See “Reading the Fuse and Lock Bits from
Software” on page 286 for details.
• Bit 2 – PGWRT: Page Write
If this bit is written to one at the same time as SPMEN, the next SPM instruction within
four clock cycles executes Page Write, with the data stored in the temporary buffer. The
page address is taken from the high part of the Z-pointer. The data in R1 and R0 are
ignored. The PGWRT bit will auto-clear upon completion of a Page Write, or if no SPM
instruction is executed within four clock cycles. The CPU is halted during the entire
Page Write operation if the NRWW section is addressed.
• Bit 1 – PGERS: Page Erase
If this bit is written to one at the same time as SPMEN, the next SPM instruction within
four clock cycles executes Page Erase. The page address is taken from the high part of
the Z-pointer. The data in R1 and R0 are ignored. The PGERS bit will auto-clear upon
completion of a Page Erase, or if no SPM instruction is executed within four clock
cycles. The CPU is halted during the entire Page Write operation if the NRWW section is
addressed.
• Bit 0 – SPMEN: Store Program Memory Enable
This bit enables the SPM instruction for the next four clock cycles. If written to one
together with either RWWSRE, BLBSET, PGWRT’ or PGERS, the following SPM
instruction will have a special meaning, see description above. If only SPMEN is written,
the following SPM instruction will store the value in R1:R0 in the temporary page buffer
addressed by the Z-pointer. The LSB of the Z-pointer is ignored. The SPMEN bit will
auto-clear upon completion of an SPM instruction, or if no SPM instruction is executed
within four clock cycles. During Page Erase and Page Write, the SPMEN bit remains
high until the operation is completed.
Writing any other combination than “10001”, “01001”, “00101”, “00011” or “00001” in the
lower five bits will have no effect.
282
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Addressing the Flash
During Self-
programming
The Z-pointer is used to address the SPM commands.
Bit
15
Z15
Z7
7
14
Z14
Z6
6
13
Z13
Z5
5
12
Z12
Z4
4
11
Z11
Z3
3
10
Z10
Z2
2
9
Z9
Z1
1
8
Z8
Z0
0
ZH (R31)
ZL (R30)
Since the Flash is organized in pages (see Table 124 on page 296), the Program
Counter can be treated as having two different sections. One section, consisting of the
least significant bits, is addressing the words within a page, while the most significant
bits are addressing the pages. This is shown in Figure 137. Note that the Page Erase
and Page Write operations are addressed independently. Therefore, it is of major impor-
tance that the Boot Loader software addresses the same page in both the Page Erase
and Page Write operation. Once a programming operation is initiated, the address is
latched and the Z-pointer can be used for other operations.
The only SPM operation that does not use the Z-pointer is Setting the Boot Loader Lock
bits. The content of the Z-pointer is ignored and will have no effect on the operation. The
LPM instruction does also use the Z-pointer to store the address. Since this instruction
addresses the Flash byte-by-byte, also the LSB (Bit Z0) of the Z-pointer is used.
Figure 137. Addressing the Flash during SPM(1)Table 2 on page 283
BIT 15
ZPCMSB
ZPAGEMSB
1
0
Z - REGISTER
0
PCMSB
PAGEMSB
PCWORD
PROGRAM
COUNTER
PCPAGE
PAGE ADDRESS
WITHIN THE FLASH
WORD ADDRESS
WITHIN A PAGE
PROGRAM MEMORY
PAGE
PAGE
INSTRUCTION WORD
PCWORD[PAGEMSB:0]:
00
01
02
PAGEEND
Notes: 1. The different variables used in Figure 137 are listed in Table 114 on page 289.
2. PCPAGE and PCWORD are listed in Table 125 on page 296.
283
2490G–AVR–03/04
Self-programming the
Flash
The program memory is updated in a page-by-page fashion. Before programming a
page with the data stored in the temporary page buffer, the page must be erased. The
temporary page buffer is filled one word at a time using SPM and the buffer can be filled
either before the Page Erase command or between a Page Erase and a Page Write
operation:
Alternative 1, fill the buffer before a Page Erase:
•
•
•
Fill temporary page buffer
Perform a Page Erase
Perform a Page Write
Alternative 2, fill the buffer after Page Erase:
•
•
•
Perform a Page Erase
Fill temporary page buffer
Perform a Page Write
If only a part of the page needs to be changed, the rest of the page must be stored (for
example in the temporary page buffer) before the erase, and then be rewritten. When
using Alternative 1, the boot loader provides an effective Read-Modify-Write feature
which allows the user software to first read the page, do the necessary changes, and
then write back the modified data. If Alternative 2 is used, it is not possible to read the
old data while loading since the page is already erased. The temporary page buffer can
be accessed in a random sequence. It is essential that the page address used in both
the Page Erase and Page Write operation is addressing the same page. See “Simple
Assembly Code Example for a Boot Loader” on page 287 for an assembly code
example.
Performing Page Erase by
SPM
To execute Page Erase, set up the address in the Z-pointer, write “X0000011” to
SPMCSR and execute SPM within four clock cycles after writing SPMCSR. The data in
R1 and R0 is ignored. The page address must be written to PCPAGE in the Z-register.
Other bits in the Z-pointer must be written zero during this operation.
•
Page Erase to the RWW section: The NRWW section can be read during the Page
Erase.
•
Page Erase to the NRWW section: The CPU is halted during the operation.
Filling the Temporary Buffer
(Page Loading)
To write an instruction word, set up the address in the Z-pointer and data in R1:R0, write
“00000001” to SPMCSR and execute SPM within four clock cycles after writing
SPMCSR. The content of PCWORD in the Z-register is used to address the data in the
temporary buffer. The temporary buffer will auto-erase after a Page Write operation or
by writing the RWWSRE bit in SPMCSR. It is also erased after a System Reset. Note
that it is not possible to write more than one time to each address without erasing the
temporary buffer.
Note:
If the EEPROM is written in the middle of an SPM Page Load operation, all data loaded
will be lost.
Performing a Page Write
To execute Page Write, set up the address in the Z-pointer, write “X0000101” to
SPMCSR and execute SPM within four clock cycles after writing SPMCSR. The data in
R1 and R0 is ignored. The page address must be written to PCPAGE. Other bits in the
Z-pointer must be written zero during this operation.
•
Page Write to the RWW section: The NRWW section can be read during the Page
Write.
•
Page Write to the NRWW section: The CPU is halted during the operation.
284
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Using the SPM Interrupt
If the SPM interrupt is enabled, the SPM interrupt will generate a constant interrupt
when the SPMEN bit in SPMCSR is cleared. This means that the interrupt can be used
instead of polling the SPMCSR Register in software. When using the SPM interrupt, the
Interrupt Vectors should be moved to the BLS section to avoid that an interrupt is
accessing the RWW section when it is blocked for reading. How to move the interrupts
is described in “Interrupts” on page 59.
Consideration While Updating Special care must be taken if the user allows the Boot Loader section to be updated by
BLS
leaving Boot Lock bit11 unprogrammed. An accidental write to the Boot Loader itself can
corrupt the entire Boot Loader, and further software updates might be impossible. If it is
not necessary to change the Boot Loader software itself, it is recommended to program
the Boot Lock bit11 to protect the Boot Loader software from any internal software
changes.
Prevent Reading the RWW
Section During Self-
programming
During Self-programming (either Page Erase or Page Write), the RWW section is
always blocked for reading. The user software itself must prevent that this section is
addressed during the Self-programming operation. The RWWSB in the SPMCSR will be
set as long as the RWW section is busy. During Self-programming the Interrupt Vector
table should be moved to the BLS as described in “Interrupts” on page 59, or the inter-
rupts must be disabled. Before addressing the RWW section after the programming is
completed, the user software must clear the RWWSB by writing the RWWSRE. See
“Simple Assembly Code Example for a Boot Loader” on page 287 for an example.
Setting the Boot Loader Lock To set the Boot Loader Lock bits, write the desired data to R0, write “X0001001” to
Bits by SPM
SPMCSR and execute SPM within four clock cycles after writing SPMCSR. The only
accessible Lock bits are the Boot Lock bits that may prevent the Application and Boot
Loader section from any software update by the MCU.
Bit
R0
7
6
5
4
3
2
1
0
1
1
BLB12
BLB11
BLB02
BLB01
1
1
See Table 109 and Table 110 for how the different settings of the Boot Loader Bits
affect the Flash access.
If bits 5..2 in R0 are cleared (zero), the corresponding Boot Lock bit will be programmed
if an SPM instruction is executed within four cycles after BLBSET and SPMEN are set in
SPMCSR. The Z-pointer is don’t care during this operation, but for future compatibility it
is recommended to load the Z-pointer with 0x0001 (same as used for reading the Lock
bits). For future compatibility It is also recommended to set bits 7, 6, 1, and 0 in R0 to “1”
when writing the Lock bits. When programming the Lock bits the entire Flash can be
read during the operation.
EEPROM Write Prevents
Writing to SPMCSR
Note that an EEPROM write operation will block all software programming to Flash.
Reading the Fuses and Lock bits from software will also be prevented during the
EEPROM write operation. It is recommended that the user checks the status bit (EEWE)
in the EECR Register and verifies that the bit is cleared before writing to the SPMCSR
Register.
285
2490G–AVR–03/04
Reading the Fuse and Lock
Bits from Software
It is possible to read both the Fuse and Lock bits from software. To read the Lock bits,
load the Z-pointer with 0x0001 and set the BLBSET and SPMEN bits in SPMCSR.
When an LPM instruction is executed within three CPU cycles after the BLBSET and
SPMEN bits are set in SPMCSR, the value of the Lock bits will be loaded in the destina-
tion register. The BLBSET and SPMEN bits will auto-clear upon completion of reading
the Lock bits or if no LPM instruction is executed within three CPU cycles or no SPM
instruction is executed within four CPU cycles. When BLBSET and SPMEN are cleared,
LPM will work as described in the AVR Instruction Set Reference Manual.
Bit
Rd
7
6
5
4
3
2
1
0
–
–
BLB12
BLB11
BLB02
BLB01
LB2
LB1
The algorithm for reading the Fuse Low bits is similar to the one described above for
reading the Lock bits. To read the Fuse Low bits, load the Z-pointer with 0x0000 and set
the BLBSET and SPMEN bits in SPMCSR. When an LPM instruction is executed within
three cycles after the BLBSET and SPMEN bits are set in the SPMCSR, the value of the
Fuse Low bits (FLB) will be loaded in the destination register as shown below. Refer to
Table 120 on page 292 for a detailed description and mapping of the Fuse Low bits.
Bit
Rd
7
6
5
4
3
2
1
0
FLB7
FLB6
FLB5
FLB4
FLB3
FLB2
FLB1
FLB0
Similarly, when reading the Fuse High bits, load 0x0003 in the Z-pointer. When an LPM
instruction is executed within three cycles after the BLBSET and SPMEN bits are set in
the SPMCSR, the value of the Fuse High bits (FHB) will be loaded in the destination
register as shown below. Refer to Table 119 on page 292 for detailed description and
mapping of the Fuse High bits.
Bit
Rd
7
6
5
4
3
2
1
0
FHB7
FHB6
FHB5
FHB4
FHB3
FHB2
FHB1
FHB0
When reading the Extended Fuse bits, load 0x0002 in the Z-pointer. When an LPM
instruction is executed within three cycles after the BLBSET and SPMEN bits are set in
the SPMCSR, the value of the Extended Fuse bits (EFB) will be loaded in the destina-
tion register as shown below. Refer to Table 118 on page 291 for detailed description
and mapping of the Fuse High bits.
Bit
Rd
7
6
5
4
3
2
1
0
–
–
–
–
–
–
EFB1
EFB0
Fuse and Lock bits that are programmed will be read as zero. Fuse and Lock bits that
are unprogrammed will be read as one.
Preventing Flash Corruption
During periods of low VCC, the Flash program can be corrupted because the supply volt-
age is too low for the CPU and the Flash to operate properly. These issues are the same
as for board level systems using the Flash, and the same design solutions should be
applied.
A Flash program corruption can be caused by two situations when the voltage is too low.
First, a regular write sequence to the Flash requires a minimum voltage to operate cor-
rectly. Second, the CPU itself can execute instructions incorrectly, if the supply voltage
for executing instructions is too low.
Flash corruption can easily be avoided by following these design recommendations (one
is sufficient):
286
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
1. If there is no need for a Boot Loader update in the system, program the Boot
Loader Lock bits to prevent any Boot Loader software updates.
2. Keep the AVR RESET active (low) during periods of insufficient power supply
voltage. This can be done by enabling the internal Brown-out Detector (BOD) if
the operating voltage matches the detection level. If not, an external low VCC
Reset Protection circuit can be used. If a reset occurs while a write operation is
in progress, the write operation will be completed provided that the power supply
voltage is sufficient.
3. Keep the AVR core in Power-down Sleep mode during periods of low VCC. This
will prevent the CPU from attempting to decode and execute instructions, effec-
tively protecting the SPMCSR Register and thus the Flash from unintentional
writes.
Programming Time for Flash
when Using SPM
The calibrated RC Oscillator is used to time Flash accesses. Table 112 shows the typi-
cal programming time for Flash accesses from the CPU.
Table 112. SPM Programming Time
Symbol
Min Programming Time Max Programming Time
3.7 ms 4.5 ms
Flash write (Page Erase, Page Write,
and write Lock bits by SPM)
Simple Assembly Code
Example for a Boot Loader
;-the routine writes one page of data from RAM to Flash
; the first data location in RAM is pointed to by the Y pointer
; the first data location in Flash is pointed to by the Z-pointer
;-error handling is not included
;-the routine must be placed inside the boot space
; (at least the Do_spm sub routine). Only code inside NRWW section can
; be read during self-programming (Page Erase and Page Write).
;-registers used: r0, r1, temp1 (r16), temp2 (r17), looplo (r24),
; loophi (r25), spmcrval (r20)
; storing and restoring of registers is not included in the routine
; register usage can be optimized at the expense of code size
;-It is assumed that either the interrupt table is moved to the Boot
; loader section or that the interrupts are disabled.
.equ PAGESIZEB = PAGESIZE*2
words
;PAGESIZEB is page size in BYTES, not
.org SMALLBOOTSTART
Write_page:
; Page Erase
ldi
spmcrval, (1<<PGERS) | (1<<SPMEN)
call Do_spm
; re-enable the RWW section
ldi
spmcrval, (1<<RWWSRE) | (1<<SPMEN)
call Do_spm
; transfer data from RAM to Flash page buffer
ldi
ldi
looplo, low(PAGESIZEB) ;init loop variable
loophi, high(PAGESIZEB) ;not required for PAGESIZEB<=256
Wrloop:
ld
r0, Y+
ld
r1, Y+
ldi
spmcrval, (1<<SPMEN)
call Do_spm
adiw ZH:ZL, 2
sbiw loophi:looplo, 2
brne Wrloop
;use subi for PAGESIZEB<=256
; execute Page Write
287
2490G–AVR–03/04
subi ZL, low(PAGESIZEB)
sbci ZH, high(PAGESIZEB)
;restore pointer
;not required for PAGESIZEB<=256
ldi
spmcrval, (1<<PGWRT) | (1<<SPMEN)
call Do_spm
; re-enable the RWW section
ldi
spmcrval, (1<<RWWSRE) | (1<<SPMEN)
call Do_spm
; read back and check, optional
ldi
ldi
looplo, low(PAGESIZEB) ;init loop variable
loophi, high(PAGESIZEB) ;not required for PAGESIZEB<=256
subi YL, low(PAGESIZEB)
sbci YH, high(PAGESIZEB)
;restore pointer
Rdloop:
lpm
ld
r0, Z+
r1, Y+
cpse r0, r1
jmp Error
sbiw loophi:looplo, 1
brne Rdloop
;use subi for PAGESIZEB<=256
; return to RWW section
; verify that RWW section is safe to read
Return:
lds
sbrs temp1, RWWSB
yet
temp1, SPMCSR
; If RWWSB is set, the RWW section is not ready
ret
; re-enable the RWW section
ldi
spmcrval, (1<<RWWSRE) | (1<<SPMEN)
call Do_spm
rjmp Return
Do_spm:
; check for previous SPM complete
Wait_spm:
lds
temp1, SPMCSR
sbrc temp1, SPMEN
rjmp Wait_spm
; input: spmcrval determines SPM action
; disable interrupts if enabled, store status
in
temp2, SREG
cli
; check that no EEPROM write access is present
Wait_ee:
sbic EECR, EEWE
rjmp Wait_ee
; SPM timed sequence
sts
spm
SPMCSR, spmcrval
; restore SREG (to enable interrupts if originally enabled)
out
ret
SREG, temp2
288
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
ATmega64 Boot Loader
Parameters
In Table 113 through Table 115, the parameters used in the description of the Self-pro-
gramming are given.
Table 113. Boot Size Configuration(1)
Boot Reset
Appli-
cation
Flash
Boot
Loader
Flash
End
Applic-
ation
Address
(Start Boot
Loader
BOOTSZ BOOTSZ Boot
1
0
Size
Pages Section
Section
Section Section)
512
words
0x0000 -
0x7DFF
0x7E00-
0x7FFF
1
1
4
0x7DFF 0x7E00
1024
words
0x0000 -
0x7BFF
0x7C00-
0x7FFF
1
0
0
0
1
0
8
0x7BFF
0x77FF
0x6FFF
0x7C00
0x7800
0x7000
2048
words
0x0000 -
0x77FF
0x7800 -
0x7FFF
16
4096
words
0x0000 -
32
0x7000 -
0x7FFF
0x6FFF
Note:
1. The different BOOTSZ Fuse configurations are shown in Figure 136
Table 114. Read-While-Write Limit(1)
Section
Pages
224
Address
Read-While-Write (RWW)
No Read-While-Write (NRWW)
0x0000 - 0x6FFF
0x7000 - 0x7FFF
32
Note:
1. For details about these two section, see “NRWW – No Read-While-Write Section” on
page 278 and “RWW – Read-While-Write Section” on page 278
Table 115. Explanation of Different Variables Used in Figure 137 and the Mapping to
the Z-pointer(1)(2)
Corresponding
Variable
Z-value
Description
14
6
Most significant bit in the Program Counter.
(The Program Counter is 15 bits PC[14:0]).
PCMSB
Most significant bit which is used to address the
words within one page (128 words in a page
requires seven bits PC [6:0]).
PAGEMSB
ZPCMSB
Z15
Z7
Bit in Z-register that is mapped to PCMSB.
Because Z0 is not used, the ZPCMSB equals
PCMSB + 1.
Bit in Z-register that is mapped to PAGEMSB.
Because Z0 is not used, the ZPAGEMSB
equals PAGEMSB + 1.
ZPAGEMSB
PCPAGE
PC[14:7]
PC[6:0]
Z15:Z8
Z7:Z1
Program Counter page address: Page select,
for Page Erase and Page Write
Program Counter word address: Word select,
for filling temporary buffer (must be zero during
Page Write operation)
PCWORD
Notes: 1. Z0: should be zero for all SPM commands, byte select for the LPM instruction.
2. See “Addressing the Flash During Self-programming” on page 283 for details about
the use of Z-pointer during Self-programming.
289
2490G–AVR–03/04
Memory
Programming
Program and Data
Memory Lock Bits
The ATmega64 provides six Lock bits which can be left unprogrammed (“1”) or can be
programmed (“0”) to obtain the additional features listed in Table 117. The Lock bits can
only be erased to “1” with the Chip Erase command.
Table 116. Lock Bit Byte(1)
Lock Bit Byte
Bit no
Description
–
Default Value
7
6
5
4
3
2
1
0
1 (unprogrammed)
1 (unprogrammed)
1 (unprogrammed)
1 (unprogrammed)
1 (unprogrammed)
1 (unprogrammed)
1 (unprogrammed)
1 (unprogrammed)
–
BLB12
BLB11
BLB02
BLB01
LB2
Boot Lock bit
Boot Lock bit
Boot Lock bit
Boot Lock bit
Lock bit
LB1
Lock bit
Note:
Table 117. Lock Bit Protection Modes(2)
Memory Lock Bits Protection Type
1. “1” means unprogrammed, “0” means programmed
LB Mode
LB2
1
LB1
1
1
2
No memory lock features enabled.
1
0
Further programming of the Flash and EEPROM is
disabled in Parallel and SPI/JTAG Serial Programming
mode. The Fuse bits are locked in both Serial and Parallel
Programming mode.(1)
3
0
0
Further programming and verification of the Flash and
EEPROM is disabled in Parallel and SPI/JTAG Serial
Programming mode. The Fuse bits are locked in both
Serial and Parallel Programming mode.(1)
BLB0 Mode BLB02 BLB01
1
1
1
No restrictions for SPM or LPM accessing the Application
section.
2
3
1
0
0
0
SPM is not allowed to write to the Application section.
SPM is not allowed to write to the Application section, and
LPM executing from the Boot Loader section is not
allowed to read from the Application section. If Interrupt
Vectors are placed in the Boot Loader section, interrupts
are disabled while executing from the Application section.
4
0
1
LPM executing from the Boot Loader section is not
allowed to read from the Application section. If Interrupt
Vectors are placed in the Boot Loader section, interrupts
are disabled while executing from the Application section.
BLB1 Mode BLB12 BLB11
290
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 117. Lock Bit Protection Modes(2) (Continued)
Memory Lock Bits
Protection Type
1
1
1
No restrictions for SPM or LPM accessing the Boot Loader
section.
2
3
1
0
0
0
SPM is not allowed to write to the Boot Loader section.
SPM is not allowed to write to the Boot Loader section,
and LPM executing from the Application section is not
allowed to read from the Boot Loader section. If Interrupt
Vectors are placed in the Application section, interrupts
are disabled while executing from the Boot Loader section.
4
0
1
LPM executing from the Application section is not allowed
to read from the Boot Loader section. If Interrupt Vectors
are placed in the Application section, interrupts are
disabled while executing from the Boot Loader section.
Notes: 1. Program the Fuse bits before programming the Lock bits.
2. “1” means unprogrammed, “0” means programmed
Fuse Bits
The ATmega64 has three fuse bytes. Table 118 - Table 120 describe briefly the func-
tionality of all the fuses and how they are mapped into the fuse bytes. Note that the
fuses are read as logical zero, “0”, if they are programmed.
Table 118. Extended Fuse Byte
Fuse Low Byte
Bit no
Description
Default Value
–
7
6
5
4
3
2
1
0
–
1
–
–
1
–
–
1
–
–
1
–
–
1
–
–
1
M103C(1)
WDTON(2)
ATmega103 compatibility mode
Watchdog Timer always on
0 (programmed)
1 (unprogrammed)
Notes: 1. See “ATmega103 and ATmega64 Compatibility” on page 4 for details.
2. See “Watchdog Timer Control Register – WDTCR” on page 55 for details.
291
2490G–AVR–03/04
Table 119. Fuse High Byte
Fuse High Byte Bit no Description
Default Value
OCDEN
7
6
5
Enable OCD
1 (unprogrammed, OCD
disabled)
JTAGEN(4)
SPIEN(1)
Enable JTAG
0 (programmed, JTAG
enabled)
Enable SPI Serial Program and
Data Downloading
0 (programmed, SPI prog.
enabled)
CKOPT(2)
EESAVE
4
3
Oscillator options
1 (unprogrammed)
EEPROM memory is preserved
through the Chip Erase
1 (unprogrammed,
EEPROM not preserved)
BOOTSZ1
BOOTSZ0
BOOTRST
2
1
0
Select Boot Size (see Table 113
for details)
0 (programmed)(3)
0 (programmed)(3)
1 (unprogrammed)
Select Boot Size (see Table 113
for details)
Select Reset Vector
Notes: 1. The SPIEN Fuse is not accessible in SPI Serial Programming mode.
2. The CKOPT Fuse functionality depends on the setting of the CKSEL bits. See “Clock
Sources” on page 36 for details.
3. The default value of BOOTSZ1..0 results in maximum Boot Size. See Table 113 on
page 289
4. If the JTAG interface is left unconnected, the JTAGEN fuse should if possible be dis-
abled. This to avoid static current at the TDO pin in the JTAG interface
Table 120. Fuse Low Byte
Fuse Low Byte Bit no Description
Default Value
BODLEVEL
7
Brown out detector trigger
level
1 (unprogrammed)
BODEN
6
Brown out detector enable
1 (unprogrammed, BOD
disabled)
SUT1
5
4
3
2
1
0
Select start-up time
Select start-up time
Select Clock source
Select Clock source
Select Clock source
Select Clock source
1 (unprogrammed)(1)
0 (programmed)(1)
0 (programmed)(2)
0 (programmed)(2)
0 (programmed)(2)
1 (unprogrammed)(2)
SUT0
CKSEL3
CKSEL2
CKSEL1
CKSEL0
Notes: 1. The default value of SUT1..0 results in maximum start-up time. See Table 14 on page
40 for details.
2. The default setting of CKSEL3..0 results in internal RC Oscillator @ 1 MHz. See
Table 6 on page 36 for details.
The status of the Fuse bits is not affected by Chip Erase. Note that the Fuse bits are
locked if Lock bit1 (LB1) is programmed. Program the Fuse bits before programming the
Lock bits.
292
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Latching of Fuses
The fuse values are latched when the device enters Programming mode and changes of
the fuse values will have no effect until the part leaves Programming mode. This does
not apply to the EESAVE Fuse which will take effect once it is programmed. The fuses
are also latched on Power-up in Normal mode.
Signature Bytes
All Atmel microcontrollers have a 3-byte signature code which identifies the device. This
code can be read in both Serial and Parallel mode, also when the device is locked. The
three bytes reside in a separate address space.
For the ATmega64 the signature bytes are:
1. 0x000: 0x1E (indicates manufactured by Atmel)
2. 0x001: 0x96 (indicates 64KB Flash memory)
3. 0x002: 0x02 (indicates ATmega64 device when 0x001 is 0x96)
Calibration Byte
The ATmega64 stores four different calibration values for the internal RC Oscillator.
These bytes resides in the signature row high byte of the addresses 0x000, 0x0001,
0x0002, and 0x0003 for 1, 2, 4, and 8 MHz respectively. During Reset, the 1 MHz value
is automatically loaded into the OSCCAL Register. If other frequencies are used, the
calibration value has to be loaded manually, see “Oscillator Calibration Register – OSC-
CAL(1)” on page 40 for details.
Parallel Programming
Parameters, Pin
Mapping, and
This section describes how to parallel program and verify Flash Program memory,
EEPROM Data memory, Memory Lock bits, and Fuse bits in the ATmega64. Pulses are
assumed to be at least 250 ns unless otherwise noted.
Commands
Signal Names
In this section, some pins of the ATmega64 are referenced by signal names describing
their functionality during parallel programming, see Figure 138 and Table 121. Pins not
described in the following table are referenced by pin names.
The XA1/XA0 pins determine the action executed when the XTAL1 pin is given a posi-
tive pulse. The bit coding is shown in Table 123.
When pulsing WR or OE, the command loaded determines the action executed. The dif-
ferent Commands are shown in Table 124.
293
2490G–AVR–03/04
Figure 138. Parallel Programming
+5V
+5V
RDY/BSY
OE
PD1
PD2
PD3
PD4
PD5
PD6
PD7
VCC
WR
AVCC
PB7 - PB0
DATA
BS1
XA0
XA1
PAGEL
+12 V
BS2
RESET
PA0
XTAL1
GND
294
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 121. Pin Name Mapping
Signal Name in
Programming Mode Pin Name I/O Function
RDY/BSY
PD1
O
0: Device is busy programming, 1: Device is ready
for new command
OE
PD2
PD3
PD4
I
I
I
Output Enable (Active low)
Write Pulse (Active low)
WR
BS1
Byte Select 1 (“0” selects low byte, “1” selects high
byte)
XA0
PD5
PD6
PD7
PA0
I
I
I
I
XTAL Action Bit 0
XA1
XTAL Action Bit 1
PAGEL
BS2
Program Memory and EEPROM data Page Load
Byte Select 2 (“0” selects low byte, “1” selects 2’nd
high byte)
DATA
PB7 - 0
I/O Bi-directional Data bus (Output when OE is low)
Table 122. Pin Values Used to Enter Programming Mode
Pin
PAGEL
XA1
Symbol
Value
Prog_enable[3]
Prog_enable[2]
Prog_enable[1]
Prog_enable[0]
0
0
0
0
XA0
BS1
Table 123. XA1 and XA0 Coding
XA1 XA0 Action when XTAL1 is Pulsed
0
0
1
1
0
1
0
1
Load Flash or EEPROM Address (High or low address byte determined by BS1)
Load Data (High or Low data byte for Flash determined by BS1)
Load Command
No Action, Idle
295
2490G–AVR–03/04
Table 124. Command Byte Bit Coding
Command Byte
1000 0000
0100 0000
0010 0000
0001 0000
0001 0001
0000 1000
0000 0100
0000 0010
0000 0011
Command Executed
Chip Erase
Write Fuse Bits
Write Lock Bits
Write Flash
Write EEPROM
Read Signature Bytes and Calibration byte
Read Fuse and Lock Bits
Read Flash
Read EEPROM
Table 125. No. of Words in a Page and no. of Pages in the Flash
Flash Size
Page Size
PCWORD
No. of Pages
PCPAGE
PCMSB
32K words (64K bytes)
128 words
PC[6:0]
256
PC[14:7]
14
Table 126. No. of Words in a Page and no. of Pages in the EEPROM
EEPROM Size
Page Size
PCWORD
No. of Pages
PCPAGE
EEAMSB
2K bytes
8 bytes
EEA[2:0]
256
EEA[10:3]
10
296
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Parallel Programming
Enter Programming Mode
The following algorithm puts the device in Parallel Programming mode:
1. Apply 4.5 - 5.5V between VCC and GND, and wait at least 100 µs.
2. Set RESET to “0” and toggle XTAL1 at least six times.
3. Set the Prog_enable pins listed in Table 122 on page 295 to “0000” and wait at
least 100 ns.
4. Apply 11.5 - 12.5V to RESET. Any activity on Prog_enable pins within 100 ns
after +12V has been applied to RESET, will cause the device to fail entering Pro-
gramming mode.
Note, if External Crystal or External RC configuration is selected, it may not be possible
to apply qualified XTAL1 pulses. In such cases, the following algorithm should be
followed:
1. Set Prog_enable pins listed in Table on page 295 to “0000”.
2. Apply 4.5 - 5.5V between VCC and GND simultaneously as 11.5 - 12.5V is
applied to RESET.
3. Wait 100 µs.
4. Re-program the fuses to ensure that External Clock is selected as clock source
(CKSEL3:0 = 0b0000) If Lock bits are programmed, a Chip Erase command
must be executed before changing the fuses.
5. Exit Programming mode by power the device down or by bringing RESET pin to
0b0.
6. Entering Programming mode with the original algorithm, as described above.
Considerations for Efficient
Programming
The loaded command and address are retained in the device during programming. For
efficient programming, the following should be considered.
•
•
•
The command needs only be loaded once when writing or reading multiple memory
locations.
Skip writing the data value 0xFF, that is the contents of the entire EEPROM (unless
the EESAVE Fuse is programmed) and Flash after a Chip Erase.
Address high byte needs only be loaded before programming or reading a new 256
word window in Flash or 256 byte EEPROM. This consideration also applies to
Signature bytes reading.
Chip Erase
The Chip Erase will erase the Flash and EEPROM(1) memories plus Lock bits. The Lock
bits are not reset until the program memory has been completely erased. The Fuse bits
are not changed. A Chip Erase must be performed before the Flash and/or the
EEPROM reprogrammed.
Note:
1. The EEPRPOM memory is preserved during chip erase if the EESAVE Fuse is
programmed.
Load Command “Chip Erase”
1. Set XA1, XA0 to “10”. This enables command loading.
2. Set BS1 to “0”.
3. Set DATA to “1000 0000”. This is the command for Chip Erase.
4. Give XTAL1 a positive pulse. This loads the command.
5. Give WR a negative pulse. This starts the Chip Erase. RDY/BSY goes low.
6. Wait until RDY/BSY goes high before loading a new command.
297
2490G–AVR–03/04
Programming the Flash
The Flash is organized in pages, see Table 124 on page 296. When programming the
Flash, the program data is latched into a page buffer. This allows one page of program
data to be programmed simultaneously. The following procedure describes how to pro-
gram the entire Flash memory:
A. Load Command “Write Flash”
1. Set XA1, XA0 to “10”. This enables command loading.
2. Set BS1 to “0”.
3. Set DATA to “0001 0000”. This is the command for Write Flash.
4. Give XTAL1 a positive pulse. This loads the command.
B. Load Address Low byte
1. Set XA1, XA0 to “00”. This enables address loading.
2. Set BS1 to “0”. This selects low address.
3. Set DATA = Address low byte (0x00 - 0xFF).
4. Give XTAL1 a positive pulse. This loads the address low byte.
C. Load Data Low Byte
1. Set XA1, XA0 to “01”. This enables data loading.
2. Set DATA = Data low byte (0x00 - 0xFF).
3. Give XTAL1 a positive pulse. This loads the data byte.
D. Load Data High Byte
1. Set BS1 to “1”. This selects high data byte.
2. Set XA1, XA0 to “01”. This enables data loading.
3. Set DATA = Data high byte (0x00 - 0xFF).
4. Give XTAL1 a positive pulse. This loads the data byte.
E. Latch Data
1. Set BS1 to “1”. This selects high data byte.
2. Give PAGEL a positive pulse. This latches the data bytes. (See Figure 140 for
signal waveforms).
F. Repeat B through E until the entire buffer is filled or until all data within the page is
loaded.
While the lower bits in the address are mapped to words within the page, the higher bits
address the pages within the Flash. This is illustrated in Figure 139 on page 299. Note
that if less than eight bits are required to address words in the page (pagesize < 256),
the most significant bit(s) in the address low byte are used to address the page when
performing a Page Write.
G. Load Address High byte
1. Set XA1, XA0 to “00”. This enables address loading.
2. Set BS1 to “1”. This selects high address.
3. Set DATA = Address high byte (0x00 - 0xFF).
4. Give XTAL1 a positive pulse. This loads the address high byte.
H. Program Page
1. Set BS1 = “0”.
2. Give WR a negative pulse. This starts programming of the entire page of data.
RDY/BSYgoes low.
298
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
3. Wait until RDY/BSY goes high. (See Figure 140 for signal waveforms.)
I. Repeat B through H until the entire Flash is programmed or until all data has been
programmed.
J. End Page Programming
1. 1. Set XA1, XA0 to “10”. This enables command loading.
2. Set DATA to “0000 0000”. This is the command for No Operation.
3. Give XTAL1 a positive pulse. This loads the command, and the internal write sig-
nals are reset.
Figure 139. Addressing the Flash which is Organized in Pages(1)
PCMSB
PAGEMSB
PROGRAM
COUNTER
PCPAGE
PCWORD
PAGE ADDRESS
WITHIN THE FLASH
WORD ADDRESS
WITHIN A PAGE
PROGRAM MEMORY
PAGE
PAGE
INSTRUCTION WORD
PCWORD[PAGEMSB:0]:
00
01
02
PAGEEND
Note:
1. PCPAGE and PCWORD are listed in Table 124 on page 296.
299
2490G–AVR–03/04
Figure 140. Programming the Flash Waveforms(1)
F
A
B
C
D
E
B
C
D
E
G
H
$10
ADDR. LOW DATA LOW
DATA HIGH
ADDR. LOW DATA LOW
DATA HIGH
XX
ADDR. HIGH
XX
XX
DATA
XA1
XA0
BS1
XTAL1
WR
RDY/BSY
RESET +12V
OE
PAGEL
BS2
Note:
1. “XX” is don’t care. The letters refer to the programming description above.
300
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Programming the EEPROM
The EEPROM is organized in pages, see Table 125 on page 296. When programming
the EEPROM, the program data is latched into a page buffer. This allows one page of
data to be programmed simultaneously. The programming algorithm for the EEPROM
data memory is as follows (refer to “Programming the Flash” on page 298 for details on
Command, Address and Data loading):
1. A: Load Command “0001 0001”.
2. G: Load Address High Byte (0x00 - 0xFF).
3. B: Load Address Low Byte (0x00 - 0xFF).
4. C: Load Data (0x00 - 0xFF).
5. E: Latch data (give PAGEL a positive pulse).
K: Repeat 3 through 5 until the entire buffer is filled.
L: Program EEPROM page
1. Set BS1 to “0”.
2. Give WR a negative pulse. This starts programming of the EEPROM page.
RDY/BSY goes low.
3. Wait until to RDY/BSY goes high before programming the next page.
(See Figure 141 for signal waveforms.)
Figure 141. Programming the EEPROM Waveforms
K
A
G
B
C
E
B
C
E
L
0x11
ADDR. HIGH ADDR. LOW
DATA
ADDR. LOW
DATA
XX
XX
DATA
XA1
XA0
BS1
XTAL1
WR
RDY/BSY
RESET +12V
OE
PAGEL
BS2
Reading the Flash
The algorithm for reading the Flash memory is as follows (refer to “Programming the
Flash” on page 298 for details on Command and Address loading):
1. A: Load Command “0000 0010”.
2. G: Load Address High Byte (0x00 - 0xFF).
3. B: Load Address Low Byte (0x00 - 0xFF).
4. Set OE to “0”, and BS1 to “0”. The Flash word low byte can now be read at DATA.
5. Set BS to “1”. The Flash word high byte can now be read at DATA.
6. Set OE to “1”.
301
2490G–AVR–03/04
Reading the EEPROM
The algorithm for reading the EEPROM memory is as follows (refer to “Programming the
Flash” on page 298 for details on Command and Address loading):
1. A: Load Command “0000 0011”.
2. G: Load Address High Byte (0x00 - 0xFF).
3. B: Load Address Low Byte (0x00 - 0xFF).
4. Set OE to “0”, and BS1 to “0”. The EEPROM Data byte can now be read at
DATA.
5. Set OE to “1”.
Programming the Fuse Low
Bits
The algorithm for programming the Fuse Low bits is as follows (refer to “Programming
the Flash” on page 298 for details on Command and Data loading):
1. A: Load Command “0100 0000”.
2. C: Load Data Low Byte. Bit n = “0” programs and bit n = “1” erases the Fuse bit.
3. Set BS1 to “0” and BS2 to “0”.
4. Give WR a negative pulse and wait for RDY/BSY to go high.
Programming the Fuse High
Bits
The algorithm for programming the Fuse High bits is as follows (refer to “Programming
the Flash” on page 298 for details on Command and Data loading):
1. A: Load Command “0100 0000”.
2. C: Load Data Low Byte. Bit n = “0” programs and bit n = “1” erases the Fuse bit.
3. Set BS1 to “1” and BS2 to “0”. This selects high data byte.
4. Give WR a negative pulse and wait for RDY/BSY to go high.
5. Set BS1 to “0”. This selects low data byte.
Programming the Extended
Fuse Bits
The algorithm for programming the Extended Fuse bits is as follows (refer to “Program-
ming the Flash” on page 298 for details on Command and Data loading):
1. A: Load Command “0100 0000”.
2. C: Load Data Low Byte. Bit n = “0” programs and bit n = “1” erases the Fuse bit.
3. Set BS2 to “1” and BS1 to “0”. This selects extended data byte.
4. Give WR a negative pulse and wait for RDY/BSY to go high.
5. Set BS2 to “0”. This selects low data byte.
302
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 142. Programming the FUSES Waveforms
Write Fuse Low byte
Write Fuse high byte
Write Extended Fuse byte
A
C
A
C
A
C
0x40
DATA
XX
0x40
DATA
XX
0x40
DATA
XX
DATA
XA1
XA0
BS1
BS2
XTAL1
WR
RDY/BSY
RESET +12V
OE
PAGEL
Programming the Lock Bits
The algorithm for programming the Lock bits is as follows (refer to “Programming the
Flash” on page 298 for details on Command and Data loading):
1. A: Load Command “0010 0000”.
2. C: Load Data Low Byte. Bit n = “0” programs the Lock bit.
3. Give WR a negative pulse and wait for RDY/BSY to go high.
The Lock bits can only be cleared by executing Chip Erase.
Reading the Fuse and Lock
Bits
The algorithm for reading the Fuse and Lock bits is as follows (refer to “Programming
the Flash” on page 298 for details on Command loading):
1. A: Load Command “0000 0100”.
2. Set OE to “0”, BS2 to “0” and BS1 to “0”. The status of the Fuse Low bits can
now be read at DATA (“0” means programmed).
3. Set OE to “0”, BS2 to “1” and BS1 to “1”. The status of the Fuse High bits can
now be read at DATA (“0” means programmed).
4. Set OE to “0”, BS2 to “1” and BS1 to “0”. The status of the Extended Fuse bits
can now be read at DATA (“0” means programmed).
5. Set OE to “0”, BS2 to “0” and BS1 to “1”. The status of the Lock bits can now be
read at DATA (“0” means programmed).
6. Set OE to “1”.
303
2490G–AVR–03/04
Figure 143. Mapping Between BS1, BS2 and the Fuse and Lock Bits during Read
0
Fuse Low Byte
Extended Fuse Byte
BS2
1
0
1
DATA
0
1
Lock Bits
BS1
Fuse High Byte
BS2
Reading the Signature Bytes
The algorithm for reading the Signature bytes is as follows (refer to “Programming the
Flash” for details on Command and Address loading):
1. A: Load Command “0000 1000”.
2. B: Load Address Low Byte (0x00 - 0x02).
3. Set OE to “0”, and BS1 to “0”. The selected Signature byte can now be read at
DATA.
4. Set OE to “1”.
Reading the Calibration Byte
The algorithm for reading the Calibration bytes is as follows (refer to “Programming the
Flash” for details on Command and Address loading):
1. A: Load Command “0000 1000”.
2. B: Load Address Low Byte, (0x00 - 0x03).
3. Set OE to “0”, and BS1 to “1”. The Calibration byte can now be read at DATA.
4. Set OE to “1”.
Parallel Programming
Characteristics
Figure 144. Parallel Programming Timing, Including some General Timing
Requirements
tXLWL
tXHXL
XTAL1
tDVXH
tXLDX
Data & Contol
(DATA, XA0/1, BS1, BS2)
tBVPH
tPLBX
t BVWL
tWLBX
PAGEL
tPHPL
tWL WH
WR
tPLWL
WLRL
RDY/BSY
tWLRH
304
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 145. Parallel Programming Timing, Loading Sequence with Timing
Requirements(1)
LOAD ADDRESS
(LOW BYTE)
LOAD DATA
(LOW BYTE)
LOAD DATA LOAD DATA
(HIGH BYTE)
LOAD ADDRESS
(LOW BYTE)
tXLPH
t XLXH
tPLXH
XTAL1
BS1
PAGEL
DATA
ADDR0 (Low Byte)
DATA (Low Byte)
DATA (High Byte)
ADDR1 (Low Byte)
XA0
XA1
Note:
1. The timing requirements shown in Figure 144 (i.e. tDVXH, tXHXL, and tXLDX) also apply
to loading operation.
Figure 146. Parallel Programming Timing, Reading Sequence (Within the Same Page)
with Timing Requirements(1)
LOAD ADDRESS
(LOW BYTE)
READ DATA
(LOW BYTE)
READ DATA
(HIGH BYTE)
LOAD ADDRESS
(LOW BYTE)
tXLOL
XTAL1
BS1
tBHDV
tOLDV
OE
tOHDZ
ADDR1 (Low Byte)
DATA (High Byte)
DATA
ADDR0 (Low Byte)
DATA (Low Byte)
XA0
XA1
Note:
1. The timing requirements shown in Figure 144 (i.e. tDVXH, tXHXL, and tXLDX) also apply
to reading operation.
305
2490G–AVR–03/04
Table 127. Parallel Programming Characteristics, VCC = 5V 10%
Symbol
VPP
Parameter
Min
Typ
Max
12.5
250
Units
V
Programming Enable Voltage
Programming Enable Current
Data and Control Valid before XTAL1 High
XTAL1 Low to XTAL1 High
XTAL1 Pulse Width High
Data and Control Hold after XTAL1 Low
XTAL1 Low to WR Low
11.5
IPP
µA
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
µs
ms
ms
ns
ns
ns
ns
tDVXH
tXLXH
tXHXL
tXLDX
tXLWL
tXLPH
tPLXH
tBVPH
tPHPL
tPLBX
tWLBX
tPLWL
tBVWL
tWLWH
tWLRL
tWLRH
tWLRH_CE
tXLOL
tBVDV
tOLDV
tOHDZ
67
200
150
67
0
XTAL1 Low to PAGEL high
PAGEL low to XTAL1 high
BS1 Valid before PAGEL High
PAGEL Pulse Width High
BS1 Hold after PAGEL Low
BS2/1 Hold after WR Low
PAGEL Low to WR Low
0
150
67
150
67
67
67
67
150
0
BS1 Valid to WR Low
WR Pulse Width Low
WR Low to RDY/BSY Low
WR Low to RDY/BSY High(1)
WR Low to RDY/BSY High for Chip Erase(2)
XTAL1 Low to OE Low
1
4.5
9
3.7
7.5
0
BS1 Valid to DATA valid
0
250
250
250
OE Low to DATA Valid
OE High to DATA Tri-stated
Notes: 1. tWLRH is valid for the Write Flash, Write EEPROM, Write Fuse Bits and Write Lock
bits commands.
2. tWLRH_CE is valid for the Chip Erase command.
Serial Downloading
Both the Flash and EEPROM memory arrays can be programmed using the serial SPI
bus while RESET is pulled to GND. The serial interface consists of pins SCK, MOSI
(input) and MISO (output). After RESET is set low, the Programming Enable instruction
needs to be executed first before program/erase operations can be executed. NOTE, in
Table 128 on page 307, the pin mapping for SPI programming is listed. Not all parts use
the SPI pins dedicated for the internal SPI interface. Note that throughout the descrip-
tion about Serial downloading, MOSI and MISO are used to describe the serial data in
and serial data out, respectively. For ATmega64, these pins are mapped to PDI and
PDO.
306
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
SPI Serial Programming
Pin Mapping
Even though the SPI Programming interface re-uses the SPI I/O module, there is one
important difference: The MOSI/MISO pins that are mapped to PB2 and PB3 in the SPI
I/O module are not used in the Programming interface. Instead, PE0 and PE1 are used
for data in SPI Programming mode as shown in Table 128.
Table 128. Pin Mapping SPI Serial Programming
Symbol
MOSI (PDI)
MISO (PDO)
SCK
Pins
PE0
PE1
PB1
I/O
Description
Serial Data In
Serial Data Out
Serial Clock
I
O
I
Figure 147. SPI Serial Programming and Verify(1)
+2.7 - 5.5V
VCC
+2.7 - 5.5V (2)
PE0
PE1
PB1
MOSI
MISO
AVCC
SCK
XTAL1
RESET
GND
Notes: 1. If the device is clocked by the internal Oscillator, it is no need to connect a clock
source to the XTAL1 pin.
2. VCC - 0.3 < AVCC < VCC + 0.3, however, AVCC should always be within 2.7 - 5.5V.
When programming the EEPROM, an auto-erase cycle is built into the self-timed pro-
gramming operation (in the Serial mode ONLY) and there is no need to first execute the
Chip Erase instruction. The Chip Erase operation turns the content of every memory
location in both the Program and EEPROM arrays into 0xFF.
Depending on CKSEL Fuses, a valid clock must be present. The minimum low and high
periods for the serial clock (SCK) input are defined as follows:
Low: > 2 CPU clock cycles for fck < 12 MHz, 3 CPU clock cycles for fck ≥ 12 MHz
High: > 2 CPU clock cycles for fck < 12 MHz, 3 CPU clock cycles for fck ≥ 12 MHz
SPI Serial Programming
Algorithm
When writing serial data to the ATmega64, data is clocked on the rising edge of SCK.
When reading data from the ATmega64, data is clocked on the falling edge of SCK. See
Figure 148 for timing details.
To program and verify the ATmega64 in the SPI Serial Programming mode, the follow-
ing sequence is recommended:
1. Power-up sequence:
Apply power between VCC and GND while RESET and SCK are set to “0”. In
some systems, the programmer cannot guarantee that SCK is held low during
Power-up. In this case, RESET must be given a positive pulse of at least two
307
2490G–AVR–03/04
CPU clock cycles duration after SCK has been set to “0”.
As an alternative to using the RESET signal, PEN can be held low during Power-
on Reset while SCK is set to “0”. In this case, only the PEN value at Power-on
Reset is important. If the programmer cannot guarantee that SCK is held low
during Power-up, the PEN method cannot be used. The device must be powered
down in order to commence normal operation when using this method.
2. Wait for at least 20 ms and enable SPI Serial Programming by sending the Pro-
gramming Enable serial instruction to pin MOSI.
3. The SPI Serial Programming instructions will not work if the communication is
out of synchronization. When in sync. the second byte (0x53), will echo back
when issuing the third byte of the Programming Enable instruction. Whether the
echo is correct or not, all four bytes of the instruction must be transmitted. If the
0x53 did not echo back, give RESET a positive pulse and issue a new Program-
ming Enable command.
4. The Flash is programmed one page at a time. The Page size is found in Table
125 on page 296. The memory page is loaded one byte at a time by supplying
the 7 LSB of the address and data together with the Load Program Memory
Page instruction. To ensure correct loading of the page, the data low byte must
be loaded before data high byte is applied for given address. The Program Mem-
ory Page is stored by loading the Write Program Memory Page instruction with
the 8 MSB of the address. If polling is not used, the user must wait at least
t
WD_FLASH before issuing the next page. (See Table 129). Accessing the SPI
Serial Programming interface before the Flash write operation completes can
result in incorrect programming.
5. The EEPROM array is programmed one byte at a time by supplying the address
and data together with the appropriate Write instruction. An EEPROM memory
location is first automatically erased before new data is written. If polling is not
used, the user must wait at least tWD_EEPROM before issuing the next byte. (See
Table 129).
6. Any memory location can be verified by using the Read instruction which returns
the content at the selected address at serial output MISO.
7. At the end of the programming session, RESET can be set high to commence
normal operation.
8. Power-off sequence (if needed):
Set RESET to “1”.
Turn VCC power off.
Note:
If other commands that polling (read) are applied before any write operation (FLASH,
EEPROM, Lock bits, Fuses) is completed, may result in incorrect programming.
Data Polling Flash
When a page is being programmed into the Flash, reading an address location within
the page being programmed will give the value 0xFF. At the time the device is ready for
a new page, the programmed value will read correctly. This is used to determine when
the next page can be written. Note that the entire page is written simultaneously and any
address within the page can be used for polling. Data polling of the Flash will not work
for the value 0xFF, so when programming this value, the user will have to wait for at
least tWD_FLASH before programming the next page. As a chip -erased device contains
0xFF in all locations, programming of addresses that are meant to contain 0xFF, can be
skipped. See Table 129 for tWD_FLASH value.
308
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Data Polling EEPROM
When a new byte has been written and is being programmed into EEPROM, reading the
address location being programmed will give the value 0xFF. At the time the device is
ready for a new byte, the programmed value will read correctly. This is used to deter-
mine when the next byte can be written. This will not work for the value 0xFF, but the
user should have the following in mind: As a chip erased device contains 0xFF in all
locations, programming of addresses that are meant to contain 0xFF, can be skipped.
This does not apply if the EEPROM is re-programmed without chip erasing the device.
In this case, data polling cannot be used for the value 0xFF, and the user will have to
wait at least tWD_EEPROM before programming the next byte. See Table 129 for
tWD_EEPROM value.
Table 129. Minimum Wait Delay before Writing the Next Flash or EEPROM Location
Symbol
tWD_FUSE
tWD_FLASH
Minimum Wait Delay
4.5 ms
(1)
4.5 ms
tWD_EEPROM
tWD_ERASE
Note: 1. Flash write: per page
9.0 ms
9.0 ms
Figure 148. SPI Serial Programming Waveforms
SERIAL DATA INPUT
(MOSI)
MSB
LSB
LSB
SERIAL DATA OUTPUT
(MISO)
MSB
SERIAL CLOCK INPUT
(SCK)
SAMPLE
309
2490G–AVR–03/04
Table 130. SPI Serial Programming Instruction Set
Instruction Format
Instruction
Byte 1
Byte 2
Byte 3
Byte4
Operation
Programming Enable
1010 1100
0101 0011
xxxx xxxx
xxxx xxxx
Enable SPI Serial Programming
after RESET goes low.
Chip Erase
1010 1100
100x xxxx
xxxx xxxx
xxxx xxxx
Chip Erase EEPROM and Flash.
Read Program Memory
0010 H000
xaaa aaaa
bbbb bbbb
oooo oooo
Read H (high or low) data o from
Program memory at word address
a:b.
Load Program Memory
Page
0100 H000
xxxx xxxx
xbbb bbbb
iiii iiii
Write H (high or low) data i to
Program Memory page at word
address b. Data low byte must be
loaded before data high byte is
applied within the same address.
Write Program Memory
Page
0100 1100
1010 0000
1100 0000
0101 1000
xaaa aaaa
xxxx xaaa
xxxx xaaa
0000 0000
bxxx xxxx
bbbb bbbb
bbbb bbbb
xxxx xxxx
xxxx xxxx
oooo oooo
iiii iiii
xxoo oooo
Write Program Memory Page at
address a:b.
Read EEPROM Memory
Write EEPROM Memory
Read Lock Bits
Read data o from EEPROM
memory at address a:b.
Write data i to EEPROM memory
at address a:b.
Read Lock bits. “0” = programmed,
“1” = unprogrammed. See Table
116 on page 290 for details.
Write Lock Bits
1010 1100
111x xxxx
xxxx xxxx
11ii iiii
Write Lock bits. Set bits = “0” to
program Lock bits. See Table 116
on page 290 for details.
Read Signature Byte
Write Fuse Bits
0011 0000
1010 1100
xxxx xxxx
1010 0000
xxxx xxbb
oooo oooo
iiii iiii
Read Signature Byte o at address
b.
xxxx xxxx
Set bits = “0” to program, “1” to
unprogram. See Table 120 on page
292 for details.
Write Fuse High Bits
Write Extended Fuse Bits
Read Fuse Bits
1010 1100
1010 1100
0101 0000
1010 1000
1010 0100
0000 0000
xxxx xxxx
xxxx xxxx
xxxx xxxx
iiii iiii
xxxx xxii
oooo oooo
Set bits = “0” to program, “1” to
unprogram. See Table 119 on page
292 for details.
Set bits = “0” to program, “1” to
unprogram. See Table 120 on page
292 for details.
Read Fuse bits. “0” = programmed,
“1” = unprogrammed. See Table
120 on page 292 for details.
310
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 130. SPI Serial Programming Instruction Set (Continued)
Instruction Format
Instruction
Byte 1
Byte 2
Byte 3
Byte4
Operation
Read Extendend Fuse
Bits
0101 0000
0000 1000
xxxx xxxx
oooo oooo
Read Extended Fuse bits. “0” =
pro-grammed, “1” =
unprogrammed. See Table 120 on
page 292 for details.
Read Fuse High Bits
Read Calibration Byte
0101 1000
0011 1000
0000 1000
00xx xxxx
xxxx xxxx
oooo oooo
oooo oooo
Read Fuse high bits. “0” = pro-
grammed, “1” = unprogrammed.
See Table 119 on page 292 for
details.
0000 00bb
Read Calibration Byte o at address
b.
Note:
a = address high bits, b = address low bits, H = 0 - Low byte, 1 - High Byte, o = data out, i = data in, x = don’t care
SPI Serial Programming
Characteristics
For characteristics of the SPI module, see “SPI Timing Characteristics” on page 331.
311
2490G–AVR–03/04
Programming Via the
JTAG Interface
Programming through the JTAG interface requires control of the four JTAG specific
pins: TCK, TMS, TDI, and TDO. Control of the reset and clock pins is not required.
To be able to use the JTAG interface, the JTAGEN Fuse must be programmed. The
device is default shipped with the fuse programmed. In addition, the JTD bit in MCUCSR
must be cleared. Alternatively, if the JTD bit is set, the External Reset can be forced low.
Then, the JTD bit will be cleared after two chip clocks, and the JTAG pins are available
for programming. This provides a means of using the JTAG pins as normal port pins in
running mode while still allowing In-System Programming via the JTAG interface. Note
that this technique can not be used when using the JTAG pins for Boundary-scan or On-
chip Debug. In these cases the JTAG pins must be dedicated for this purpose.
As a definition in this data sheet, the LSB is shifted in and out first of all Shift Registers.
Programming Specific JTAG
Instructions
The instruction register is 4-bit wide, supporting up to 16 instructions. The JTAG instruc-
tions useful for Programming are listed below.
The OPCODE for each instruction is shown behind the instruction name in hex format.
The text describes which data register is selected as path between TDI and TDO for
each instruction.
The Run-Test/Idle state of the TAP controller is used to generate internal clocks. It can
also be used as an idle state between JTAG sequences. The state machine sequence
for changing the instruction word is shown in Figure 149.
312
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 149. State Machine Sequence for Changing the Instruction Word
1
Test-Logic-Reset
0
1
1
1
0
Run-Test/Idle
Select-DR Scan
Select-IR Scan
0
0
1
1
Capture-DR
Capture-IR
0
0
Shift-DR
0
Shift-IR
0
1
Exit1-DR
0
1
1
1
Exit1-IR
0
Pause-IR
1
Pause-DR
1
0
0
0
0
Exit2-DR
1
Exit2-IR
1
Update-DR
Update-IR
1
1
0
0
AVR_RESET (0xC)
The AVR specific public JTAG instruction for setting the AVR device in the Reset mode
or taking the device out from the Reset mode. The TAP controller is not reset by this
instruction. The one bit Reset Register is selected as Data Register. Note that the reset
will be active as long as there is a logic 'one' in the Reset Chain. The output from this
chain is not latched.
The active states are:
•
Shift-DR: The Reset Register is shifted by the TCK input.
313
2490G–AVR–03/04
PROG_ENABLE (0x4)
The AVR specific public JTAG instruction for enabling programming via the JTAG port.
The 16-bit Programming Enable Register is selected as data register. The active states
are the following:
•
•
Shift-DR: the Programming enable signature is shifted into the data register.
Update-DR: The programming enable signature is compared to the correct value,
and programming mode is entered if the signature is valid.
PROG_COMMANDS (0x5)
The AVR specific public JTAG instruction for entering programming commands via the
JTAG port. The 15-bit Programming Command Register is selected as data register.
The active states are the following:
•
•
Capture-DR: The result of the previous command is loaded into the data register.
Shift-DR: The data register is shifted by the TCK input, shifting out the result of the
previous command and shifting in the new command.
•
•
Update-DR: The programming command is applied to the Flash inputs
Run-Test/Idle: One clock cycle is generated, executing the applied command (not
always required, see Table 131 on page 317).
PROG_PAGELOAD (0x6)
The AVR specific public JTAG instruction to directly load the Flash data page via the
JTAG port. The 1024-bit Virtual Flash Page Load Register is selected as data register.
This is a virtual scan chain with length equal to the number of bits in one Flash page.
Internally the Shift Register is 8-bit. Unlike most JTAG instructions, the Update-DR state
is not used to transfer data from the Shift Register. The data are automatically trans-
ferred to the Flash page buffer byte-by-byte in the Shift-DR state by an internal state
machine. This is the only active state:
•
Shift-DR: Flash page data are shifted in from TDI by the TCK input, and
automatically loaded into the Flash page one byte at a time.
Note:
The JTAG instruction PROG_PAGELOAD can only be used if the AVR device is the first
device in JTAG scan chain. If the AVR cannot be the first device in the scan chain, the
byte-wise programming algorithm must be used.
PROG_PAGEREAD (0x7)
The AVR specific public JTAG instruction to read one full Flash data page via the JTAG
port. The 1032-bit Virtual Flash Page Read Register is selected as data register. This is
a virtual scan chain with length equal to the number of bits in one Flash page plus eight.
Internally the Shift Register is 8-bit. Unlike most JTAG instructions, the Capture-DR
state is not used to transfer data to the Shift Register. The data are automatically trans-
ferred from the Flash page buffer byte-by-byte in the Shift-DR state by an internal state
machine. This is the only active state:
•
Shift-DR: Flash data are automatically read one byte at a time and shifted out on
TDO by the TCK input. The TDI input is ignored.
Note:
The JTAG instruction PROG_PAGEREAD can only be used if the AVR device is the first
device in JTAG scan chain. If the AVR cannot be the first device in the scan chain, the
byte-wise programming algorithm must be used.
314
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Data Registers
The data registers are selected by the JTAG instruction registers described in section
“Programming Specific JTAG Instructions” on page 312. The data registers relevant for
programming operations are:
•
•
•
•
•
Reset Register
Programming Enable Register
Programming Command Register
Virtual Flash Page Load Register
Virtual Flash Page Read Register
Reset Register
The Reset Register is a Test Data Register used to reset the part during programming. It
is required to reset the part before entering programming mode.
A high value in the Reset Register corresponds to pulling the External Reset low. The
part is reset as long as there is a high value present in the Reset Register. Depending
on the Fuse settings for the clock options, the part will remain reset for a Reset Time-out
Period (refer to “Clock Sources” on page 36) after releasing the Reset Register. The
output from this data register is not latched, so the reset will take place immediately, as
shown in Figure 126 on page 256.
Programming Enable Register The Programming Enable Register is a 16-bit register. The contents of this register is
compared to the programming enable signature, binary code 1010_0011_0111_0000.
When the contents of the register is equal to the programming enable signature, pro-
gramming via the JTAG port is enabled. The register is reset to 0 on Power-on Reset,
and should always be reset when leaving Programming mode.
Figure 150. Programming Enable Register
TDI
$A370
D
D
Q
A
T
A
Programming Enable
=
ClockDR & PROG_ENABLE
TDO
315
2490G–AVR–03/04
Programming Command
Register
The Programming Command Register is a 15-bit register. This register is used to seri-
ally shift in programming commands, and to serially shift out the result of the previous
command, if any. The JTAG Programming Instruction Set is shown in Table 131. The
state sequence when shifting in the programming commands is illustrated in Figure 152.
Figure 151. Programming Command Register
TDI
S
T
R
O
B
E
S
Flash
A
EEPROM
Fuses
Lock Bits
D
D
R
E
S
S
/
D
A
T
A
TDO
316
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 131. JTAG Programming Instruction Set
a = address high bits, b = address low bits, H = 0 - Low byte, 1 - High Byte, o = data out, i = data in, x = don’t care
Instruction
TDI sequence
TDO sequence
Notes
1a. Chip Erase
0100011_10000000
0110001_10000000
0110011_10000000
0110011_10000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
1b. Poll for Chip Erase Complete
2a. Enter Flash Write
0110011_10000000
0100011_00010000
0000111_aaaaaaaa
0000011_bbbbbbbb
0010011_iiiiiiii
xxxxxox_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(2)
(9)
2b. Load Address High Byte
2c. Load Address Low Byte
2d. Load Data Low Byte
2e. Load Data High Byte
2f. Latch Data
0010111_iiiiiiii
0110111_00000000
1110111_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
(1)
2g. Write Flash Page
0110111_00000000
0110101_00000000
0110111_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
2h. Poll for Page Write Complete
3a. Enter Flash Read
0110111_00000000
0100011_00000010
0000111_aaaaaaaa
0000011_bbbbbbbb
xxxxxox_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(2)
(9)
3b. Load Address High Byte
3c. Load Address Low Byte
3d. Read Data Low and High Byte
0110010_00000000
0110110_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
xxxxxxx_oooooooo
low byte
high byte
4a. Enter EEPROM Write
4b. Load Address High Byte
4c. Load Address Low Byte
4d. Load Data Byte
0100011_00010001
0000111_aaaaaaaa
0000011_bbbbbbbb
0010011_iiiiiiii
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(9)
4e. Latch Data
0110111_00000000
1110111_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
(1)
4f. Write EEPROM Page
0110011_00000000
0110001_00000000
0110011_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
4g. Poll for Page Write Complete
5a. Enter EEPROM Read
0110011_00000000
0100011_00000011
0000111_aaaaaaaa
0000011_bbbbbbbb
xxxxxox_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(2)
(9)
5b. Load Address High Byte
5c. Load Address Low Byte
317
2490G–AVR–03/04
Table 131. JTAG Programming Instruction Set (Continued)
a = address high bits, b = address low bits, H = 0 - Low byte, 1 - High Byte, o = data out, i = data in, x = don’t care
Instruction
TDI sequence
TDO sequence
Notes
5d. Read Data Byte
0110011_bbbbbbbb
0110010_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
6a. Enter Fuse Write
0100011_01000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
6b. Load Data Low Byte(6)
0010011_iiiiiiii
(3)
(1)
6c. Write Fuse Extended Byte
0111011_00000000
0111001_00000000
0111011_00000000
0111011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
6d. Poll for Fuse Write Complete
6e. Load Data Low Byte(7)
6f. Write Fuse High Byte
0111011_00000000
xxxxxox_xxxxxxxx
(2)
(3)
(1)
0010011_iiiiiiii
xxxxxxx_xxxxxxxx
0110111_00000000
0110101_00000000
0110111_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
6g. Poll for Fuse Write Complete
6h. Load Data Low Byte(8)
6i. Write Fuse Low byte
0110111_00000000
xxxxxox_xxxxxxxx
(2)
(3)
(1)
0010011_iiiiiiii
xxxxxxx_xxxxxxxx
0110011_00000000
0110001_00000000
0110011_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
6j. Poll for Fuse Write Complete
7a. Enter Lock Bit Write
7b. Load Data Byte(9)
0110011_00000000
0100011_00100000
0010011_11iiiiii
xxxxxox_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(2)
(4)
(1)
7c. Write Lock Bits
0110011_00000000
0110001_00000000
0110011_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
7d. Poll for Lock Bit Write Complete
8a. Enter Fuse/Lock Bit Read
8b. Read Fuse Extended Byte(6)
0110011_00000000
0100011_00000100
xxxxxox_xxxxxxxx
(2)
xxxxxxx_xxxxxxxx
0111010_00000000
0111111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
8c. Read Fuse High Byte(7)
8d. Read Fuse Low Byte(8)
8e. Read Lock Bits(9)
0111110_00000000
0111111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
0110010_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
0110110_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
(5)
xxxxxxx_xxoooooo
318
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 131. JTAG Programming Instruction Set (Continued)
a = address high bits, b = address low bits, H = 0 - Low byte, 1 - High Byte, o = data out, i = data in, x = don’t care
Instruction
TDI sequence
TDO sequence
Notes
8f. Read Fuses and Lock Bits
0111010_00000000
0111110_00000000
0110010_00000000
0110110_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
xxxxxxx_oooooooo
xxxxxxx_oooooooo
xxxxxxx_oooooooo
(5)
Fuse Ext. byte
Fuse High byte
Fuse Low byte
Lock bits
9a. Enter Signature Byte Read
9b. Load Address Byte
0100011_00001000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
0000011_bbbbbbbb
9c. Read Signature Byte
0110010_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
10a. Enter Calibration Byte Read
10b. Load Address Byte
0100011_00001000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
0000011_bbbbbbbb
10c. Read Calibration Byte
0110110_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
Notes: 1. This command sequence is not required if the seven MSB are correctly set by the previous command sequence (which is
normally the case).
2. Repeat until o = “1”.
3. Set bits to “0” to program the corresponding fuse, “1” to unprogram the fuse.
4. Set bits to “0” to program the corresponding Lock bit, “1” to leave the Lock bit unchanged.
5. “0” = programmed, “1” = unprogrammed.
6. The bit mapping for Fuses Extended byte is listed in Table 118 on page 291.
7. The bit mapping for Fuses High byte is listed in Table 119 on page 292.
8. The bit mapping for Fuses Low byte is listed in Table 120 on page 292.
9. The bit mapping for Lock bits byte is listed in Table 116 on page 290.
10. Address bits exceeding PCMSB and EEAMSB (Table 124 and Table 125) are don’t care.
319
2490G–AVR–03/04
Figure 152. State Machine Sequence for Changing/Reading the Data Word
1
Test-Logic-Reset
0
1
1
1
0
Run-Test/Idle
Select-DR Scan
Select-IR Scan
0
0
1
1
Capture-DR
Capture-IR
0
0
Shift-IR
1
Shift-DR
0
0
1
Exit1-DR
0
1
1
Exit1-IR
0
Pause-DR
1
0
Pause-IR
1
0
0
0
Exit2-DR
1
Exit2-IR
1
Update-DR
Update-IR
1
1
0
0
320
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Virtual Flash Page Load
Register
The Virtual Flash Page Load Register is a virtual scan chain with length equal to the
number of bits in one Flash page. Internally the Shift Register is 8-bit, and the data are
automatically transferred to the Flash page buffer byte-by-byte. Shift in all instruction
words in the page, starting with the LSB of the first instruction in the page and ending
with the MSB of the last instruction in the page. This provides an efficient way to load the
entire Flash page buffer before executing Page Write.
Figure 153. Virtual Flash Page Load Register
STROBES
State
Machine
ADDRESS
Flash
TDI
EEPROM
Fuses
Lock Bits
D
A
T
A
TDO
Virtual Flash Page Read
Register
The Virtual Flash Page Read Register is a virtual scan chain with length equal to the
number of bits in one Flash page plus eight. Internally the Shift Register is 8-bit, and the
data are automatically transferred from the Flash data page byte-by-byte. The first eight
cycles are used to transfer the first byte to the internal Shift Register, and the bits that
are shifted out during these eight cycles should be ignored. Following this initialization,
data are shifted out starting with the LSB of the first instruction in the page and ending
with the MSB of the last instruction in the page. This provides an efficient way to read
one full Flash page to verify programming.
321
2490G–AVR–03/04
Figure 154. Virtual Flash Page Read Register
STROBES
State
Machine
ADDRESS
Flash
EEPROM
Fuses
TDI
Lock Bits
D
A
T
A
TDO
Programming Algorithm
All references below of type “1a”, “1b”, and so on, refer to Table 131.
1. Enter JTAG instruction AVR_RESET and shift 1 in the Reset Register.
Entering Programming Mode
2. Enter instruction PROG_ENABLE and shift 1010_0011_0111_0000 in the Pro-
gramming Enable Register.
Leaving Programming Mode
Performing Chip Erase
1. Enter JTAG instruction PROG_COMMANDS.
2. Disable all programming instructions by using no operation instruction 11a.
3. Enter instruction PROG_ENABLE and shift 0000_0000_0000_0000 in the Pro-
gramming Enable Register.
4. Enter JTAG instruction AVR_RESET and shift 0 in the Reset Register.
1. Enter JTAG instruction PROG_COMMANDS.
2. Start chip erase using programming instruction 1a.
3. Poll for chip erase complete using programming instruction 1b, or wait for
t
WLRH_CE (refer to Table 1 on page 305).
322
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Programming the Flash
Before programming the Flash, a Chip Erase must be performed. See “Performing Chip
Erase” on page 322.
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Flash write using programming instruction 2a.
3. Load address high byte using programming instruction 2b.
4. Load address low byte using programming instruction 2c.
5. Load data using programming instructions 2d, 2e and 2f.
6. Repeat steps 4 and 5 for all instruction words in the page.
7. Write the page using programming instruction 2g.
8. Poll for Flash write complete using programming instruction 2h, or wait for
t
WLRH_FLASH (refer to Table 1 on page 305).
9. Repeat steps 3 to 7 until all data have been programmed.
A more efficient data transfer can be achieved using the PROG_PAGELOAD
instruction:
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Flash write using programming instruction 2a.
3. Load the page address using programming instructions 2b and 2c. PCWORD
(refer to Table 124 on page 296) is used to address within one page and must be
written as 0.
4. Enter JTAG instruction PROG_PAGELOAD.
5. Load the entire page by shifting in all instruction words in the page, starting with
the LSB of the first instruction in the page and ending with the MSB of the last
instruction in the page.
6. Enter JTAG instruction PROG_COMMANDS.
7. Write the page using programming instruction 2g.
8. Poll for Flash write complete using programming instruction 2h, or wait for
t
WLRH_FLASH (refer to Table 1 on page 305).
9. Repeat steps 3 to 8 until all data have been programmed.
323
2490G–AVR–03/04
Reading the Flash
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Flash read using programming instruction 3a.
3. Load address using programming instructions 3b and 3c.
4. Read data using programming instruction 3d.
5. Repeat steps 3 and 4 until all data have been read.
A more efficient data transfer can be achieved using the PROG_PAGEREAD
instruction:
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Flash read using programming instruction 3a.
3. Load the page address using programming instructions 3b and 3c. PCWORD
(refer to Table 124 on page 296) is used to address within one page and must be
written as 0.
4. Enter JTAG instruction PROG_PAGEREAD.
5. Read the entire page by shifting out all instruction words in the page, starting
with the LSB of the first instruction in the page and ending with the MSB of the
last instruction in the page. Remember that the first eight bits shifted out should
be ignored.
6. Enter JTAG instruction PROG_COMMANDS.
7. Repeat steps 3 to 6 until all data have been read.
Programming the EEPROM
Before programming the EEPROM, a Chip Erase must be performed. See “Performing
Chip Erase” on page 322.
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable EEPROM write using programming instruction 4a.
3. Load address high byte using programming instruction 4b.
4. Load address low byte using programming instruction 4c.
5. Load data using programming instructions 4d and 4e.
6. Repeat steps 4 and 5 for all data bytes in the page.
7. Write the data using programming instruction 4f.
8. Poll for EEPROM write complete using programming instruction 4g, or wait for
t
WLRH (refer to Table 1 on page 305).
9. Repeat steps 3 to 8 until all data have been programmed.
Note that the PROG_PAGELOAD instruction can not be used when programming the
EEPROM.
Reading the EEPROM
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable EEPROM read using programming instruction 5a.
3. Load address using programming instructions 5b and 5c.
4. Read data using programming instruction 5d.
5. Repeat steps 3 and 4 until all data have been read.
Note that the PROG_PAGEREAD instruction can not be used when reading the
EEPROM
324
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Programming the Fuses
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Fuse write using programming instruction 6a.
3. Load data Low byte using programming instructions 6b. A bit value of “0” will pro-
gram the corresponding fuse, a “1” will unprogram the fuse.
4. Write Fuse Extended byte using programming instruction 6c.
5. Poll for Fuse write complete using programming instruction 6d, or wait for tWLRH
(refer to Table 1 on page 305).
6. Load data Low byte using programming instructions 6e. A bit value of “0” will pro-
gram the corresponding fuse, a “1” will unprogram the fuse.
7. Write Fuse High byte using programming instruction 6f.
8. Poll for Fuse write complete using programming instruction 6g, or wait for tWLRH
(refer to Table 1 on page 305).
9. Load data low byte using programming instructions 6h. A “0” will program the
fuse, a “1” will unprogram the fuse.
10. Write Fuse low byte using programming instruction 6i.
11. Poll for Fuse write complete using programming instruction 6j, or wait for tWLRH
(refer to Table 1 on page 305).
Programming the Lock Bits
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Lock bit write using programming instruction 7a.
3. Load data using programming instructions 7b. A bit value of “0” will program the
corresponding Lock bit, a “1” will leave the Lock bit unchanged.
4. Write Lock bits using programming instruction 7c.
5. Poll for Lock bit write complete using programming instruction 7d, or wait for
t
WLRH (refer to Table 1 on page 305).
Reading the Fuses and Lock
Bits
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Fuse/Lock bit read using programming instruction 8a.
3. To read all Fuses and Lock bits, use programming instruction 8f.
To only read Fuse Extended byte, use programming instruction 8b.
To only read Fuse High byte, use programming instruction 8c.
To only read Fuse Low byte, use programming instruction 8d.
To only read Lock bits, use programming instruction 8e.
Reading the Signature Bytes
Reading the Calibration Byte
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Signature byte read using programming instruction 9a.
3. Load address 0x00 using programming instruction 9b.
4. Read first signature byte using programming instruction 9c.
5. Repeat steps 3 and 4 with address 0x01 and address 0x02 to read the second
and third signature bytes, respectively.
1. Enter JTAG instruction PROG_COMMANDS.
2. Enable Calibration byte read using programming instruction 10a.
3. Load address 0x00 using programming instruction 10b.
4. Read the calibration byte using programming instruction 10c.
325
2490G–AVR–03/04
Electrical Characteristics
Absolute Maximum Ratings*
*NOTICE:
Stresses beyond those listed under “Absolute
Maximum Ratings” may cause permanent dam-
age to the device. This is a stress rating only and
functional operation of the device at these or
other conditions beyond those indicated in the
operational sections of this specification is not
implied. Exposure to absolute maximum rating
conditions for extended periods may affect device
reliability.
Operating Temperature.................................. -55°C to +125°C
Storage Temperature..................................... -65°C to +150°C
Voltage on any Pin except RESET
with respect to Ground ................................-1.0V to VCC+0.5V
Voltage on RESET with respect to Ground......-1.0V to +13.0V
Maximum Operating Voltage ............................................ 6.0V
DC Current per I/O Pin ............................................... 40.0 mA
DC Current VCC and GND Pins................................ 200.0 mA
DC Characteristics
TA = -40°C to 85°C, VCC = 2.7V to 5.5V (unless otherwise noted)
Symbol
Parameter
Condition
Min
Typ
Max
Units
Except XTAL1 and
RESET pins
(1)
VIL
Input Low Voltage
-0.5
0.2 VCC
V
XTAL1 pin, External
Clock Selected
(1)
(1)
VIL1
VIL2
VIH
Input Low Voltage
Input Low Voltage
Input High Voltage
-0.5
-0.5
0.1 VCC
0.2 VCC
V
V
V
RESET pin
Except XTAL1 and
RESET pins
(2)
0.6 VCC
0.7 VCC
VCC + 0.5
XTAL1 pin, External
Clock Selected
(2)
VIH1
VIH2
VOL
Input High Voltage
Input High Voltage
VCC + 0.5
VCC + 0.5
V
V
(2)
RESET pin
0.85 VCC
Output Low Voltage(3)
(Ports A,B,C,D, E, F, G)
I
OL = 20 mA, VCC = 5V
0.7
0.5
V
V
IOL = 10 mA, VCC = 3V
Output High Voltage(4)
(Ports A,B,C,D)
I
OH = -20 mA, VCC = 5V
4.0
2.2
V
V
VOH
IIL
IOH = -10 mA, VCC = 3V
Input Leakage
Current I/O Pin
Vcc = 5.5V, pin low
(absolute value)
8.0
8.0
µA
µA
Input Leakage
Current I/O Pin
Vcc = 5.5V, pin high
(absolute value)
IIH
RRST
RPEN
RPU
Reset Pull-up Resistor
PEN Pull-up Resistor
I/O Pin Pull-up Resistor
30
25
20
100
100
100
kΩ
kΩ
kΩ
326
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
DC Characteristics
TA = -40°C to 85°C, VCC = 2.7V to 5.5V (unless otherwise noted) (Continued)
Symbol
Parameter
Condition
Min
Typ
Max
Units
Active 4 MHz, VCC = 3V
(ATmega64L)
5
mA
Active 8 MHz, VCC = 5V
(ATmega64)
20
2
mA
mA
mA
Power Supply Current
Idle 4 MHz, VCC = 3V
(ATmega64L)
ICC
Idle 8 MHz, VCC = 5V
(ATmega64)
12
WDT enabled, VCC = 3V
WDT disabled, VCC = 3V
< 25
< 10
40
25
µA
µA
Power-down mode(5)
Analog Comparator
Input Offset Voltage
VCC = 5V
Vin = VCC/2
VACIO
IACLK
40
50
mV
nA
Analog Comparator
Input Leakage Current
VCC = 5V
Vin = VCC/2
-50
Notes: 1. “Max” means the highest value where the pin is guaranteed to be read as low
2. “Min” means the lowest value where the pin is guaranteed to be read as high
3. Although each I/O port can sink more than the test conditions (20 mA at VCC = 5V, 10 mA at VCC = 3V) under steady state
conditions (non-transient), the following must be observed:
TQFP and MLF Package:
1] The sum of all IOL, for all ports, should not exceed 400 mA.
2] The sum of all IOL, for ports A0 - A7, G2, C3 - C7 should not exceed 300 mA.
3] The sum of all IOL, for ports C0 - C2, G0 - G1, D0 - D7, XTAL2 should not exceed 150 mA.
4] The sum of all IOL, for ports B0 - B7, G3 - G4, E0 - E7 should not exceed 150 mA.
5] The sum of all IOL, for ports F0 - F7, should not exceed 200 mA.
If IOL exceeds the test condition, VOL may exceed the related specification. Pins are not guaranteed to sink current greater
than the listed test condition.
4. Although each I/O port can source more than the test conditions (20 mA at Vcc = 5V, 10 mA at Vcc = 3V) under steady state
conditions (non-transient), the following must be observed:
TQFP and MLF Package:
1] The sum of all IOH, for all ports, should not exceed 400 mA.
2] The sum of all IOH, for ports A0 - A7, G2, C3 - C7 should not exceed 300 mA.
3] The sum of all IOH, for ports C0 - C2, G0 - G1, D0 - D7, XTAL2 should not exceed 150 mA.
4] The sum of all IOH, for ports B0 - B7, G3 - G4, E0 - E7 should not exceed 150 mA.
5] The sum of all IOH, for ports F0 - F7, should not exceed 200 mA.
If IOH exceeds the test condition, VOH may exceed the related specification. Pins are not guaranteed to source current
greater than the listed test condition.
5. Minimum VCC for Power-down is 2.5V.
327
2490G–AVR–03/04
External Clock Drive
Waveforms
Figure 155. External Clock Drive Waveforms
V
IH1
V
IL1
External Clock Drive
Table 132. External Clock Drive(1)
VCC = 2.7V to 5.5V
VCC = 4.5V to 5.5V
Symbol
1/tCLCL
tCLCL
Parameter
Oscillator Frequency
Clock Period
High Time
Min
0
Max
Min
0
Max
Units
MHz
ns
8
16
125
50
62.5
25
tCHCX
tCLCX
ns
Low Time
50
25
ns
tCLCH
Rise Time
1.6
1.6
0.5
0.5
µs
tCHCL
Fall Time
µs
Change in period from
one clock cycle to the
next
∆tCLCL
2
2
%
Note:
1. Refer to “External Clock” on page 41 for details.
Table 133. External RC Oscillator, Typical Frequencies
R [kΩ](1)
100
C [pF]
70
f
TBD
TBD
TBD
31.5
20
6.5
20
Note:
1. R should be in the range 3kΩ - 100kΩ, and C should be at least 20 pF. The C values
given in the table includes pin capacitance. This will vary with package type.
328
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Two-wire Serial Interface Characteristics
Table 134 describes the requirements for devices connected to the Two-wire Serial Bus. The ATmega64 Two-wire Serial
Interface meets or exceeds these requirements under the noted conditions.
Timing symbols refer to Figure 156.
Table 134. Two-wire Serial Bus Requirements
Symbol Parameter
Condition
Min
-0.5
Max
0.3 VCC
VCC + 0.5
–
Units
V
VIL
Input Low-voltage
VIH
Vhys
Input High-voltage
0.7 VCC
V
(1)
(1)
(2)
Hysteresis of Schmitt Trigger Inputs
Output Low-voltage
0.05 VCC
V
VOL
tr(1)
3 mA sink current
10 pF < Cb < 400 pF(3)
0.1VCC < Vi < 0.9VCC
0
0.4
V
(3)(2)
(3)(2)
Rise Time for both SDA and SCL
Output Fall Time from VIHmin to VILmax
Spikes Suppressed by Input Filter
Input Current each I/O Pin
Capacitance for each I/O Pin
SCL Clock Frequency
20 + 0.1Cb
300
ns
ns
ns
µA
pF
kHz
(1)
tof
20 + 0.1Cb
250
tSP
Ii
Ci(1)
fSCL
0
-10
–
50(2)
(1)
10
10
fCK(4) > max(16fSCL, 250 kHz)(5)
0
400
V
CC – 0,4V
----------------------------
3mA
f
SCL ≤ 100 kHz
1000ns
Cb
-------------------
Ω
Ω
Rp
Value of Pull-up resistor
V
CC – 0,4V
3mA
f
SCL > 100 kHz
300ns
----------------------------
---------------
Cb
f
SCL ≤ 100 kHz
4.0
0.6
4.7
1.3
4.0
0.6
4.7
0.6
0
–
µs
µs
µs
µs
µs
µs
µs
µs
µs
µs
ns
ns
µs
µs
tHD;STA
Hold Time (repeated) START Condition
Low Period of the SCL Clock
High period of the SCL clock
Set-up time for a repeated START condition
Data hold time
fSCL > 100 kHz
–
f
SCL ≤ 100 kHz(6)
–
tLOW
f
SCL > 100 kHz(7)
SCL ≤ 100 kHz
fSCL > 100 kHz
SCL ≤ 100 kHz
fSCL > 100 kHz
SCL ≤ 100 kHz
SCL > 100 kHz
SCL ≤ 100 kHz
fSCL > 100 kHz
SCL ≤ 100 kHz
fSCL > 100 kHz
–
f
–
tHIGH
–
f
–
tSU;STA
tHD;DAT
tSU;DAT
–
f
3.45
0.9
–
f
0
f
250
100
4.0
0.6
Data setup time
–
f
–
tSU;STO
Setup time for STOP condition
–
Bus free time between a STOP and START
condition
tBUF
fSCL ≤ 100 kHz
4.7
–
µs
Notes: 1. In ATmega64, this parameter is characterized and not 100% tested.
2. Required only for fSCL > 100 kHz.
3. Cb = capacitance of one bus line in pF.
4. fCK = CPU clock frequency
329
2490G–AVR–03/04
5. This requirement applies to all ATmega64 Two-wire Serial Interface operation. Other devices connected to the Two-wire
Serial Bus need only obey the general fSCL requirement.
6. The actual low period generated by the ATmega64 Two-wire Serial Interface is (1/fSCL - 2/fCK), thus fCK must be greater than
6 MHz for the low time requirement to be strictly met at fSCL = 100 kHz.
7. The actual low period generated by the ATmega64 Two-wire Serial Interface is (1/fSCL - 2/fCK), thus the low time requirement
will not be strictly met for fSCL > 308 kHz when fCK = 8 MHz. Still, ATmega64 devices connected to the bus may communicate
at full speed (400 kHz) with other ATmega64 devices, as well as any other device with a proper tLOW acceptance margin.
Figure 156. Two-wire Serial Bus Timing
t
HIGH
t
t
r
of
t
t
LOW
LOW
SCL
SDA
t
t
t
HD;DAT
SU;STA
HD;STA
t
SU;DAT
t
SU;STO
t
BUF
330
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
SPI Timing
Characteristics
Table 135. SPI Timing Parameters
Description Mode
SCK period
Min
Typ
Max
1
2
Master
Master
Master
Master
Master
Master
Master
Master
Slave
See Table 72
SCK high/low
Rise/Fall time
Setup
50% duty cycle
3
TBD
10
4
5
Hold
10
6
Out to SCK
SCK to out
SCK to out high
SS low to out
SCK period
SCK high/low(1)
Rise/Fall time
Setup
0.5 • tsck
10
7
8
10
9
15
ns
10
11
12
13
14
15
16
17
18
Slave
4 • tck
2 • tck
Slave
Slave
TBD
Slave
10
tck
Hold
Slave
SCK to out
SCK to SS high
SS high to tri-state
SS low to SCK
Slave
15
10
Slave
20
Slave
Slave
20
Note:
1. In SPI Programming mode the minimum SCK high/low period is:
- 2 tCLCL for fCK < 12 MHz
- 3 tCLCL for fCK >12 MHz
Figure 157. SPI Interface Timing Requirements (Master Mode)
SS
6
1
SCK
(CPOL = 0)
2
2
SCK
(CPOL = 1)
4
5
3
MISO
(Data Input)
MSB
...
LSB
7
8
MOSI
(Data Output)
MSB
...
LSB
331
2490G–AVR–03/04
Figure 158. SPI Interface Timing Requirements (Slave Mode)
18
SS
10
16
9
SCK
(CPOL = 0)
11
11
SCK
(CPOL = 1)
13
14
12
MOSI
(Data Input)
MSB
...
LSB
15
17
MISO
(Data Output)
MSB
...
LSB
X
332
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
ADC Characteristics – Preliminary Data
Table 136. ADC Characteristics, Single Ended Channels, -40°C – 85°C
Symbol Parameter
Condition
Min(1)
Typ(1)
Max(1)
Units
Resolution
Single Ended Conversion
10
Bits
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
1.5
3
LSB
LSB
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 1 MHz
Absolute Accuracy
Single Ended Conversion
(Including INL, DNL, Quantization Error, Gain
and Offset Error)
VREF = 4V, VCC = 4V
1.5
3
LSB
LSB
ADC clock = 200 kHz
Noise Reduction mode
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 1 MHz
Noise Reduction mode
Single Ended Conversion
Integral Non-Linearity (INL)
Differential Non-Linearity (DNL)
Gain Error
V
REF = 4V, VCC = 4V
0.75
0.25
0.75
0.75
LSB
LSB
LSB
LSB
ADC clock = 200 kHz
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
Single Ended Conversion
Offset error
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
Clock Frequency
50
13
1000
260
kHz
µs
Conversion Time
AVCC
VREF
VIN
Analog Supply Voltage
Reference Voltage
V
CC–0.3(2)
VCC + 0.3(3)
AVCC – 0.5
VREF
V
2.0
V
Input Voltage
GND
0
V
ADC Conversion Output
Input Bandwidth
1023
LSB
kHz
V
38.5
2.56
32
VINT
RREF
RAIN
Internal Voltage Reference
Reference Input Resistance
Analog Input Resistance
2.3
2.7
kΩ
MΩ
100
Note:
1. Values are guidelines only. Actual values are TBD.
2. Minimum for AVCC is 2.7V.
3. Maximum for AVCC is 5.5V.
333
2490G–AVR–03/04
Table 137. ADC Characteristics, Differential Channels, -40°C – 85°C
Symbol Parameter
Condition
Gain = 1x
Gain = 10x
Gain = 200x
Min(1)
Typ(1)
Max(1)
10
Units
Bits
Resolution
10
Bits
10
Bits
Gain = 1x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
16
16
LSB
LSB
LSB
LSB
LSB
LSB
Gain = 10x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
Absolute Accuracy
Gain = 200x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
8
Gain = 1x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
0.75
0.75
2.5
Integral Non-Linearity (INL)
Gain = 10x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
(Accuracy after Calibration for Offset and
Gain Error)
Gain = 200x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
Gain = 1x
Gain = 10x
Gain = 200x
1.6
1.6
0.3
%
%
%
Gain Error
Gain = 1x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
1.5
1
LSB
LSB
LSB
Gain = 10x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
Offset Error
Gain = 200x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
6
Clock Frequency
50
65
200
260
kHz
µs
V
Conversion Time
AVCC
VREF
VIN
Analog Supply Voltage
Reference Voltage
Input Voltage
VCC–0.3(2)
VCC + 0.3(3)
AVCC – 0.5
VCC
2.0
V
GND
V
VDIFF
Input Differential Voltage
ADC Conversion Output
Input Bandwidth
-VREF/Gain
-511
VREF/Gain
511
V
LSB
kHz
4
334
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 137. ADC Characteristics, Differential Channels, -40°C – 85°C (Continued)
Symbol Parameter Condition
Min(1)
Typ(1)
2.56
32
Max(1)
Units
V
VINT
RREF
RAIN
Internal Voltage Reference
2.3
2.7
Reference Input Resistance
Analog Input Resistance
kΩ
100
MΩ
Notes: 1. Values are guidelines only. Actual values are TBD.
2. Minimum for AVCC is 2.7V.
3. Maximum for AVCC is 5.5V.
335
2490G–AVR–03/04
External Data Memory Timing
Table 138. External Data Memory Characteristics, 4.5 - 5.5 Volts, No Wait-state
8 MHz Oscillator
Variable Oscillator
Symbol
1/tCLCL
tLHLL
Parameter
Min
Max
Min
Max
Unit
MHz
ns
0
1
2
Oscillator Frequency
ALE Pulse Width
0.0
16
115
1.0tCLCL-10
0.5tCLCL-5(1)
tAVLL
Address Valid A to ALE Low
57.5
ns
Address Hold After ALE Low,
write access
5
5
5
3a
tLLAX_ST
ns
Address Hold after ALE Low,
read access
5
3b
4
tLLAX_LD
tAVLLC
tAVRL
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
Address Valid C to ALE Low
Address Valid to RD Low
Address Valid to WR Low
ALE Low to WR Low
ALE Low to RD Low
57.5
115
115
47.5
47.5
40
0.5tCLCL-5(1)
1.0tCLCL-10
1.0tCLCL-10
0.5tCLCL-15(2)
0.5tCLCL-15(2)
40
5
6
tAVWL
tLLWL
7
67.5
67.5
0.5tCLCL+5(2)
0.5tCLCL+5(2)
8
tLLRL
9
tDVRH
tRLDV
tRHDX
tRLRH
tDVWL
tWHDX
tDVWH
tWLWH
Data Setup to RD High
Read Low to Data Valid
Data Hold After RD High
RD Pulse Width
10
11
12
13
14
15
16
75
1.0tCLCL-50
0
0
115
42.5
115
125
115
1.0tCLCL-10
0.5tCLCL-20(1)
1.0tCLCL-10
1.0tCLCL
Data Setup to WR Low
Data Hold After WR High
Data Valid to WR High
WR Pulse Width
1.0tCLCL-10
Notes: 1. This assumes 50% clock duty cycle. The half period is actually the high time of the external clock, XTAL1.
2. This assumes 50% clock duty cycle. The half period is actually the low time of the external clock, XTAL1.
Table 139. External Data Memory Characteristics, 4.5 - 5.5 Volts, 1 Cycle Wait-state
8 MHz Oscillator
Variable Oscillator
Min Max
Symbol
Parameter
Min
Max
Unit
MHz
ns
0
1/tCLCL
Oscillator Frequency
Read Low to Data Valid
RD Pulse Width
0.0
16
10 tRLDV
12 tRLRH
15 tDVWH
16 tWLWH
200
2.0tCLCL-50
240
240
240
2.0tCLCL-10
2.0tCLCL
ns
Data Valid to WR High
WR Pulse Width
ns
2.0tCLCL-10
ns
336
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table 140. External Data Memory Characteristics, 4.5 - 5.5 Volts, SRWn1 = 1, SRWn0 = 0
4 MHz Oscillator Variable Oscillator
Symbol
Parameter
Min
Max
Min
Max
16
Unit
MHz
ns
0
1/tCLCL
Oscillator Frequency
Read Low to Data Valid
RD Pulse Width
0.0
10 tRLDV
12 tRLRH
15 tDVWH
16 tWLWH
325
3.0tCLCL-50
365
375
365
3.0tCLCL-10
3.0tCLCL
ns
Data Valid to WR High
WR Pulse Width
ns
3.0tCLCL-10
ns
Table 141. External Data Memory Characteristics, 4.5 - 5.5 Volts, SRWn1 = 1, SRWn0 = 1
4 MHz Oscillator Variable Oscillator
Symbol
Parameter
Min
Max
Min
Max
16
Unit
MHz
ns
0
1/tCLCL
Oscillator Frequency
Read Low to Data Valid
RD Pulse Width
0.0
10 tRLDV
12 tRLRH
14 tWHDX
15 tDVWH
16 tWLWH
325
3.0tCLCL-50
365
240
375
365
3.0tCLCL-10
2.0tCLCL-10
3.0tCLCL
ns
Data Hold After WR High
Data Valid to WR High
WR Pulse Width
ns
ns
3.0tCLCL-10
ns
Table 142. External Data Memory Characteristics, 2.7 - 5.5 Volts, No Wait-state
4 MHz Oscillator
Variable Oscillator
Symbol
1/tCLCL
tLHLL
Parameter
Min
Max
Min
0.0
Max
Unit
MHz
ns
0
1
2
Oscillator Frequency
ALE Pulse Width
8
235
115
tCLCL-15
0.5tCLCL-10(1)
tAVLL
Address Valid A to ALE Low
ns
Address Hold After ALE Low,
write access
5
5
5
3a tLLAX_ST
ns
Address Hold after ALE Low,
read access
5
3b tLLAX_LD
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
4
5
6
7
8
9
tAVLLC
tAVRL
tAVWL
tLLWL
tLLRL
Address Valid C to ALE Low
Address Valid to RD Low
Address Valid to WR Low
ALE Low to WR Low
115
235
235
115
115
45
0.5tCLCL-10(1)
1.0tCLCL-15
1.0tCLCL-15
0.5tCLCL-10(2)
0.5tCLCL-10(2)
45
130
130
0.5tCLCL+5(2)
0.5tCLCL+5(2)
ALE Low to RD Low
tDVRH
Data Setup to RD High
Read Low to Data Valid
Data Hold After RD High
RD Pulse Width
10 tRLDV
11 tRHDX
12 tRLRH
190
1.0tCLCL-60
0
0
235
1.0tCLCL-15
337
2490G–AVR–03/04
Table 142. External Data Memory Characteristics, 2.7 - 5.5 Volts, No Wait-state (Continued)
4 MHz Oscillator Variable Oscillator
Min Max
Symbol
13 tDVWL
Parameter
Min
Max
Unit
ns
Data Setup to WR Low
Data Hold After WR High
Data Valid to WR High
WR Pulse Width
105
235
250
235
0.5tCLCL-20(1)
1.0tCLCL-15
1.0tCLCL
14 tWHDX
15 tDVWH
16 tWLWH
ns
ns
1.0tCLCL-15
ns
Notes: 1. This assumes 50% clock duty cycle. The half period is actually the high time of the external clock, XTAL1.
2. This assumes 50% clock duty cycle. The half period is actually the low time of the external clock, XTAL1.
Table 143. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 0, SRWn0 = 1
4 MHz Oscillator
Variable Oscillator
Min Max
Symbol
Parameter
Min
Max
Unit
MHz
ns
0
1/tCLCL
Oscillator Frequency
Read Low to Data Valid
RD Pulse Width
0.0
8
10 tRLDV
12 tRLRH
15 tDVWH
16 tWLWH
440
2.0tCLCL-60
485
500
485
2.0tCLCL-15
2.0tCLCL
ns
Data Valid to WR High
WR Pulse Width
ns
2.0tCLCL-15
ns
Table 144. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 1, SRWn0 = 0
4 MHz Oscillator Variable Oscillator
Symbol
Parameter
Min
Max
Min
Max
8
Unit
MHz
ns
0
1/tCLCL
Oscillator Frequency
Read Low to Data Valid
RD Pulse Width
0.0
10 tRLDV
12 tRLRH
15 tDVWH
16 tWLWH
690
3.0tCLCL-60
735
750
735
3.0tCLCL-15
3.0tCLCL
ns
Data Valid to WR High
WR Pulse Width
ns
3.0tCLCL-15
ns
Table 145. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 1, SRWn0 = 1
4 MHz Oscillator Variable Oscillator
Symbol
Parameter
Min
Max
Min
Max
8
Unit
MHz
ns
0
1/tCLCL
Oscillator Frequency
Read Low to Data Valid
RD Pulse Width
0.0
10 tRLDV
12 tRLRH
14 tWHDX
15 tDVWH
16 tWLWH
690
3.0tCLCL-60
735
485
750
735
3.0tCLCL-15
2.0tCLCL-15
3.0tCLCL
ns
Data Hold After WR High
Data Valid to WR High
WR Pulse Width
ns
ns
3.0tCLCL-15
ns
338
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Figure 159. External Memory Timing (SRWn1 = 0, SRWn0 = 0
T1
T2
T3
T4
System Clock (CLKCPU
)
1
ALE
4
2
7
A15:8 Prev. addr.
Address
15
3a
3b
13
DA7:0 Prev. data
Address
6
XX
Data
16
14
WR
9
11
DA7:0 (XMBK = 0)
Address
5
Data
10
8
12
RD
Figure 160. External Memory Timing (SRWn1 = 0, SRWn0 = 1)
T1
T2
T3
T4
T5
System Clock (CLKCPU
)
1
ALE
4
2
7
A15:8 Prev. addr.
Address
15
3a
3b
13
Data
16
DA7:0 Prev. data
Address
6
XX
14
WR
9
11
DA7:0 (XMBK = 0)
Address
5
Data
10
8
12
RD
339
2490G–AVR–03/04
Figure 161. External Memory Timing (SRWn1 = 1, SRWn0 = 0)
T1
T2
T3
T4
T5
T6
System Clock (CLKCPU
)
1
ALE
4
2
7
Address
15
A15:8 Prev. addr.
3a
3b
13
DA7:0 Prev. data
Address
6
XX
Data
16
14
WR
9
11
DA7:0 (XMBK = 0)
Address
5
Data
10
8
12
RD
Figure 162. External Memory Timing (SRWn1 = 1, SRWn0 = 1)(1)
T1
T2
T3
T4
T5
T6
T7
System Clock (CLKCPU
)
1
ALE
4
2
7
Address
15
A15:8 Prev. addr.
3a
3b
13
DA7:0 Prev. data
Address
6
XX
Data
16
14
WR
9
11
DA7:0 (XMBK = 0)
Address
5
Data
10
8
12
RD
Note:
1. The ALE pulse in the last period (T4-T7) is only present if the next instruction
accesses the RAM (internal or external).
340
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
ATmega64 Typical
Characteristics –
Preliminary Data
The following charts show typical behavior. These figures are not tested during manu-
facturing. All current consumption measurements are performed with all I/O pins
configured as inputs and with internal pull-ups enabled. A sine wave generator with rail-
to-rail output is used as clock source.
The power consumption in Power-down mode is independent of clock selection.
The current consumption is a function of several factors such as: Operating voltage,
operating frequency, loading of I/O pins, switching rate of I/O pins, code executed and
ambient temperature. The dominating factors are operating voltage and frequency.
The current drawn from capacitive loaded pins may be estimated (for one pin) as
CL*VCC*f where CL = load capacitance, VCC = operating voltage and f = average switch-
ing frequency of I/O pin.
The parts are characterized at frequencies higher than test limits. Parts are not guaran-
teed to function properly at frequencies higher than the ordering code indicates.
The difference between current consumption in Power-down mode with Watchdog
Timer enabled and Power-down mode with Watchdog Timer disabled represents the dif-
ferential current drawn by the Watchdog Timer.
341
2490G–AVR–03/04
Register Summary
Address
Name
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
Page
(0xFF)
..
(0x9E)
Reserved
Reserved
Reserved
UCSR1C
UDR1
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
(0x9D)
(0x9C)
(0x9B)
(0x9A)
(0x99)
(0x98)
(0x97)
(0x96)
(0x95)
(0x94)
(0x93)
(0x92)
(0x91)
UMSEL1
UPM11
UPM10
USART1 I/O Data Register
FE1
RXEN1
USART1 Baud Rate Register Low
– USART1 Baud Rate Register High
USBS1
UCSZ11
UCSZ10
UCPOL1
189
186
187
188
191
191
UCSR1A
UCSR1B
UBRR1L
UBRR1H
Reserved
Reserved
UCSR0C
Reserved
Reserved
Reserved
Reserved
UBRR0H
Reserved
ADCSRB
Reserved
TCCR3C
TCCR3A
TCCR3B
TCNT3H
TCNT3L
OCR3AH
OCR3AL
OCR3BH
OCR3BL
OCR3CH
OCR3CL
ICR3H
RXC1
RXCIE1
TXC1
TXCIE1
UDRE1
UDRIE1
DOR1
TXEN1
UPE1
UCSZ12
U2X1
RXB81
MPCM1
TXB81
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
UMSEL0
UPM01
–
–
–
–
–
–
–
UPM00
–
–
–
–
–
–
–
USBS0
UCSZ01
UCSZ00
UCPOL0
189
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
(0x90)
(0x8F)
(0x8E)
–
–
–
USART0 Baud Rate Register High
191
247
–
–
–
–
–
ADTS2
–
–
ADTS1
–
–
ADTS0
–
(0x8D)
(0x8C)
(0x8B)
(0x8A)
(0x89)
(0x88)
(0x87)
(0x86)
(0x85)
(0x84)
(0x83)
(0x82)
(0x81)
(0x80)
(0x7F)
(0x7E)
(0x7D)
(0x7C)
(0x7B)
(0x7A)
(0x79)
(0x78)
(0x77)
(0x76)
(0x75)
(0x74)
(0x73)
(0x72)
(0x71)
(0x70)
(0x6F)
(0x6E)
(0x6D)
(0x6C)
(0x6B)
(0x6A)
(0x69)
(0x68)
(0x67)
(0x66)
(0x65)
(0x64)
(0x63)
–
–
–
–
–
FOC3A
COM3A1
ICNC3
FOC3B
COM3A0
ICES3
FOC3C
COM3B1
–
–
–
–
136
131
134
136
136
137
137
137
137
137
137
138
138
COM3B0
WGM33
COM3C1
WGM32
COM3C0
CS32
WGM31
CS31
WGM30
CS30
Timer/Counter3 – Counter Register High Byte
Timer/Counter3 – Counter Register Low Byte
Timer/Counter3 – Output Compare Register A High Byte
Timer/Counter3 – Output Compare Register A Low Byte
Timer/Counter3 – Output Compare Register B High Byte
Timer/Counter3 – Output Compare Register B Low Byte
Timer/Counter3 – Output Compare Register C High Byte
Timer/Counter3 – Output Compare Register C Low Byte
Timer/Counter3 – Input Capture Register High Byte
Timer/Counter3 – Input Capture Register Low Byte
ICR3L
Reserved
Reserved
ETIMSK
ETIFR
Reserved
TCCR1C
OCR1CH
OCR1CL
Reserved
Reserved
Reserved
TWCR
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
OCIE3A
OCF3A
–
–
OCIE3B
OCF3B
–
–
TOIE3
TOV3
–
–
OCIE3C
OCF3C
–
–
OCIE1C
OCF1C
–
TICIE3
ICF3
–
139
140
FOC1A
FOC1B
FOC1C
–
–
–
–
–
135
137
137
Timer/Counter1 – Output Compare Register C High Byte
Timer/Counter1 – Output Compare Register C Low Byte
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
TWIE
205
207
207
206
205
40
TWDR
TWAR
TWSR
TWBR
Two-wire Serial Interface Data Register
TWA6
TWS7
TWA5
TWS6
TWA4
TWS5
TWA3
TWS4
TWA2
TWS3
TWA1
–
TWA0
TWPS1
TWGCE
TWPS0
Two-wire Serial Interface Bit Rate Register
Oscillator Calibration Register
OSCCAL
Reserved
XMCRA
XMCRB
Reserved
EICRA
Reserved
SPMCSR
Reserved
Reserved
PORTG
DDRG
–
–
–
–
–
SRL0
–
–
SRW01
–
–
SRW00
XMM2
–
ISC10
–
PGWRT
–
–
PORTG2
DDG2
PING2
PORTF2
DDF2
–
SRW11
XMM1
–
ISC01
–
PGERS
–
–
PORTG1
DDG1
PING1
PORTF1
DDF1
–
SRL2
–
–
ISC30
–
RWWSB
SRL1
–
–
ISC21
–
–
–
–
–
–
–
30
32
XMBK
–
ISC31
–
SPMIE
–
–
–
XMM0
–
ISC00
–
SPMEN
–
–
PORTG0
DDG0
PING0
PORTF0
DDF0
–
–
ISC20
–
RWWSRE
–
ISC11
–
BLBSET
–
88
281
–
–
–
–
–
–
–
PORTG4
DDG4
PING4
PORTF4
DDF4
PORTG3
DDG3
PING3
PORTF3
DDF3
87
87
87
86
87
–
–
PING
PORTF
DDRF
(0x62)
(0x61)
PORTF7
DDF7
PORTF6
DDF6
PORTF5
DDF5
342
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Register Summary (Continued)
Address
Name
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
Page
(0x60)
Reserved
SREG
–
I
–
T
–
H
–
S
–
V
–
N
–
Z
–
C
0x3F (0x5F)
0x3E (0x5E)
0x3D (0x5D)
0x3C (0x5C)
0x3B (0x5B)
0x3A (0x5A)
0x39 (0x59)
0x38 (0x58)
0x37 (0x57)
0x36 (0x56)
0x35 (0x55)
0x34 (0x54)
0x33 (0x53)
0x32 (0x52)
0x31 (0x51)
0x30 (0x50)
0x2F (0x4F)
0x2E (0x4E)
0x2D (0x4D)
0x2C (0x4C)
0x2B (0x4B)
0x2A (0x4A)
0x29 (0x49)
0x28 (0x48)
0x27 (0x47)
0x26 (0x46)
0x25 (0x45)
0x24 (0x44)
0x23 (0x43)
10
12
12
43
SPH
SPL
XDIV
SP15
SP7
XDIVEN
–
ISC71
INT7
INTF7
OCIE2
OCF2
SRE
JTD
SP14
SP6
XDIV6
–
ISC70
INT6
INTF6
TOIE2
TOV2
SRW10
–
SP13
SP5
XDIV5
–
ISC61
INT5
INTF5
TICIE1
ICF1
SE
SP12
SP4
XDIV4
–
ISC60
INT4
INTF4
OCIE1A
OCF1A
SM1
SP11
SP3
XDIV3
–
ISC51
INT3
INTF3
OCIE1B
OCF1B
SM0
SP10
SP2
XDIV2
–
ISC50
INT2
INTF
TOIE1
TOV1
SM2
BORF
CS02
SP9
SP1
XDIV1
–
ISC41
INT1
INTF1
OCIE0
OCF0
IVSEL
EXTRF
CS01
SP8
SP0
XDIV0
–
ISC40
INT0
INTF0
TOIE0
TOV0
IVCE
PORF
CS00
Reserved
EICRB
EIMSK
EIFR
TIMSK
TIFR
MCUCR
MCUCSR
TCCR0
TCNT0
OCR0
89
90
90
107, 138, 158
107, 140, 158
30, 44, 62
53, 256
102
–
JTRF
COM00
Timer/Counter0 (8 Bit)
WDRF
WGM01
FOC0
WGM00
COM01
104
104
105
131
134
136
136
137
137
137
137
138
138
155
157
158
Timer/Counter0 Output Compare Register
ASSR
–
–
–
COM1B1
–
–
AS0
COM1C1
WGM12
TCN0UB
COM1C0
CS12
OCR0UB
WGM11
CS11
TCR0UB
WGM10
CS10
TCCR1A
TCCR1B
TCNT1H
TCNT1L
OCR1AH
OCR1AL
OCR1BH
OCR1BL
ICR1H
COM1A1
ICNC1
COM1A0
ICES1
COM1B0
WGM13
Timer/Counter1 – Counter Register High Byte
Timer/Counter1 – Counter Register Low Byte
Timer/Counter1 – Output Compare Register A High Byte
Timer/Counter1 – Output Compare Register A Low Byte
Timer/Counter1 – Output Compare Register B High Byte
Timer/Counter1 – Output Compare Register B Low Byte
Timer/Counter1 – Input Capture Register High Byte
Timer/Counter1 – Input Capture Register Low Byte
ICR1L
TCCR2
TCNT2
OCR2
FOC2
WGM20
OCDR6
COM21
COM20
Timer/Counter2 (8 Bit)
Timer/Counter2 Output Compare Register
WGM21
CS22
CS21
CS20
IDRD/
0x22 (0x42)
OCDR
OCDR5
OCDR4
OCDR3
OCDR2
OCDR1
OCDR0
253
OCDR7
0x21 (0x41)
0x20 (0x40)
0x1F (0x3F)
0x1E (0x3E)
0x1D (0x3D)
0x1C (0x3C)
0x1B (0x3B)
0x1A (0x3A)
0x19 (0x39)
0x18 (0x38)
0x17 (0x37)
0x16 (0x36)
0x15 (0x35)
0x14 (0x34)
0x13 (0x33)
0x12 (0x32)
0x11 (0x31)
0x10 (0x30)
0x0F (0x2F)
0x0E (0x2E)
0x0D (0x2D)
0x0C (0x2C)
0x0B (0x2B)
0x0A (0x2A)
0x09 (0x29)
0x08 (0x28)
0x07 (0x27)
0x06 (0x26)
0x05 (0x25)
0x04 (0x24)
0x03 (0x23)
0x02 (0x22)
0x01 (0x21)
WDTCR
SFIOR
EEARH
EEARL
EEDR
–
TSM
–
–
–
–
–
–
–
WDCE
–
–
WDE
ACME
–
WDP2
PUD
WDP1
PSR0
WDP0
PSR321
55
70, 109, 143, 227
EEPROM Address Register High Byte
20
20
20
20
85
85
85
85
85
85
85
85
86
86
86
86
167
167
165
186
187
188
191
228
243
245
246
246
86
EEPROM Address Register Low Byte
EEPROM Data Register
EECR
–
–
–
–
EERIE
PORTA3
DDA3
PINA3
PORTB3
DDB3
PINB3
PORTC3
DDC3
PINC3
PORTD3
DDD3
EEMWE
PORTA2
DDA2
PINA2
PORTB2
DDB2
PINB2
PORTC2
DDC2
PINC2
PORTD2
DDD2
EEWE
PORTA1
DDA1
PINA1
PORTB1
DDB1
PINB1
PORTC1
DDC1
PINC1
PORTD1
DDD1
EERE
PORTA0
DDA0
PINA0
PORTB0
DDB0
PINB0
PORTC0
DDC0
PINC0
PORTD0
DDD0
PORTA
DDRA
PINA
PORTB
DDRB
PINB
PORTC
DDRC
PINC
PORTD
DDRD
PIND
SPDR
SPSR
SPCR
UDR0
UCSR0A
UCSR0B
UBRR0L
ACSR
ADMUX
ADCSRA
ADCH
ADCL
PORTA7
DDA7
PINA7
PORTB7
DDB7
PINB7
PORTC7
DDC7
PINC7
PORTD7
DDD7
PORTA6
DDA6
PINA6
PORTB6
DDB6
PINB6
PORTC6
DDC6
PINC6
PORTD6
DDD6
PORTA5
DDA5
PINA5
PORTB5
DDB5
PINB5
PORTC5
DDC5
PINC5
PORTD5
DDD5
PORTA4
DDA4
PINA4
PORTB4
DDB4
PINB4
PORTC4
DDC4
PINC4
PORTD4
DDD4
PIND7
PIND6
PIND5
PIND4
PIND3
PIND2
PIND1
PIND0
SPI Data Register
SPIF
SPIE
WCOL
SPE
–
–
–
–
–
SPI2X
SPR0
DORD
MSTR
CPOL
CPHA
SPR1
USART0 I/O Data Register
FE0
RXEN0
RXC0
RXCIE0
TXC0
TXCIE0
UDRE0
UDRIE0
DOR0
TXEN0
UPE0
UCSZ02
U2X0
RXB80
MPCM0
TXB80
USART0 Baud Rate Register Low
ACD
REFS1
ADEN
ACBG
REFS0
ADSC
ACO
ADLAR
ADATE
ACI
MUX4
ADIF
ACIE
MUX3
ADIE
ACIC
MUX2
ADPS2
ACIS1
MUX1
ADPS1
ACIS0
MUX0
ADPS0
ADC Data Register High Byte
ADC Data Register Low byte
PORTE4
DDE4
PORTE
DDRE
PINE
PORTE7
DDE7
PINE7
PORTE6
DDE6
PINE6
PORTE5
DDE5
PINE5
PORTE3
DDE3
PINE3
PORTE2
DDE2
PINE2
PORTE1
DDE1
PINE1
PORTE0
DDE0
PINE0
86
86
PINE4
343
2490G–AVR–03/04
Register Summary (Continued)
Address
Name
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
Page
0x00 (0x20)
PINF
PINF7
PINF6
PINF5
PINF4
PINF3
PINF2
PINF1
PINF0
87
Notes: 1. For compatibility with future devices, reserved bits should be written to zero if accessed. Reserved I/O memory addresses
should never be written.
2. Some of the status flags are cleared by writing a logical one 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 0x00 to 0x1F only.
344
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Instruction Set Summary
Mnemonics
Operands
Description
Operation
Flags
#Clocks
ARITHMETIC AND LOGIC INSTRUCTIONS
ADD
Rd, Rr
Rd, Rr
Rdl,K
Rd, Rr
Rd, K
Rd, Rr
Rd, K
Rdl,K
Rd, Rr
Rd, K
Rd, Rr
Rd, K
Rd, Rr
Rd
Add two Registers
Rd ← Rd + Rr
Z,C,N,V,H
Z,C,N,V,H
Z,C,N,V,S
Z,C,N,V,H
Z,C,N,V,H
Z,C,N,V,H
Z,C,N,V,H
Z,C,N,V,S
Z,N,V
Z,N,V
Z,N,V
Z,N,V
Z,N,V
Z,C,N,V
Z,C,N,V,H
Z,N,V
Z,N,V
Z,N,V
Z,N,V
Z,N,V
Z,N,V
None
1
1
2
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
ADC
ADIW
SUB
SUBI
SBC
SBCI
SBIW
AND
ANDI
OR
Add with Carry two Registers
Add Immediate to Word
Subtract two Registers
Subtract Constant from Register
Subtract with Carry two Registers
Subtract with Carry Constant from Reg.
Subtract Immediate from Word
Logical AND Registers
Logical AND Register and Constant
Logical OR Registers
Logical OR Register and Constant
Exclusive OR Registers
One’s Complement
Two’s Complement
Set Bit(s) in Register
Rd ← Rd + Rr + C
Rdh:Rdl ← Rdh:Rdl + K
Rd ← Rd - Rr
Rd ← Rd - K
Rd ← Rd - Rr - C
Rd ← Rd - K - C
Rdh:Rdl ← Rdh:Rdl - K
Rd ← Rd • Rr
Rd ← Rd • K
Rd ← Rd v Rr
ORI
Rd ← Rd v K
Rd ← Rd ⊕ Rr
EOR
COM
NEG
SBR
CBR
INC
Rd ← 0xFF − Rd
Rd ← 0x00 − Rd
Rd ← Rd v K
Rd ← Rd • (0xFF - K)
Rd ← Rd + 1
Rd ← Rd − 1
Rd ← Rd • Rd
Rd ← Rd ⊕ Rd
Rd ← 0xFF
R1:R0 ← Rd x Rr
R1:R0 ← Rd x Rr
R1:R0 ← Rd x Rr
R1:R0 ¨ (Rd x Rr) << 1
R1:R0 ¨ (Rd x Rr) << 1
R1:R0 ¨ (Rd x Rr) << 1
Rd
Rd,K
Rd,K
Rd
Clear Bit(s) in Register
Increment
DEC
Rd
Decrement
TST
CLR
Rd
Rd
Test for Zero or Minus
Clear Register
SER
Rd
Set Register
MUL
Rd, Rr
Rd, Rr
Rd, Rr
Rd, Rr
Rd, Rr
Rd, Rr
Multiply Unsigned
Multiply Signed
Multiply Signed with Unsigned
Fractional Multiply Unsigned
Fractional Multiply Signed
Fractional Multiply Signed with Unsigned
Z,C
Z,C
Z,C
Z,C
Z,C
Z,C
MULS
MULSU
FMUL
FMULS
FMULSU
BRANCH INSTRUCTIONS
RJMP
IJMP
JMP
RCALL
ICALL
CALL
RET
k
Relative Jump
Indirect Jump to (Z)
Direct Jump
Relative Subroutine Call
Indirect Call to (Z)
Direct Subroutine Call
Subroutine Return
Interrupt Return
PC ← PC + k + 1
PC ← Z
PC ← k
PC ← PC + k + 1
PC ← Z
PC ← k
PC ← STACK
PC ← STACK
None
None
None
None
None
None
None
I
2
2
3
3
3
4
4
4
k
k
k
RETI
CPSE
CP
CPC
Rd,Rr
Rd,Rr
Rd,Rr
Rd,K
Rr, b
Rr, b
P, b
P, b
s, k
s, k
k
k
k
k
k
k
k
k
k
k
k
k
k
Compare, Skip if Equal
Compare
Compare with Carry
if (Rd = Rr) PC ← PC + 2 or 3
Rd − Rr
Rd − Rr − C
None
Z, N,V,C,H
Z, N,V,C,H
Z, N,V,C,H
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
1/2/3
1
1
CPI
Compare Register with Immediate
Skip if Bit in Register Cleared
Skip if Bit in Register is Set
Skip if Bit in I/O Register Cleared
Skip if Bit in I/O Register is Set
Branch if Status Flag Set
Branch if Status Flag Cleared
Branch if Equal
Branch if Not Equal
Branch if Carry Set
Branch if Carry Cleared
Branch if Same or Higher
Branch if Lower
Rd − K
1
SBRC
SBRS
SBIC
SBIS
if (Rr(b)=0) PC ← PC + 2 or 3
if (Rr(b)=1) PC ← PC + 2 or 3
if (P(b)=0) PC ← PC + 2 or 3
if (P(b)=1) PC ← PC + 2 or 3
if (SREG(s) = 1) then PC←PC+k + 1
if (SREG(s) = 0) then PC←PC+k + 1
if (Z = 1) then PC ← PC + k + 1
if (Z = 0) then PC ← PC + k + 1
if (C = 1) then PC ← PC + k + 1
if (C = 0) then PC ← PC + k + 1
if (C = 0) then PC ← PC + k + 1
if (C = 1) then PC ← PC + k + 1
if (N = 1) then PC ← PC + k + 1
if (N = 0) then PC ← PC + k + 1
if (N ⊕ V= 0) then PC ← PC + k + 1
if (N ⊕ V= 1) then PC ← PC + k + 1
if (H = 1) then PC ← PC + k + 1
if (H = 0) then PC ← PC + k + 1
if (T = 1) then PC ← PC + k + 1
if (T = 0) then PC ← PC + k + 1
if (V = 1) then PC ← PC + k + 1
if (V = 0) then PC ← PC + k + 1
1/2/3
1/2/3
1/2/3
1/2/3
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
BRBS
BRBC
BREQ
BRNE
BRCS
BRCC
BRSH
BRLO
BRMI
BRPL
BRGE
BRLT
BRHS
BRHC
BRTS
BRTC
BRVS
BRVC
Branch if Minus
Branch if Plus
Branch if Greater or Equal, Signed
Branch if Less Than Zero, Signed
Branch if Half Carry Flag Set
Branch if Half Carry Flag Cleared
Branch if T Flag Set
Branch if T Flag Cleared
Branch if Overflow Flag is Set
Branch if Overflow Flag is Cleared
k
k
k
345
2490G–AVR–03/04
Instruction Set Summary (Continued)
BRIE
k
Branch if Interrupt Enabled
if ( I = 1) then PC ← PC + k + 1
if ( I = 0) then PC ← PC + k + 1
None
None
1/2
1/2
BRID
k
Branch if Interrupt Disabled
DATA TRANSFER INSTRUCTIONS
MOV
MOVW
LDI
LD
LD
LD
LD
LD
LD
LDD
LD
LD
LD
LDD
LDS
ST
ST
ST
ST
ST
ST
STD
ST
ST
ST
STD
STS
LPM
LPM
LPM
SPM
IN
OUT
PUSH
POP
Rd, Rr
Rd, Rr
Rd, K
Move Between Registers
Copy Register Word
Load Immediate
Rd ← Rr
Rd+1:Rd ← Rr+1:Rr
Rd ← K
Rd ← (X)
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
-
Rd, X
Load Indirect
Rd, X+
Rd, - X
Rd, Y
Rd, Y+
Rd, - Y
Rd,Y+q
Rd, Z
Rd, Z+
Rd, -Z
Rd, Z+q
Rd, k
Load Indirect and Post-Inc.
Load Indirect and Pre-Dec.
Load Indirect
Load Indirect and Post-Inc.
Load Indirect and Pre-Dec.
Load Indirect with Displacement
Load Indirect
Load Indirect and Post-Inc.
Load Indirect and Pre-Dec.
Load Indirect with Displacement
Load Direct from SRAM
Store Indirect
Store Indirect and Post-Inc.
Store Indirect and Pre-Dec.
Store Indirect
Store Indirect and Post-Inc.
Store Indirect and Pre-Dec.
Store Indirect with Displacement
Store Indirect
Store Indirect and Post-Inc.
Store Indirect and Pre-Dec.
Store Indirect with Displacement
Store Direct to SRAM
Load Program Memory
Load Program Memory
Load Program Memory and Post-Inc
Store Program Memory
In Port
Rd ← (X), X ← X + 1
X ← X - 1, Rd ← (X)
Rd ← (Y)
Rd ← (Y), Y ← Y + 1
Y ← Y - 1, Rd ← (Y)
Rd ← (Y + q)
Rd ← (Z)
Rd ← (Z), Z ← Z+1
Z ← Z - 1, Rd ← (Z)
Rd ← (Z + q)
Rd ← (k)
X, Rr
(X) ← Rr
X+, Rr
- X, Rr
Y, Rr
Y+, Rr
- Y, Rr
Y+q,Rr
Z, Rr
Z+, Rr
-Z, Rr
Z+q,Rr
k, Rr
(X) ← Rr, X ← X + 1
X ← X - 1, (X) ← Rr
(Y) ← Rr
(Y) ← Rr, Y ← Y + 1
Y ← Y - 1, (Y) ← Rr
(Y + q) ← Rr
(Z) ← Rr
(Z) ← Rr, Z ← Z + 1
Z ← Z - 1, (Z) ← Rr
(Z + q) ← Rr
(k) ← Rr
R0 ← (Z)
Rd ← (Z)
Rd ← (Z), Z ← Z+1
(Z) ← R1:R0
Rd ← P
P ← Rr
Rd, Z
Rd, Z+
Rd, P
P, Rr
Rr
1
1
2
2
Out Port
Push Register on Stack
Pop Register from Stack
STACK ← Rr
Rd ← STACK
Rd
BIT AND BIT-TEST INSTRUCTIONS
SBI
CBI
LSL
P,b
P,b
Rd
Rd
Rd
Rd
Rd
Rd
s
Set Bit in I/O Register
Clear Bit in I/O Register
Logical Shift Left
I/O(P,b) ← 1
I/O(P,b) ← 0
Rd(n+1) ← Rd(n), Rd(0) ← 0
Rd(n) ← Rd(n+1), Rd(7) ← 0
Rd(0)←C,Rd(n+1)← Rd(n),C←Rd(7)
Rd(7)←C,Rd(n)← Rd(n+1),C←Rd(0)
Rd(n) ← Rd(n+1), n=0..6
Rd(3..0)←Rd(7..4),Rd(7..4)←Rd(3..0)
SREG(s) ← 1
SREG(s) ← 0
T ← Rr(b)
Rd(b) ← T
C ← 1
C ← 0
N ← 1
N ← 0
Z ← 1
Z ← 0
I ← 1
I ← 0
S ← 1
S ← 0
V ← 1
V ← 0
T ← 1
None
None
Z,C,N,V
Z,C,N,V
Z,C,N,V
Z,C,N,V
Z,C,N,V
None
SREG(s)
SREG(s)
T
None
C
C
N
N
Z
Z
I
I
S
S
V
V
T
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
LSR
ROL
ROR
ASR
SWAP
BSET
BCLR
BST
BLD
SEC
CLC
SEN
CLN
SEZ
CLZ
SEI
Logical Shift Right
Rotate Left Through Carry
Rotate Right Through Carry
Arithmetic Shift Right
Swap Nibbles
Flag Set
Flag Clear
Bit Store from Register to T
Bit load from T to Register
Set Carry
Clear Carry
Set Negative Flag
Clear Negative Flag
Set Zero Flag
Clear Zero Flag
Global Interrupt Enable
Global Interrupt Disable
Set Signed Test Flag
Clear Signed Test Flag
Set Twos Complement Overflow.
Clear Twos Complement Overflow
Set T in SREG
s
Rr, b
Rd, b
CLI
SES
CLS
SEV
CLV
SET
CLT
SEH
Clear T in SREG
Set Half Carry Flag in SREG
T ← 0
H ← 1
T
H
346
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Instruction Set Summary (Continued)
CLH
Clear Half Carry Flag in SREG
H ← 0
H
1
MCU CONTROL INSTRUCTIONS
NOP
SLEEP
WDR
No Operation
Sleep
Watchdog Reset
Break
None
None
None
None
1
1
1
(see specific descr. for Sleep function)
(see specific descr. for WDR/timer)
For On-chip Debug Only
BREAK
N/A
347
2490G–AVR–03/04
Ordering Information
Speed (MHz)
Power Supply
Ordering Code
Package
Operation Range
8
2.7 - 5.5
ATmega64L-8AC
ATmega64L-8MC
64A
Commercial
64M1
(0°C to 70°C)
ATmega64L-8AI
ATmega64L-8MI
64A
Industrial
64M1
(-40°C to 85°C)
16
4.5 - 5.5
ATmega64-16AC
ATmega64-16MC
64A
Commercial
64M1
(0°C to 70°C)
ATmega64-16AI
ATmega64-16MI
64A
Industrial
64M1
(-40°C to 85°C)
Note:
1. This device can also be supplied in wafer form. Please contact your local Atmel sales office for detailed ordering information
and minimum quantities.
Package Type
64A
64-lead, Thin (1.0 mm) Plastic Gull Wing Quad Flat Package (TQFP)
64-pad, 9 x 9 x 1.0 mm body, lead pitch 0.50 mm, Micro Lead Frame Package (MLF)
64M1
348
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Packaging Information
64A
PIN 1
B
PIN 1 IDENTIFIER
E1
E
e
D1
D
C
0˚~7˚
A2
A
A1
L
COMMON DIMENSIONS
(Unit of Measure = mm)
MIN
–
MAX
1.20
NOM
NOTE
SYMBOL
A
–
–
A1
A2
D
0.05
0.95
15.75
13.90
15.75
13.90
0.30
0.09
0.45
0.15
1.00
16.00
14.00
16.00
14.00
–
1.05
16.25
D1
E
14.10 Note 2
16.25
Notes:
1. This package conforms to JEDEC reference MS-026, Variation AEB.
2. Dimensions D1 and E1 do not include mold protrusion. Allowable
protrusion is 0.25 mm per side. Dimensions D1 and E1 are maximum
plastic body size dimensions including mold mismatch.
E1
B
14.10 Note 2
0.45
C
–
0.20
3. Lead coplanarity is 0.10 mm maximum.
L
–
0.75
e
0.80 TYP
10/5/2001
TITLE
DRAWING NO.
REV.
2325 Orchard Parkway
San Jose, CA 95131
64A, 64-lead, 14 x 14 mm Body Size, 1.0 mm Body Thickness,
0.8 mm Lead Pitch, Thin Profile Plastic Quad Flat Package (TQFP)
64A
R
B
349
2490G–AVR–03/04
64M1
D
Marked Pin# 1 ID
E
SEATING PLANE
C
A1
TOP VIEW
A
0.08
C
L
Pin #1 Corner
SIDE VIEW
D2
1
2
3
COMMON DIMENSIONS
(Unit of Measure = mm)
MIN
0.80
–
MAX
1.00
0.05
0.28
NOM
0.90
NOTE
SYMBOL
E2
A
A1
b
0.02
0.23
0.25
D
9.00 BSC
5.40
D2
E
5.20
5.20
0.35
5.60
5.60
0.45
9.00 BSC
5.40
e
b
E2
e
BOTTOM VIEW
0.50 BSC
0.40
L
Notes: 1. JEDEC Standard MO-220, Fig. 1, VMMD.
01/15/03
DRAWING NO. REV.
TITLE
2325 Orchard Parkway
San Jose, CA 95131
64M1, 64-pad, 9 x 9 x 1.0 mm Body, Lead Pitch 0.50 mm
Micro Lead Frame Package (MLF)
64M1
C
R
350
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Errata
The revision letter in this section refers to the revision of the ATmega64 device.
ATmega64, all rev.
• Stabilizing time needed when changing XDIV Register
• Stabilizing time needed when changing OSCCAL Register
1. Stabilizing time needed when changing XDIV Register
After increasing the source clock frequency more than 2% with settings in the XDIV
register, the device may execute some of the subsequent instructions incorrectly.
Problem Fix / Workaround
The NOP instruction will always be executed correctly also right after a frequency
change. Thus, the next 8 instructions after the change should be NOP instructions.
To ensure this, follow this procedure:
1.Clear the I bit in the SREG Register.
2.Set the new pre-scaling factor in XDIV register.
3.Execute 8 NOP instructions
4.Set the I bit in SREG
This will ensure that all subsequent instructions will execute correctly.
Assembly Code Example:
CLI
; clear global interrupt enable
; set new prescale value
; no operation
OUT XDIV, temp
NOP
NOP
NOP
NOP
NOP
NOP
NOP
NOP
SEI
; no operation
; no operation
; no operation
; no operation
; no operation
; no operation
; no operation
; clear global interrupt enable
2. Stabilizing time needed when changing OSCCAL Register
After increasing the source clock frequency more than 2% with settings in the OSC-
CAL register, the device may execute some of the subsequent instructions
incorrectly.
Problem Fix / Workaround
The behavior follows errata number 1., and the same Fix / Workaround is applicable
on this errata.
A proposal for solving problems regarding the JTAG instruction IDCODE is presented
below.
IDCODE masks data from TDI input
The public but optional JTAG instruction IDCODE is not implemented correctly
according to IEEE1149.1; a logic one is scanned into the shift register instead of the
TDI input while shifting the Device ID Register. Hence, captured data from the pre-
ceding devices in the boundary scan chain are lost and replaced by all-ones, and
data to succeeding devices are replaced by all-ones during Update-DR.
If ATmega64 is the only device in the scan chain, the problem is not visible.
351
2490G–AVR–03/04
Problem Fix / Workaround
Select the Device ID Register of the ATmega64 (Either by issuing the IDCODE
instruction or by entering the Test-Logic-Reset state of the TAP controller) to read
out the contents of its Device ID Register and possibly data from succeeding
devices of the scan chain. Note that data to succeeding devices cannot be entered
during this scan, but data to preceding devices can. Issue the BYPASS instruction
to the ATmega64 to select its Bypass Register while reading the Device ID Regis-
ters of preceding devices of the boundary scan chain. Never read data from
succeeding devices in the boundary scan chain or upload data to the succeeding
devices while the Device ID Register is selected for the ATmega64. Note that the
IDCODE instruction is the default instruction selected by the Test-Logic-Reset state
of the TAP-controller.
Alternative Problem Fix / Workaround
If the Device IDs of all devices in the boundary scan chain must be captured simul-
taneously (for instance if blind interrogation is used), the boundary scan chain can
be connected in such way that the ATmega64 is the fist device in the chain. Update-
DR will still not work for the succeeding devices in the boundary scan chain as long
as IDCODE is present in the JTAG Instruction Register, but the Device ID registered
cannot be uploaded in any case.
352
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Datasheet Change
Log for ATmega64
Please note that the referring page numbers in this section are referred to this docu-
ment. The referring revision in this section are referring to the document revision.
Changes from Rev.
2490F-12/03 to Rev.
2490G-03/04
1. Updated “Errata” on page 351.
Changes from Rev.
2490E-09/03 to Rev.
2490F-12/03
1. Updated “Calibrated Internal RC Oscillator” on page 40.
Changes from Rev.
2490D-02/03 to Rev.
2490E-09/03
1. Updated note in “XTAL Divide Control Register – XDIV” on page 43.
2. Updated “JTAG Interface and On-chip Debug System” on page 48.
3. Updated “Test Access Port – TAP” on page 248 regarding JTAGEN.
4. Updated description for the JTD bit on page 258.
5. Added a note regarding JTAGEN fuse to Table 119 on page 292.
6. Updated RPU values in “DC Characteristics” on page 326.
7. Updated “ADC Characteristics – Preliminary Data” on page 333.
8. Added a proposal for solving problems regarding the JTAG instruction
IDCODE in “Errata” on page 351.
Changes from Rev.
2490C-09/02 to Rev.
2490D-02/03
1. Added reference to Table 125 on page 296 from both SPI Serial Programming
and Self Programming to inform about the Flash page size.
2. Added Chip Erase as a first step under “Programming the Flash” on page 323
and “Programming the EEPROM” on page 324.
3. Corrected OCn waveforms in Figure 52 on page 124.
4. Various minor Timer1 corrections.
5. Improved the description in “Phase Correct PWM Mode” on page 99 and on
page 152.
6. Various minor TWI corrections.
7. Added note under "Filling the Temporary Buffer (Page Loading)" about writ-
ing to the EEPROM during an SPM page load.
8. Removed ADHSM completely.
9. Added note about masking out unused bits when reading the Program
Counter in “Stack Pointer” on page 12.
353
2490G–AVR–03/04
10. Added section “EEPROM Write During Power-down Sleep Mode” on page 23.
11. Changed VHYST value to 120 in Table 19 on page 50.
12. Added information about conversion time for Differential mode with Auto
Triggering on page 234.
13. Added tWD_FUSE in Table 129 on page 309.
14. Updated “Packaging Information” on page 349.
Changes from Rev.
2490B-09/02 to Rev.
2490C-09/02
1. Changed the Endurance on the Flash to 10,000 Write/Erase Cycles.
Changes from Rev.
2490A-10/01 to Rev.
2490B-09/02
1. Added 64-pad MLF Package and updated “Ordering Information” on page 348.
2. Added the section “Using all Locations of External Memory Smaller than 64
KB” on page 33.
3. Added the section “Default Clock Source” on page 36.
4. Renamed SPMCR to SPMCSR in entire document.
5. Added Some Preliminary Test Limits and Characterization Data
Removed some of the TBD's and corrected data in the following tables and pages:
Table 2 on page 22, Table 7 on page 36, Table 9 on page 38, Table 10 on page 38,
Table 12 on page 39, Table 14 on page 40, Table 16 on page 41, Table 19 on page
50, Table 20 on page 54, Table 22 on page 56, “DC Characteristics” on page 326,
Table 132 on page 328, Table 135 on page 331, Table 137 on page 334, and Table
138 - Table 145.
6. Removed Alternative Algortihm for Leaving JTAG Programming Mode.
See “Leaving Programming Mode” on page 322.
7. Improved description on how to do a polarity check of the ADC diff results in
“ADC Conversion Result” on page 242.
8. Updated Programming Figures:
Figure 138 on page 294 and Figure 147 on page 307 are updated to also reflect that
AVCC must be connected during Programming mode. Figure 142 on page 303
added to illustrate how to program the fuses.
9. Added a note regarding usage of the “PROG_PAGELOAD (0x6)” and
“PROG_PAGEREAD (0x7)” instructions on page 314.
10. Updated “Two-wire Serial Interface” on page 196.
More details regarding use of the TWI Power-down operation and using the TWI as
master with low TWBRR values are added into the data sheet. Added the note at
the end of the “Bit Rate Generator Unit” on page 202. Added the description at the
end of “Address Match Unit” on page 203.
354
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
11. Updated Description of OSCCAL Calibration Byte.
In the data sheet, it was not explained how to take advantage of the calibration
bytes for 2, 4, and 8 MHz Oscillator selections. This is now added in the following
sections:
Improved description of “Oscillator Calibration Register – OSCCAL(1)” on page 40
and “Calibration Byte” on page 293.
12. When using external clock there are some limitations regards to change of
frequency. This is descried in “External Clock” on page 41 and Table 132 on
page 328.
13. Added a sub section regarding OCD-system and power consumption in the
section “Minimizing Power Consumption” on page 47.
14. Corrected typo (WGM-bit setting) for:
–
–
–
–
“Fast PWM Mode” on page 97 (Timer/Counter0).
“Phase Correct PWM Mode” on page 99 (Timer/Counter0).
“Fast PWM Mode” on page 150 (Timer/Counter2).
“Phase Correct PWM Mode” on page 152 (Timer/Counter2).
15. Corrected Table 81 on page 190 (USART).
16. Corrected Table 103 on page 262 (Boundary-Scan)
355
2490G–AVR–03/04
356
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
Table of Contents
Features................................................................................................ 1
Pin Configuration................................................................................. 2
Disclaimer............................................................................................................. 2
Overview............................................................................................... 3
Block Diagram ...................................................................................................... 3
ATmega103 and ATmega64 Compatibility........................................................... 4
Pin Descriptions.................................................................................................... 5
About Code Examples......................................................................... 7
AVR CPU Core ..................................................................................... 8
Introduction........................................................................................................... 8
Architectural Overview.......................................................................................... 8
ALU – Arithmetic Logic Unit.................................................................................. 9
Status Register ................................................................................................... 10
General Purpose Register File ........................................................................... 11
Stack Pointer ...................................................................................................... 12
Instruction Execution Timing............................................................................... 13
Reset and Interrupt Handling.............................................................................. 13
AVR ATmega64 Memories ................................................................ 16
In-System Reprogrammable Flash Program Memory ........................................ 16
SRAM Data Memory........................................................................................... 17
EEPROM Data Memory...................................................................................... 19
I/O Memory......................................................................................................... 24
External Memory Interface.................................................................................. 25
XMEM Register Description................................................................................ 30
System Clock and Clock Options .................................................... 35
Clock Systems and their Distribution.................................................................. 35
Clock Sources..................................................................................................... 36
Default Clock Source.......................................................................................... 36
Crystal Oscillator................................................................................................. 37
Low-frequency Crystal Oscillator........................................................................ 38
External RC Oscillator ........................................................................................ 39
Calibrated Internal RC Oscillator ........................................................................ 40
External Clock..................................................................................................... 41
Timer/Counter Oscillator..................................................................................... 43
Power Management and Sleep Modes............................................. 44
Idle Mode............................................................................................................ 45
ADC Noise Reduction Mode............................................................................... 45
Power-down Mode.............................................................................................. 45
Power-save Mode............................................................................................... 45
i
2490G–AVR–03/04
Standby Mode..................................................................................................... 46
Extended Standby Mode .................................................................................... 46
Minimizing Power Consumption ......................................................................... 47
System Control and Reset................................................................ 49
Internal Voltage Reference................................................................................. 54
Watchdog Timer ................................................................................................. 54
Timed Sequences for Changing the Configuration of the Watchdog Timer ....... 58
Interrupts............................................................................................ 59
Interrupt Vectors in ATmega64........................................................................... 59
I/O Ports.............................................................................................. 64
Introduction......................................................................................................... 64
Ports as General Digital I/O................................................................................ 65
Alternate Port Functions ..................................................................................... 69
Register Description for I/O Ports....................................................................... 85
External Interrupts............................................................................. 88
8-bit Timer/Counter0 with PWM and Asynchronous Operation.... 91
Overview............................................................................................................. 91
Timer/Counter Clock Sources............................................................................. 92
Counter Unit........................................................................................................ 93
Output Compare Unit.......................................................................................... 93
Compare Match Output Unit............................................................................... 95
Modes of Operation ............................................................................................ 96
Timer/Counter Timing Diagrams....................................................................... 100
8-bit Timer/Counter Register Description ......................................................... 102
Asynchronous Operation of the Timer/Counter ................................................ 105
Timer/Counter Prescaler................................................................................... 108
16-bit Timer/Counter (Timer/Counter1 and Timer/Counter3)....... 110
Overview........................................................................................................... 110
Accessing 16-bit Registers ............................................................................... 113
Timer/Counter Clock Sources........................................................................... 115
Counter Unit...................................................................................................... 116
Input Capture Unit............................................................................................. 117
Output Compare Units...................................................................................... 119
Compare Match Output Unit............................................................................. 120
Modes of Operation .......................................................................................... 122
Timer/Counter Timing Diagrams....................................................................... 129
16-bit Timer/Counter Register Description ....................................................... 131
Timer/Counter3, Timer/Counter2 and Timer/Counter1 Prescalers ....
142
ii
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
8-bit Timer/Counter2 with PWM...................................................... 144
Overview........................................................................................................... 144
Timer/Counter Clock Sources........................................................................... 145
Counter Unit...................................................................................................... 146
Output Compare Unit........................................................................................ 146
Compare Match Output Unit............................................................................. 148
Modes of Operation .......................................................................................... 149
Timer/Counter Timing Diagrams....................................................................... 153
8-bit Timer/Counter Register Description ......................................................... 155
Output Compare Modulator (OCM1C2).......................................... 159
Overview........................................................................................................... 159
Description........................................................................................................ 159
Serial Peripheral Interface – SPI..................................................... 161
SS Pin Functionality.......................................................................................... 165
Data Modes ...................................................................................................... 168
USART .............................................................................................. 169
Overview........................................................................................................... 169
Clock Generation.............................................................................................. 170
Frame Formats ................................................................................................. 173
USART Initialization.......................................................................................... 174
Data Transmission – The USART Transmitter ................................................. 175
Data Reception – The USART Receiver .......................................................... 179
Asynchronous Data Reception ......................................................................... 182
Multi-processor Communication Mode ............................................................. 185
USART Register Description ............................................................................ 186
Examples of Baud Rate Setting........................................................................ 191
Two-wire Serial Interface ................................................................ 196
Features............................................................................................................ 196
Two-wire Serial Interface Bus Definition........................................................... 196
Data Transfer and Frame Format..................................................................... 197
Multi-master Bus Systems, Arbitration and Synchronization............................ 200
Overview of the TWI Module ............................................................................ 202
TWI Register Description.................................................................................. 205
Using the TWI................................................................................................... 208
Transmission Modes......................................................................................... 212
Multi-master Systems and Arbitration............................................................... 225
Analog Comparator ......................................................................... 227
Analog Comparator Multiplexed Input .............................................................. 229
Analog to Digital Converter ............................................................ 230
Features............................................................................................................ 230
iii
2490G–AVR–03/04
Operation.......................................................................................................... 231
Starting a Conversion ....................................................................................... 232
Prescaling and Conversion Timing................................................................... 233
Changing Channel or Reference Selection ...................................................... 236
ADC Noise Canceler......................................................................................... 238
ADC Conversion Result.................................................................................... 242
JTAG Interface and On-chip Debug System ................................. 248
Features............................................................................................................ 248
Overview........................................................................................................... 248
Test Access Port – TAP.................................................................................... 248
TAP Controller .................................................................................................. 250
Using the Boundary -scan Chain...................................................................... 251
Using the On-chip Debug system..................................................................... 251
On-chip Debug Specific JTAG Instructions ...................................................... 252
On-chip Debug Related Register in I/O Memory .............................................. 253
Using the JTAG Programming Capabilities ...................................................... 253
Bibliography...................................................................................................... 253
IEEE 1149.1 (JTAG) Boundary-scan .............................................. 254
Features............................................................................................................ 254
System Overview.............................................................................................. 254
Data Registers.................................................................................................. 254
Boundary-scan Specific JTAG Instructions ...................................................... 256
Boundary-scan Related Register in I/O Memory .............................................. 258
Boundary-scan Chain ....................................................................................... 258
ATmega64 Boundary-scan Order..................................................................... 270
Boundary-scan Description Language Files..................................................... 276
Boot Loader Support – Read-While-Write Self-programming..... 277
Features............................................................................................................ 277
Application and Boot Loader Flash Sections.................................................... 277
Read-While-Write and No Read-While-Write Flash Sections........................... 277
Boot Loader Lock Bits....................................................................................... 279
Entering the Boot Loader Program................................................................... 281
Addressing the Flash During Self-programming............................................... 283
Self-programming the Flash ............................................................................. 284
Memory Programming..................................................................... 290
Program and Data Memory Lock Bits............................................................... 290
Fuse Bits........................................................................................................... 291
Signature Bytes ................................................................................................ 293
Calibration Byte ................................................................................................ 293
Parallel Programming Parameters, Pin Mapping, and Commands .................. 293
Parallel Programming ....................................................................................... 297
Serial Downloading........................................................................................... 306
iv
ATmega64(L)
2490G–AVR–03/04
ATmega64(L)
SPI Serial Programming Pin Mapping .............................................................. 307
Programming Via the JTAG Interface............................................................... 312
Electrical Characteristics................................................................ 326
Absolute Maximum Ratings*............................................................................. 326
DC Characteristics............................................................................................ 326
External Clock Drive Waveforms...................................................................... 328
External Clock Drive ......................................................................................... 328
Two-wire Serial Interface Characteristics ......................................................... 329
SPI Timing Characteristics ............................................................................... 331
ADC Characteristics – Preliminary Data........................................................... 333
External Data Memory Timing .......................................................................... 336
ATmega64 Typical Characteristics – Preliminary Data................ 341
Register Summary........................................................................... 342
Instruction Set Summary ................................................................ 345
Ordering Information....................................................................... 348
Packaging Information.................................................................... 349
64A ................................................................................................................... 349
64M1................................................................................................................. 350
Errata ................................................................................................ 351
ATmega64, all rev............................................................................................. 351
Datasheet Change Log for ATmega64........................................... 353
Changes from Rev. 2490F-12/03 to Rev. 2490G-03/04................................... 353
Changes from Rev. 2490E-09/03 to Rev. 2490F-12/03 ................................... 353
Changes from Rev. 2490D-02/03 to Rev. 2490E-09/03................................... 353
Changes from Rev. 2490C-09/02 to Rev. 2490D-02/03................................... 353
Changes from Rev. 2490B-09/02 to Rev. 2490C-09/02................................... 354
Changes from Rev. 2490A-10/01 to Rev. 2490B-09/02 ................................... 354
Table of Contents ................................................................................. i
v
2490G–AVR–03/04
vi
ATmega64(L)
2490G–AVR–03/04
Atmel Corporation
Atmel Operations
2325 Orchard Parkway
San Jose, CA 95131, USA
Tel: 1(408) 441-0311
Fax: 1(408) 487-2600
Memory
RF/Automotive
Theresienstrasse 2
Postfach 3535
74025 Heilbronn, Germany
Tel: (49) 71-31-67-0
Fax: (49) 71-31-67-2340
2325 Orchard Parkway
San Jose, CA 95131, USA
Tel: 1(408) 441-0311
Fax: 1(408) 436-4314
Regional Headquarters
Microcontrollers
2325 Orchard Parkway
San Jose, CA 95131, USA
Tel: 1(408) 441-0311
Fax: 1(408) 436-4314
1150 East Cheyenne Mtn. Blvd.
Colorado Springs, CO 80906, USA
Tel: 1(719) 576-3300
Europe
Atmel Sarl
Route des Arsenaux 41
Case Postale 80
CH-1705 Fribourg
Switzerland
Tel: (41) 26-426-5555
Fax: (41) 26-426-5500
Fax: 1(719) 540-1759
Biometrics/Imaging/Hi-Rel MPU/
High Speed Converters/RF Datacom
Avenue de Rochepleine
La Chantrerie
BP 70602
44306 Nantes Cedex 3, France
Tel: (33) 2-40-18-18-18
Fax: (33) 2-40-18-19-60
BP 123
38521 Saint-Egreve Cedex, France
Tel: (33) 4-76-58-30-00
Fax: (33) 4-76-58-34-80
Asia
Room 1219
Chinachem Golden Plaza
77 Mody Road Tsimshatsui
East Kowloon
Hong Kong
Tel: (852) 2721-9778
Fax: (852) 2722-1369
ASIC/ASSP/Smart Cards
Zone Industrielle
13106 Rousset Cedex, France
Tel: (33) 4-42-53-60-00
Fax: (33) 4-42-53-60-01
1150 East Cheyenne Mtn. Blvd.
Colorado Springs, CO 80906, USA
Tel: 1(719) 576-3300
Japan
9F, Tonetsu Shinkawa Bldg.
1-24-8 Shinkawa
Chuo-ku, Tokyo 104-0033
Japan
Tel: (81) 3-3523-3551
Fax: (81) 3-3523-7581
Fax: 1(719) 540-1759
Scottish Enterprise Technology Park
Maxwell Building
East Kilbride G75 0QR, Scotland
Tel: (44) 1355-803-000
Fax: (44) 1355-242-743
Literature Requests
www.atmel.com/literature
Disclaimer: Atmel Corporation makes no warranty for the use of its products, other than those expressly contained in the Company’s standard
warranty which is detailed in Atmel’s Terms and Conditions located on the Company’s web site. The Company assumes no responsibility for any
errors which may appear in this document, reserves the right to change devices or specifications detailed herein at any time without notice, and
does not make any commitment to update the information contained herein. No licenses to patents or other intellectual property of Atmel are
granted by the Company in connection with the sale of Atmel products, expressly or by implication. Atmel’s products are not authorized for use
as critical components in life support devices or systems.
© Atmel Corporation 2003. All rights reserved. Atmel® and combinations thereof, AVR®, and AVR Studio® are the registered trademarks of
Atmel Corporation or its subsidiaries. Microsoft®, Windows®, Windows NT®, and Windows XP® are the registered trademarks of Microsoft Corpo-
ration. Other terms and product names may be the trademarks of others
Printed on recycled paper.
2490G–AVR–03/04
相关型号:
ATMEGA64L-8AJ
RISC Microcontroller, 8-Bit, FLASH, 8MHz, CMOS, PQFP64, 14 X 14 MM, 1 MM HEIGHT, 0.80 MM PITCH, PLASTIC, MS-026AEB, TQFP-64
ATMEL
ATMEGA64L-8AL
RISC Microcontroller, 8-Bit, FLASH, 8MHz, CMOS, PQFP64, 14 X 14 MM, 1 MM HEIGHT, 0.80 MM PITCH, PLASTIC, MS-026AEB, TQFP-64
ATMEL
ATMEGA64L-8ML
RISC Microcontroller, 8-Bit, FLASH, 8MHz, CMOS, 9 X 9 MM, 1 MM HEIGHT, 0.50 MM PITCH, MO-220VMMD, MLF-64
ATMEL
©2020 ICPDF网 联系我们和版权申明