ATMEL ATMEGA64-16AU 8-bit microcontroller with 64k bytes in-system programmable flash Datasheet

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
High Endurance Non-volatile Memory segments
– 64K Bytes of In-System Reprogrammable Flash program memory
– 2K Bytes EEPROM
– 4K Bytes Internal SRAM
– Write/Erase Cycles: 10,000 Flash/100,000 EEPROM
– Data retention: 20 years at 85°C/100 years at 25°C(1)
– Optional Boot Code Section with Independent Lock Bits
In-System Programming by On-chip Boot Program
True Read-While-Write Operation
– 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
– 6 PWM Channels with Programmable Resolution from 1 to 16 Bits
– 8-channel, 10-bit ADC
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 QFN/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
8-bit
Microcontroller
with 64K Bytes
In-System
Programmable
Flash
ATmega64
ATmega64L
Figure 1. Pinout ATmega64
TQFP/MLF
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
PA3 (AD3)
PA4 (AD4)
PA5 (AD5)
PA6 (AD6)
PA7 (AD7)
PG2(ALE)
PC7 (A15)
PC6 (A14)
PC5 (A13)
PC4 (A12)
PC3 (A11)
PC2 (A10
PC1 (A9)
PC0 (A8)
PG1(RD)
PG0(WR)
(OC2/OC1C) PB7
TOSC2/PG3
TOSC1/PG4
RESET
VCC
GND
XTAL2
XTAL1
(SCL/INT0) PD0
(SDA/INT1) PD1
(RXD1/INT2) PD2
(TXD1/INT3) PD3
(ICP1) PD4
(XCK1) PD5
(T1) PD6
(T2) PD7
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
PEN
RXD0/(PDI) PE0
(TXD0/PDO) PE1
(XCK0/AIN0) PE2
(OC3A/AIN1) PE3
(OC3B/INT4) PE4
(OC3C/INT5) PE5
(T3/INT6) PE6
(ICP3/INT7) PE7
(SS) PB0
(SCK) PB1
(MOSI) PB2
(MISO) PB3
(OC0) PB4
(OC1A) PB5
(OC1B) PB6
64
63
62
61
60
59
58
57
56
55
54
53
52
51
50
49
AVCC
GND
AREF
PF0 (ADC0)
PF1 (ADC1)
PF2 (ADC2)
PF3 (ADC3)
PF4 (ADC4/TCK)
PF5 (ADC5/TMS)
PF6 (ADC6/TDO)
PF7 (ADC7/TDI)
GND
VCC
PA0 (AD0)
PA1 (AD1)
PA2 (AD2)
Pin
Configuration
Note:
Disclaimer
2
The bottom pad under the QFN/MLF package should be soldered to ground.
Typical values contained in this data sheet are based on simulations and characterization of
other AVR microcontrollers manufactured on the same process technology. Min and Max values
will be available after the device is characterized.
ATmega64(L)
2490N–AVR–05/08
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 DIR.
REG. PORTF
DATA REGISTER
PORTF
DATA REGISTER
PORTA
DATA DIR.
REG. PORTA
DATA DIR.
REG. PORTC
DATA REGISTER
PORTC
8-BIT DATA BUS
XTAL1
AREF
CALIB. OSC
INTERNAL
OSCILLATOR
ADC
XTAL2
OSCILLATOR
PROGRAM
COUNTER
STACK
POINTER
WATCHDOG
TIMER
ON-CHIP DEBUG
PROGRAM
FLASH
SRAM
MCU CONTROL
REGISTER
BOUNDARYSCAN
INSTRUCTION
REGISTER
JTAG TAP
OSCILLATOR
TIMING AND
CONTROL
RESET
PEN
PROGRAMMING
LOGIC
INSTRUCTION
DECODER
CONTROL
LINES
TIMER/
COUNTERS
GENERAL
PURPOSE
REGISTERS
X
Y
Z
INTERRUPT
UNIT
ALU
EEPROM
STATUS
REGISTER
SPI
+
-
ANALOG
COMPARATOR
USART0
DATA REGISTER
PORTE
DATA DIR.
REG. PORTE
PORTE DRIVERS
PE0 - PE7
DATA REGISTER
PORTB
DATA DIR.
REG. PORTB
PORTB DRIVERS
PB0 - PB7
USART1
2-WIRE SERIAL
INTERFACE
DATA REGISTER
PORTD
DATA DIR.
REG. PORTD
DATA REG. DATA DIR.
PORTG REG. PORTG
PORTD DRIVERS
PORTG DRIVERS
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
2490N–AVR–05/08
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 general 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 oriented 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 asynchronous 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 asynchronous 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. Software 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-Circuit 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 notes “Replacing ATmega103 by ATmega128”
and “Migration between ATmega64 and ATmega128” describes what the user should be aware
of replacing the ATmega103 by an ATmega128 or ATmega64.
4
ATmega64(L)
2490N–AVR–05/08
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. However, 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.
GND
Ground.
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
73.
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
74.
5
2490N–AVR–05/08
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 77. 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 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
78.
Port E (PE7..PE0)
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
81.
Port F (PF7..PF0)
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 resistors 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)
2490N–AVR–05/08
ATmega64(L)
RESET
Reset input. A low level on this pin for longer than the minimum pulse length will generate a
reset, even if the clock is not running. The minimum pulse length is given in Table 19 on page
52. Shorter pulses are not guaranteed to generate a reset.
XTAL1
Input to the inverting Oscillator amplifier and input to the internal clock operating circuit.
XTAL2
Output from the inverting Oscillator amplifier.
AVCC
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 connected to VCC
through a low-pass filter.
AREF
AREF is the analog reference pin for the A/D Converter.
PEN
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 is internally pulled high. The pullup is shown in Figure 22 on page 52 and its value is given in Section
“DC Characteristics” on page 325. PEN has no function during normal operation.
7
2490N–AVR–05/08
Resources
A comprehensive set of development tools, application notes and datasheetsare available for
download on http://www.atmel.com/avr.
Note:
Data Retention
8
1.
Reliability Qualification results show that the projected data retention failure rate is much less
than 1 PPM over 20 years at 85°C or 100 years at 25°C.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 definitions in the header files and
interrupt handling in C is compiler dependent. Please confirm with the C compiler documentation
for more details.
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”.
9
2490N–AVR–05/08
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
Flash
Program
Memory
Program
Counter
Status
and Control
32 x 8
General
Purpose
Registrers
Control Lines
Direct Addressing
Instruction
Decoder
Indirect Addressing
Instruction
Register
Interrupt
Unit
SPI
Unit
Watchdog
Timer
ALU
Analog
Comparator
I/O Module1
Data
SRAM
I/O Module 2
I/O Module n
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 executed, the next instruction is pre-fetched from the program memory. This concept enables instructions to be executed
in every clock cycle. The program memory is In-System 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.
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
10
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
can also be used as an address pointer for look up tables in Flash program 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 constant 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-functions. Some implementations of the
architecture also provide a powerful multiplier supporting both signed/unsigned multiplication
and fractional format. See the “Instruction Set” section for a detailed description.
11
2490N–AVR–05/08
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.
SREG – AVR Status
Register
The AVR Status Register – SREG – is defined as:
Bit
7
6
5
4
3
2
1
0
0x3F (0x5F)
I
T
H
S
V
N
Z
C
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
SREG
• Bit 7 – I: Global Interrupt Enable
The Global Interrupt Enable bit must be set for the interrupts to be enabled. The individual interrupt enable control is then performed in separate control registers. If the Global Interrupt Enable
Register is cleared, 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 Complement
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.
12
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
• Bit 0 – C: Carry Flag
The Carry Flag C indicates a carry in an arithmetic or logic operation. See the “Instruction 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.
R0
0x00
R1
0x01
R2
0x02
…
R13
0x0D
General
R14
0x0E
Purpose
R15
0x0F
Working
R16
0x10
Registers
R17
0x11
…
R26
0x1A
X-register Low Byte
R27
0x1B
X-register High Byte
R28
0x1C
Y-register Low Byte
R29
0x1D
Y-register High Byte
R30
0x1E
Z-register Low Byte
R31
0x1F
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 physically 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.
13
2490N–AVR–05/08
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
X - register
XH
XL
7
0
R27 (0x1B)
YH
YL
7
0
R29 (0x1D)
0
7
0
R28 (0x1C)
15
Z - register
0
R26 (0x1A)
15
Y - register
0
7
ZH
ZL
7
0
R31 (0x1F)
0
7
0
R30 (0x1E)
In the different addressing modes these address registers have functions as fixed displacement,
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 Register 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 interrupt
Stacks are located. This Stack space in the data SRAM must be defined by the program before
any subroutine calls are executed or interrupts are enabled. The Stack Pointer must be set to
point above 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 number 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
14
13
12
11
10
9
8
0x3E (0x5E)
SP15
SP14
SP13
SP12
SP11
SP10
SP9
SP8
SPH
0x3D (0x5D)
SP7
SP6
SP5
SP4
SP3
SP2
SP1
SP0
SPL
7
6
5
4
3
2
1
0
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Read/Write
Initial Value
Instruction
Execution Timing
14
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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 pipelining concept
to obtain up to 1 MIPS per MHz with the corresponding unique results for functions per cost,
functions per clocks, and functions per power-unit.
Figure 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 61. 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 61 for more information.
The Reset Vector can also be 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.
15
2490N–AVR–05/08
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 interrupts. 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 remembered 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 corresponding 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 disappears 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 execute one
more instruction before any pending interrupt is served.
Note that the Status Register is not automatically stored when entering an interrupt routine, 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 simultaneously 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
cli
; store SREG value
; disable interrupts during timed sequence
sbi EECR, EEMWE
; start EEPROM write
sbi EECR, EEWE
out SREG, r16
; 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) */
16
ATmega64(L)
2490N–AVR–05/08
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.
17
2490N–AVR–05/08
AVR 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 memory 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 organized 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 contains 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 Execution Timing” on page 14.
Figure 8. Program Memory Map
$0000
Application Flash Section
Boot Flash Section
$7FFF
18
ATmega64(L)
2490N–AVR–05/08
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
Configuration
Internal SRAM
Data Memory
External SRAM
Data Memory
Normal mode
4096
up to 64K
ATmega103 compatibility mode
4000
up to 64K
Figure 9 on page 20 shows how the ATmega64 SRAM Memory is organized.
The ATmega64 is a complex microcontroller with more peripheral units than can be supported
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 27 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.
The five different addressing modes for the data memory cover: Direct, Indirect with Displacement, Indirect, Indirect with Pre-decrement, and Indirect with Post-increment. In the Register
File, registers R26 to R31 feature the indirect addressing pointer registers.
The direct addressing reaches the entire data space.
19
2490N–AVR–05/08
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 Registers, 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
13.
Figure 9. Data Memory Map
Memory Configuration A
Data Memory
32 Registers
64 I/O Registers
160 Ext I/O Reg.
Memory Configuration B
Data Memory
$0000 - $001F
$0020 - $005F
$0060 - $00FF
$0100
Internal SRAM
(4096 x 8)
32 Registers
64 I/O Registers
Internal SRAM
(4000 x 8)
$0FFF
$1000
$10FF
$1100
External SRAM
(0 - 64K x 8)
External SRAM
(0 - 64K x 8)
$FFFF
20
$0000 - $001F
$0020 - $005F
$0060
$FFFF
ATmega64(L)
2490N–AVR–05/08
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
Compute Address
Address Valid
Write
Data
WR
Read
Data
RD
Memory Access Instruction
EEPROM Data
Memory
Next Instruction
The ATmega64 contains 2K bytes of data EEPROM memory. It is organized as a separate data
space, in which single bytes can be read and written. The EEPROM has an endurance of at
least 100,000 write/erase cycles. The access between the EEPROM and the CPU is described
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 programming
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 24. 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 Corruption” on page 26. for details on how to avoid problems in
these situations.
In order to prevent unintentional EEPROM writes, a specific write procedure must be followed.
Refer to the description of the EEPROM Control Register for details on this.
When the EEPROM is 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.
21
2490N–AVR–05/08
EEARH and EEARL –
EEPROM Address
Register
Bit
15
14
13
12
11
10
9
8
0x1F (0x3F)
–
–
–
–
–
EEAR10
EEAR9
EEAR8
EEARH
0x1E (0x3E)
EEAR7
EEAR6
EEAR5
EEAR4
EEAR3
EEAR2
EEAR1
EEAR0
EEARL
7
6
5
4
3
2
1
0
Read/Write
Initial Value
R
R
R
R
R
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
0
0
0
0
0
X
X
X
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 location,
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.
EEDR – EEPROM Data
Register
Bit
7
6
5
4
3
2
1
0
0x1D (0x3D)
MSB
LSB
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
EEDR
• Bits 7..0 – EEDR7.0: EEPROM Data
For the EEPROM write operation, the EEDR Register contains the data to be written to the
EEPROM in the address given by the EEAR Register. For the EEPROM read operation, the
EEDR contains the data read out from the EEPROM at the address given by EEAR.
EECR – EEPROM
Control Register
Bit
7
6
5
4
3
2
1
0
0x1C (0x3C)
–
–
–
–
EERIE
EEMWE
EEWE
EERE
Read/Write
R
R
R
R
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
X
0
EECR
• 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.
22
ATmega64(L)
2490N–AVR–05/08
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.
23
2490N–AVR–05/08
Table 2. EEPROM Programming Time(1)
Symbol
Number of Calibrated RC
Oscillator Cycles
Typ Programming Time
8448
8.4 ms
EEPROM write (from CPU)
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 interrupts 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);
}
24
ATmega64(L)
2490N–AVR–05/08
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-down Sleep
Mode
When entering Power-down Sleep mode while an EEPROM write operation is active, the
EEPROM write operation will continue, and will complete before the Write Access time has
passed. However, when the write operation is completed, the 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 completed before entering Power-down.
25
2490N–AVR–05/08
Preventing EEPROM
Corruption
During periods of low VCC, the EEPROM data can be corrupted because the supply voltage is
too low for the CPU and the EEPROM to operate properly. These issues are the same as for
board level systems using 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 voltage. 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 370.
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 instructions. 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 registers 0x00 to 0x1F only.
The I/O and peripherals control registers are explained in later sections.
26
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
External Memory
Interface
Overview
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).
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 73, Table 33 on page 77, and Table 45 on page 85). 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
Note:
0xFFFF
1. ATmega64 in non ATmega103 compatibility mode: Memory Configuration A is available (Memory Configuration B N/A).
ATmega64 in mega103 compatibility mode: Memory Configuration B is available (Memory
Configuration A N/A).
27
2490N–AVR–05/08
ATmega103
Compatibility
Using the External
Memory Interface
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).
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 66. 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 normal 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 137 to Table 144 on page 337). 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 before G low (tsu) must not exceed
address valid to ALE low (tAVLLC) minus PCB wiring delay (dependent on the capacitive load).
28
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 12. External SRAM Connected to the AVR
D[7:0]
AD7:0
D
ALE
G
AVR
A15:8
RD
WR
Pull-up and Bus
Keeper
Q
A[7:0]
SRAM
A[15:8]
RD
WR
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 disable 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 “XMCRB – External Memory Control Register
B” on page 34. When enabled, the Bus Keeper will ensure a defined logic level (zero or one) on
the AD7:0 bus when these lines would otherwise be tri-stated by the XMEM interface.
Timing
External memory devices have different timing requirements. To meet these requirements, 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 waitstate. 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 137 to Table 144 on
page 337). The different wait states are set up in software. As an additional feature, it is possible
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 137 to Table 144.
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.
29
2490N–AVR–05/08
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
Address
DA7:0 (XMBK = 0)
Prev. data
Address
DA7:0 (XMBK = 1)
Prev. data
Address
XX
Write
Address
Data
WR
XXXXX
Data
XXXXXXXX
Read
Data
RD
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
Address
DA7:0 (XMBK = 0)
Prev. Data
Address
DA7:0 (XMBK = 1)
Prev. Data
XX
Data
Write
Address
WR
Data
Read
Address
Data
RD
Note:
30
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).
ATmega64(L)
2490N–AVR–05/08
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
Address
DA7:0 (XMBK = 0)
Prev. Data
Address
DA7:0 (XMBK = 1)
Prev. Data
XX
Write
Address
Data
WR
Address
Read
Data
Data
RD
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
Address XX
DA7:0 (XMBK = 0)
Prev. Data
Address
DA7:0 (XMBK = 1)
Prev. Data
Write
Address
Data
WR
Data
Read
Address
Data
RD
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).
31
2490N–AVR–05/08
XMEM Register
Description
MCUCR – MCU
Control Register
Bit
7
6
5
4
3
2
1
0
0x35 (0x55)
SRE
SRW10
SE
SM1
SM0
SM2
IVSEL
IVCE
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
MCUCR
• 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 overrides 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 settings are used.
• Bit 6 – SRW10: Wait State Select Bit
For a detailed description in non ATmega103 compatibility mode, see common description 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.
XMCRA – External
Memory Control
Register A
Bit
7
6
5
4
3
2
1
0
(0x6D)
–
SRL2
SRL1
SRL0
SRW01
SRW00
SRW11
–
Read/Write
R
R/W
R/W
R/W
R/W
R/W
R/W
R
Initial Value
0
0
0
0
0
0
0
0
XMCRA
• 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.
32
ATmega64(L)
2490N–AVR–05/08
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
1
Lower sector = 0x1100 - 0x1FFF
Upper sector = 0x2000 - 0xFFFF
0
1
0
Lower sector = 0x1100 - 0x3FFF
Upper sector = 0x4000 - 0xFFFF
0
1
1
Lower sector = 0x1100 - 0x5FFF
Upper sector = 0x6000 - 0xFFFF
1
0
0
Lower sector = 0x1100 - 0x7FFF
Upper sector = 0x8000 - 0xFFFF
1
0
1
Lower sector = 0x1100 - 0x9FFF
Upper sector = 0xA000 - 0xFFFF
1
1
0
Lower sector = 0x1100 - 0xBFFF
Upper sector = 0xC000 - 0xFFFF
1
1
1
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
0
0
No wait states
0
1
Wait one cycle during read/write strobe
1
0
Wait two cycles during read/write strobe
1
1
Wait two cycles during read/write and wait one cycle before driving out
new address
Note:
Wait States
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.
33
2490N–AVR–05/08
XMCRB – External
Memory Control
Register B
Bit
7
6
5
4
3
2
1
0
XMBK
–
–
–
–
XMM2
XMM1
XMM0
Read/Write
R/W
R
R
R
R
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
(0x6C)
XMCRB
• 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, it will ensure a defined logic level (zero or one) on AD7:0 when they would otherwise
be tri-stated. 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 location,
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 memory, 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 36, 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
34
XMM2
XMM1
XMM0
# Bits for External Memory Address
Released Port Pins
0
0
0
8 (Full 60 KB space)
None
0
0
1
7
PC7
0
1
0
6
PC7 - PC6
0
1
1
5
PC7 - PC5
1
0
0
4
PC7 - PC4
1
0
1
3
PC7 - PC3
1
1
0
2
PC7 - PC2
1
1
1
No Address high bits
Full Port C
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Using all Locations of
External Memory
Smaller than 64 KB
Since the external memory is mapped after the internal memory as shown in Figure 11, the
external memory is not addressed when addressing the first 4,352 bytes of data 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 external 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 location 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
0x0000
External 32K SRAM
AVR Memory Map
0x0000
Internal Memory
0x10FF
0x1100
0x7FFF
0x8000
0x10FF
0x1100
External
Memory
0x90FF
0x9100
0x7FFF
0x0000
0x0FFF
0x1000
0x7FFF
0x8000
0x0000
Internal Memory
External
0x0FFF
0x1000
0x7FFF
Memory
0x8FFF
0x9000
(Unused)
0xFFFF
External 32K SRAM
(Unused)
0xFFFF
35
2490N–AVR–05/08
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
r16, 0xFF
out
DDRC, r16
ldi
r16, 0x00
out
PORTC, r16
; release PC7:5
ldi
r16, (1<<XMM1)|(1<<XMM0)
sts
XMCRB, r16
; write 0xAA to address 0x0001 of external
; memory
ldi
r16, 0xaa
sts
0x0001+OFFSET, r16
; re-enable PC7:5 for external memory
ldi
r16, (0<<XMM1)|(0<<XMM0)
sts
XMCRB, r16
; store 0x55 to address (OFFSET + 1) of
; external memory
ldi
r16, 0x55
sts
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. See “About Code Examples” on page 9.
Care must be exercised using this option as most of the memory is masked away.
36
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
System Clock
and Clock
Options
Clock Systems
and their
Distribution
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 clocks to modules
not being used can be halted by using different sleep modes, as described in “Power Management and Sleep Modes” on page 46. The clock systems are detailed below.
Figure 18. Clock Distribution
Asynchronous
Timer/Counter
General I/O
Modules
ADC
CPU Core
RAM
Flash and
EEPROM
clkADC
clkI/O
clkCPU
AVR Clock
Control Unit
clkASY
clkFLASH
Reset Logic
Source Clock
Watchdog Timer
Watchdog Clock
Clock
Multiplexer
Timer/Counter
Oscillator
External RC
Oscillator
External Clock
Watchdog
Oscillator
Crystal
Oscillator
Low-frequency
Crystal Oscillator
Calibrated RC
Oscillator
CPU Clock – clkCPU
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 Register and the
data memory holding the Stack Pointer. Halting the CPU clock inhibits the core from performing
general operations and calculations.
I/O Clock – clkI/O
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 reception 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.
37
2490N–AVR–05/08
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 accurate 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
CKSEL3..0
External Crystal/Ceramic Resonator
1111 - 1010
External Low-frequency Crystal
1001
External RC Oscillator
1000 - 0101
Calibrated Internal RC Oscillator
0100 - 0001
External Clock
Note:
0000
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 startup, 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 Oscillator is voltage dependent as shown in the “Typical Characteristics” on page 340.
Table 7. Number of Watchdog Oscillator Cycles
38
Typ Time-out (VCC = 5.0V)
Typ Time-out (VCC = 3.0V)
Number of Cycles
4.1 ms
4.3 ms
4K (4,096)
65 ms
69 ms
64K (65,536)
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
XDIV – XTAL Divide
Control Register
The XTAL Divide Control Register is used to divide the source clock frequency by a 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
6
5
4
3
2
1
0
XDIVEN
XDIV6
XDIV5
XDIV4
XDIV3
XDIV2
XDIV1
XDIV0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
0x3C (0x5C)
XDIV
• 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:
clockf CLK = Source
--------------------------------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 division 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.
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 setting ensures that all
users can make their desired clock source setting using an In-System or Parallel Programmer.
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 different Oscillator
amplifier modes. When CKOPT is programmed, the Oscillator output will oscillate a full rail-torail 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 environment. 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.
39
2490N–AVR–05/08
Figure 19. Crystal Oscillator Connections
C2
C1
XTAL2
XTAL1
GND
The Oscillator can operate in three different modes, each optimized for a specific frequency
range. The operating mode is selected by the fuses CKSEL3..1 as shown in Table 8.
Table 8. Crystal Oscillator Operating Modes
CKOPT
CKSEL3..1
Frequency Range
(MHz)
Recommended Range for Capacitors
C1 and C2 for Use with Crystals (pF)
1
101(1)
0.4 - 0.9
–
1
110
0.9 - 3.0
12 - 22
1
111
3.0 - 8.0
12 - 22
0
101, 110, 111
1.0 -
12 - 22
Note:
1. 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.
40
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 9. Start-up Times for the Crystal Oscillator Clock Selection
CKSEL0
SUT1..0
Start-up Time from
Power-down and
Power-save
0
00
258 CK(1)
4.1 ms
Ceramic resonator, fast
rising power
0
01
258 CK(1)
65 ms
Ceramic resonator,
slowly rising power
0
10
1K CK(2)
–
Ceramic resonator, BOD
enabled
0
11
1K CK(2)
4.1 ms
Ceramic resonator, fast
rising power
1
00
1K CK(2)
65 ms
Ceramic resonator,
slowly rising power
1
01
16K CK
–
Crystal Oscillator, BOD
enabled
1
10
16K CK
4.1 ms
Crystal Oscillator, fast
rising power
1
11
16K CK
65 ms
Crystal Oscillator, slowly
rising power
Notes:
Low-frequency
Crystal Oscillator
Additional Delay
from Reset
(VCC = 5.0V)
Recommended Usage
1. These options should only be used when not operating close to the maximum frequency 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 frequency 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.
To use a 32.768 kHz watch crystal as the clock source for the device, the Low-frequency 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
SUT1..0
Start-up Time
from Power-down
and Power-save
Additional Delay
from Reset
(VCC = 5.0V)
Recommended Usage
1K CK
(1)
4.1 ms
Fast rising power or BOD enabled
01
1K CK
(1)
65 ms
Slowly rising power
10
32K CK
65 ms
Stable frequency at start-up
00
11
Note:
Reserved
1. These options should only be used if frequency stability at start-up is not important for the
application.
41
2490N–AVR–05/08
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.
Figure 20. External RC Configuration
VCC
R
NC
XTAL2
XTAL1
C
GND
The Oscillator can operate in four different modes, each optimized for a specific frequency
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
Frequency Range (MHz)
0101
0.1 - 0.9
0110
0.9 - 3.0
0111
3.0 - 8.0
1000
8.0 - 12.0
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
SUT1..0
Start-up Time from
Power-down and
Power-save
Additional Delay
from Reset
(VCC = 5.0V)
00
18 CK
–
01
18 CK
4.1 ms
Fast rising power
10
18 CK
65 ms
Slowly rising power
11
(1)
4.1 ms
Fast rising power or BOD enabled
Note:
42
6 CK
Recommended Usage
BOD enabled
1. This option should not be used when operating close to the maximum frequency of the device.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Calibrated Internal 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 system clock by
RC Oscillator
programming the CKSEL Fuses as shown in Table 13. If selected, it will operate with no external
components. The CKOPT Fuse should always be unprogrammed 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% accuracy 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
Note:
CKSEL3..0
Nominal Frequency (MHz)
0001(1)
1.0
0010
2.0
0011
4.0
0100
8.0
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
SUT1..0
Start-up Time from Powerdown and Power-save
Additional Delay from
Reset (VCC = 5.0V)
00
6 CK
–
01
6 CK
4.1 ms
Fast rising power
6 CK
65 ms
Slowly rising power
(1)
10
11
Note:
OSCCAL – Oscillator
Calibration Register(1)
BOD enabled
Reserved
1. The device is shipped with this option selected.
Bit
(0x6F)
Read/Write
Initial Value
Note:
Recommended Usage
7
6
5
4
3
2
1
0
CAL7
CAL6
CAL5
CAL4
CAL3
CAL2
CAL1
CAL0
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
OSCCAL
Device Specific Calibration Value
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 process 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 value can be read by software and loaded into
the OSCCAL Register. When OSCCAL is zero, the lowest available frequency is chosen. Writing
43
2490N–AVR–05/08
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
External Clock
OSCCAL Value
Min Frequency in Percentage of
Nominal Frequency (%)
Max Frequency in Percentage of
Nominal Frequency (%)
0x00
50
100
0x7F
75
150
0xFF
100
200
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 programmed 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
SUT1..0
Start-up Time from Powerdown and Power-save
Additional Delay from
Reset (VCC = 5.0V)
00
6 CK
–
01
6 CK
4.1 ms
Fast rising power
10
6 CK
65 ms
Slowly rising power
11
Recommended Usage
BOD enabled
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 required to ensure that the
MCU is kept in Reset during such changes in the clock frequency.
44
ATmega64(L)
2490N–AVR–05/08
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.
Note:
The Timer/Counter Oscillator uses the same type of crystal oscillator as Low-Frequency Oscillator
and the internal capacitors have the same nominal value of 36 pF.
45
2490N–AVR–05/08
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 instruction 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 37 presents the different clock systems in the ATmega64, and their distribution. This figure is helpful in selecting an appropriate sleep mode.
MCUCR – MCU
Control Register
The MCU Control Register contains control bits for power management.
Bit
7
6
5
4
3
2
1
0
0x35 (0x55)
SRE
SRW10
SE
SM1
SM0
SM2
IVSEL
IVCE
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
MCUCR
• 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 waking 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
Note:
46
SM2
SM1
SM0
Sleep Mode
0
0
0
Idle
0
0
1
ADC Noise Reduction
0
1
0
Power-down
0
1
1
Power-save
1
0
0
Reserved
1
0
1
Reserved
1
1
0
Standby(1)
1
1
1
Extended Standby(1)
1. Standby mode and Extended Standby mode are only available with external crystals or
resonators.
ATmega64(L)
2490N–AVR–05/08
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 Status 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 interrupts, 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 measurements. 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 interrupt, 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 Powerdown 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 Interrupts” on page 90
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 38.
Power-save Mode
When the SM2..0 bits are written to 011, the SLEEP instruction makes the MCU enter Powersave 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 Overflow 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 recommended
instead of Power-save mode because the contents of the registers in the asynchronous timer
should be considered undefined after wake-up in Power-save mode if AS0 is 0.
This sleep mode basically halts all clocks except clkASY, allowing operation only of asynchronous
modules, including Timer/Counter0 if clocked asynchronously.
47
2490N–AVR–05/08
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
Timer
Osc
Enabled
I
N
TWI
Address
Match
Timer0
SPM/
EEPROM
Ready
A
Other
I/O
X
X
X
X(2)
X
X
X
X
X
X
X
X
X
X(2)
X(3)
X
X
X
X
X(3)
X
X(3)
X
X(3)
X
X(3)
X
Main
Clock
Source
Enabled
clkASY
X
ADC
Noise
Reduction
Powerdown
Powersave
X(2)
Standby(1)
Extended
Standby(1)
Notes:
48
Wake Up Sources
clkADC
Idle
Oscillators
clkIO
Sleep
Mode
clkFLASH
clkCPU
Active Clock Domains
X(2)
X
X(2)
X
X(2)
X(2)
X(2)
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.
ATmega64(L)
2490N–AVR–05/08
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 possible, 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
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 Converter” on page 230
for details on ADC operation.
Analog Comparator
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 reference 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 49 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 consuming 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 56 for details on the start-up time.
Watchdog Timer
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 56 for details on how to configure the Watchdog Timer.
Port Pins
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 70 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.
49
2490N–AVR–05/08
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.
50
ATmega64(L)
2490N–AVR–05/08
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 execution
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 Timeout 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 38.
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.
51
2490N–AVR–05/08
Figure 22. Reset Logic
DATA BUS
D
Q
L
Q
MCU Control and Status
Register (MCUCSR)
PORF
BORF
EXTRF
WDRF
JTRF
PEN
Pull-up Resistor
Power-On Reset
Circuit
Brown-Out
Reset Circuit
BODEN
BODLEVEL
Pull-up Resistor
SPIKE
FILTER
JTAG Reset
Register
Reset Circuit
COUNTER RESET
RESET
Watchdog
Timer
Watchdog
Oscillator
Clock
Generator
CK
Delay Counters
TIMEOUT
CKSEL[3:0]
SUT[1:0]
Table 19. Reset Characteristics
Symbol
VPOT
Condition
Min
Typ
Max
Units
Power-on Reset
Threshold Voltage
(rising)
1.4
2.3
V
Power-on Reset
Threshold Voltage
(falling)(1)
1.3
2.3
V
0.85 VCC
V
1.5
µs
VRST
RESET Pin Threshold
Voltage
tRST
Minimum pulse width on
RESET Pin
Brown-out Reset
Threshold Voltage(2)
BODLEVEL = 1
2.5
2.7
2.9
VBOT
BODLEVEL = 0
3.6
4.0
4.2
Minimum low voltage
period for Brown-out
Detection
BODLEVEL = 1
2
µs
tBOD
BODLEVEL = 0
2
µs
VHYST
Brown-out Detector
hysteresis
120
mV
Notes:
52
Parameter
0.2 VCC
V
1. The Power-on Reset will not work unless the supply voltage has been below VPOT (falling).
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 detection 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. Reaching 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
VCC
RESET
VPOT
VRST
tTOUT
TIME-OUT
INTERNAL
RESET
Figure 24. MCU Start-up, RESET Extended Externally
VCC
RESET
TIME-OUT
VPOT
VRST
tTOUT
INTERNAL
RESET
53
2490N–AVR–05/08
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 VBOT+ = 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
VCC
VBOT-
VBOT+
RESET
TIME-OUT
tTOUT
INTERNAL
RESET
54
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Watchdog Reset
When the Watchdog times out, it will generate a short reset pulse of one CK cycle duration. On
the falling edge of this pulse, the delay timer starts counting the Time-out period tTOUT. Refer to
page 56 for details on operation of the Watchdog Timer.
Figure 27. Watchdog Reset During Operation
CC
CK
MCUCSR – MCU
Control and Status
Register(1)
The MCU Control and Status Register provides information on which reset source caused an
MCU Reset.
Bit
7
6
5
4
3
2
1
0
0x34 (0x54)
JTD
–
–
JTRF
WDRF
BORF
EXTRF
PORF
Read/Write
R/W
R
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
Note:
MCUCSR
See Bit Description
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.
55
2490N–AVR–05/08
• 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 Comparator 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
Watchdog Timer
Parameter
Min
Typ
Max
Units
VBG
Bandgap reference voltage
1.15
1.23
1.35
V
tBG
Bandgap reference start-up time
40
70
µs
IBG
Bandgap reference current consumption
10
µA
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 VCC levels. By
controlling the Watchdog Timer prescaler, the Watchdog Reset interval can be adjusted as
shown in Table 22 on page 58. The WDR – Watchdog Reset – instruction resets the Watchdog
Timer. The Watchdog Timer is also reset when it is disabled 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 55.
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 60 for details.
56
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 21. WDT Configuration as a Function of the Fuse Settings of M103C and WDTON
Safety
Level
WDT
Initial
State
How to
Disable the
WDT
How to
Change
Time-out
M103C
WDTON
Unprogrammed
Unprogrammed
1
Disabled
Timed
sequence
Timed
sequence
Unprogrammed
Programmed
2
Enabled
Always
enabled
Timed
sequence
Programmed
Unprogrammed
0
Disabled
Timed
sequence
No restriction
Programmed
Programmed
2
Enabled
Always
enabled
Timed
sequence
Figure 28. Watchdog Timer
WATCHDOG
OSCILLATOR
WDTCR – Watchdog
Timer Control Register
Bit
7
6
5
4
3
2
1
0
0x21 (0x41)
–
–
–
WDCE
WDE
WDP2
WDP1
WDP0
Read/Write
R
R
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
WDTCR
• 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 60.
• 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
57
2490N–AVR–05/08
has logic level one. To disable an enabled Watchdog Timer, the following 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 algorithm
described above. See “Timed Sequences for Changing the Configuration of the Watchdog
Timer” on page 60.
• 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
58
WDP2
WDP1
WDP0
Number of WDT
Oscillator Cycles
Typical Time-out
at VCC = 3.0V
Typical Time-out
at VCC = 5.0V
0
0
0
16K (16,384)
17.1 ms
16.3 ms
0
0
1
32K (32,768)
34.3 ms
32.5 ms
0
1
0
64K (65,536)
68.5 ms
65 ms
0
1
1
128K (131,072)
0.14 s
0.13 s
1
0
0
256K (262,144)
0.27 s
0.26 s
1
0
1
512K (524,288)
0.55 s
0.52 s
1
1
0
1,024K (1,048,576)
1.1 s
1.0 s
1
1
1
2,048K (2,097,152)
2.2 s
2.1 s
ATmega64(L)
2490N–AVR–05/08
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:
; reset WDT
wdr
in r16, WDTCR
ldi r16, (1<<WDCE)|(1<<WDE)
; Write logical one to WDCE and WDE
ori
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)
{
/* Reset WDT*/
_WDRC();
/* Write logical one to WDCE and WDE */
WDTCR |= (1<<WDCE) | (1<<WDE);
/* Turn off WDT */
WDTCR = 0x00;
}
59
2490N–AVR–05/08
Timed Sequences
for Changing the
Configuration of
the Watchdog
Timer
The sequence for changing configuration differs slightly between the three safety levels. Separate procedures are described for each level.
Safety Level 0
This mode is compatible with the Watchdog operation found in ATmega103. The Watchdog
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 57 (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 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 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.
60
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Interrupts
Interrupt Vectors
in ATmega64
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 15.
Table 23. Reset and Interrupt Vectors
Vector
No.
1
Program
Address(2)
(1)
0x0000
Source
Interrupt Definition
RESET
External Pin, Power-on Reset, Brown-out Reset,
Watchdog Reset, and JTAG AVR Reset
2
0x0002
INT0
External Interrupt Request 0
3
0x0004
INT1
External Interrupt Request 1
4
0x0006
INT2
External Interrupt Request 2
5
0x0008
INT3
External Interrupt Request 3
6
0x000A
INT4
External Interrupt Request 4
7
0x000C
INT5
External Interrupt Request 5
8
0x000E
INT6
External Interrupt Request 6
9
0x0010
INT7
External Interrupt Request 7
10
0x0012
TIMER2 COMP
Timer/Counter2 Compare Match
11
0x0014
TIMER2 OVF
Timer/Counter2 Overflow
12
0x0016
TIMER1 CAPT
Timer/Counter1 Capture Event
13
0x0018
TIMER1 COMPA
Timer/Counter1 Compare Match A
14
0x001A
TIMER1 COMPB
Timer/Counter1 Compare Match B
15
0x001C
TIMER1 OVF
Timer/Counter1 Overflow
16
0x001E
TIMER0 COMP
Timer/Counter0 Compare Match
17
0x0020
TIMER0 OVF
Timer/Counter0 Overflow
18
0x0022
SPI, STC
SPI Serial Transfer Complete
19
0x0024
USART0, RX
USART0, Rx Complete
20
0x0026
USART0, UDRE
USART0 Data Register Empty
21
0x0028
USART0, TX
USART0, Tx Complete
22
0x002A
ADC
ADC Conversion Complete
23
0x002C
EE READY
EEPROM Ready
24
0x002E
ANALOG COMP
Analog Comparator
25
0x0030(3)
TIMER1 COMPC
Timer/Countre1 Compare Match C
26
(3)
TIMER3 CAPT
Timer/Counter3 Capture Event
(3)
TIMER3 COMPA
Timer/Counter3 Compare Match A
28
(3)
0x0036
TIMER3 COMPB
Timer/Counter3 Compare Match B
29
0x0038(3)
TIMER3 COMPC
Timer/Counter3 Compare Match C
30
(3)
TIMER3 OVF
Timer/Counter3 Overflow
(3)
USART1, RX
USART1, Rx Complete
27
31
0x0032
0x0034
0x003A
0x003C
61
2490N–AVR–05/08
Table 23. Reset and Interrupt Vectors (Continued)
Vector
No.
Program
Address(2)
32
Source
Interrupt Definition
0x003E(3)
USART1, UDRE
USART1 Data Register Empty
33
0x0040(3)
USART1, TX
USART1, Tx Complete
34
(3)
TWI
Two-wire Serial Interface
(3)
SPM READY
Store Program Memory Ready
0x0042
35
0x0044
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 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.
Table 24. Reset and Interrupt Vectors Placement(1)
BOOTRST
IVSEL
1
Note:
Reset Address
Interrupt Vectors Start Address
0
0x0000
0x0002
1
1
0x0000
Boot Reset Address + 0x0002
0
0
Boot Reset Address
0x0002
0
1
Boot Reset Address
Boot Reset Address + 0x0002
1. The Boot Reset Address is shown in Table 112 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:
Address Labels Code
62
Comments
0x0000
jmp
RESET
; Reset Handler
0x0002
jmp
EXT_INT0
; IRQ0 Handler
0x0004
jmp
EXT_INT1
; IRQ1 Handler
0x0006
jmp
EXT_INT2
; IRQ2 Handler
0x0008
jmp
EXT_INT3
; IRQ3 Handler
0x000A
jmp
EXT_INT4
; IRQ4 Handler
0x000C
jmp
EXT_INT5
; IRQ5 Handler
0x000E
jmp
EXT_INT6
; IRQ6 Handler
0x0010
jmp
EXT_INT7
; IRQ7 Handler
0x0012
jmp
TIM2_COMP
; Timer2 Compare Handler
0x0014
jmp
TIM2_OVF
; Timer2 Overflow Handler
0x0016
jmp
TIM1_CAPT
; Timer1 Capture Handler
0x0018
jmp
TIM1_COMPA
; Timer1 CompareA Handler
0x001A
jmp
TIM1_COMPB
; Timer1 CompareB Handler
0x001C
jmp
TIM1_OVF
; Timer1 Overflow Handler
0x001E
jmp
TIM0_COMP
; Timer0 Compare Handler
0x0020
jmp
TIM0_OVF
; Timer0 Overflow Handler
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
0x0022
jmp
SPI_STC
; SPI Transfer Complete Handler
0x0024
jmp
USART0_RXC
; USART0 RX Complete Handler
0x0026
jmp
USART0_DRE
; USART0,UDR Empty Handler
0x0028
jmp
USART0_TXC
; USART0 TX Complete Handler
0x002A
jmp
ADC
; ADC Conversion Complete Handler
0x002C
jmp
EE_RDY
; EEPROM Ready Handler
0x002E
jmp
ANA_COMP
; Analog Comparator Handler
0x0030
jmp
TIM1_COMPC
; Timer1 CompareC Handler
0x0032
jmp
TIM3_CAPT
; Timer3 Capture Handler
0x0034
jmp
TIM3_COMPA
; Timer3 CompareA Handler
0x0036
jmp
TIM3_COMPB
; Timer3 CompareB Handler
0x0038
jmp
TIM3_COMPC
; Timer3 CompareC Handler
0x003A
jmp
TIM3_OVF
; Timer3 Overflow Handler
0x003C
jmp
USART1_RXC
; USART1 RX Complete Handler
0x003E
jmp
USART1_DRE
; USART1,UDR Empty Handler
0x0040
jmp
USART1_TXC
; USART1 TX Complete Handler
0x0042
jmp
TWI
; Two-wire Serial Interface Handler
0x0044
jmp
SPM_RDY
; SPM Ready Handler
;
0x0046 RESET: ldi
r16, high(RAMEND); Main program start
0x0047
out
SPH,r16
0x0048
ldi
r16, low(RAMEND)
0x0049
0x004A
out
sei
SPL,r16
0x004B
<instr>
...
...
...
; Set Stack Pointer to top of RAM
; Enable interrupts
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:
Address Labels Code
Comments
0x0000 RESET: ldi
r16,high(RAMEND); Main program start
0x0001
out
SPH,r16
0x0002
ldi
r16,low(RAMEND)
0x0003
0x0004
out
sei
SPL,r16
0x0005
<instr>
; Set Stack Pointer to top of RAM
; Enable interrupts
xxx
;
.org 0x7002
0x7002
jmp
EXT_INT0
; IRQ0 Handler
0x7004
jmp
EXT_INT1
; IRQ1 Handler
...
...
...
0x7044
jmp
SPM_RDY
;
; Store Program Memory Ready Handler
63
2490N–AVR–05/08
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:
Address Labels Code
Comments
.org 0x0002
0x0002
jmp
EXT_INT0 ; IRQ0 Handler
0x0004
jmp
EXT_INT1 ; IRQ1 Handler
...
...
...
;
0x0044
jmp
SPM_RDY
; Store Program Memory Ready Handler
;
.org 0x7000
0x7000 RESET: ldi
r16,high(RAMEND); Main program start
0x7001
out
SPH,r16
0x7002
ldi
r16,low(RAMEND)
0x7003
0x7004
out
sei
SPL,r16
0x7005
<instr>
; Set Stack Pointer to top of RAM
; Enable interrupts
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 typical and general
program setup for the Reset and Interrupt Vector Addresses is:
Address Labels Code
Comments
;
Moving Interrupts
Between Application
and Boot Space
MCUCR – MCU
Control Register
.org 0x7000
0x7000
0x7002
jmp
jmp
RESET
; Reset handler
EXT_INT0 ; IRQ0 Handler
0x7004
jmp
EXT_INT1 ; IRQ1 Handler
...
...
...
0x7044
jmp
SPM_RDY ; Store Program Memory Ready Handler
;
0x7046 RESET: ldi
r16,high(RAMEND); Main program start
0x7047
out
SPH,r16
0x7048
ldi
r16,low(RAMEND)
0x7049
0x704A
out
sei
SPL,r16
0x704B
<instr>
; Set Stack Pointer to top of RAM
; Enable interrupts
xxx
The General Interrupt Control Register controls the placement of the Interrupt Vector table.
Bit
7
6
5
4
3
2
1
0
0x35 (0x55)
SRE
SRW10
SE
SM1
SM0
SM2
IVSEL
IVCE
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
MCUCR
• 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 beginning of the Boot
64
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 unintentional changes of Interrupt Vector
tables, a special write procedure must be followed to change the IVSEL bit:
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 following 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 programmed,
interrupts are disabled while executing from the Application section. If Interrupt Vectors are placed
in the Application section and Boot Lock bit BLB12 is programed, interrupts are disabled while
executing from the Boot Loader section. Refer to the section “Boot Loader Support – Read-WhileWrite 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);
}
65
2490N–AVR–05/08
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 unintentionally 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 VCC and Ground as indicated in Figure 29. Refer to “Electrical Characteristics” on page 325 for a complete list of parameters.
Figure 29. I/O Pin Equivalent Schematic
Rpu
Logic
Pxn
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 87.
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 66.
Most port pins are multiplexed with alternate functions for the peripheral features on the device.
How each alternate function interferes with the port pin is described in “Alternate Port Functions”
on page 71. 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.
Ports as General
Digital I/O
66
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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 30. General Digital I/O(1)
PUD
Q
D
DDxn
Q CLR
RESET
WDx
Q
Pxn
D
PORTxn
Q CLR
WPx
DATA BUS
RDx
RESET
SLEEP
RRx
SYNCHRONIZER
D
Q
L
Q
D
RPx
Q
PINxn
Q
clk I/O
PUD:
SLEEP:
clkI/O:
Note:
Configuring the Pin
PULLUP DISABLE
SLEEP CONTROL
I/O CLOCK
WDx:
RDx:
WPx:
RRx:
RPx:
WRITE DDRx
READ DDRx
WRITE PORTx
READ PORTx REGISTER
READ PORTx PIN
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.
Each port pin consists of three register bits: DDxn, PORTxn, and PINxn. As shown in “Register
Description for I/O Ports” on page 87, 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 configured 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 output pin, the port
pin is driven low (zero).
67
2490N–AVR–05/08
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 difference 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
Reading the Pin Value
DDxn
PORTxn
PUD
(in SFIOR)
I/O
Pull-up
0
0
X
Input
No
Tri-state (Hi-Z)
0
1
0
Input
Yes
Pxn will source current if ext. pulled
low.
0
1
1
Input
No
Tri-state (Hi-Z)
1
0
X
Output
No
Output Low (Sink)
1
1
X
Output
No
Output High (Source)
Comment
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
INSTRUCTIONS
XXX
in r17, PINx
XXX
SYNC LATCH
PINxn
r17
0xFF
0x00
tpd, max
tpd, min
68
ATmega64(L)
2490N–AVR–05/08
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 succeeding 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
INSTRUCTIONS
0xFF
out PORTx, r16
nop
in r17, PINx
SYNC LATCH
PINxn
r17
0x00
0xFF
tpd
69
2490N–AVR–05/08
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
r16,(1<<PB7)|(1<<PB6)|(1<<PB1)|(1<<PB0)
ldi
r17,(1<<DDB3)|(1<<DDB2)|(1<<DDB1)|(1<<DDB0)
out
PORTB,r16
out
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:
Digital Input Enable
and Sleep Modes
1. For the assembly program, two temporary registers are used to minimize the time from pullups 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.
As shown in Figure 30, the digital input signal can be clamped to ground at the input of 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 Functions” on page 71.
If a logic high level (“one”) is present on an asynchronous External Interrupt pin configured 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.
Unconnected Pins
70
If some pins are unused, it is recommended to ensure that these pins have a defined level. Even
though most of the digital inputs are disabled in the deep sleep modes as described above, float-
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
ing inputs should be avoided to reduce current consumption in all other modes where the digital
inputs are enabled (Reset, Active mode and Idle mode).
The simplest method to ensure a defined level of an unused pin, is to enable the internal pull-up.
In this case, the pull-up will be disabled during reset. If low power consumption during reset is
important, it is recommended to use an external pull-up or pull-down. Connecting unused pins
directly to VCC or GND is not recommended, since this may cause excessive currents if the pin
is accidentally configured as an output.
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 overridden 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 microcontroller family.
Figure 33. Alternate Port Functions(1)
PUOExn
PUOVxn
1
PUD
0
DDOExn
DDOVxn
1
Q D
DDxn
0
Q CLR
WDx
PVOExn
RESET
1
Pxn
Q
0
D
PORTxn
Q CLR
DIEOExn
WPx
DIEOVxn
DATA BUS
RDx
PVOVxn
RESET
1
0
RRx
SLEEP
SYNCHRONIZER
D
SET
Q
RPx
Q
D
PINxn
L
CLR
Q
CLR
Q
clk I/O
DIxn
AIOxn
PUOExn:
PUOVxn:
DDOExn:
DDOVxn:
PVOExn:
PVOVxn:
DIEOExn:
DIEOVxn:
SLEEP:
Note:
Pxn PULL-UP OVERRIDE ENABLE
Pxn PULL-UP OVERRIDE VALUE
Pxn DATA DIRECTION OVERRIDE ENABLE
Pxn DATA DIRECTION OVERRIDE VALUE
Pxn PORT VALUE OVERRIDE ENABLE
Pxn PORT VALUE OVERRIDE VALUE
Pxn DIGITAL INPUT-ENABLE OVERRIDE ENABLE
Pxn DIGITAL INPUT-ENABLE OVERRIDE VALUE
SLEEP CONTROL
PUD:
WDx:
RDx:
RRx:
WPx:
RPx:
clkI/O:
DIxn:
AIOxn:
PULLUP DISABLE
WRITE DDRx
READ DDRx
READ PORTx REGISTER
WRITE PORTx
READ PORTx PIN
I/O CLOCK
DIGITAL INPUT PIN n ON PORTx
ANALOG INPUT/OUTPUT PIN n ON PORTx
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 generated internally in
the modules having the alternate function.
71
2490N–AVR–05/08
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
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.
DDOE
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.
DDOV
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.
PVOE
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
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).
DI
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.
SFIOR – Special
Function IO Register
Bit
7
6
5
4
3
2
1
0
0x20 (0x40)
TSM
–
–
–
ACME
PUD
PSR0
PSR321
Read/Write
R/W
R
R
R
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
SFIOR
• 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 67 for more details about this feature.
72
ATmega64(L)
2490N–AVR–05/08
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
Alternate Function
PA7
AD7 (External memory interface address and data bit 7)
PA6
AD6 (External memory interface address and data bit 6)
PA5
AD5 (External memory interface address and data bit 5)
PA4
AD4 (External memory interface address and data bit 4)
PA3
AD3 (External memory interface address and data bit 3)
PA2
AD2 (External memory interface address and data bit 2)
PA1
AD1 (External memory interface address and data bit 1)
PA0
AD0 (External memory interface address and data bit 0)
Table 28 and Table 29 relates the alternate functions of Port A to the overriding signals shown in
Figure 33 on page 71.
Table 28. Overriding Signals for Alternate Functions in PA7..PA4
Signal
Name
PA7/AD7
PA6/AD6
PA5/AD5
PA4/AD4
PUOE
SRE
SRE
SRE
SRE
(1)
PUOV
~(WR | ADA ) •
PORTA7 • PUD
~(WR | ADA) •
PORTA6 • PUD
~(WR | ADA) •
PORTA5 • PUD
~(WR | ADA) •
PORTA4 • PUD
DDOE
SRE
SRE
SRE
SRE
DDOV
WR | ADA
WR | ADA
WR | ADA
WR | ADA
PVOE
SRE
SRE
SRE
SRE
PVOV
A7 • ADA | D7
OUTPUT • WR
A6 • ADA | D6
OUTPUT • WR
A5 • ADA | D5
OUTPUT • WR
A4 • ADA | D4
OUTPUT • WR
DIEOE
0
0
0
0
DIEOV
0
0
0
0
D7 INPUT
D6 INPUT
D5 INPUT
D4 INPUT
–
–
–
–
DI
AIO
73
2490N–AVR–05/08
Table 29. Overriding Signals for Alternate Functions in PA3..PA0(1)
Signal
Name
PA3/AD3
PA2/AD2
PA1/AD1
PA0/AD0
PUOE
SRE
SRE
SRE
SRE
PUOV
~(WR | ADA) •
PORTA3 • PUD
~(WR | ADA) •
PORTA2 • PUD
~(WR | ADA) •
PORTA1 • PUD
~(WR | ADA) •
PORTA0 • PUD
DDOE
SRE
SRE
SRE
SRE
DDOV
WR | ADA
WR | ADA
WR | ADA
WR | ADA
PVOE
SRE
SRE
SRE
SRE
PVOV
A3 • ADA | D3
OUTPUT • WR
A2• ADA | D2
OUTPUT • WR
A1 • ADA | D1
OUTPUT • WR
A0 • ADA | D0
OUTPUT • WR
DIEOE
0
0
0
0
DIEOV
0
0
0
0
D3 INPUT
D2 INPUT
D1 INPUT
D0 INPUT
–
–
–
–
DI
AIO
Note:
Alternate Functions of
Port B
1. ADA is short for ADdress Active and represents the time when address is output. See “External Memory Interface” on page 27 for details.
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)
PB6
OC1B (Output Compare and PWM Output B for Timer/Counter1)
PB5
OC1A (Output Compare and PWM Output A for Timer/Counter1)
PB4
OC0 (Output Compare and PWM Output for Timer/Counter0)
PB3
MISO (SPI Bus Master Input/Slave Output)
PB2
MOSI (SPI Bus Master Output/Slave Input)
PB1
SCK (SPI Bus Serial Clock)
PB0
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.
74
ATmega64(L)
2490N–AVR–05/08
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 controlled 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 71. SPI MSTR INPUT and SPI SLAVE OUTPUT constitute the MISO signal,
while MOSI is divided into SPI MSTR OUTPUT and SPI SLAVE INPUT.
75
2490N–AVR–05/08
Table 31. Overriding Signals for Alternate Functions in PB7..PB4
Signal
Name
PB7/OC2/OC1C
PB6/OC1B
PB5/OC1A
PB4/OC0
PUOE
0
0
0
0
PUOV
0
0
0
0
DDOE
0
0
0
0
DDOV
0
0
0
0
OC1B ENABLE
OC1A ENABLE
OC0 ENABLE
OC1B
OC1A
OC0B
PVOE
OC2/OC1C ENABLE
(1)
(1)
PVOV
OC2/OC1C
DIEOE
0
0
0
0
DIEOV
0
0
0
0
DI
–
–
–
–
AIO
–
–
–
–
Note:
1. See “Output Compare Modulator (OCM1C2)” on page 161 for details. OC1C does not exist in
ATmega103 compatibility mode.
Table 32. Overriding Signals for Alternate Functions in PB3..PB0
76
Signal
Name
PB3/MISO
PB2/MOSI
PB1/SCK
PB0/SS
PUOE
SPE • MSTR
SPE • MSTR
SPE • MSTR
SPE • MSTR
PUOV
PORTB3 • PUD
PORTB2 • PUD
PORTB1 • PUD
PORTB0 • PUD
DDOE
SPE • MSTR
SPE • MSTR
SPE • MSTR
SPE • MSTR
DDOV
0
0
0
0
PVOE
SPE • MSTR
SPE • MSTR
SPE • MSTR
0
PVOV
SPI SLAVE OUTPUT
SPI MSTR OUTPUT
SCK OUTPUT
0
DIEOE
0
0
0
0
DIEOV
0
0
0
0
DI
SPI MSTR INPUT
SPI SLAVE INPUT
SCK INPUT
SPI SS
AIO
–
–
–
–
ATmega64(L)
2490N–AVR–05/08
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
Alternate Function
PC7
A15
PC6
A14
PC5
A13
PC4
A12
PC3
A11
PC2
A10
PC1
A9
PC0
A8
Table 34 and Table 35 relate the alternate functions of Port C to the overriding signals shown in
Figure 33 on page 71.
Table 34. Overriding Signals for Alternate Functions in PC7..PC4
Signal Name
PC7/A15
(1)
PC6/A14
PC5/A13
PC4/A12
PUOE
SRE • (XMM <1)
SRE • (XMM<2)
SRE • (XMM<3)
SRE • (XMM<4)
PUOV
0
0
0
0
DDOE
SRE • (XMM<1)
SRE • (XMM<2)
SRE • (XMM<3)
SRE • (XMM<4)
DDOV
1
1
1
1
PVOE
SRE • (XMM<1)
SRE • (XMM<2)
SRE • (XMM<3)
SRE • (XMM<4)
PVOV
A11
A10
A9
A8
DIEOE
0
0
0
0
DIEOV
0
0
0
0
DI
–
–
–
–
AIO
–
–
–
–
77
2490N–AVR–05/08
Table 35. Overriding Signals for Alternate Functions in PC3..PC0(1)
Signal Name
PC3/A11
PC2/A10
PC1/A9
PC0/A8
PUOE
SRE • (XMM<5)
SRE • (XMM<6)
SRE • (XMM<7)
SRE • (XMM<7)
PUOV
0
0
0
0
DDOE
SRE • (XMM<5)
SRE • (XMM<6)
SRE • (XMM<7)
SRE • (XMM<7)
DDOV
1
1
1
1
PVOE
SRE • (XMM<5)
SRE • (XMM<6)
SRE • (XMM<7)
SRE • (XMM<7)
PVOV
A11
A10
A9
A8
DIEOE
0
0
0
0
DIEOV
0
0
0
0
DI
–
–
–
–
AIO
–
–
–
–
Note:
Alternate Functions of
Port D
1. XMM = 0 in ATmega103 compatibility mode.
The Port D pins with alternate functions are shown in Table 36.
Table 36. Port D Pins Alternate Functions
Port Pin
Alternate Function
PD7
T2 (Timer/Counter2 Clock Input)
PD6
T1 (Timer/Counter1 Clock Input)
PD5
XCK1(1) (USART1 External Clock Input/Output)
PD4
ICP1 (Timer/Counter1 Input Capture Pin)
PD3
INT3/TXD1(1) (External Interrupt3 Input or UART1 Transmit Pin)
PD2
INT2/RXD1(1) (External Interrupt2 Input or UART1 Receive Pin)
PD1
INT1/SDA(1) (External Interrupt1 Input or TWI Serial DAta)
PD0
INT0/SCL(1) (External Interrupt0 Input or TWI Serial CLock)
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.
• ICP1 – Port D, Bit 4
ICP1 – Input Capture Pin1: The PD4 pin can act as an Input Capture pin for Timer/Counter1.
78
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
• 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 71.
79
2490N–AVR–05/08
Table 37. Overriding Signals for Alternate Functions PD7..PD4
Signal Name
PD7/T2
PD6/T1
PD5/XCK1
PD4/ICP1
PUOE
0
0
0
0
PUOV
0
0
0
0
DDOE
0
0
0
0
DDOV
0
0
0
0
PVOE
0
0
UMSEL1
0
PVOV
0
0
XCK1 OUTPUT
0
DIEOE
0
0
0
0
DIEOV
0
0
0
0
DI
T2 INPUT
T1 INPUT
XCK1 INPUT
ICP1 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
TXEN1
RXEN1
TWEN
TWEN
PUOV
0
PORTD2 • PUD
PORTD1 • PUD
PORTD0 • PUD
DDOE
TXEN1
RXEN1
TWEN
TWEN
DDOV
1
0
SDA_OUT
SCL_OUT
PVOE
TXEN1
0
TWEN
TWEN
PVOV
TXD1
0
0
0
DIEOE
INT3 ENABLE
INT2 ENABLE
INT1 ENABLE
INT0 ENABLE
DIEOV
1
1
1
1
DI
INT3 INPUT
INT2 INPUT/RXD1
INT1 INPUT
INT0 INPUT
AIO
–
–
SDA INPUT
SCL INPUT
Note:
80
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 connected between
the AIO outputs shown in the port figure and the digital logic of the TWI module.
ATmega64(L)
2490N–AVR–05/08
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
INT7/ICP3(1) (External Interrupt 7 Input or Timer/Counter3 Input Capture Pin)
PE6
INT6/ T3(1) (External Interrupt 6 Input or Timer/Counter3 Clock Input)
PE5
INT5/OC3C(1) (External Interrupt 5 Input or Output Compare and PWM Output C
for Timer/Counter3)
PE4
INT4/OC3B(1) (External Interrupt 4 Input or Output Compare and PWM Output B for
Timer/Counter3)
PE3
AIN1/OC3A (1) (Analog Comparator Negative Input or Output Compare and PWM
Output A for Timer/Counter3)
PE2
AIN0/XCK0(1) (Analog Comparator Positive Input or USART0 external clock
input/output)
PE1
PDO/TXD0 (Programming Data Output or UART0 Transmit Pin)
PE0
PDI/RXD0 (Programming Data Input or UART0 Receive Pin)
Note:
1. ICP3, T3, OC3C, OC3B, OC3B, OC3A, and XCK0 not applicable in ATmega103 compatibility
mode.
• INT7/ICP3 – Port E, Bit 7
INT7, External Interrupt Source 7: The PE7 pin can serve as an External Interrupt source.
ICP3 – 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.
• 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.
81
2490N–AVR–05/08
• 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 71.
Table 40. Overriding Signals for Alternate Functions PE7..PE4
82
Signal
Name
PE7/INT7/ICP3
PE6/INT6/T3
PE5/INT5/OC3C
PE4/INT4/OC3B
PUOE
0
0
0
0
PUOV
0
0
0
0
DDOE
0
0
0
0
DDOV
0
0
0
0
PVOE
0
0
OC3C ENABLE
OC3B ENABLE
PVOV
0
0
OC3C
OC3B
DIEOE
INT7 ENABLE
INT6 ENABLE
INT5 ENABLE
INT4 ENABLE
DIEOV
1
1
1
1
DI
INT7 INPUT/ICP3
INPUT
INT7 INPUT/T3
INPUT
INT5 INPUT
INT4 INPUT
AIO
–
–
–
–
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 41. Overriding Signals for Alternate Functions in PE3..PE0
Alternate Functions of
Port F
Signal
Name
PE3/AIN1/OC3A
PE2/AIN0/XCK0
PE1/PDO/TXD0
PE0/PDI/RXD0
PUOE
0
0
TXEN0
RXEN0
PUOV
0
0
0
PORTE0 • PUD
DDOE
0
0
TXEN0
RXEN0
DDOV
0
0
1
0
PVOE
OC3B ENABLE
UMSEL0
TXEN0
0
PVOV
OC3B
XCK0 OUTPUT
TXD0
0
DIEOE
0
0
0
0
DIEOV
0
0
0
0
DI
0
XCK0 INPUT
–
RXD0
AIO
AIN1 INPUT
AIN0 INPUT
–
–
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
Alternate Function
PF7
ADC7/TDI (ADC input channel 7 or JTAG Test Data Input)
PF6
ADC6/TDO (ADC input channel 6 or JTAG Test Data Output)
PF5
ADC5/TMS (ADC input channel 5 or JTAG Test mode Select)
PF4
ADC4/TCK (ADC input channel 4 or JTAG Test Clock)
PF3
ADC3 (ADC input channel 3)
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 Register. 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.
83
2490N–AVR–05/08
• 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 interface 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
84
Signal
Name
PF7/ADC7/TDI
PF6/ADC6/TDO
PF5/ADC5/TMS
PF4/ADC4/TCK
PUOE
JTAGEN
JTAGEN
JTAGEN
JTAGEN
PUOV
1
0
1
1
DDOE
JTAGEN
JTAGEN
JTAGEN
JTAGEN
DDOV
0
SHIFT_IR +
SHIFT_DR
0
0
PVOE
0
JTAGEN
0
0
PVOV
0
TDO
0
0
DIEOE
JTAGEN
JTAGEN
JTAGEN
JTAGEN
DIEOV
0
0
0
0
DI
–
–
–
–
AIO
TDI/ADC7 INPUT
ADC6 INPUT
TMS/ADC5
INPUT
TCKADC4 INPUT
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 44. Overriding Signals for Alternate Functions in PF3..PF0
Alternate Functions of
Port G
Signal
Name
PF3/ADC3
PF2/ADC2
PF1/ADC1
PF0/ADC0
PUOE
0
0
0
0
PUOV
0
0
0
0
DDOE
0
0
0
0
DDOV
0
0
0
0
PVOE
0
0
0
0
PVOV
0
0
0
0
DIEOE
0
0
0
0
DIEOV
0
0
0
0
DI
–
–
–
–
AIO
ADC3 INPUT
ADC2 INPUT
ADC1 INPUT
ADC0 INPUT
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 configuration is as follows:
Table 45. Port G Pins Alternate Functions
Port Pin
Alternate Function
PG4
TOSC1 (RTC Oscillator Timer/Counter0)
PG3
TOSC2 (RTC Oscillator Timer/Counter0)
PG2
ALE (Address Latch Enable to external memory)
PG1
RD (Read strobe to external memory)
PG0
WR (Write strobe to external memory)
• TOSC1 – Port G, Bit 4
TOSC2, Timer Oscillator pin 1: When the AS0 bit in ASSR is set (one) to enable asynchronous
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 asynchronous
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.
85
2490N–AVR–05/08
• 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 71.
Table 46. Overriding Signals for Alternate Functions in PG4..PG1
Signal Name
PG4/TOSC1
PG3/TOSC2
PG2/ALE
PG1/RD
PUOE
AS0
AS0
SRE
SRE
PUOV
0
0
0
0
DDOE
AS0
AS0
SRE
SRE
DDOV
0
0
1
1
PVOE
0
0
SRE
SRE
PVOV
0
0
ALE
RD
DIEOE
AS0
AS0
0
0
DIEOV
0
0
0
0
DI
–
–
–
–
AIO
T/C0 OSC INPUT
T/C0 OSC OUTPUT
–
–
Table 47. Overriding Signals for Alternate Functions in PG0
86
Signal Name
PG0/WR
PUOE
SRE
PUOV
0
DDOE
SRE
DDOV
1
PVOE
SRE
PVOV
WR
DIEOE
0
DIEOV
0
DI
–
AIO
–
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Register
Description for I/O
Ports
PORTA – Port A Data
Register
DDRA – Port A Data
Direction Register
PINA – Port A Input
Pins Address
PORTB – Port B Data
Register
DDRB – Port B Data
Direction Register
PINB – Port B Input
Pins Address
PORTC – Port C Data
Register
DDRC – Port C Data
Direction Register
Bit
7
6
5
4
3
2
1
0
0x1B (0x3B)
PORTA7
PORTA6
PORTA5
PORTA4
PORTA3
PORTA2
PORTA1
PORTA0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x1A (0x3A)
DDA7
DDA6
DDA5
DDA4
DDA3
DDA2
DDA1
DDA0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x19 (0x39)
PINA7
PINA6
PINA5
PINA4
PINA3
PINA2
PINA1
PINA0
Read/Write
R
R
R
R
R
R
R
R
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Bit
7
6
5
4
3
2
1
0
0x18 (0x38)
PORTB7
PORTB6
PORTB5
PORTB4
PORTB3
PORTB2
PORTB1
PORTB0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x17 (0x37)
DDB7
DDB6
DDB5
DDB4
DDB3
DDB2
DDB1
DDB0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x16 (0x36)
PINB7
PINB6
PINB5
PINB4
PINB3
PINB2
PINB1
PINB0
Read/Write
R
R
R
R
R
R
R
R
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Bit
7
6
5
4
3
2
1
0
0x15 (0x35)
PORTC7
PORTC6
PORTC5
PORTC4
PORTC3
PORTC2
PORTC1
PORTC0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x14 (0x34)
DDC7
DDC6
DDC5
DDC4
DDC3
DDC2
DDC1
DDC0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
PORTA
DDRA
PINA
PORTB
DDRB
PINB
PORTC
DDRC
87
2490N–AVR–05/08
PINC – Port C Input
Pins Address
Bit
7
6
5
4
3
2
1
0
0x13 (0x33)
PINC7
PINC6
PINC5
PINC4
PINC3
PINC2
PINC1
PINC0
Read/Write
R
R
R
R
R
R
R
R
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
PINC
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 compatibility mode, and should not
be used for 100% backward compatibility.
PORTD – Port D Data
Register
DDRD – Port D Data
Direction Register
PIND – Port D Input
Pins Address
PORTE – Port E Data
Register
DDRE – Port E Data
Direction Register
PINE – Port E Input
Pins Address
PORTF – Port F Data
Register
Bit
7
6
5
4
3
2
1
0
0x12 (0x32)
PORTD7
PORTD6
PORTD5
PORTD4
PORTD3
PORTD2
PORTD1
PORTD0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x11 (0x31)
DDD7
DDD6
DDD5
DDD4
DDD3
DDD2
DDD1
DDD0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x10 (0x30)
PIND7
PIND6
PIND5
PIND4
PIND3
PIND2
PIND1
PIND0
Read/Write
R
R
R
R
R
R
R
R
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Bit
7
6
5
4
3
2
1
0
0x03 (0x23)
PORTE7
PORTE6
PORTE5
PORTE4
PORTE3
PORTE2
PORTE1
PORTE0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x02 (0x22)
DDE7
DDE6
DDE5
DDE4
DDE3
DDE2
DDE1
DDE0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x01 (0x21)
PINE7
PINE6
PINE5
PINE4
PINE3
PINE2
PINE1
PINE0
Read/Write
R
R
R
R
R
R
R
R
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Bit
7
6
5
4
3
2
1
0
PORTF7
PORTF6
PORTF5
PORTF4
PORTF3
PORTF2
PORTF1
PORTF0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
(0x62)
88
PORTD
DDRD
PIND
PORTE
DDRE
PINF
PORTF
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
DDRF – Port F Data
Direction Register
Bit
7
6
5
4
3
2
1
0
DDF7
DDF6
DDF5
DDF4
DDF3
DDF2
DDF1
DDF0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
(0x61)
PINF – Port F Input
Pins Address
Bit
7
6
5
4
3
2
1
0
0x00 (0x20)
PINF7
PINF6
PINF5
PINF4
PINF3
PINF2
PINF1
PINF0
Read/Write
R
R
R
R
R
R
R
R
Initial Value
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
DDRF
PINF
Note that PORTF and DDRF Registers are not available in ATmega103 compatibility mode
where Port F serves as digital input only.
PORTG – Port G Data
Register
DDRG – Port G Data
Direction Register
PING – Port G Input
Pins Address
Bit
7
6
5
4
3
2
1
0
(0x65)
–
–
–
PORTG4
PORTG3
PORTG2
PORTG1
PORTG0
Read/Write
R
R
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
(0x64)
–
–
–
DDG4
DDG3
DDG2
DDG1
DDG0
Read/Write
R
R
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
(0x63)
–
–
–
PING4
PING3
PING2
PING1
PING0
Read/Write
R
R
R
R
R
R
R
R
Initial Value
0
0
0
N/A
N/A
N/A
N/A
N/A
PORTG
DDRG
PING
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).
89
2490N–AVR–05/08
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 provides 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 interrupts on INT7:4 requires the presence of an I/O clock,
described in “Clock Systems and their Distribution” on page 37. 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 325. 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 37. 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.
EICRA – External
Interrupt Control
Register A
Bit
7
6
5
4
3
2
1
0
ISC31
ISC30
ISC21
ISC20
ISC11
ISC10
ISC01
ISC00
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
(0x6A)
EICRA
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.
90
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 48. Interrupt Sense Control(1)
ISCn1
ISCn0
0
0
The low level of INTn generates an interrupt request.
0
1
Reserved
1
0
The falling edge of INTn generates asynchronously an interrupt request.
1
1
The rising edge of INTn generates asynchronously an interrupt request.
Note:
Description
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
Condition
Min
Typ
Minimum pulse width for
asynchronous External Interrupt
tINT
EICRB – External
Interrupt Control
Register B
Parameter
Bit
Max
Units
50
ns
7
6
5
4
3
2
1
0
0x3A (0x5A)
ISC71
ISC70
ISC61
ISC60
ISC51
ISC50
ISC41
ISC40
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
EICRB
• 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
0
0
The low level of INTn generates an interrupt request.
0
1
Any logical change on INTn generates an interrupt request
1
0
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:
Description
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.
91
2490N–AVR–05/08
EIMSK – External
Interrupt Mask
Register
Bit
7
6
5
4
3
2
1
0
0x39 (0x59)
INT7
INT6
INT5
INT4
INT3
INT2
INT1
INT0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
EIMSK
• 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 Control 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.
EIFR – External
Interrupt Flag Register
Bit
7
6
5
4
3
2
1
0
0x38 (0x58)
INTF7
INTF6
INTF5
INTF4
INTF3
INTF2
INTF1
INTF0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
EIFR
• 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 internal signals which will set the INTF3:0 flags. See “Digital Input
Enable and Sleep Modes” on page 70 for more information.
92
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
8-bit
Timer/Counter0
with PWM and
Asynchronous
Operation
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)
• 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 104.
Figure 34. 8-bit Timer/Counter Block Diagram
TCCRn
count
TOVn
(Int. Req.)
clear
Control Logic
direction
clkTn
TOSC1
BOTTOM
TOP
Prescaler
T/C
Oscillator
TOSC2
Timer/Counter
TCNTn
=0
= 0xFF
OCn
(Int. Req.)
Waveform
Generation
=
clkI/O
OCn
DATABUS
OCRn
Synchronized Status Flags
clkI/O
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 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 decrement) its value. The Timer/Counter is inac-
93
2490N–AVR–05/08
tive 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 Waveform Generator to generate a PWM or variable frequency output on the Output Compare pin (OC0). See “Output
Compare Unit” on page 95. 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
Timer/Counter
Clock Sources
94
BOTTOM
The counter reaches the BOTTOM when it becomes zero (0x00).
MAX
The counter reaches its MAXimum when it becomes 0xFF (decimal 255).
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 the
fixed value 0xFF (MAX) or the value stored in the OCR0 Register. The
assignment is dependent on the mode of operation.
The Timer/Counter can be clocked by an internal synchronous or an external asynchronous
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 “ASSR
– Asynchronous Status Register” on page 107. For details on clock sources and prescaler, see
“Timer/Counter Prescaler” on page 110.
ATmega64(L)
2490N–AVR–05/08
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
TCNTn
clear
clk Tn
Control Logic
Prescaler
T/C
Oscillator
direction
bottom
TOSC2
top
clkI/O
Signal description (internal signals):
count
Increment or decrement TCNT0 by 1.
direction
Selects between increment and decrement.
clear
Clear TCNT0 (set all bits to zero).
clkT0
Timer/Counter clock.
top
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 decremented
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 98.
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 Generator 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 signals 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 98). Figure 36 shows a block diagram of the Output Compare unit.
95
2490N–AVR–05/08
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 synchronization 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 Output 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.
96
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Compare Match
Output Unit
The Compare Output mode (COM01:0) bits have two functions. The Waveform Generator 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
COMn0
FOCn
Waveform
Generator
D
Q
1
OCn
DATA BUS
D
0
OCn
Pin
Q
PORT
D
Q
DDR
clk I/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 direction (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 104.
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 compare output actions in the nonPWM modes refer to Table 53 on page 105. For fast PWM mode, refer to Table 54 on page 105,
and for phase correct PWM refer to Table 55 on page 105.
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.
97
2490N–AVR–05/08
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 97.).
For detailed timing information refer to “Timer/Counter Timing Diagrams” on page 102.
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)
Period
(COMn1:0 = 1)
1
2
3
4
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
98
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 frequency 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 frequency
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 BOTTOM. 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.
99
2490N–AVR–05/08
Figure 39. Fast PWM Mode, Timing Diagram
OCRn Interrupt Flag Set
OCRn Update and
TOVn Interrupt Flag Set
TCNTn
OCn
(COMn1:0 = 2)
OCn
(COMn1:0 = 3)
Period
1
2
3
4
5
6
7
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 compare 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 105). 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 setting) 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 output 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.
100
ATmega64(L)
2490N–AVR–05/08
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 noninverting 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 represent 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
OCn
(COMn1:0 = 2)
OCn
(COMn1:0 = 3)
Period
1
2
3
The Timer/Counter Overflow Flag (TOV0) is set each time the counter reaches BOTTOM. 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 105). 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 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 between OCR0 and TCNT0 when the counter decrements.
101
2490N–AVR–05/08
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 noninverted 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.
Timer/Counter
Timing Diagrams
•
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 upcounting 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.
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
MAX - 1
MAX
BOTTOM
BOTTOM + 1
TOVn
Figure 42 shows the same timing data, but with the prescaler enabled.
102
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 42. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O /8)
TCNTn
MAX - 1
MAX
BOTTOM
BOTTOM + 1
TOVn
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 - 1
OCRn
OCRn
OCRn + 1
OCRn + 2
OCRn Value
OCFn
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)
OCRn
TOP - 1
TOP
BOTTOM
BOTTOM + 1
TOP
OCFn
103
2490N–AVR–05/08
8-bit
Timer/Counter
Register
Description
TCCR0 –
Timer/Counter Control
Register
Bit
7
6
5
4
3
2
1
0
0x33 (0x53)
FOC0
WGM00
COM01
COM00
WGM01
CS02
CS01
CS00
Read/Write
W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
TCCR0
• 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 operation 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 98.
Table 52. Waveform Generation Mode Bit Description(1)
Mode
WGM01
(CTC0)
WGM00
(PWM0)
Timer/Counter Mode of
Operation
TOP
Update of
OCR0 at
TOV0 Flag
Set on
0
0
0
Normal
0xFF
Immediate
MAX
1
0
1
PWM, Phase Correct
0xFF
TOP
BOTTOM
2
1
0
CTC
OCR0
Immediate
MAX
3
1
1
Fast PWM
0xFF
BOTTOM
MAX
Note:
1. The CTC0 and PWM0 bit definition names are now obsolete. Use the WGM01:0 definitions.
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 corresponding 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).
104
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 53. Compare Output Mode, non-PWM Mode
COM01
COM00
Description
0
0
Normal port operation, OC0 disconnected.
0
1
Toggle OC0 on Compare Match.
1
0
Clear OC0 on Compare Match.
1
1
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
0
0
Normal port operation, OC0 disconnected.
0
1
Reserved
1
0
Clear OC0 on Compare Match, set OC0 at BOTTOM,
(non-inverting mode).
1
1
Set OC0 on Compare Match, clear OC0 at BOTTOM,
(inverting mode).
Note:
Description
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 BOTTOM. See “Fast PWM Mode” on page 99
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
0
0
Normal port operation, OC0 disconnected.
0
1
Reserved.
1
0
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:
Description
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
101 for more details.
105
2490N–AVR–05/08
• 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
TCNT0 –
Timer/Counter
Register
CS02
CS01
CS00
0
0
0
No clock source (Timer/counter stopped)
0
0
1
clkT0S/(No prescaling)
0
1
0
clkT0S/8 (From prescaler)
0
1
1
clkT0S/32 (From prescaler)
1
0
0
clkT0S/64 (From prescaler)
1
0
1
clkT0S/128 (From prescaler)
1
1
0
clkT0S/256 (From prescaler)
1
1
1
clkT0S/1024 (From prescaler)
Bit
7
6
Description
5
0x32 (0x52)
4
3
2
1
0
TCNT0[7:0]
TCNT0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
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.
OCR0 – Output
Compare Register
Bit
7
6
5
0x31 (0x51)
4
3
2
1
0
OCR0[7:0]
OCR0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
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.
106
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Asynchronous
Operation of the
Timer/Counter
ASSR – Asynchronous
Status Register
Bit
7
6
5
4
3
2
1
0
0x30 (0x50)
–
–
–
–
AS0
TCN0UB
OCR0UB
TCR0UB
Read/Write
R
R
R
R
R/W
R
R
R
Initial Value
0
0
0
0
0
0
0
0
ASSR
• 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.
6. Enable interrupts, if needed.
107
2490N–AVR–05/08
•
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 reentering 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 Powersave 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.
108
•
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 Powersave mode, and the I/O clock (clkI/O) again becomes active, TCNT0 will 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:
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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.
•
TIMSK –
Timer/Counter
Interrupt Mask
Register
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.
Bit
7
6
5
4
3
2
1
0
0x37 (0x57)
OCIE2
TOIE2
TICIE1
OCIE1A
OCIE1B
TOIE1
OCIE0
TOIE0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
TIMSK
• 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.
TIFR – Timer/Counter
Interrupt Flag Register
Bit
7
6
5
4
3
2
1
0
0x36 (0x56)
OCF2
TOV2
ICF1
OCF1A
OCF1B
TOV1
OCF0
TOV0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
TIFR
• 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 Compare Match Interrupt Enable), and
OCF0 are set (one), the Timer/Counter0 Compare Match Interrupt is executed.
109
2490N–AVR–05/08
• 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.
Figure 45. Prescaler for Timer/Counter0
PSR0
clkT0S/1024
clkT0S/256
AS0
clkT0S/128
10-BIT T/C PRESCALER
Clear
clkT0S/64
TOSC1
clkT0S
clkT0S/32
clkOSC
clkT0S/8
Timer/Counter
Prescaler
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 asynchronously 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: clk T0S /8, clk T0S /32, clk T0S /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.
110
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
SFIOR – Special
Function IO Register
Bit
7
6
5
4
3
2
1
0
0x20 (0x40)
TSM
–
–
–
ACME
PUD
PSR0
PSR321
Read/Write
R/W
R
R
R
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
SFIOR
• 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 corresponding 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.
111
2490N–AVR–05/08
16-bit
Timer/Counter
(Timer/Counter
1 and
Timer/Counter3
)
The 16-bit Timer/Counter unit allows accurate program execution timing (event management),
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
• 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 132.
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.
112
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 46. 16-bit Timer/Counter Block Diagram(1)
Count
Clear
Direction
TOVx
(Int.Req.)
Control Logic
TCLK
Clock Select
Edge
Detector
TOP
Tx
BOTTOM
( From Prescaler )
Timer/Counter
TCNTx
=
=0
OCFxA
(Int.Req.)
Waveform
Generation
=
OCxA
OCRxA
OCFxB
(Int.Req.)
Fixed
TOP
Values
Waveform
Generation
DATA BUS
=
OCxB
OCRxB
OCFxC
(Int.Req.)
Waveform
Generation
=
OCRxC
OCxC
( From Analog
Comparator Ouput )
ICFx (Int.Req.)
Edge
Detector
ICRx
Noise
Canceler
ICPx
TCCRxA
Note:
Registers
TCCRxB
TCCRxC
1. Refer to Figure 1 on page 2, Table 30 on page 74, and Table 39 on page 81 for
Timer/Counter1 and 3 pin placement and description.
The Timer/Counter (TCNTn), Output Compare Registers (OCRnA/B/C), and Input Capture Register (ICRn) are all 16-bit registers. Special procedures must be followed when accessing the 16bit registers. These procedures are described in the section “Accessing 16-bit Registers” on
page 115. 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 (clkTn).
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).
113
2490N–AVR–05/08
See “Output Compare Units” on page 121. The Compare Match event will also set the Compare
Match Flag (OCFnA/B/C) which can be used to generate an Output 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 Comparator 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
Compatibility
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.
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.
•
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.
114
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 storing of the high byte of the 16bit 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 register 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. See “About Code Examples” on page 9.
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 interrupt
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 register, the main code must disable the
interrupts during the 16-bit access.
115
2490N–AVR–05/08
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:
116
1. See “About Code Examples” on page 9.
The assembly code example returns the TCNTn value in the r17:r16 register pair.
ATmega64(L)
2490N–AVR–05/08
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. See “About Code Examples” on page 9.
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 144.
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.
117
2490N–AVR–05/08
Figure 47. Counter Unit Block Diagram
DATA BUS
(8-bit)
TOVn
(Int.Req.)
TEMP (8-bit)
Clock Select
Count
TCNTnH (8-bit)
TCNTnL (8-bit)
TCNTn (16-bit Counter)
Clear
Direction
Control Logic
clkTn
Edge
Detector
Tn
( 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).
clkTn
Timer/counter clock.
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 important 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 decremented
at each Timer Clock (clkT n). The clk T n can be generated from an external or 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 overrides (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 Operation” on page 124.
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.
118
ATmega64(L)
2490N–AVR–05/08
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. Alternatively 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 elements 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)
WRITE
ICRnL (8-bit)
TCNTnH (8-bit)
ICRn (16-bit Register)
ACO*
Analog
Comparator
ACIC*
TCNTnL (8-bit)
TCNTn (16-bit Counter)
ICNC
ICES
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 written to the ICRn
119
2490N–AVR–05/08
Register. When writing the ICRn Register the high byte must be written 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 115.
Input Capture 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 setting 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 capture. 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 144). 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 capture 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 relatively 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).
120
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Output Compare
Units
The 16-bit comparator continuously compares TCNTn with the Output Compare Register
(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 Generator 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 124.)
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 register 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)
OCRnx Buffer (16-bit Register)
OCRnxH (8-bit)
TCNTnL (8-bit)
TCNTn (16-bit Counter)
OCRnxL (8-bit)
OCRnx (16-bit Register)
= (16-bit Comparator )
OCFnx (Int.Req.)
TOP
BOTTOM
Waveform Generator
WGMn3:0
OCnx
COMnx1:0
The OCRnx Register is double buffered when using any of the twelve 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 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 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
121
2490N–AVR–05/08
automatically as the TCNTn – and ICRn Register). Therefore 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 115.
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
122
The Compare Output mode (COMnx1:0) bits have two functions. The Waveform Generator 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. Figure 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”.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 50. Compare Match Output Unit, Schematic
COMnx1
COMnx0
FOCnx
Waveform
Generator
D
Q
1
OCnx
DATA BUS
D
0
OCnx
Pin
Q
PORT
D
Q
DDR
clk I/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 132.
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 compare output actions in
the non-PWM modes refer to Table 58 on page 134. For fast PWM mode refer to Table 59 on
page 134, and for phase correct and phase and frequency correct PWM refer to Table 60 on
page 135.
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.
123
2490N–AVR–05/08
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 output should be set, cleared or toggle at a Compare
Match (See “Compare Match Output Unit” on page 122.)
For detailed timing information refer to “Timer/Counter Timing Diagrams” on page 131.
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 Overflow 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 maximum
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.
Figure 51. CTC Mode, Timing Diagram
OCnA Interrupt Flag Set
or ICFn Interrupt Flag Set
(Interrupt on TOP)
TCNTn
OCnA
(Toggle)
Period
124
(COMnA1:0 = 1)
1
2
3
4
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 fOCnA = fclk_I/O/2 when OCRnA is set to zero (0x0000). 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) provides 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 cleared
on the Compare Match between TCNTn and OCRnx, and set at BOTTOM. In inverting Compare
Output mode 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 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 frequency 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 illustrating 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 Compare Match occurs.
125
2490N–AVR–05/08
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
OCnx
(COMnx1:0 = 2)
OCnx
(COMnx1:0 = 3)
Period
1
2
3
4
5
6
7
8
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 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 starting 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 134). 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 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).
126
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 polarity 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 waveform generated will have
a maximum frequency of fOCnA = fclk_I/O/2 when OCRnA is set to zero (0x0000). This feature is
similar to the OCnA toggle in CTC mode, except the double 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 dualslope 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.
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 represent Compare Matches between OCRnx and TCNTn. The OCnx interrupt flag will be set when a Compare Match occurs.
127
2490N–AVR–05/08
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
OCnx
(COMnx1:0 = 2)
OCnx
(COMnx1:0 = 3)
Period
1
2
3
4
The Timer/Counter Overflow Flag (TOVn) is set each time the counter reaches BOTTOM. 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 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. 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 at TOP, the PWM period starts and ends at TOP. This implies
that the length of the falling 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 135).
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 generated 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 following equation:
f clk_I/O
f OCnxPCPWM = --------------------------2 ⋅ N ⋅ TOP
128
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 PWM Mode
The phase and frequency correct Pulse Width Modulation, or phase and frequency correct 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 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 Compare Output
mode, the operation is inverted. The dual-slope operation gives a lower maximum operation frequency compared 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 )
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 direction. 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 noninverted 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
Compare Match occurs.
129
2490N–AVR–05/08
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
OCnx
(COMnx1:0 = 2)
OCnx
(COMnx1:0 = 3)
Period
1
2
3
4
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, symmetrical
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 non-inverted PWM and
an inverted PWM output can be generated by setting the COMnx1:0 to three (See Table 60 on
page 135). 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 generated 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
and frequency 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
130
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
output will be continuously low and if set equal to TOP the output will be set to high for noninverted 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 - 1
OCRnx
OCRnx
OCRnx + 1
OCRnx + 2
OCRnx Value
OCFnx
Figure 56 shows the same timing data, but with the prescaler enabled.
Figure 56. Timer/Counter Timing Diagram, Setting of OCFnx, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O /8)
TCNTn
OCRnx
OCRnx - 1
OCRnx
OCRnx + 1
OCRnx + 2
OCRnx Value
OCFnx
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.
131
2490N–AVR–05/08
Figure 57. Timer/Counter Timing Diagram, no Prescaling
clkI/O
clkTn
(clkI/O /1)
TCNTn
(CTC and FPWM)
TCNTn
(PC and PFC PWM)
TOP - 1
TOP
BOTTOM
BOTTOM + 1
TOP - 1
TOP
TOP - 1
TOP - 2
TOVn (FPWM)
and ICFn (if used
as TOP)
OCRnx
Old OCRnx Value
(Update at TOP)
New OCRnx Value
Figure 58 shows the same timing data, but with the prescaler enabled.
Figure 58. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O /8)
TCNTn
(CTC and FPWM)
TCNTn
(PC and PFC PWM)
TOP - 1
TOP
BOTTOM
BOTTOM + 1
TOP - 1
TOP
TOP - 1
TOP - 2
TOVn (FPWM)
and ICFn (if used
as TOP)
OCRnx
Old OCRnx Value
(Update at TOP)
New OCRnx Value
16-bit
Timer/Counter
Register
Description
TCCR1A –
Timer/Counter1
Control Register A
TCCR3A –
Timer/Counter3
Control Register A
Bit
7
6
5
4
3
2
1
0
0x2F (0x4F)
COM1A1
COM1A0
COM1B1
COM1B0
COM1C1
COM1C0
WGM11
WGM10
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
(0x8B)
Read/Write
132
7
6
5
4
3
2
1
0
COM3A1
COM3A0
COM3B1
COM3B0
COM3C1
COM3C0
WGM31
WGM30
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
TCCR1A
TCCR3A
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Initial Value
0
0
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 written 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 functionality 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 functionality when
the WGMn3:0 bits are set to a Normal or a CTC mode (non-PWM).
133
2490N–AVR–05/08
Table 58. Compare Output Mode, non-PWM
COMnA1/
COMnB1/
COMnC1
COMnA0/
COMnB0/
COMnC0
0
0
Normal port operation, OCnA/OCnB/OCnC disconnected.
0
1
Toggle OCnA/OCnB/OCnC on Compare Match.
1
0
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).
Description
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
0
0
Normal port operation, OCnA/OCnB/OCnC disconnected.
0
1
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
0
Clear OCnA/OCnB/OCnC on Compare Match, set
OCnA/OCnB/OCnC at BOTTOM (non-inverting mode).
1
1
Set OCnA/OCnB/OCnC on Compare Match, clear
OCnA/OCnB/OCnC at BOTTOM (inverting mode).
Note:
Description
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 BOTTOM. See “Fast PWM Mode” on page 125. 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.
134
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 60. Compare Output Mode, Phase Correct and Phase and Frequency Correct PWM(1)
COMnA1/
COMnB1/
COMnC1
COMnA0/
COMnB0/
COMnC0
0
0
Normal port operation, OCnA/OCnB/OCnC disconnected.
0
1
WGMn3:0 = 9 or 11: Toggle OCnA on Compare Match,
OCnB/OCnC disconnected (normal port operation).
Forr all other WGMn settings, normal port operation,
OCnA/OCnB/OCnC disconnected.
1
0
Clear OCnA/OCnB/OCnC on Compare Match when upcounting. Set OCnA/OCnB/OCnC on Compare Match when
downcounting.
1
1
Set OCnA/OCnB/OCnC on Compare Match when up-counting.
Clear OCnA/OCnB/OCnC on Compare Match when
downcounting.
Note:
Description
1. A
special
case
occurs
when
OCRnA/OCRnB/OCRnC
equals
TOP
and
COMnA1/COMnB1/COMnC1 is set. See “Phase Correct PWM Mode” on page 127. 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 supported 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 124.)
135
2490N–AVR–05/08
Table 61. Waveform Generation Mode Bit Description
Mode
WGMn3
WGMn2
(CTCn)
WGMn1
(PWMn1)
WGMn0
(PWMn0)
Timer/Counter Mode of
Operation
TOP
Update of
OCRnx at
TOVn Flag
Set on
0
0
0
0
0
Normal
0xFFFF
Immediate
MAX
1
0
0
0
1
PWM, Phase Correct, 8-bit
0x00FF
TOP
BOTTOM
2
0
0
1
0
PWM, Phase Correct, 9-bit
0x01FF
TOP
BOTTOM
3
0
0
1
1
PWM, Phase Correct, 10-bit
0x03FF
TOP
BOTTOM
4
0
1
0
0
CTC
OCRnA
Immediate
MAX
5
0
1
0
1
Fast PWM, 8-bit
0x00FF
BOTTOM
TOP
6
0
1
1
0
Fast PWM, 9-bit
0x01FF
BOTTOM
TOP
7
0
1
1
1
Fast PWM, 10-bit
0x03FF
BOTTOM
TOP
8
1
0
0
0
PWM, Phase and Frequency
Correct
ICRn
BOTTOM
BOTTOM
9
1
0
0
1
PWM, Phase and Frequency
Correct
OCRnA
BOTTOM
BOTTOM
10
1
0
1
0
PWM, Phase Correct
ICRn
TOP
BOTTOM
11
1
0
1
1
PWM, Phase Correct
OCRnA
TOP
BOTTOM
12
1
1
0
0
CTC
ICRn
Immediate
MAX
13
1
1
0
1
(Reserved)
–
–
–
14
1
1
1
0
Fast PWM
ICRn
BOTTOM
TOP
15
1
1
1
1
Fast PWM
OCRnA
BOTTOM
TOP
Note:
The CTCn and PWMn1:0 bit definition names are obsolete. Use the WGMn2:0 definitions. However, the functionality and location of these bits are compatible with previous versions of the timer.
TCCR1B –
Timer/Counter1
Control Register B
TCCR3B –
Timer/Counter3
Control Register B
Bit
7
6
5
4
3
2
1
0
0x2E (0x4E)
ICNC1
ICES1
–
WGM13
WGM12
CS12
CS11
CS10
Read/Write
R/W
R/W
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
ICNC3
ICES3
–
WGM33
WGM32
CS32
CS31
CS30
Read/Write
R/W
R/W
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
(0x8A)
TCCR1B
TCCR3B
• Bit 7 – ICNCn: Input Capture Noise Canceler
Setting this bit (to one) activates the Input Capture Noise Canceler. When the noise canceler 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.
136
ATmega64(L)
2490N–AVR–05/08
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 capture
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
No clock source (Timer/counter stopped).
0
0
1
clkI/O/1 (No prescaling)
0
1
0
clkI/O/8 (From prescaler)
0
1
1
clkI/O/64 (From prescaler)
1
0
0
clkI/O/256 (From prescaler)
1
0
1
clkI/O/1024 (From prescaler)
1
1
0
External clock source on Tn pin. Clock on falling edge.
1
1
1
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.
TCCR1C –
Timer/Counter1
Control Register C
Bit
7
6
5
4
3
2
1
FOC1A
FOC1B
FOC1C
–
–
–
–
–
Read/Write
W
W
W
R
R
R
R
R
Initial Value
0
0
0
0
0
0
0
0
(0x7A)
0
TCCR1C
137
2490N–AVR–05/08
TCCR3C –
Timer/Counter3
Control Register C
Bit
7
6
5
4
3
2
1
FOC3A
FOC3B
FOC3C
–
–
–
–
–
Read/Write
W
W
W
R
R
R
R
R
Initial Value
0
0
0
0
0
0
0
0
(0x8C)
0
TCCR3C
• 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.
TCNT1H and TCNT1L
– Timer/Counter1
TCNT3H and TCNT3L
– Timer/Counter3
Bit
7
6
5
4
3
0x2D (0x4D)
TCNT1[15:8]
0x2C (0x4C)
TCNT1[7:0]
2
1
0
TCNT1H
TCNT1L
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
(0x89)
TCNT3[15:8]
(0x88)
TCNT3[7:0]
TCNT3H
TCNT3L
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
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 temporary High Byte Register
(TEMP). This temporary register is shared by all the other 16-bit registers. See “Accessing 16-bit
Registers” on page 115.
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.
138
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
OCR1AH and OCR1AL
–Output Compare
Register 1 A
OCR1BH and OCR1BL
– Output Compare
Register 1 B
OCR1CH and OCR1CL
– Output Compare
Register 1 C
OCR3AH and OCR3AL
– Output Compare
Register 3 A
OCR3BH and OCR3BL
– Output Compare
Register 3 B
OCR3CH and OCR3CL
– Output Compare
Register 3 C
Bit
7
6
5
4
3
0x2B (0x4B)
OCR1A[15:8]
0x2A (0x4A)
OCR1A[7:0]
2
1
0
OCR1AH
OCR1AL
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
0x29 (0x49)
OCR1B[15:8]
0x28 (0x48)
OCR1B[7:0]
OCR1BH
OCR1BL
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
(0x79)
OCR1C[15:8]
(0x78)
OCR1C[7:0]
OCR1CH
OCR1CL
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
(0x87)
OCR3A[15:8]
(0x86)
OCR3A[7:0]
OCR3AH
OCR3AL
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
(0x85)
OCR3B[15:8]
(0x84)
OCR3B[7:0]
OCR3BH
OCR3BL
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
(0x83)
OCR3C[15:8]
(0x82)
OCR3C[7:0]
OCR3CH
OCR3CL
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
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 16bit registers. See “Accessing 16-bit Registers” on page 115.
139
2490N–AVR–05/08
ICR1H and ICR1L –
Input Capture Register
1
ICR3H and ICR3L –
Input Capture Register
3
Bit
7
6
5
4
3
0x27 (0x47)
ICR1[15:8]
0x26 (0x46)
ICR1[7:0]
2
1
0
ICR1H
ICR1L
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Bit
7
6
5
4
3
2
1
0
(0x81)
ICR3[15:8]
(0x80)
ICR3[7:0]
ICR3H
ICR3L
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
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 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 115.
TIMSK –
Timer/Counter
Interrupt Mask
Register(1)
Bit
7
6
5
4
3
2
1
0
0x37 (0x57)
OCIE2
TOIE2
TICIE1
OCIE1A
OCIE1B
TOIE1
OCIE0
TOIE0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Note:
TIMSK
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 globally
enabled), the Timer/Counter1 Input Capture interrupt is enabled. The corresponding Interrupt
Vector (see “Interrupts” on page 61) 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 globally
enabled), the Timer/Counter1 Output Compare A Match interrupt is enabled. The corresponding
Interrupt Vector (see “Interrupts” on page 61) 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 globally
enabled), the Timer/Counter1 Output Compare B Match interrupt is enabled. The corresponding
Interrupt Vector (see “Interrupts” on page 61) is executed when the OCF1B flag, located in TIFR,
is set.
140
ATmega64(L)
2490N–AVR–05/08
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 globally
enabled), the Timer/Counter1 Overflow Interrupt is enabled. The corresponding Interrupt Vector
(see “Interrupts” on page 61) is executed when the TOV1 flag, located in TIFR, is set.
ETIMSK – Extended
Timer/Counter
Interrupt Mask
Register(1)
Bit
7
6
5
4
3
2
1
0
(0x7D)
–
–
TICIE3
OCIE3A
OCIE3B
TOIE3
OCIE3C
OCIE1C
Read/Write
R
R
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Note:
ETIMSK
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 globally
enabled), the Timer/Counter3 Input Capture interrupt is enabled. The corresponding Interrupt
Vector (see “Interrupts” on page 61) 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 globally
enabled), the Timer/Counter3 Output Compare A Match interrupt is enabled. The corresponding
Interrupt Vector (see “Interrupts” on page 61) 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 globally
enabled), the Timer/Counter3 Output Compare B Match interrupt is enabled. The corresponding
Interrupt Vector (see “Interrupts” on page 61) 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 globally
enabled), the Timer/Counter3 Overflow Interrupt is enabled. The corresponding Interrupt Vector
(see “Interrupts” on page 61) 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 globally
enabled), the Timer/Counter3 Output Compare C Match interrupt is enabled. The corresponding
Interrupt Vector (see “Interrupts” on page 61) is executed when the OCF3C flag, located in
ETIFR, is set.
• 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 globally
enabled), the Timer/Counter1 Output Compare C Match interrupt is enabled. The corresponding
Interrupt Vector (see “Interrupts” on page 61) is executed when the OCF1C flag, located in
ETIFR, is set.
141
2490N–AVR–05/08
TIFR – Timer/Counter
Interrupt Flag
Register(1)
Bit
7
6
5
4
3
2
1
0
0x36 (0x56)
OCF2
TOV2
ICF1
OCF1A
OCF1B
TOV1
OCF0
TOV0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
Note:
TIFR
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. Alternatively,
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 Output
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 Output
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 136 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.
ETIFR – Extended
Timer/Counter
Interrupt Flag Register
Bit
7
6
5
4
3
2
1
0
(0x7C)
–
–
ICF3
OCF3A
OCF3B
TOV3
OCF3C
OCF1C
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
ETIFR
• 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.
142
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
• 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 Output
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 Output
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 104 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 Output
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 Output
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.
143
2490N–AVR–05/08
Timer/Counter3,
Timer/Counter2
and
Timer/Counter1
Prescalers
Timer/Counter3, Timer/Counter2 and Timer/Counter1 share the same prescaler module, 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 synchronization logic. The synchronized (sampled) signal is then passed through the edge detector. Figure
59 shows a functional equivalent block diagram of the Tn synchronization 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 clkT1/clkT2/clkT3 pulse for each positive (CSn2:0 = 7) or negative (CSn2:0 = 6) edge it detects.
Figure 59. Tn Pin Sampling
Tn
D Q
D Q
D
Tn_sync
(To Clock
Select Logic)
Q
LE
clk I/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.
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 sys-
144
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
tem 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 variation 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
PSR321
T3
T2
0
SFIOR – Special
Function IO Register
T1
0
0
CS30
CS20
CS31
CS21
CS10
CS11
CS32
CS22
CS12
TIMER/COUNTER3 CLOCK SOURCE
clkT3
Note:
CK/1024
CK/64
CK/8
CK/256
10-BIT T/C PRESCALER
Clear
TIMER/COUNTER2 CLOCK SOURCE
clkT2
TIMER/COUNTER1 CLOCK SOURCE
clkT1
1. The synchronization logic on the input pins (T3/T2/T1) is shown in Figure 59.
Bit
7
6
5
4
3
2
1
0
0x20 (0x40)
TSM
–
–
–
ACME
PUD
PSR0
PSR321
Read/Write
R/W
R
R
R
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
SFIOR
• 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 corresponding 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 prescaler 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.
145
2490N–AVR–05/08
8-bit
Timer/Counter2
with PWM
Timer/Counter2 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)
• 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 157.
Figure 61. 8-bit Timer/Counter Block Diagram
TCCRn
count
TOVn
(Int.Req.)
clear
Control Logic
direction
clk Tn
Clock Select
Edge
Detector
DATA BUS
BOTTOM
Tn
TOP
( 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).
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 Waveform Generator to gener-
146
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
ate a PWM or variable frequency output on the Output Compare pin (OC2). For details, see
“Output Compare Unit” on page 148. 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
Timer/Counter
Clock Sources
BOTTOM
The counter reaches the BOTTOM when it becomes 0x00.
MAX
The counter reaches its MAXimum when it becomes 0xFF (decimal 255).
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 the
fixed value 0xFF (MAX) or the value stored in the OCR2 Register. The
assignment is dependent on the mode of operation.
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 144.
147
2490N–AVR–05/08
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
TCNTn
clear
Control Logic
clkTn
Edge
Detector
Tn
direction
( From Prescaler )
bottom
top
Signal description (internal signals):
count
Increment or decrement TCNT2 by 1.
direction
Select between increment and decrement.
clear
Clear TCNT2 (set all bits to zero).
clkTn
Timer/counter clock, referred to as clkT0 in the following.
top
Signalize that TCNT2 has reached maximum value.
bottom
Signalize that TCNT2 has reached minimum value (zero).
Depending of the mode of operation used, the counter is cleared, incremented, or decremented
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 151.
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
148
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 Generator for handling the special cases of the extreme values in some
modes of operation (see “Modes of Operation” on page 151). Figure 63 shows a block diagram
of the Output Compare unit.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 63. Output Compare Unit, Block Diagram
DATA BUS
OCRn
TCNTn
= (8-bit Comparator )
OCFn (Int.Req.)
top
bottom
Waveform Generator
OCn
FOCn
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 synchronization 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.
149
2490N–AVR–05/08
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 Output 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 Generator 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
COMn0
FOCn
Waveform
Generator
D
Q
1
OCn
DATA BUS
D
0
OCn
Pin
Q
PORT
D
Q
DDR
clk I/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 direction (input or output)
is still controlled by the Data Direction Register (DDR) for the port pin. The Data Direction Regis150
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
ter 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 157.
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 compare output actions in the nonPWM modes refer to Table 65 on page 158. For fast PWM mode, refer to Table 66 on page 158,
and for phase correct PWM refer to Table 67 on page 159.
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 150).
For detailed timing information refer to Figure 68, Figure 69, Figure 70, and Figure 71 in
“Timer/Counter Timing Diagrams” on page 155.
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.
151
2490N–AVR–05/08
Figure 65. CTC Mode, Timing Diagram
OCn Interrupt Flag Set
TCNTn
OCn
(Toggle)
Period
(COMn1:0 = 1)
1
2
3
4
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 frequency 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 frequency
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 BOTTOM. 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 non-inverted and inverted
PWM outputs. The small horizontal line marks on the TCNT2 slopes represent Compare
Matches between OCR2 and TCNT2.
152
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 66. Fast PWM Mode, Timing Diagram
OCRn Interrupt Flag Set
OCRn Update and
TOVn Interrupt Flag Set
TCNTn
OCn
(COMn1:0 = 2)
OCn
(COMn1:0 = 3)
Period
1
2
3
4
5
6
7
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 compare 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 158). 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 setting) 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 output 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.
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 noninverting 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 downcount-
153
2490N–AVR–05/08
ing. 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 represent 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
OCn
(COMn1:0 = 2)
OCn
(COMn1:0 = 3)
Period
1
2
3
The Timer/Counter Overflow Flag (TOV2) is set each time the counter reaches BOTTOM. 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 159). 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 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 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
154
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 noninverted 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.
Timer/Counter
Timing Diagrams
•
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 upcounting 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.
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
MAX - 1
MAX
BOTTOM
BOTTOM + 1
TOVn
Figure 69 shows the same timing data, but with the prescaler enabled.
155
2490N–AVR–05/08
Figure 69. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)
clkI/O
clkTn
(clkI/O /8)
TCNTn
MAX - 1
MAX
BOTTOM
BOTTOM + 1
TOVn
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
OCRn - 1
OCRn
OCRn + 1
OCRn + 2
OCRn Value
OCFn
Figure 71 shows the setting of OCF2 and the clearing of TCNT2 in CTC mode.
156
ATmega64(L)
2490N–AVR–05/08
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
OCRn
BOTTOM + 1
TOP
OCFn
8-bit
Timer/Counter
Register
Description
TCCR2 –
Timer/Counter Control
Register
Bit
7
6
5
4
3
2
1
0
0x25 (0x45)
FOC2
WGM20
COM21
COM20
WGM21
CS22
CS21
CS20
Read/Write
W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
TCCR2
• 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 operation 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 151.
157
2490N–AVR–05/08
Table 64. Waveform Generation Mode Bit Description(1)
Mode
WGM21
(CTC2)
WGM20
(PWM2)
Timer/Counter Mode
of Operation
TOP
Update of
OCR2
TOV2 Flag
Set on
0
0
0
Normal
0xFF
Immediate
MAX
1
0
1
PWM, Phase Correct
0xFF
TOP
BOTTOM
2
1
0
CTC
OCR2
Immediate
MAX
3
1
1
Fast PWM
0xFF
BOTTOM
MAX
Note:
1. The CTC2 and PWM2 bit definition names are now obsolete. Use the WGM21:0 definitions.
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 corresponding 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
Normal port operation, OC2 disconnected.
0
1
Toggle OC2 on Compare Match.
1
0
Clear OC2 on Compare Match.
1
1
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
0
0
Normal port operation, OC2 disconnected.
0
1
Reserved
1
0
Clear OC2 on Compare Match, set OC2 at BOTTOM,
(non-inverting mode).
1
1
Set OC2 on Compare Match, clear OC2 at BOTTOM,
(inverting mode).
Note:
Description
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 BOTTOM. See “Fast PWM Mode” on page 152
for more details.
Table 67 shows the COM21:0 bit functionality when the WGM21:0 bits are set to phase correct
PWM mode.
158
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 67. Compare Output Mode, Phase Correct PWM Mode(1)
COM21
COM20
0
0
Normal port operation, OC2 disconnected.
0
1
Reserved
1
0
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:
Description
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
153 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
No clock source (Timer/counter stopped).
0
0
1
clkI/O/(No prescaling)
0
1
0
clkI/O/8 (From prescaler)
0
1
1
clkI/O/64 (From prescaler)
1
0
0
clkI/O/256 (From prescaler)
1
0
1
clkI/O/1024 (From prescaler)
1
1
0
External clock source on T2 pin. Clock on falling edge.
1
1
1
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.
TCNT2 –
Timer/Counter
Register
Bit
7
6
5
0x24 (0x44)
4
3
2
1
0
TCNT2[7:0]
TCNT2
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
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.
159
2490N–AVR–05/08
OCR2 – Output
Compare Register
Bit
7
6
5
0x23 (0x43)
4
3
2
1
0
OCR2[7:0]
OCR2
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
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.
TIMSK –
Timer/Counter
Interrupt Mask
Register
Bit
7
6
5
4
3
2
1
0
0x37 (0x57)
OCIE2
TOIE2
TICIE1
OCIE1A
OCIE1B
TOIE1
OCIE0
TOIE0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
TIMSK
• 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.
TIFR – Timer/Counter
Interrupt Flag Register
Bit
7
6
5
4
3
2
1
0
0x36 (0x56)
OCF2
TOV2
ICF1
OCF1A
OCF1B
TOV1
OCF0
TOV0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
TIFR
• 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 Compare 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.
160
ATmega64(L)
2490N–AVR–05/08
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 112 and “8-bit Timer/Counter2 with PWM” on page 146. Note that this feature is not
available in ATmega103 compatibility mode.
Figure 72. Output Compare Modulator, Block Diagram
Timer/Counter1
OC1C
Pin
Timer/Counter2
OC2
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 automatically enabled.
The functional equivalent schematic of the modulator is shown on Figure 73. The schematic
includes part of the Timer/Counter units and the Port B pin 7 output driver circuit.
161
2490N–AVR–05/08
Figure 73. Output Compare Modulator, Schematic
COM21
COM20
Vcc
COM1C1
COM1C0
Modulator
0
( From Waveform Generator )
D
1
Q
1
OC1C
Pin
0
( From Waveform Generator )
D
OC1C /
OC2 / PB7
Q
OC2
D
Q
D
PORTB7
Q
DDRB7
DATA BUS
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
clk I/O
OC1C
(FPWM Mode)
OC2
(CTC Mode)
PB7
(PORTB7 = 0)
PB7
(PORTB7 = 1)
(Period)
1
2
3
In this example, Timer/Counter2 provides the carrier, while the modulating signal is generated
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.
162
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
SPI – Serial
Peripheral
Interface
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)
SPI2X
SPI2X
DIVIDER
/2/4/8/16/32/64/128
Note:
1. Refer to Figure 1 on page 2, and Table 30 on page 74 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 Master 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 interchange 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 done, writing a
byte to the SPI Data Register starts the SPI clock generator, and the hardware shifts the eight
163
2490N–AVR–05/08
bits into the Slave. After shifting one byte, the SPI clock generator stops, setting the end of
transmission flag (SPIF). If the SPI interrupt enable bit (SPIE) in the SPCR Register is set, an
interrupt is requested. The 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
MISO
MISO
8 BIT SHIFT REGISTER
MSB
SLAVE
LSB
8 BIT SHIFT REGISTER
MOSI
MOSI
SHIFT
ENABLE
SPI
CLOCK GENERATOR
SCK
SS
VCC
SCK
SS
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 minimum low and high period should be:
Low periods: Longer than 2 CPU clock cycles.
High periods: Longer than 2 CPU clock cycles.
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 71.
Table 69. SPI Pin Overrides(1)
Pin
Direction, Master SPI
Direction, Slave SPI
MOSI
User Defined
Input
MISO
Input
User Defined
SCK
User Defined
Input
SS
User Defined
Input
Note:
164
1. See “Alternate Functions of Port B” on page 74 for a detailed description of how to define the
direction of the user defined SPI pins.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
The following code examples show how to initialize the SPI as a Master and how to perform 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
r17,(1<<DD_MOSI)|(1<<DD_SCK)
out
DDR_SPI,r17
; Enable SPI, Master, set clock rate fck/16
ldi
r17,(1<<SPE)|(1<<MSTR)|(1<<SPR0)
out
SPCR,r17
ret
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. See “About Code Examples” on page 9.
165
2490N–AVR–05/08
The following code examples show how to initialize the SPI as a Slave and how to perform a
simple reception.
Assembly Code Example(1)
SPI_SlaveInit:
; Set MISO output, all others input
ldi
r17,(1<<DD_MISO)
out
DDR_SPI,r17
; Enable SPI
ldi
r17,(1<<SPE)
out
SPCR,r17
ret
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:
166
1. See “About Code Examples” on page 9.
ATmega64(L)
2490N–AVR–05/08
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 except MISO which can be user
configured as an output, 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 synchronous
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.
SPCR – SPI Control
Register
Bit
7
6
5
4
3
2
1
0
0x0D (0x2D)
SPIE
SPE
DORD
MSTR
CPOL
CPHA
SPR1
SPR0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
SPCR
• 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.
167
2490N–AVR–05/08
• 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 functionality is summarized below:
Table 70. CPOL Functionality
CPOL
Leading Edge
Trailing Edge
0
Rising
Falling
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
Trailing Edge
0
Sample
Setup
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
168
SPI2X
SPR1
SPR0
0
0
0
0
0
1
0
1
0
0
1
1
1
0
0
1
0
1
1
1
0
1
1
1
SCK Frequency
fosc/4
fosc/16
fosc/64
fosc/128
fosc/2
fosc/8
fosc/32
fosc/64
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
SPSR – SPI Status
Register
Bit
7
6
5
4
3
2
1
0
0x0E (0x2E)
SPIF
WCOL
–
–
–
–
–
SPI2X
Read/Write
R
R
R
R
R
R
R
R/W
Initial Value
0
0
0
0
0
0
0
0
SPSR
• 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 hardware 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 guaranteed 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 305 for SPI Serial Programming and verification.
SPDR – SPI Data
Register
Bit
7
6
5
4
3
2
1
0
0x0F (0x2F)
MSB
LSB
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
X
X
X
X
X
X
X
X
SPDR
Undefined
The SPI Data Register is a read/write register used for data transfer between the Register File
and the SPI Shift Register. Writing to the register initiates data transmission. Reading the register causes the Shift Register Receive buffer to be read.
169
2490N–AVR–05/08
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 opposite 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
Trailing Edge
SPI Mode
CPOL = 0, CPHA = 0
Sample (Rising)
Setup (Falling)
0
CPOL = 0, CPHA = 1
Setup (Rising)
Sample (Falling)
1
CPOL = 1, CPHA = 0
Sample (Falling)
Setup (Rising)
2
CPOL = 1, CPHA = 1
Setup (Falling)
Sample (Rising)
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)
170
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
ATmega64(L)
2490N–AVR–05/08
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 370. 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 accessible 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
CONTROL
XCK
DATA BUS
Transmitter
PARITY
GENERATOR
PIN
CONTROL
TRANSMIT SHIFT REGISTER
TxD
Receiver
UCSRA
Note:
TX
CONTROL
UDR (Transmit)
CLOCK
RECOVERY
RX
CONTROL
RECEIVE SHIFT REGISTER
DATA
RECOVERY
PIN
CONTROL
UDR (Receive)
PARITY
CHECKER
UCSRB
RxD
UCSRC
1. Refer to Figure 1 on page 2, Table 36 on page 78, and Table 39 on page 81 for USART pin
placement.
171
2490N–AVR–05/08
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 Transmitter 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 continuous 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 recovery 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 (UDRn). The Receiver supports the same frame formats as the Transmitter, 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 UDRn must only be read once for each incoming data! More important
is the fact that the error flags (FEn and DORn) and the ninth data bit (RXB8n) are buffered
with the data in the receive buffer. Therefore the status bits must always be read before the
UDRn 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 Over
Run (DORn) error conditions.
The following control bits have changed name, but have same functionality and register location:
Clock Generation
•
CHR9 is changed to UCSZn2.
•
OR is changed to DORn.
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 UMSELn bit in USART
Control and Status Register n C (UCSRnC) selects between asynchronous and synchronous
operation. Double Speed (asynchronous mode only) is controlled by the U2Xn found in the
UCSRnB Register. When using synchronous mode (UMSELn = 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.
172
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 80. Clock Generation Logic, Block Diagram
UBRR
U2X
fosc
Prescaling
Down-Counter
UBRR+1
/2
/4
/2
0
1
0
OSC
DDR_XCK
xcki
XCK
Pin
Sync
Register
Edge
Detector
0
UCPOL
txclk
UMSEL
1
xcko
DDR_XCK
1
1
0
rxclk
Signal description:
Internal Clock
Generation – The
Baud Rate Generator
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 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 n (UBRRn) and the down-counter connected to it function as a
programmable prescaler or baud rate generator. The down-counter, running at system clock
(fOSC), is loaded with the UBRRn value each time the counter has counted down to zero or when
the UBRRnL Register is written. A clock is generated each time the counter reaches zero. This
clock is the baud rate generator clock output (= fOSC/(UBRRn+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
UMSELn, U2Xn and DDR_XCK bits.
Table 74 contains equations for calculating the baud rate (in bits per second) and for calculating
the UBRRn value for each mode of operation using an internally generated clock source.
173
2490N–AVR–05/08
Table 74. Equations for Calculating Baud Rate Register Setting
Operating Mode
Equation for Calculating
Baud Rate(1)
Equation for Calculating
UBRR Value
Asynchronous Normal
mode (U2Xn = 0)
f OSC
BAUD = ----------------------------------------16 ( UBRR + 1n )
f OSC
UBRRn = -----------------------–1
16BAUD
Asynchronous Double
Speed mode (U2Xn = 1)
f OSC
BAUD = -------------------------------------8 ( UBRRn + 1 )
f OSC
-–1
UBRRn = ------------------8BAUD
Synchronous Master
mode
f OSC
BAUD = -------------------------------------2 ( UBRR + 1n )
f OSC
-–1
UBRRn = ------------------2BAUD
Note:
1. The baud rate is defined to be the transfer rate in bit per second (bps).
BAUD Baud rate (in bits per second, bps)
fOSC
System Oscillator clock frequency
UBRR Contents of the UBRRnH and UBRRnL Registers, (0 - 4095)
Some examples of UBRRn values for some system clock frequencies are found in Table 82 on
page 194 to Table 85 on page 197.
Double Speed
Operation (U2Xn)
The transfer rate can be doubled by setting the U2Xn bit in UCSRnB. 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 minimize 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 maximum external XCK clock frequency
is limited by the following equation:
f OSC
f XCK < ----------4
Note that fosc depends on the stability of the system clock source. It is therefore recommended to
add some margin to avoid possible loss of data due to frequency variations.
Synchronous Clock
Operation
174
When synchronous mode is used (UMSELn = 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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 81. Synchronous Mode XCK Timing
UCPOLn = 1
XCK
RxD / TxD
Sample
UCPOLn = 0
XCK
RxD / TxD
Sample
The UCPOLn bit UCRSnC selects which XCK clock edge is used for data sampling and which is
used for data change. As Figure 81 shows, when UCPOLn is zero the data will be changed at
rising XCK edge and sampled at falling XCK edge. If UCPOLn 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 complete 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
Start bit, always low.
(n)
Data bits (0 to 8).
P
Parity bit. Can be odd or even.
Sp
Stop bit, always high.
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 UCSZn2:0, UPMn1:0 and USBSn bits in
UCSRnB and UCSRnC. The Receiver and Transmitter use the same setting. Note that changing
175
2490N–AVR–05/08
the setting of any of these bits will corrupt all ongoing communication for both the Receiver and
Transmitter.
The USART Character Size (UCSZn2:0) bits select the number of data bits in the frame. The
USART Parity mode (UPMn1:0) bits enable and set the type of parity bit. The selection between
one or two stop bits is done by the USART Stop Bit Select n (USBSn) bit. The receiver ignores
the second stop bit. An FEn (Frame Error n) 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 even = d n – 1 ⊕ … ⊕ d 3 ⊕ d 2 ⊕ d 1 ⊕ d 0 ⊕ 0
P odd = d n – 1 ⊕ … ⊕ d 3 ⊕ d 2 ⊕ d 1 ⊕ d 0 ⊕ 1
Peven
Parity bit using even parity
Podd
Parity bit using odd parity
dn
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 initialization 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 TXCn flag can be used
to check that the Transmitter has completed all transfers, and the RXCn flag can be used to
check that there are no unread data in the receive buffer. Note that the TXCn flag must be
cleared before each transmission (before UDRn 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 operation 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.
176
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Assembly Code Example(1)
USART_Init:
; Set baud rate
out
UBRRnH, r17
out
UBRRnL, r16
; Enable receiver and transmitter
ldi
r16, (1<<RXENn)|(1<<TXENn)
out
UCSRnB,r16
; Set frame format: 8data, 2stop bit
ldi
r16, (1<<USBSn)|(3<<UCSZn0)
out
UCSRnC,r16
ret
C Code Example(1)
#define FOSC 1843200// Clock Speed
#define BAUD 9600
#define MYUBRR FOSC/16/BAUD-1
void main( void )
{
...
USART_Init ( MYUBRR );
...
}
void USART_Init( unsigned int ubrr )
{
/* Set baud rate */
UBRRnH = (unsigned char)(ubrr>>8);
UBRRnL = (unsigned char)ubrr;
/* Enable receiver and transmitter */
UCSRnB = (1<<RXENn)|(1<<TXENn);
/* Set frame format: 8data, 2stop bit */
UCSRnC = (1<<USBSn)|(3<<UCSZn0);
}
Note:
1. See “About Code Examples” on page 9.
More advanced initialization routines can be made that include frame format as parameters, 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.
177
2490N–AVR–05/08
Data Transmission The USART Transmitter is enabled by setting the Transmit Enable (TXENn) bit in the UCSRnB
Register. When the Transmitter is enabled, the normal port operation of the TxD pin is overrid– The USART
den by the USART and given the function as the transmitter’s serial output. The baud rate, mode
Transmitter
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 transmitted. The
CPU can load the transmit buffer by writing to the UDRn 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 previous 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,
U2Xn bit or by XCK depending on mode of operation.
The following code examples show a simple USART transmit function based on polling of the
Data Register Empty (UDREn) flag. When using frames with less than eight bits, the most significant bits written to the UDRn 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 UCSRnA,UDREn
rjmp USART_Transmit
; Put data (r16) into buffer, sends the data
out
UDRn,r16
ret
C Code Example(1)
void USART_Transmit( unsigned char data )
{
/* Wait for empty transmit buffer */
while ( !( UCSRnA & (1<<UDREn)) )
;
/* Put data into buffer, sends the data */
UDRn = data;
}
Note:
1. See “About Code Examples” on page 9.
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 UDREn 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.
178
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Sending Frames with
9 Data Bits
If 9-bit characters are used (UCSZn = 7), the ninth bit must be written to the TXB8n bit in UCSRnB before the low byte of the character is written to UDRn. 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 UCSRnA,UDREn
rjmp USART_Transmit
; Copy ninth bit from r17 to TXB8
cbi
UCSRnB,TXB8n
sbrc r17,0
sbi
UCSRnB,TXB8n
; Put LSB data (r16) into buffer, sends the data
out
UDRn,r16
ret
C Code Example(1)
void USART_Transmit( unsigned int data )
{
/* Wait for empty transmit buffer */
while ( !( UCSRnA & (1<<UDREn)) )
;
/* Copy ninth bit to TXB8 */
UCSRnB &= ~(1<<TXB8n);
if ( data & 0x0100 )
UCSRnB |= (1<<TXB8n);
/* Put data into buffer, sends the data */
UDRn = data;
}
Note:
1. These transmit functions are written to be general functions. They can be optimized if the contents of the UCSRnB is static. For example, only the TXB8n bit of the UCSRnB 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.
179
2490N–AVR–05/08
Transmitter Flags and
Interrupts
The USART Transmitter has two flags that indicate its state: USART Data Register Empty
(UDREn) and Transmit Complete (TXCn). Both flags can be used for generating interrupts.
The Data Register Empty (UDREn) 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 UCSRnA Register.
When the Data Register empty Interrupt Enable (UDRIEn) bit in UCSRnB is written to one, the
USART Data Register Empty Interrupt will be executed as long as UDREn is set (provided that
global interrupts are enabled). UDREn is cleared by writing UDRn. When interrupt-driven data
transmission is used, the Data Register Empty Interrupt routine must either write new data to
UDRn in order to clear UDREn or disable the Data Register Empty Interrupt, otherwise a new
interrupt will occur once the interrupt routine terminates.
The Transmit Complete n (TXCn) 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 TXCn 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 TXCn 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 (TXCIEn) bit in UCSRnB is set, the USART
Transmit Complete Interrupt will be executed when the TXCn 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 TXCn 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
(UPMn1 = 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 TXENn 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 disabled, the transmitter
will no longer override the TxD pin.
180
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Data Reception –
The USART
Receiver
The USART Receiver is enabled by writing the Receive Enable n (RXENn) bit in the UCSRnB
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 UDRn I/O location.
The following code example shows a simple USART receive function based on polling of the
Receive Complete n (RXCn) flag. When using frames with less than eight bits the most significant bits of the data read from the UDRn 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 UCSRnA, RXCn
rjmp USART_Receive
; Get and return received data from buffer
in
r16, UDRn
ret
C Code Example(1)
unsigned char USART_Receive( void )
{
/* Wait for data to be received */
while ( !(UCSRnA & (1<<RXCn)) )
;
/* Get and return received data from buffer */
return UDRn;
}
Note:
1. See “About Code Examples” on page 9.
The function simply waits for data to be present in the receive buffer by checking the RXCn flag,
before reading the buffer and returning the value.
181
2490N–AVR–05/08
Receiving Frames with
9 Data Bits
If 9-bit characters are used (UCSZn=7) the ninth bit must be read from the RXB8n bit in UCSRnB before reading the low bits from the UDR. This rule applies to the FEn, DORn, and UPE
status flags as well. Read status from UCSRnA, then data from UDRn. Reading the UDRn I/O
location will change the state of the receive buffer FIFO and consequently the TXB8n, FEn,
DORn, and UPEn bits, which all are stored in the FIFO, will change. The following 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 UCSRnA, RXCn
rjmp USART_Receive
; Get status and ninth bit, then data from buffer
in
r18, UCSRnA
in
r17, UCSRnB
in
r16, UDRn
; If error, return -1
andi r18,(1<<FEn)|(1<<DORn)|(1<<UPEn)
breq USART_ReceiveNoError
ldi
r17, HIGH(-1)
ldi
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 ( !(UCSRnA & (1<<RXCn)) )
;
/* Get status and ninth bit, then data */
/* from buffer */
status = UCSRnA;
resh = UCSRnB;
resl = UDRn;
/* If error, return -1 */
if ( status & (1<<FEn)|(1<<DORn)|(1<<UPEn) )
return -1;
/* Filter the ninth bit, then return */
resh = (resh >> 1) & 0x01;
return ((resh << 8) | resl);
}
Note:
182
1. See “About Code Examples” on page 9.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 n (RXCn) 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 (RXENn = 0),
the receive buffer will be flushed and consequently the RXCn bit will become zero.
When the Receive Complete Interrupt Enable n (RXCIEn) in UCSRnB is set, the USART
Receive Complete Interrupt will be executed as long as the RXCn 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 UDRn in order to clear the RXCn flag, otherwise a new
interrupt will occur once the interrupt routine terminates.
Receiver Error Flags
The USART Receiver has three error flags: Frame Error n (FEn), Data OverRun n (DORn) and
USART Parity Error n (UPEn). All can be accessed by reading UCSRnA. 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 UCSRnA must be read before the
receive buffer (UDRn), since reading the UDRn 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 UCSRnA is written for upward
compatibility of future USART implementations. None of the error flags can generate interrupts.
The Frame Error n (FEn) flag indicates the state of the first stop bit of the next readable frame
stored in the receive buffer. The FEn flag is zero when the stop bit was correctly read (as one),
and the FEn 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 FEn flag
is not affected by the setting of the USBSn bit in UCSRnC 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 UCSRnA.
The Data OverRun n (DORn) 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 character waiting in the Receive Shift Register, and a new start bit is detected. If the DORn flag is set there
was one or more serial frame lost between the frame last read from UDRn, and the next frame
read from UDRn. For compatibility with future devices, always write this bit to zero when writing
to UCSRnA. The DORn flag is cleared when the frame received was successfully moved from
the Shift Register to the receive buffer.
The USART Parity Error n (UPEn) flag indicates that the next frame in the receive buffer had a
Parity Error when received. If parity check is not enabled the UPEn bit will always be read zero.
For compatibility with future devices, always set this bit to zero when writing to UCSRnA. For
more details see “Parity Bit Calculation” on page 176 and “Parity Checker” on page 183.
Parity Checker
The Parity Checker is active when the high USART Parity mode (UPMn1) bit is set. Type of parity check to be performed (odd or even) is selected by the UPMn0 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 stored in the receive buffer together
with the received data and stop bits. The Parity Error n (UPEn) flag can then be read by software
to check if the frame had a Parity Error.
The UPEn 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 (UPMn1 = 1). This bit is
valid until the receive buffer (UDRn) is read.
183
2490N–AVR–05/08
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 RXENn 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 UDRn I/O location until the RXCn flag
is cleared. The following code examples show how to flush the receive buffer.
Assembly Code Example(1)
USART_Flush:
sbis UCSRnA, RXCn
ret
in
r16, UDRn
rjmp USART_Flush
C Code Example(1)
void USART_Flush( void )
{
unsigned char dummy;
while ( UCSRnA & (1<<RXCn) ) dummy = UDRn;
}
Note:
1. See “About Code Examples” on page 9.
Asynchronous
Data Reception
The USART includes a clock recovery and a data recovery unit for handling asynchronous 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 improving 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. Figure 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 (U2Xn = 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
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1
2
3
Sample
(U2X = 1)
0
1
2
3
4
5
6
7
8
1
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-sample as shown in
184
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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-transition. 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 sampling 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
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1
Sample
(U2X = 1)
1
2
3
4
5
6
7
8
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.
Figure 85. Stop Bit Sampling and Next Start Bit Sampling
RxD
STOP 1
(A)
(B)
(C)
Sample
(U2X = 0)
1
2
3
4
5
6
7
8
9
10
0/1
0/1
0/1
Sample
(U2X = 1)
1
2
3
4
5
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 n (FEn) 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 detection influences the operational range of
the Receiver.
185
2490N–AVR–05/08
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
R slow = ------------------------------------------S – 1 + D ⋅ S + SF
( D + 2 )S
R fast = ----------------------------------( D + 1 )S + S M
D
Sum of character size and parity size (D = 5 to 10 bit)
S
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.
186
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 75. Recommended Maximum Receiver Baud Rate Error for Normal Speed Mode (U2Xn
= 0)
D
# (Data+Parity Bit)
Rslow (%)
Rfast (%)
Max Total
Error (%)
Recommended Max
Receiver Error (%)
5
93.20
106.67
+6.67/-6.8
±3.0
6
94.12
105.79
+5.79/-5.88
±2.5
7
94.81
105.11
+5.11/-5.19
±2.0
8
95.36
104.58
+4.58/-4.54
±2.0
9
95.81
104.14
+4.14/-4.19
±1.5
10
96.17
103.78
+3.78/-3.83
±1.5
Table 76. Recommended Maximum Receiver Baud Rate Error for Double Speed Mode (U2Xn =
1)
D
# (Data+Parity Bit)
Rslow (%)
Rfast (%)
Max Total
Error (%)
Recommended Max
Receiver Error (%)
5
94.12
105.66
+5.66/-5.88
±2.5
6
94.92
104.92
+4.92/-5.08
±2.0
7
95.52
104.35
+4.35/-4.48
±1.5
8
96.00
103.90
+3.90/-4.00
±1.5
9
96.39
103.53
+3.53/-3.61
±1.5
10
96.70
103.23
+3.23/-3.30
±1.0
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 n (MPCMn) bit in UCSRnA enables a filtering
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 Transmitter is unaffected by the MPCMn
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 (RXB8n) 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.
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
187
2490N–AVR–05/08
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 (UCSZn = 7). The
ninth bit (TXB8n) must be set when an address frame (TXB8n = 1) or cleared when a data frame
(TXBn = 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 Communication
mode:
1. All Slave MCUs are in Multi-processor Communication mode (MPCMn in UCSRnA is
set).
2. The Master MCU sends an address frame, and all slaves receive and read this frame. In
the Slave MCUs, the RXCn flag in UCSRnA will be set as normal.
3. Each Slave MCU reads the UDRn Register and determines if it has been selected. If so,
it clears the MPCMn bit in UCSRnA, otherwise it waits for the next address byte and
keeps the MPCMn 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 MPCMn bit set, will ignore the data frames.
5. When the last data frame is received by the addressed MCU, the addressed MCU sets
the MPCMn 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 character size setting. If 5- to 8-bit character frames are used, the Transmitter must be set to use two stop bit
(USBSn = 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
MPCMn bit shares the same I/O location as the TXCn flag and this might accidentally be cleared
when using SBI or CBI instructions.
USART Register
Description
UDRn – USART I/O
Data Register
Bit
7
6
5
4
3
2
1
0
RXB[7:0]
UDnR (Read)
TXB[7:0]
UDnR (Write)
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
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 UDRn. The Transmit Data Buffer Register (TXBn) will be the destination for data written to the UDRn Register location. Reading the
UDRn Register location will return the contents of the Receive Data Buffer Register (RXBn).
For 5-, 6-, or 7-bit characters the upper unused bits will be ignored by the Transmitter and set to
zero by the Receiver.
The transmit buffer can only be written when the UDREn flag in the UCSRnA Register is set.
Data written to UDRn when the UDREn 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
188
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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.
UCSRnA – USART
Control and Status
Register A
Bit
7
6
5
4
3
2
1
0
RXCn
TXCn
UDREn
FEn
DORn
UPEn
U2Xn
MPCMn
Read/Write
R
R/W
R
R
R
R
R/W
R/W
Initial Value
0
0
1
0
0
0
0
0
UCSRnA
• Bit 7 – RXCn: 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 disabled, the receive
buffer will be flushed and consequently the RXCn bit will become zero. The RXCn flag can be
used to generate a Receive Complete interrupt (see description of the RXCIEn bit).
• Bit 6 – TXCn: 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 – UDREn: USART Data Register Empty
The UDREn flag indicates if the transmit buffer (UDR) is ready to receive new data. If UDREn is
one, the buffer is empty, and therefore ready to be written. The UDREn flag can generate a Data
Register Empty interrupt (see description of the UDRIEn bit).
UDREn is set after a reset to indicate that the Transmitter is ready.
• Bit 4 – FEn: 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 – DORn: 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 (UDRn) is read. Always set this
bit to zero when writing to UCSRnA.
• Bit 2 – UPEn: 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
(UDRn) is read. Always set this bit to zero when writing to UCSRnA.
189
2490N–AVR–05/08
• Bit 1 – U2Xn: 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 effectively doubling the transfer rate for asynchronous communication.
• Bit 0 – MPCMn: Multi-processor Communication Mode
This bit enables the Multi-processor Communication Mode. When the MPCMn bit is written 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 MPCMn setting. For more detailed
information see “Multi-processor Communication Mode” on page 187.
UCSRnB – USART
Control and Status
Register B
Bit
7
6
5
4
3
2
1
0
RXCIEn
TXCIEn
UDRIEn
RXENn
TXENn
UCSZn2
RXB8n
TXB8n
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R
R/W
Initial Value
0
0
0
0
0
0
0
0
UCSRnB
• Bit 7 – RXCIEn: RX Complete Interrupt Enable
Writing this bit to one enables interrupt on the RXCn flag. A USART Receive Complete interrupt
will be generated only if the RXCIEn bit is written to one, the Global Interrupt Flag in SREG is
written to one and the RXCn bit in UCSRnA is set.
• Bit 6 – TXCIEn: TX Complete Interrupt Enable
Writing this bit to one enables interrupt on the TXCn flag. A USART Transmit Complete interrupt
will be generated only if the TXCIEn bit is written to one, the Global Interrupt Flag in SREG is
written to one and the TXCn bit in UCSRnA is set.
• Bit 5 – UDRIEn: USART Data Register Empty Interrupt Enable
Writing this bit to one enables interrupt on the UDREn flag. A Data Register Empty interrupt will
be generated only if the UDRIEn bit is written to one, the Global Interrupt Flag in SREG is written
to one and the UDREn bit in UCSRnA is set.
• Bit 4 – RXENn: 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 FEn, DORn, and UPEn flags.
• Bit 3 – TXENn: Transmitter Enable
Writing this bit to one enables the USART Transmitter. The Transmitter will override normal 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 transmissions are completed, i.e., when the
Transmit Shift Register and Transmit Buffer Register do not contain data to be transmitted.
When disabled, the Transmitter will no longer override the TxD port.
• Bit 2 – UCSZn2: Character Size
The UCSZn2 bits combined with the UCSZn1:0 bit in UCSRC sets the number of data bits
(Character Size) in a frame the Receiver and Transmitter use.
190
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
• Bit 1 – RXB8n: Receive Data Bit 8
RXB8n 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 UDRn.
• Bit 0 – TXB8n: Transmit Data Bit 8
TXB8n 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 UDRn.
UCSRnC – USART
Control and Status
Register C(1)
Bit
7
6
5
4
3
2
1
0
–
UMSELn
UPMn1
UPMn0
USBSn
UCSZn1
UCSZn0
UCPOLn
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
1
1
0
Note:
UCSRnC
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 – UMSELn: USART Mode Select
This bit selects between asynchronous and synchronous mode of operation.
Table 77. UMSEL Bit Settings
UMSELn
Mode
0
Asynchronous Operation
1
Synchronous Operation
• Bit 5:4 – UPMn1:0: Parity Mode
These bits enable and set type of parity generation and check. If enabled, the Transmitter 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 compare it to the UPMn0 setting.
If a mismatch is detected, the UPEn flag in UCSRnB will be set.
Table 78. UPM Bits Settings
UPMn1
UPMn0
Parity Mode
0
0
Disabled
0
1
Reserved
1
0
Enabled, Even Parity
1
1
Enabled, Odd Parity
191
2490N–AVR–05/08
• Bit 3 – USBSn: 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
USBSn
Stop Bit(s)
0
1-bit
1
2-bit
• Bit 2:1 – UCSZn1:0: Character Size
The UCSZn1:0 bits combined with the UCSZn2 bit in UCSRnB sets the number of data bits
(Character Size) in a frame the Receiver and Transmitter use.
Table 80. UCSZ Bits Settings
UCSZn2
UCSZn1
UCSZn0
Character Size
0
0
0
5-bit
0
0
1
6-bit
0
1
0
7-bit
0
1
1
8-bit
1
0
0
Reserved
1
0
1
Reserved
1
1
0
Reserved
1
1
1
9-bit
• Bit 0 – UCPOLn: Clock Polarity
This bit is used for synchronous mode only. Write this bit to zero when asynchronous mode is
used. The UCPOLn 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)
0
Rising XCK Edge
Falling XCK Edge
1
Falling XCK Edge
Rising XCK Edge
UCPOLn
192
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
UBRRnL and UBRRnH
– USART Baud Rate
Registers(1)
Bit
15
14
13
12
–
–
–
–
11
10
9
8
UBRRn[11:8]
UBRRnH
UBRRn[7:0]
7
Read/Write
Initial Value
Note:
6
5
UBRRnL
4
3
2
1
0
R
R
R
R
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
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 – UBRRn11:0: USART Baud Rate Register
This is a 12-bit register which contains the USART baud rate. The UBRRnH contains the four
most significant bits, and the UBRRnL contains the eight least significant bits of the USART
baud rate. Ongoing transmissions by the Transmitter and Receiver will be corrupted if the baud
rate is changed. Writing UBRRnL 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 UBRRn settings in Table 82 to Table 85.
UBRRn 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 186). The error values are calculated using the following
equation:
BaudRate Closest Match
- – 1⎞⎠ • 100%
Error[%] = ⎛⎝ ------------------------------------------------------BaudRate
193
2490N–AVR–05/08
Table 82. Examples of UBRR Settings for Commonly Used Oscillator Frequencies
fosc = 1.0000 MHz
fosc = 1.8432 MHz
fosc = 2.0000 MHz
Baud
Rate
(bps)
UBRRn
Error
UBRRn
Errorn
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
2400
25
0.2%
51
0.2%
47
0.0%
95
0.0%
51
0.2%
103
0.2%
4800
12
0.2%
25
0.2%
23
0.0%
47
0.0%
25
0.2%
51
0.2%
9600
6
-7.0%
12
0.2%
11
0.0%
23
0.0%
12
0.2%
25
0.2%
14.4k
3
8.5%
8
-3.5%
7
0.0%
15
0.0%
8
-3.5%
16
2.1%
19.2k
2
8.5%
6
-7.0%
5
0.0%
11
0.0%
6
-7.0%
12
0.2%
28.8k
1
8.5%
3
8.5%
3
0.0%
7
0.0%
3
8.5%
8
-3.5%
38.4k
1
-18.6%
2
8.5%
2
0.0%
5
0.0%
2
8.5%
6
-7.0%
57.6k
0
8.5%
1
8.5%
1
0.0%
3
0.0%
1
8.5%
3
8.5%
76.8k
–
–
1
-18.6%
1
-25.0%
2
0.0%
1
-18.6%
2
8.5%
115.2k
–
–
0
8.5%
0
0.0%
1
0.0%
0
8.5%
1
8.5%
230.4k
–
–
–
–
–
–
0
0.0%
–
–
–
–
250k
–
–
–
–
–
–
–
–
–
–
0
0.0%
Max
1.
194
(1)
U2X = 0
U2X = 1
62.5 kbps
125 kbps
U2X = 0
U2X = 1
115.2 kbps
U2X = 0
230.4 kbps
125 kbps
U2X = 1
250 kbps
UBRR = 0, Error = 0.0%
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 83. Examples of UBRR Settings for Commonly Used Oscillator Frequencies (Continued)
fosc = 3.6864 MHz
fosc = 4.0000 MHz
fosc = 7.3728 MHz
Baud
Rate
(bps)
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
2400
95
0.0%
191
0.0%
103
0.2%
207
0.2%
191
0.0%
383
0.0%
4800
47
0.0%
95
0.0%
51
0.2%
103
0.2%
95
0.0%
191
0.0%
9600
23
0.0%
47
0.0%
25
0.2%
51
0.2%
47
0.0%
95
0.0%
14.4k
15
0.0%
31
0.0%
16
2.1%
34
-0.8%
31
0.0%
63
0.0%
19.2k
11
0.0%
23
0.0%
12
0.2%
25
0.2%
23
0.0%
47
0.0%
28.8k
7
0.0%
15
0.0%
8
-3.5%
16
2.1%
15
0.0%
31
0.0%
38.4k
5
0.0%
11
0.0%
6
-7.0%
12
0.2%
11
0.0%
23
0.0%
57.6k
3
0.0%
7
0.0%
3
8.5%
8
-3.5%
7
0.0%
15
0.0%
76.8k
2
0.0%
5
0.0%
2
8.5%
6
-7.0%
5
0.0%
11
0.0%
115.2k
1
0.0%
3
0.0%
1
8.5%
3
8.5%
3
0.0%
7
0.0%
230.4k
0
0.0%
1
0.0%
0
8.5%
1
8.5%
1
0.0%
3
0.0%
250k
0
-7.8%
1
-7.8%
0
0.0%
1
0.0%
1
-7.8%
3
-7.8%
0.5M
–
–
0
-7.8%
–
–
0
0.0%
0
-7.8%
1
-7.8%
1M
–
–
–
–
–
–
–
–
–
–
0
-7.8%
Max
1.
(1)
U2X = 0
U2X = 1
230.4 kbps
U2X = 0
460.8 kbps
250 kbps
U2X = 1
0.5 Mbps
U2X = 0
U2X = 1
460.8 kbps
921.6 kbps
UBRR = 0, Error = 0.0%
195
2490N–AVR–05/08
Table 84. Examples of UBRR Settings for Commonly Used Oscillator Frequencies (Continued)
fosc = 11.0592 MHz
fosc = 8.0000 MHz
fosc = 14.7456 MHz
Baud
Rate
(bps)
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
2400
207
0.2%
416
-0.1%
287
0.0%
575
0.0%
383
0.0%
767
0.0%
4800
103
0.2%
207
0.2%
143
0.0%
287
0.0%
191
0.0%
383
0.0%
9600
51
0.2%
103
0.2%
71
0.0%
143
0.0%
95
0.0%
191
0.0%
14.4k
34
-0.8%
68
0.6%
47
0.0%
95
0.0%
63
0.0%
127
0.0%
19.2k
25
0.2%
51
0.2%
35
0.0%
71
0.0%
47
0.0%
95
0.0%
28.8k
16
2.1%
34
-0.8%
23
0.0%
47
0.0%
31
0.0%
63
0.0%
38.4k
12
0.2%
25
0.2%
17
0.0%
35
0.0%
23
0.0%
47
0.0%
57.6k
8
-3.5%
16
2.1%
11
0.0%
23
0.0%
15
0.0%
31
0.0%
76.8k
6
-7.0%
12
0.2%
8
0.0%
17
0.0%
11
0.0%
23
0.0%
115.2k
3
8.5%
8
-3.5%
5
0.0%
11
0.0%
7
0.0%
15
0.0%
230.4k
1
8.5%
3
8.5%
2
0.0%
5
0.0%
3
0.0%
7
0.0%
250k
1
0.0%
3
0.0%
2
-7.8%
5
-7.8%
3
-7.8%
6
5.3%
0.5M
0
0.0%
1
0.0%
–
–
2
-7.8%
1
-7.8%
3
-7.8%
1M
–
–
0
0.0%
–
–
–
–
0
-7.8%
1
-7.8%
Max
1.
196
(1)
U2X = 0
U2X = 1
0.5 Mbps
1 Mbps
U2X = 0
U2X = 1
691.2 kbps
U2X = 0
1.3824 Mbps
921.6 kbps
U2X = 1
1.8432 Mbps
UBRR = 0, Error = 0.0%
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 85. Examples of UBRR Settings for Commonly Used Oscillator Frequencies (Continued)
fosc = 16.0000 MHz
fosc = 18.4320 MHz
fosc = 20.0000 MHz
Baud
Rate
(bps)
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
UBRRn
Error
2400
416
-0.1%
832
0.0%
479
0.0%
959
0.0%
520
0.0%
1041
0.0%
4800
207
0.2%
416
-0.1%
239
0.0%
479
0.0%
259
0.2%
520
0.0%
9600
103
0.2%
207
0.2%
119
0.0%
239
0.0%
129
0.2%
259
0.2%
14.4k
68
0.6%
138
-0.1%
79
0.0%
159
0.0%
86
-0.2%
173
-0.2%
19.2k
51
0.2%
103
0.2%
59
0.0%
119
0.0%
64
0.2%
129
0.2%
28.8k
34
-0.8%
68
0.6%
39
0.0%
79
0.0%
42
0.9%
86
-0.2%
38.4k
25
0.2%
51
0.2%
29
0.0%
59
0.0%
32
-1.4%
64
0.2%
57.6k
16
2.1%
34
-0.8%
19
0.0%
39
0.0%
21
-1.4%
42
0.9%
76.8k
12
0.2%
25
0.2%
14
0.0%
29
0.0%
15
1.7%
32
-1.4%
115.2k
8
-3.5%
16
2.1%
9
0.0%
19
0.0%
10
-1.4%
21
-1.4%
230.4k
3
8.5%
8
-3.5%
4
0.0%
9
0.0%
4
8.5%
10
-1.4%
250k
3
0.0%
7
0.0%
4
-7.8%
8
2.4%
4
0.0%
9
0.0%
0.5M
1
0.0%
3
0.0%
–
–
4
-7.8%
–
–
4
0.0%
1M
0
0.0%
1
0.0%
–
–
–
–
–
–
–
–
Max
1.
U2X = 0
(1)
U2X = 1
1 Mbps
2 Mbps
U2X = 0
U2X = 1
1.152 Mbps
U2X = 0
2.304 Mbps
U2X = 1
1.25 Mbps
2.5 Mbps
UBRR = 0, Error = 0.0%
197
2490N–AVR–05/08
TWI – Two-wire
Serial Interface
Features
•
•
•
•
•
•
•
•
•
•
Two-wire Serial
Interface Bus
Definition
The Two-wire Serial Interface (TWI) is ideally suited for typical microcontroller applications. The
TWI protocol allows the systems designer to interconnect up to 128 different 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.
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
Figure 86. TWI Bus Interconnection
VCC
Device 1
Device 2
Device 3
........
Device n
R1
R2
SDA
SCL
TWI Terminology
The following definitions are frequently encountered in this section.
Table 86. TWI Terminology
198
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.
Transmitter
The device placing data on the bus.
Receiver
The device reading data from the bus.
ATmega64(L)
2490N–AVR–05/08
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 operation 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 Characteristics” on page 328. 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 condition. 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.
199
2490N–AVR–05/08
Figure 88. START, REPEATED START, and STOP Conditions
SDA
SCL
START
Address Packet
Format
REPEATED START
STOP START
STOP
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 performed. 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 consisting 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 allocated 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 message 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
7
8
9
SDA
SCL
1
2
START
200
ATmega64(L)
2490N–AVR–05/08
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 signalled. 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
8
9
Aggregate
SDA
SDA from
Transmitter
SDA from
Receiver
SCL from
Master
1
2
7
SLA+R/W
Combining Address
and Data Packets Into
a Transmission
STOP, REPEATED
START, or Next
Data Byte
Data Byte
A transmission basically consists of a START condition, a SLA+R/W, one or more data 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 transmitted
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
7
8
9
1
Data LSB
ACK
8
9
SDA
SCL
1
START
2
SLA+R/W
2
7
Data Byte
STOP
201
2490N–AVR–05/08
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-master 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
Masters Start
Counting Low Period
TBhigh
Masters Start
Counting High Period
Arbitration is carried out by all masters continuously monitoring the SDA line after outputting
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.
202
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 93. Arbitration between Two Masters
START
SDA from
Master A
Master A Loses
Arbitration, SDAA SDA
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.
203
2490N–AVR–05/08
Overview of the
TWI Module
The TWI module is comprised of several submodules, as shown in Figure 94. All registers drawn
in a thick line are accessible through the AVR data bus.
Figure 94. Overview of the TWI Module
Slew-rate
Control
SDA
Spike
Filter
Slew-rate
Control
Spike
Filter
Bus Interface Unit
START / STOP
Control
Spike Suppression
Arbitration Detection
Address/Data Shift
Register (TWDR)
Bit Rate Generator
Prescaler
Address Match Unit
Address Register
(TWAR)
Bit Rate Register
(TWBR)
Ack
Control Unit
Status Register
(TWSR)
Control Register
(TWCR)
TWI Unit
SCL
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 internal 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 Prescaler 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.
•
TWPS = Value of the prescaler bits in the TWI Status Register.
Note:
204
Pull-up resistor values should be selected according to the SCL frequency and the capacitive bus
line load. See Table 133 on page 328 for value of pull-up resistor."
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Bus Interface Unit
This unit contains the Data and Address Shift Register (TWDR), a START/STOP Controller 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 application 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 settings 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 identifying 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.
•
When a bus error has occurred due to an illegal START or STOP condition.
205
2490N–AVR–05/08
TWI Register
Description
TWBR –TWI Bit Rate
Register
Bit
7
6
5
4
3
2
1
0
TWBR7
TWBR6
TWBR5
TWBR4
TWBR3
TWBR2
TWBR1
TWBR0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
(0x70)
TWBR
• 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 204 for calculating bit rates.
TWCR – TWI Control
Register
Bit
7
6
5
4
3
2
1
0
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Read/Write
R/W
R/W
R/W
R/W
R
R/W
R
R/W
Initial Value
0
0
0
0
0
0
0
0
(0x74)
TWCR
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 application
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 routine. 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 Register (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 written 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 Twowire 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
206
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
is detected, and then generates a new START condition to claim 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.
TWSR – TWI Status
Register
Bit
7
6
5
4
3
2
1
0
TWS7
TWS6
TWS5
TWS4
TWS3
–
TWPS1
TWPS0
Read/Write
R
R
R
R
R
R
R/W
R/W
Initial Value
1
1
1
1
1
0
0
0
(0x71)
TWSR
• Bits 7..3 – TWS: TWI Status
These five bits reflect the status of the TWI logic and the Two-wire Serial Bus. The different 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.
207
2490N–AVR–05/08
• 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
0
1
4
1
0
16
1
1
64
To calculate bit rates, see “Bit Rate Generator Unit” on page 204. The value of TWPS1..0 is
used in the equation.
TWDR – TWI Data
Register
Bit
7
6
5
4
3
2
1
0
TWD7
TWD6
TWD5
TWD4
TWD3
TWD2
TWD1
TWD0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
1
1
1
1
1
1
1
1
(0x73)
TWDR
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.
TWAR – TWI (Slave)
Address Register
Bit
7
6
5
4
3
2
1
0
TWA6
TWA5
TWA4
TWA3
TWA2
TWA1
TWA0
TWGCE
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
1
1
1
1
1
1
1
0
(0x72)
TWAR
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.
208
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
• 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 application
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 description is quite
abstract, a more detailed explanation follows later in this section. A simple code example implementing the desired behavior is also presented.
Application
Action
Figure 95. Interfacing the Application to the TWI in a Typical Transmission
1. Application writes
to TWCR to initiate
transmission of
START
TWI
Hardware
Action
TWI bus
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
TWSTA is written to zero.
START
SLA+W
2. TWINT set.
Status code indicates
START condition sent
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
A
4. TWINT set.
Status code indicates
SLA+W sent, ACK
received
Data
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
A
6. TWINT set.
Status code indicates
data sent, ACK received
STOP
Indicates
TWINT set
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.
3. The application software should now examine the value of TWSR, to make sure that the
START condition was successfully transmitted. If TWSR indicates otherwise, the application software might take some special action, like calling an error routine. Assuming that
209
2490N–AVR–05/08
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 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 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 successfully 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 application 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 transmissions.
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 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)
2490N–AVR–05/08
ATmega64(L)
Assembly code example(1)
1
ldi
r16, (1<<TWINT)|(1<<TWSTA)|
(1<<TWEN)
out
2
TWCR = (1<<TWINT)|(1<<TWSTA)|
Comments
Send START condition
(1<<TWEN)
TWCR, r16
wait1:
in
C example(1)
r16,TWCR
while (!(TWCR & (1<<TWINT)))
;
sbrs r16,TWINT
Wait for TWINT flag set. This
indicates that the START
condition has been transmitted
rjmp wait1
3
in
r16,TWSR
andi r16, 0xF8
cpi
if ((TWSR & 0xF8) != START)
ERROR();
r16, START
brne ERROR
4
ldi
r16, SLA_W
TWDR = SLA_W;
out
TWDR, r16
TWCR = (1<<TWINT) | (1<<TWEN);
ldi
r16, (1<<TWINT) | (1<<TWEN)
out
TWCR, r16
wait2:
in
r16,TWCR
while (!(TWCR & (1<<TWINT)))
;
sbrs r16,TWINT
rjmp wait2
5
in
r16,TWSR
andi r16, 0xF8
cpi
if ((TWSR & 0xF8) != MT_SLA_ACK)
ERROR();
r16, MT_SLA_ACK
brne ERROR
6
ldi
r16, DATA
TWDR = DATA;
out
TWDR, r16
TWCR = (1<<TWINT) | (1<<TWEN);
ldi
r16, (1<<TWINT) | (1<<TWEN)
out
TWCR, r16
wait3:
in
r16,TWCR
while (!(TWCR & (1<<TWINT)))
;
sbrs r16,TWINT
rjmp wait3
7
in
r16,TWSR
andi r16, 0xF8
cpi
if ((TWSR & 0xF8) != MT_DATA_ACK)
ERROR();
r16, MT_DATA_ACK
brne ERROR
ldi
r16, (1<<TWINT)|(1<<TWEN)|
(1<<TWSTO)
out
Note:
TWCR = (1<<TWINT)|(1<<TWEN)|
Check value of TWI Status
Register. Mask prescaler bits. If
status different from START go to
ERROR
Load SLA_W into TWDR
Register. Clear TWINT bit in
TWCR to start transmission of
address
Wait for TWINT flag set. This
indicates that the SLA+W has
been transmitted, and
ACK/NACK has been received.
Check value of TWI Status
Register. Mask prescaler bits. If
status different from
MT_SLA_ACK go to ERROR
Load DATA into TWDR Register.
Clear TWINT bit in TWCR to
start transmission of data
Wait for TWINT flag set. This
indicates that the DATA has been
transmitted, and ACK/NACK has
been received.
Check value of TWI Status
Register. Mask prescaler bits. If
status different from
MT_DATA_ACK go to ERROR
Transmit STOP condition
(1<<TWSTO);
TWCR, r16
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
2490N–AVR–05/08
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 figures contain the
following abbreviations:
S:
START condition
Rs:
REPEATED START condition
R:
Read bit (high level at SDA)
W:
Write bit (low level at SDA)
A:
Acknowledge bit (low level at SDA)
A:
Not acknowledge bit (high level at SDA)
Data:
8-bit data byte
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 appropriate 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)
2490N–AVR–05/08
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
Device 2
MASTER
TRANSMITTER
SLAVE
RECEIVER
Device 3
........
Device n
R1
R2
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 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 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 status 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
2490N–AVR–05/08
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 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
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)
Prescaler Bits
are 0
Application Software Response
Status of the Two-wire Serial
Bus and Two-wire Serial Interface Hardware
To TWCR
To/from TWDR
STA
STO
TWINT
TWEA
Next Action Taken by TWI Hardware
0x08
A START condition has been
transmitted
Load SLA+W
0
0
1
X
SLA+W will be transmitted;
ACK or NOT ACK will be received
0x10
A repeated START condition
has been transmitted
Load SLA+W or
0
0
1
X
Load SLA+R
0
0
1
X
SLA+W will be transmitted;
ACK or NOT ACK will be received
SLA+R will be transmitted;
Logic will switch to Master Receiver mode
Load data byte or
0
0
1
X
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
No TWDR action
1
1
1
X
Load data byte or
0
0
1
X
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
No TWDR action
1
1
1
X
Load data byte or
0
0
1
X
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
No TWDR action
1
1
1
X
Load data byte or
0
0
1
X
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
No TWDR action
1
1
1
X
No TWDR action or
0
0
1
X
No TWDR action
1
0
1
X
0x18
0x20
0x28
0x30
0x38
214
SLA+W has been transmitted;
ACK has been received
SLA+W has been transmitted;
NOT ACK has been received
Data byte has been transmitted;
ACK has been received
Data byte has been transmitted;
NOT ACK has been received
Arbitration lost in SLA+W or
data bytes
Data byte will be transmitted and ACK or NOT ACK will
be received
Repeated START will be transmitted
STOP condition will be transmitted and
TWSTO flag will be reset
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
Data byte will be transmitted and ACK or NOT ACK will
be received
Repeated START will be transmitted
STOP condition will be transmitted and
TWSTO flag will be reset
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
Data byte will be transmitted and ACK or NOT ACK will
be received
Repeated START will be transmitted
STOP condition will be transmitted and
TWSTO flag will be reset
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
Data byte will be transmitted and ACK or NOT ACK will
be received
Repeated START will be transmitted
STOP condition will be transmitted and
TWSTO flag will be reset
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
Two-wire Serial Bus will be released and not addressed slave mode entered
A START condition will be transmitted when the bus
becomes free
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 97. Formats and States in the Master Transmitter Mode
MT
Successfull
transmission
to a slave
receiver
S
SLA
$08
W
A
DATA
$18
A
P
$28
Next transfer
started with a
repeated start
condition
RS
SLA
W
$10
Not acknowledge
received after the
slave address
A
R
P
$20
MR
Not acknowledge
received after a data
byte
A
P
$30
Arbitration lost in slave
address or data byte
A or A
Other master
continues
$38
Arbitration lost and
addressed as slave
A
$68
From master to slave
From slave to master
A or A
Other master
continues
$38
Other master
continues
$78
DATA
To corresponding
states in slave mode
$B0
A
n
Any number of data bytes
and their associated acknowledge bits
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
215
2490N–AVR–05/08
Master Receiver Mode
In the Master Receiver mode, a number of data bytes are received from a slave transmitter (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
Device 2
MASTER
RECEIVER
SLAVE
TRANSMITTER
Device 3
........
Device n
R1
R2
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 status 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:
216
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
Value
1
X
1
0
X
1
0
X
ATmega64(L)
2490N–AVR–05/08
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)
Prescaler Bits
are 0
Application Software Response
Status of the Two-wire Serial
Bus and Two-wire Serial Interface Hardware
To TWCR
To/from TWDR
STA
STO
TWINT
TWEA
Next Action Taken by TWI Hardware
0x08
A START condition has been
transmitted
Load SLA+R
0
0
1
X
SLA+R will be transmitted
ACK or NOT ACK will be received
0x10
A repeated START condition
has been transmitted
Load SLA+R or
0
0
1
X
Load SLA+W
0
0
1
X
SLA+R will be transmitted
ACK or NOT ACK will be received
SLA+W will be transmitted
Logic will switch to Master Transmitter mode
No TWDR action or
0
0
1
X
No TWDR action
1
0
1
X
0
0
1
0
0x38
Arbitration lost in SLA+R or
NOT ACK bit
0x40
SLA+R has been transmitted;
ACK has been received
No TWDR action or
No TWDR action
0
0
1
1
0x48
SLA+R has been transmitted;
NOT ACK has been received
No TWDR action or
No TWDR action or
1
0
0
1
1
1
X
X
No TWDR action
1
1
1
X
0
0x50
Data byte has been received;
ACK has been returned
Read data byte or
0
0
1
Read data byte
0
0
1
1
0x58
Data byte has been received;
NOT ACK has been returned
Read data byte or
Read data byte or
1
0
0
1
1
1
X
X
Read data byte
1
1
1
X
Two-wire Serial Bus will be released and not addressed Slave mode will be entered
A START condition will be transmitted when the bus
becomes free
Data byte will be received and NOT ACK will be
returned
Data byte will be received and ACK will be returned
Repeated START will be transmitted
STOP condition will be transmitted and TWSTO flag
will be reset
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
Data byte will be received and NOT ACK will be
returned
Data byte will be received and ACK will be returned
Repeated START will be transmitted
STOP condition will be transmitted and TWSTO flag
will be reset
STOP condition followed by a START condition will be
transmitted and TWSTO flag will be reset
217
2490N–AVR–05/08
Figure 99. Formats and States in the Master Receiver Mode
MR
Successfull
reception
from a slave
receiver
S
SLA
$08
R
A
DATA
$40
A
DATA
$50
A
P
$58
Next transfer
started with a
repeated start
condition
RS
SLA
R
$10
Not acknowledge
received after the
slave address
A
W
P
$48
MT
Arbitration lost in slave
address or data byte
A or A
Other master
continues
$38
Arbitration lost and
addressed as slave
A
$68
From master to slave
From slave to master
218
A
Other master
continues
$38
Other master
continues
$78
DATA
To corresponding
states in slave mode
$B0
A
n
Any number of data bytes
and their associated acknowledge bits
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
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Slave Receiver Mode
In the Slave Receiver mode, a number of data bytes are received from a master transmitter (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
Device 2
SLAVE
RECEIVER
MASTER
TRANSMITTER
Device 3
........
Device n
R1
R2
SDA
SCL
To initiate the Slave Receiver mode, TWAR and TWCR must be initialized as follows:
TWAR
TWA6
TWA5
Value
TWA4
TWA3
TWA2
TWA1
TWA0
TWGCE
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 carried out as normal, with the AVR clocks
219
2490N–AVR–05/08
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)
Prescaler Bits
Are 0
Application Software Response
Status of the Two-wire Serial Bus
and Two-wire Serial Interface
Hardware
To TWCR
STA
STO
TWINT
TWEA
No TWDR action or
X
0
1
0
To/from TWDR
0x60
Own SLA+W has been received;
ACK has been returned
No TWDR action
X
0
1
1
0x68
Arbitration lost in SLA+R/W as
master; own SLA+W has been
received; ACK has been returned
No TWDR action or
X
0
1
0
No TWDR action
X
0
1
1
0x70
General call address has been
received; ACK has been returned
No TWDR action or
X
0
1
0
No TWDR action
X
0
1
1
0x78
Arbitration lost in SLA+R/W as
master; General call address has
been received; ACK has been
returned
No TWDR action or
X
0
1
0
No TWDR action
X
0
1
1
0x80
Previously addressed with own
SLA+W; data has been received;
ACK has been returned
Read data byte or
X
0
1
0
Read data byte
X
0
1
1
0x88
Previously addressed with own
SLA+W; data has been received;
NOT ACK has been returned
Read data byte or
0
0
1
0
Read data byte or
0
0
1
1
Read data byte or
1
0
1
0
Read data byte
1
0
1
1
0
1
0
0x90
Previously addressed with
general call; data has been received; ACK has been returned
Read data byte or
X
Read data byte
X
0
1
1
0x98
Previously addressed with
general call; data has been
received; NOT ACK has been
returned
Read data byte or
0
0
1
0
Read data byte or
0
0
1
1
Read data byte or
1
0
1
0
Read data byte
1
0
1
1
No Action
0
0
1
0
0
0
1
1
1
0
1
0
1
0
1
1
0xA0
220
A STOP condition or repeated
START condition has been
received while still addressed as
slave
Next Action Taken by TWI Hardware
Data byte will be received and NOT ACK will be
returned
Data byte will be received and ACK will be returned
Data byte will be received and NOT ACK will be
returned
Data byte will be received and ACK will be returned
Data byte will be received and NOT ACK will be
returned
Data byte will be received and ACK will be returned
Data byte will be received and NOT ACK will be
returned
Data byte will be received and ACK will be returned
Data byte will be received and NOT ACK will be
returned
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;
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
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
Data byte will be received and NOT ACK will be
returned
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;
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
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
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
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
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 101. Formats and States in the Slave Receiver Mode
Reception of the own
slave address and one or
more data bytes. All are
acknowledged
S
SLA
W
A
DATA
$60
A
DATA
$80
Last data byte received
is not acknowledged
A
P or S
$80
$A0
A
P or S
$88
Arbitration lost as master
and addressed as slave
A
$68
Reception of the general call
address and one or more data
bytes
General Call
A
DATA
$70
A
DATA
$90
Last data byte received is
not acknowledged
A
P or S
$90
$A0
A
P or S
$98
Arbitration lost as master and
addressed as slave by general call
A
$78
From master to slave
From slave to master
DATA
A
n
Any number of data bytes
and their associated acknowledge bits
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
221
2490N–AVR–05/08
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
Device 2
SLAVE
TRANSMITTER
MASTER
RECEIVER
Device 3
........
Device n
R1
R2
SDA
SCL
To initiate the Slave Transmitter mode, TWAR and TWCR must be initialized as follows:
TWAR
TWA6
TWA5
Value
TWA4
TWA3
TWA2
TWA1
TWA0
TWGCE
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 transmitted 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)
2490N–AVR–05/08
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)
Prescaler
Bits are 0
0xA8
0xB0
0xB8
0xC0
0xC8
Application Software Response
Status of the Two-wire Serial Bus
and Two-wire Serial Interface
Hardware
To TWCR
To/from TWDR
STA
STO
TWINT
TWEA
Load data byte or
X
0
1
0
Load data byte
X
0
1
1
Arbitration lost in SLA+R/W as
master; own SLA+R has been
received; ACK has been returned
Load data byte or
X
0
1
0
Load data byte
X
0
1
1
Data byte in TWDR has been
transmitted; ACK has been
received
Load data byte or
X
0
1
0
Load data byte
X
0
1
1
Data byte in TWDR has been
transmitted; NOT ACK has been
received
No TWDR action or
0
0
1
0
No TWDR action or
0
0
1
1
No TWDR action or
1
0
1
0
No TWDR action
1
0
1
1
No TWDR action or
0
0
1
0
No TWDR action or
0
0
1
1
No TWDR action or
1
0
1
0
No TWDR action
1
0
1
1
Own SLA+R has been received;
ACK has been returned
Last data byte in TWDR has been
transmitted (TWEA = “0”); ACK
has been received
Next Action Taken by TWI Hardware
Last data byte will be transmitted and NOT ACK should
be received
Data byte will be transmitted and ACK should be received
Last data byte will be transmitted and NOT ACK should
be received
Data byte will be transmitted and ACK should be received
Last data byte will be transmitted and NOT ACK should
be received
Data byte will be transmitted and ACK should be received
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
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
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
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
2490N–AVR–05/08
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
$A8
Arbitration lost as master
and addressed as slave
A
DATA
$B8
A
P or S
$C0
A
$B0
Last data byte transmitted.
Switched to not addressed
slave (TWEA = '0')
A
All 1's
P or S
$C8
DATA
From master to slave
From slave to master
Miscellaneous States
Any number of data bytes
and their associated acknowledge bits
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
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 transfer. 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
(TWSR)
Prescaler Bits
are 0
Application Software Response
Status of the Two-wire Serial
Bus and Two-wire Serial Interface hardware
To TWCR
To/from TWDR
0xF8
No relevant state information
available; TWINT = “0”
No TWDR action
0x00
Bus error due to an illegal
START or STOP condition
No TWDR action
224
STA
STO
TWINT
TWEA
No TWCR action
0
1
1
Next Action Taken by TWI Hardware
Wait or proceed current transfer
X
Only the internal hardware is affected, no STOP condition is sent on the bus. In all cases, the bus is released
and TWSTO is cleared.
ATmega64(L)
2490N–AVR–05/08
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. Subsequently, 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 principle 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
S
SLA+W
A
ADDRESS
S = START
Master Receiver
A
Rs
A
DATA
Rs = REPEATED START
Transmitted from master to slave
Multi-master
Systems and
Arbitration
SLA+R
A
P
P = STOP
Transmitted from slave to master
If multiple masters are connected to the same bus, transmissions may be initiated simultaneously 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 transfer, 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
Device 2
Device 3
MASTER
TRANSMITTER
MASTER
TRANSMITTER
SLAVE
RECEIVER
........
Device n
R1
R2
SDA
SCL
Several different scenarios may arise during arbitration, as described below:
225
2490N–AVR–05/08
•
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
Arbitration lost in SLA
Own
Address / General Call
received
No
STOP
Arbitration lost in Data
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
Yes
Direction
Write
68/78
Read
B0
226
Data byte will be received and NOT ACK will be returned
Data byte will be received and ACK will be returned
Last data byte will be transmitted and NOT ACK should be received
Data byte will be transmitted and ACK should be received
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Analog
Comparator
The Analog Comparator compares the input values on the positive pin AIN0 and negative 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 diagram 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:
SFIOR – Special
Function IO Register
1. See Table 94 on page 229.
2. Refer to Figure 1 on page 2 and Table 30 on page 74 for Analog Comparator pin placement.
Bit
7
6
5
4
3
2
1
0
0x20
(0x40)
TSM
–
–
–
ACME
PUD
PSR2
PSR10
Read/Write
R/W
R
R
R
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
SFIOR
• 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 Comparator. For a detailed
description of this bit, see “Analog Comparator Multiplexed Input” on page 229.
227
2490N–AVR–05/08
ACSR – Analog
Comparator Control
and Status Register
Bit
7
6
5
4
3
2
1
0
0x08 (0x28)
ACD
ACBG
ACO
ACI
ACIE
ACIC
ACIS1
ACIS0
Read/Write
R/W
R/W
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
N/A
0
0
0
0
0
ACSR
• 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 Comparator 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 56.
• 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 executing 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 Analog 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 interrupt, the TICIE1 bit in the Timer Interrupt Mask
Register (TIMSK) must be set.
228
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
• Bits 1, 0 – ACIS1, ACIS0: Analog Comparator Interrupt Mode Select
These bits determine which comparator events that trigger the Analog Comparator interrupt. The
different settings are shown in Table 93.
Table 93. ACIS1/ACIS0 Settings
ACIS1
ACIS0
Interrupt Mode
0
0
Comparator Interrupt on Output Toggle.
0
1
Reserved
1
0
Comparator Interrupt on Falling Output Edge.
1
1
Comparator Interrupt on Rising Output Edge.
When changing the ACIS1/ACIS0 bits, the Analog Comparator Interrupt must be disabled by
clearing its Interrupt Enable bit in the ACSR Register. Otherwise an interrupt can occur when the
bits are changed.
Analog
Comparator
Multiplexed Input
It is possible to select any of the ADC7..0 pins to replace the negative input to the Analog 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 Analog 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
Analog Comparator Negative Input
0
x
xxx
AIN1
1
1
xxx
AIN1
1
0
000
ADC0
1
0
001
ADC1
1
0
010
ADC2
1
0
011
ADC3
1
0
100
ADC4
1
0
101
ADC5
1
0
110
ADC6
1
0
111
ADC7
229
2490N–AVR–05/08
Analog to
Digital
Converter
Features
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
10-bit Resolution
0.75 LSB Integral Non-linearity
±1.5 LSB Absolute Accuracy
13 - 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
2.7 - VCC Differential ADC 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 connected 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 differential 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 differential 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 237 on how to connect this
pin.
Internal reference voltages of nominally 2.56V or AVCC are provided On-chip. The voltage reference may be externally decoupled at the AREF pin by a capacitor for better noise performance.
230
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 108. Analog to Digital Converter Block Schematic
ADC CONVERSION
COMPLETE IRQ
INTERRUPT
FLAGS
ADTS[2:0]
TRIGGER
SELECT
ADC[9:0]
ADPS1
0
ADC DATA REGISTER
(ADCH/ADCL)
ADPS0
ADPS2
ADIF
ADATE
ADEN
ADSC
MUX1
15
ADC CTRL. & STATUS
REGISTER (ADCSRA)
MUX0
MUX3
MUX2
MUX4
ADLAR
REFS1
REFS0
ADC MULTIPLEXER
SELECT (ADMUX)
ADIE
ADIF
8-BIT DATA BUS
AVCC
PRESCALER
START
GAIN SELECTION
CHANNEL SELECTION
MUX DECODER
CONVERSION LOGIC
INTERNAL 2.56V
REFERENCE
SAMPLE & HOLD
COMPARATOR
AREF
10-BIT DAC
+
GND
BANDGAP
REFERENCE
ADC7
SINGLE ENDED / DIFFERENTIAL SELECTION
ADC6
ADC5
ADC MULTIPLEXER
OUTPUT
POS.
INPUT
MUX
ADC4
ADC3
+
ADC2
GAIN
AMPLIFIER
ADC1
ADC0
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 reference 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 reference, 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 difference
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
2490N–AVR–05/08
The ADC is enabled by setting the ADC Enable bit, ADEN in ADCSRA. Voltage reference 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 conversion 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 Triggering 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
trigger a new conversion at the next interrupt event.
Figure 109. ADC Auto Trigger Logic
ADTS[2:0]
PRESCALER
START
ADIF
CLKADC
ADATE
SOURCE 1
.
.
.
.
SOURCE n
CONVERSION
LOGIC
EDGE
DETECTOR
ADSC
232
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 Running 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 conversion was started.
Prescaling and
Figure 110. ADC Prescaler
Conversion Timing
ADEN
START
Reset
7-BIT ADC PRESCALER
CK/64
CK/128
CK/32
CK/8
CK/16
CK/4
CK/2
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
2490N–AVR–05/08
The ADC module contains a prescaler, which generates an acceptable ADC clock frequency
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 conversion
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 conversion 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 conversion 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
Conversion
First 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
ADCH
MSB of Result
ADCL
LSB of Result
MUX and REFS
Update
234
Sample & Hold
Conversion
Complete
MUX and REFS
Update
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 112. ADC Timing Diagram, Single Conversion
One Conversion
Cycle Number
1
2
3
4
5
6
7
8
Next Conversion
10
9
11
12
13
1
2
3
ADC Clock
ADSC
ADIF
ADCH
MSB of Result
ADCL
LSB of Result
Sample & Hold
Conversion
Complete
MUX and REFS
Update
MUX and REFS
Update
Figure 113. ADC Timing Diagram, Auto Triggered Conversion
One Conversion
Cycle Number
1
2
3
4
5
6
7
8
Next Conversion
9
10
11
12
13
1
2
ADC Clock
Trigger
Source
ADATE
ADIF
ADCH
MSB of Result
ADCL
LSB of Result
Prescaler
Reset
Sample &
Hold
Conversion
Complete
Prescaler
Reset
MUX and REFS
Update
Figure 114. ADC Timing Diagram, Free Running Conversion
One Conversion
Cycle Number
11
12
Next Conversion
13
1
2
3
4
ADC Clock
ADSC
ADIF
ADCH
MSB of Result
ADCL
LSB of Result
Conversion
Complete
Sample & Hold
MUX and REFS
Update
235
2490N–AVR–05/08
Table 95. ADC Conversion Time
Sample & Hold (Cycles from
Start of Conversion)
Conversion Time
(Cycles)
First conversion
13.5
25
Normal conversions, single ended
1.5
13
2
13.5
1.5/2.5
13/14
Condition
Auto Triggered conversions
Normal conversions, differential
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 initiated 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 mechanism. 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 frequencies 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 bandwidth. 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 dependent 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 The MUXn and REFS1:0 bits in the ADMUX Register are single buffered through a temporary
register to which the CPU has random access. This ensures that the channels and reference
or Reference
selection only takes place at a safe point during the conversion. The channel and reference
Selection
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.
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:
236
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 overridden.
Refer to Table 42, “Port F Pins Alternate Functions,” on page 83.
ADC Input Channels
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 automatically, 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 circuitry. The user
should preferably disregard the first conversion result.
ADC Voltage
Reference
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 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 discard this result.
If differential channels are used, the selected reference should not be closer to AVCC than
indicated in Table 136 on page 333.
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
237
2490N–AVR–05/08
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 Conversion
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 conversion
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 conversion 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 conversions, 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 analog
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 channel 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 recommended 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.
Figure 115. Analog Input Circuitry
IIH
ADCn
1..100 kΩ
CS/H= 14 pF
IIL
VCC/2
238
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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
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
VCC
10 μΗ
52
GND
53
(ADC7) PF7
54
(ADC6) PF6
55
(ADC5) PF5
56
(ADC4) PF4
57
(ADC3) PF3
58
(ADC2) PF2
59
(ADC1) PF1
60
(ADC0) PF0
61
AREF
62
GND
AVCC
63
64
1
PEN
100 nF
51
Offset Compensation
Schemes
The gain stage has a built-in offset cancellation circuitry that nulls the offset of differential 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 V REF in 2 n 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.
239
2490N–AVR–05/08
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
Output Code
Gain
Error
Ideal ADC
Actual ADC
VREF Input Voltage
•
240
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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 119. Integral Non-linearity (INL)
Output Code
INL
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
2490N–AVR–05/08
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 IN ⋅ 1024
ADC = -------------------------V REF
where VIN is the voltage on the selected input pin and VREF the selected voltage reference (see
Table 97 on page 243 and Table 98 on page 244). 0x000 represents ground, and 0x3FF represents the selected reference voltage minus one LSB.
If differential channels are used, the result is
( V POS – V NEG ) ⋅ GAIN ⋅ 512
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
- V REF/GAIN
0x3FF
0
VREF/GAIN
Differential Input
Voltage (Volts)
0x200
242
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 96. Correlation Between Input Voltage and Output Codes
VADCn
Read Code
Corresponding Decimal Value
VADCm + VREF/GAIN
0x1FF
511
VADCm + 511/512 VREF/GAIN
0x1FF
511
VADCm + 510/512 VREF/GAIN
0x1FE
510
...
...
...
VADCm + 1/512 VREF/GAIN
0x001
1
VADCm
0x000
0
VADCm - 1/512 VREF/GAIN
0x3FF
-1
...
...
...
VADCm - 511/512VREF/GAIN
0x201
-511
VADCm - VREF/GAIN
0x200
-512
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.
ADMUX – ADC
Multiplexer Selection
Register
Bit
7
6
5
4
3
2
1
0
0x07 (0x27)
REFS1
REFS0
ADLAR
MUX4
MUX3
MUX2
MUX1
MUX0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
ADMUX
• 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
AREF, Internal Vref turned off.
0
1
AVCC with external capacitor at AREF pin.
1
0
Reserved
1
1
Internal 2.56V Voltage Reference with external capacitor at AREF pin.
243
2490N–AVR–05/08
•
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 “ADCL and ADCH – The ADC Data Register” 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
MUX4..0
Single Ended Input
00000
ADC0
00001
ADC1
00010
ADC2
00011
ADC3
00100
ADC4
00101
ADC5
00110
ADC6
00111
ADC7
Negative Differential
Input
Gain
N/A
01000
ADC0
ADC0
10x
01001
ADC1
ADC0
10x
01010
ADC0
ADC0
200x
01011
ADC1
ADC0
200x
01100
ADC2
ADC2
10x
01101
ADC3
ADC2
10x
01110
ADC2
ADC2
200x
01111
ADC3
ADC2
200x
10000
ADC0
ADC1
1x
10001
ADC1
ADC1
1x
ADC2
ADC1
1x
10011
ADC3
ADC1
1x
10100
ADC4
ADC1
1x
10101
ADC5
ADC1
1x
10110
ADC6
ADC1
1x
10111
ADC7
ADC1
1x
11000
ADC0
ADC2
1x
11001
ADC1
ADC2
1x
10010
244
Positive Differential
Input
N/A
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 98. Input Channel and Gain Selections (Continued)
Positive Differential
Input
Negative Differential
Input
Gain
11010
ADC2
ADC2
1x
11011
ADC3
ADC2
1x
11100
ADC4
ADC2
1x
11101
ADC5
ADC2
1x
MUX4..0
ADCSRA – ADC
Control and Status
Register A
Single Ended Input
11110
1.22V (VBG)
11111
0V (GND)
Bit
N/A
7
6
5
4
3
2
1
0
0x06 (0x26)
ADEN
ADSC
ADATE
ADIF
ADIE
ADPS2
ADPS1
ADPS0
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
ADCSRA
• Bit 7 – ADEN: ADC Enable
Writing this bit to one enables the ADC. By writing it to zero, the ADC is turned off. Turning the
ADC off while a conversion is in progress, will terminate this conversion.
• Bit 6 – ADSC: ADC Start Conversion
In Single Conversion mode, write this bit to one to start each conversion. In Free Running 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 disabled. 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 Complete Interrupt is activated.
245
2490N–AVR–05/08
• 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
2
0
0
1
2
0
1
0
4
0
1
1
8
1
0
0
16
1
0
1
32
1
1
0
64
1
1
1
128
ADCL and ADCH –
The ADC Data
Register
ADLAR = 0
Bit
15
14
13
12
11
10
9
8
0x05 (0x25)
–
–
–
–
–
–
ADC9
ADC8
ADCH
0x04 (0x24)
ADC7
ADC6
ADC5
ADC4
ADC3
ADC2
ADC1
ADC0
ADCL
7
6
5
4
3
2
1
0
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Read/Write
Initial Value
ADLAR = 1
Bit
15
14
13
12
11
10
9
8
0x05 (0x25)
ADC9
ADC8
ADC7
ADC6
ADC5
ADC4
ADC3
ADC2
ADCH
0x04 (0x24)
ADC1
ADC0
–
–
–
–
–
–
ADCL
7
6
5
4
3
2
1
0
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Read/Write
Initial Value
When an ADC conversion is complete, the result is found in these two registers. If differential
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. Consequently, 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)
2490N–AVR–05/08
ATmega64(L)
• ADC9:0: ADC Conversion Result
These bits represent the result from the conversion, as detailed in “ADC Conversion Result” on
page 242.
ADCSRB – ADC
Control and Status
Register B
Bit
7
6
5
4
3
2
1
0
(0x8E)
–
–
–
–
–
ADTS2
ADTS1
ADTS0
Read/Write
R
R
R
R
R
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
ADCSRB
• 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 trigger event, even if the ADC Interrupt Flag is set.
Figure 122. ADC Auto Trigger Source Selections
ADTS2
ADTS1
ADTS0
Trigger Source
0
0
0
Free Running mode
0
0
1
Analog Comparator
0
1
0
External Interrupt Request 0
0
1
1
Timer/Counter0 Compare Match
1
0
0
Timer/Counter0 Overflow
1
0
1
Timer/Counter1 Compare Match B
1
1
0
Timer/Counter1 Overflow
1
1
1
Timer/Counter1 Capture Event
247
2490N–AVR–05/08
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 Programming via
the JTAG interface, and using the Boundary-scan chain can be found in the sections “Programming Via the JTAG Interface” on page 311 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 system. 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 registers 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.
TAP – Test Access
Port
248
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.
ATmega64(L)
2490N–AVR–05/08
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 connected 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 collectors 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
M
U
X
BREAKPOINT
UNIT
BYPASS
REGISTER
INTERNAL
SCAN
CHAIN
PC
Instruction
FLOW CONTROL
UNIT
DIGITAL
PERIPHERAL
UNITS
BREAKPOINT
SCAN CHAIN
ADDRESS
DECODER
JTAG / AVR CORE
COMMUNICATION
INTERFACE
OCD STATUS
AND CONTROL
ANALOG
PERIPHERIAL
UNITS
Analog inputs
ID
REGISTER
Address
Data
Control & Clock lines
INSTRUCTION
REGISTER
FLASH
MEMORY
I/O PORT n
249
2490N–AVR–05/08
Figure 124. TAP Controller State Diagram
1
Test-Logic-Reset
0
0
Run-Test/Idle
1
Select-DR Scan
1
Select-IR Scan
0
0
1
1
Capture-DR
Capture-IR
0
0
0
Shift-DR
1
1
Exit1-DR
0
0
Pause-DR
0
Pause-IR
1
1
0
Exit2-DR
Exit2-IR
1
1
Update-DR
TAP Controller
1
Exit1-IR
0
1
0
Shift-IR
1
0
1
Update-IR
0
1
0
The TAP controller is a 16-state finite state machine that controls the operation of the Boundaryscan 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 TestLogic-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 interface is:
•
250
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 pin, the captured IR-state 0x01 is shifted out
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 “Bibliography”
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 system
As shown in Figure 123, the hardware support for On-chip Debugging consists mainly of:
•
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.
•
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”).
251
2490N–AVR–05/08
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 Specific JTAG
Instructions” on page 252.
The JTAGEN Fuse must be programmed to enable the JTAG Test Access Port. In addition, 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 programs assembled with Atmel AVR
Assembler and C programs compiled with third party vendors’ compilers.
AVR Studio runs under Microsoft® Windows® 95/98/2000/XP/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 distributed within
ATMEL and to selected third party vendors only. Instruction opcodes are listed for reference.
PRIVATE0; 0x8
Private JTAG instruction for accessing On-chip Debug system.
PRIVATE1; 0x9
Private JTAG instruction for accessing On-chip Debug system.
PRIVATE2; 0xA
Private JTAG instruction for accessing On-chip Debug system.
PRIVATE3; 0xB
Private JTAG instruction for accessing On-chip Debug system.
252
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
On-chip Debug
Related Register in
I/O Memory
OCDR – On-chip
Debug Register
Bit
7
6
5
4
3
2
1
0
0x22 (0x42)
MSB/IDRD
LSB
Read/Write
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
OCDR
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 311.
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
2490N–AVR–05/08
IEEE 1149.1
(JTAG)
Boundary-scan
Features
•
•
•
•
•
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 interconnections and integrity of components on Printed Circuits
Boards by using the four TAP signals only.
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
The four IEEE 1149.1 defined mandatory JTAG instructions IDCODE, BYPASS, SAMPLE/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/PRELOAD 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
Bypass Register
254
The data registers relevant for Boundary-scan operations are:
•
Bypass Register
•
Device Identification Register
•
Reset Register
•
Boundary-scan Chain
The Bypass Register consists of a single Shift Register stage. When the Bypass Register is
selected as path between TDI and TDO, the register is reset to 0 when leaving the Capture-DR
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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
MSB
Bit
Device ID
31
28
27
12
11
1
0
Version
Part Number
Manufacturer ID
1
4 bits
16 bits
11 bits
1-bit
Version
Version is a 4-bit number identifying the revision of the component. The JTAG version number
follows the revision of the device. Revision A is 0x0, revision B is 0x1 and so on.
Part Number
The part number is a 16-bit code identifying the component. The JTAG Part Number for
ATmega64 is listed in Table 100.
Table 100. AVR JTAG Part Number
Part Number
ATmega64
Manufacturer ID
JTAG Part Number (Hex)
0x9602
The Manufacturer ID is a 11-bit code identifying the manufacturer. The JTAG manufacturer ID
for Atmel is listed in Table 101.
Table 101. Manufacturer ID
Manufacturer
Atmel
JTAG Man. ID (Hex)
0x01F
255
2490N–AVR–05/08
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 38) 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
D
Q
Internal Reset
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 JTAG
Instructions
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 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 circuits 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 IRRegister is loaded with the EXTEST instruction.
The active states are:
256
•
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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
•
IDCODE; 0x1
Update-DR: Data from the scan chain is applied to output pins.
Optional JTAG instruction selecting the 32-bit ID-Register as data register. The ID-Register 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:
SAMPLE_PRELOAD;
0x2
•
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.
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:
AVR_RESET; 0xC
•
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.
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:
•
BYPASS; 0xF
Shift-DR: The Reset Register is shifted by the TCK input.
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
2490N–AVR–05/08
Boundary-scan
Related Register in
I/O Memory
MCUCSR – MCU
Control and Status
Register
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
5
4
3
2
1
0
0x34 (0x54)
JTD
–
–
JTRF
WDRF
BORF
EXTRF
PORF
Read/Write
R/W
R
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
MCUCSR
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 function. 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 66. 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 connection 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)
2490N–AVR–05/08
ATmega64(L)
Figure 127. Boundary-scan Cell for Bi-directional Port Pin with Pull-up Function
Pullup Enable (PUE)
ShiftDR
To Next Cell
EXTEST
Vcc
0
FF2
LD2
1
0
D
Q
D
Q
1
Output Control (OC)
G
FF1
LD1
0
D
Q
D
Q
0
1
1
0
1
FF0
LD0
0
D
Q
D
1
Q
0
1
Port Pin (PXn)
Output Data (OD)
G
Input Data (ID)
G
From Last Cell
ClockDR
UpdateDR
259
2490N–AVR–05/08
Figure 128. General Port Pin Schematic Diagram
See Boundary-scan Description
for Details!
PUExn
PUD
Q
D
DDxn
Q CLR
RESET
OCxn
WDx
Q
Pxn
ODxn
D
PORTxn
Q CLR
WPx
IDxn
DATA BUS
RDx
RESET
SLEEP
RRx
SYNCHRONIZER
D
Q
L
Q
D
RPx
Q
PINxn
Q
CLK I/O
PUD:
PUExn:
OCxn:
ODxn:
IDxn:
SLEEP:
Boundary-scan and
the Two-wire Interface
WDx:
RDx:
WPx:
RRx:
RPx:
CLK I/O :
WRITE DDRx
READ DDRx
WRITE PORTx
READ PORTx REGISTER
READ PORTx PIN
I/O CLOCK
The two Two-wire Interface pins SCL and SDA have one additional control signal in the scanchain; 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:
260
PULLUP DISABLE
PULLUP ENABLE for pin Pxn
OUTPUT CONTROL for pin Pxn
OUTPUT DATA to pin Pxn
INPUT DATA from pin Pxn
SLEEP CONTROL
1. A separate scan chain for the 50 ns spike filter on the input is not provided. The ordinary 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 boundaryscan.
2. Make sure the OC and TWIEN signals are not asserted simultaneously, as this will lead to
drive contention.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 129. Additional Scan Signal for the Two-wire Interface
PUExn
OCxn
ODxn
Pxn
TWIEN
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 Figure 130 is
inserted both for the 5V reset signal; RSTT, and the 12V reset signal; RSTHV.
To
Next
Cell
ShiftDR
FF1
0
D
To System Logic
From System Pin
Figure 130. Observe-only Cell
Q
1
From
Previous
Cell
ClockDR
261
2490N–AVR–05/08
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-frequency 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 Oscillator is not scanned, as this
Oscillator does not have external connections.
Figure 131. Boundary-scan Cells for Oscillators and Clock Options
To
Next
Cell
From Digital Logic
ShiftDR
XTAL2/TOSC2
Oscillator
EXTEST
To
Next
Cell
ShiftDR
0
ENABLE
OUTPUT
1
FF1
0
D
Q
D
Q
0
1
D
G
From
Previous
Cell
ClockDR
To System Logic
XTAL1/TOSC1
Q
1
UpdateDR
From
Previous
Cell
ClockDR
Table 102 summaries the scan registers for the external clock pin XTAL1, oscillators with
XTAL1/XTAL2 connections as well as 32 kHz Timer Oscillator.
Table 102. Scan Signals for the Oscillators(1)(2)(3)
Enable
Signal
Scanned Clock
Line
Clock Option
Scanned Clock Line
when Not Used
EXTCLKEN
EXTCLK (XTAL1)
External Clock
0
OSCON
OSCCK
External Crystal
External Ceramic Resonator
0
RCOSCEN
RCCK
External RC
0
OSC32EN
OSC32CK
Low Freq. External Crystal
1
TOSKON
TOSCK
32 kHz Timer Oscillator
0
Notes:
262
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 in sleep modes, thereby dis-
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
connecting 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 signals are
described in Table 103.
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
2490N–AVR–05/08
To
Next
Cell
ShiftDR
EXTEST
0
1
0
D
Q
D
Q
1
To Analog Circuitry/
To Digital Logic
From Digital Logic/
From Analog Ciruitry
Figure 133. General Boundary-scan Cell used for Signals for Comparator and ADC
G
From
Previous
Cell
ClockDR
UpdateDR
Table 103. Boundary-scan Signals for the Analog Comparator
264
Signal
Name
Direction as
Seen from the
Comparator
Recommended
Input when Not
in Use
Output Values when
Recommended
Inputs are Used
AC_IDLE
Input
Turns off Analog
Comparator
when true
1
Depends upon µC
code being executed
ACO
Output
Analog
Comparator
Output
Will become
input to µC code
being executed
0
ACME
Input
Uses output
signal from ADC
mux when true
0
Depends upon µC
code being executed
ACBG
Input
Bandgap
Reference
enable
0
Depends upon µC
code being executed
Description
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Scanning the ADC
Figure 134 shows a block diagram of the ADC with all relevant control and observe signals. 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
1.22V
Ref
EXTCH
MUXEN_3
ADC_3
MUXEN_2
ADC_2
MUXEN_1
ADC_1
MUXEN_0
ADC_0
NEGSEL_2
NEGSEL_1
NEGSEL_0
ADCBGEN
SCTEST
PRECH
AREF
AREF
DACOUT
DAC_9..0
10-bit DAC
G10
G20
ADCEN
+
-
COMP
ACTEN
+
+
20x
10x
-
ADC_2
GNDEN
ADC_1
ADC_0
HOLD
-
ST
ACLK
AMPEN
The signals are described briefly in Table 104.
265
2490N–AVR–05/08
Table 104. Boundary-scan Signals for the ADC(1)
266
Signal
Name
Direction
as Seen
from the
ADC
Recommended
Input when not
in Use
Output Values when
Recommended Inputs
are Used, and CPU is
not Using the ADC
Description
COMP
Output
Comparator Output
0
0
ACLK
Input
Clock signal to gain
stages implemented
as Switch-cap filters
0
0
ACTEN
Input
Enable path from
gain stages to the
Comparator
0
0
ADCBGEN
Input
Enable Band-gap
reference as
negative input to
Comparator
0
0
ADCEN
Input
Power-on signal to
the ADC
0
0
AMPEN
Input
Power-on signal to
the gain stages
0
0
DAC_9
Input
Bit nine of digital
value to DAC
1
1
DAC_8
Input
Bit eight of digital
value to DAC
0
0
DAC_7
Input
Bit seven of digital
value to DAC
0
0
DAC_6
Input
Bit six of digital
value to DAC
0
0
DAC_5
Input
Bit five of digital
value to DAC
0
0
DAC_4
Input
Bit four of digital
value to DAC
0
0
DAC_3
Input
Bit three of digital
value to DAC
0
0
DAC_2
Input
Bit two of digital
value to DAC
0
0
DAC_1
Input
Bit 1 of digital value
to DAC
0
0
DAC_0
Input
Bit 0 of digital value
to DAC
0
0
EXTCH
Input
Connect ADC
channels 0 - 3 to
bypass path around
gain stages
1
1
G10
Input
Enable 10x gain
0
0
G20
Input
Enable 20x gain
0
0
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 104. Boundary-scan Signals for the ADC(1) (Continued)
Signal
Name
Direction
as Seen
from the
ADC
GNDEN
Input
Ground the negative
input to comparator
when true
0
0
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.
1
1
IREFEN
Input
Enables Band-gap
reference as AREF
signal to DAC
0
0
MUXEN_7
Input
Input Mux bit 7
0
0
MUXEN_6
Input
Input Mux bit 6
0
0
MUXEN_5
Input
Input Mux bit 5
0
0
MUXEN_4
Input
Input Mux bit 4
0
0
MUXEN_3
Input
Input Mux bit 3
0
0
MUXEN_2
Input
Input Mux bit 2
0
0
MUXEN_1
Input
Input Mux bit 1
0
0
MUXEN_0
Input
Input Mux bit 0
1
1
NEGSEL_2
Input
Input Mux for
negative input for
differential signal,
bit 2
0
0
NEGSEL_1
Input
Input Mux for
negative input for
differential signal,
bit 1
0
0
NEGSEL_0
Input
Input Mux for
negative input for
differential signal,
bit 0
0
0
PASSEN
Input
Enable pass-gate of
gain stages.
1
1
PRECH
Input
Precharge output
latch of comparator
(Active low)
1
1
Description
Recommended
Input when not
in Use
Output Values when
Recommended Inputs
are Used, and CPU is
not Using the ADC
267
2490N–AVR–05/08
Table 104. Boundary-scan Signals for the ADC(1) (Continued)
Signal
Name
Direction
as Seen
from the
ADC
SCTEST
Input
Switch-cap TEST
enable. Output from
x10 gain stage send
out to Port Pin
having ADC_4
0
0
ST
Input
Output of gain
stages will settle
faster if this signal is
high first two ACLK
periods after
AMPEN goes high.
0
0
VCCREN
Input
Selects Vcc as the
ACC reference
voltage.
0
0
Note:
Description
Recommended
Input when not
in Use
Output Values when
Recommended Inputs
are Used, and CPU is
not Using the ADC
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 104 should
be used. The user is recommended not to use the Differential Gain stages during scan. Switchcap 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 comparator 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).
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:
268
1024 ⋅ 1,5V ⋅ 0,95 ⁄ 5V = 291 = 0x123
1024 ⋅ 1,5V ⋅ 1,05 ⁄ 5V = 323 = 0x143
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
The recommended values from Table 104 are used unless other values are given in the algorithm in Table 105. 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 105. Algorithm for Using the ADC(1)
ADCEN
DAC
MUXEN
HOLD
PRECH
PA3.
Data
PA3.
Control
PA3.
Pullup_
Enable
SAMPLE_PRELOAD
1
0x200
0x08
1
1
0
0
0
EXTEST
1
0x200
0x08
0
1
0
0
0
3
1
0x200
0x08
1
1
0
0
0
4
1
0x123
0x08
1
1
0
0
0
5
1
0x123
0x08
1
0
0
0
0
1
0x200
0x08
1
1
0
0
0
7
1
0x200
0x08
0
1
0
0
0
8
1
0x200
0x08
1
1
0
0
0
9
1
0x143
0x08
1
1
0
0
0
10
1
0x143
0x08
1
0
0
0
0
1
0x200
0x08
1
1
0
0
0
Ste
p
Actions
1
2
6
11
Note:
Verify the COMP bit scanned out to be 0
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 maximum hold time, thold,max.
269
2490N–AVR–05/08
ATmega64
Boundary-scan
Order
Table 106 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 106. ATmega64 Boundary-scan Order
270
Bit Number
Signal Name
Module
204
AC_IDLE
Comparator
203
ACO
202
ACME
201
AINBG
200
COMP
ADC
(1)
199
PRIVATE_SIGNAL1
198
ACLK
197
ACTEN
196
PRIVATE_SIGNAL2(2)
195
ADCBGEN
194
ADCEN
193
AMPEN
192
DAC_9
191
DAC_8
190
DAC_7
189
DAC_6
188
DAC_5
187
DAC_4
186
DAC_3
185
DAC_2
184
DAC_1
183
DAC_0
182
EXTCH
181
G10
180
G20
179
GNDEN
178
HOLD
177
IREFEN
176
MUXEN_7
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 106. ATmega64 Boundary-scan Order (Continued)
Bit Number
Signal Name
Module
175
MUXEN_6
ADC
174
MUXEN_5
173
MUXEN_4
172
MUXEN_3
171
MUXEN_2
170
MUXEN_1
169
MUXEN_0
168
NEGSEL_2
167
NEGSEL_1
166
NEGSEL_0
165
PASSEN
164
PRECH
163
SCTEST
162
ST
161
VCCREN
160
PEN
Programming Enable (Observe-only)
159
PE0.Data
Port E
158
PE0.Control
157
PE0.Pullup_Enable
156
PE1.Data
155
PE1.Control
154
PE1.Pullup_Enable
153
PE2.Data
152
PE2.Control
151
PE2.Pullup_Enable
150
PE3.Data
149
PE3.Control
148
PE3.Pullup_Enable
147
PE4.Data
146
PE4.Control
145
PE4.Pullup_Enable
144
PE5.Data
143
PE5.Control
142
PE5.Pullup_Enable
141
PE6.Data
140
PE6.Control
271
2490N–AVR–05/08
Table 106. ATmega64 Boundary-scan Order (Continued)
272
Bit Number
Signal Name
Module
139
PE6.Pullup_Enable
Port E
138
PE7.Data
137
PE7.Control
136
PE7.Pullup_Enable
135
PB0.Data
134
PB0.Control
133
PB0.Pullup_Enable
132
PB1.Data
131
PB1.Control
130
PB1.Pullup_Enable
129
PB2.Data
128
PB2.Control
127
PB2.Pullup_Enable
126
PB3.Data
125
PB3.Control
124
PB3.Pullup_Enable
123
PB4.Data
122
PB4.Control
121
PB4.Pullup_Enable
120
PB5.Data
119
PB5.Control
118
PB5.Pullup_Enable
117
PB6.Data
116
PB6.Control
115
PB6.Pullup_Enable
114
PB7.Data
113
PB7.Control
112
PB7.Pullup_Enable
111
PG3.Data
110
PG3.Control
109
PG3.Pullup_Enable
108
PG4.Data
107
PG4.Control
106
PG4.Pullup_Enable
105
TOSC
104
TOSCON
Port B
Port G
32 kHz Timer Oscillator
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 106. ATmega64 Boundary-scan Order (Continued)
Bit Number
Signal Name
Module
103
RSTT
102
RSTHV
Reset Logic
(Observe-only)
101
EXTCLKEN
100
OSCON
99
RCOSCEN
98
OSC32EN
97
EXTCLK (XTAL1)
96
OSCCK
95
RCCK
94
OSC32CK
93
TWIEN
TWI
92
PD0.Data
Port D
91
PD0.Control
90
PD0.Pullup_Enable
89
PD1.Data
88
PD1.Control
87
PD1.Pullup_Enable
86
PD2.Data
85
PD2.Control
84
PD2.Pullup_Enable
83
PD3.Data
82
PD3.Control
81
PD3.Pullup_Enable
80
PD4.Data
79
PD4.Control
78
PD4.Pullup_Enable
77
PD5.Data
76
PD5.Control
75
PD5.Pullup_Enable
74
PD6.Data
73
PD6.Control
72
PD6.Pullup_Enable
71
PD7.Data
70
PD7.Control
69
PD7.Pullup_Enable
68
PG0.Data
Enable Signals for Main Clock/Oscillators
Clock Input and Oscillators for the Main Clock
(Observe-only)
Port G
273
2490N–AVR–05/08
Table 106. ATmega64 Boundary-scan Order (Continued)
274
Bit Number
Signal Name
Module
67
PG0.Control
Port G
66
PG0.Pullup_Enable
65
PG1.Data
64
PG1.Control
63
PG1.Pullup_Enable
62
PC0.Data
61
PC0.Control
60
PC0.Pullup_Enable
59
PC1.Data
58
PC1.Control
57
PC1.Pullup_Enable
56
PC2.Data
55
PC2.Control
54
PC2.Pullup_Enable
53
PC3.Data
52
PC3.Control
51
PC3.Pullup_Enable
50
PC4.Data
49
PC4.Control
48
PC4.Pullup_Enable
47
PC5.Data
46
PC5.Control
45
PC5.Pullup_Enable
44
PC6.Data
43
PC6.Control
42
PC6.Pullup_Enable
41
PC7.Data
40
PC7.Control
39
PC7.Pullup_Enable
38
PG2.Data
37
PG2.Control
36
PG2.Pullup_Enable
35
PA7.Data
34
PA7.Control
33
PA7.Pullup_Enable
32
PA6.Data
Port C
Port G
Port A
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 106. ATmega64 Boundary-scan Order (Continued)
Bit Number
Signal Name
Module
31
PA6.Control
Port A
30
PA6.Pullup_Enable
29
PA5.Data
28
PA5.Control
27
PA5.Pullup_Enable
26
PA4.Data
25
PA4.Control
24
PA4.Pullup_Enable
23
PA3.Data
22
PA3.Control
21
PA3.Pullup_Enable
20
PA2.Data
19
PA2.Control
18
PA2.Pullup_Enable
17
PA1.Data
16
PA1.Control
15
PA1.Pullup_Enable
14
PA0.Data
13
PA0.Control
12
PA0.Pullup_Enable
11
PF3.Data
10
PF3.Control
9
PF3.Pullup_Enable
8
PF2.Data
7
PF2.Control
6
PF2.Pullup_Enable
5
PF1.Data
4
PF1.Control
3
PF1.Pullup_Enable
2
PF0.Data
1
PF0.Control
0
PF0.Pullup_Enable
Notes:
Port F
1. PRIVATE_SIGNAL1 should always scanned in as zero.
2. PRIVATE_SIGNAL2 should always scanned in as zero.
275
2490N–AVR–05/08
Boundary-scan
Description
Language Files
276
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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Boot Loader
Support – ReadWhile-Write
Selfprogramming
The Boot Loader Support provides a real Read-While-Write Self-programming mechanism for
downloading and uploading program code by the MCU itself. This feature allows flexible application software updates controlled by the MCU using a Flash-resident Boot Loader program. The
Boot Loader program can use any available data 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 123 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 112 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 108 on page 280. The Application section can never store any
Boot Loader code since the SPM instruction is disabled when executed 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 programming 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 109 on page 280.
Read-While-Write
and No ReadWhile-Write Flash
Sections
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-WhileWrite (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:
•
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.
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 section” refers to which
277
2490N–AVR–05/08
section that is being programmed (erased or written), not which section that actually is being
read during a Boot Loader software update.
RWW – Read-WhileWrite 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 section. 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
“SPMCSR – Store Program Memory Control Register” on page 281. for details on how to clear
RWWSB.
NRWW – No ReadWhile-Write Section
The code located in the NRWW section can be read when the Boot Loader software is 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 107. Read-While-Write Features
Which Section does the Zpointer Address During the
Programming?
Which Section Can
be Read During
Programming?
Is the
CPU
Halted?
Read-WhileWrite
Supported?
RWW section
NRWW section
No
Yes
NRWW section
None
Yes
No
Figure 135. Read-While-Write vs. No Read-While-Write
Read-While-Write
(RWW) Section
Z-pointer
Addresses RWW
Section
Z-pointer
Addresses NRWW
Section
No Read-While-Write
(NRWW) Section
CPU is Halted
During the Operation
Code Located in
NRWW Section
Can be Read During
the Operation
278
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 136. Memory Sections(1)
Program Memory
BOOTSZ = '10'
Program Memory
BOOTSZ = '11'
$0000
Read-While-Write Section
Application Flash Section
End RWW
Start NRWW
Application Flash Section
Boot Loader Flash Section
End Application
Start Boot Loader
Flashend
No Read-While-Write Section
No Read-While-Write Section
Read-While-Write Section
$0000
Program Memory
BOOTSZ = '01'
Application Flash Section
End RWW
Start NRWW
Application Flash Section
End Application
Start Boot Loader
Boot Loader Flash Section
Flashend
Program Memory
BOOTSZ = '00'
$0000
Note:
Boot Loader Lock
Bits
Read-While-Write Section
Application Flash Section
End RWW
Start NRWW
Application Flash Section
End Application
Start Boot Loader
Boot Loader Flash Section
Flashend
No Read-While-Write Section
No Read-While-Write Section
Read-While-Write Section
$0000
Application flash Section
End RWW, End Application
Start NRWW, Start Boot Loader
Boot Loader Flash Section
Flashend
1. The parameters are given in Table 112 on page 289.
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 independently. 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 108 and Table 109 for further details. The Boot Lock bits can be set in software 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 mem-
279
2490N–AVR–05/08
ory by SPM instruction. Similarly, the general Read/Write Lock (Lock bit mode 3) does not
control reading nor writing by LPM/SPM, if it is attempted.
Table 108. Boot Lock Bit0 Protection Modes (Application Section)(1)
BLB0 Mode
BLB02
BLB01
1
1
1
No restrictions for SPM or LPM accessing the Application
section.
2
1
0
SPM is not allowed to write to the Application section.
3
0
0
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:
Protection
1. “1” means unprogrammed, “0” means programmed
Table 109. Boot Lock Bit1 Protection Modes (Boot Loader Section)(1)
BLB1 Mode
BLB12
BLB11
1
1
1
No restrictions for SPM or LPM accessing the Boot Loader
section.
2
1
0
SPM is not allowed to write to the Boot Loader section.
3
0
0
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:
280
Protection
1. “1” means unprogrammed, “0” means programmed
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Entering the Boot
Loader Program
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 interface. 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 executing 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 110. Boot Reset Fuse(1)
BOOTRST
Note:
SPMCSR – Store
Program Memory
Control Register
Reset Address
1
Reset Vector = Application Reset (address 0x0000)
0
Reset Vector = Boot Loader Reset (see Table 112 on page 289)
1. “1” means unprogrammed, “0” means programmed
The Store Program Memory Control Register contains the control bits needed to control the Boot
Loader operations.
Bit
7
6
5
4
3
2
1
0
SPMIE
RWWSB
–
RWWSRE
BLBSET
PGWRT
PGERS
SPMEN
Read/Write
R/W
R
R
R/W
R/W
R/W
R/W
R/W
Initial Value
0
0
0
0
0
0
0
0
(0x68)
SPMCSR
• 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
2490N–AVR–05/08
• 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 Zpointer 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)
2490N–AVR–05/08
ATmega64(L)
Addressing the
Flash During Selfprogramming
The Z-pointer is used to address the SPM commands.
Bit
15
14
13
12
11
10
9
8
ZH (R31)
Z15
Z14
Z13
Z12
Z11
Z10
Z9
Z8
ZL (R30)
Z7
Z6
Z5
Z4
Z3
Z2
Z1
Z0
7
6
5
4
3
2
1
0
Since the Flash is organized in pages (see Table 123 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 importance 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
Z - REGISTER
1 0
0
PCMSB
PROGRAM
COUNTER
PAGEMSB
PCPAGE
PAGE ADDRESS
WITHIN THE FLASH
PROGRAM MEMORY
PAGE
PCWORD
WORD ADDRESS
WITHIN A PAGE
PAGE
INSTRUCTION WORD
PCWORD[PAGEMSB:0]:
00
01
02
PAGEEND
Notes:
1. The different variables used in Figure 137 are listed in Table 113 on page 289.
2. PCPAGE and PCWORD are listed in Table 124 on page 296.
283
2490N–AVR–05/08
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
Filling the Temporary
Buffer (Page Loading)
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.
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:
Performing a Page
Write
284
If the EEPROM is written in the middle of an SPM Page Load operation, all data loaded will be
lost.
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.
ATmega64(L)
2490N–AVR–05/08
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 61.
Consideration While
Updating BLS
Special care must be taken if the user allows the Boot Loader section to be updated by 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 61, or the interrupts 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 Bits by
SPM
To set the Boot Loader Lock bits, write the desired data to R0, write “X0001001” to 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
7
6
5
4
3
2
1
0
R0
1
1
BLB12
BLB11
BLB02
BLB01
1
1
See Table 108 and Table 109 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
2490N–AVR–05/08
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 destination 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
7
6
5
4
3
2
1
0
Rd
–
–
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 119 on page 292 for a detailed
description and mapping of the Fuse Low bits.
Bit
7
6
5
4
3
2
1
0
Rd
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 118 on page 292 for detailed description and mapping of the Fuse High bits.
Bit
7
6
5
4
3
2
1
0
Rd
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 destination register as shown below.
Refer to Table 117 on page 291 for detailed description and mapping of the Fuse High bits.
Bit
7
6
5
4
3
2
1
0
Rd
–
–
–
–
–
–
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 voltage 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 correctly. 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):
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
286
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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, effectively 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 111 shows the typical programming time for Flash accesses from the CPU.
Table 111. SPM Programming Time
Symbol
Flash write (Page Erase, Page Write,
and write Lock bits by SPM)
Simple Assembly
Code Example for a
Boot Loader
Min Programming Time
Max Programming Time
3.7 ms
4.5 ms
;-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
;PAGESIZEB is page size in BYTES, not words
.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
looplo, low(PAGESIZEB) ;init loop variable
ldi
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
;use subi for PAGESIZEB<=256
brne Wrloop
; execute Page Write
subi ZL, low(PAGESIZEB)
;restore pointer
sbci ZH, high(PAGESIZEB)
;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
287
2490N–AVR–05/08
; read back and check, optional
ldi
looplo, low(PAGESIZEB) ;init loop variable
ldi
loophi, high(PAGESIZEB) ;not required for PAGESIZEB<=256
subi YL, low(PAGESIZEB)
;restore pointer
sbci YH, high(PAGESIZEB)
Rdloop:
lpm
r0, Z+
ld
r1, Y+
cpse r0, r1
jmp
Error
sbiw loophi:looplo, 1
;use subi for PAGESIZEB<=256
brne Rdloop
; return to RWW section
; verify that RWW section is safe to read
Return:
lds
temp1, SPMCSR
sbrs temp1, RWWSB
; If RWWSB is set, the RWW section is not ready yet
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
SPMCSR, spmcrval
spm
; restore SREG (to enable interrupts if originally enabled)
out
SREG, temp2
ret
288
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
ATmega64 Boot
Loader Parameters
In Table 112 through Table 114, the parameters used in the description of the Self-programming
are given.
Table 112. Boot Size Configuration(1)
Application
Flash
Section
Boot
Loader
Flash
Section
End
Application
Section
Boot Reset
Address
(Start Boot
Loader
Section)
BOOTSZ
1
BOOTSZ
0
Boot
Size
1
1
512
words
4
0x0000 0x7DFF
0x7E00 0x7FFF
0x7DFF
0x7E00
1
0
1024
words
8
0x0000 0x7BFF
0x7C00 0x7FFF
0x7BFF
0x7C00
0
1
2048
words
16
0x0000 0x77FF
0x7800 0x7FFF
0x77FF
0x7800
0
0
4096
words
32
0x0000 0x6FFF
0x7000 0x7FFF
0x6FFF
0x7000
Note:
Pages
1. The different BOOTSZ Fuse configurations are shown in Figure 136
Table 113. Read-While-Write Limit(1)
Section
Pages
Address
Read-While-Write (RWW)
224
0x0000 - 0x6FFF
No Read-While-Write (NRWW)
32
0x7000 - 0x7FFF
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 114. Explanation of Different Variables Used in Figure 137 and the Mapping to the Zpointer(1)(2)
Corresponding
Z-value
Variable
PCMSB
14
Most significant bit in the Program Counter.
(The Program Counter is 15 bits PC[14:0]).
6
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
Z15
Bit in Z-register that is mapped to PCMSB.
Because Z0 is not used, the ZPCMSB equals
PCMSB + 1.
Z7
Bit in Z-register that is mapped to PAGEMSB.
Because Z0 is not used, the ZPAGEMSB
equals PAGEMSB + 1.
PC[14:7]
Z15:Z8
Program Counter page address: Page select,
for Page Erase and Page Write
PC[6:0]
Z7:Z1
Program Counter word address: Word select,
for filling temporary buffer (must be zero during
Page Write operation)
ZPCMSB
ZPAGEMSB
PCPAGE
PCWORD
Notes:
Description
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
2490N–AVR–05/08
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 116. The Lock bits can only be
erased to “1” with the Chip Erase command.
Table 115. Lock Bit Byte(1)
Lock Bit Byte
Description
Default Value
7
–
1 (unprogrammed)
6
–
1 (unprogrammed)
BLB12
5
Boot Lock bit
1 (unprogrammed)
BLB11
4
Boot Lock bit
1 (unprogrammed)
BLB02
3
Boot Lock bit
1 (unprogrammed)
BLB01
2
Boot Lock bit
1 (unprogrammed)
LB2
1
Lock bit
1 (unprogrammed)
LB1
0
Lock bit
1 (unprogrammed)
Note:
Bit no
1. “1” means unprogrammed, “0” means programmed
Table 116. Lock Bit Protection Modes(2)
Memory Lock Bits
290
Protection Type
LB Mode
LB2
LB1
1
1
1
No memory lock features enabled.
2
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
1
0
SPM is not allowed to write to the Application section.
3
0
0
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
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 116. Lock Bit Protection Modes(2) (Continued)
Memory Lock Bits
1
1
1
No restrictions for SPM or LPM accessing the Boot Loader
section.
2
1
0
SPM is not allowed to write to the Boot Loader section.
3
0
0
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:
Fuse Bits
Protection Type
1. Program the Fuse bits before programming the Lock bits.
2. “1” means unprogrammed, “0” means programmed
The ATmega64 has three fuse bytes. Table 117 - Table 119 describe briefly the functionality 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 117. Extended Fuse Byte
Fuse Low Byte
Bit no
Description
Default Value
–
7
–
1
–
6
–
1
–
5
–
1
–
4
–
1
–
3
–
1
2
–
1
1
ATmega103 compatibility mode
0 (programmed)
0
Watchdog Timer always on
1 (unprogrammed)
–
(1)
M103C
(2)
WDTON
Notes:
1. See “ATmega103 and ATmega64 Compatibility” on page 4 for details.
2. See “WDTCR – Watchdog Timer Control Register” on page 57 for details.
291
2490N–AVR–05/08
Table 118. Fuse High Byte
Fuse High Byte
Bit no
Description
Default Value
OCDEN
7
Enable OCD
1 (unprogrammed, OCD
disabled)
JTAGEN(4)
6
Enable JTAG
0 (programmed, JTAG
enabled)
SPIEN(1)
5
Enable SPI Serial Program and
Data Downloading
0 (programmed, SPI prog.
enabled)
CKOPT(2)
4
Oscillator options
1 (unprogrammed)
EESAVE
3
EEPROM memory is preserved
through the Chip Erase
1 (unprogrammed,
EEPROM not preserved)
BOOTSZ1
2
Select Boot Size (see Table 112
for details)
0 (programmed)(3)
BOOTSZ0
1
Select Boot Size (see Table 112
for details)
0 (programmed)(3)
BOOTRST
0
Select Reset Vector
1 (unprogrammed)
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 38 for details.
3. The default value of BOOTSZ1..0 results in maximum Boot Size. See Table 112 on page 289
4. If the JTAG interface is left unconnected, the JTAGEN fuse should if possible be disabled. This
to avoid static current at the TDO pin in the JTAG interface
Table 119. 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
Select start-up time
1 (unprogrammed)(1)
SUT0
4
Select start-up time
0 (programmed)(1)
CKSEL3
3
Select Clock source
0 (programmed)(2)
CKSEL2
2
Select Clock source
0 (programmed)(2)
CKSEL1
1
Select Clock source
0 (programmed)(2)
CKSEL0
0
Select Clock source
1 (unprogrammed)(2)
Notes:
1. The default value of SUT1..0 results in maximum start-up time. See Table 14 on page 43 for
details.
2. The default setting of CKSEL3..0 results in internal RC Oscillator @ 1 MHz. See Table 6 on
page 38 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)
2490N–AVR–05/08
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 “OSCCAL – Oscillator Calibration Register(1)” on page 43 for details.
Parallel
Programming
Parameters, Pin
Mapping, and
Commands
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.
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 120. 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 positive pulse.
The bit coding is shown in Table 122.
When pulsing WR or OE, the command loaded determines the action executed. The different
Commands are shown in Table 123.
293
2490N–AVR–05/08
Figure 138. Parallel Programming
+5V
RDY/BSY
PD1
OE
PD2
WR
PD3
BS1
PD4
XA0
PD5
XA1
PD6
PAGEL
PD7
+12 V
BS2
VCC
+5V
AVCC
PB7 - PB0
DATA
RESET
PA0
XTAL1
GND
294
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 120. 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
I
Output Enable (Active low)
WR
PD3
I
Write Pulse (Active low)
BS1
PD4
I
Byte Select 1 (“0” selects low byte, “1” selects high
byte)
XA0
PD5
I
XTAL Action Bit 0
XA1
PD6
I
XTAL Action Bit 1
PAGEL
PD7
I
Program Memory and EEPROM data Page Load
BS2
PA0
I
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 121. Pin Values Used to Enter Programming Mode
Pin
Symbol
Value
PAGEL
Prog_enable[3]
0
XA1
Prog_enable[2]
0
XA0
Prog_enable[1]
0
BS1
Prog_enable[0]
0
Table 122. XA1 and XA0 Coding
XA1
XA0
Action when XTAL1 is Pulsed
0
0
Load Flash or EEPROM Address (High or low address byte determined by BS1)
0
1
Load Data (High or Low data byte for Flash determined by BS1)
1
0
Load Command
1
1
No Action, Idle
295
2490N–AVR–05/08
Table 123. Command Byte Bit Coding
Command Byte
Command Executed
1000 0000
Chip Erase
0100 0000
Write Fuse Bits
0010 0000
Write Lock Bits
0001 0000
Write Flash
0001 0001
Write EEPROM
0000 1000
Read Signature Bytes and Calibration byte
0000 0100
Read Fuse and Lock Bits
0000 0010
Read Flash
0000 0011
Read EEPROM
Table 124. 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 125. No. of Words in a Page and no. of Pages in the EEPROM
296
EEPROM Size
Page Size
PCWORD
No. of Pages
PCPAGE
EEAMSB
2K bytes
8 bytes
EEA[2:0]
256
EEA[10:3]
10
ATmega64(L)
2490N–AVR–05/08
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 121 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 Programming 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
Chip Erase
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.
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.
Programming the
Flash
The Flash is organized in pages, see Table 123 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 pro-
297
2490N–AVR–05/08
grammed simultaneously. The following procedure describes how to program 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.
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.
298
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 signals are
reset.
Figure 139. Addressing the Flash which is Organized in Pages(1)
PCMSB
PROGRAM
COUNTER
PAGEMSB
PCPAGE
PCWORD
PAGE ADDRESS
WITHIN THE FLASH
WORD ADDRESS
WITHIN A PAGE
PROGRAM MEMORY
PAGE
PAGE
PCWORD[PAGEMSB:0]:
00
INSTRUCTION WORD
01
02
PAGEEND
Note:
1. PCPAGE and PCWORD are listed in Table 123 on page 296.
Figure 140. Programming the Flash Waveforms(1)
F
A
DATA
$10
B
ADDR. LOW
C
DATA LOW
D
E
B
DATA HIGH
XX
ADDR. LOW
C
DATA LOW
D
E
G
DATA HIGH
XX
ADDR. HIGH
H
XX
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.
299
2490N–AVR–05/08
Programming the
EEPROM
The EEPROM is organized in pages, see Table 124 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 297 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
DATA
A
G
B
0x11
ADDR. HIGH
ADDR. LOW
C
E
B
DATA
XX
ADDR. LOW
C
E
DATA
XX
L
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 297 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”.
300
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Reading the EEPROM
The algorithm for reading the EEPROM memory is as follows (refer to “Programming the Flash”
on page 297 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 297 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 297 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 “Programming the
Flash” on page 297 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.
Figure 142. Programming the FUSES Waveforms
Write Fuse Low byte
DATA
A
C
0x40
DATA
XX
Write Fuse high byte
A
C
0x40
DATA
XX
Write Extended Fuse byte
A
C
0x40
DATA
XX
XA1
XA0
BS1
BS2
XTAL1
WR
RDY/BSY
RESET +12V
OE
PAGEL
301
2490N–AVR–05/08
Programming the Lock
Bits
The algorithm for programming the Lock bits is as follows (refer to “Programming the Flash” on
page 297 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 297 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”.
Figure 143. Mapping Between BS1, BS2 and the Fuse and Lock Bits during Read
Fuse Low Byte
0
Extended Fuse Byte
1
0
BS2
1
0
Lock Bits
DATA
BS1
Fuse High Byte
1
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
302
The algorithm for reading the Calibration bytes is as follows (refer to “Programming the Flash”
for details on Command and Address loading):
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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
t XLWL
t XHXL
XTAL1
t DVXH
t XLDX
Data & Contol
(DATA, XA0/1, BS1, BS2)
t PLBX
t BVPH
PAGEL
t BVWL
t WLBX
t PHPL
t WL
WR
WH
t PLWL
WLRL
RDY/BSY
t WLRH
Figure 145. Parallel Programming Timing, Loading Sequence with Timing Requirements(1)
LOAD ADDRESS
(LOW BYTE)
LOAD DATA LOAD DATA
(HIGH BYTE)
LOAD DATA
(LOW BYTE)
t XLXH
t XLPH
LOAD ADDRESS
(LOW BYTE)
t PLXH
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.
303
2490N–AVR–05/08
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)
t XLOL
XTAL1
t BHDV
BS1
t OLDV
OE
DATA
t OHDZ
ADDR0 (Low Byte)
DATA (Low Byte)
DATA (High Byte)
ADDR1 (Low Byte)
XA0
XA1
Note:
304
1. The timing requirements shown in Figure 144 (i.e. tDVXH, tXHXL, and tXLDX) also apply to reading operation.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 126. Parallel Programming Characteristics, VCC = 5V ± 10%
Symbol
Parameter
Min
VPP
Programming Enable Voltage
11.5
IPP
Programming Enable Current
tDVXH
Data and Control Valid before XTAL1 High
67
ns
tXLXH
XTAL1 Low to XTAL1 High
200
ns
tXHXL
XTAL1 Pulse Width High
150
ns
tXLDX
Data and Control Hold after XTAL1 Low
67
ns
tXLWL
XTAL1 Low to WR Low
0
ns
tXLPH
XTAL1 Low to PAGEL high
0
ns
tPLXH
PAGEL low to XTAL1 high
150
ns
tBVPH
BS1 Valid before PAGEL High
67
ns
tPHPL
PAGEL Pulse Width High
150
ns
tPLBX
BS1 Hold after PAGEL Low
67
ns
tWLBX
BS2/1 Hold after WR Low
67
ns
tPLWL
PAGEL Low to WR Low
67
ns
tBVWL
BS1 Valid to WR Low
67
ns
tWLWH
WR Pulse Width Low
150
ns
tWLRL
WR Low to RDY/BSY Low
Max
Units
12.5
V
250
μA
tWLRH
0
1
μs
WR Low to RDY/BSY High(1)
3.7
4.5
ms
tWLRH_CE
WR Low to RDY/BSY High for Chip Erase(2)
7.5
9
ms
tXLOL
XTAL1 Low to OE Low
0
tBVDV
BS1 Valid to DATA valid
0
tOLDV
tOHDZ
Notes:
Serial
Downloading
Typ
ns
250
ns
OE Low to DATA Valid
250
ns
OE High to DATA Tri-stated
250
ns
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.
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 127 on page 306, 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 description 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.
305
2490N–AVR–05/08
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 127.
Table 127. Pin Mapping SPI Serial Programming
Symbol
Pins
I/O
Description
MOSI (PDI)
PE0
I
Serial Data In
MISO (PDO)
PE1
O
Serial Data Out
SCK
PB1
I
Serial Clock
Figure 147. SPI Serial Programming and Verify(1)
+2.7 - 5.5V
VCC
MOSI
PE0
MISO
PE1
SCK
PB1
+2.7 - 5.5V (2)
AVCC
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 programming
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 following
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 CPU clock cycles duration
306
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 Programming
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 Programming Enable command.
4. The Flash is programmed one page at a time. The Page size is found in Table 124 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 Memory 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 tWD_FLASH before issuing the next page. (See Table 128). 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 128).
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:
Data Polling Flash
If other commands that polling (read) are applied before any write operation (FLASH, EEPROM,
Lock bits, Fuses) is completed, may result in incorrect programming.
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 128 for tWD_FLASH value.
307
2490N–AVR–05/08
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 determine 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 128 for tWD_EEPROM value.
Table 128. Minimum Wait Delay before Writing the Next Flash or EEPROM Location
Symbol
Minimum Wait Delay
tWD_FUSE
4.5 ms
tWD_FLASH(1)
4.5 ms
tWD_EEPROM
9.0 ms
tWD_ERASE
9.0 ms
Note:
1. Flash write: per page
Figure 148. SPI Serial Programming Waveforms
SERIAL DATA INPUT
(MOSI)
MSB
LSB
SERIAL DATA OUTPUT
(MISO)
MSB
LSB
SERIAL CLOCK INPUT
(SCK)
SAMPLE
308
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 129. 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
xaaa aaaa
bxxx xxxx
xxxx xxxx
Write Program Memory Page at
address a:b.
Read EEPROM Memory
1010 0000
xxxx xaaa
bbbb bbbb
oooo oooo
Read data o from EEPROM
memory at address a:b.
Write EEPROM Memory
1100 0000
xxxx xaaa
bbbb bbbb
iiii iiii
Write data i to EEPROM memory
at address a:b.
Read Lock Bits
0101 1000
0000 0000
xxxx xxxx
xxoo oooo
Read Lock bits. “0” = programmed,
“1” = unprogrammed. See Table
115 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 115
on page 290 for details.
Read Signature Byte
0011 0000
xxxx xxxx
xxxx xxbb
oooo oooo
Read Signature Byte o at address
b.
Write Fuse Bits
1010 1100
1010 0000
xxxx xxxx
iiii iiii
Set bits = “0” to program, “1” to
unprogram. See Table 119 on page
292 for details.
Write Fuse High Bits
1010 1100
1010 1000
xxxx xxxx
iiii iiii
Set bits = “0” to program, “1” to
unprogram. See Table 118 on page
292 for details.
Write Extended Fuse Bits
1010 1100
1010 0100
xxxx xxxx
xxxx xxii
Set bits = “0” to program, “1” to
unprogram. See Table 119 on page
292 for details.
Read Fuse Bits
0101 0000
0000 0000
xxxx xxxx
oooo oooo
Read Fuse bits. “0” = programmed,
“1” = unprogrammed. See Table
119 on page 292 for details.
309
2490N–AVR–05/08
Table 129. SPI Serial Programming Instruction Set (Continued)
Instruction Format
Instruction
Byte 1
Byte 2
Byte 3
Byte4
Read Extendend Fuse
Bits
0101 0000
0000 1000
xxxx xxxx
oooo oooo
Read Extended Fuse bits. “0” =
pro-grammed, “1” =
unprogrammed. See Table 119 on
page 292 for details.
Read Fuse High Bits
0101 1000
0000 1000
xxxx xxxx
oooo oooo
Read Fuse high bits. “0” = programmed, “1” = unprogrammed.
See Table 118 on page 292 for
details.
Read Calibration Byte
0011 1000
00xx xxxx
0000 00bb
oooo oooo
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
310
Operation
For characteristics of the SPI module, see “SPI Timing Characteristics” on page 330.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 instructions
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.
311
2490N–AVR–05/08
Figure 149. State Machine Sequence for Changing the Instruction Word
1
Test-Logic-Reset
0
0
Run-Test/Idle
1
Select-DR Scan
1
Select-IR Scan
0
1
0
1
Capture-DR
Capture-IR
0
0
Shift-DR
Shift-IR
0
1
Exit1-DR
0
Pause-DR
0
0
Pause-IR
1
1
0
Exit2-DR
Exit2-IR
1
1
Update-DR
AVR_RESET (0xC)
1
Exit1-IR
0
1
0
1
1
0
1
Update-IR
0
1
0
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:
•
312
Shift-DR: The Reset Register is shifted by the TCK input.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
PROG_ENABLE (0x4)
PROG_COMMANDS
(0x5)
PROG_PAGELOAD
(0x6)
The AVR specific public JTAG instruction for enabling programming via the JTAG port. The 16bit 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.
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 130 on page 316).
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 8bit. Unlike most JTAG instructions, the Update-DR state is not used to transfer data from the
Shift Register. The data are automatically transferred 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:
PROG_PAGEREAD
(0x7)
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.
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 transferred 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.
313
2490N–AVR–05/08
Data Registers
The data registers are selected by the JTAG instruction registers described in section “Programming Specific JTAG Instructions” on page 311. The data registers relevant for programming
operations are:
•
Reset Register
Reset Register
•
Programming Enable Register
•
Programming Command Register
•
Virtual Flash Page Load Register
•
Virtual Flash Page Read 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 38) 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, programming 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
D
A
T
A
$A370
=
D
Q
Programming Enable
ClockDR & PROG_ENABLE
TDO
314
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Programming
Command Register
The Programming Command Register is a 15-bit register. This register is used to serially 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 130. 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
A
D
D
R
E
S
S
/
D
A
T
A
Flash
EEPROM
Fuses
Lock Bits
TDO
315
2490N–AVR–05/08
Table 130. 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
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
0110011_10000000
xxxxxox_xxxxxxxx
2a. Enter Flash Write
0100011_00010000
xxxxxxx_xxxxxxxx
2b. Load Address High Byte
0000111_aaaaaaaa
xxxxxxx_xxxxxxxx
2c. Load Address Low Byte
0000011_bbbbbbbb
xxxxxxx_xxxxxxxx
2d. Load Data Low Byte
0010011_iiiiiiii
xxxxxxx_xxxxxxxx
2e. Load Data High Byte
0010111_iiiiiiii
xxxxxxx_xxxxxxxx
2f. Latch Data
0110111_00000000
1110111_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
2g. Write Flash Page
0110111_00000000
0110101_00000000
0110111_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
2h. Poll for Page Write Complete
0110111_00000000
xxxxxox_xxxxxxxx
(2)
3a. Enter Flash Read
0100011_00000010
xxxxxxx_xxxxxxxx
3b. Load Address High Byte
0000111_aaaaaaaa
xxxxxxx_xxxxxxxx
3c. Load Address Low Byte
0000011_bbbbbbbb
xxxxxxx_xxxxxxxx
3d. Read Data Low and High Byte
0110010_00000000
0110110_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
xxxxxxx_oooooooo
4a. Enter EEPROM Write
0100011_00010001
xxxxxxx_xxxxxxxx
4b. Load Address High Byte
0000111_aaaaaaaa
xxxxxxx_xxxxxxxx
4c. Load Address Low Byte
0000011_bbbbbbbb
xxxxxxx_xxxxxxxx
4d. Load Data Byte
0010011_iiiiiiii
xxxxxxx_xxxxxxxx
4e. Latch Data
0110111_00000000
1110111_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
4f. Write EEPROM Page
0110011_00000000
0110001_00000000
0110011_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
4g. Poll for Page Write Complete
0110011_00000000
xxxxxox_xxxxxxxx
(2)
5a. Enter EEPROM Read
0100011_00000011
xxxxxxx_xxxxxxxx
5b. Load Address High Byte
0000111_aaaaaaaa
xxxxxxx_xxxxxxxx
5c. Load Address Low Byte
0000011_bbbbbbbb
xxxxxxx_xxxxxxxx
316
Notes
(2)
(9)
(9)
low byte
high byte
(9)
(9)
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 130. 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
5d. Read Data Byte
0110011_bbbbbbbb
0110010_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
0100011_01000000
xxxxxxx_xxxxxxxx
6b. Load Data Low Byte
0010011_iiiiiiii
xxxxxxx_xxxxxxxx
(3)
6c. Write Fuse Extended Byte
0111011_00000000
0111001_00000000
0111011_00000000
0111011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
6d. Poll for Fuse Write Complete
0111011_00000000
xxxxxox_xxxxxxxx
(2)
0010011_iiiiiiii
xxxxxxx_xxxxxxxx
(3)
6f. Write Fuse High Byte
0110111_00000000
0110101_00000000
0110111_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
6g. Poll for Fuse Write Complete
0110111_00000000
xxxxxox_xxxxxxxx
(2)
6h. Load Data Low Byte
0010011_iiiiiiii
xxxxxxx_xxxxxxxx
(3)
6i. Write Fuse Low byte
0110011_00000000
0110001_00000000
0110011_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
6j. Poll for Fuse Write Complete
0110011_00000000
xxxxxox_xxxxxxxx
(2)
7a. Enter Lock Bit Write
0100011_00100000
xxxxxxx_xxxxxxxx
7b. Load Data Byte(9)
0010011_11iiiiii
xxxxxxx_xxxxxxxx
(4)
7c. Write Lock Bits
0110011_00000000
0110001_00000000
0110011_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
xxxxxxx_xxxxxxxx
(1)
7d. Poll for Lock Bit Write Complete
0110011_00000000
xxxxxox_xxxxxxxx
(2)
8a. Enter Fuse/Lock Bit Read
0100011_00000100
xxxxxxx_xxxxxxxx
8b. Read Fuse Extended Byte
0111010_00000000
0111111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
8c. Read Fuse High Byte(7)
0111110_00000000
0111111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
8d. Read Fuse Low Byte(8)
0110010_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
8e. Read Lock Bits(9)
0110110_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_xxoooooo
6a. Enter Fuse Write
(6)
6e. Load Data Low Byte
(7)
(8)
(6)
Notes
(5)
317
2490N–AVR–05/08
Table 130. 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
0100011_00001000
xxxxxxx_xxxxxxxx
9b. Load Address Byte
0000011_bbbbbbbb
xxxxxxx_xxxxxxxx
9c. Read Signature Byte
0110010_00000000
0110011_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
10a. Enter Calibration Byte Read
0100011_00001000
xxxxxxx_xxxxxxxx
10b. Load Address Byte
0000011_bbbbbbbb
xxxxxxx_xxxxxxxx
10c. Read Calibration Byte
0110110_00000000
0110111_00000000
xxxxxxx_xxxxxxxx
xxxxxxx_oooooooo
Notes:
318
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 117 on page 291.
7. The bit mapping for Fuses High byte is listed in Table 118 on page 292.
8. The bit mapping for Fuses Low byte is listed in Table 119 on page 292.
9. The bit mapping for Lock bits byte is listed in Table 115 on page 290.
10. Address bits exceeding PCMSB and EEAMSB (Table 123 and Table 124) are don’t care.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 152. State Machine Sequence for Changing/Reading the Data Word
1
Test-Logic-Reset
0
0
Run-Test/Idle
1
Select-DR Scan
1
Select-IR Scan
0
1
0
1
Capture-DR
Capture-IR
0
0
Shift-DR
Shift-IR
0
1
Exit1-DR
1
Exit1-IR
0
0
Pause-DR
0
0
Pause-IR
1
1
0
Exit2-DR
Exit2-IR
1
1
Update-DR
1
0
1
1
0
1
Update-IR
0
1
0
319
2490N–AVR–05/08
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
TDI
Flash
EEPROM
Fuses
Lock Bits
D
A
T
A
TDO
Virtual Flash Page
Read Register
320
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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 154. Virtual Flash Page Read Register
STROBES
State
Machine
ADDRESS
TDI
Flash
EEPROM
Fuses
Lock Bits
D
A
T
A
TDO
Programming
Algorithm
All references below of type “1a”, “1b”, and so on, refer to Table 130.
Entering Programming
Mode
1. Enter JTAG instruction AVR_RESET and shift 1 in the Reset Register.
Leaving Programming
Mode
1. Enter JTAG instruction PROG_COMMANDS.
2. Enter instruction PROG_ENABLE and shift 1010_0011_0111_0000 in the Programming
Enable Register.
2. Disable all programming instructions by using no operation instruction 11a.
3. Enter instruction PROG_ENABLE and shift 0000_0000_0000_0000 in the Programming
Enable Register.
4. Enter JTAG instruction AVR_RESET and shift 0 in the Reset Register.
Performing Chip Erase 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 tWLRH_CE (refer
to Table 1 on page 304).
321
2490N–AVR–05/08
Programming the
Flash
Before programming the Flash, a Chip Erase must be performed. See “Performing Chip Erase”
on page 321.
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 tWLRH_FLASH
(refer to Table 1 on page 304).
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 123 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 tWLRH_FLASH
(refer to Table 1 on page 304).
9. Repeat steps 3 to 8 until all data have been programmed.
322
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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 123 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 321.
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 tWLRH
(refer to Table 1 on page 304).
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
323
2490N–AVR–05/08
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 program 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 304).
6. Load data Low byte using programming instructions 6e. A bit value of “0” will program 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 304).
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 304).
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 tWLRH (refer
to Table 1 on page 304).
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
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.
Reading the
Calibration Byte
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.
324
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Electrical Characteristics
Absolute Maximum Ratings*
Operating Temperature.................................. -55°C to +125°C
*NOTICE:
Storage Temperature ..................................... -65°C to +150°C
Voltage on any Pin except RESET
with respect to Ground ................................-0.5V to VCC+0.5V
Voltage on RESET with respect to Ground......-0.5V to +13.0V
Stresses beyond those listed under “Absolute
Maximum Ratings” may cause permanent damage 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.
Maximum Operating Voltage ............................................ 6.0V
DC Current per I/O Pin ............................................... 40.0 mA
DC Current VCC and GND Pins..................... 200.0 - 400.0mA
DC Characteristics
TA = -40°C to 85°C, VCC = 2.7V to 5.5V (unless otherwise noted)
Symbol
Parameter
Condition
Min
VIL
Input Low Voltage except
XTAL1 and RESET pins
VCC=2.7V - 5.5V
VIH
Input High Voltage except
XTAL1 and RESET pins
VIL1
Typ
Max
Units
-0.5
0.2 VCC(1)
V
VCC=2.7V - 5.5V
0.6 VCC(2)
VCC + 0.5
V
Input Low Voltage
XTAL1 pin
VCC=2.7V - 5.5V
-0.5
0.1 VCC(1)
V
VIH1
Input High Voltage
XTAL1 pin
VCC=2.7V - 5.5V
0.7 VCC(2)
VCC + 0.5
V
VIL2
Input Low Voltage
RESET pin
VCC=2.7V - 5.5V
-0.5
0.2 VCC(1)
V
VIH2
Input High Voltage
RESET pin
VCC=2.7V - 5.5V
0.85 VCC(2)
VCC + 0.5
V
VOL
Output Low Voltage(3)
(Ports A,B,C,D, E, F, G)
IOL = 20 mA, VCC = 5V
IOL = 10 mA, VCC = 3V
0.7
0.5
V
V
VOH
Output High Voltage(4)
(Ports A,B,C,D, E, F, G))
IOH = -20 mA, VCC = 5V
IOH = -10 mA, VCC = 3V
IIL
Input Leakage
Current I/O Pin
Vcc = 5.5V, pin low
(absolute value)
1.0
µA
IIH
Input Leakage
Current I/O Pin
Vcc = 5.5V, pin high
(absolute value)
1.0
µA
RRST
Reset Pull-up Resistor
30
60
kΩ
RPEN
PEN Pull-up Resistor
30
60
kΩ
RPU
I/O Pin Pull-up Resistor
20
50
kΩ
4.2
2.2
V
V
325
2490N–AVR–05/08
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
mA
Idle 4 MHz, VCC = 3V
(ATmega64L)
2
mA
Idle 8 MHz, VCC = 5V
(ATmega64)
12
mA
Power Supply Current
ICC
Power-down mode(5)
WDT enabled, VCC = 3V
< 10
20
µA
WDT disabled, VCC = 3V
<4
10
µA
VACIO
Analog Comparator
Input Offset Voltage
VCC = 5V
Vin = VCC/2
-40
40
mV
IACLK
Analog Comparator
Input Leakage Current
VCC = 5V
Vin = VCC/2
-50
50
nA
tACPD
Analog Comparator
Propagation Delay
VCC = 2.7V
VCC = 4.0V
Notes:
326
750
500
ns
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 QFN/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 100 mA.
3] The sum of all IOL, for ports C0 - C2, G0 - G1, D0 - D7, XTAL2 should not exceed 100 mA.
4] The sum of all IOL, for ports B0 - B7, G3 - G4, E0 - E7 should not exceed 100 mA.
5] The sum of all IOL, for ports F0 - F7, should not exceed 100 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 QFN/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 100 mA.
3] The sum of all IOH, for ports C0 - C2, G0 - G1, D0 - D7, XTAL2 should not exceed 100 mA.
4] The sum of all IOH, for ports B0 - B7, G3 - G4, E0 - E7 should not exceed 100 mA.
5] The sum of all IOH, for ports F0 - F7, should not exceed 100 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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
External Clock
Drive Waveforms
Figure 155. External Clock Drive Waveforms
V IH1
V IL1
External Clock
Drive
Table 131. External Clock Drive(1)
VCC = 2.7V to 5.5V
VCC = 4.5V to 5.5V
Min
Max
Min
Max
Units
0
8
0
16
MHz
Symbol
Parameter
1/tCLCL
Oscillator Frequency
tCLCL
Clock Period
125
62.5
ns
tCHCX
High Time
50
25
ns
tCLCX
Low Time
50
25
ns
tCLCH
Rise Time
1.6
0.5
μs
tCHCL
Fall Time
1.6
0.5
μs
ΔtCLCL
Change in period from
one clock cycle to the
next
2
2
%
Note:
1. Refer to “External Clock” on page 44 for details.
Table 132. External RC Oscillator, Typical Frequencies
Notes:
R [kΩ](1)
C [pF]
f(2)
31.5
20
650 kHz
6.5
20
2.0 MHz
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.
2. The frequency will vary with package type and board layout.
327
2490N–AVR–05/08
Two-wire Serial Interface Characteristics
Table 133 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 133. Two-wire Serial Bus Requirements
Symbol
Parameter
VIL
VIH
Vhys
(1)
Min
Max
Units
Input Low-voltage
-0.5
0.3 VCC
V
Input High-voltage
0.7 VCC
VCC + 0.5
V
–
V
Hysteresis of Schmitt Trigger Inputs
VOL(1)
Output Low-voltage
tr(1)
Rise Time for both SDA and SCL
tof(1)
Output Fall Time from VIHmin to VILmax
tSP(1)
Spikes Suppressed by Input Filter
Ii
Input Current each I/O Pin
Ci(1)
Capacitance for each I/O Pin
fSCL
SCL Clock Frequency
Rp
Hold Time (repeated) START Condition
tLOW
Low Period of the SCL Clock
tHIGH
High period of the SCL clock
tSU;STA
Set-up time for a repeated START condition
tHD;DAT
Data hold time
tSU;DAT
Data setup time
tSU;STO
Setup time for STOP condition
tBUF
Bus free time between a STOP and START
condition
328
0.05 VCC
3 mA sink current
10 pF < Cb < 400 pF(3)
1.
2.
3.
4.
(2)
0
0.4
V
(3)(2)
300
ns
20 + 0.1Cb(3)(2)
250
ns
(2)
ns
20 + 0.1Cb
0
0.1VCC < Vi < 0.9VCC
50
-10
10
µA
–
10
pF
fCK(4) > max(16fSCL, 250 kHz)(5)
0
400
kHz
fSCL ≤ 100 kHz
V CC – 0,4V
---------------------------3mA
1000ns
------------------Cb
Ω
fSCL > 100 kHz
V CC – 0,4V
---------------------------3mA
300ns
---------------Cb
Ω
fSCL ≤ 100 kHz
4.0
–
µs
fSCL > 100 kHz
Value of Pull-up resistor
tHD;STA
Notes:
Condition
0.6
–
µs
(6)
fSCL ≤ 100 kHz
4.7
–
µs
fSCL > 100 kHz(7)
1.3
–
µs
fSCL ≤ 100 kHz
4.0
–
µs
fSCL > 100 kHz
0.6
–
µs
fSCL ≤ 100 kHz
4.7
–
µs
fSCL > 100 kHz
0.6
–
µs
fSCL ≤ 100 kHz
0
3.45
µs
fSCL > 100 kHz
0
0.9
µs
fSCL ≤ 100 kHz
250
–
ns
fSCL > 100 kHz
100
–
ns
fSCL ≤ 100 kHz
4.0
–
µs
fSCL > 100 kHz
0.6
–
µs
fSCL ≤ 100 kHz
4.7
–
µs
In ATmega64, this parameter is characterized and not 100% tested.
Required only for fSCL > 100 kHz.
Cb = capacitance of one bus line in pF.
fCK = CPU clock frequency
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
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
tof
tHIGH
tLOW
tr
tLOW
SCL
tSU;STA
SDA
tHD;STA
tHD;DAT
tSU;DAT
tSU;STO
tBUF
329
2490N–AVR–05/08
SPI Timing
Characteristics
See Figure 157 on page 330 and Figure 158 on page 331 for details.
Table 134. SPI Timing Parameters
Description
Mode
Min
Typ
Max
1
SCK period
Master
See Table 72
2
SCK high/low
Master
50% duty cycle
3
Rise/Fall time
Master
3.6
4
Setup
Master
10
5
Hold
Master
10
6
Out to SCK
Master
0.5 • tsck
7
SCK to out
Master
10
8
SCK to out high
Master
10
9
SS low to out
Slave
15
10
SCK period
Slave
4 • tck
Slave
2 • tck
11
SCK high/low
(1)
12
Rise/Fall time
Slave
13
Setup
Slave
10
14
Hold
Slave
tck
15
SCK to out
Slave
16
SCK to SS high
Slave
17
SS high to tri-state
Slave
18
SS low to SCK
Slave
Note:
ns
1.6
µs
15
ns
20
10
20
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
MISO
(Data Input)
5
3
MSB
...
LSB
7
MOSI
(Data Output)
330
MSB
8
...
LSB
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 158. SPI Interface Timing Requirements (Slave Mode)
18
SS
10
9
16
SCK
(CPOL = 0)
11
11
SCK
(CPOL = 1)
13
MOSI
(Data Input)
14
12
MSB
...
LSB
15
MISO
(Data Output)
MSB
17
...
LSB
X
331
2490N–AVR–05/08
ADC Characteristics
Table 135. ADC Characteristics, Single Ended Channels, -40°C – 85°C
Symbol
Parameter
Condition
Resolution
Single Ended Conversion
Units
10
Bits
LSB
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 1 MHz
3
LSB
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
Noise Reduction mode
1.5
LSB
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 1 MHz
Noise Reduction mode
3
LSB
Integral Non-Linearity (INL)
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
0.75
LSB
Differential Non-Linearity (DNL)
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
0.25
LSB
Gain Error
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
0.75
LSB
Offset error
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
0.75
LSB
Clock Frequency
50
Conversion Time
13
Analog Supply Voltage
VREF
Reference Voltage
Input Voltage
ADC Conversion Output
VINT
Internal Voltage Reference
RREF
Reference Input Resistance
RAIN
Analog Input Resistance
1000
kHz
260
µs
(2)
(1)
VCC –0.3
VCC + 0.3
V
2.0
AVCC
V
GND
VREF
V
0
1023
LSB
Input Bandwidth
332
Max
1.5
AVCC
Notes:
Typ
Single Ended Conversion
VREF = 4V, VCC = 4V
ADC clock = 200 kHz
Absolute Accuracy
(Including INL, DNL, Quantization Error, Gain
and Offset Error)
VIN
Min
38.5
2.4
2.56
kHz
2.8
32
V
kΩ
100
MΩ
1. Minimum for AVCC is 2.7V.
2. Maximum for AVCC is 5.5V.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 136. ADC Characteristics, Differential Channels, -40°C – 85°C
Symbol
Parameter
Condition
Max
Units
1x
10
Bits
Gain = 10x
10
Bits
Gain = 200x
10
Bits
Gain =
Resolution
Absolute Accuracy
Integral Non-Linearity (INL)
(Accuracy after Calibration for Offset and
Gain Error)
Offset Error
16
LSB
Gain = 10x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
16
LSB
Gain = 200x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
8
LSB
Gain = 1x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
0.75
LSB
Gain = 10x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
0.75
LSB
Gain = 200x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
2.5
LSB
1x
1.6
%
Gain = 10x
1.6
%
Gain = 200x
0.3
%
Gain = 1x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
1.5
LSB
Gain = 10x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
1
LSB
Gain = 200x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
6
LSB
Clock Frequency
50
1000
Conversion Time
13
260
µs
VCC + 0.3
V
2.0
AVCC – 0.5
V
GND
VCC
V
Input Differential Voltage
-VREF/Gain
VREF/Gain
V
ADC Conversion Output
-511
511
LSB
Analog Supply Voltage
VREF
Reference Voltage
VDIFF
kHz
(2)
AVCC
VIN
Typ
Gain = 1x
VREF = 4V, VCC = 5V
ADC clock = 50 - 200 kHz
Gain =
Gain Error
Min
Input Voltage
Input Bandwidth
VCC –0.3(1)
4
kHz
333
2490N–AVR–05/08
Table 136. ADC Characteristics, Differential Channels, -40°C – 85°C (Continued)
Symbol
Parameter
Condition
Min
Typ
Max
Units
2.3
2.56
2.7
V
VINT
Internal Voltage Reference
RREF
Reference Input Resistance
32
kΩ
RAIN
Analog Input Resistance
100
MΩ
Notes:
334
1. Minimum for AVCC is 2.7V.
2. Maximum for AVCC is 5.5V.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
External Data Memory Timing
Table 137. External Data Memory Characteristics, 4.5 - 5.5 Volts, No Wait-state
8 MHz Oscillator
Min
Max
Variable Oscillator
Symbol
Parameter
Min
Max
Unit
0
1/tCLCL
Oscillator Frequency
0.0
16
MHz
1
tLHLL
ALE Pulse Width
115
1.0tCLCL-10
ns
2
tAVLL
Address Valid A to ALE Low
57.5
0.5tCLCL-5(1)
ns
3a
tLLAX_ST
Address Hold After ALE Low,
write access
5
5
3b
tLLAX_LD
Address Hold after ALE Low,
read access
5
5
ns
ns
(1)
4
tAVLLC
Address Valid C to ALE Low
57.5
0.5tCLCL-5
ns
5
tAVRL
Address Valid to RD Low
115
1.0tCLCL-10
ns
6
tAVWL
Address Valid to WR Low
115
1.0tCLCL-10
ns
7
tLLWL
ALE Low to WR Low
47.5
8
tLLRL
ALE Low to RD Low
9
tDVRH
Data Setup to RD High
10
tRLDV
Read Low to Data Valid
11
tRHDX
Data Hold After RD High
12
tRLRH
RD Pulse Width
47.5
67.5
0.5tCLCL-15(2)
0.5tCLCL+5(2)
ns
67.5
(2)
(2)
ns
40
0.5tCLCL-15
0.5tCLCL+5
40
ns
75
1.0tCLCL-50
0
0
115
1.0tCLCL-10
ns
ns
13
tDVWL
Data Setup to WR Low
42.5
14
tWHDX
Data Hold After WR High
115
1.0tCLCL-10
ns
15
tDVWH
Data Valid to WR High
125
1.0tCLCL
ns
16
tWLWH
WR Pulse Width
115
1.0tCLCL-10
ns
Notes:
0.5tCLCL-20
ns
(1)
ns
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 138. External Data Memory Characteristics, 4.5 - 5.5 Volts, 1 Cycle Wait-state
8 MHz Oscillator
Min
Max
Variable Oscillator
Symbol
Parameter
Min
Max
Unit
0
1/tCLCL
Oscillator Frequency
0.0
16
MHz
10
tRLDV
Read Low to Data Valid
12
tRLRH
RD Pulse Width
240
2.0tCLCL-10
ns
15
tDVWH
Data Valid to WR High
240
2.0tCLCL
ns
16
tWLWH
WR Pulse Width
240
2.0tCLCL-10
ns
200
2.0tCLCL-50
ns
335
2490N–AVR–05/08
Table 139. External Data Memory Characteristics, 4.5 - 5.5 Volts, SRWn1 = 1, SRWn0 = 0
4 MHz Oscillator
Min
Max
Variable Oscillator
Symbol
Parameter
Min
Max
Unit
0
1/tCLCL
Oscillator Frequency
0.0
16
MHz
10
tRLDV
Read Low to Data Valid
12
tRLRH
RD Pulse Width
365
3.0tCLCL-10
ns
15
tDVWH
Data Valid to WR High
375
3.0tCLCL
ns
16
tWLWH
WR Pulse Width
365
3.0tCLCL-10
ns
325
3.0tCLCL-50
ns
Table 140. External Data Memory Characteristics, 4.5 - 5.5 Volts, SRWn1 = 1, SRWn0 = 1
4 MHz Oscillator
Min
Max
Variable Oscillator
Symbol
Parameter
Min
Max
Unit
0
1/tCLCL
Oscillator Frequency
0.0
16
MHz
10
tRLDV
Read Low to Data Valid
12
tRLRH
RD Pulse Width
365
3.0tCLCL-10
ns
14
tWHDX
Data Hold After WR High
240
2.0tCLCL-10
ns
15
tDVWH
Data Valid to WR High
375
3.0tCLCL
ns
16
tWLWH
WR Pulse Width
365
3.0tCLCL-10
ns
325
3.0tCLCL-50
ns
Table 141. External Data Memory Characteristics, 2.7 - 5.5 Volts, No Wait-state
4 MHz Oscillator
Symbol
Parameter
0
1/tCLCL
Oscillator Frequency
1
tLHLL
ALE Pulse Width
Min
Max
235
Variable Oscillator
Min
Max
Unit
0.0
8
MHz
tCLCL-15
ns
(1)
2
tAVLL
Address Valid A to ALE Low
115
3a
tLLAX_ST
Address Hold After ALE Low,
write access
5
5
3b
tLLAX_LD
Address Hold after ALE Low,
read access
5
5
4
tAVLLC
Address Valid C to ALE Low
115
0.5tCLCL-10(1)
ns
5
tAVRL
Address Valid to RD Low
235
1.0tCLCL-15
ns
6
tAVWL
Address Valid to WR Low
235
1.0tCLCL-15
0.5tCLCL-10
ns
ns
(2)
7
tLLWL
ALE Low to WR Low
115
130
0.5tCLCL-10
8
tLLRL
ALE Low to RD Low
115
130
0.5tCLCL-10(2)
9
tDVRH
Data Setup to RD High
45
10
tRLDV
Read Low to Data Valid
11
tRHDX
Data Hold After RD High
12
tRLRH
RD Pulse Width
336
ns
ns
(2)
ns
0.5tCLCL+5(2)
ns
0.5tCLCL+5
45
190
ns
1.0tCLCL-60
ns
0
0
ns
235
1.0tCLCL-15
ns
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Table 141. External Data Memory Characteristics, 2.7 - 5.5 Volts, No Wait-state (Continued)
4 MHz Oscillator
Symbol
Parameter
Min
Max
Variable Oscillator
Min
0.5tCLCL-20
Max
(1)
Unit
13
tDVWL
Data Setup to WR Low
105
ns
14
tWHDX
Data Hold After WR High
235
1.0tCLCL-15
ns
15
tDVWH
Data Valid to WR High
250
1.0tCLCL
ns
16 tWLWH
WR Pulse Width
235
1.0tCLCL-15
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.
ns
Table 142. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 0, SRWn0 = 1
4 MHz Oscillator
Min
Max
Variable Oscillator
Symbol
Parameter
Min
Max
Unit
0
1/tCLCL
Oscillator Frequency
0.0
8
MHz
10
tRLDV
Read Low to Data Valid
12
tRLRH
RD Pulse Width
485
2.0tCLCL-15
ns
15
tDVWH
Data Valid to WR High
500
2.0tCLCL
ns
16
tWLWH
WR Pulse Width
485
2.0tCLCL-15
ns
440
2.0tCLCL-60
ns
Table 143. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 1, SRWn0 = 0
4 MHz Oscillator
Min
Max
Variable Oscillator
Symbol
Parameter
Min
Max
Unit
0
1/tCLCL
Oscillator Frequency
0.0
8
MHz
10
tRLDV
Read Low to Data Valid
12
tRLRH
RD Pulse Width
735
3.0tCLCL-15
ns
15
tDVWH
Data Valid to WR High
750
3.0tCLCL
ns
16
tWLWH
WR Pulse Width
735
3.0tCLCL-15
ns
690
3.0tCLCL-60
ns
Table 144. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 1, SRWn0 = 1
4 MHz Oscillator
Min
Max
Variable Oscillator
Symbol
Parameter
Min
Max
Unit
0
1/tCLCL
Oscillator Frequency
0.0
8
MHz
10
tRLDV
Read Low to Data Valid
12
tRLRH
RD Pulse Width
735
3.0tCLCL-15
ns
14
tWHDX
Data Hold After WR High
485
2.0tCLCL-15
ns
15
tDVWH
Data Valid to WR High
750
3.0tCLCL
ns
16
tWLWH
WR Pulse Width
735
3.0tCLCL-15
ns
690
3.0tCLCL-60
ns
337
2490N–AVR–05/08
Figure 159. External Memory Timing (SRWn1 = 0, SRWn0 = 0
T1
T2
T3
T4
System Clock (CLKCPU )
1
ALE
4
A15:8
7
Prev. addr.
Address
15
3a
DA7:0
Prev. data
Address
13
XX
Data
14
16
6
Write
2
WR
3b
DA7:0 (XMBK = 0)
11
9
Data
5
Read
Address
10
8
12
RD
Figure 160. External Memory Timing (SRWn1 = 0, SRWn0 = 1)
T1
T2
T3
T4
T5
System Clock (CLKCPU )
1
ALE
4
A15:8
7
Prev. addr.
Address
15
DA7:0
Prev. data
3a
Address
13
Data
XX
14
16
6
Write
2
WR
9
3b
Address
11
Data
5
Read
DA7:0 (XMBK = 0)
10
8
12
RD
338
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 161. External Memory Timing (SRWn1 = 1, SRWn0 = 0)
T1
T2
T3
T5
T4
T6
System Clock (CLKCPU )
1
ALE
4
A15:8
7
Address
Prev. addr.
15
DA7:0
Prev. data
3a
Address
13
XX
Data
14
16
6
Write
2
WR
9
3b
DA7:0 (XMBK = 0)
Address
11
5
Read
Data
10
8
12
RD
Figure 162. External Memory Timing (SRWn1 = 1, SRWn0 = 1)(1)
T1
T2
T3
T4
T6
T5
T7
System Clock (CLKCPU )
1
ALE
4
A15:8
7
Address
Prev. addr.
15
3a
DA7:0
Prev. data
Address
13
XX
Data
14
16
6
Write
2
WR
9
3b
Address
11
Data
5
Read
DA7:0 (XMBK = 0)
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).
339
2490N–AVR–05/08
Typical
Characteristics
The following charts show typical behavior. These figures are not tested during manufacturing.
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 switching frequency of I/O pin.
The parts are characterized at frequencies higher than test limits. Parts are not guaranteed 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 differential current drawn by the Watchdog Timer.
Active Supply Current
Figure 163. Active Supply Current vs. Frequency (0.1 - 1.0 MHz)
ACTIVE SUPPLY CURRENT vs. FREQUENCY
0.1 - 1.0 MHz
2.5
5.5 V
5.0 V
2
Icc(m A)
4.5 V
4.0 V
1.5
3.6 V
3.3 V
2.7 V
1
0.5
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Frequency (MHz)
340
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 164. Active Supply Current vs. Frequency (1 - 20 MHz)
ACTIVE SUPPLY CURRENT vs. FREQUENCY
1 - 20 MHz
45
5.5 V
40
5.0 V
35
Icc (mA)
30
25
4.5 V
20
15
4.0 V
10
3.6 V
3.0 V
5
0
0
2
4
6
8
10
12
14
16
18
20
Frequency (MHz)
Figure 165. Active Supply Current vs. VCC (Internal RC Oscillator, 1 MHz)
ACTIVE SUPPLY CURRENT vs. Vcc
INTERNAL RC OSCILLATOR. 1 MHz
2.8
85 °C
-40 °C
2.6
2.4
Icc (mA)
2.2
2
1.8
1.6
1.4
1.2
1
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
341
2490N–AVR–05/08
Figure 166. Active Supply Current vs. VCC (Internal RC Oscillator, 2 MHz)
ACTIVE SUPPLY CURRENT vs. Vcc
INTERNAL RC OSCILLATOR. 2 MHz
5.5
85 °C
25 °C
-40 °C
5
Icc (mA)
4.5
4
3.5
3
2.5
2
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Figure 167. Active Supply Current vs. VCC (Internal RC Oscillator, 4 MHz)
ACTIVE SUPPLY CURRENT vs. Vcc
INTERNAL RC OSCILLATOR. 4 MHz
11
25 °C
-40 °C
85 °C
10
Icc (mA)
9
8
7
6
5
4
3
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
342
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 168. Active Supply Current vs. VCC (Internal RC Oscillator, 8 MHz)
ACTIVE SUPPLY CURRENT vs. Vcc
INTERNAL RC OSCILLATOR. 8 MHz
-40 °C
25 °C
85 °C
20
18
Icc (mA)
16
14
12
10
8
6
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Figure 169. Active Supply Current vs. VCC (32 kHz External Oscillator)
ACTIVE SUPPLY CURRENT vs. Vcc
32kHz EXTERNAL OSCILLATOR
130
120
25 °C
110
Icc (uA)
100
90
80
70
60
50
40
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
343
2490N–AVR–05/08
Idle Supply Current
Figure 170. Idle Supply Current vs. Frequency (0.1 - 1.0 MHz)
IDLE SUPPLY CURRENT vs. FREQUENCY
0.1 - 1.0 MHz
Icc (mA)
1.4
5.5 V
1.2
5.0 V
1
4.5 V
0.8
4.0 V
3.6 V
0.6
3.3 V
0.4
2.7 V
0.2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Frequency (MHz)
Figure 171. Idle Supply Current vs. Frequency (1 - 20 MHz)
IDLE SUPPLY CURRENT vs. FREQUENCY
1 - 20 MHz
25
5.5V
5.0V
20
Icc (mA)
4.5V
15
10
4.0V
5
0
0
2
4
6
8
10
12
14
16
18
20
Frequency (MHz)
344
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 172. Idle Supply Current vs. VCC (Internal RC Oscillator, 1 MHz)
IDLE SUPPLY CURRENT vs. Vcc
INTERNAL RC OSCILLATOR. 1 MHz
1.6
85 °C
25 °C
-40 °C
1.4
Icc (mA)
1.2
1
0.8
0.6
0.4
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Figure 173. Idle Supply Current vs. VCC (Internal RC Oscillator, 2 MHz)
IDLE SUPPLY CURRENT vs. Vcc
INTERNAL RC OSCILLATOR. 2 MHz
3
85 °C
25 °C
-40 °C
Icc (mA)
2.5
2
1.5
1
0.5
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
345
2490N–AVR–05/08
Figure 174. Idle Supply Current vs. VCC (Internal RC Oscillator, 4 MHz)
IDLE SUPPLY CURRENT vs. Vcc
INTERNAL RC OSCILLATOR. 4 MHz
6
25 °C
85 °C
-40 °C
5.5
5
Icc (mA)
4.5
4
3.5
3
2.5
2
1.5
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Figure 175. Idle Supply Current vs. VCC (Internal RC Oscillator, 8 MHz)
IDLE SUPPLY CURRENT vs. Vcc
INTERNAL RC OSCILLATOR. 8 MHz
-40 °C
25 °C
85 °C
11
10
9
Icc (mA)
8
7
6
5
4
3
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
346
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 176. Idle Supply Current vs. VCC (32 kHz External Oscillator)
IDLE SUPPLY CURRENT vs. Vcc
32kHz EXTERNAL OSCILLATOR
80
75
25 °C
70
65
Icc (uA)
60
55
50
45
40
35
30
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Power-Down Supply
Current
Figure 177. Power-Down Supply Current vs. VCC (Watchdog Timer Disabled)
POWER-DOWN SUPPLY CURRENT vs. Vcc
WATCHDOG TIMER DISABLED
4
85 °C
3.5
3
Icc (uA )
2.5
2
-40 °C
1.5
25 °C
1
0.5
0
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
347
2490N–AVR–05/08
Figure 178. Power-Down Supply Current vs. VCC (Watchdog Timer Enabled)
POWER-DOWN SUPPLY CURRENT vs. Vcc
WATCHDOG TIMER ENABLED
25
85 °C
20
25 °C
-40 °C
Icc (uA)
15
10
5
0
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Power-Save Supply
Current
Figure 179. Power-Save Supply Current vs. VCC (Watchdog Timer Disabled)
POWER-SAVE SUPPLY CURRENT vs. Vcc
WATCHDOG TIMER DISABLED
14
25 °C
12
Icc (uA)
10
8
6
4
2
0
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
348
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Standby Supply
Current
Figure 180. Standby Supply Current vs. VCC
STANDBY SUPPLY CURRENT vs. Vcc
0.2
6MHz Xtal
0.18
6MHz Res
0.16
Icc (mA)
0.14
4MHz Res
4MHz Xtal
0.12
0.1
0.08
2MHz Xtal
2MHz Res
0.06
455KHz Res
1MHz Res
0.04
0.02
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Figure 181. Standby Supply Current vs. VCC (CKOPT Programmed)
STANDBY SUPPLY CURRENT vs. Vcc
CKOPT programmed
3
16MHz Xtal
2.5
12MHz Xtal
Icc (mA)
2
6MHz Xtal
1.5
4MHz Xtal
1
0.5
0
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
349
2490N–AVR–05/08
Pin Pull-up
Figure 182. I/O Pin Pull-up Resistor Current vs. Input Voltage (VCC = 5V)
I/O PIN PULL-UP RESISTOR CURRENT vs. INPUT VOLTAGE
V cc = 5V
160
140
25 °C
120
85 °C
-40 °C
IOP (uA)
100
80
60
40
20
0
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
VOP(V)
Figure 183. I/O Pin Pull-up Resistor Current vs. Input Voltage (VCC = 2.7V)
I/O PIN PULL-UP RESISTOR CURRENT vs. INPUT VOLTAGE
Vcc = 2.7V
80
-40 °C
70
25 °C
60
85 °C
IOP (uA)
50
40
30
20
10
0
0
0.5
1
1.5
2
2.5
3
VOP(V)
350
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 184. Reset Pull-up Resistor Current vs. Reset Pin Voltage (VCC = 5V)
RESET PULL-UP RESISTOR CURRENT vs. RESET PIN VOLTAGE
Vcc = 5V
120
25 °C
-40 °C
100
85 °C
IRESET (uA)
80
60
40
20
0
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
VRESET (V)
Figure 185. Reset Pull-up Resistor Current vs. Reset Pin Voltage (VCC = 2.7V)
RESET PULL-UP RESISTOR CURRENT vs. RESET PIN VOLTAGE
Vcc = 2.7V
60
-40 °C
25 °C
50
85 °C
IRESET (uA)
40
30
20
10
0
0
0.5
1
1.5
2
2.5
3
VRESET (V)
351
2490N–AVR–05/08
Figure 186. PEN Pull-up Resistor Current vs. PEN Pin Voltage (VCC = 5V)
PEN PULL-UP RESISTOR CURRENT vs. PEN PIN VOLTAGE
Vcc = 5V
140
25 °C
-40 °C
120
85 °C
IPEN (uA)
100
80
60
40
20
0
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
VPEN (V)
Figure 187. PEN Pull-up Resistor Current vs. PEN Pin Voltage (VCC = 2.7V)
PEN PULL-UP RESISTOR CURRENT vs. PEN PIN VOLTAGE
VCC = 2.7V
80
25 °C
-40 °C
70
85 °C
IPEN (uA)
60
50
40
30
20
10
0
0
0.5
1
1.5
2
2.5
3
VPEN (V)
352
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Pin Driver Strength
Figure 188. I/O Pin Source Current vs. Output Voltage (VCC = 5V)
I/O PIN SOURCE CURRENT vs. OUTPUT VOLTAGE
Vcc = 5V
80
-40 °C
70
25 °C
85 °C 60
IOH (mA)
50
40
30
20
10
0
3
3.2
3.4
3.6
3.8
4
4.2
4.4
4.6
4.8
5
VOH (V)
Figure 189. I/O Pin Source Current vs. Output Voltage (VCC = 2.7V)
I/O PIN SOURCE CURRENT vs. OUTPUT VOLTAGE
Vcc = 2.7V
30
25
-40 °C
25 °C
85 °C
IOH (mA)
20
15
10
5
0
0.5
1
1.5
2
2.5
3
VOH (V)
353
2490N–AVR–05/08
Figure 190. I/O Pin Sink Current vs. Output Voltage (VCC = 5V)
I/O PIN SINK CURRENT vs. OUTPUT VOLTAGE
Vcc = 5V
80
-40 °C
70
25 °C
60
85 °C
IOL (mA)
50
40
30
20
10
0
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
VOL (V)
Figure 191. I/O Pin Sink Current vs. Output Voltage (VCC = 2.7V)
I/O PIN SINK CURRENT vs. OUTPUT VOLTAGE
Vcc = 2.7V
30
-40 °C
25 °C
25
85 °C
IOL (mA)
20
15
10
5
0
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
VOL (V)
354
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Pin Thresholds and
Hysteresis
Figure 192. I/O Pin Input Threshold Voltage vs. VCC (VIH, I/O Pin Read as '1')
I/O PIN INPUT THRESHOLD VOLTAGE vs. Vcc
VIH. IO PIN READ AS '1'
2.4
2.2
-40 °C
85 °C
25 °C
Threshold (V)
2
1.8
1.6
1.4
1.2
1
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Figure 193. I/O Pin Input Threshold Voltage vs. VCC (VIL, I/O Pin Read as '0')
I/O PIN INPUT THRESHOLD VOLTAGE vs. Vcc
VIL. IO PIN READ AS '0'
1.5
-40 °C
25 °C
85 °C
1.4
Threshold (V)
1.3
1.2
1.1
1
0.9
0.8
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
355
2490N–AVR–05/08
Figure 194. I/O Pin Input Hysteresis vs. VCC
I/O PIN INPUT HYSTERESIS vs. Vcc
0.8
-40 °C
25 °C
85 °C
Threshold (V)
0.6
0.4
0.2
0
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Figure 195. Reset Input Threshold Voltage vs. VCC (VIH, Reset Pin Read as '1')
RESET INPUT THRESHOLD VOLTAGE vs. Vcc
VIH. IO PIN READ AS '1'
2.4
2.2
Threshold (V)
2
1.8
1.6
-40 °C
1.4
25 °C
1.2
85 °C
1
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
356
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 196. Reset Input Threshold Voltage vs. VCC (VIL, Reset Pin Read as '0')
RESET INPUT THRESHOLD VOLTAGE vs. Vcc
VIL. IO PIN READ AS '0'
85 °C
25 °C
-40 °C
2.4
2.2
Threshold (V)
2
1.8
1.6
1.4
1.2
1
2.5
3
3.5
4
4.5
5
5.5
5
5.5
V cc (V)
Figure 197. Reset Input Pin Hysteresis vs. VCC
RESET INPUT PIN HYSTERESIS vs. Vcc
0.35
-40 °C
0.3
Hysteresis (V)
0.25
0.2
25 °C
0.15
0.1
0.05
85 °C
0
2.5
3
3.5
4
4.5
Vcc (V)
357
2490N–AVR–05/08
BOD Thresholds and
Analog Comparator
Offset
Figure 198. BOD Thresholds vs. Temperature (BODLEVEL is 4.0V)
BOD THRESHOLDS vs. TEMPERATURE
BOD LEVEL IS 4.0 V
4
Rising Vcc
Threshold (V)
3.95
3.9
3.85
Falling Vcc
3.8
3.75
-40
-30
-20
-10
0
10
20
30
40
50
60
70
80
Temperature (°C)
Figure 199. BOD Thresholds vs. Temperature (BODLEVEL is 2.7V)
BOD THRESHOLDS vs. TEMPERATURE
BOD LEVEL IS 2.7 V
2.8
2.78
2.76
Rising Vcc
Thres hold (V)
2.74
2.72
2.7
2.68
2.66
2.64
Falling Vcc
2.62
2.6
-40
-20
0
20
40
60
80
Temperature (°C)
358
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 200. Bandgap Voltage vs. VCC
BANDGAP VOLTAGE vs. VCC
1.275
85 °C
1.27
Bandgap Voltage (V)
-40 °C
1.265
25 °C
1.26
1.255
1.25
2.5
3
3.5
4
4.5
5
5.5
VCC (V)
Internal Oscillator
Speed
Figure 201. Watchdog Oscillator Frequency vs. VCC
WATCHDOG OSCILLATOR FREQUENCY vs. Vcc
1060
-40 °C
25 °C
1050
1040
85 °C
1030
FRC (k Hz)
1020
1010
1000
990
980
970
960
950
2.5
3
3.5
4
4.5
5
5.5
V cc (V)
359
2490N–AVR–05/08
Figure 202. Calibrated 1 MHz RC Oscillator Frequency vs. Temperature
CALIBRATED 1MHz RC OSCILLATOR FREQUENCY vs. TEMPERATURE
1.02
1
5.5 V
5.0 V
FRC (MHz )
0.98
4.5 V
4.0 V
3.6 V
3.3 V
0.96
0.94
2.7 V
0.92
0.9
-40
-30
-20
-10
0
10
20
30
40
50
60
70
80
Temperature (°C)
Figure 203. Calibrated 1 MHz RC Oscillator Frequency vs. VCC
CALIBRATED 1MHz RC OSCILLATOR FREQUENCY vs. Vcc
1.02
-40 °C
25 °C
1
85 °C
F RC (MHz )
0.98
0.96
0.94
0.92
0.9
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
360
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 204. Calibrated 1 MHz RC Oscillator Frequency vs. Osccal Value
CALIBRATED 1MHz RC OSCILLATOR FREQUENCY vs. OSCCAL VALUE
1.5
25 °C
1.4
1.3
1.2
FRC (MHz)
1.1
1
0.9
0.8
0.7
0.6
0.5
0.4
0
16
32
48
64
80
96
112
128
144
160
176
192
208
224
240
256
OSCCAL VALUE
Figure 205. Calibrated 2 MHz RC Oscillator Frequency vs. Temperature
CALIBRATED 2MHz RC OSCILLATOR FREQUENCY vs. TEMPERATURE
2.05
2
5.5 V
5.0 V
FRC (MHz)
1.95
4.5 V
4.0 V
3.6 V
3.3 V
1.9
1.85
2.7 V
1.8
1.75
-40
-30
-20
-10
0
10
20
30
40
50
60
70
80
Temperature (°C)
361
2490N–AVR–05/08
Figure 206. Calibrated 2 MHz RC Oscillator Frequency vs. VCC
CALIBRATED 2MHz RC OSCILLATOR FREQUENCY vs. Vcc
2.05
-40 °C
25 °C
FRC (MHz)
2
85 °C
1.95
1.9
1.85
1.8
2.5
3
3.5
4
4.5
5
5.5
Vcc (V)
Figure 207. Calibrated 2 MHz RC Oscillator Frequency vs. Osccal Value
CALIBRATED 2MHz RC OSCILLATOR FREQUENCY vs. OSCCAL VALUE
25 °C
2.75
FRC (MHz)
2.25
1.75
1.25
0.75
0
16
32
48
64
80
96
112
128
144
160
176
192
208
224
240
256
OSCCAL VALUE
362
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 208. Calibrated 4 MHz RC Oscillator Frequency vs. Temperature
CALIBRATED 4MHz RC OSCILLATOR FREQUENCY vs. TEMPERATURE
4,1
4,05
4
5.5 V
3,95
F RC (M Hz)
5.0 V
3,9
4.5 V
3,85
4.0 V
3,8
3.6 V
3,75
3.3 V
3,7
2.7 V
3,65
3,6
-40
-30
-20
-10
0
10
20
30
40
50
60
70
80
Temperature (°C)
Figure 209. Calibrated 4 MHz RC Oscillator Frequency vs. VCC
CALIBRATED 4MHz RC OSCILLATOR FREQUENCY vs. Vcc
4.1
-40 °C
25 °C
4.05
4
85 °C
FRC (MHz )
3.95
3.9
3.85
3.8
3.75
3.7
3.65
3.6
2.5
3
3.5
4
4.5
5
5.5
V cc (V)
363
2490N–AVR–05/08
Figure 210. Calibrated 4 MHz RC Oscillator Frequency vs. Osccal Value
CALIBRATED 4MHz RC OSCILLATOR FREQUENCY vs. OSCCAL VALUE
6.5
25 °C
6
5.5
FRC (MHz)
5
4.5
4
3.5
3
2.5
2
1.5
0
16
32
48
64
80
96
112
128
144
160
176
192
208
224
240
256
OSCCAL VALUE
Figure 211. Calibrated 8 MHz RC Oscillator Frequency vs. Temperature
CALIBRATED 8MHz RC OSCILLATOR FREQUENCY vs. TEMPERATURE
8.3
8.1
7.9
5.5 V
5.0 V
FRC (MHz)
7.7
4.5 V
7.5
4.0 V
7.3
3.6 V
3.3 V
7.1
6.9
2.7 V
6.7
6.5
-40
-20
0
20
40
60
80
Temperature (°C)
364
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 212. Calibrated 8 MHz RC Oscillator Frequency vs. VCC
CALIBRATED 8MHz RC OSCILLATOR FREQUENCY vs. Vcc
8.4
-40 °C
8.2
25 °C
FRC (MHz)
8
85 °C
7.8
7.6
7.4
7.2
7
6.8
6.6
2.5
3
3.5
4
4.5
5
5.5
V cc (V)
Figure 213. Calibrated 8 MHz RC Oscillator Frequency vs. Osccal Value
CALIBRATED 8MHz RC OSCILLATOR FREQUENCY vs. OSCCAL VALUE
13
25 °C
12
11
FRC (MHz )
10
9
8
7
6
5
4
3
0
16
32
48
64
80
96
112
128
144
160
176
192
208
224
240
256
OSCCAL VALUE
365
2490N–AVR–05/08
Current Consumption
of Peripheral Units
Figure 214. Brownout Detector Current vs. VCC
BROWNOUT DETECTOR CURRENT vs. Vcc
16
14
-40 °C
12
25 °C
Icc (uA)
10
8
6
85 °C
4
2
0
2.5
3
3.5
4
4.5
5
5.5
V cc (V)
Figure 215. ADC Current vs. VCC (ADC CLK = 50 kHz)
ADC CURRENT vs. Vcc
ADC CLK = 50KHz
330
-40 °C
310
25 °C
290
Icc (uA)
270
85 °C
250
230
210
190
170
150
130
2.5
3
3.5
4
4.5
5
5.5
V cc (V)
366
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 216. Aref Current vs. VCC
AREF CURRENT vs. Vcc
170
85 °C
25 °C
-40 °C
160
150
Icc (uA)
140
130
120
110
100
90
80
70
2.5
3
3.5
4
4.5
5
5.5
V cc (V)
Figure 217. Analog Comparator Current vs. VCC
ANALOG COMPARATOR CURRENT vs. Vcc
90
85 °C
80
Icc (uA)
70
25 °C
60
-40 °C
50
40
30
2,5
3
3,5
4
4,5
5
5,5
V cc (V)
367
2490N–AVR–05/08
Figure 218. Programming Current vs. VCC
PROGRAMMING CURRENT vs. Vcc
9
-40 °C
Icc (mA)
8
7
25 °C
6
85 °C
5
4
3
2
1
0
2.5
3
3.5
4
4.5
5
5.5
V cc (V)
Current Consumption
in Reset and Reset
Pulse width
Figure 219. Reset Supply Current vs. VCC (0.1 - 1.0 MHz, Excluding Current through the Reset
Pull-up)
RESET SUPPLY CURRENT vs. Vcc
0.1 - 1.0 MHz. EXCLUDING CURRENT THROUGH THE RESET PULLUP
3.5
5.5 V
3
5.0 V
ICC (m A)
2.5
4.5 V
2
4.0 V
1.5
3.6 V
3.3 V
2.7 V
1
0.5
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Frequency (MHz)
368
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Figure 220. Reset Supply Current vs. VCC (1 - 20 MHz, Excluding Current through the Reset
Pull-up)
RESET SUPPLY CURRENT vs. Vcc
1- 20MHz. EXCLUDING CURRENT THROUGH THE RESET PULLUP
Icc (mA)
40
35
5.5 V
30
5.0 V
25
4.5 V
4.0 V
20
3.6 V
3.3 V
15
2.7 V
10
5
0
0
2
4
6
8
10
12
14
16
18
20
Frequency (MHz)
Figure 221. Minimum Reset Pulse Width vs. VCC
MINIMUM RESET PULSE WIDTH vs. Vcc
1200
Pulsewidth (ns)
1000
800
600
85 °C
25 °C
400
-40 °C
200
0
2.5
3
3.5
4
4.5
5
5.5
V cc (V)
369
2490N–AVR–05/08
Register Summary
Address
Name
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
(0xFF)
Reserved
–
–
–
–
–
–
–
–
..
(0x9E)
Reserved
–
–
–
–
–
–
–
–
Reserved
–
–
–
–
–
–
–
–
(0x9D)
UCSR1C
–
UMSEL1
UPM11
UPM10
USBS1
UCSZ11
UCSZ10
UCPOL1
191
(0x9C)
UDR1
(0x9B)
UCSR1A
RXC1
TXC1
UDRE1
(0x9A)
UCSR1B
RXCIE1
TXCIE1
UDRIE1
(0x99)
UBRR1L
(0x98)
(0x97)
UBRR1H
–
–
–
–
Reserved
–
–
–
–
–
–
–
(0x96)
Reserved
–
–
–
–
–
–
–
–
(0x95)
(0x94)
UCSR0C
–
UMSEL0
UPM01
UPM00
USBS0
UCSZ01
UCSZ00
UCPOL0
Reserved
–
–
–
–
–
–
–
–
(0x93)
Reserved
–
–
–
–
–
–
–
–
(0x92)
Reserved
–
–
–
–
–
–
–
–
(0x91)
Reserved
–
–
–
–
–
–
–
–
(0x90)
(0x8F)
UBRR0H
–
–
–
–
Reserved
–
–
–
–
–
–
–
–
370
USART1 I/O Data Register
Page
188
FE1
DOR1
UPE1
U2X1
MPCM1
189
RXEN1
TXEN1
UCSZ12
RXB81
TXB81
190
USART1 Baud Rate Register Low
193
USART1 Baud Rate Register High
193
–
USART0 Baud Rate Register High
191
193
(0x8E)
ADCSRB
–
–
–
–
–
ADTS2
ADTS1
ADTS0
(0x8D)
Reserved
–
–
–
–
–
–
–
–
(0x8C)
TCCR3C
FOC3A
FOC3B
FOC3C
–
–
–
–
–
138
(0x8B)
TCCR3A
COM3A1
COM3A0
COM3B1
COM3B0
COM3C1
COM3C0
WGM31
WGM30
132
ICNC3
ICES3
–
WGM33
WGM32
CS32
CS31
CS30
136
247
(0x8A)
TCCR3B
(0x89)
TCNT3H
Timer/Counter3 – Counter Register High Byte
(0x88)
TCNT3L
Timer/Counter3 – Counter Register Low Byte
138
(0x87)
OCR3AH
Timer/Counter3 – Output Compare Register A High Byte
139
138
(0x86)
OCR3AL
Timer/Counter3 – Output Compare Register A Low Byte
139
(0x85)
OCR3BH
Timer/Counter3 – Output Compare Register B High Byte
139
(0x84)
OCR3BL
Timer/Counter3 – Output Compare Register B Low Byte
139
(0x83)
OCR3CH
Timer/Counter3 – Output Compare Register C High Byte
139
(0x82)
OCR3CL
Timer/Counter3 – Output Compare Register C Low Byte
139
(0x81)
ICR3H
Timer/Counter3 – Input Capture Register High Byte
140
(0x80)
(0x7F)
ICR3L
Timer/Counter3 – Input Capture Register Low Byte
Reserved
–
–
–
–
–
–
140
–
–
(0x7E)
Reserved
–
–
–
–
–
–
–
–
(0x7D)
ETIMSK
–
–
TICIE3
OCIE3A
OCIE3B
TOIE3
OCIE3C
OCIE1C
141
(0x7C)
(0x7B)
ETIFR
–
–
ICF3
OCF3A
OCF3B
TOV3
OCF3C
OCF1C
142
Reserved
–
–
–
–
–
–
–
–
(0x7A)
TCCR1C
FOC1A
FOC1B
FOC1C
–
–
–
–
–
(0x79)
OCR1CH
Timer/Counter1 – Output Compare Register C High Byte
139
(0x78)
(0x77)
OCR1CL
Timer/Counter1 – Output Compare Register C Low Byte
139
Reserved
–
–
–
–
–
–
–
–
(0x76)
Reserved
–
–
–
–
–
–
–
–
(0x75)
Reserved
–
–
–
–
–
–
–
–
(0x74)
TWCR
TWINT
TWEA
TWSTA
TWSTO
TWWC
TWEN
–
TWIE
(0x73)
TWDR
(0x72)
TWAR
TWA6
TWA5
TWA4
TWA3
TWA2
TWA1
TWA0
TWGCE
208
(0x71)
TWSR
TWS7
TWS6
TWS5
TWS4
TWS3
–
TWPS1
TWPS0
207
Two-wire Serial Interface Data Register
206
208
(0x70)
TWBR
Two-wire Serial Interface Bit Rate Register
(0x6F)
(0x6E)
OSCCAL
Oscillator Calibration Register
Reserved
137
206
43
–
–
–
–
–
–
–
(0x6D)
XMCRA
–
SRL2
SRL1
SRL0
SRW01
SRW00
SRW11
(0x6C)
XMCRB
XMBK
–
–
–
–
XMM2
XMM1
–
32
XMM0
34
(0x6B)
Reserved
–
–
–
–
–
–
–
–
(0x6A)
(0x69)
EICRA
ISC31
ISC30
ISC21
ISC20
ISC11
ISC10
ISC01
ISC00
Reserved
–
–
–
–
–
–
–
–
(0x68)
SPMCSR
SPMIE
RWWSB
–
RWWSRE
BLBSET
PGWRT
PGERS
SPMEN
(0x67)
Reserved
–
–
–
–
–
–
–
–
(0x66)
Reserved
–
–
–
–
–
–
–
–
(0x65)
PORTG
–
–
–
PORTG4
PORTG3
PORTG2
PORTG1
PORTG0
89
(0x64)
DDRG
–
–
–
DDG4
DDG3
DDG2
DDG1
DDG0
89
90
281
(0x63)
PING
–
–
–
PING4
PING3
PING2
PING1
PING0
89
(0x62)
PORTF
PORTF7
PORTF6
PORTF5
PORTF4
PORTF3
PORTF2
PORTF1
PORTF0
88
(0x61)
DDRF
DDF7
DDF6
DDF5
DDF4
DDF3
DDF2
DDF1
DDF0
89
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Register Summary (Continued)
Address
Name
Bit 7
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0
(0x60)
Reserved
–
–
–
–
–
–
–
–
Page
0x3F (0x5F)
SREG
I
T
H
S
V
N
Z
C
12
0x3E (0x5E)
SPH
SP15
SP14
SP13
SP12
SP11
SP10
SP9
SP8
14
0x3D (0x5D)
SPL
SP7
SP6
SP5
SP4
SP3
SP2
SP1
SP0
14
0x3C (0x5C)
XDIV
XDIVEN
XDIV6
XDIV5
XDIV4
XDIV3
XDIV2
XDIV1
XDIV0
39
0x3B (0x5B)
Reserved
–
–
–
–
–
–
–
–
0x3A (0x5A)
EICRB
ISC71
ISC70
ISC61
ISC60
ISC51
ISC50
ISC41
ISC40
91
0x39 (0x59)
EIMSK
INT7
INT6
INT5
INT4
INT3
INT2
INT1
INT0
92
0x38 (0x58)
EIFR
INTF7
INTF6
INTF5
INTF4
INTF3
INTF
INTF1
INTF0
92
0x37 (0x57)
TIMSK
OCIE2
TOIE2
TICIE1
OCIE1A
OCIE1B
TOIE1
OCIE0
TOIE0
109, 140, 160
0x36 (0x56)
TIFR
OCF2
TOV2
ICF1
OCF1A
OCF1B
TOV1
OCF0
TOV0
109, 142, 160
0x35 (0x55)
MCUCR
SRE
SRW10
SE
SM1
SM0
SM2
IVSEL
IVCE
32, 46, 64
0x34 (0x54)
MCUCSR
JTD
–
–
JTRF
WDRF
BORF
EXTRF
PORF
55, 256
0x33 (0x53)
TCCR0
FOC0
WGM00
COM01
COM00
WGM01
CS02
CS01
CS00
0x32 (0x52)
TCNT0
0x31 (0x51)
OCR0
0x30 (0x50)
ASSR
–
–
–
–
AS0
TCN0UB
OCR0UB
TCR0UB
107
Timer/Counter0 (8 Bit)
104
106
Timer/Counter0 Output Compare Register
106
0x2F (0x4F)
TCCR1A
COM1A1
COM1A0
COM1B1
COM1B0
COM1C1
COM1C0
WGM11
WGM10
132
0x2E (0x4E)
TCCR1B
ICNC1
ICES1
–
WGM13
WGM12
CS12
CS11
CS10
136
0x2D (0x4D)
TCNT1H
Timer/Counter1 – Counter Register High Byte
138
0x2C (0x4C)
TCNT1L
Timer/Counter1 – Counter Register Low Byte
138
0x2B (0x4B)
OCR1AH
Timer/Counter1 – Output Compare Register A High Byte
139
0x2A (0x4A)
OCR1AL
Timer/Counter1 – Output Compare Register A Low Byte
139
0x29 (0x49)
OCR1BH
Timer/Counter1 – Output Compare Register B High Byte
139
0x28 (0x48)
OCR1BL
Timer/Counter1 – Output Compare Register B Low Byte
139
0x27 (0x47)
ICR1H
Timer/Counter1 – Input Capture Register High Byte
140
0x26 (0x46)
ICR1L
0x25 (0x45)
TCCR2
Timer/Counter1 – Input Capture Register Low Byte
0x24 (0x44)
TCNT2
Timer/Counter2 (8 Bit)
0x23 (0x43)
OCR2
Timer/Counter2 Output Compare Register
0x22 (0x42)
OCDR
0x21 (0x41)
WDTCR
IDRD/
OCDR7
–
0x20 (0x40)
SFIOR
TSM
0x1F (0x3F)
EEARH
–
0x1E (0x3E)
EEARL
FOC2
WGM20
COM21
COM20
OCDR6
OCDR5
OCDR4
–
–
–
–
–
–
WGM21
CS22
140
CS21
CS20
157
159
160
OCDR3
OCDR2
OCDR1
OCDR0
WDCE
WDE
WDP2
WDP1
WDP0
57
–
ACME
PUD
PSR0
PSR321
72, 111, 145, 227
–
–
EEPROM Address Register High Byte
EEPROM Address Register Low Byte
253
22
22
0x1D (0x3D)
EEDR
0x1C (0x3C)
EECR
–
–
–
EEPROM Data Register
–
EERIE
EEMWE
EEWE
EERE
22
0x1B (0x3B)
PORTA
PORTA7
PORTA6
PORTA5
PORTA4
PORTA3
PORTA2
PORTA1
PORTA0
87
0x1A (0x3A)
DDRA
DDA7
DDA6
DDA5
DDA4
DDA3
DDA2
DDA1
DDA0
87
22
0x19 (0x39)
PINA
PINA7
PINA6
PINA5
PINA4
PINA3
PINA2
PINA1
PINA0
87
0x18 (0x38)
PORTB
PORTB7
PORTB6
PORTB5
PORTB4
PORTB3
PORTB2
PORTB1
PORTB0
87
0x17 (0x37)
DDRB
DDB7
DDB6
DDB5
DDB4
DDB3
DDB2
DDB1
DDB0
87
0x16 (0x36)
PINB
PINB7
PINB6
PINB5
PINB4
PINB3
PINB2
PINB1
PINB0
87
0x15 (0x35)
PORTC
PORTC7
PORTC6
PORTC5
PORTC4
PORTC3
PORTC2
PORTC1
PORTC0
87
0x14 (0x34)
DDRC
DDC7
DDC6
DDC5
DDC4
DDC3
DDC2
DDC1
DDC0
87
0x13 (0x33)
PINC
PINC7
PINC6
PINC5
PINC4
PINC3
PINC2
PINC1
PINC0
88
0x12 (0x32)
PORTD
PORTD7
PORTD6
PORTD5
PORTD4
PORTD3
PORTD2
PORTD1
PORTD0
88
0x11 (0x31)
DDRD
DDD7
DDD6
DDD5
DDD4
DDD3
DDD2
DDD1
DDD0
88
0x10 (0x30)
PIND
PIND7
PIND6
PIND5
PIND4
PIND3
PIND2
PIND1
PIND0
0x0F (0x2F)
SPDR
SPI Data Register
88
169
0x0E (0x2E)
SPSR
SPIF
WCOL
–
–
–
–
–
SPI2X
0x0D (0x2D)
SPCR
SPIE
SPE
DORD
MSTR
CPOL
CPHA
SPR1
SPR0
0x0C (0x2C)
UDR0
0x0B (0x2B)
UCSR0A
RXC0
TXC0
UDRE0
FE0
DOR0
UPE0
U2X0
MPCM0
189
0x0A (0x2A)
UCSR0B
RXCIE0
TXCIE0
UDRIE0
RXEN0
TXEN0
UCSZ02
RXB80
TXB80
190
0x09 (0x29)
UBRR0L
0x08 (0x28)
ACSR
ACD
ACBG
ACIC
ACIS1
ACIS0
228
0x07 (0x27)
ADMUX
REFS1
0x06 (0x26)
ADCSRA
ADEN
0x05 (0x25)
ADCH
ADC Data Register High Byte
246
0x04 (0x24)
ADCL
ADC Data Register Low byte
246
0x03 (0x23)
PORTE
PORTE7
PORTE6
PORTE5
PORTE4
PORTE3
PORTE2
PORTE1
PORTE0
88
0x02 (0x22)
DDRE
DDE7
DDE6
DDE5
DDE4
DDE3
DDE2
DDE1
DDE0
88
0x01 (0x21)
PINE
PINE7
PINE6
PINE5
PINE4
PINE3
PINE2
PINE1
PINE0
88
USART0 I/O Data Register
169
167
188
USART0 Baud Rate Register Low
193
ACO
ACI
ACIE
REFS0
ADLAR
MUX4
MUX3
MUX2
MUX1
MUX0
243
ADSC
ADATE
ADIF
ADIE
ADPS2
ADPS1
ADPS0
245
371
2490N–AVR–05/08
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
89
Notes:
372
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.
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Instruction Set Summary
Mnemonics
Operands
Description
Operation
Flags
#Clocks
ARITHMETIC AND LOGIC INSTRUCTIONS
ADD
Rd, Rr
Add two Registers
Rd ← Rd + Rr
Z,C,N,V,H
ADC
Rd, Rr
Add with Carry two Registers
Rd ← Rd + Rr + C
Z,C,N,V,H
1
ADIW
Rdl,K
Add Immediate to Word
Rdh:Rdl ← Rdh:Rdl + K
Z,C,N,V,S
2
SUB
Rd, Rr
Subtract two Registers
Rd ← Rd - Rr
Z,C,N,V,H
1
SUBI
Rd, K
Subtract Constant from Register
Rd ← Rd - K
Z,C,N,V,H
1
SBC
Rd, Rr
Subtract with Carry two Registers
Rd ← Rd - Rr - C
Z,C,N,V,H
1
1
SBCI
Rd, K
Subtract with Carry Constant from Reg.
Rd ← Rd - K - C
Z,C,N,V,H
1
SBIW
Rdl,K
Subtract Immediate from Word
Rdh:Rdl ← Rdh:Rdl - K
Z,C,N,V,S
2
1
AND
Rd, Rr
Logical AND Registers
Rd ← Rd • Rr
Z,N,V
ANDI
Rd, K
Logical AND Register and Constant
Rd ← Rd • K
Z,N,V
1
OR
Rd, Rr
Logical OR Registers
Rd ← Rd v Rr
Z,N,V
1
ORI
Rd, K
Logical OR Register and Constant
Rd ← Rd v K
Z,N,V
1
EOR
Rd, Rr
Exclusive OR Registers
Rd ← Rd ⊕ Rr
Z,N,V
1
COM
Rd
One’s Complement
Rd ← 0xFF − Rd
Z,C,N,V
1
NEG
Rd
Two’s Complement
Rd ← 0x00 − Rd
Z,C,N,V,H
1
SBR
Rd,K
Set Bit(s) in Register
Rd ← Rd v K
Z,N,V
1
CBR
Rd,K
Clear Bit(s) in Register
Rd ← Rd • (0xFF - K)
Z,N,V
1
INC
Rd
Increment
Rd ← Rd + 1
Z,N,V
1
DEC
Rd
Decrement
Rd ← Rd − 1
Z,N,V
1
TST
Rd
Test for Zero or Minus
Rd ← Rd • Rd
Z,N,V
1
CLR
Rd
Clear Register
Rd ← Rd ⊕ Rd
Z,N,V
1
SER
Rd
Set Register
Rd ← 0xFF
None
1
MUL
Rd, Rr
Multiply Unsigned
R1:R0 ← Rd x Rr
Z,C
2
MULS
Rd, Rr
Multiply Signed
R1:R0 ← Rd x Rr
Z,C
2
MULSU
Rd, Rr
Multiply Signed with Unsigned
R1:R0 ← Rd x Rr
Z,C
2
FMUL
Rd, Rr
Fractional Multiply Unsigned
R1:R0 ¨ (Rd x Rr) << 1
Z,C
2
FMULS
Rd, Rr
Fractional Multiply Signed
R1:R0 ¨ (Rd x Rr) << 1
Z,C
2
FMULSU
Rd, Rr
Fractional Multiply Signed with Unsigned
R1:R0 ¨ (Rd x Rr) << 1
Z,C
2
Relative Jump
PC ← PC + k + 1
None
2
Indirect Jump to (Z)
PC ← Z
None
2
3
BRANCH INSTRUCTIONS
RJMP
k
IJMP
JMP
k
Direct Jump
PC ← k
None
RCALL
k
Relative Subroutine Call
PC ← PC + k + 1
None
3
Indirect Call to (Z)
PC ← Z
None
3
ICALL
Direct Subroutine Call
PC ← k
None
4
RET
Subroutine Return
PC ← STACK
None
4
RETI
Interrupt Return
PC ← STACK
I
if (Rd = Rr) PC ← PC + 2 or 3
None
CALL
k
4
CPSE
Rd,Rr
Compare, Skip if Equal
1/2/3
CP
Rd,Rr
Compare
Rd − Rr
Z, N,V,C,H
1
CPC
Rd,Rr
Compare with Carry
Rd − Rr − C
Z, N,V,C,H
1
CPI
Rd,K
Compare Register with Immediate
Rd − K
Z, N,V,C,H
SBRC
Rr, b
Skip if Bit in Register Cleared
if (Rr(b)=0) PC ← PC + 2 or 3
None
SBRS
Rr, b
Skip if Bit in Register is Set
if (Rr(b)=1) PC ← PC + 2 or 3
None
1/2/3
SBIC
P, b
Skip if Bit in I/O Register Cleared
if (P(b)=0) PC ← PC + 2 or 3
None
1/2/3
SBIS
P, b
Skip if Bit in I/O Register is Set
if (P(b)=1) PC ← PC + 2 or 3
None
1/2/3
BRBS
s, k
Branch if Status Flag Set
if (SREG(s) = 1) then PC←PC+k + 1
None
1/2
BRBC
s, k
Branch if Status Flag Cleared
if (SREG(s) = 0) then PC←PC+k + 1
None
1/2
BREQ
k
Branch if Equal
if (Z = 1) then PC ← PC + k + 1
None
1/2
BRNE
k
Branch if Not Equal
if (Z = 0) then PC ← PC + k + 1
None
1/2
BRCS
k
Branch if Carry Set
if (C = 1) then PC ← PC + k + 1
None
1/2
BRCC
k
Branch if Carry Cleared
if (C = 0) then PC ← PC + k + 1
None
1/2
BRSH
k
Branch if Same or Higher
if (C = 0) then PC ← PC + k + 1
None
1/2
BRLO
k
Branch if Lower
if (C = 1) then PC ← PC + k + 1
None
1/2
BRMI
k
Branch if Minus
if (N = 1) then PC ← PC + k + 1
None
1/2
BRPL
k
Branch if Plus
if (N = 0) then PC ← PC + k + 1
None
1/2
BRGE
k
Branch if Greater or Equal, Signed
if (N ⊕ V= 0) then PC ← PC + k + 1
None
1/2
1
1/2/3
BRLT
k
Branch if Less Than Zero, Signed
if (N ⊕ V= 1) then PC ← PC + k + 1
None
1/2
BRHS
k
Branch if Half Carry Flag Set
if (H = 1) then PC ← PC + k + 1
None
1/2
BRHC
k
Branch if Half Carry Flag Cleared
if (H = 0) then PC ← PC + k + 1
None
1/2
BRTS
k
Branch if T Flag Set
if (T = 1) then PC ← PC + k + 1
None
1/2
BRTC
k
Branch if T Flag Cleared
if (T = 0) then PC ← PC + k + 1
None
1/2
BRVS
k
Branch if Overflow Flag is Set
if (V = 1) then PC ← PC + k + 1
None
1/2
BRVC
k
Branch if Overflow Flag is Cleared
if (V = 0) then PC ← PC + k + 1
None
1/2
373
2490N–AVR–05/08
Instruction Set Summary (Continued)
BRIE
k
Branch if Interrupt Enabled
if ( I = 1) then PC ← PC + k + 1
None
1/2
BRID
k
Branch if Interrupt Disabled
if ( I = 0) then PC ← PC + k + 1
None
1/2
None
1
None
1
DATA TRANSFER INSTRUCTIONS
MOV
Rd, Rr
Move Between Registers
MOVW
Rd, Rr
Copy Register Word
Rd ← Rr
Rd+1:Rd ← Rr+1:Rr
LDI
Rd, K
Load Immediate
Rd ← K
None
1
LD
Rd, X
Load Indirect
Rd ← (X)
None
2
LD
Rd, X+
Load Indirect and Post-Inc.
Rd ← (X), X ← X + 1
None
2
LD
Rd, - X
Load Indirect and Pre-Dec.
X ← X - 1, Rd ← (X)
None
2
2
LD
Rd, Y
Load Indirect
Rd ← (Y)
None
LD
Rd, Y+
Load Indirect and Post-Inc.
Rd ← (Y), Y ← Y + 1
None
2
LD
Rd, - Y
Load Indirect and Pre-Dec.
Y ← Y - 1, Rd ← (Y)
None
2
LDD
Rd,Y+q
Load Indirect with Displacement
Rd ← (Y + q)
None
2
LD
Rd, Z
Load Indirect
Rd ← (Z)
None
2
LD
Rd, Z+
Load Indirect and Post-Inc.
Rd ← (Z), Z ← Z+1
None
2
LD
Rd, -Z
Load Indirect and Pre-Dec.
Z ← Z - 1, Rd ← (Z)
None
2
LDD
Rd, Z+q
Load Indirect with Displacement
Rd ← (Z + q)
None
2
2
LDS
Rd, k
Load Direct from SRAM
Rd ← (k)
None
ST
X, Rr
Store Indirect
(X) ← Rr
None
2
ST
X+, Rr
Store Indirect and Post-Inc.
(X) ← Rr, X ← X + 1
None
2
ST
- X, Rr
Store Indirect and Pre-Dec.
X ← X - 1, (X) ← Rr
None
2
ST
Y, Rr
Store Indirect
(Y) ← Rr
None
2
ST
Y+, Rr
Store Indirect and Post-Inc.
(Y) ← Rr, Y ← Y + 1
None
2
ST
- Y, Rr
Store Indirect and Pre-Dec.
Y ← Y - 1, (Y) ← Rr
None
2
STD
Y+q,Rr
Store Indirect with Displacement
(Y + q) ← Rr
None
2
ST
Z, Rr
Store Indirect
(Z) ← Rr
None
2
ST
Z+, Rr
Store Indirect and Post-Inc.
(Z) ← Rr, Z ← Z + 1
None
2
ST
-Z, Rr
Store Indirect and Pre-Dec.
Z ← Z - 1, (Z) ← Rr
None
2
STD
Z+q,Rr
Store Indirect with Displacement
(Z + q) ← Rr
None
2
STS
k, Rr
Store Direct to SRAM
(k) ← Rr
None
2
Load Program Memory
R0 ← (Z)
None
3
LPM
LPM
Rd, Z
Load Program Memory
Rd ← (Z)
None
3
LPM
Rd, Z+
Load Program Memory and Post-Inc
Rd ← (Z), Z ← Z+1
None
3
Store Program Memory
(Z) ← R1:R0
None
-
In Port
Rd ← P
None
1
SPM
IN
Rd, P
OUT
P, Rr
Out Port
P ← Rr
None
1
PUSH
Rr
Push Register on Stack
STACK ← Rr
None
2
POP
Rd
Pop Register from Stack
Rd ← STACK
None
2
BIT AND BIT-TEST INSTRUCTIONS
SBI
P,b
Set Bit in I/O Register
I/O(P,b) ← 1
None
2
CBI
P,b
Clear Bit in I/O Register
I/O(P,b) ← 0
None
2
LSL
Rd
Logical Shift Left
Rd(n+1) ← Rd(n), Rd(0) ← 0
Z,C,N,V
1
LSR
Rd
Logical Shift Right
Rd(n) ← Rd(n+1), Rd(7) ← 0
Z,C,N,V
1
ROL
Rd
Rotate Left Through Carry
Rd(0)←C,Rd(n+1)← Rd(n),C←Rd(7)
Z,C,N,V
1
ROR
Rd
Rotate Right Through Carry
Rd(7)←C,Rd(n)← Rd(n+1),C←Rd(0)
Z,C,N,V
1
ASR
Rd
Arithmetic Shift Right
Rd(n) ← Rd(n+1), n=0..6
Z,C,N,V
1
SWAP
Rd
Swap Nibbles
Rd(3..0)←Rd(7..4),Rd(7..4)←Rd(3..0)
None
1
BSET
s
Flag Set
SREG(s) ← 1
SREG(s)
1
BCLR
s
Flag Clear
SREG(s) ← 0
SREG(s)
1
BST
Rr, b
Bit Store from Register to T
T ← Rr(b)
T
1
BLD
Rd, b
Bit load from T to Register
Rd(b) ← T
None
1
1
SEC
Set Carry
C←1
C
CLC
Clear Carry
C←0
C
1
SEN
Set Negative Flag
N←1
N
1
CLN
Clear Negative Flag
N←0
N
1
SEZ
Set Zero Flag
Z←1
Z
1
CLZ
Clear Zero Flag
Z←0
Z
1
SEI
Global Interrupt Enable
I←1
I
1
CLI
Global Interrupt Disable
I←0
I
1
1
SES
Set Signed Test Flag
S←1
S
CLS
Clear Signed Test Flag
S←0
S
1
SEV
CLV
Set Twos Complement Overflow.
Clear Twos Complement Overflow
V←1
V←0
V
V
1
1
SET
Set T in SREG
T←1
T
1
CLT
Clear T in SREG
T←0
T
1
SEH
Set Half Carry Flag in SREG
H←1
H
1
374
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Instruction Set Summary (Continued)
CLH
Clear Half Carry Flag in SREG
H←0
H
1
MCU CONTROL INSTRUCTIONS
NOP
No Operation
None
1
SLEEP
Sleep
(see specific descr. for Sleep function)
None
1
WDR
BREAK
Watchdog Reset
Break
(see specific descr. for WDR/timer)
For On-chip Debug Only
None
None
1
N/A
375
2490N–AVR–05/08
Ordering Information
Speed (MHz)
8
16
Note:
Power Supply
Ordering Code(2)
Package(1)
2.7 - 5.5
ATmega64L-8AU
ATmega64L-8MU
64A
64M1
4.5 - 5.5
ATmega64-16AU
ATmega64-16MU
64A
64M1
Operation Range
Industrial
(-40°C to 85°C)
1. This device can also be supplied in wafer form. Please contact your local Atmel sales office for detailed ordering information
and minimum quantities.
2. Pb-free packaging complies to the European Directive for Restriction of Hazardous Substances (RoHS directive). Also
Halide free and fully Green.
Package Type
64A
64-lead, Thin (1.0 mm) Plastic Gull Wing Quad Flat Package (TQFP)
64M1
64-pad, 9 x 9 x 1.0 mm body, lead pitch 0.50 mm, Quad Flat No-Lead/Micro Lead Frame Package (QFN/MLF)
376
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Packaging Information
64A
PIN 1
B
PIN 1 IDENTIFIER
E1
e
E
D1
D
C
0°~7°
A1
A2
A
L
COMMON DIMENSIONS
(Unit of Measure = mm)
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.
3. Lead coplanarity is 0.10 mm maximum.
SYMBOL
MIN
NOM
MAX
A
–
–
1.20
A1
0.05
–
0.15
A2
0.95
1.00
1.05
D
15.75
16.00
16.25
D1
13.90
14.00
14.10
E
15.75
16.00
16.25
E1
13.90
14.00
14.10
B
0.30
–
0.45
C
0.09
–
0.20
L
0.45
–
0.75
e
NOTE
Note 2
Note 2
0.80 TYP
10/5/2001
R
2325 Orchard Parkway
San Jose, CA 95131
TITLE
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)
DRAWING NO.
REV.
64A
B
377
2490N–AVR–05/08
64M1
D
Marked Pin# 1 ID
E
C
SEATING PLANE
A1
TOP VIEW
A
K
0.08 C
L
Pin #1 Corner
D2
1
2
3
Option A
SIDE VIEW
Pin #1
Triangle
COMMON DIMENSIONS
(Unit of Measure = mm)
E2
Option B
K
Option C
b
e
Pin #1
Chamfer
(C 0.30)
Pin #1
Notch
(0.20 R)
BOTTOM VIEW
Note: 1. JEDEC Standard MO-220, (SAW Singulation) Fig. 1, VMMD.
2. Dimension and tolerance conform to ASMEY14.5M-1994.
SYMBOL
MIN
NOM
MAX
A
0.80
0.90
1.00
0.05
A1
–
0.02
b
0.18
0.25
0.30
D
8.90
9.00
9.10
D2
5.20
5.40
5.60
E
8.90
9.00
9.10
E2
5.20
5.40
5.60
e
NOTE
0.50 BSC
L
0.35
0.40
0.45
K
1.25
1.40
1.55
5/25/06
R
378
2325 Orchard Parkway
San Jose, CA 95131
TITLE
64M1, 64-pad, 9 x 9 x 1.0 mm Body, Lead Pitch 0.50 mm,
5.40 mm Exposed Pad, Micro Lead Frame Package (MLF)
DRAWING NO.
64M1
REV.
G
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Errata
The revision letter in this section refers to the revision of the ATmega64 device.
ATmega64, rev. A
to C
•
•
•
•
•
•
First Analog Comparator conversion may be delayed
Interrupts may be lost when writing the timer registers in the asynchronous timer
Stabilizing time needed when changing XDIV Register
Stabilizing time needed when changing OSCCAL Register
IDCODE masks data from TDI input
Reading EEPROM by using ST or STS to set EERE bit triggers unexpected interrupt request
1. First Analog Comparator conversion may be delayed
If the device is powered by a slow rising VCC, the first Analog Comparator conversion will
take longer than expected on some devices.
Problem Fix/Workaround
When the device has been powered or reset, disable then enable theAnalog Comparator
before the first conversion.
2. Interrupts may be lost when writing the timer registers in the asynchronous timer
If one of the timer registers which is synchronized to the asynchronous timer2 clock is written in the cycle before a overflow interrupt occurs, the interrupt may be lost.
Problem Fix/Workaround
Always check that the Timer2 Timer/Counter register, TCNT2, does not have the value 0xFF
before writing the Timer2 Control Register, TCCR2, or Output Compare Register, OCR2
3. 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
OUT
; clear global interrupt enable
XDIV, temp
; set new prescale value
NOP
; no operation
NOP
; no operation
NOP
; no operation
NOP
; no operation
NOP
; no operation
NOP
; no operation
NOP
; no operation
NOP
; no operation
SEI
; clear global interrupt enable
379
2490N–AVR–05/08
4. Stabilizing time needed when changing OSCCAL Register
After increasing the source clock frequency more than 2% with settings in the OSCCAL register, the device may execute some of the subsequent instructions incorrectly.
Problem Fix / Workaround
The behavior follows errata number 3., and the same Fix / Workaround is applicable on this
errata.
5. IDCODE masks data from TDI input
The JTAG instruction IDCODE is not working correctly. Data to succeeding devices are
replaced by all-ones during Update-DR.
Problem Fix / Workaround
–
If ATmega64 is the only device in the scan chain, the problem is not visible.
–
Select the Device ID Register of the ATmega64 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. Issue the BYPASS instruction to the ATmega64 while reading the Device ID
Registers of preceding devices of the boundary scan chain.
–
If the Device IDs of all devices in the boundary scan chain must be captured
simultaneously, the ATmega64 must be the first device in the chain.
6. Reading EEPROM by using ST or STS to set EERE bit triggers unexpected interrupt
request.
Reading EEPROM by using the ST or STS command to set the EERE bit in the EECR register triggers an unexpected EEPROM interrupt request.
Problem Fix / Workaround
Always use OUT or SBI to set EERE in EECR.
380
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Datasheet
Revision
History
Please note that the referring page numbers in this section are referred to this document. The
referring revision in this section are referring to the document revision.
Changes from Rev. 1. Updated “PEN” on page 7.
2490M-08/07 to
2. Updated “Ordering Information” on page 376.
Rev. 2490N-05/08
Changes from Rev. 1. Updated “Features” on page 1.
2490L-10/06 to
2. Added “Data Retention” on page 8.
Rev. 2490M-08/07
3. Updated “Errata” on page 379.
4. Updated “Assembly Code Example(1)” on page 177.
5. Updated “Slave Mode” on page 167.
Changes from Rev. 1. Added note to “Timer/Counter Oscillator” on page 45.
2490K-04/06 to
2. Updated “Fast PWM Mode” on page 125.
Rev. 2490L-10/06
3. Updated Table 52 on page 104, Table 54 on page 105, Table 59 on page 134, Table 61
on page 136, Table 64 on page 158, and Table 66 on page 158.
4. Updated “Errata” on page 379.
Changes from Rev. 1. Updated Figure 2 on page 3.
2490J-03/05 to
2. Added “Resources” on page 8.
Rev. 2490K-04/06
3. Added Addresses in Register Descriptions.
4. Updated “SPI – Serial Peripheral Interface” on page 163.
5. Updated Register- and bit names in “USART” on page 171.
6. Updated note in “Bit Rate Generator Unit” on page 204.
7. Updated Features in “Analog to Digital Converter” on page 230.
Changes from Rev. 1. MLF-package alternative changed to “Quad Flat No-Lead/Micro Lead Frame Package
QFN/MLF”.
2490I-10/04 to Rev.
2490J-03/05
2. Updated “Electrical Characteristics” on page 325
3. Updated “Ordering Information” on page 376
381
2490N–AVR–05/08
Changes from Rev. 1. Removed “Preliminary” and TBD’s.
2490H-10/04 to
2. Updated Table 8 on page 40, Table 11 on page 42, Table 19 on page 52, Table 132 on
Rev. 2490I-11/04
page 327, Table 134 on page 330.
3. Updated features in “Analog to Digital Converter” on page 230.
4. Updated “Electrical Characteristics” on page 325.
Changes from Rev. 1. Removed references to Analog Ground, IC1/IC3 changed to ICP1/ICP3, Input Capture
Trigger changed to Input Capture Pin.
2490G-03/04 to
Rev. 2490H-10/04
2. Updated “ATmega103 and ATmega64 Compatibility” on page 4.
3. Updated “External Memory Interface” on page 27
4. Updated “XDIV – XTAL Divide Control Register” to “Clock Sources” on page 38.
5. Updated code example in “WDTCR – Watchdog Timer Control Register” on page 57.
6. Added section “Unconnected Pins” on page 70.
7. Updated Table 19 on page 52, Table 20 on page 56, Table 95 on page 236, and
Table 60 on page 135.
8. Updated Figure 116 on page 239.
9. Updated “Version” on page 255.
10. Updated “DC Characteristics” on page 325.
11. Updated “Typical Characteristics” on page 340.
12. Updated features in“Analog to Digital Converter” on page 230 and Table 136 on page
333.
13. Updated “Ordering Information” on page 376.
Changes from Rev. 1. Updated “Errata” on page 379.
2490F-12/03 to
Rev. 2490G-03/04
Changes from Rev. 1. Updated “Calibrated Internal RC Oscillator” on page 43.
2490E-09/03 to
Rev. 2490F-12/03
Changes from Rev. 1. Updated note in “XDIV – XTAL Divide Control Register” on page 39.
2490D-02/03 to
2. Updated “JTAG Interface and On-chip Debug System” on page 50.
Rev. 2490E-09/03
3. Updated “TAP – Test Access Port” on page 248 regarding JTAGEN.
382
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
4. Updated description for the JTD bit on page 258.
5. Added a note regarding JTAGEN fuse to Table 118 on page 292.
6. Updated RPU values in “DC Characteristics” on page 325.
7. Updated “ADC Characteristics” on page 332.
8. Added a proposal for solving problems regarding the JTAG instruction IDCODE in
“Errata” on page 379.
Changes from Rev. 1. Added reference to Table 124 on page 296 from both SPI Serial Programming and Self
Programming to inform about the Flash page size.
2490C-09/02 to
Rev. 2490D-02/03
2. Added Chip Erase as a first step under “Programming the Flash” on page 322 and
“Programming the EEPROM” on page 323.
3. Corrected OCn waveforms in Figure 52 on page 126.
4. Various minor Timer1 corrections.
5. Improved the description in “Phase Correct PWM Mode” on page 101 and on page
153.
6. Various minor TWI corrections.
7. Added note under "Filling the Temporary Buffer (Page Loading)" about writing 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 14.
10. Added section “EEPROM Write During Power-down Sleep Mode” on page 25.
11. Changed VHYST value to 120 in Table 19 on page 52.
12. Added information about conversion time for Differential mode with Auto Triggering
on page 234.
13. Added tWD_FUSE in Table 128 on page 308.
14. Updated “Packaging Information” on page 377.
Changes from Rev. 1. Changed the Endurance on the Flash to 10,000 Write/Erase Cycles.
2490B-09/02 to
Rev. 2490C-09/02
Changes from Rev. 1. Added 64-pad QFN/MLF Package and updated “Ordering Information” on page 376.
2490A-10/01 to
2. Added the section “Using all Locations of External Memory Smaller than 64 KB” on
Rev. 2490B-09/02
page 35.
383
2490N–AVR–05/08
3. Added the section “Default Clock Source” on page 39.
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 24, Table 7 on page 38, Table 9 on page 41, Table 10 on page 41, Table
12 on page 42, Table 14 on page 43, Table 16 on page 44, Table 19 on page 52, Table 20
on page 56, Table 22 on page 58, “DC Characteristics” on page 325, Table 131 on page
327, Table 134 on page 330, Table 136 on page 333, and Table 137 - Table 144.
6. Removed Alternative Algortihm for Leaving JTAG Programming Mode.
See “Leaving Programming Mode” on page 321.
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 306 are updated to also reflect that AVCC
must be connected during Programming mode. Figure 142 on page 301 added to illustrate
how to program the fuses.
9. Added a note regarding usage of the “PROG_PAGELOAD
“PROG_PAGEREAD (0x7)” instructions on page 313.
(0x6)”
and
10. Updated “TWI – Two-wire Serial Interface” on page 198.
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 204. Added the description at the end of “Address Match Unit”
on page 205.
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 “OSCCAL – Oscillator Calibration Register(1)” on page 43 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 44 and Table 131 on page 327.
13. Added a sub section regarding OCD-system and power consumption in the section
“Minimizing Power Consumption” on page 49.
14. Corrected typo (WGM-bit setting) for:
–
“Fast PWM Mode” on page 99 (Timer/Counter0).
–
“Phase Correct PWM Mode” on page 101 (Timer/Counter0).
–
“Fast PWM Mode” on page 152 (Timer/Counter2).
–
“Phase Correct PWM Mode” on page 153 (Timer/Counter2).
15. Corrected Table 81 on page 192 (USART).
384
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
16. Corrected Table 102 on page 262 (Boundary-Scan)
385
2490N–AVR–05/08
386
ATmega64(L)
2490N–AVR–05/08
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
Resources 8
Data Retention 8
About Code Examples 9
AVR CPU Core 10
Introduction 10
Architectural Overview 10
ALU – Arithmetic Logic Unit 11
Status Register 12
General Purpose Register File 13
Stack Pointer 14
Instruction Execution Timing 14
Reset and Interrupt Handling 15
AVR Memories 18
In-System Reprogrammable Flash Program Memory 18
SRAM Data Memory 19
EEPROM Data Memory 21
I/O Memory 26
External Memory Interface 27
XMEM Register Description 32
System Clock and Clock Options 37
Clock Systems and their Distribution 37
Clock Sources 38
Default Clock Source 39
Crystal Oscillator 39
Low-frequency Crystal Oscillator 41
External RC Oscillator 42
Calibrated Internal RC Oscillator 43
External Clock 44
Timer/Counter Oscillator 45
Power Management and Sleep Modes 46
i
2490N–AVR–05/08
Idle Mode 47
ADC Noise Reduction Mode 47
Power-down Mode 47
Power-save Mode 47
Standby Mode 48
Extended Standby Mode 48
Minimizing Power Consumption 49
System Control and Reset 51
Internal Voltage Reference 56
Watchdog Timer 56
Timed Sequences for Changing the Configuration of the Watchdog Timer 60
Interrupts 61
Interrupt Vectors in ATmega64 61
I/O Ports 66
Introduction 66
Ports as General Digital I/O 66
Alternate Port Functions 71
Register Description for I/O Ports 87
External Interrupts 90
8-bit Timer/Counter0 with PWM and Asynchronous Operation 93
Overview 93
Timer/Counter Clock Sources 94
Counter Unit 95
Output Compare Unit 95
Compare Match Output Unit 97
Modes of Operation 98
Timer/Counter Timing Diagrams 102
8-bit Timer/Counter Register Description 104
Asynchronous Operation of the Timer/Counter 107
Timer/Counter Prescaler 110
16-bit Timer/Counter (Timer/Counter1 and Timer/Counter3) 112
Overview 112
Accessing 16-bit Registers 115
Timer/Counter Clock Sources 117
Counter Unit 117
Input Capture Unit 119
Output Compare Units 121
Compare Match Output Unit 122
Modes of Operation 124
Timer/Counter Timing Diagrams 131
ii
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
16-bit Timer/Counter Register Description 132
Timer/Counter3, Timer/Counter2 and Timer/Counter1 Prescalers 144
8-bit Timer/Counter2 with PWM 146
Overview 146
Timer/Counter Clock Sources 147
Counter Unit 148
Output Compare Unit 148
Compare Match Output Unit 150
Modes of Operation 151
Timer/Counter Timing Diagrams 155
8-bit Timer/Counter Register Description 157
Output Compare Modulator (OCM1C2) 161
Overview 161
Description 161
SPI – Serial Peripheral Interface 163
SS Pin Functionality 167
Data Modes 170
USART 171
Overview 171
Clock Generation 172
Frame Formats 175
USART Initialization 176
Data Transmission – The USART Transmitter 178
Data Reception – The USART Receiver 181
Asynchronous Data Reception 184
Multi-processor Communication Mode 187
USART Register Description 188
Examples of Baud Rate Setting 193
TWI – Two-wire Serial Interface 198
Features 198
Two-wire Serial Interface Bus Definition 198
Data Transfer and Frame Format 199
Multi-master Bus Systems, Arbitration and Synchronization 202
Overview of the TWI Module 204
TWI Register Description 206
Using the TWI 209
Transmission Modes 212
Multi-master Systems and Arbitration 225
Analog Comparator 227
iii
2490N–AVR–05/08
Analog Comparator Multiplexed Input 229
Analog to Digital Converter 230
Features 230
Operation 231
Starting a Conversion 232
Prescaling and Conversion Timing 233
Changing Channel or Reference Selection 236
ADC Noise Canceler 237
ADC Conversion Result 242
JTAG Interface and On-chip Debug System 248
Features 248
Overview 248
TAP – Test Access Port 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
iv
ATmega64(L)
2490N–AVR–05/08
ATmega64(L)
Calibration Byte 293
Parallel Programming Parameters, Pin Mapping, and Commands 293
Parallel Programming 297
Serial Downloading 305
SPI Serial Programming Pin Mapping 306
Programming Via the JTAG Interface 311
Electrical Characteristics 325
Absolute Maximum Ratings* 325
DC Characteristics 325
External Clock Drive Waveforms 327
External Clock Drive 327
Two-wire Serial Interface Characteristics 328
SPI Timing Characteristics 330
ADC Characteristics 332
External Data Memory Timing 335
Typical Characteristics 340
Register Summary 370
Instruction Set Summary 373
Ordering Information 376
Packaging Information 377
64A 377
64M1 378
Errata 379
ATmega64, rev. A to C 379
Datasheet Revision History 381
Changes from Rev. 2490M-08/07 to Rev. 2490N-05/08 381
Changes from Rev. 2490L-10/06 to Rev. 2490M-08/07 381
Changes from Rev. 2490K-04/06 to Rev. 2490L-10/06 381
Changes from Rev. 2490J-03/05 to Rev. 2490K-04/06 381
Changes from Rev. 2490I-10/04 to Rev. 2490J-03/05 381
Changes from Rev. 2490H-10/04 to Rev. 2490I-11/04 382
Changes from Rev. 2490G-03/04 to Rev. 2490H-10/04 382
Changes from Rev. 2490F-12/03 to Rev. 2490G-03/04 382
Changes from Rev. 2490E-09/03 to Rev. 2490F-12/03 382
Changes from Rev. 2490D-02/03 to Rev. 2490E-09/03 382
Changes from Rev. 2490C-09/02 to Rev. 2490D-02/03 383
Changes from Rev. 2490B-09/02 to Rev. 2490C-09/02 383
Changes from Rev. 2490A-10/01 to Rev. 2490B-09/02 383
v
2490N–AVR–05/08
Table of Contents i
vi
ATmega64(L)
2490N–AVR–05/08
Headquarters
International
Atmel Corporation
2325 Orchard Parkway
San Jose, CA 95131
USA
Tel: 1(408) 441-0311
Fax: 1(408) 487-2600
Atmel Asia
Room 1219
Chinachem Golden Plaza
77 Mody Road Tsimshatsui
East Kowloon
Hong Kong
Tel: (852) 2721-9778
Fax: (852) 2722-1369
Atmel Europe
Le Krebs
8, Rue Jean-Pierre Timbaud
BP 309
78054 Saint-Quentin-enYvelines Cedex
France
Tel: (33) 1-30-60-70-00
Fax: (33) 1-30-60-71-11
Atmel 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
Technical Support
[email protected]
Sales Contact
www.atmel.com/contacts
Product Contact
Web Site
www.atmel.com
Literature Requests
www.atmel.com/literature
Disclaimer: The information in this document is provided in connection with Atmel products. No license, express or implied, by estoppel or otherwise, to any
intellectual property right is granted by this document or in connection with the sale of Atmel products. EXCEPT AS SET FORTH IN ATMEL’S TERMS AND CONDITIONS OF SALE LOCATED ON ATMEL’S WEB SITE, ATMEL ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED OR STATUTORY
WARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
PURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL ATMEL BE LIABLE FOR ANY DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF PROFITS, BUSINESS INTERRUPTION, OR LOSS OF INFORMATION) ARISING OUT OF
THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF ATMEL HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Atmel makes no
representations or warranties with respect to the accuracy or completeness of the contents of this document and reserves the right to make changes to specifications
and product descriptions at any time without notice. Atmel does not make any commitment to update the information contained herein. Unless specifically provided
otherwise, Atmel products are not suitable for, and shall not be used in, automotive applications. Atmel’s products are not intended, authorized, or warranted for use
as components in applications intended to support or sustain life.
© 2008 Atmel Corporation. All rights reserved. Atmel ®, logo and combinations thereof, AVR ® and others are registered trademarks or trademarks of Atmel Corporation or its subsidiaries. Other terms and product names may be trademarks of others.
2490N–AVR–05/08